Training in progress, step 2000

adapter_config.json

@@ -14,13 +14,13 @@
   "r": 64,
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "o_proj",
     "k_proj",
+    "gate_proj",
     "up_proj",
     "down_proj",
     "v_proj",
-    "gate_proj"
+    "o_proj",
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fed8cf9fdfe60dcc647042b17e4ebd6e41d3a5eadcdacaf6a453fd896353160a
+oid sha256:f0d503b7d54953f192ce4ff9ad18dcf0998ec24528a28bec217d0e467556066e
 size 871609293

all_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 2.53,
+    "eval_loss": 0.5517624616622925,
+    "eval_runtime": 396.834,
+    "eval_samples_per_second": 2.52,
+    "eval_steps_per_second": 1.26,
+    "train_loss": 0.5183539204340842,
+    "train_runtime": 30592.645,
+    "train_samples_per_second": 1.308,
+    "train_steps_per_second": 0.163
+}

checkpoint-1200/adapter_model/adapter_model/README.md

@@ -26,6 +26,28 @@ The following `bitsandbytes` quantization config was used during training:
 - bnb_4bit_use_double_quant: True
 - bnb_4bit_compute_dtype: bfloat16
 
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: bfloat16
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: bfloat16
+
 The following `bitsandbytes` quantization config was used during training:
 - load_in_8bit: False
 - load_in_4bit: True
@@ -38,6 +60,8 @@ The following `bitsandbytes` quantization config was used during training:
 - bnb_4bit_compute_dtype: bfloat16
 ### Framework versions
 
+- PEFT 0.4.0
+- PEFT 0.4.0
 - PEFT 0.4.0
 - PEFT 0.4.0
 

checkpoint-1200/adapter_model/adapter_model/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c95e2606f0ea854dfd026727cf0a6e19ec69747d71b6c9ef24174461f88c2a50
+oid sha256:a81d8a71f1e9899f8fa76ffc1b78c7d6d5b2dff5171933aafd9e8d6fe27a9050
 size 871609293

checkpoint-2000/README.md

@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: bfloat16
+### Framework versions
+
+
+- PEFT 0.4.0

checkpoint-2000/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "alpindale/CodeLlama-34B-hf",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16.0,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "gate_proj",
+    "up_proj",
+    "down_proj",
+    "v_proj",
+    "o_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-2000/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0d503b7d54953f192ce4ff9ad18dcf0998ec24528a28bec217d0e467556066e
+size 871609293

checkpoint-2000/added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "[PAD]": 32000
+}

checkpoint-2000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6ecda594d48793e4abfdfb6a6cc7365a13442a5e76033859f3dd9f5e88202d5
+size 3485867229

checkpoint-2000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75a9a1f8204dc0a51013c680de5a90fa63e71ad88749bd702001ef34b8678d0e
+size 14511

checkpoint-2000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:720bafe9a34b3041027dccef7d9c03bb23064e1fcdd80ae5f27e717d41b7cf31
+size 627

checkpoint-2000/special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "[PAD]",
+  "unk_token": "<unk>"
+}

checkpoint-2000/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

checkpoint-2000/tokenizer_config.json

@@ -0,0 +1,37 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "bos_token": {
+    "__type": "AddedToken",
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "clean_up_tokenization_spaces": false,
+  "eos_token": {
+    "__type": "AddedToken",
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "legacy": null,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": null,
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": {
+    "__type": "AddedToken",
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "use_default_system_prompt": true
+}

checkpoint-2000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b62cb6f6463a7093dc0569db5eddc6507b6f1bfc6a2b3aecea1a80a972ec794
+size 5947

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.53,
+    "eval_loss": 0.5517624616622925,
+    "eval_runtime": 396.834,
+    "eval_samples_per_second": 2.52,
+    "eval_steps_per_second": 1.26
+}

metrics.json

@@ -0,0 +1 @@
+{"run_name": "codellama34b_unnatural", "train_runtime": 30592.645, "train_samples_per_second": 1.308, "train_steps_per_second": 0.163, "train_loss": 0.5183539204340842, "epoch": 2.53, "eval_loss": 0.5517624616622925, "eval_runtime": 396.834, "eval_samples_per_second": 2.52, "eval_steps_per_second": 1.26}

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.53,
+    "train_loss": 0.5183539204340842,
+    "train_runtime": 30592.645,
+    "train_samples_per_second": 1.308,
+    "train_steps_per_second": 0.163
+}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:85d09a9f54aa026a0295bc7d334655520e9bf512c90be29e98b4ed206613fba6
+oid sha256:4b62cb6f6463a7093dc0569db5eddc6507b6f1bfc6a2b3aecea1a80a972ec794
 size 5947