Upload model

README.md

@@ -0,0 +1,155 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: mistralai/Mistral-7B-vp0.3
+language: en
+datasets:
+- Word2Li/MiddOptimized
+tags:
+- llama-factory
+- full
+pipeline_tag: text-generation
+model-index:
+- name: Mistral-7B-v0.3-Middo-Alpaca
+  results:
+    - task:
+        type: text-generation
+      dataset:
+        name: MMLU
+        type: MMLU
+      metrics:
+        - name: Weighted Avg.
+          type: Weighted Avg.
+          value: 38.73
+          verified: true
+    - task:
+        type: text-generation
+      dataset:
+        name: IFEval
+        type: IFEval
+      metrics:
+        - name: Avg.
+          type: Avg.
+          value: 44.01
+          verified: true
+    - task:
+        type: text-generation
+      dataset:
+        name: GSM8K
+        type: GSM8K
+      metrics:
+        - name: pass@1
+          type: pass@1
+          value: 34.80
+          verified: true
+    - task:
+        type: text-generation
+      dataset:
+        name: MATH
+        type: MATH
+      metrics:
+        - name: pass@1
+          type: pass@1
+          value: 6.64
+          verified: true
+    - task:
+        type: text-generation
+      dataset:
+        name: HumanEval
+        type: HumanEval
+      metrics:
+        - name: pass@1
+          type: pass@1
+          value: 26.22
+          verified: true
+    - task:
+        type: text-generation
+      dataset:
+        name: MBPP
+        type: MBPP
+      metrics:
+        - name: pass@1
+          type: pass@1
+          value: 31.40
+          verified: true
+    - task:
+        type: text-generation
+      dataset:
+        name: Hellaswag
+        type: Hellaswag
+      metrics:
+        - name: pass@1
+          type: pass@1
+          value: 44.86
+          verified: true
+    - task:
+        type: text-generation
+      dataset:
+        name: GPQA
+        type: GPQA
+      metrics:
+        - name: pass@1
+          type: pass@1
+          value: 11.11
+          verified: true
+metrics:
+- accuracy
+---
+
+# Mistral-7B-v0.3-Middo-Alpaca
+
+Paper: [Middo: Model-Informed Dynamic Data Optimization for Enhanced LLM Fine-Tuning via Closed-Loop Learning](https://arxiv.org/abs/2508.21589)
+
+Code: https://github.com/Word2VecT/Middo
+
+## Model description
+
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the [MiddOptimzed/llama_alpaca](https://huggingface.co/datasets/Word2Li/MiddOptimized/viewer/default/mistral_alpaca) dataset.
+
+## Training and evaluation data
+
+### Training data
+
+Middo optimized [Alpaca](https://huggingface.co/datasets/tatsu-lab/alpaca) on [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3).
+
+### Evaluation data
+
+- General
+  - MMLU
+  - IFEval
+- Math
+  - GSM8K
+  - MATH
+- Code
+  - HumanEval
+  - MBPP
+- Reasoning
+  - Hellaswag
+  - GPQA
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+
+- learning_rate: 1e-05
+- train_batch_size: 4
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 256
+- total_eval_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.03
+- num_epochs: 1.0
+
+### Framework versions
+
+- Transformers 4.45.2
+- Pytorch 2.5.1+cu121
+- Datasets 2.21.0
+- Tokenizers 0.20.1

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "total_flos": 1.387246770039292e+18,
+    "train_loss": 1.0000714727661066,
+    "train_runtime": 2851.6353,
+    "train_samples_per_second": 20.461,
+    "train_steps_per_second": 0.08
+}

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "/mnt/petrelfs/tangzinan/LLaMA-Factory/models/Mistral-7B-v0.3",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "pad_token_id": 770,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.2",
+  "unsloth_version": "2024.9",
+  "use_cache": false,
+  "vocab_size": 32768
+}

generation_config.json

@@ -0,0 +1,8 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "max_length": 32768,
+  "pad_token_id": 770,
+  "transformers_version": "4.45.2"
+}

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9481b6ea50d5ca8ec517345454497a746e99176d875febc07cf6b9985c35974
+size 4949453792

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a79e6255f4d51b625ae5e75a6c8fdb63212eb49f809402be247a219e9b2ac35
+size 4999819336

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca3c7a2fdad22f2a0d72ddb944186c03ee1b795f8cb008dbd0456da44c49da36
+size 4546807800

model.safetensors.index.json

@@ -0,0 +1,298 @@
+{
+  "metadata": {
+    "total_size": 14496047104
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00003-of-00003.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.norm.weight": "model-00003-of-00003.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[control_768]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37f00374dea48658ee8f5d0f21895b9bc55cb0103939607c8185bfd1c6ca1f89
+size 587404

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "total_flos": 1.387246770039292e+18,
+    "train_loss": 1.0000714727661066,
+    "train_runtime": 2851.6353,
+    "train_samples_per_second": 20.461,
+    "train_steps_per_second": 0.08
+}

trainer_log.jsonl

@@ -0,0 +1,229 @@
+{"current_steps": 1, "total_steps": 228, "loss": 1.2955, "lr": 1.4285714285714286e-06, "epoch": 0.0043859649122807015, "percentage": 0.44, "elapsed_time": "0:00:14", "remaining_time": "0:55:23"}
+{"current_steps": 2, "total_steps": 228, "loss": 1.3459, "lr": 2.8571428571428573e-06, "epoch": 0.008771929824561403, "percentage": 0.88, "elapsed_time": "0:00:27", "remaining_time": "0:50:57"}
+{"current_steps": 3, "total_steps": 228, "loss": 1.2305, "lr": 4.2857142857142855e-06, "epoch": 0.013157894736842105, "percentage": 1.32, "elapsed_time": "0:00:39", "remaining_time": "0:49:00"}
+{"current_steps": 4, "total_steps": 228, "loss": 1.0947, "lr": 5.7142857142857145e-06, "epoch": 0.017543859649122806, "percentage": 1.75, "elapsed_time": "0:00:52", "remaining_time": "0:49:07"}
+{"current_steps": 5, "total_steps": 228, "loss": 1.0768, "lr": 7.1428571428571436e-06, "epoch": 0.021929824561403508, "percentage": 2.19, "elapsed_time": "0:01:04", "remaining_time": "0:48:14"}
+{"current_steps": 6, "total_steps": 228, "loss": 1.0448, "lr": 8.571428571428571e-06, "epoch": 0.02631578947368421, "percentage": 2.63, "elapsed_time": "0:01:17", "remaining_time": "0:47:47"}
+{"current_steps": 7, "total_steps": 228, "loss": 1.0738, "lr": 1e-05, "epoch": 0.03070175438596491, "percentage": 3.07, "elapsed_time": "0:01:30", "remaining_time": "0:47:44"}
+{"current_steps": 8, "total_steps": 228, "loss": 1.0353, "lr": 9.999494817970498e-06, "epoch": 0.03508771929824561, "percentage": 3.51, "elapsed_time": "0:01:42", "remaining_time": "0:47:04"}
+{"current_steps": 9, "total_steps": 228, "loss": 0.9942, "lr": 9.997979373965542e-06, "epoch": 0.039473684210526314, "percentage": 3.95, "elapsed_time": "0:01:53", "remaining_time": "0:46:13"}
+{"current_steps": 10, "total_steps": 228, "loss": 1.0467, "lr": 9.995453974215164e-06, "epoch": 0.043859649122807015, "percentage": 4.39, "elapsed_time": "0:02:06", "remaining_time": "0:46:06"}
+{"current_steps": 11, "total_steps": 228, "loss": 0.9998, "lr": 9.991919129033994e-06, "epoch": 0.04824561403508772, "percentage": 4.82, "elapsed_time": "0:02:19", "remaining_time": "0:45:44"}
+{"current_steps": 12, "total_steps": 228, "loss": 0.9997, "lr": 9.987375552718133e-06, "epoch": 0.05263157894736842, "percentage": 5.26, "elapsed_time": "0:02:31", "remaining_time": "0:45:31"}
+{"current_steps": 13, "total_steps": 228, "loss": 1.0602, "lr": 9.981824163400827e-06, "epoch": 0.05701754385964912, "percentage": 5.7, "elapsed_time": "0:02:44", "remaining_time": "0:45:13"}
+{"current_steps": 14, "total_steps": 228, "loss": 1.043, "lr": 9.975266082866923e-06, "epoch": 0.06140350877192982, "percentage": 6.14, "elapsed_time": "0:02:56", "remaining_time": "0:44:55"}
+{"current_steps": 15, "total_steps": 228, "loss": 1.0331, "lr": 9.967702636326195e-06, "epoch": 0.06578947368421052, "percentage": 6.58, "elapsed_time": "0:03:08", "remaining_time": "0:44:37"}
+{"current_steps": 16, "total_steps": 228, "loss": 1.0376, "lr": 9.959135352145552e-06, "epoch": 0.07017543859649122, "percentage": 7.02, "elapsed_time": "0:03:19", "remaining_time": "0:44:06"}
+{"current_steps": 17, "total_steps": 228, "loss": 1.0244, "lr": 9.9495659615402e-06, "epoch": 0.07456140350877193, "percentage": 7.46, "elapsed_time": "0:03:31", "remaining_time": "0:43:43"}
+{"current_steps": 18, "total_steps": 228, "loss": 0.9812, "lr": 9.938996398223802e-06, "epoch": 0.07894736842105263, "percentage": 7.89, "elapsed_time": "0:03:45", "remaining_time": "0:43:45"}
+{"current_steps": 19, "total_steps": 228, "loss": 1.0288, "lr": 9.927428798017738e-06, "epoch": 0.08333333333333333, "percentage": 8.33, "elapsed_time": "0:03:57", "remaining_time": "0:43:37"}
+{"current_steps": 20, "total_steps": 228, "loss": 1.0403, "lr": 9.91486549841951e-06, "epoch": 0.08771929824561403, "percentage": 8.77, "elapsed_time": "0:04:10", "remaining_time": "0:43:24"}
+{"current_steps": 21, "total_steps": 228, "loss": 1.046, "lr": 9.901309038130392e-06, "epoch": 0.09210526315789473, "percentage": 9.21, "elapsed_time": "0:04:22", "remaining_time": "0:43:07"}
+{"current_steps": 22, "total_steps": 228, "loss": 1.0609, "lr": 9.886762156542428e-06, "epoch": 0.09649122807017543, "percentage": 9.65, "elapsed_time": "0:04:33", "remaining_time": "0:42:44"}
+{"current_steps": 23, "total_steps": 228, "loss": 1.03, "lr": 9.871227793184893e-06, "epoch": 0.10087719298245613, "percentage": 10.09, "elapsed_time": "0:04:45", "remaining_time": "0:42:26"}
+{"current_steps": 24, "total_steps": 228, "loss": 1.0388, "lr": 9.854709087130261e-06, "epoch": 0.10526315789473684, "percentage": 10.53, "elapsed_time": "0:04:56", "remaining_time": "0:42:00"}
+{"current_steps": 25, "total_steps": 228, "loss": 1.0588, "lr": 9.837209376359918e-06, "epoch": 0.10964912280701754, "percentage": 10.96, "elapsed_time": "0:05:08", "remaining_time": "0:41:47"}
+{"current_steps": 26, "total_steps": 228, "loss": 1.0688, "lr": 9.81873219708962e-06, "epoch": 0.11403508771929824, "percentage": 11.4, "elapsed_time": "0:05:20", "remaining_time": "0:41:31"}
+{"current_steps": 27, "total_steps": 228, "loss": 1.0311, "lr": 9.79928128305494e-06, "epoch": 0.11842105263157894, "percentage": 11.84, "elapsed_time": "0:05:32", "remaining_time": "0:41:18"}
+{"current_steps": 28, "total_steps": 228, "loss": 1.0086, "lr": 9.778860564756769e-06, "epoch": 0.12280701754385964, "percentage": 12.28, "elapsed_time": "0:05:45", "remaining_time": "0:41:07"}
+{"current_steps": 29, "total_steps": 228, "loss": 1.0627, "lr": 9.757474168667072e-06, "epoch": 0.12719298245614036, "percentage": 12.72, "elapsed_time": "0:05:57", "remaining_time": "0:40:55"}
+{"current_steps": 30, "total_steps": 228, "loss": 1.0349, "lr": 9.73512641639504e-06, "epoch": 0.13157894736842105, "percentage": 13.16, "elapsed_time": "0:06:10", "remaining_time": "0:40:43"}
+{"current_steps": 31, "total_steps": 228, "loss": 1.1047, "lr": 9.711821823813812e-06, "epoch": 0.13596491228070176, "percentage": 13.6, "elapsed_time": "0:06:22", "remaining_time": "0:40:28"}
+{"current_steps": 32, "total_steps": 228, "loss": 1.0646, "lr": 9.68756510014794e-06, "epoch": 0.14035087719298245, "percentage": 14.04, "elapsed_time": "0:06:33", "remaining_time": "0:40:12"}
+{"current_steps": 33, "total_steps": 228, "loss": 1.0211, "lr": 9.66236114702178e-06, "epoch": 0.14473684210526316, "percentage": 14.47, "elapsed_time": "0:06:46", "remaining_time": "0:40:02"}
+{"current_steps": 34, "total_steps": 228, "loss": 1.03, "lr": 9.636215057469009e-06, "epoch": 0.14912280701754385, "percentage": 14.91, "elapsed_time": "0:06:58", "remaining_time": "0:39:49"}
+{"current_steps": 35, "total_steps": 228, "loss": 1.0495, "lr": 9.609132114903458e-06, "epoch": 0.15350877192982457, "percentage": 15.35, "elapsed_time": "0:07:11", "remaining_time": "0:39:39"}
+{"current_steps": 36, "total_steps": 228, "loss": 1.0653, "lr": 9.581117792051487e-06, "epoch": 0.15789473684210525, "percentage": 15.79, "elapsed_time": "0:07:23", "remaining_time": "0:39:27"}
+{"current_steps": 37, "total_steps": 228, "loss": 1.0085, "lr": 9.552177749846083e-06, "epoch": 0.16228070175438597, "percentage": 16.23, "elapsed_time": "0:07:36", "remaining_time": "0:39:14"}
+{"current_steps": 38, "total_steps": 228, "loss": 1.0548, "lr": 9.522317836282949e-06, "epoch": 0.16666666666666666, "percentage": 16.67, "elapsed_time": "0:07:48", "remaining_time": "0:39:02"}
+{"current_steps": 39, "total_steps": 228, "loss": 1.0704, "lr": 9.491544085238778e-06, "epoch": 0.17105263157894737, "percentage": 17.11, "elapsed_time": "0:08:00", "remaining_time": "0:38:50"}
+{"current_steps": 40, "total_steps": 228, "loss": 1.0463, "lr": 9.459862715251973e-06, "epoch": 0.17543859649122806, "percentage": 17.54, "elapsed_time": "0:08:14", "remaining_time": "0:38:45"}
+{"current_steps": 41, "total_steps": 228, "loss": 1.0771, "lr": 9.427280128266049e-06, "epoch": 0.17982456140350878, "percentage": 17.98, "elapsed_time": "0:08:27", "remaining_time": "0:38:33"}
+{"current_steps": 42, "total_steps": 228, "loss": 1.0615, "lr": 9.393802908335978e-06, "epoch": 0.18421052631578946, "percentage": 18.42, "elapsed_time": "0:08:39", "remaining_time": "0:38:20"}
+{"current_steps": 43, "total_steps": 228, "loss": 1.0431, "lr": 9.359437820297716e-06, "epoch": 0.18859649122807018, "percentage": 18.86, "elapsed_time": "0:08:52", "remaining_time": "0:38:10"}
+{"current_steps": 44, "total_steps": 228, "loss": 1.0271, "lr": 9.324191808401235e-06, "epoch": 0.19298245614035087, "percentage": 19.3, "elapsed_time": "0:09:03", "remaining_time": "0:37:54"}
+{"current_steps": 45, "total_steps": 228, "loss": 1.0765, "lr": 9.288071994907262e-06, "epoch": 0.19736842105263158, "percentage": 19.74, "elapsed_time": "0:09:15", "remaining_time": "0:37:38"}
+{"current_steps": 46, "total_steps": 228, "loss": 1.0526, "lr": 9.251085678648072e-06, "epoch": 0.20175438596491227, "percentage": 20.18, "elapsed_time": "0:09:28", "remaining_time": "0:37:27"}
+{"current_steps": 47, "total_steps": 228, "loss": 1.0502, "lr": 9.213240333552589e-06, "epoch": 0.20614035087719298, "percentage": 20.61, "elapsed_time": "0:09:41", "remaining_time": "0:37:17"}
+{"current_steps": 48, "total_steps": 228, "loss": 0.9878, "lr": 9.174543607136111e-06, "epoch": 0.21052631578947367, "percentage": 21.05, "elapsed_time": "0:09:54", "remaining_time": "0:37:07"}
+{"current_steps": 49, "total_steps": 228, "loss": 0.9989, "lr": 9.135003318954954e-06, "epoch": 0.2149122807017544, "percentage": 21.49, "elapsed_time": "0:10:06", "remaining_time": "0:36:53"}
+{"current_steps": 50, "total_steps": 228, "loss": 1.0816, "lr": 9.094627459026326e-06, "epoch": 0.21929824561403508, "percentage": 21.93, "elapsed_time": "0:10:17", "remaining_time": "0:36:38"}
+{"current_steps": 51, "total_steps": 228, "loss": 1.0509, "lr": 9.053424186213776e-06, "epoch": 0.2236842105263158, "percentage": 22.37, "elapsed_time": "0:10:30", "remaining_time": "0:36:28"}
+{"current_steps": 52, "total_steps": 228, "loss": 0.9849, "lr": 9.011401826578492e-06, "epoch": 0.22807017543859648, "percentage": 22.81, "elapsed_time": "0:10:44", "remaining_time": "0:36:22"}
+{"current_steps": 53, "total_steps": 228, "loss": 0.9975, "lr": 8.968568871696847e-06, "epoch": 0.2324561403508772, "percentage": 23.25, "elapsed_time": "0:10:56", "remaining_time": "0:36:08"}
+{"current_steps": 54, "total_steps": 228, "loss": 1.0747, "lr": 8.924933976944474e-06, "epoch": 0.23684210526315788, "percentage": 23.68, "elapsed_time": "0:11:10", "remaining_time": "0:35:59"}
+{"current_steps": 55, "total_steps": 228, "loss": 1.028, "lr": 8.880505959747245e-06, "epoch": 0.2412280701754386, "percentage": 24.12, "elapsed_time": "0:11:22", "remaining_time": "0:35:47"}
+{"current_steps": 56, "total_steps": 228, "loss": 0.9659, "lr": 8.835293797799517e-06, "epoch": 0.24561403508771928, "percentage": 24.56, "elapsed_time": "0:11:34", "remaining_time": "0:35:33"}
+{"current_steps": 57, "total_steps": 228, "loss": 1.0228, "lr": 8.789306627249985e-06, "epoch": 0.25, "percentage": 25.0, "elapsed_time": "0:11:47", "remaining_time": "0:35:21"}
+{"current_steps": 58, "total_steps": 228, "loss": 1.0338, "lr": 8.742553740855507e-06, "epoch": 0.2543859649122807, "percentage": 25.44, "elapsed_time": "0:11:59", "remaining_time": "0:35:07"}
+{"current_steps": 59, "total_steps": 228, "loss": 1.0344, "lr": 8.695044586103297e-06, "epoch": 0.25877192982456143, "percentage": 25.88, "elapsed_time": "0:12:12", "remaining_time": "0:34:57"}
+{"current_steps": 60, "total_steps": 228, "loss": 1.0423, "lr": 8.646788763301842e-06, "epoch": 0.2631578947368421, "percentage": 26.32, "elapsed_time": "0:12:24", "remaining_time": "0:34:45"}
+{"current_steps": 61, "total_steps": 228, "loss": 1.0372, "lr": 8.59779602364094e-06, "epoch": 0.2675438596491228, "percentage": 26.75, "elapsed_time": "0:12:37", "remaining_time": "0:34:32"}
+{"current_steps": 62, "total_steps": 228, "loss": 1.0935, "lr": 8.548076267221258e-06, "epoch": 0.2719298245614035, "percentage": 27.19, "elapsed_time": "0:12:48", "remaining_time": "0:34:18"}
+{"current_steps": 63, "total_steps": 228, "loss": 0.9915, "lr": 8.497639541053769e-06, "epoch": 0.27631578947368424, "percentage": 27.63, "elapsed_time": "0:13:01", "remaining_time": "0:34:07"}
+{"current_steps": 64, "total_steps": 228, "loss": 1.0252, "lr": 8.446496037029555e-06, "epoch": 0.2807017543859649, "percentage": 28.07, "elapsed_time": "0:13:17", "remaining_time": "0:34:02"}
+{"current_steps": 65, "total_steps": 228, "loss": 0.9989, "lr": 8.394656089860274e-06, "epoch": 0.2850877192982456, "percentage": 28.51, "elapsed_time": "0:13:28", "remaining_time": "0:33:48"}
+{"current_steps": 66, "total_steps": 228, "loss": 1.0872, "lr": 8.342130174989819e-06, "epoch": 0.2894736842105263, "percentage": 28.95, "elapsed_time": "0:13:40", "remaining_time": "0:33:33"}
+{"current_steps": 67, "total_steps": 228, "loss": 1.0545, "lr": 8.288928906477497e-06, "epoch": 0.29385964912280704, "percentage": 29.39, "elapsed_time": "0:13:52", "remaining_time": "0:33:20"}
+{"current_steps": 68, "total_steps": 228, "loss": 1.0128, "lr": 8.235063034853228e-06, "epoch": 0.2982456140350877, "percentage": 29.82, "elapsed_time": "0:14:05", "remaining_time": "0:33:08"}
+{"current_steps": 69, "total_steps": 228, "loss": 1.0294, "lr": 8.180543444945154e-06, "epoch": 0.3026315789473684, "percentage": 30.26, "elapsed_time": "0:14:17", "remaining_time": "0:32:55"}
+{"current_steps": 70, "total_steps": 228, "loss": 1.0445, "lr": 8.125381153680103e-06, "epoch": 0.30701754385964913, "percentage": 30.7, "elapsed_time": "0:14:29", "remaining_time": "0:32:41"}
+{"current_steps": 71, "total_steps": 228, "loss": 1.0168, "lr": 8.069587307857377e-06, "epoch": 0.31140350877192985, "percentage": 31.14, "elapsed_time": "0:14:41", "remaining_time": "0:32:29"}
+{"current_steps": 72, "total_steps": 228, "loss": 1.0207, "lr": 8.013173181896283e-06, "epoch": 0.3157894736842105, "percentage": 31.58, "elapsed_time": "0:14:54", "remaining_time": "0:32:18"}
+{"current_steps": 73, "total_steps": 228, "loss": 1.0401, "lr": 7.95615017555788e-06, "epoch": 0.3201754385964912, "percentage": 32.02, "elapsed_time": "0:15:05", "remaining_time": "0:32:01"}
+{"current_steps": 74, "total_steps": 228, "loss": 1.0347, "lr": 7.898529811641393e-06, "epoch": 0.32456140350877194, "percentage": 32.46, "elapsed_time": "0:15:17", "remaining_time": "0:31:49"}
+{"current_steps": 75, "total_steps": 228, "loss": 1.061, "lr": 7.84032373365578e-06, "epoch": 0.32894736842105265, "percentage": 32.89, "elapsed_time": "0:15:31", "remaining_time": "0:31:40"}
+{"current_steps": 76, "total_steps": 228, "loss": 0.988, "lr": 7.781543703466881e-06, "epoch": 0.3333333333333333, "percentage": 33.33, "elapsed_time": "0:15:44", "remaining_time": "0:31:28"}
+{"current_steps": 77, "total_steps": 228, "loss": 1.0442, "lr": 7.722201598920673e-06, "epoch": 0.33771929824561403, "percentage": 33.77, "elapsed_time": "0:15:57", "remaining_time": "0:31:16"}
+{"current_steps": 78, "total_steps": 228, "loss": 1.0412, "lr": 7.662309411443084e-06, "epoch": 0.34210526315789475, "percentage": 34.21, "elapsed_time": "0:16:09", "remaining_time": "0:31:04"}
+{"current_steps": 79, "total_steps": 228, "loss": 1.0205, "lr": 7.601879243616838e-06, "epoch": 0.34649122807017546, "percentage": 34.65, "elapsed_time": "0:16:22", "remaining_time": "0:30:52"}
+{"current_steps": 80, "total_steps": 228, "loss": 1.0298, "lr": 7.540923306735868e-06, "epoch": 0.3508771929824561, "percentage": 35.09, "elapsed_time": "0:16:34", "remaining_time": "0:30:40"}
+{"current_steps": 81, "total_steps": 228, "loss": 1.0174, "lr": 7.479453918337733e-06, "epoch": 0.35526315789473684, "percentage": 35.53, "elapsed_time": "0:16:47", "remaining_time": "0:30:27"}
+{"current_steps": 82, "total_steps": 228, "loss": 1.0281, "lr": 7.417483499714589e-06, "epoch": 0.35964912280701755, "percentage": 35.96, "elapsed_time": "0:16:59", "remaining_time": "0:30:14"}
+{"current_steps": 83, "total_steps": 228, "loss": 1.0045, "lr": 7.355024573403174e-06, "epoch": 0.36403508771929827, "percentage": 36.4, "elapsed_time": "0:17:11", "remaining_time": "0:30:01"}
+{"current_steps": 84, "total_steps": 228, "loss": 1.0253, "lr": 7.292089760654352e-06, "epoch": 0.3684210526315789, "percentage": 36.84, "elapsed_time": "0:17:23", "remaining_time": "0:29:48"}
+{"current_steps": 85, "total_steps": 228, "loss": 1.0245, "lr": 7.2286917788826926e-06, "epoch": 0.37280701754385964, "percentage": 37.28, "elapsed_time": "0:17:35", "remaining_time": "0:29:35"}
+{"current_steps": 86, "total_steps": 228, "loss": 1.0149, "lr": 7.1648434390966356e-06, "epoch": 0.37719298245614036, "percentage": 37.72, "elapsed_time": "0:17:47", "remaining_time": "0:29:22"}
+{"current_steps": 87, "total_steps": 228, "loss": 1.0019, "lr": 7.100557643309732e-06, "epoch": 0.3815789473684211, "percentage": 38.16, "elapsed_time": "0:17:59", "remaining_time": "0:29:10"}
+{"current_steps": 88, "total_steps": 228, "loss": 1.0147, "lr": 7.035847381933494e-06, "epoch": 0.38596491228070173, "percentage": 38.6, "elapsed_time": "0:18:11", "remaining_time": "0:28:56"}
+{"current_steps": 89, "total_steps": 228, "loss": 0.9572, "lr": 6.970725731152389e-06, "epoch": 0.39035087719298245, "percentage": 39.04, "elapsed_time": "0:18:23", "remaining_time": "0:28:42"}
+{"current_steps": 90, "total_steps": 228, "loss": 0.9651, "lr": 6.905205850281502e-06, "epoch": 0.39473684210526316, "percentage": 39.47, "elapsed_time": "0:18:35", "remaining_time": "0:28:30"}
+{"current_steps": 91, "total_steps": 228, "loss": 1.0471, "lr": 6.8393009791073895e-06, "epoch": 0.3991228070175439, "percentage": 39.91, "elapsed_time": "0:18:46", "remaining_time": "0:28:15"}
+{"current_steps": 92, "total_steps": 228, "loss": 1.0062, "lr": 6.773024435212678e-06, "epoch": 0.40350877192982454, "percentage": 40.35, "elapsed_time": "0:18:57", "remaining_time": "0:28:01"}
+{"current_steps": 93, "total_steps": 228, "loss": 1.0101, "lr": 6.706389611284953e-06, "epoch": 0.40789473684210525, "percentage": 40.79, "elapsed_time": "0:19:10", "remaining_time": "0:27:49"}
+{"current_steps": 94, "total_steps": 228, "loss": 1.0208, "lr": 6.639409972410446e-06, "epoch": 0.41228070175438597, "percentage": 41.23, "elapsed_time": "0:19:21", "remaining_time": "0:27:36"}
+{"current_steps": 95, "total_steps": 228, "loss": 1.0014, "lr": 6.57209905335312e-06, "epoch": 0.4166666666666667, "percentage": 41.67, "elapsed_time": "0:19:34", "remaining_time": "0:27:24"}
+{"current_steps": 96, "total_steps": 228, "loss": 0.9726, "lr": 6.504470455819651e-06, "epoch": 0.42105263157894735, "percentage": 42.11, "elapsed_time": "0:19:45", "remaining_time": "0:27:10"}
+{"current_steps": 97, "total_steps": 228, "loss": 1.0462, "lr": 6.436537845710904e-06, "epoch": 0.42543859649122806, "percentage": 42.54, "elapsed_time": "0:19:58", "remaining_time": "0:26:58"}
+{"current_steps": 98, "total_steps": 228, "loss": 0.997, "lr": 6.368314950360416e-06, "epoch": 0.4298245614035088, "percentage": 42.98, "elapsed_time": "0:20:10", "remaining_time": "0:26:45"}
+{"current_steps": 99, "total_steps": 228, "loss": 1.0351, "lr": 6.299815555760478e-06, "epoch": 0.4342105263157895, "percentage": 43.42, "elapsed_time": "0:20:24", "remaining_time": "0:26:35"}
+{"current_steps": 100, "total_steps": 228, "loss": 1.04, "lr": 6.231053503776363e-06, "epoch": 0.43859649122807015, "percentage": 43.86, "elapsed_time": "0:20:36", "remaining_time": "0:26:22"}
+{"current_steps": 101, "total_steps": 228, "loss": 0.9802, "lr": 6.1620426893492645e-06, "epoch": 0.44298245614035087, "percentage": 44.3, "elapsed_time": "0:20:48", "remaining_time": "0:26:09"}
+{"current_steps": 102, "total_steps": 228, "loss": 1.006, "lr": 6.092797057688496e-06, "epoch": 0.4473684210526316, "percentage": 44.74, "elapsed_time": "0:21:01", "remaining_time": "0:25:57"}
+{"current_steps": 103, "total_steps": 228, "loss": 0.9833, "lr": 6.0233306014535505e-06, "epoch": 0.4517543859649123, "percentage": 45.18, "elapsed_time": "0:21:12", "remaining_time": "0:25:44"}
+{"current_steps": 104, "total_steps": 228, "loss": 1.0409, "lr": 5.953657357926569e-06, "epoch": 0.45614035087719296, "percentage": 45.61, "elapsed_time": "0:21:24", "remaining_time": "0:25:31"}
+{"current_steps": 105, "total_steps": 228, "loss": 1.0191, "lr": 5.883791406175775e-06, "epoch": 0.4605263157894737, "percentage": 46.05, "elapsed_time": "0:21:37", "remaining_time": "0:25:19"}
+{"current_steps": 106, "total_steps": 228, "loss": 0.9992, "lr": 5.813746864210489e-06, "epoch": 0.4649122807017544, "percentage": 46.49, "elapsed_time": "0:21:50", "remaining_time": "0:25:08"}
+{"current_steps": 107, "total_steps": 228, "loss": 1.0023, "lr": 5.743537886128258e-06, "epoch": 0.4692982456140351, "percentage": 46.93, "elapsed_time": "0:22:02", "remaining_time": "0:24:55"}
+{"current_steps": 108, "total_steps": 228, "loss": 0.9909, "lr": 5.673178659254698e-06, "epoch": 0.47368421052631576, "percentage": 47.37, "elapsed_time": "0:22:14", "remaining_time": "0:24:42"}
+{"current_steps": 109, "total_steps": 228, "loss": 0.9598, "lr": 5.6026834012766155e-06, "epoch": 0.4780701754385965, "percentage": 47.81, "elapsed_time": "0:22:25", "remaining_time": "0:24:29"}
+{"current_steps": 110, "total_steps": 228, "loss": 0.9879, "lr": 5.532066357369012e-06, "epoch": 0.4824561403508772, "percentage": 48.25, "elapsed_time": "0:22:37", "remaining_time": "0:24:16"}
+{"current_steps": 111, "total_steps": 228, "loss": 1.0184, "lr": 5.46134179731651e-06, "epoch": 0.4868421052631579, "percentage": 48.68, "elapsed_time": "0:22:50", "remaining_time": "0:24:04"}
+{"current_steps": 112, "total_steps": 228, "loss": 1.0496, "lr": 5.390524012629824e-06, "epoch": 0.49122807017543857, "percentage": 49.12, "elapsed_time": "0:23:02", "remaining_time": "0:23:51"}
+{"current_steps": 113, "total_steps": 228, "loss": 1.0399, "lr": 5.319627313657829e-06, "epoch": 0.4956140350877193, "percentage": 49.56, "elapsed_time": "0:23:14", "remaining_time": "0:23:39"}
+{"current_steps": 114, "total_steps": 228, "loss": 0.9906, "lr": 5.248666026695835e-06, "epoch": 0.5, "percentage": 50.0, "elapsed_time": "0:23:26", "remaining_time": "0:23:26"}
+{"current_steps": 115, "total_steps": 228, "loss": 0.9836, "lr": 5.177654491090627e-06, "epoch": 0.5043859649122807, "percentage": 50.44, "elapsed_time": "0:23:37", "remaining_time": "0:23:13"}
+{"current_steps": 116, "total_steps": 228, "loss": 1.0083, "lr": 5.1066070563428736e-06, "epoch": 0.5087719298245614, "percentage": 50.88, "elapsed_time": "0:23:49", "remaining_time": "0:23:00"}
+{"current_steps": 117, "total_steps": 228, "loss": 1.0034, "lr": 5.035538079207488e-06, "epoch": 0.5131578947368421, "percentage": 51.32, "elapsed_time": "0:24:01", "remaining_time": "0:22:47"}
+{"current_steps": 118, "total_steps": 228, "loss": 1.0244, "lr": 4.964461920792512e-06, "epoch": 0.5175438596491229, "percentage": 51.75, "elapsed_time": "0:24:13", "remaining_time": "0:22:35"}
+{"current_steps": 119, "total_steps": 228, "loss": 0.9473, "lr": 4.893392943657127e-06, "epoch": 0.5219298245614035, "percentage": 52.19, "elapsed_time": "0:24:26", "remaining_time": "0:22:22"}
+{"current_steps": 120, "total_steps": 228, "loss": 0.93, "lr": 4.822345508909376e-06, "epoch": 0.5263157894736842, "percentage": 52.63, "elapsed_time": "0:24:40", "remaining_time": "0:22:12"}
+{"current_steps": 121, "total_steps": 228, "loss": 1.0262, "lr": 4.751333973304166e-06, "epoch": 0.5307017543859649, "percentage": 53.07, "elapsed_time": "0:24:53", "remaining_time": "0:22:00"}
+{"current_steps": 122, "total_steps": 228, "loss": 1.0236, "lr": 4.680372686342173e-06, "epoch": 0.5350877192982456, "percentage": 53.51, "elapsed_time": "0:25:05", "remaining_time": "0:21:48"}
+{"current_steps": 123, "total_steps": 228, "loss": 0.9891, "lr": 4.609475987370177e-06, "epoch": 0.5394736842105263, "percentage": 53.95, "elapsed_time": "0:25:17", "remaining_time": "0:21:35"}
+{"current_steps": 124, "total_steps": 228, "loss": 0.9761, "lr": 4.53865820268349e-06, "epoch": 0.543859649122807, "percentage": 54.39, "elapsed_time": "0:25:29", "remaining_time": "0:21:22"}
+{"current_steps": 125, "total_steps": 228, "loss": 0.9847, "lr": 4.467933642630989e-06, "epoch": 0.5482456140350878, "percentage": 54.82, "elapsed_time": "0:25:41", "remaining_time": "0:21:10"}
+{"current_steps": 126, "total_steps": 228, "loss": 0.9916, "lr": 4.397316598723385e-06, "epoch": 0.5526315789473685, "percentage": 55.26, "elapsed_time": "0:25:53", "remaining_time": "0:20:57"}
+{"current_steps": 127, "total_steps": 228, "loss": 0.9535, "lr": 4.326821340745304e-06, "epoch": 0.5570175438596491, "percentage": 55.7, "elapsed_time": "0:26:05", "remaining_time": "0:20:45"}
+{"current_steps": 128, "total_steps": 228, "loss": 0.9249, "lr": 4.256462113871741e-06, "epoch": 0.5614035087719298, "percentage": 56.14, "elapsed_time": "0:26:17", "remaining_time": "0:20:32"}
+{"current_steps": 129, "total_steps": 228, "loss": 0.9885, "lr": 4.186253135789511e-06, "epoch": 0.5657894736842105, "percentage": 56.58, "elapsed_time": "0:26:28", "remaining_time": "0:20:19"}
+{"current_steps": 130, "total_steps": 228, "loss": 1.0087, "lr": 4.116208593824227e-06, "epoch": 0.5701754385964912, "percentage": 57.02, "elapsed_time": "0:26:40", "remaining_time": "0:20:06"}
+{"current_steps": 131, "total_steps": 228, "loss": 0.9654, "lr": 4.046342642073433e-06, "epoch": 0.5745614035087719, "percentage": 57.46, "elapsed_time": "0:26:52", "remaining_time": "0:19:53"}
+{"current_steps": 132, "total_steps": 228, "loss": 0.991, "lr": 3.976669398546451e-06, "epoch": 0.5789473684210527, "percentage": 57.89, "elapsed_time": "0:27:03", "remaining_time": "0:19:40"}
+{"current_steps": 133, "total_steps": 228, "loss": 0.9702, "lr": 3.907202942311506e-06, "epoch": 0.5833333333333334, "percentage": 58.33, "elapsed_time": "0:27:15", "remaining_time": "0:19:27"}
+{"current_steps": 134, "total_steps": 228, "loss": 0.9451, "lr": 3.837957310650738e-06, "epoch": 0.5877192982456141, "percentage": 58.77, "elapsed_time": "0:27:27", "remaining_time": "0:19:15"}
+{"current_steps": 135, "total_steps": 228, "loss": 0.9498, "lr": 3.7689464962236367e-06, "epoch": 0.5921052631578947, "percentage": 59.21, "elapsed_time": "0:27:39", "remaining_time": "0:19:03"}
+{"current_steps": 136, "total_steps": 228, "loss": 0.9828, "lr": 3.700184444239524e-06, "epoch": 0.5964912280701754, "percentage": 59.65, "elapsed_time": "0:27:51", "remaining_time": "0:18:50"}
+{"current_steps": 137, "total_steps": 228, "loss": 0.956, "lr": 3.6316850496395863e-06, "epoch": 0.6008771929824561, "percentage": 60.09, "elapsed_time": "0:28:03", "remaining_time": "0:18:38"}
+{"current_steps": 138, "total_steps": 228, "loss": 0.9553, "lr": 3.563462154289098e-06, "epoch": 0.6052631578947368, "percentage": 60.53, "elapsed_time": "0:28:14", "remaining_time": "0:18:24"}
+{"current_steps": 139, "total_steps": 228, "loss": 1.0269, "lr": 3.49552954418035e-06, "epoch": 0.6096491228070176, "percentage": 60.96, "elapsed_time": "0:28:28", "remaining_time": "0:18:13"}
+{"current_steps": 140, "total_steps": 228, "loss": 0.9888, "lr": 3.4279009466468825e-06, "epoch": 0.6140350877192983, "percentage": 61.4, "elapsed_time": "0:28:40", "remaining_time": "0:18:01"}
+{"current_steps": 141, "total_steps": 228, "loss": 1.0003, "lr": 3.3605900275895565e-06, "epoch": 0.618421052631579, "percentage": 61.84, "elapsed_time": "0:28:52", "remaining_time": "0:17:48"}
+{"current_steps": 142, "total_steps": 228, "loss": 0.9982, "lr": 3.2936103887150484e-06, "epoch": 0.6228070175438597, "percentage": 62.28, "elapsed_time": "0:29:04", "remaining_time": "0:17:36"}
+{"current_steps": 143, "total_steps": 228, "loss": 0.9645, "lr": 3.226975564787322e-06, "epoch": 0.6271929824561403, "percentage": 62.72, "elapsed_time": "0:29:16", "remaining_time": "0:17:24"}
+{"current_steps": 144, "total_steps": 228, "loss": 0.9443, "lr": 3.1606990208926125e-06, "epoch": 0.631578947368421, "percentage": 63.16, "elapsed_time": "0:29:28", "remaining_time": "0:17:11"}
+{"current_steps": 145, "total_steps": 228, "loss": 1.0069, "lr": 3.0947941497184985e-06, "epoch": 0.6359649122807017, "percentage": 63.6, "elapsed_time": "0:29:40", "remaining_time": "0:16:59"}
+{"current_steps": 146, "total_steps": 228, "loss": 0.9718, "lr": 3.0292742688476125e-06, "epoch": 0.6403508771929824, "percentage": 64.04, "elapsed_time": "0:29:52", "remaining_time": "0:16:46"}
+{"current_steps": 147, "total_steps": 228, "loss": 0.9981, "lr": 2.964152618066508e-06, "epoch": 0.6447368421052632, "percentage": 64.47, "elapsed_time": "0:30:03", "remaining_time": "0:16:33"}
+{"current_steps": 148, "total_steps": 228, "loss": 0.994, "lr": 2.899442356690271e-06, "epoch": 0.6491228070175439, "percentage": 64.91, "elapsed_time": "0:30:16", "remaining_time": "0:16:21"}
+{"current_steps": 149, "total_steps": 228, "loss": 0.9605, "lr": 2.835156560903365e-06, "epoch": 0.6535087719298246, "percentage": 65.35, "elapsed_time": "0:30:26", "remaining_time": "0:16:08"}
+{"current_steps": 150, "total_steps": 228, "loss": 0.9965, "lr": 2.771308221117309e-06, "epoch": 0.6578947368421053, "percentage": 65.79, "elapsed_time": "0:30:38", "remaining_time": "0:15:55"}
+{"current_steps": 151, "total_steps": 228, "loss": 1.0161, "lr": 2.7079102393456503e-06, "epoch": 0.6622807017543859, "percentage": 66.23, "elapsed_time": "0:30:49", "remaining_time": "0:15:43"}
+{"current_steps": 152, "total_steps": 228, "loss": 0.9696, "lr": 2.6449754265968263e-06, "epoch": 0.6666666666666666, "percentage": 66.67, "elapsed_time": "0:31:01", "remaining_time": "0:15:30"}
+{"current_steps": 153, "total_steps": 228, "loss": 0.9674, "lr": 2.5825165002854124e-06, "epoch": 0.6710526315789473, "percentage": 67.11, "elapsed_time": "0:31:13", "remaining_time": "0:15:18"}
+{"current_steps": 154, "total_steps": 228, "loss": 0.9933, "lr": 2.5205460816622684e-06, "epoch": 0.6754385964912281, "percentage": 67.54, "elapsed_time": "0:31:24", "remaining_time": "0:15:05"}
+{"current_steps": 155, "total_steps": 228, "loss": 0.977, "lr": 2.4590766932641353e-06, "epoch": 0.6798245614035088, "percentage": 67.98, "elapsed_time": "0:31:36", "remaining_time": "0:14:53"}
+{"current_steps": 156, "total_steps": 228, "loss": 0.9729, "lr": 2.3981207563831633e-06, "epoch": 0.6842105263157895, "percentage": 68.42, "elapsed_time": "0:31:48", "remaining_time": "0:14:41"}
+{"current_steps": 157, "total_steps": 228, "loss": 0.914, "lr": 2.3376905885569185e-06, "epoch": 0.6885964912280702, "percentage": 68.86, "elapsed_time": "0:32:01", "remaining_time": "0:14:28"}
+{"current_steps": 158, "total_steps": 228, "loss": 0.9641, "lr": 2.2777984010793264e-06, "epoch": 0.6929824561403509, "percentage": 69.3, "elapsed_time": "0:32:12", "remaining_time": "0:14:16"}
+{"current_steps": 159, "total_steps": 228, "loss": 0.9074, "lr": 2.2184562965331203e-06, "epoch": 0.6973684210526315, "percentage": 69.74, "elapsed_time": "0:32:25", "remaining_time": "0:14:04"}
+{"current_steps": 160, "total_steps": 228, "loss": 0.938, "lr": 2.159676266344222e-06, "epoch": 0.7017543859649122, "percentage": 70.18, "elapsed_time": "0:32:38", "remaining_time": "0:13:52"}
+{"current_steps": 161, "total_steps": 228, "loss": 0.9387, "lr": 2.1014701883586087e-06, "epoch": 0.706140350877193, "percentage": 70.61, "elapsed_time": "0:32:51", "remaining_time": "0:13:40"}
+{"current_steps": 162, "total_steps": 228, "loss": 0.9931, "lr": 2.043849824442124e-06, "epoch": 0.7105263157894737, "percentage": 71.05, "elapsed_time": "0:33:03", "remaining_time": "0:13:28"}
+{"current_steps": 163, "total_steps": 228, "loss": 0.9578, "lr": 1.9868268181037186e-06, "epoch": 0.7149122807017544, "percentage": 71.49, "elapsed_time": "0:33:15", "remaining_time": "0:13:15"}
+{"current_steps": 164, "total_steps": 228, "loss": 0.9615, "lr": 1.9304126921426235e-06, "epoch": 0.7192982456140351, "percentage": 71.93, "elapsed_time": "0:33:27", "remaining_time": "0:13:03"}
+{"current_steps": 165, "total_steps": 228, "loss": 0.952, "lr": 1.8746188463198983e-06, "epoch": 0.7236842105263158, "percentage": 72.37, "elapsed_time": "0:33:40", "remaining_time": "0:12:51"}
+{"current_steps": 166, "total_steps": 228, "loss": 0.9422, "lr": 1.8194565550548477e-06, "epoch": 0.7280701754385965, "percentage": 72.81, "elapsed_time": "0:33:53", "remaining_time": "0:12:39"}
+{"current_steps": 167, "total_steps": 228, "loss": 0.9556, "lr": 1.764936965146773e-06, "epoch": 0.7324561403508771, "percentage": 73.25, "elapsed_time": "0:34:05", "remaining_time": "0:12:26"}
+{"current_steps": 168, "total_steps": 228, "loss": 0.9609, "lr": 1.7110710935225055e-06, "epoch": 0.7368421052631579, "percentage": 73.68, "elapsed_time": "0:34:17", "remaining_time": "0:12:14"}
+{"current_steps": 169, "total_steps": 228, "loss": 0.9945, "lr": 1.6578698250101828e-06, "epoch": 0.7412280701754386, "percentage": 74.12, "elapsed_time": "0:34:29", "remaining_time": "0:12:02"}
+{"current_steps": 170, "total_steps": 228, "loss": 0.9526, "lr": 1.6053439101397257e-06, "epoch": 0.7456140350877193, "percentage": 74.56, "elapsed_time": "0:34:42", "remaining_time": "0:11:50"}
+{"current_steps": 171, "total_steps": 228, "loss": 0.953, "lr": 1.5535039629704467e-06, "epoch": 0.75, "percentage": 75.0, "elapsed_time": "0:34:54", "remaining_time": "0:11:38"}
+{"current_steps": 172, "total_steps": 228, "loss": 0.9418, "lr": 1.502360458946232e-06, "epoch": 0.7543859649122807, "percentage": 75.44, "elapsed_time": "0:35:07", "remaining_time": "0:11:26"}
+{"current_steps": 173, "total_steps": 228, "loss": 0.9842, "lr": 1.451923732778745e-06, "epoch": 0.7587719298245614, "percentage": 75.88, "elapsed_time": "0:35:19", "remaining_time": "0:11:13"}
+{"current_steps": 174, "total_steps": 228, "loss": 0.9483, "lr": 1.4022039763590595e-06, "epoch": 0.7631578947368421, "percentage": 76.32, "elapsed_time": "0:35:32", "remaining_time": "0:11:01"}
+{"current_steps": 175, "total_steps": 228, "loss": 1.0225, "lr": 1.3532112366981598e-06, "epoch": 0.7675438596491229, "percentage": 76.75, "elapsed_time": "0:35:45", "remaining_time": "0:10:49"}
+{"current_steps": 176, "total_steps": 228, "loss": 0.9787, "lr": 1.3049554138967052e-06, "epoch": 0.7719298245614035, "percentage": 77.19, "elapsed_time": "0:35:57", "remaining_time": "0:10:37"}
+{"current_steps": 177, "total_steps": 228, "loss": 0.9812, "lr": 1.257446259144494e-06, "epoch": 0.7763157894736842, "percentage": 77.63, "elapsed_time": "0:36:09", "remaining_time": "0:10:25"}
+{"current_steps": 178, "total_steps": 228, "loss": 0.9686, "lr": 1.210693372750017e-06, "epoch": 0.7807017543859649, "percentage": 78.07, "elapsed_time": "0:36:21", "remaining_time": "0:10:12"}
+{"current_steps": 179, "total_steps": 228, "loss": 0.9909, "lr": 1.1647062022004845e-06, "epoch": 0.7850877192982456, "percentage": 78.51, "elapsed_time": "0:36:32", "remaining_time": "0:10:00"}
+{"current_steps": 180, "total_steps": 228, "loss": 0.9682, "lr": 1.1194940402527566e-06, "epoch": 0.7894736842105263, "percentage": 78.95, "elapsed_time": "0:36:44", "remaining_time": "0:09:47"}
+{"current_steps": 181, "total_steps": 228, "loss": 0.9218, "lr": 1.075066023055527e-06, "epoch": 0.793859649122807, "percentage": 79.39, "elapsed_time": "0:36:57", "remaining_time": "0:09:35"}
+{"current_steps": 182, "total_steps": 228, "loss": 0.9753, "lr": 1.0314311283031531e-06, "epoch": 0.7982456140350878, "percentage": 79.82, "elapsed_time": "0:37:10", "remaining_time": "0:09:23"}
+{"current_steps": 183, "total_steps": 228, "loss": 0.9746, "lr": 9.885981734215094e-07, "epoch": 0.8026315789473685, "percentage": 80.26, "elapsed_time": "0:37:23", "remaining_time": "0:09:11"}
+{"current_steps": 184, "total_steps": 228, "loss": 0.9479, "lr": 9.465758137862264e-07, "epoch": 0.8070175438596491, "percentage": 80.7, "elapsed_time": "0:37:35", "remaining_time": "0:08:59"}
+{"current_steps": 185, "total_steps": 228, "loss": 0.9734, "lr": 9.053725409736752e-07, "epoch": 0.8114035087719298, "percentage": 81.14, "elapsed_time": "0:37:48", "remaining_time": "0:08:47"}
+{"current_steps": 186, "total_steps": 228, "loss": 0.9145, "lr": 8.649966810450472e-07, "epoch": 0.8157894736842105, "percentage": 81.58, "elapsed_time": "0:38:01", "remaining_time": "0:08:35"}
+{"current_steps": 187, "total_steps": 228, "loss": 0.9302, "lr": 8.254563928638892e-07, "epoch": 0.8201754385964912, "percentage": 82.02, "elapsed_time": "0:38:14", "remaining_time": "0:08:23"}
+{"current_steps": 188, "total_steps": 228, "loss": 0.9509, "lr": 7.86759666447412e-07, "epoch": 0.8245614035087719, "percentage": 82.46, "elapsed_time": "0:38:26", "remaining_time": "0:08:10"}
+{"current_steps": 189, "total_steps": 228, "loss": 0.9081, "lr": 7.489143213519301e-07, "epoch": 0.8289473684210527, "percentage": 82.89, "elapsed_time": "0:38:38", "remaining_time": "0:07:58"}
+{"current_steps": 190, "total_steps": 228, "loss": 0.9404, "lr": 7.119280050927407e-07, "epoch": 0.8333333333333334, "percentage": 83.33, "elapsed_time": "0:38:49", "remaining_time": "0:07:45"}
+{"current_steps": 191, "total_steps": 228, "loss": 0.9482, "lr": 6.758081915987669e-07, "epoch": 0.8377192982456141, "percentage": 83.77, "elapsed_time": "0:39:02", "remaining_time": "0:07:33"}
+{"current_steps": 192, "total_steps": 228, "loss": 0.9161, "lr": 6.405621797022848e-07, "epoch": 0.8421052631578947, "percentage": 84.21, "elapsed_time": "0:39:13", "remaining_time": "0:07:21"}
+{"current_steps": 193, "total_steps": 228, "loss": 0.9351, "lr": 6.061970916640236e-07, "epoch": 0.8464912280701754, "percentage": 84.65, "elapsed_time": "0:39:25", "remaining_time": "0:07:08"}
+{"current_steps": 194, "total_steps": 228, "loss": 0.9633, "lr": 5.727198717339511e-07, "epoch": 0.8508771929824561, "percentage": 85.09, "elapsed_time": "0:39:39", "remaining_time": "0:06:56"}
+{"current_steps": 195, "total_steps": 228, "loss": 0.9923, "lr": 5.401372847480285e-07, "epoch": 0.8552631578947368, "percentage": 85.53, "elapsed_time": "0:39:50", "remaining_time": "0:06:44"}
+{"current_steps": 196, "total_steps": 228, "loss": 0.9049, "lr": 5.084559147612244e-07, "epoch": 0.8596491228070176, "percentage": 85.96, "elapsed_time": "0:40:01", "remaining_time": "0:06:32"}
+{"current_steps": 197, "total_steps": 228, "loss": 0.9344, "lr": 4.776821637170525e-07, "epoch": 0.8640350877192983, "percentage": 86.4, "elapsed_time": "0:40:14", "remaining_time": "0:06:19"}
+{"current_steps": 198, "total_steps": 228, "loss": 0.9594, "lr": 4.4782225015391754e-07, "epoch": 0.868421052631579, "percentage": 86.84, "elapsed_time": "0:40:25", "remaining_time": "0:06:07"}
+{"current_steps": 199, "total_steps": 228, "loss": 0.9519, "lr": 4.1888220794851386e-07, "epoch": 0.8728070175438597, "percentage": 87.28, "elapsed_time": "0:40:37", "remaining_time": "0:05:55"}
+{"current_steps": 200, "total_steps": 228, "loss": 0.9539, "lr": 3.908678850965425e-07, "epoch": 0.8771929824561403, "percentage": 87.72, "elapsed_time": "0:40:49", "remaining_time": "0:05:42"}
+{"current_steps": 201, "total_steps": 228, "loss": 0.9506, "lr": 3.6378494253099307e-07, "epoch": 0.881578947368421, "percentage": 88.16, "elapsed_time": "0:41:01", "remaining_time": "0:05:30"}
+{"current_steps": 202, "total_steps": 228, "loss": 0.9309, "lr": 3.3763885297822153e-07, "epoch": 0.8859649122807017, "percentage": 88.6, "elapsed_time": "0:41:14", "remaining_time": "0:05:18"}
+{"current_steps": 203, "total_steps": 228, "loss": 0.9497, "lr": 3.1243489985206097e-07, "epoch": 0.8903508771929824, "percentage": 89.04, "elapsed_time": "0:41:27", "remaining_time": "0:05:06"}
+{"current_steps": 204, "total_steps": 228, "loss": 0.9707, "lr": 2.8817817618618846e-07, "epoch": 0.8947368421052632, "percentage": 89.47, "elapsed_time": "0:41:40", "remaining_time": "0:04:54"}
+{"current_steps": 205, "total_steps": 228, "loss": 0.953, "lr": 2.648735836049615e-07, "epoch": 0.8991228070175439, "percentage": 89.91, "elapsed_time": "0:41:53", "remaining_time": "0:04:42"}
+{"current_steps": 206, "total_steps": 228, "loss": 0.9465, "lr": 2.4252583133292927e-07, "epoch": 0.9035087719298246, "percentage": 90.35, "elapsed_time": "0:42:05", "remaining_time": "0:04:29"}
+{"current_steps": 207, "total_steps": 228, "loss": 0.9527, "lr": 2.2113943524323167e-07, "epoch": 0.9078947368421053, "percentage": 90.79, "elapsed_time": "0:42:18", "remaining_time": "0:04:17"}
+{"current_steps": 208, "total_steps": 228, "loss": 0.9498, "lr": 2.007187169450603e-07, "epoch": 0.9122807017543859, "percentage": 91.23, "elapsed_time": "0:42:29", "remaining_time": "0:04:05"}
+{"current_steps": 209, "total_steps": 228, "loss": 0.9582, "lr": 1.8126780291038037e-07, "epoch": 0.9166666666666666, "percentage": 91.67, "elapsed_time": "0:42:43", "remaining_time": "0:03:53"}
+{"current_steps": 210, "total_steps": 228, "loss": 0.9169, "lr": 1.6279062364008446e-07, "epoch": 0.9210526315789473, "percentage": 92.11, "elapsed_time": "0:42:54", "remaining_time": "0:03:40"}
+{"current_steps": 211, "total_steps": 228, "loss": 0.9789, "lr": 1.4529091286973994e-07, "epoch": 0.9254385964912281, "percentage": 92.54, "elapsed_time": "0:43:07", "remaining_time": "0:03:28"}
+{"current_steps": 212, "total_steps": 228, "loss": 0.9705, "lr": 1.2877220681510927e-07, "epoch": 0.9298245614035088, "percentage": 92.98, "elapsed_time": "0:43:20", "remaining_time": "0:03:16"}
+{"current_steps": 213, "total_steps": 228, "loss": 0.9249, "lr": 1.1323784345757205e-07, "epoch": 0.9342105263157895, "percentage": 93.42, "elapsed_time": "0:43:32", "remaining_time": "0:03:03"}
+{"current_steps": 214, "total_steps": 228, "loss": 0.9967, "lr": 9.869096186961025e-08, "epoch": 0.9385964912280702, "percentage": 93.86, "elapsed_time": "0:43:43", "remaining_time": "0:02:51"}
+{"current_steps": 215, "total_steps": 228, "loss": 0.9782, "lr": 8.513450158049109e-08, "epoch": 0.9429824561403509, "percentage": 94.3, "elapsed_time": "0:43:57", "remaining_time": "0:02:39"}
+{"current_steps": 216, "total_steps": 228, "loss": 0.9853, "lr": 7.257120198226219e-08, "epoch": 0.9473684210526315, "percentage": 94.74, "elapsed_time": "0:44:10", "remaining_time": "0:02:27"}
+{"current_steps": 217, "total_steps": 228, "loss": 0.8929, "lr": 6.100360177619946e-08, "epoch": 0.9517543859649122, "percentage": 95.18, "elapsed_time": "0:44:23", "remaining_time": "0:02:14"}
+{"current_steps": 218, "total_steps": 228, "loss": 0.9297, "lr": 5.0434038459801213e-08, "epoch": 0.956140350877193, "percentage": 95.61, "elapsed_time": "0:44:36", "remaining_time": "0:02:02"}
+{"current_steps": 219, "total_steps": 228, "loss": 0.9525, "lr": 4.086464785444777e-08, "epoch": 0.9605263157894737, "percentage": 96.05, "elapsed_time": "0:44:48", "remaining_time": "0:01:50"}
+{"current_steps": 220, "total_steps": 228, "loss": 0.9839, "lr": 3.229736367380498e-08, "epoch": 0.9649122807017544, "percentage": 96.49, "elapsed_time": "0:45:00", "remaining_time": "0:01:38"}
+{"current_steps": 221, "total_steps": 228, "loss": 0.9349, "lr": 2.4733917133077378e-08, "epoch": 0.9692982456140351, "percentage": 96.93, "elapsed_time": "0:45:13", "remaining_time": "0:01:25"}
+{"current_steps": 222, "total_steps": 228, "loss": 0.9716, "lr": 1.8175836599173545e-08, "epoch": 0.9736842105263158, "percentage": 97.37, "elapsed_time": "0:45:25", "remaining_time": "0:01:13"}
+{"current_steps": 223, "total_steps": 228, "loss": 0.9661, "lr": 1.2624447281867625e-08, "epoch": 0.9780701754385965, "percentage": 97.81, "elapsed_time": "0:45:36", "remaining_time": "0:01:01"}
+{"current_steps": 224, "total_steps": 228, "loss": 0.9614, "lr": 8.080870966008513e-09, "epoch": 0.9824561403508771, "percentage": 98.25, "elapsed_time": "0:45:48", "remaining_time": "0:00:49"}
+{"current_steps": 225, "total_steps": 228, "loss": 0.9655, "lr": 4.546025784837316e-09, "epoch": 0.9868421052631579, "percentage": 98.68, "elapsed_time": "0:46:02", "remaining_time": "0:00:36"}
+{"current_steps": 226, "total_steps": 228, "loss": 0.9867, "lr": 2.0206260344590724e-09, "epoch": 0.9912280701754386, "percentage": 99.12, "elapsed_time": "0:46:12", "remaining_time": "0:00:24"}
+{"current_steps": 227, "total_steps": 228, "loss": 0.9372, "lr": 5.051820295032262e-10, "epoch": 0.9956140350877193, "percentage": 99.56, "elapsed_time": "0:46:23", "remaining_time": "0:00:12"}
+{"current_steps": 228, "total_steps": 228, "loss": 0.9237, "lr": 0.0, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "0:46:37", "remaining_time": "0:00:00"}
+{"current_steps": 228, "total_steps": 228, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "0:47:29", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,1638 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 228,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0043859649122807015,
+      "grad_norm": 49.592716217041016,
+      "learning_rate": 1.4285714285714286e-06,
+      "loss": 1.2955,
+      "step": 1
+    },
+    {
+      "epoch": 0.008771929824561403,
+      "grad_norm": 52.511619567871094,
+      "learning_rate": 2.8571428571428573e-06,
+      "loss": 1.3459,
+      "step": 2
+    },
+    {
+      "epoch": 0.013157894736842105,
+      "grad_norm": 23.82323455810547,
+      "learning_rate": 4.2857142857142855e-06,
+      "loss": 1.2305,
+      "step": 3
+    },
+    {
+      "epoch": 0.017543859649122806,
+      "grad_norm": 11.829014778137207,
+      "learning_rate": 5.7142857142857145e-06,
+      "loss": 1.0947,
+      "step": 4
+    },
+    {
+      "epoch": 0.021929824561403508,
+      "grad_norm": 20.558698654174805,
+      "learning_rate": 7.1428571428571436e-06,
+      "loss": 1.0768,
+      "step": 5
+    },
+    {
+      "epoch": 0.02631578947368421,
+      "grad_norm": 14.469958305358887,
+      "learning_rate": 8.571428571428571e-06,
+      "loss": 1.0448,
+      "step": 6
+    },
+    {
+      "epoch": 0.03070175438596491,
+      "grad_norm": 7.883849620819092,
+      "learning_rate": 1e-05,
+      "loss": 1.0738,
+      "step": 7
+    },
+    {
+      "epoch": 0.03508771929824561,
+      "grad_norm": 7.827476978302002,
+      "learning_rate": 9.999494817970498e-06,
+      "loss": 1.0353,
+      "step": 8
+    },
+    {
+      "epoch": 0.039473684210526314,
+      "grad_norm": 8.651590347290039,
+      "learning_rate": 9.997979373965542e-06,
+      "loss": 0.9942,
+      "step": 9
+    },
+    {
+      "epoch": 0.043859649122807015,
+      "grad_norm": 6.32966423034668,
+      "learning_rate": 9.995453974215164e-06,
+      "loss": 1.0467,
+      "step": 10
+    },
+    {
+      "epoch": 0.04824561403508772,
+      "grad_norm": 5.549691677093506,
+      "learning_rate": 9.991919129033994e-06,
+      "loss": 0.9998,
+      "step": 11
+    },
+    {
+      "epoch": 0.05263157894736842,
+      "grad_norm": 5.89638614654541,
+      "learning_rate": 9.987375552718133e-06,
+      "loss": 0.9997,
+      "step": 12
+    },
+    {
+      "epoch": 0.05701754385964912,
+      "grad_norm": 6.868354320526123,
+      "learning_rate": 9.981824163400827e-06,
+      "loss": 1.0602,
+      "step": 13
+    },
+    {
+      "epoch": 0.06140350877192982,
+      "grad_norm": 5.4902496337890625,
+      "learning_rate": 9.975266082866923e-06,
+      "loss": 1.043,
+      "step": 14
+    },
+    {
+      "epoch": 0.06578947368421052,
+      "grad_norm": 5.782011032104492,
+      "learning_rate": 9.967702636326195e-06,
+      "loss": 1.0331,
+      "step": 15
+    },
+    {
+      "epoch": 0.07017543859649122,
+      "grad_norm": 6.002492427825928,
+      "learning_rate": 9.959135352145552e-06,
+      "loss": 1.0376,
+      "step": 16
+    },
+    {
+      "epoch": 0.07456140350877193,
+      "grad_norm": 5.092071056365967,
+      "learning_rate": 9.9495659615402e-06,
+      "loss": 1.0244,
+      "step": 17
+    },
+    {
+      "epoch": 0.07894736842105263,
+      "grad_norm": 5.242514133453369,
+      "learning_rate": 9.938996398223802e-06,
+      "loss": 0.9812,
+      "step": 18
+    },
+    {
+      "epoch": 0.08333333333333333,
+      "grad_norm": 5.504293918609619,
+      "learning_rate": 9.927428798017738e-06,
+      "loss": 1.0288,
+      "step": 19
+    },
+    {
+      "epoch": 0.08771929824561403,
+      "grad_norm": 5.18280029296875,
+      "learning_rate": 9.91486549841951e-06,
+      "loss": 1.0403,
+      "step": 20
+    },
+    {
+      "epoch": 0.09210526315789473,
+      "grad_norm": 5.409468173980713,
+      "learning_rate": 9.901309038130392e-06,
+      "loss": 1.046,
+      "step": 21
+    },
+    {
+      "epoch": 0.09649122807017543,
+      "grad_norm": 4.993797779083252,
+      "learning_rate": 9.886762156542428e-06,
+      "loss": 1.0609,
+      "step": 22
+    },
+    {
+      "epoch": 0.10087719298245613,
+      "grad_norm": 4.722639083862305,
+      "learning_rate": 9.871227793184893e-06,
+      "loss": 1.03,
+      "step": 23
+    },
+    {
+      "epoch": 0.10526315789473684,
+      "grad_norm": 5.58522367477417,
+      "learning_rate": 9.854709087130261e-06,
+      "loss": 1.0388,
+      "step": 24
+    },
+    {
+      "epoch": 0.10964912280701754,
+      "grad_norm": 5.170425891876221,
+      "learning_rate": 9.837209376359918e-06,
+      "loss": 1.0588,
+      "step": 25
+    },
+    {
+      "epoch": 0.11403508771929824,
+      "grad_norm": 5.433144569396973,
+      "learning_rate": 9.81873219708962e-06,
+      "loss": 1.0688,
+      "step": 26
+    },
+    {
+      "epoch": 0.11842105263157894,
+      "grad_norm": 4.71676778793335,
+      "learning_rate": 9.79928128305494e-06,
+      "loss": 1.0311,
+      "step": 27
+    },
+    {
+      "epoch": 0.12280701754385964,
+      "grad_norm": 4.437475681304932,
+      "learning_rate": 9.778860564756769e-06,
+      "loss": 1.0086,
+      "step": 28
+    },
+    {
+      "epoch": 0.12719298245614036,
+      "grad_norm": 5.034616947174072,
+      "learning_rate": 9.757474168667072e-06,
+      "loss": 1.0627,
+      "step": 29
+    },
+    {
+      "epoch": 0.13157894736842105,
+      "grad_norm": 4.68194055557251,
+      "learning_rate": 9.73512641639504e-06,
+      "loss": 1.0349,
+      "step": 30
+    },
+    {
+      "epoch": 0.13596491228070176,
+      "grad_norm": 4.9859700202941895,
+      "learning_rate": 9.711821823813812e-06,
+      "loss": 1.1047,
+      "step": 31
+    },
+    {
+      "epoch": 0.14035087719298245,
+      "grad_norm": 4.754051685333252,
+      "learning_rate": 9.68756510014794e-06,
+      "loss": 1.0646,
+      "step": 32
+    },
+    {
+      "epoch": 0.14473684210526316,
+      "grad_norm": 4.727588653564453,
+      "learning_rate": 9.66236114702178e-06,
+      "loss": 1.0211,
+      "step": 33
+    },
+    {
+      "epoch": 0.14912280701754385,
+      "grad_norm": 4.892414569854736,
+      "learning_rate": 9.636215057469009e-06,
+      "loss": 1.03,
+      "step": 34
+    },
+    {
+      "epoch": 0.15350877192982457,
+      "grad_norm": 4.797459602355957,
+      "learning_rate": 9.609132114903458e-06,
+      "loss": 1.0495,
+      "step": 35
+    },
+    {
+      "epoch": 0.15789473684210525,
+      "grad_norm": 4.844034194946289,
+      "learning_rate": 9.581117792051487e-06,
+      "loss": 1.0653,
+      "step": 36
+    },
+    {
+      "epoch": 0.16228070175438597,
+      "grad_norm": 4.3364057540893555,
+      "learning_rate": 9.552177749846083e-06,
+      "loss": 1.0085,
+      "step": 37
+    },
+    {
+      "epoch": 0.16666666666666666,
+      "grad_norm": 4.658485412597656,
+      "learning_rate": 9.522317836282949e-06,
+      "loss": 1.0548,
+      "step": 38
+    },
+    {
+      "epoch": 0.17105263157894737,
+      "grad_norm": 4.726075649261475,
+      "learning_rate": 9.491544085238778e-06,
+      "loss": 1.0704,
+      "step": 39
+    },
+    {
+      "epoch": 0.17543859649122806,
+      "grad_norm": 4.422685623168945,
+      "learning_rate": 9.459862715251973e-06,
+      "loss": 1.0463,
+      "step": 40
+    },
+    {
+      "epoch": 0.17982456140350878,
+      "grad_norm": 4.876737117767334,
+      "learning_rate": 9.427280128266049e-06,
+      "loss": 1.0771,
+      "step": 41
+    },
+    {
+      "epoch": 0.18421052631578946,
+      "grad_norm": 4.657575607299805,
+      "learning_rate": 9.393802908335978e-06,
+      "loss": 1.0615,
+      "step": 42
+    },
+    {
+      "epoch": 0.18859649122807018,
+      "grad_norm": 4.246818542480469,
+      "learning_rate": 9.359437820297716e-06,
+      "loss": 1.0431,
+      "step": 43
+    },
+    {
+      "epoch": 0.19298245614035087,
+      "grad_norm": 4.417186737060547,
+      "learning_rate": 9.324191808401235e-06,
+      "loss": 1.0271,
+      "step": 44
+    },
+    {
+      "epoch": 0.19736842105263158,
+      "grad_norm": 4.741846084594727,
+      "learning_rate": 9.288071994907262e-06,
+      "loss": 1.0765,
+      "step": 45
+    },
+    {
+      "epoch": 0.20175438596491227,
+      "grad_norm": 5.002007007598877,
+      "learning_rate": 9.251085678648072e-06,
+      "loss": 1.0526,
+      "step": 46
+    },
+    {
+      "epoch": 0.20614035087719298,
+      "grad_norm": 4.708237171173096,
+      "learning_rate": 9.213240333552589e-06,
+      "loss": 1.0502,
+      "step": 47
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 4.335841655731201,
+      "learning_rate": 9.174543607136111e-06,
+      "loss": 0.9878,
+      "step": 48
+    },
+    {
+      "epoch": 0.2149122807017544,
+      "grad_norm": 4.573909282684326,
+      "learning_rate": 9.135003318954954e-06,
+      "loss": 0.9989,
+      "step": 49
+    },
+    {
+      "epoch": 0.21929824561403508,
+      "grad_norm": 4.54152774810791,
+      "learning_rate": 9.094627459026326e-06,
+      "loss": 1.0816,
+      "step": 50
+    },
+    {
+      "epoch": 0.2236842105263158,
+      "grad_norm": 4.57687520980835,
+      "learning_rate": 9.053424186213776e-06,
+      "loss": 1.0509,
+      "step": 51
+    },
+    {
+      "epoch": 0.22807017543859648,
+      "grad_norm": 4.246875762939453,
+      "learning_rate": 9.011401826578492e-06,
+      "loss": 0.9849,
+      "step": 52
+    },
+    {
+      "epoch": 0.2324561403508772,
+      "grad_norm": 4.673925876617432,
+      "learning_rate": 8.968568871696847e-06,
+      "loss": 0.9975,
+      "step": 53
+    },
+    {
+      "epoch": 0.23684210526315788,
+      "grad_norm": 4.580467224121094,
+      "learning_rate": 8.924933976944474e-06,
+      "loss": 1.0747,
+      "step": 54
+    },
+    {
+      "epoch": 0.2412280701754386,
+      "grad_norm": 4.450971603393555,
+      "learning_rate": 8.880505959747245e-06,
+      "loss": 1.028,
+      "step": 55
+    },
+    {
+      "epoch": 0.24561403508771928,
+      "grad_norm": 4.807090759277344,
+      "learning_rate": 8.835293797799517e-06,
+      "loss": 0.9659,
+      "step": 56
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 4.357285022735596,
+      "learning_rate": 8.789306627249985e-06,
+      "loss": 1.0228,
+      "step": 57
+    },
+    {
+      "epoch": 0.2543859649122807,
+      "grad_norm": 4.60875940322876,
+      "learning_rate": 8.742553740855507e-06,
+      "loss": 1.0338,
+      "step": 58
+    },
+    {
+      "epoch": 0.25877192982456143,
+      "grad_norm": 4.6360321044921875,
+      "learning_rate": 8.695044586103297e-06,
+      "loss": 1.0344,
+      "step": 59
+    },
+    {
+      "epoch": 0.2631578947368421,
+      "grad_norm": 4.603052139282227,
+      "learning_rate": 8.646788763301842e-06,
+      "loss": 1.0423,
+      "step": 60
+    },
+    {
+      "epoch": 0.2675438596491228,
+      "grad_norm": 4.477334976196289,
+      "learning_rate": 8.59779602364094e-06,
+      "loss": 1.0372,
+      "step": 61
+    },
+    {
+      "epoch": 0.2719298245614035,
+      "grad_norm": 4.819328784942627,
+      "learning_rate": 8.548076267221258e-06,
+      "loss": 1.0935,
+      "step": 62
+    },
+    {
+      "epoch": 0.27631578947368424,
+      "grad_norm": 4.351193428039551,
+      "learning_rate": 8.497639541053769e-06,
+      "loss": 0.9915,
+      "step": 63
+    },
+    {
+      "epoch": 0.2807017543859649,
+      "grad_norm": 9.282684326171875,
+      "learning_rate": 8.446496037029555e-06,
+      "loss": 1.0252,
+      "step": 64
+    },
+    {
+      "epoch": 0.2850877192982456,
+      "grad_norm": 4.520321369171143,
+      "learning_rate": 8.394656089860274e-06,
+      "loss": 0.9989,
+      "step": 65
+    },
+    {
+      "epoch": 0.2894736842105263,
+      "grad_norm": 4.605741500854492,
+      "learning_rate": 8.342130174989819e-06,
+      "loss": 1.0872,
+      "step": 66
+    },
+    {
+      "epoch": 0.29385964912280704,
+      "grad_norm": 4.512912750244141,
+      "learning_rate": 8.288928906477497e-06,
+      "loss": 1.0545,
+      "step": 67
+    },
+    {
+      "epoch": 0.2982456140350877,
+      "grad_norm": 4.024132251739502,
+      "learning_rate": 8.235063034853228e-06,
+      "loss": 1.0128,
+      "step": 68
+    },
+    {
+      "epoch": 0.3026315789473684,
+      "grad_norm": 4.560809135437012,
+      "learning_rate": 8.180543444945154e-06,
+      "loss": 1.0294,
+      "step": 69
+    },
+    {
+      "epoch": 0.30701754385964913,
+      "grad_norm": 4.295252799987793,
+      "learning_rate": 8.125381153680103e-06,
+      "loss": 1.0445,
+      "step": 70
+    },
+    {
+      "epoch": 0.31140350877192985,
+      "grad_norm": 4.614228248596191,
+      "learning_rate": 8.069587307857377e-06,
+      "loss": 1.0168,
+      "step": 71
+    },
+    {
+      "epoch": 0.3157894736842105,
+      "grad_norm": 4.256739139556885,
+      "learning_rate": 8.013173181896283e-06,
+      "loss": 1.0207,
+      "step": 72
+    },
+    {
+      "epoch": 0.3201754385964912,
+      "grad_norm": 4.458586692810059,
+      "learning_rate": 7.95615017555788e-06,
+      "loss": 1.0401,
+      "step": 73
+    },
+    {
+      "epoch": 0.32456140350877194,
+      "grad_norm": 4.6360392570495605,
+      "learning_rate": 7.898529811641393e-06,
+      "loss": 1.0347,
+      "step": 74
+    },
+    {
+      "epoch": 0.32894736842105265,
+      "grad_norm": 4.286371231079102,
+      "learning_rate": 7.84032373365578e-06,
+      "loss": 1.061,
+      "step": 75
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 4.306332588195801,
+      "learning_rate": 7.781543703466881e-06,
+      "loss": 0.988,
+      "step": 76
+    },
+    {
+      "epoch": 0.33771929824561403,
+      "grad_norm": 4.2425947189331055,
+      "learning_rate": 7.722201598920673e-06,
+      "loss": 1.0442,
+      "step": 77
+    },
+    {
+      "epoch": 0.34210526315789475,
+      "grad_norm": 4.27526330947876,
+      "learning_rate": 7.662309411443084e-06,
+      "loss": 1.0412,
+      "step": 78
+    },
+    {
+      "epoch": 0.34649122807017546,
+      "grad_norm": 4.224104881286621,
+      "learning_rate": 7.601879243616838e-06,
+      "loss": 1.0205,
+      "step": 79
+    },
+    {
+      "epoch": 0.3508771929824561,
+      "grad_norm": 4.250307559967041,
+      "learning_rate": 7.540923306735868e-06,
+      "loss": 1.0298,
+      "step": 80
+    },
+    {
+      "epoch": 0.35526315789473684,
+      "grad_norm": 4.223860263824463,
+      "learning_rate": 7.479453918337733e-06,
+      "loss": 1.0174,
+      "step": 81
+    },
+    {
+      "epoch": 0.35964912280701755,
+      "grad_norm": 4.442715644836426,
+      "learning_rate": 7.417483499714589e-06,
+      "loss": 1.0281,
+      "step": 82
+    },
+    {
+      "epoch": 0.36403508771929827,
+      "grad_norm": 4.697696685791016,
+      "learning_rate": 7.355024573403174e-06,
+      "loss": 1.0045,
+      "step": 83
+    },
+    {
+      "epoch": 0.3684210526315789,
+      "grad_norm": 4.343409061431885,
+      "learning_rate": 7.292089760654352e-06,
+      "loss": 1.0253,
+      "step": 84
+    },
+    {
+      "epoch": 0.37280701754385964,
+      "grad_norm": 4.392301082611084,
+      "learning_rate": 7.2286917788826926e-06,
+      "loss": 1.0245,
+      "step": 85
+    },
+    {
+      "epoch": 0.37719298245614036,
+      "grad_norm": 5.410698890686035,
+      "learning_rate": 7.1648434390966356e-06,
+      "loss": 1.0149,
+      "step": 86
+    },
+    {
+      "epoch": 0.3815789473684211,
+      "grad_norm": 4.763819694519043,
+      "learning_rate": 7.100557643309732e-06,
+      "loss": 1.0019,
+      "step": 87
+    },
+    {
+      "epoch": 0.38596491228070173,
+      "grad_norm": 4.71998405456543,
+      "learning_rate": 7.035847381933494e-06,
+      "loss": 1.0147,
+      "step": 88
+    },
+    {
+      "epoch": 0.39035087719298245,
+      "grad_norm": 4.432498931884766,
+      "learning_rate": 6.970725731152389e-06,
+      "loss": 0.9572,
+      "step": 89
+    },
+    {
+      "epoch": 0.39473684210526316,
+      "grad_norm": 4.130496978759766,
+      "learning_rate": 6.905205850281502e-06,
+      "loss": 0.9651,
+      "step": 90
+    },
+    {
+      "epoch": 0.3991228070175439,
+      "grad_norm": 4.6582441329956055,
+      "learning_rate": 6.8393009791073895e-06,
+      "loss": 1.0471,
+      "step": 91
+    },
+    {
+      "epoch": 0.40350877192982454,
+      "grad_norm": 4.263586521148682,
+      "learning_rate": 6.773024435212678e-06,
+      "loss": 1.0062,
+      "step": 92
+    },
+    {
+      "epoch": 0.40789473684210525,
+      "grad_norm": 4.377040386199951,
+      "learning_rate": 6.706389611284953e-06,
+      "loss": 1.0101,
+      "step": 93
+    },
+    {
+      "epoch": 0.41228070175438597,
+      "grad_norm": 4.429991722106934,
+      "learning_rate": 6.639409972410446e-06,
+      "loss": 1.0208,
+      "step": 94
+    },
+    {
+      "epoch": 0.4166666666666667,
+      "grad_norm": 4.125415325164795,
+      "learning_rate": 6.57209905335312e-06,
+      "loss": 1.0014,
+      "step": 95
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 4.075379848480225,
+      "learning_rate": 6.504470455819651e-06,
+      "loss": 0.9726,
+      "step": 96
+    },
+    {
+      "epoch": 0.42543859649122806,
+      "grad_norm": 4.166284084320068,
+      "learning_rate": 6.436537845710904e-06,
+      "loss": 1.0462,
+      "step": 97
+    },
+    {
+      "epoch": 0.4298245614035088,
+      "grad_norm": 3.857902765274048,
+      "learning_rate": 6.368314950360416e-06,
+      "loss": 0.997,
+      "step": 98
+    },
+    {
+      "epoch": 0.4342105263157895,
+      "grad_norm": 4.08658504486084,
+      "learning_rate": 6.299815555760478e-06,
+      "loss": 1.0351,
+      "step": 99
+    },
+    {
+      "epoch": 0.43859649122807015,
+      "grad_norm": 4.2274651527404785,
+      "learning_rate": 6.231053503776363e-06,
+      "loss": 1.04,
+      "step": 100
+    },
+    {
+      "epoch": 0.44298245614035087,
+      "grad_norm": 4.353753089904785,
+      "learning_rate": 6.1620426893492645e-06,
+      "loss": 0.9802,
+      "step": 101
+    },
+    {
+      "epoch": 0.4473684210526316,
+      "grad_norm": 4.267554759979248,
+      "learning_rate": 6.092797057688496e-06,
+      "loss": 1.006,
+      "step": 102
+    },
+    {
+      "epoch": 0.4517543859649123,
+      "grad_norm": 4.590907573699951,
+      "learning_rate": 6.0233306014535505e-06,
+      "loss": 0.9833,
+      "step": 103
+    },
+    {
+      "epoch": 0.45614035087719296,
+      "grad_norm": 4.248331546783447,
+      "learning_rate": 5.953657357926569e-06,
+      "loss": 1.0409,
+      "step": 104
+    },
+    {
+      "epoch": 0.4605263157894737,
+      "grad_norm": 3.9719252586364746,
+      "learning_rate": 5.883791406175775e-06,
+      "loss": 1.0191,
+      "step": 105
+    },
+    {
+      "epoch": 0.4649122807017544,
+      "grad_norm": 4.340267181396484,
+      "learning_rate": 5.813746864210489e-06,
+      "loss": 0.9992,
+      "step": 106
+    },
+    {
+      "epoch": 0.4692982456140351,
+      "grad_norm": 4.42672872543335,
+      "learning_rate": 5.743537886128258e-06,
+      "loss": 1.0023,
+      "step": 107
+    },
+    {
+      "epoch": 0.47368421052631576,
+      "grad_norm": 4.119699478149414,
+      "learning_rate": 5.673178659254698e-06,
+      "loss": 0.9909,
+      "step": 108
+    },
+    {
+      "epoch": 0.4780701754385965,
+      "grad_norm": 4.342530250549316,
+      "learning_rate": 5.6026834012766155e-06,
+      "loss": 0.9598,
+      "step": 109
+    },
+    {
+      "epoch": 0.4824561403508772,
+      "grad_norm": 4.101733684539795,
+      "learning_rate": 5.532066357369012e-06,
+      "loss": 0.9879,
+      "step": 110
+    },
+    {
+      "epoch": 0.4868421052631579,
+      "grad_norm": 4.340119361877441,
+      "learning_rate": 5.46134179731651e-06,
+      "loss": 1.0184,
+      "step": 111
+    },
+    {
+      "epoch": 0.49122807017543857,
+      "grad_norm": 4.37206506729126,
+      "learning_rate": 5.390524012629824e-06,
+      "loss": 1.0496,
+      "step": 112
+    },
+    {
+      "epoch": 0.4956140350877193,
+      "grad_norm": 4.118902683258057,
+      "learning_rate": 5.319627313657829e-06,
+      "loss": 1.0399,
+      "step": 113
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 4.071263790130615,
+      "learning_rate": 5.248666026695835e-06,
+      "loss": 0.9906,
+      "step": 114
+    },
+    {
+      "epoch": 0.5043859649122807,
+      "grad_norm": 4.421329975128174,
+      "learning_rate": 5.177654491090627e-06,
+      "loss": 0.9836,
+      "step": 115
+    },
+    {
+      "epoch": 0.5087719298245614,
+      "grad_norm": 4.517954349517822,
+      "learning_rate": 5.1066070563428736e-06,
+      "loss": 1.0083,
+      "step": 116
+    },
+    {
+      "epoch": 0.5131578947368421,
+      "grad_norm": 4.462676525115967,
+      "learning_rate": 5.035538079207488e-06,
+      "loss": 1.0034,
+      "step": 117
+    },
+    {
+      "epoch": 0.5175438596491229,
+      "grad_norm": 4.097958087921143,
+      "learning_rate": 4.964461920792512e-06,
+      "loss": 1.0244,
+      "step": 118
+    },
+    {
+      "epoch": 0.5219298245614035,
+      "grad_norm": 4.527233600616455,
+      "learning_rate": 4.893392943657127e-06,
+      "loss": 0.9473,
+      "step": 119
+    },
+    {
+      "epoch": 0.5263157894736842,
+      "grad_norm": 4.666014671325684,
+      "learning_rate": 4.822345508909376e-06,
+      "loss": 0.93,
+      "step": 120
+    },
+    {
+      "epoch": 0.5307017543859649,
+      "grad_norm": 4.115242958068848,
+      "learning_rate": 4.751333973304166e-06,
+      "loss": 1.0262,
+      "step": 121
+    },
+    {
+      "epoch": 0.5350877192982456,
+      "grad_norm": 4.032143592834473,
+      "learning_rate": 4.680372686342173e-06,
+      "loss": 1.0236,
+      "step": 122
+    },
+    {
+      "epoch": 0.5394736842105263,
+      "grad_norm": 4.429379463195801,
+      "learning_rate": 4.609475987370177e-06,
+      "loss": 0.9891,
+      "step": 123
+    },
+    {
+      "epoch": 0.543859649122807,
+      "grad_norm": 4.283740043640137,
+      "learning_rate": 4.53865820268349e-06,
+      "loss": 0.9761,
+      "step": 124
+    },
+    {
+      "epoch": 0.5482456140350878,
+      "grad_norm": 4.0465545654296875,
+      "learning_rate": 4.467933642630989e-06,
+      "loss": 0.9847,
+      "step": 125
+    },
+    {
+      "epoch": 0.5526315789473685,
+      "grad_norm": 4.005406856536865,
+      "learning_rate": 4.397316598723385e-06,
+      "loss": 0.9916,
+      "step": 126
+    },
+    {
+      "epoch": 0.5570175438596491,
+      "grad_norm": 4.05433988571167,
+      "learning_rate": 4.326821340745304e-06,
+      "loss": 0.9535,
+      "step": 127
+    },
+    {
+      "epoch": 0.5614035087719298,
+      "grad_norm": 4.201546669006348,
+      "learning_rate": 4.256462113871741e-06,
+      "loss": 0.9249,
+      "step": 128
+    },
+    {
+      "epoch": 0.5657894736842105,
+      "grad_norm": 4.1509904861450195,
+      "learning_rate": 4.186253135789511e-06,
+      "loss": 0.9885,
+      "step": 129
+    },
+    {
+      "epoch": 0.5701754385964912,
+      "grad_norm": 4.355051040649414,
+      "learning_rate": 4.116208593824227e-06,
+      "loss": 1.0087,
+      "step": 130
+    },
+    {
+      "epoch": 0.5745614035087719,
+      "grad_norm": 4.163605690002441,
+      "learning_rate": 4.046342642073433e-06,
+      "loss": 0.9654,
+      "step": 131
+    },
+    {
+      "epoch": 0.5789473684210527,
+      "grad_norm": 4.231169700622559,
+      "learning_rate": 3.976669398546451e-06,
+      "loss": 0.991,
+      "step": 132
+    },
+    {
+      "epoch": 0.5833333333333334,
+      "grad_norm": 4.092093467712402,
+      "learning_rate": 3.907202942311506e-06,
+      "loss": 0.9702,
+      "step": 133
+    },
+    {
+      "epoch": 0.5877192982456141,
+      "grad_norm": 4.0646233558654785,
+      "learning_rate": 3.837957310650738e-06,
+      "loss": 0.9451,
+      "step": 134
+    },
+    {
+      "epoch": 0.5921052631578947,
+      "grad_norm": 4.397289276123047,
+      "learning_rate": 3.7689464962236367e-06,
+      "loss": 0.9498,
+      "step": 135
+    },
+    {
+      "epoch": 0.5964912280701754,
+      "grad_norm": 3.946786403656006,
+      "learning_rate": 3.700184444239524e-06,
+      "loss": 0.9828,
+      "step": 136
+    },
+    {
+      "epoch": 0.6008771929824561,
+      "grad_norm": 4.302344799041748,
+      "learning_rate": 3.6316850496395863e-06,
+      "loss": 0.956,
+      "step": 137
+    },
+    {
+      "epoch": 0.6052631578947368,
+      "grad_norm": 4.430941581726074,
+      "learning_rate": 3.563462154289098e-06,
+      "loss": 0.9553,
+      "step": 138
+    },
+    {
+      "epoch": 0.6096491228070176,
+      "grad_norm": 4.306185722351074,
+      "learning_rate": 3.49552954418035e-06,
+      "loss": 1.0269,
+      "step": 139
+    },
+    {
+      "epoch": 0.6140350877192983,
+      "grad_norm": 4.095279216766357,
+      "learning_rate": 3.4279009466468825e-06,
+      "loss": 0.9888,
+      "step": 140
+    },
+    {
+      "epoch": 0.618421052631579,
+      "grad_norm": 4.261977195739746,
+      "learning_rate": 3.3605900275895565e-06,
+      "loss": 1.0003,
+      "step": 141
+    },
+    {
+      "epoch": 0.6228070175438597,
+      "grad_norm": 3.9479899406433105,
+      "learning_rate": 3.2936103887150484e-06,
+      "loss": 0.9982,
+      "step": 142
+    },
+    {
+      "epoch": 0.6271929824561403,
+      "grad_norm": 4.058523654937744,
+      "learning_rate": 3.226975564787322e-06,
+      "loss": 0.9645,
+      "step": 143
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 4.375962734222412,
+      "learning_rate": 3.1606990208926125e-06,
+      "loss": 0.9443,
+      "step": 144
+    },
+    {
+      "epoch": 0.6359649122807017,
+      "grad_norm": 4.618311405181885,
+      "learning_rate": 3.0947941497184985e-06,
+      "loss": 1.0069,
+      "step": 145
+    },
+    {
+      "epoch": 0.6403508771929824,
+      "grad_norm": 4.422805309295654,
+      "learning_rate": 3.0292742688476125e-06,
+      "loss": 0.9718,
+      "step": 146
+    },
+    {
+      "epoch": 0.6447368421052632,
+      "grad_norm": 3.93886399269104,
+      "learning_rate": 2.964152618066508e-06,
+      "loss": 0.9981,
+      "step": 147
+    },
+    {
+      "epoch": 0.6491228070175439,
+      "grad_norm": 3.937767744064331,
+      "learning_rate": 2.899442356690271e-06,
+      "loss": 0.994,
+      "step": 148
+    },
+    {
+      "epoch": 0.6535087719298246,
+      "grad_norm": 4.434424877166748,
+      "learning_rate": 2.835156560903365e-06,
+      "loss": 0.9605,
+      "step": 149
+    },
+    {
+      "epoch": 0.6578947368421053,
+      "grad_norm": 4.723587989807129,
+      "learning_rate": 2.771308221117309e-06,
+      "loss": 0.9965,
+      "step": 150
+    },
+    {
+      "epoch": 0.6622807017543859,
+      "grad_norm": 4.1226301193237305,
+      "learning_rate": 2.7079102393456503e-06,
+      "loss": 1.0161,
+      "step": 151
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 3.8456857204437256,
+      "learning_rate": 2.6449754265968263e-06,
+      "loss": 0.9696,
+      "step": 152
+    },
+    {
+      "epoch": 0.6710526315789473,
+      "grad_norm": 3.9130499362945557,
+      "learning_rate": 2.5825165002854124e-06,
+      "loss": 0.9674,
+      "step": 153
+    },
+    {
+      "epoch": 0.6754385964912281,
+      "grad_norm": 4.2803473472595215,
+      "learning_rate": 2.5205460816622684e-06,
+      "loss": 0.9933,
+      "step": 154
+    },
+    {
+      "epoch": 0.6798245614035088,
+      "grad_norm": 3.8558075428009033,
+      "learning_rate": 2.4590766932641353e-06,
+      "loss": 0.977,
+      "step": 155
+    },
+    {
+      "epoch": 0.6842105263157895,
+      "grad_norm": 4.1506452560424805,
+      "learning_rate": 2.3981207563831633e-06,
+      "loss": 0.9729,
+      "step": 156
+    },
+    {
+      "epoch": 0.6885964912280702,
+      "grad_norm": 4.3258843421936035,
+      "learning_rate": 2.3376905885569185e-06,
+      "loss": 0.914,
+      "step": 157
+    },
+    {
+      "epoch": 0.6929824561403509,
+      "grad_norm": 4.091325759887695,
+      "learning_rate": 2.2777984010793264e-06,
+      "loss": 0.9641,
+      "step": 158
+    },
+    {
+      "epoch": 0.6973684210526315,
+      "grad_norm": 4.333226203918457,
+      "learning_rate": 2.2184562965331203e-06,
+      "loss": 0.9074,
+      "step": 159
+    },
+    {
+      "epoch": 0.7017543859649122,
+      "grad_norm": 3.8289241790771484,
+      "learning_rate": 2.159676266344222e-06,
+      "loss": 0.938,
+      "step": 160
+    },
+    {
+      "epoch": 0.706140350877193,
+      "grad_norm": 4.058959007263184,
+      "learning_rate": 2.1014701883586087e-06,
+      "loss": 0.9387,
+      "step": 161
+    },
+    {
+      "epoch": 0.7105263157894737,
+      "grad_norm": 3.7621591091156006,
+      "learning_rate": 2.043849824442124e-06,
+      "loss": 0.9931,
+      "step": 162
+    },
+    {
+      "epoch": 0.7149122807017544,
+      "grad_norm": 4.214111804962158,
+      "learning_rate": 1.9868268181037186e-06,
+      "loss": 0.9578,
+      "step": 163
+    },
+    {
+      "epoch": 0.7192982456140351,
+      "grad_norm": 4.251298427581787,
+      "learning_rate": 1.9304126921426235e-06,
+      "loss": 0.9615,
+      "step": 164
+    },
+    {
+      "epoch": 0.7236842105263158,
+      "grad_norm": 3.980278730392456,
+      "learning_rate": 1.8746188463198983e-06,
+      "loss": 0.952,
+      "step": 165
+    },
+    {
+      "epoch": 0.7280701754385965,
+      "grad_norm": 4.493079662322998,
+      "learning_rate": 1.8194565550548477e-06,
+      "loss": 0.9422,
+      "step": 166
+    },
+    {
+      "epoch": 0.7324561403508771,
+      "grad_norm": 4.167688846588135,
+      "learning_rate": 1.764936965146773e-06,
+      "loss": 0.9556,
+      "step": 167
+    },
+    {
+      "epoch": 0.7368421052631579,
+      "grad_norm": 3.890090227127075,
+      "learning_rate": 1.7110710935225055e-06,
+      "loss": 0.9609,
+      "step": 168
+    },
+    {
+      "epoch": 0.7412280701754386,
+      "grad_norm": 4.315472602844238,
+      "learning_rate": 1.6578698250101828e-06,
+      "loss": 0.9945,
+      "step": 169
+    },
+    {
+      "epoch": 0.7456140350877193,
+      "grad_norm": 4.024162769317627,
+      "learning_rate": 1.6053439101397257e-06,
+      "loss": 0.9526,
+      "step": 170
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 4.042996883392334,
+      "learning_rate": 1.5535039629704467e-06,
+      "loss": 0.953,
+      "step": 171
+    },
+    {
+      "epoch": 0.7543859649122807,
+      "grad_norm": 4.670426845550537,
+      "learning_rate": 1.502360458946232e-06,
+      "loss": 0.9418,
+      "step": 172
+    },
+    {
+      "epoch": 0.7587719298245614,
+      "grad_norm": 3.7296359539031982,
+      "learning_rate": 1.451923732778745e-06,
+      "loss": 0.9842,
+      "step": 173
+    },
+    {
+      "epoch": 0.7631578947368421,
+      "grad_norm": 4.069624423980713,
+      "learning_rate": 1.4022039763590595e-06,
+      "loss": 0.9483,
+      "step": 174
+    },
+    {
+      "epoch": 0.7675438596491229,
+      "grad_norm": 3.8240842819213867,
+      "learning_rate": 1.3532112366981598e-06,
+      "loss": 1.0225,
+      "step": 175
+    },
+    {
+      "epoch": 0.7719298245614035,
+      "grad_norm": 4.527555465698242,
+      "learning_rate": 1.3049554138967052e-06,
+      "loss": 0.9787,
+      "step": 176
+    },
+    {
+      "epoch": 0.7763157894736842,
+      "grad_norm": 3.748631238937378,
+      "learning_rate": 1.257446259144494e-06,
+      "loss": 0.9812,
+      "step": 177
+    },
+    {
+      "epoch": 0.7807017543859649,
+      "grad_norm": 3.8050689697265625,
+      "learning_rate": 1.210693372750017e-06,
+      "loss": 0.9686,
+      "step": 178
+    },
+    {
+      "epoch": 0.7850877192982456,
+      "grad_norm": 3.9663491249084473,
+      "learning_rate": 1.1647062022004845e-06,
+      "loss": 0.9909,
+      "step": 179
+    },
+    {
+      "epoch": 0.7894736842105263,
+      "grad_norm": 3.8755075931549072,
+      "learning_rate": 1.1194940402527566e-06,
+      "loss": 0.9682,
+      "step": 180
+    },
+    {
+      "epoch": 0.793859649122807,
+      "grad_norm": 3.51550030708313,
+      "learning_rate": 1.075066023055527e-06,
+      "loss": 0.9218,
+      "step": 181
+    },
+    {
+      "epoch": 0.7982456140350878,
+      "grad_norm": 3.9786829948425293,
+      "learning_rate": 1.0314311283031531e-06,
+      "loss": 0.9753,
+      "step": 182
+    },
+    {
+      "epoch": 0.8026315789473685,
+      "grad_norm": 4.108859539031982,
+      "learning_rate": 9.885981734215094e-07,
+      "loss": 0.9746,
+      "step": 183
+    },
+    {
+      "epoch": 0.8070175438596491,
+      "grad_norm": 4.208093166351318,
+      "learning_rate": 9.465758137862264e-07,
+      "loss": 0.9479,
+      "step": 184
+    },
+    {
+      "epoch": 0.8114035087719298,
+      "grad_norm": 4.333961009979248,
+      "learning_rate": 9.053725409736752e-07,
+      "loss": 0.9734,
+      "step": 185
+    },
+    {
+      "epoch": 0.8157894736842105,
+      "grad_norm": 3.7324233055114746,
+      "learning_rate": 8.649966810450472e-07,
+      "loss": 0.9145,
+      "step": 186
+    },
+    {
+      "epoch": 0.8201754385964912,
+      "grad_norm": 3.5640878677368164,
+      "learning_rate": 8.254563928638892e-07,
+      "loss": 0.9302,
+      "step": 187
+    },
+    {
+      "epoch": 0.8245614035087719,
+      "grad_norm": 4.192229270935059,
+      "learning_rate": 7.86759666447412e-07,
+      "loss": 0.9509,
+      "step": 188
+    },
+    {
+      "epoch": 0.8289473684210527,
+      "grad_norm": 4.028424263000488,
+      "learning_rate": 7.489143213519301e-07,
+      "loss": 0.9081,
+      "step": 189
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 3.9033594131469727,
+      "learning_rate": 7.119280050927407e-07,
+      "loss": 0.9404,
+      "step": 190
+    },
+    {
+      "epoch": 0.8377192982456141,
+      "grad_norm": 4.0438361167907715,
+      "learning_rate": 6.758081915987669e-07,
+      "loss": 0.9482,
+      "step": 191
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 4.058148384094238,
+      "learning_rate": 6.405621797022848e-07,
+      "loss": 0.9161,
+      "step": 192
+    },
+    {
+      "epoch": 0.8464912280701754,
+      "grad_norm": 4.070837020874023,
+      "learning_rate": 6.061970916640236e-07,
+      "loss": 0.9351,
+      "step": 193
+    },
+    {
+      "epoch": 0.8508771929824561,
+      "grad_norm": 3.934864044189453,
+      "learning_rate": 5.727198717339511e-07,
+      "loss": 0.9633,
+      "step": 194
+    },
+    {
+      "epoch": 0.8552631578947368,
+      "grad_norm": 3.696377992630005,
+      "learning_rate": 5.401372847480285e-07,
+      "loss": 0.9923,
+      "step": 195
+    },
+    {
+      "epoch": 0.8596491228070176,
+      "grad_norm": 4.334061622619629,
+      "learning_rate": 5.084559147612244e-07,
+      "loss": 0.9049,
+      "step": 196
+    },
+    {
+      "epoch": 0.8640350877192983,
+      "grad_norm": 3.781756639480591,
+      "learning_rate": 4.776821637170525e-07,
+      "loss": 0.9344,
+      "step": 197
+    },
+    {
+      "epoch": 0.868421052631579,
+      "grad_norm": 3.749248504638672,
+      "learning_rate": 4.4782225015391754e-07,
+      "loss": 0.9594,
+      "step": 198
+    },
+    {
+      "epoch": 0.8728070175438597,
+      "grad_norm": 3.8009796142578125,
+      "learning_rate": 4.1888220794851386e-07,
+      "loss": 0.9519,
+      "step": 199
+    },
+    {
+      "epoch": 0.8771929824561403,
+      "grad_norm": 3.914064645767212,
+      "learning_rate": 3.908678850965425e-07,
+      "loss": 0.9539,
+      "step": 200
+    },
+    {
+      "epoch": 0.881578947368421,
+      "grad_norm": 3.75138521194458,
+      "learning_rate": 3.6378494253099307e-07,
+      "loss": 0.9506,
+      "step": 201
+    },
+    {
+      "epoch": 0.8859649122807017,
+      "grad_norm": 3.7791149616241455,
+      "learning_rate": 3.3763885297822153e-07,
+      "loss": 0.9309,
+      "step": 202
+    },
+    {
+      "epoch": 0.8903508771929824,
+      "grad_norm": 3.780644655227661,
+      "learning_rate": 3.1243489985206097e-07,
+      "loss": 0.9497,
+      "step": 203
+    },
+    {
+      "epoch": 0.8947368421052632,
+      "grad_norm": 4.0235676765441895,
+      "learning_rate": 2.8817817618618846e-07,
+      "loss": 0.9707,
+      "step": 204
+    },
+    {
+      "epoch": 0.8991228070175439,
+      "grad_norm": 3.802351951599121,
+      "learning_rate": 2.648735836049615e-07,
+      "loss": 0.953,
+      "step": 205
+    },
+    {
+      "epoch": 0.9035087719298246,
+      "grad_norm": 3.995856523513794,
+      "learning_rate": 2.4252583133292927e-07,
+      "loss": 0.9465,
+      "step": 206
+    },
+    {
+      "epoch": 0.9078947368421053,
+      "grad_norm": 4.330905914306641,
+      "learning_rate": 2.2113943524323167e-07,
+      "loss": 0.9527,
+      "step": 207
+    },
+    {
+      "epoch": 0.9122807017543859,
+      "grad_norm": 4.164973258972168,
+      "learning_rate": 2.007187169450603e-07,
+      "loss": 0.9498,
+      "step": 208
+    },
+    {
+      "epoch": 0.9166666666666666,
+      "grad_norm": 3.576552391052246,
+      "learning_rate": 1.8126780291038037e-07,
+      "loss": 0.9582,
+      "step": 209
+    },
+    {
+      "epoch": 0.9210526315789473,
+      "grad_norm": 3.625762939453125,
+      "learning_rate": 1.6279062364008446e-07,
+      "loss": 0.9169,
+      "step": 210
+    },
+    {
+      "epoch": 0.9254385964912281,
+      "grad_norm": 3.5316288471221924,
+      "learning_rate": 1.4529091286973994e-07,
+      "loss": 0.9789,
+      "step": 211
+    },
+    {
+      "epoch": 0.9298245614035088,
+      "grad_norm": 3.8600916862487793,
+      "learning_rate": 1.2877220681510927e-07,
+      "loss": 0.9705,
+      "step": 212
+    },
+    {
+      "epoch": 0.9342105263157895,
+      "grad_norm": 3.9807686805725098,
+      "learning_rate": 1.1323784345757205e-07,
+      "loss": 0.9249,
+      "step": 213
+    },
+    {
+      "epoch": 0.9385964912280702,
+      "grad_norm": 3.589660406112671,
+      "learning_rate": 9.869096186961025e-08,
+      "loss": 0.9967,
+      "step": 214
+    },
+    {
+      "epoch": 0.9429824561403509,
+      "grad_norm": 3.8679358959198,
+      "learning_rate": 8.513450158049109e-08,
+      "loss": 0.9782,
+      "step": 215
+    },
+    {
+      "epoch": 0.9473684210526315,
+      "grad_norm": 4.188008785247803,
+      "learning_rate": 7.257120198226219e-08,
+      "loss": 0.9853,
+      "step": 216
+    },
+    {
+      "epoch": 0.9517543859649122,
+      "grad_norm": 4.046226978302002,
+      "learning_rate": 6.100360177619946e-08,
+      "loss": 0.8929,
+      "step": 217
+    },
+    {
+      "epoch": 0.956140350877193,
+      "grad_norm": 7.8870110511779785,
+      "learning_rate": 5.0434038459801213e-08,
+      "loss": 0.9297,
+      "step": 218
+    },
+    {
+      "epoch": 0.9605263157894737,
+      "grad_norm": 3.9931716918945312,
+      "learning_rate": 4.086464785444777e-08,
+      "loss": 0.9525,
+      "step": 219
+    },
+    {
+      "epoch": 0.9649122807017544,
+      "grad_norm": 3.876042604446411,
+      "learning_rate": 3.229736367380498e-08,
+      "loss": 0.9839,
+      "step": 220
+    },
+    {
+      "epoch": 0.9692982456140351,
+      "grad_norm": 4.057157516479492,
+      "learning_rate": 2.4733917133077378e-08,
+      "loss": 0.9349,
+      "step": 221
+    },
+    {
+      "epoch": 0.9736842105263158,
+      "grad_norm": 3.735743522644043,
+      "learning_rate": 1.8175836599173545e-08,
+      "loss": 0.9716,
+      "step": 222
+    },
+    {
+      "epoch": 0.9780701754385965,
+      "grad_norm": 3.7397193908691406,
+      "learning_rate": 1.2624447281867625e-08,
+      "loss": 0.9661,
+      "step": 223
+    },
+    {
+      "epoch": 0.9824561403508771,
+      "grad_norm": 3.6132941246032715,
+      "learning_rate": 8.080870966008513e-09,
+      "loss": 0.9614,
+      "step": 224
+    },
+    {
+      "epoch": 0.9868421052631579,
+      "grad_norm": 4.1698784828186035,
+      "learning_rate": 4.546025784837316e-09,
+      "loss": 0.9655,
+      "step": 225
+    },
+    {
+      "epoch": 0.9912280701754386,
+      "grad_norm": 3.830890655517578,
+      "learning_rate": 2.0206260344590724e-09,
+      "loss": 0.9867,
+      "step": 226
+    },
+    {
+      "epoch": 0.9956140350877193,
+      "grad_norm": 3.9525303840637207,
+      "learning_rate": 5.051820295032262e-10,
+      "loss": 0.9372,
+      "step": 227
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 4.012087345123291,
+      "learning_rate": 0.0,
+      "loss": 0.9237,
+      "step": 228
+    },
+    {
+      "epoch": 1.0,
+      "step": 228,
+      "total_flos": 1.387246770039292e+18,
+      "train_loss": 1.0000714727661066,
+      "train_runtime": 2851.6353,
+      "train_samples_per_second": 20.461,
+      "train_steps_per_second": 0.08
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 228,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.387246770039292e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}