Initial commit

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+assets/mstar-logo.png filter=lfs diff=lfs merge=lfs -text

config.json

@@ -0,0 +1,53 @@
+{
+  "_name_or_path": "openbmb/MiniCPM-Llama3-V-2_5",
+  "architectures": [
+    "MiniCPMV"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_minicpm.MiniCPMVConfig",
+    "AutoModel": "modeling_minicpmv.MiniCPMV",
+    "AutoModelForCausalLM": "modeling_minicpmv.MiniCPMV"
+  },
+  "batch_vision_input": true,
+  "bos_token_id": 128000,
+  "drop_vision_last_layer": false,
+  "eos_token_id": 128001,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "image_size": 448,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 8192,
+  "mm_use_im_start_end": true,
+  "model_type": "minicpmv",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "patch_size": 14,
+  "pretraining_tp": 1,
+  "query_num": 96,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 500000.0,
+  "slice_config": {
+    "max_slice_nums": 9,
+    "model_type": "minicpmv"
+  },
+  "slice_mode": true,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.40.0",
+  "use_cache": true,
+  "vision_config": {
+    "hidden_size": 1152,
+    "image_size": 980,
+    "intermediate_size": 4304,
+    "model_type": "idefics2",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 27,
+    "patch_size": 14
+  },
+  "vocab_size": 128256
+}

configuration_minicpm.py

@@ -0,0 +1,113 @@
+# coding=utf-8
+# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" MiniCPM model configuration"""
+import os
+from typing import Union
+
+from transformers.utils import logging
+from transformers import LlamaConfig, PretrainedConfig
+from transformers.models.idefics2.modeling_idefics2 import Idefics2VisionConfig
+
+logger = logging.get_logger(__name__)
+
+
+class MiniCPMVSliceConfig(PretrainedConfig):
+    model_type = "minicpmv"
+
+    def __init__(
+        self,
+        patch_size=14,
+        max_slice_nums=9,
+        scale_resolution=448,
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+        self.patch_size = patch_size
+        self.max_slice_nums = max_slice_nums
+        self.scale_resolution = scale_resolution
+
+    @classmethod
+    def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os.PathLike], **kwargs) -> "PretrainedConfig":
+        cls._set_token_in_kwargs(kwargs)
+
+        config_dict, kwargs = cls.get_config_dict(pretrained_model_name_or_path, **kwargs)
+
+        if config_dict.get("model_type") == "minicpmv":
+            config_dict = config_dict["slice_config"]
+
+        if "model_type" in config_dict and hasattr(cls, "model_type") and config_dict["model_type"] != cls.model_type:
+            logger.warning(
+                f"You are using a model of type {config_dict['model_type']} to instantiate a model of type "
+                f"{cls.model_type}. This is not supported for all configurations of models and can yield errors."
+            )
+
+        return cls.from_dict(config_dict, **kwargs)
+
+
+
+class MiniCPMVConfig(LlamaConfig):
+    model_type = "minicpmv"
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    default_vision_config = {
+        "hidden_size": 1152,
+        "image_size": 980,
+        "intermediate_size": 4304,
+        "model_type": "idefics2",
+        "num_attention_heads": 16,
+        "num_hidden_layers": 27,
+        "patch_size": 14,
+    }
+
+    def __init__(
+        self,
+        use_cache=True,
+        query_num=64,
+        image_size=448,
+        drop_vision_last_layer=True,
+        batch_vision_input=True,
+        slice_config=None,
+        vision_config=None,
+        **kwargs,
+    ):
+        self.use_cache = use_cache
+        self.query_num = query_num
+        self.image_size = image_size
+        self.drop_vision_last_layer = drop_vision_last_layer
+        self.batch_vision_input = batch_vision_input
+
+        if slice_config is None:
+            self.slice_config = MiniCPMVSliceConfig(max_slice_nums=1)
+        else:
+            self.slice_config = MiniCPMVSliceConfig(**slice_config)
+        self.slice_mode = True
+
+        # same as HuggingFaceM4/siglip-so400m-14-980-flash-attn2-navit
+        if vision_config is None:
+            self.vision_config = Idefics2VisionConfig(**self.default_vision_config)
+            logger.info("vision_config is None, using default vision config")
+        elif isinstance(vision_config, dict):
+            self.vision_config = Idefics2VisionConfig(**vision_config)
+        elif isinstance(vision_config, Idefics2VisionConfig):
+            self.vision_config = vision_config
+
+        self.patch_size = self.vision_config.patch_size
+
+        super().__init__(**kwargs)

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 128000,
+  "eos_token_id": 128001,
+  "transformers_version": "4.42.4"
+}

latest

@@ -0,0 +1 @@
+global_step8435

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ccf45f7c689ed51ab2e7aee9f4937a2bbd557f6908fffb9714c7b810bb1c94a
+size 4976698920

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a605061cb58e3afdd0893f89c16ae594236e3f77687d2bf23cd0a6dd16b76f71
+size 4999803032

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a51c38176f5e4eff73f806df1a581eb6bf1c74e16156477580f8ccb9f6334aaa
+size 4915916480

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8e8f8e6e94fabef155749e140ef352749a3af18e566151a2e977d40b43653cf
+size 2181852600

model.safetensors.index.json

@@ -0,0 +1,748 @@
+{
+  "metadata": {
+    "total_size": 17074184672
+  },
+  "weight_map": {
+    "llm.lm_head.weight": "model-00004-of-00004.safetensors",
+    "llm.model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "llm.model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "llm.model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "llm.model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "llm.model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "llm.model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "llm.model.norm.weight": "model-00004-of-00004.safetensors",
+    "resampler.attn.in_proj_bias": "model-00004-of-00004.safetensors",
+    "resampler.attn.in_proj_weight": "model-00004-of-00004.safetensors",
+    "resampler.attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "resampler.attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "resampler.kv_proj.weight": "model-00004-of-00004.safetensors",
+    "resampler.ln_kv.bias": "model-00004-of-00004.safetensors",
+    "resampler.ln_kv.weight": "model-00004-of-00004.safetensors",
+    "resampler.ln_post.bias": "model-00004-of-00004.safetensors",
+    "resampler.ln_post.weight": "model-00004-of-00004.safetensors",
+    "resampler.ln_q.bias": "model-00004-of-00004.safetensors",
+    "resampler.ln_q.weight": "model-00004-of-00004.safetensors",
+    "resampler.proj": "model-00004-of-00004.safetensors",
+    "resampler.query": "model-00004-of-00004.safetensors",
+    "vpm.embeddings.patch_embedding.bias": "model-00004-of-00004.safetensors",
+    "vpm.embeddings.patch_embedding.weight": "model-00004-of-00004.safetensors",
+    "vpm.embeddings.position_embedding.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.0.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.1.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.10.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.11.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.12.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.13.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.14.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.15.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.16.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.17.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.18.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.19.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.2.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.20.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.21.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.22.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.23.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.24.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.25.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.26.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.3.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.4.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.5.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.6.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.7.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.8.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.layer_norm1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.layer_norm1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.layer_norm2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.layer_norm2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.mlp.fc1.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.mlp.fc1.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.mlp.fc2.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.mlp.fc2.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.self_attn.out_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.self_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "vpm.encoder.layers.9.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "vpm.post_layernorm.bias": "model-00004-of-00004.safetensors",
+    "vpm.post_layernorm.weight": "model-00004-of-00004.safetensors"
+  }
+}

modeling_minicpmv.py

@@ -0,0 +1,702 @@
+import math
+from typing import List, Optional
+import json
+import torch
+import torchvision
+from threading import Thread
+from copy import deepcopy
+from PIL import Image
+from torchvision import transforms
+from transformers import LlamaTokenizer, LlamaPreTrainedModel, LlamaForCausalLM, AutoModel, PreTrainedTokenizerFast, TextIteratorStreamer
+from transformers.models.idefics2.modeling_idefics2 import Idefics2VisionTransformer
+
+from .configuration_minicpm import MiniCPMVConfig
+from .resampler import Resampler
+
+IMAGENET_INCEPTION_MEAN = (0.5, 0.5, 0.5) # timm.data.IMAGENET_INCEPTION_MEAN
+IMAGENET_INCEPTION_STD = (0.5, 0.5, 0.5)  # timm.data.IMAGENET_INCEPTION_STD
+
+class MiniCPMVPreTrainedModel(LlamaPreTrainedModel):
+    config_class = MiniCPMVConfig
+
+
+class MiniCPMV(MiniCPMVPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+
+        self.llm = LlamaForCausalLM(config)
+        self.vpm = self.init_vision_module()
+        self.vision_dim = self.vpm.embed_dim
+        self.embed_dim = self.llm.config.hidden_size
+        self.resampler = self.init_resampler(self.embed_dim, self.vision_dim)
+        self.transform = self.init_transform()
+
+    def init_vision_module(self):
+        # same as HuggingFaceM4/siglip-so400m-14-980-flash-attn2-navit
+        model = Idefics2VisionTransformer(self.config.vision_config)
+        if self.config.drop_vision_last_layer:
+            model.encoder.layers = model.encoder.layers[:-1]
+
+        setattr(model, 'embed_dim', model.embeddings.embed_dim)
+        setattr(model, 'patch_size', model.embeddings.patch_size)
+
+        return model
+
+    def init_resampler(self, embed_dim, vision_dim,):
+        return Resampler(
+            num_queries=self.config.query_num,
+            embed_dim=embed_dim,
+            num_heads=embed_dim // 128,
+            kv_dim=vision_dim,
+            adaptive=True,
+        )
+
+    def init_transform(self):
+        return transforms.Compose(
+            [
+                transforms.ToTensor(),
+                transforms.Normalize(
+                    mean=IMAGENET_INCEPTION_MEAN, std=IMAGENET_INCEPTION_STD
+                ),
+            ]
+        )
+    
+    def get_input_embeddings(self):
+        return self.llm.get_input_embeddings()
+
+    def set_input_embeddings(self, value):
+        self.llm.embed_tokens = value
+        
+    def get_vllm_embedding(self, data):
+        if 'vision_hidden_states' not in data:
+            dtype = self.llm.model.embed_tokens.weight.dtype
+            device = self.llm.model.embed_tokens.weight.device
+            tgt_sizes = data['tgt_sizes']
+            pixel_values_list = data['pixel_values']
+            vision_hidden_states = []
+            all_pixel_values = []
+            img_cnt = []
+            for pixel_values in pixel_values_list:
+                img_cnt.append(len(pixel_values))
+                all_pixel_values.extend([i.flatten(end_dim=1).permute(1, 0) for i in pixel_values])
+
+            # exist image
+            if all_pixel_values:
+                tgt_sizes = torch.vstack(tgt_sizes).type(torch.int32)
+
+                if self.config.batch_vision_input:
+                    max_patches = torch.max(tgt_sizes[:, 0] * tgt_sizes[:, 1])
+
+                    all_pixel_values = torch.nn.utils.rnn.pad_sequence(all_pixel_values, batch_first=True,
+                                                                       padding_value=0.0)
+                    B, L, _ = all_pixel_values.shape
+                    all_pixel_values = all_pixel_values.permute(0, 2, 1).reshape(B, 3, -1, L)
+
+                    patch_attn_mask = torch.zeros((B, 1, max_patches), dtype=torch.bool, device=device)
+                    for i in range(B):
+                        patch_attn_mask[i, :tgt_sizes[i][0] * tgt_sizes[i][1]] = True
+
+                    vision_embedding = self.vpm(all_pixel_values.type(dtype), patch_attention_mask=patch_attn_mask).last_hidden_state
+                    vision_embedding = self.resampler(vision_embedding, tgt_sizes)
+                else:
+                    # get vision_embedding foreach
+                    vision_embedding = []
+                    for single_tgt_size, single_pixel_values in zip(tgt_sizes, all_pixel_values):
+                        single_pixel_values = single_pixel_values.unsqueeze(0)
+                        B, L, _ = single_pixel_values.shape
+                        single_pixel_values = single_pixel_values.permute(0, 2, 1).reshape(B, 3, -1, L)
+                        single_vision_embedding = self.vpm(single_pixel_values.type(dtype)).last_hidden_state
+                        single_vision_embedding = self.resampler(single_vision_embedding, single_tgt_size.unsqueeze(0))
+                        
+                        vision_embedding.append(single_vision_embedding)
+                    vision_embedding = torch.vstack(vision_embedding)
+
+                start = 0
+                for pixel_values in pixel_values_list:
+                    img_cnt = len(pixel_values)
+                    if img_cnt > 0:
+                        vision_hidden_states.append(vision_embedding[start: start + img_cnt])
+                        start += img_cnt
+                    else:
+                        vision_hidden_states.append([])
+            else: # no image
+                if self.training:
+                    dummy_image = torch.zeros(
+                        (1, 3, 224, 224),
+                        device=device, dtype=dtype
+                    )
+                    tgt_sizes = torch.Tensor([[(224 // self.config.patch_size), math.ceil(224 / self.config.patch_size)]]).type(torch.int32)
+                    dummy_feature = self.resampler(self.vpm(dummy_image).last_hidden_state, tgt_sizes)
+                else:
+                    dummy_feature = []
+                for _ in range(len(pixel_values_list)):
+                    vision_hidden_states.append(dummy_feature)
+
+        else:
+            vision_hidden_states = data['vision_hidden_states']
+
+        if hasattr(self.llm.config, 'scale_emb'):
+            vllm_embedding = self.llm.model.embed_tokens(data['input_ids']) * self.llm.config.scale_emb
+        else:
+            vllm_embedding = self.llm.model.embed_tokens(data['input_ids'])
+
+        vision_hidden_states = [i.type(vllm_embedding.dtype) if isinstance(
+            i, torch.Tensor) else i for i in vision_hidden_states]
+
+        bs = len(data['input_ids'])
+        for i in range(bs):
+            cur_vs_hs = vision_hidden_states[i]
+            if len(cur_vs_hs) > 0:
+                cur_vllm_emb = vllm_embedding[i]
+                cur_image_bound = data['image_bound'][i]
+                if len(cur_image_bound) > 0:
+                    image_indices = torch.stack(
+                        [torch.arange(r[0], r[1], dtype=torch.long) for r in cur_image_bound]
+                    ).to(vllm_embedding.device)
+                    cur_vllm_emb.scatter_(0, image_indices.view(-1, 1).repeat(1, cur_vllm_emb.shape[-1]),
+                                          cur_vs_hs.view(-1, cur_vs_hs.shape[-1]))
+                elif self.training:
+                    cur_vllm_emb += cur_vs_hs[0].mean() * 0
+
+        return vllm_embedding, vision_hidden_states
+        
+    def forward(self, data, **kwargs):
+        vllm_embedding, vision_hidden_states = self.get_vllm_embedding(data)
+        position_ids = data["position_ids"]
+        if position_ids.dtype != torch.int64:
+            position_ids = position_ids.long()
+
+        return self.llm(
+            input_ids=None,
+            position_ids=position_ids,
+            inputs_embeds=vllm_embedding,
+            **kwargs
+        )
+
+    def _convert_to_tensors(
+        self, tokenizer, input_ids, max_inp_length: Optional[int] = None
+    ):
+        if max_inp_length is not None:
+            input_ids = input_ids[:max_inp_length]
+        input_ids = torch.tensor(input_ids, dtype=torch.int32)
+
+        image_start_tokens = torch.where(input_ids == tokenizer.im_start_id)[0]
+        # 跳过 im_start
+        image_start_tokens += 1
+        image_end_tokens = torch.where(input_ids == tokenizer.im_end_id)[0]
+        valid_image_nums = max(len(image_start_tokens), len(image_end_tokens))
+        image_bound = torch.hstack(
+            [
+                image_start_tokens[:valid_image_nums].unsqueeze(-1),
+                image_end_tokens[:valid_image_nums].unsqueeze(-1),
+            ]
+        )
+
+        model_input = {}
+        model_input["input_ids"] = input_ids.unsqueeze(0).to(self.device)
+        model_input["image_bound"] = image_bound
+
+        return model_input
+
+    def _process_list(
+        self, tokenizer, input_id_list, max_inp_length: Optional[int] = None
+    ):
+        pad_keys = ["input_ids"]
+        input_tensors = []
+        for input_ids in input_id_list:
+            input_tensors.append(
+                self._convert_to_tensors(tokenizer, input_ids, max_inp_length)
+            )
+        padded = {}
+        for key in pad_keys:
+            padded[key] = pad(input_tensors, key, padding_side="left").to(self.device)
+        padded["image_bound"] = [i["image_bound"] for i in input_tensors]
+        return padded
+
+    def _decode(self, inputs_embeds, tokenizer, **kwargs):
+        terminators = [
+            tokenizer.eos_token_id,
+            tokenizer.convert_tokens_to_ids("<|eot_id|>")
+        ]
+        output = self.llm.generate(
+            inputs_embeds=inputs_embeds,
+            pad_token_id=0,
+            eos_token_id=terminators,
+            **kwargs
+        )
+        return self._decode_text(output, tokenizer)
+    
+    def _decode_stream(self, inputs_embeds, tokenizer, **kwargs):
+        terminators = [
+            tokenizer.eos_token_id,
+            tokenizer.convert_tokens_to_ids("<|eot_id|>")
+        ]
+        streamer = TextIteratorStreamer(tokenizer=tokenizer)
+        generation_kwargs = {
+            'inputs_embeds': inputs_embeds,
+            'pad_token_id': 0,
+            'eos_token_id': terminators,
+            'streamer': streamer
+        }
+        generation_kwargs.update(kwargs)
+
+        thread = Thread(target=self.llm.generate, kwargs=generation_kwargs)
+        thread.start()
+    
+        return streamer
+
+    def _decode_text(self, result_ids, tokenizer):
+        result_text = []
+        for result in result_ids:
+            result = result[result != 0]
+            if result[0] == tokenizer.bos_id:
+                result = result[1:]
+            if result[-1] == tokenizer.eos_id or result[-1] == tokenizer.eot_id:
+                result = result[:-1]
+            result_text.append(tokenizer.decode(result).strip())
+        return result_text
+
+    def slice_image(self, image):
+        return slice_image(
+            image,
+            self.config.slice_config.max_slice_nums,
+            self.config.slice_config.scale_resolution,
+            self.config.slice_config.patch_size,
+        )
+
+    def get_slice_image_placeholder(self, image, tokenizer):
+        image_placeholder = (
+            tokenizer.im_start
+            + tokenizer.unk_token * self.config.query_num
+            + tokenizer.im_end
+        )
+
+        slice_images = []
+
+        source_image, patches, best_grid = slice_image(
+            image,
+            self.config.slice_config.max_slice_nums,
+            self.config.slice_config.scale_resolution,
+            self.config.slice_config.patch_size,
+        )
+
+        slice_images.append(source_image)
+        final_placeholder = image_placeholder
+
+        if len(patches) > 0:
+            for i in range(len(patches)):
+                for j in range(len(patches[0])):
+                    slice_images.append(patches[i][j])
+
+            final_placeholder += get_grid_placeholder(
+                tokenizer, best_grid, self.config.query_num
+            )
+
+        return slice_images, final_placeholder
+
+    def reshape_by_patch(self, image_tensor):
+        """
+        :param image_tensor: shape [3, H, W]
+        :param patch_size:
+        :return: [3, patch_size, HW/patch_size]
+        """
+        patch_size = self.config.patch_size
+        patches = torch.nn.functional.unfold(
+            image_tensor,
+            (patch_size, patch_size),
+            stride=(patch_size, patch_size)
+        )
+
+        patches = patches.reshape(image_tensor.size(0), patch_size, patch_size, -1)
+        patches = patches.permute(0, 1, 3, 2).reshape(image_tensor.size(0), patch_size, -1)
+        return patches
+
+    def generate(
+        self,
+        input_id_list=None,
+        img_list=None,
+        tgt_sizes=None,
+        tokenizer=None,
+        max_inp_length: Optional[int] = None,
+        vision_hidden_states=None,
+        return_vision_hidden_states=False,
+        stream=False,
+        **kwargs
+    ):
+
+        assert input_id_list is not None
+        bs = len(input_id_list)
+        if img_list == None:
+            img_list = [[] for i in range(bs)]
+        assert bs == len(img_list)
+
+        model_inputs = self._process_list(tokenizer, input_id_list, max_inp_length)
+
+        if vision_hidden_states is None:
+            pixel_values = []
+            for i in range(bs):
+                img_inps = []
+                for img in img_list[i]:
+                    img_inps.append(img.to(self.device))
+                if img_inps:
+                    pixel_values.append(img_inps)
+                else:
+                    pixel_values.append([])
+            model_inputs["pixel_values"] = pixel_values
+            model_inputs['tgt_sizes'] = tgt_sizes
+        else:
+            model_inputs["vision_hidden_states"] = vision_hidden_states
+
+        with torch.inference_mode():
+            (
+                model_inputs["inputs_embeds"],
+                vision_hidden_states,
+            ) = self.get_vllm_embedding(model_inputs)
+
+            if stream:
+                result = self._decode_stream(model_inputs["inputs_embeds"], tokenizer, **kwargs)
+            else:
+                result = self._decode(model_inputs["inputs_embeds"], tokenizer, **kwargs)
+
+        if return_vision_hidden_states:
+            return result, vision_hidden_states
+
+        return result
+
+    def chat(
+        self,
+        image,
+        msgs,
+        tokenizer,
+        vision_hidden_states=None,
+        max_new_tokens=1024,
+        sampling=True,
+        max_inp_length=2048,
+        system_prompt='',
+        stream=False,
+        **kwargs
+    ):
+        if isinstance(msgs, str):
+            msgs = json.loads(msgs)
+
+        copy_msgs = deepcopy(msgs)
+        assert len(copy_msgs) > 0, 'msgs is empty'
+        assert sampling or not stream, 'if use stream mode, make sure sampling=True'
+
+        if image is not None and isinstance(copy_msgs[0]['content'], str):
+            copy_msgs[0]['content'] = [image, copy_msgs[0]['content']]
+
+        images = []
+        tgt_sizes = []
+        for i, msg in enumerate(copy_msgs):
+            role = msg["role"]
+            content = msg["content"]
+            assert role in ["user", "assistant"]
+            if i == 0:
+                assert role == "user", "The role of first msg should be user"
+            if isinstance(content, str):
+                content = [content]
+
+            cur_msgs = []
+            for c in content:
+                if isinstance(c, Image.Image):
+                    image = c
+                    if self.config.slice_mode:
+                        slice_images, image_placeholder = self.get_slice_image_placeholder(
+                            image, tokenizer
+                        )
+                        cur_msgs.append(image_placeholder)
+                        for slice_image in slice_images:
+                            slice_image = self.transform(slice_image)
+                            H, W = slice_image.shape[1:]
+                            images.append(self.reshape_by_patch(slice_image))
+                            tgt_sizes.append(torch.Tensor([H // self.config.patch_size, W // self.config.patch_size]).type(torch.int32))
+                    else:
+                        images.append(self.transform(image))
+                        cur_msgs.append(
+                            tokenizer.im_start
+                            + tokenizer.unk_token * self.config.query_num
+                            + tokenizer.im_end
+                        )
+                elif isinstance(c, str):
+                    cur_msgs.append(c)
+
+
+            msg['content'] = '\n'.join(cur_msgs)
+        if tgt_sizes:
+            tgt_sizes = torch.vstack(tgt_sizes)
+
+        if system_prompt:
+            sys_msg = {'role': 'system', 'content': system_prompt}
+            copy_msgs = [sys_msg] + copy_msgs
+
+        input_ids = tokenizer.apply_chat_template(copy_msgs, tokenize=True, add_generation_prompt=False)
+
+        if sampling:
+            generation_config = {
+                "top_p": 0.8,
+                "top_k": 100,
+                "temperature": 0.7,
+                "do_sample": True,
+                "repetition_penalty": 1.05
+            }
+        else:
+            generation_config = {
+                "num_beams": 3,
+                "repetition_penalty": 1.2,
+            }
+
+        generation_config.update(
+            (k, kwargs[k]) for k in generation_config.keys() & kwargs.keys()
+        )
+
+        with torch.inference_mode():
+            res, vision_hidden_states = self.generate(
+                input_id_list=[input_ids],
+                max_inp_length=max_inp_length,
+                img_list=[images],
+                tgt_sizes=[tgt_sizes],
+                tokenizer=tokenizer,
+                max_new_tokens=max_new_tokens,
+                vision_hidden_states=vision_hidden_states,
+                return_vision_hidden_states=True,
+                stream=stream,
+                **generation_config
+            )
+
+        if stream:
+            def stream_gen():
+                for text in res:
+                    text = text.replace(tokenizer.eot_token, '').replace(tokenizer.eos_token, '')
+                    yield text
+            return stream_gen()
+
+        else:
+            answer = res[0]
+            return answer
+
+
+class PreTrainedTokenizerFastWrapper(PreTrainedTokenizerFast):
+    def __init__(self, **kwargs):
+        super().__init__(**kwargs)
+        self.eot_token = "<|eot_id|>"
+        self.im_start = "<image>"
+        self.im_end = "</image>"
+        self.ref_start = "<ref>"
+        self.ref_end = "</ref>"
+        self.box_start = "<box>"
+        self.box_end = "</box>"
+        self.quad_start = "<quad>"
+        self.quad_end = "</quad>"
+        self.slice_start = "<slice>"
+        self.slice_end = "</slice>"
+
+    @property
+    def eos_id(self):
+        return self.eos_token_id
+
+    @property
+    def bos_id(self):
+        return self.bos_token_id
+
+    @property
+    def unk_id(self):
+        return self.unk_token_id
+
+    @property
+    def eot_id(self):
+        return self.convert_tokens_to_ids(self.eot_token)
+
+    @property
+    def im_start_id(self):
+        return self.convert_tokens_to_ids(self.im_start)
+
+    @property
+    def im_end_id(self):
+        return self.convert_tokens_to_ids(self.im_end)
+
+    @staticmethod
+    def escape(text: str) -> str:
+        return text
+
+    @staticmethod
+    def unescape(text: str) -> str:
+        return text
+
+
+def pad(orig_items, key, max_length=None, padding_value=0, padding_side="left"):
+    items = []
+    if isinstance(orig_items[0][key], list):
+        assert isinstance(orig_items[0][key][0], torch.Tensor)
+        for it in orig_items:
+            for tr in it[key]:
+                items.append({key: tr})
+    else:
+        assert isinstance(orig_items[0][key], torch.Tensor)
+        items = orig_items
+
+    batch_size = len(items)
+    shape = items[0][key].shape
+    dim = len(shape)
+    assert dim <= 3
+    if max_length is None:
+        max_length = 0
+    max_length = max(max_length, max(item[key].shape[-1] for item in items))
+    min_length = min(item[key].shape[-1] for item in items)
+    dtype = items[0][key].dtype
+
+    if dim == 1:
+        return torch.cat([item[key] for item in items], dim=0)
+    elif dim == 2:
+        if max_length == min_length:
+            return torch.cat([item[key] for item in items], dim=0)
+        tensor = torch.zeros((batch_size, max_length), dtype=dtype) + padding_value
+    else:
+        tensor = (
+            torch.zeros((batch_size, max_length, shape[-1]), dtype=dtype)
+            + padding_value
+        )
+
+    for i, item in enumerate(items):
+        if dim == 2:
+            if padding_side == "left":
+                tensor[i, -len(item[key][0]) :] = item[key][0].clone()
+            else:
+                tensor[i, : len(item[key][0])] = item[key][0].clone()
+        elif dim == 3:
+            if padding_side == "left":
+                tensor[i, -len(item[key][0]) :, :] = item[key][0].clone()
+            else:
+                tensor[i, : len(item[key][0]), :] = item[key][0].clone()
+
+    return tensor
+
+
+def slice_image(
+    image, max_slice_nums=9, scale_resolution=448, patch_size=14, never_split=False
+):
+    original_size = image.size
+    original_width, original_height = original_size
+    log_ratio = math.log(original_width / original_height)
+    ratio = original_width * original_height / (scale_resolution * scale_resolution)
+    multiple = min(math.ceil(ratio), max_slice_nums)
+
+    source_image = None
+    best_grid = None
+    patches = []
+
+    if multiple <= 1 or never_split:
+        # dont need to slice, upsample
+        best_size = find_best_resize(
+            original_size, scale_resolution, patch_size, allow_upscale=True
+        )
+        source_image = image.resize(best_size, Image.Resampling.BICUBIC)
+    else:
+        candidate_split_grids_nums = []
+        for i in [multiple - 1, multiple, multiple + 1]:
+            if i == 1 or i > max_slice_nums:
+                continue
+            candidate_split_grids_nums.append(i)
+
+        # source image, down-sampling and ensure divided by patch_size
+        best_resize = find_best_resize(original_size, scale_resolution, patch_size)
+        source_image = image.copy().resize(best_resize, Image.Resampling.BICUBIC)
+        candidate_grids = []
+
+        # find best grid
+        for split_grids_nums in candidate_split_grids_nums:
+            m = 1
+            while m <= split_grids_nums:
+                if split_grids_nums % m == 0:
+                    candidate_grids.append([m, split_grids_nums // m])
+                m += 1
+
+        best_grid = [1, 1]
+        min_error = float("inf")
+        for grid in candidate_grids:
+            error = abs(log_ratio - math.log(grid[0] / grid[1]))
+            if error < min_error:
+                best_grid = grid
+                min_error = error
+
+        refine_size = get_refine_size(
+            original_size, best_grid, scale_resolution, patch_size, allow_upscale=True
+        )
+
+        refine_image = image.resize(refine_size, Image.Resampling.BICUBIC)
+        patches = split_to_patches(refine_image, best_grid)
+
+    return source_image, patches, best_grid
+
+
+def ensure_divide(length, patch_size):
+    return max(round(length / patch_size) * patch_size, patch_size)
+
+
+def find_best_resize(original_size, scale_resolution, patch_size, allow_upscale=False):
+    width, height = original_size
+    if (width * height > scale_resolution * scale_resolution) or allow_upscale:
+        r = width / height
+        height = int(scale_resolution / math.sqrt(r))
+        width = int(height * r)
+    best_width = ensure_divide(width, patch_size)
+    best_height = ensure_divide(height, patch_size)
+    return (best_width, best_height)
+
+
+def get_refine_size(
+    original_size, grid, scale_resolution, patch_size, allow_upscale=False
+):
+    width, height = original_size
+    grid_x, grid_y = grid
+
+    refine_width = ensure_divide(width, grid_x)
+    refine_height = ensure_divide(height, grid_y)
+
+    grid_width = refine_width / grid_x
+    grid_height = refine_height / grid_y
+
+    best_grid_size = find_best_resize(
+        (grid_width, grid_height),
+        scale_resolution,
+        patch_size,
+        allow_upscale=allow_upscale,
+    )
+
+    refine_size = (best_grid_size[0] * grid_x, best_grid_size[1] * grid_y)
+
+    return refine_size
+
+
+def split_to_patches(image, grid):
+    patches = []
+    width, height = image.size
+    grid_x = int(width / grid[0])
+    grid_y = int(height / grid[1])
+
+    for i in range(0, height, grid_y):
+        images = []
+        for j in range(0, width, grid_x):
+            box = (j, i, j + grid_x, i + grid_y)
+            patch = image.crop(box)
+            images.append(patch)
+        patches.append(images)
+
+    return patches
+
+
+def get_grid_placeholder(tokenizer, grid, query_num):
+    image_placeholder = (
+        tokenizer.im_start + tokenizer.unk_token * query_num + tokenizer.im_end
+    )
+
+    cols = grid[0]
+    rows = grid[1]
+    slices = []
+    for i in range(rows):
+        lines = []
+        for j in range(cols):
+            lines.append(image_placeholder)
+        slices.append("".join(lines))
+    slice_placeholder = tokenizer.slice_start + "\n".join(slices) + tokenizer.slice_end
+    return slice_placeholder

resampler.py

@@ -0,0 +1,812 @@
+from functools import partial
+import numpy as np
+import warnings
+from typing import Optional, Tuple
+import torch
+from torch import nn
+from torch import Tensor
+import torch.nn.functional as F 
+from torch.nn.functional import *
+from torch.nn.modules.activation import *
+from torch.nn.init import trunc_normal_
+from torch.nn.init import constant_, xavier_normal_, xavier_uniform_
+from transformers import PreTrainedModel
+from transformers.integrations import is_deepspeed_zero3_enabled
+
+def get_2d_sincos_pos_embed(embed_dim, image_size):
+    """
+    image_size: image_size or (image_height, image_width)
+    return:
+    pos_embed: [image_height, image_width, embed_dim]
+    """
+    if isinstance(image_size, int):
+        grid_h_size, grid_w_size = image_size, image_size
+    else:
+        grid_h_size, grid_w_size = image_size[0], image_size[1]
+
+    grid_h = np.arange(grid_h_size, dtype=np.float32)
+    grid_w = np.arange(grid_w_size, dtype=np.float32)
+    grid = np.meshgrid(grid_w, grid_h)  # here w goes first
+    grid = np.stack(grid, axis=0)
+
+    pos_embed = get_2d_sincos_pos_embed_from_grid(embed_dim, grid)
+    return pos_embed
+
+
+def get_2d_sincos_pos_embed_from_grid(embed_dim, grid):
+    assert embed_dim % 2 == 0
+
+    # use half of dimensions to encode grid_h
+    emb_h = get_1d_sincos_pos_embed_from_grid_new(embed_dim // 2, grid[0])  # (H, W, D/2)
+    emb_w = get_1d_sincos_pos_embed_from_grid_new(embed_dim // 2, grid[1])  # (H, W, D/2)
+
+    emb = np.concatenate([emb_h, emb_w], axis=-1)  # (H, W, D)
+    return emb
+
+
+def get_1d_sincos_pos_embed_from_grid_new(embed_dim, pos):
+    """
+    embed_dim: output dimension for each position
+    pos: a list of positions to be encoded: size (H, W)
+    out: (H, W, D)
+    """
+    assert embed_dim % 2 == 0
+    omega = np.arange(embed_dim // 2, dtype=np.float32)
+    omega /= embed_dim / 2.
+    omega = 1. / 10000 ** omega  # (D/2,)
+
+    out = np.einsum('hw,d->hwd', pos, omega)  # (H, W, D/2), outer product
+
+    emb_sin = np.sin(out)  # (H, W, D/2)
+    emb_cos = np.cos(out)  # (H, W, D/2)
+
+    emb = np.concatenate([emb_sin, emb_cos], axis=-1)  # (H, W, D)
+    return emb
+
+    
+class Resampler(nn.Module):
+    """
+    A 2D perceiver-resampler network with one cross attention layers by
+       given learnable queries and 2d sincos pos_emb
+    Outputs:
+        A tensor with the shape of (batch_size, num_queries, embed_dim)
+    """
+
+    def __init__(
+            self,
+            num_queries,
+            embed_dim,
+            num_heads,
+            kv_dim=None,
+            norm_layer=partial(nn.LayerNorm, eps=1e-6),
+            adaptive=False,
+            max_size=(70, 70),
+    ):
+        super().__init__()
+        self.num_queries = num_queries
+        self.embed_dim = embed_dim
+        self.num_heads = num_heads
+        self.adaptive = adaptive
+        self.max_size = max_size
+
+        self.query = nn.Parameter(torch.zeros(self.num_queries, embed_dim))
+
+        if kv_dim is not None and kv_dim != embed_dim:
+            self.kv_proj = nn.Linear(kv_dim, embed_dim, bias=False)
+        else:
+            self.kv_proj = nn.Identity()
+
+        self.attn = MultiheadAttention(embed_dim, num_heads)
+        self.ln_q = norm_layer(embed_dim)
+        self.ln_kv = norm_layer(embed_dim)
+
+        self.ln_post = norm_layer(embed_dim)
+        self.proj = nn.Parameter((embed_dim ** -0.5) * torch.randn(embed_dim, embed_dim))
+
+        self._set_2d_pos_cache(self.max_size)
+
+    def _set_2d_pos_cache(self, max_size, device='cpu'):
+        if is_deepspeed_zero3_enabled():
+            device='cuda'
+        pos_embed = torch.from_numpy(get_2d_sincos_pos_embed(self.embed_dim, max_size)).float().to(device)
+        self.register_buffer("pos_embed", pos_embed, persistent=False)
+
+    def _adjust_pos_cache(self, tgt_sizes, device):
+        max_h = torch.max(tgt_sizes[:, 0])
+        max_w = torch.max(tgt_sizes[:, 1])
+        if max_h > self.max_size[0] or max_w > self.max_size[1]:
+            self.max_size = [max(max_h, self.max_size[0]), max(max_w, self.max_size[1])]
+            self._set_2d_pos_cache(self.max_size, device)
+
+    def _init_weights(self, m):
+        if isinstance(m, nn.Linear):
+            trunc_normal_(m.weight, std=.02)
+            if isinstance(m, nn.Linear) and m.bias is not None:
+                nn.init.constant_(m.bias, 0)
+        elif isinstance(m, nn.LayerNorm):
+            nn.init.constant_(m.bias, 0)
+            nn.init.constant_(m.weight, 1.0)
+
+    def forward(self, x, tgt_sizes=None):
+        assert x.shape[0] == tgt_sizes.shape[0]
+        bs = x.shape[0]
+
+        device = x.device
+        dtype = x.dtype
+
+        patch_len = tgt_sizes[:, 0] * tgt_sizes[:, 1]
+
+        self._adjust_pos_cache(tgt_sizes, device=device)
+
+        max_patch_len = torch.max(patch_len)
+        key_padding_mask = torch.zeros((bs, max_patch_len), dtype=torch.bool, device=device)
+
+        pos_embed = []
+        for i in range(bs):
+            tgt_h, tgt_w = tgt_sizes[i]
+            pos_embed.append(self.pos_embed[:tgt_h, :tgt_w, :].reshape((tgt_h * tgt_w, -1)).to(dtype))  # patches * D
+            key_padding_mask[i, patch_len[i]:] = True
+
+        pos_embed = torch.nn.utils.rnn.pad_sequence(
+            pos_embed, batch_first=True, padding_value=0.0).permute(1, 0, 2)  # BLD => L * B * D
+
+        x = self.kv_proj(x)  # B * L * D
+        x = self.ln_kv(x).permute(1, 0, 2)  # L * B * D
+
+        q = self.ln_q(self.query)  # Q * D
+
+        out = self.attn(
+            self._repeat(q, bs),  # Q * B * D
+            x + pos_embed,  # L * B * D +  L * B * D
+            x,
+            key_padding_mask=key_padding_mask)[0]
+        #  out: Q * B * D
+        x = out.permute(1, 0, 2)  # B * Q * D
+
+        x = self.ln_post(x)
+        x = x @ self.proj
+        return x
+
+    def _repeat(self, query, N: int):
+        return query.unsqueeze(1).repeat(1, N, 1)
+
+
+class MultiheadAttention(nn.MultiheadAttention):
+    def __init__(self, embed_dim, num_heads, dropout=0., bias=True, add_bias_kv=False, 
+                 add_zero_attn=False, kdim=None, vdim=None, batch_first=False, device=None, dtype=None):
+        super().__init__(embed_dim, num_heads, dropout, bias, add_bias_kv, add_zero_attn, kdim, vdim, batch_first, device, dtype)
+
+        # rewrite out_proj layer，with nn.Linear
+        self.out_proj = nn.Linear(embed_dim, embed_dim, bias=bias, device=device, dtype=dtype)
+
+    def forward(
+                self,
+                query: Tensor,
+                key: Tensor,
+                value: Tensor,
+                key_padding_mask: Optional[Tensor] = None,
+                need_weights: bool = True,
+                attn_mask: Optional[Tensor] = None,
+                average_attn_weights: bool = True,
+                is_causal : bool = False) -> Tuple[Tensor, Optional[Tensor]]:
+        why_not_fast_path = ''
+        if ((attn_mask is not None and torch.is_floating_point(attn_mask))
+           or (key_padding_mask is not None) and torch.is_floating_point(key_padding_mask)):
+            why_not_fast_path = "floating-point masks are not supported for fast path."
+
+        is_batched = query.dim() == 3
+
+        key_padding_mask = F._canonical_mask(
+            mask=key_padding_mask,
+            mask_name="key_padding_mask",
+            other_type=F._none_or_dtype(attn_mask),
+            other_name="attn_mask",
+            target_type=query.dtype
+        )
+
+        attn_mask = F._canonical_mask(
+            mask=attn_mask,
+            mask_name="attn_mask",
+            other_type=None,
+            other_name="",
+            target_type=query.dtype,
+            check_other=False,
+        )
+
+
+        if not is_batched:
+            why_not_fast_path = f"input not batched; expected query.dim() of 3 but got {query.dim()}"
+        elif query is not key or key is not value:
+            # When lifting this restriction, don't forget to either
+            # enforce that the dtypes all match or test cases where
+            # they don't!
+            why_not_fast_path = "non-self attention was used (query, key, and value are not the same Tensor)"
+        elif self.in_proj_bias is not None and query.dtype != self.in_proj_bias.dtype:
+            why_not_fast_path = f"dtypes of query ({query.dtype}) and self.in_proj_bias ({self.in_proj_bias.dtype}) don't match"
+        elif self.in_proj_weight is None:
+            why_not_fast_path = "in_proj_weight was None"
+        elif query.dtype != self.in_proj_weight.dtype:
+            # this case will fail anyway, but at least they'll get a useful error message.
+            why_not_fast_path = f"dtypes of query ({query.dtype}) and self.in_proj_weight ({self.in_proj_weight.dtype}) don't match"
+        elif self.training:
+            why_not_fast_path = "training is enabled"
+        elif (self.num_heads % 2) != 0:
+            why_not_fast_path = "self.num_heads is not even"
+        elif not self.batch_first:
+            why_not_fast_path = "batch_first was not True"
+        elif self.bias_k is not None:
+            why_not_fast_path = "self.bias_k was not None"
+        elif self.bias_v is not None:
+            why_not_fast_path = "self.bias_v was not None"
+        elif self.add_zero_attn:
+            why_not_fast_path = "add_zero_attn was enabled"
+        elif not self._qkv_same_embed_dim:
+            why_not_fast_path = "_qkv_same_embed_dim was not True"
+        elif query.is_nested and (key_padding_mask is not None or attn_mask is not None):
+            why_not_fast_path = "supplying both src_key_padding_mask and src_mask at the same time \
+                                 is not supported with NestedTensor input"
+        elif torch.is_autocast_enabled():
+            why_not_fast_path = "autocast is enabled"
+
+        if not why_not_fast_path:
+            tensor_args = (
+                query,
+                key,
+                value,
+                self.in_proj_weight,
+                self.in_proj_bias,
+                self.out_proj.weight,
+                self.out_proj.bias,
+            )
+            # We have to use list comprehensions below because TorchScript does not support
+            # generator expressions.
+            if torch.overrides.has_torch_function(tensor_args):
+                why_not_fast_path = "some Tensor argument has_torch_function"
+            elif _is_make_fx_tracing():
+                why_not_fast_path = "we are running make_fx tracing"
+            elif not all(_check_arg_device(x) for x in tensor_args):
+                why_not_fast_path = ("some Tensor argument's device is neither one of "
+                                     f"cpu, cuda or {torch.utils.backend_registration._privateuse1_backend_name}")
+            elif torch.is_grad_enabled() and any(_arg_requires_grad(x) for x in tensor_args):
+                why_not_fast_path = ("grad is enabled and at least one of query or the "
+                                     "input/output projection weights or biases requires_grad")
+            if not why_not_fast_path:
+                merged_mask, mask_type = self.merge_masks(attn_mask, key_padding_mask, query)
+
+                if self.in_proj_bias is not None and self.in_proj_weight is not None:
+                    return torch._native_multi_head_attention(
+                        query,
+                        key,
+                        value,
+                        self.embed_dim,
+                        self.num_heads,
+                        self.in_proj_weight,
+                        self.in_proj_bias,
+                        self.out_proj.weight,
+                        self.out_proj.bias,
+                        merged_mask,
+                        need_weights,
+                        average_attn_weights,
+                        mask_type)
+
+        any_nested = query.is_nested or key.is_nested or value.is_nested
+        assert not any_nested, ("MultiheadAttention does not support NestedTensor outside of its fast path. " +
+                                f"The fast path was not hit because {why_not_fast_path}")
+
+        if self.batch_first and is_batched:
+            # make sure that the transpose op does not affect the "is" property
+            if key is value:
+                if query is key:
+                    query = key = value = query.transpose(1, 0)
+                else:
+                    query, key = (x.transpose(1, 0) for x in (query, key))
+                    value = key
+            else:
+                query, key, value = (x.transpose(1, 0) for x in (query, key, value))
+        
+        if not self._qkv_same_embed_dim:
+            attn_output, attn_output_weights = self.multi_head_attention_forward(
+                query, key, value, self.embed_dim, self.num_heads,
+                self.in_proj_weight, self.in_proj_bias,
+                self.bias_k, self.bias_v, self.add_zero_attn,
+                self.dropout, self.out_proj.weight, self.out_proj.bias,
+                training=self.training,
+                key_padding_mask=key_padding_mask, need_weights=need_weights,
+                attn_mask=attn_mask,
+                use_separate_proj_weight=True,
+                q_proj_weight=self.q_proj_weight, k_proj_weight=self.k_proj_weight,
+                v_proj_weight=self.v_proj_weight,
+                average_attn_weights=average_attn_weights,
+                is_causal=is_causal)
+        else:
+            attn_output, attn_output_weights = self.multi_head_attention_forward(
+                query, key, value, self.embed_dim, self.num_heads,
+                self.in_proj_weight, self.in_proj_bias,
+                self.bias_k, self.bias_v, self.add_zero_attn,
+                self.dropout, self.out_proj.weight, self.out_proj.bias,
+                training=self.training,
+                key_padding_mask=key_padding_mask,
+                need_weights=need_weights,
+                attn_mask=attn_mask,
+                average_attn_weights=average_attn_weights,
+                is_causal=is_causal)
+        if self.batch_first and is_batched:
+            return attn_output.transpose(1, 0), attn_output_weights
+        else:
+            return attn_output, attn_output_weights
+            
+    def multi_head_attention_forward(
+        self,
+        query: Tensor,
+        key: Tensor,
+        value: Tensor,
+        embed_dim_to_check: int,
+        num_heads: int,
+        in_proj_weight: Optional[Tensor],
+        in_proj_bias: Optional[Tensor],
+        bias_k: Optional[Tensor],
+        bias_v: Optional[Tensor],
+        add_zero_attn: bool,
+        dropout_p: float,
+        out_proj_weight: Tensor,
+        out_proj_bias: Optional[Tensor],
+        training: bool = True,
+        key_padding_mask: Optional[Tensor] = None,
+        need_weights: bool = True,
+        attn_mask: Optional[Tensor] = None,
+        use_separate_proj_weight: bool = False,
+        q_proj_weight: Optional[Tensor] = None,
+        k_proj_weight: Optional[Tensor] = None,
+        v_proj_weight: Optional[Tensor] = None,
+        static_k: Optional[Tensor] = None,
+        static_v: Optional[Tensor] = None,
+        average_attn_weights: bool = True,
+        is_causal: bool = False,
+    ) -> Tuple[Tensor, Optional[Tensor]]:
+        tens_ops = (query, key, value, in_proj_weight, in_proj_bias, bias_k, bias_v, out_proj_weight, out_proj_bias)
+        if has_torch_function(tens_ops):
+            return handle_torch_function(
+                multi_head_attention_forward,
+                tens_ops,
+                query,
+                key,
+                value,
+                embed_dim_to_check,
+                num_heads,
+                in_proj_weight,
+                in_proj_bias,
+                bias_k,
+                bias_v,
+                add_zero_attn,
+                dropout_p,
+                out_proj_weight,
+                out_proj_bias,
+                training=training,
+                key_padding_mask=key_padding_mask,
+                need_weights=need_weights,
+                attn_mask=attn_mask,
+                is_causal=is_causal,
+                use_separate_proj_weight=use_separate_proj_weight,
+                q_proj_weight=q_proj_weight,
+                k_proj_weight=k_proj_weight,
+                v_proj_weight=v_proj_weight,
+                static_k=static_k,
+                static_v=static_v,
+                average_attn_weights=average_attn_weights,
+            )
+    
+        is_batched = _mha_shape_check(query, key, value, key_padding_mask, attn_mask, num_heads)
+    
+        # For unbatched input, we unsqueeze at the expected batch-dim to pretend that the input
+        # is batched, run the computation and before returning squeeze the
+        # batch dimension so that the output doesn't carry this temporary batch dimension.
+        if not is_batched:
+            # unsqueeze if the input is unbatched
+            query = query.unsqueeze(1)
+            key = key.unsqueeze(1)
+            value = value.unsqueeze(1)
+            if key_padding_mask is not None:
+                key_padding_mask = key_padding_mask.unsqueeze(0)
+    
+        # set up shape vars
+        tgt_len, bsz, embed_dim = query.shape
+        src_len, _, _ = key.shape
+    
+        key_padding_mask = _canonical_mask(
+            mask=key_padding_mask,
+            mask_name="key_padding_mask",
+            other_type=_none_or_dtype(attn_mask),
+            other_name="attn_mask",
+            target_type=query.dtype
+        )
+    
+        if is_causal and attn_mask is None:
+            raise RuntimeError(
+                "Need attn_mask if specifying the is_causal hint. "
+                "You may use the Transformer module method "
+                "`generate_square_subsequent_mask` to create this mask."
+            )
+    
+        if is_causal and key_padding_mask is None and not need_weights:
+            # when we have a kpm or need weights, we need attn_mask
+            # Otherwise, we use the is_causal hint go as is_causal
+            # indicator to SDPA.
+            attn_mask = None
+        else:
+            attn_mask = _canonical_mask(
+                mask=attn_mask,
+                mask_name="attn_mask",
+                other_type=None,
+                other_name="",
+                target_type=query.dtype,
+                check_other=False,
+            )
+    
+            if key_padding_mask is not None:
+                # We have the attn_mask, and use that to merge kpm into it.
+                # Turn off use of is_causal hint, as the merged mask is no
+                # longer causal.
+                is_causal = False
+    
+        assert embed_dim == embed_dim_to_check, \
+            f"was expecting embedding dimension of {embed_dim_to_check}, but got {embed_dim}"
+        if isinstance(embed_dim, torch.Tensor):
+            # embed_dim can be a tensor when JIT tracing
+            head_dim = embed_dim.div(num_heads, rounding_mode='trunc')
+        else:
+            head_dim = embed_dim // num_heads
+        assert head_dim * num_heads == embed_dim, f"embed_dim {embed_dim} not divisible by num_heads {num_heads}"
+        if use_separate_proj_weight:
+            # allow MHA to have different embedding dimensions when separate projection weights are used
+            assert key.shape[:2] == value.shape[:2], \
+                f"key's sequence and batch dims {key.shape[:2]} do not match value's {value.shape[:2]}"
+        else:
+            assert key.shape == value.shape, f"key shape {key.shape} does not match value shape {value.shape}"
+    
+        #
+        # compute in-projection
+        #
+        if not use_separate_proj_weight:
+            assert in_proj_weight is not None, "use_separate_proj_weight is False but in_proj_weight is None"
+            q, k, v = _in_projection_packed(query, key, value, in_proj_weight, in_proj_bias)
+        else:
+            assert q_proj_weight is not None, "use_separate_proj_weight is True but q_proj_weight is None"
+            assert k_proj_weight is not None, "use_separate_proj_weight is True but k_proj_weight is None"
+            assert v_proj_weight is not None, "use_separate_proj_weight is True but v_proj_weight is None"
+            if in_proj_bias is None:
+                b_q = b_k = b_v = None
+            else:
+                b_q, b_k, b_v = in_proj_bias.chunk(3)
+            q, k, v = _in_projection(query, key, value, q_proj_weight, k_proj_weight, v_proj_weight, b_q, b_k, b_v)
+    
+        # prep attention mask
+    
+        if attn_mask is not None:
+            # ensure attn_mask's dim is 3
+            if attn_mask.dim() == 2:
+                correct_2d_size = (tgt_len, src_len)
+                if attn_mask.shape != correct_2d_size:
+                    raise RuntimeError(f"The shape of the 2D attn_mask is {attn_mask.shape}, but should be {correct_2d_size}.")
+                attn_mask = attn_mask.unsqueeze(0)
+            elif attn_mask.dim() == 3:
+                correct_3d_size = (bsz * num_heads, tgt_len, src_len)
+                if attn_mask.shape != correct_3d_size:
+                    raise RuntimeError(f"The shape of the 3D attn_mask is {attn_mask.shape}, but should be {correct_3d_size}.")
+            else:
+                raise RuntimeError(f"attn_mask's dimension {attn_mask.dim()} is not supported")
+    
+        # add bias along batch dimension (currently second)
+        if bias_k is not None and bias_v is not None:
+            assert static_k is None, "bias cannot be added to static key."
+            assert static_v is None, "bias cannot be added to static value."
+            k = torch.cat([k, bias_k.repeat(1, bsz, 1)])
+            v = torch.cat([v, bias_v.repeat(1, bsz, 1)])
+            if attn_mask is not None:
+                attn_mask = pad(attn_mask, (0, 1))
+            if key_padding_mask is not None:
+                key_padding_mask = pad(key_padding_mask, (0, 1))
+        else:
+            assert bias_k is None
+            assert bias_v is None
+    
+        #
+        # reshape q, k, v for multihead attention and make em batch first
+        #
+        q = q.view(tgt_len, bsz * num_heads, head_dim).transpose(0, 1)
+        if static_k is None:
+            k = k.view(k.shape[0], bsz * num_heads, head_dim).transpose(0, 1)
+        else:
+            # TODO finish disentangling control flow so we don't do in-projections when statics are passed
+            assert static_k.size(0) == bsz * num_heads, \
+                f"expecting static_k.size(0) of {bsz * num_heads}, but got {static_k.size(0)}"
+            assert static_k.size(2) == head_dim, \
+                f"expecting static_k.size(2) of {head_dim}, but got {static_k.size(2)}"
+            k = static_k
+        if static_v is None:
+            v = v.view(v.shape[0], bsz * num_heads, head_dim).transpose(0, 1)
+        else:
+            # TODO finish disentangling control flow so we don't do in-projections when statics are passed
+            assert static_v.size(0) == bsz * num_heads, \
+                f"expecting static_v.size(0) of {bsz * num_heads}, but got {static_v.size(0)}"
+            assert static_v.size(2) == head_dim, \
+                f"expecting static_v.size(2) of {head_dim}, but got {static_v.size(2)}"
+            v = static_v
+    
+        # add zero attention along batch dimension (now first)
+        if add_zero_attn:
+            zero_attn_shape = (bsz * num_heads, 1, head_dim)
+            k = torch.cat([k, torch.zeros(zero_attn_shape, dtype=k.dtype, device=k.device)], dim=1)
+            v = torch.cat([v, torch.zeros(zero_attn_shape, dtype=v.dtype, device=v.device)], dim=1)
+            if attn_mask is not None:
+                attn_mask = pad(attn_mask, (0, 1))
+            if key_padding_mask is not None:
+                key_padding_mask = pad(key_padding_mask, (0, 1))
+    
+        # update source sequence length after adjustments
+        src_len = k.size(1)
+    
+        # merge key padding and attention masks
+        if key_padding_mask is not None:
+            assert key_padding_mask.shape == (bsz, src_len), \
+                f"expecting key_padding_mask shape of {(bsz, src_len)}, but got {key_padding_mask.shape}"
+            key_padding_mask = key_padding_mask.view(bsz, 1, 1, src_len).   \
+                expand(-1, num_heads, -1, -1).reshape(bsz * num_heads, 1, src_len)
+            if attn_mask is None:
+                attn_mask = key_padding_mask
+            else:
+                attn_mask = attn_mask + key_padding_mask
+    
+        # adjust dropout probability
+        if not training:
+            dropout_p = 0.0
+    
+        #
+        # (deep breath) calculate attention and out projection
+        #
+    
+        if need_weights:
+            B, Nt, E = q.shape
+            q_scaled = q / math.sqrt(E)
+    
+            assert not (is_causal and attn_mask is None), "FIXME: is_causal not implemented for need_weights"
+    
+            if attn_mask is not None:
+                attn_output_weights = torch.baddbmm(attn_mask, q_scaled, k.transpose(-2, -1))
+            else:
+                attn_output_weights = torch.bmm(q_scaled, k.transpose(-2, -1))
+            attn_output_weights = softmax(attn_output_weights, dim=-1)
+            if dropout_p > 0.0:
+                attn_output_weights = dropout(attn_output_weights, p=dropout_p)
+    
+            attn_output = torch.bmm(attn_output_weights, v)
+    
+            attn_output = attn_output.transpose(0, 1).contiguous().view(tgt_len * bsz, embed_dim)
+            attn_output = self.out_proj(attn_output)
+            attn_output = attn_output.view(tgt_len, bsz, attn_output.size(1))
+    
+            # optionally average attention weights over heads
+            attn_output_weights = attn_output_weights.view(bsz, num_heads, tgt_len, src_len)
+            if average_attn_weights:
+                attn_output_weights = attn_output_weights.mean(dim=1)
+    
+            if not is_batched:
+                # squeeze the output if input was unbatched
+                attn_output = attn_output.squeeze(1)
+                attn_output_weights = attn_output_weights.squeeze(0)
+            return attn_output, attn_output_weights
+        else:
+            # attn_mask can be either (L,S) or (N*num_heads, L, S)
+            # if attn_mask's shape is (1, L, S) we need to unsqueeze to (1, 1, L, S)
+            # in order to match the input for SDPA of (N, num_heads, L, S)
+            if attn_mask is not None:
+                if attn_mask.size(0) == 1 and attn_mask.dim() == 3:
+                    attn_mask = attn_mask.unsqueeze(0)
+                else:
+                    attn_mask = attn_mask.view(bsz, num_heads, -1, src_len)
+    
+            q = q.view(bsz, num_heads, tgt_len, head_dim)
+            k = k.view(bsz, num_heads, src_len, head_dim)
+            v = v.view(bsz, num_heads, src_len, head_dim)
+    
+            attn_output = F.scaled_dot_product_attention(q, k, v, attn_mask, dropout_p, is_causal)
+            attn_output = attn_output.permute(2, 0, 1, 3).contiguous().view(bsz * tgt_len, embed_dim)
+    
+            attn_output = self.out_proj(attn_output)
+            attn_output = attn_output.view(tgt_len, bsz, attn_output.size(1))
+            if not is_batched:
+                # squeeze the output if input was unbatched
+                attn_output = attn_output.squeeze(1)
+            return attn_output, None
+
+
+def _mha_shape_check(query: Tensor, key: Tensor, value: Tensor,
+                     key_padding_mask: Optional[Tensor], attn_mask: Optional[Tensor], num_heads: int):
+    # Verifies the expected shape for `query, `key`, `value`, `key_padding_mask` and `attn_mask`
+    # and returns if the input is batched or not.
+    # Raises an error if `query` is not 2-D (unbatched) or 3-D (batched) tensor.
+
+    # Shape check.
+    if query.dim() == 3:
+        # Batched Inputs
+        is_batched = True
+        assert key.dim() == 3 and value.dim() == 3, \
+            ("For batched (3-D) `query`, expected `key` and `value` to be 3-D"
+             f" but found {key.dim()}-D and {value.dim()}-D tensors respectively")
+        if key_padding_mask is not None:
+            assert key_padding_mask.dim() == 2, \
+                ("For batched (3-D) `query`, expected `key_padding_mask` to be `None` or 2-D"
+                 f" but found {key_padding_mask.dim()}-D tensor instead")
+        if attn_mask is not None:
+            assert attn_mask.dim() in (2, 3), \
+                ("For batched (3-D) `query`, expected `attn_mask` to be `None`, 2-D or 3-D"
+                 f" but found {attn_mask.dim()}-D tensor instead")
+    elif query.dim() == 2:
+        # Unbatched Inputs
+        is_batched = False
+        assert key.dim() == 2 and value.dim() == 2, \
+            ("For unbatched (2-D) `query`, expected `key` and `value` to be 2-D"
+             f" but found {key.dim()}-D and {value.dim()}-D tensors respectively")
+
+        if key_padding_mask is not None:
+            assert key_padding_mask.dim() == 1, \
+                ("For unbatched (2-D) `query`, expected `key_padding_mask` to be `None` or 1-D"
+                 f" but found {key_padding_mask.dim()}-D tensor instead")
+
+        if attn_mask is not None:
+            assert attn_mask.dim() in (2, 3), \
+                ("For unbatched (2-D) `query`, expected `attn_mask` to be `None`, 2-D or 3-D"
+                 f" but found {attn_mask.dim()}-D tensor instead")
+            if attn_mask.dim() == 3:
+                expected_shape = (num_heads, query.shape[0], key.shape[0])
+                assert attn_mask.shape == expected_shape, \
+                    (f"Expected `attn_mask` shape to be {expected_shape} but got {attn_mask.shape}")
+    else:
+        raise AssertionError(
+            f"query should be unbatched 2D or batched 3D tensor but received {query.dim()}-D query tensor")
+
+    return is_batched
+
+
+def _canonical_mask(
+        mask: Optional[Tensor],
+        mask_name: str,
+        other_type: Optional[DType],
+        other_name: str,
+        target_type: DType,
+        check_other: bool = True,
+) -> Optional[Tensor]:
+
+    if mask is not None:
+        _mask_dtype = mask.dtype
+        _mask_is_float = torch.is_floating_point(mask)
+        if _mask_dtype != torch.bool and not _mask_is_float:
+            raise AssertionError(
+                f"only bool and floating types of {mask_name} are supported")
+        if check_other and other_type is not None:
+            if _mask_dtype != other_type:
+                warnings.warn(
+                    f"Support for mismatched {mask_name} and {other_name} "
+                    "is deprecated. Use same type for both instead."
+                )
+        if not _mask_is_float:
+            mask = (
+                torch.zeros_like(mask, dtype=target_type)
+                .masked_fill_(mask, float("-inf"))
+            )
+    return mask
+
+
+def _none_or_dtype(input: Optional[Tensor]) -> Optional[DType]:
+    if input is None:
+        return None
+    elif isinstance(input, torch.Tensor):
+        return input.dtype
+    raise RuntimeError("input to _none_or_dtype() must be None or torch.Tensor")
+
+def _in_projection_packed(
+    q: Tensor,
+    k: Tensor,
+    v: Tensor,
+    w: Tensor,
+    b: Optional[Tensor] = None,
+) -> List[Tensor]:
+    r"""
+    Performs the in-projection step of the attention operation, using packed weights.
+    Output is a triple containing projection tensors for query, key and value.
+    Args:
+        q, k, v: query, key and value tensors to be projected. For self-attention,
+            these are typically the same tensor; for encoder-decoder attention,
+            k and v are typically the same tensor. (We take advantage of these
+            identities for performance if they are present.) Regardless, q, k and v
+            must share a common embedding dimension; otherwise their shapes may vary.
+        w: projection weights for q, k and v, packed into a single tensor. Weights
+            are packed along dimension 0, in q, k, v order.
+        b: optional projection biases for q, k and v, packed into a single tensor
+            in q, k, v order.
+    Shape:
+        Inputs:
+        - q: :math:`(..., E)` where E is the embedding dimension
+        - k: :math:`(..., E)` where E is the embedding dimension
+        - v: :math:`(..., E)` where E is the embedding dimension
+        - w: :math:`(E * 3, E)` where E is the embedding dimension
+        - b: :math:`E * 3` where E is the embedding dimension
+        Output:
+        - in output list :math:`[q', k', v']`, each output tensor will have the
+            same shape as the corresponding input tensor.
+    """
+    E = q.size(-1)
+    if k is v:
+        if q is k:
+            # self-attention
+            proj = linear(q, w, b)
+            # reshape to 3, E and not E, 3 is deliberate for better memory coalescing and keeping same order as chunk()
+            proj = proj.unflatten(-1, (3, E)).unsqueeze(0).transpose(0, -2).squeeze(-2).contiguous()
+            return proj[0], proj[1], proj[2]
+        else:
+            # encoder-decoder attention
+            w_q, w_kv = w.split([E, E * 2])
+            if b is None:
+                b_q = b_kv = None
+            else:
+                b_q, b_kv = b.split([E, E * 2])
+            q_proj = linear(q, w_q, b_q)
+            kv_proj = linear(k, w_kv, b_kv)
+            # reshape to 2, E and not E, 2 is deliberate for better memory coalescing and keeping same order as chunk()
+            kv_proj = kv_proj.unflatten(-1, (2, E)).unsqueeze(0).transpose(0, -2).squeeze(-2).contiguous()
+            return (q_proj, kv_proj[0], kv_proj[1])
+    else:
+        w_q, w_k, w_v = w.chunk(3)
+        if b is None:
+            b_q = b_k = b_v = None
+        else:
+            b_q, b_k, b_v = b.chunk(3)
+        return linear(q, w_q, b_q), linear(k, w_k, b_k), linear(v, w_v, b_v)
+
+
+def _in_projection(
+    q: Tensor,
+    k: Tensor,
+    v: Tensor,
+    w_q: Tensor,
+    w_k: Tensor,
+    w_v: Tensor,
+    b_q: Optional[Tensor] = None,
+    b_k: Optional[Tensor] = None,
+    b_v: Optional[Tensor] = None,
+) -> Tuple[Tensor, Tensor, Tensor]:
+    r"""
+    Performs the in-projection step of the attention operation. This is simply
+    a triple of linear projections, with shape constraints on the weights which
+    ensure embedding dimension uniformity in the projected outputs.
+    Output is a triple containing projection tensors for query, key and value.
+    Args:
+        q, k, v: query, key and value tensors to be projected.
+        w_q, w_k, w_v: weights for q, k and v, respectively.
+        b_q, b_k, b_v: optional biases for q, k and v, respectively.
+    Shape:
+        Inputs:
+        - q: :math:`(Qdims..., Eq)` where Eq is the query embedding dimension and Qdims are any
+            number of leading dimensions.
+        - k: :math:`(Kdims..., Ek)` where Ek is the key embedding dimension and Kdims are any
+            number of leading dimensions.
+        - v: :math:`(Vdims..., Ev)` where Ev is the value embedding dimension and Vdims are any
+            number of leading dimensions.
+        - w_q: :math:`(Eq, Eq)`
+        - w_k: :math:`(Eq, Ek)`
+        - w_v: :math:`(Eq, Ev)`
+        - b_q: :math:`(Eq)`
+        - b_k: :math:`(Eq)`
+        - b_v: :math:`(Eq)`
+        Output: in output triple :math:`(q', k', v')`,
+         - q': :math:`[Qdims..., Eq]`
+         - k': :math:`[Kdims..., Eq]`
+         - v': :math:`[Vdims..., Eq]`
+    """
+    Eq, Ek, Ev = q.size(-1), k.size(-1), v.size(-1)
+    assert w_q.shape == (Eq, Eq), f"expecting query weights shape of {(Eq, Eq)}, but got {w_q.shape}"
+    assert w_k.shape == (Eq, Ek), f"expecting key weights shape of {(Eq, Ek)}, but got {w_k.shape}"
+    assert w_v.shape == (Eq, Ev), f"expecting value weights shape of {(Eq, Ev)}, but got {w_v.shape}"
+    assert b_q is None or b_q.shape == (Eq,), f"expecting query bias shape of {(Eq,)}, but got {b_q.shape}"
+    assert b_k is None or b_k.shape == (Eq,), f"expecting key bias shape of {(Eq,)}, but got {b_k.shape}"
+    assert b_v is None or b_v.shape == (Eq,), f"expecting value bias shape of {(Eq,)}, but got {b_v.shape}"
+    return linear(q, w_q, b_q), linear(k, w_k, b_k), linear(v, w_v, b_v)

rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4b60c6e67e78f9e67a8aba1a0e479de7999ecc20bf17df7d424f8358e079819
+size 15920

rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6946050881f78246d61e7a2d5ee922e733c8d80738866c6a8fb5bee23223418
+size 15920

rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6ed1d9f52641d1754d5c68ca71beb3cbe40a66ee1b549941ee0ade7a1e161ec
+size 15920

rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1792653f7a4196db595bcbeeac1bda44ee9de0b7ae480ac1b6d12db4c8a09d44
+size 15920

rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94c340c8e9cbfd35b4e80e5cd674b2b2ad00a02f185011593784c2929de77d01
+size 15920

rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb0f8cad0d4eda7f130b5e18b346a97dd862e1a5d57b68c3cac83c9cb76853b6
+size 15920

rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8df90dc44d85d90cdca0705c0ea6fc0ae82caacb319014d4aad49a71f23684ee
+size 15920

rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8143910fe0f03885eaf75ca28ca0a562fbf8c31cf1e9b881667b4381a1e3947
+size 15920

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c955503654e0db190bc167dfb2c816a4abfb4cd787ced5f65952f61734cb4ce9
+size 1064

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|end_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "!",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,2080 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "!",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<image>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "</image>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<ref>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "</ref>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<box>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "</box>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<quad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "</quad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<point>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "</point>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<slice>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "</slice>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "auto_map": {
+    "AutoTokenizer": [
+      "modeling_minicpmv.PreTrainedTokenizerFastWrapper",
+      null
+    ]
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|end_of_text|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "!",
+  "padding_side": "right",
+  "tokenizer_class": "PreTrainedTokenizerFastWrapper",
+  "truncation_side": "right",
+  "unk_token": "<unk>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a78ecf29549624f8a589728340238ba127296f2f6dc18c6e5d4a7595ae5f690b
+size 7480

zero_to_fp32.py

@@ -0,0 +1,587 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example: python zero_to_fp32.py . pytorch_model.bin
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel()
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in param_shapes.items():
+
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+    print(f"Saving fp32 state dict to {output_file}")
+    torch.save(state_dict, output_file)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument(
+        "output_file",
+        type=str,
+        help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, args.output_file, tag=args.tag)