Upload folder using huggingface_hub

README.md

@@ -1,7 +1,7 @@
 ---
 library_name: transformers
-license: apache-2.0
-base_model: Qwen/Qwen2.5-1.5B-Instruct
+license: other
+base_model: Qwen/Qwen2.5-3B-Instruct
 tags:
 - generated_from_trainer
 datasets:
@@ -19,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 axolotl version: `0.6.0`
 ```yaml
-base_model: Qwen/Qwen2.5-1.5B-Instruct
+base_model: Qwen/Qwen2.5-3B-Instruct
 model_type: AutoModelForCausalLM
 tokenizer_type: AutoTokenizer
 trust_remote_code: false
@@ -29,10 +29,7 @@ load_in_4bit: false
 strict: false
 
 output_dir: ./outputs/out
-remove_unused_columns: false
-
 chat_template: qwen_25
-# chat_template: qwen_25
 datasets:
   - path: train.jsonl
     type: chat_template
@@ -40,17 +37,19 @@ datasets:
     message_field_role: role
     message_field_content: content
     roles:
+      system:
+        - system
       user:
         - user
       assistant:
         - assistant
 
-dataset_prepared_path: mr1-sft-1
-# dataset_prepared_path: ko_r1
+dataset_prepared_path: last_run_prepared
 val_set_size: 0.005
+output_dir: ./outputs/out
 eval_sample_packing: False
 
-sequence_len: 512
+sequence_len: 8192
 sample_packing: False
 pad_to_sequence_len: False
 
@@ -59,6 +58,7 @@ wandb_entity:
 wandb_watch:
 wandb_name:
 wandb_log_model:
+# hub_model_id: amphora/merged-bench-qwen-full
 
 plugins:
   - axolotl.integrations.liger.LigerPlugin
@@ -67,8 +67,8 @@ liger_rms_norm: true
 liger_swiglu: true
 liger_fused_linear_cross_entropy: true
 
-gradient_accumulation_steps: 1
-micro_batch_size: 128
+gradient_accumulation_steps: 4
+micro_batch_size: 8
 eval_batch_size: 4
 num_epochs: 3
 optimizer: paged_adamw_8bit
@@ -78,7 +78,7 @@ learning_rate: 2e-5
 train_on_inputs: false
 group_by_length: false
 bf16: auto
-fp16: 
+fp16:
 tf32: false
 
 gradient_checkpointing: true
@@ -90,7 +90,7 @@ logging_steps: 1
 xformers_attention:
 flash_attention: true
 
-warmup_steps: 10
+warmup_steps: 30
 evals_per_epoch: 3
 eval_max_new_tokens: 128
 eval_table_size:
@@ -101,16 +101,15 @@ weight_decay: 0.01
 fsdp:
 fsdp_config:
 special_tokens:
-  eos_token: 
 ```
 
 </details><br>
 
 # outputs/out
 
-This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the train.jsonl dataset.
+This model is a fine-tuned version of [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) on the train.jsonl dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3103
+- Loss: 0.2783
 
 ## Model description
 
@@ -130,31 +129,32 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 128
+- train_batch_size: 8
 - eval_batch_size: 4
 - seed: 42
 - distributed_type: multi-GPU
 - num_devices: 2
-- total_train_batch_size: 256
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 64
 - total_eval_batch_size: 8
 - optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: cosine
-- lr_scheduler_warmup_steps: 10
+- lr_scheduler_warmup_steps: 30
 - num_epochs: 3.0
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 4.6099        | 0.0079 | 1    | 3.1001          |
-| 0.0071        | 0.3386 | 43   | 0.3896          |
-| 0.0098        | 0.6772 | 86   | 0.3527          |
-| 0.0026        | 1.0157 | 129  | 0.3306          |
-| 0.0128        | 1.3543 | 172  | 0.3166          |
-| 0.0042        | 1.6929 | 215  | 0.3484          |
-| 0.0019        | 2.0315 | 258  | 0.2931          |
-| 0.0039        | 2.3701 | 301  | 0.3032          |
-| 0.0           | 2.7087 | 344  | 0.3103          |
+| 1.3989        | 0.0041 | 1    | 1.7111          |
+| 0.2969        | 0.3350 | 82   | 0.3192          |
+| 0.3027        | 0.6701 | 164  | 0.2914          |
+| 0.177         | 1.0082 | 246  | 0.2854          |
+| 0.1735        | 1.3432 | 328  | 0.2857          |
+| 0.1684        | 1.6782 | 410  | 0.2805          |
+| 0.1109        | 2.0163 | 492  | 0.2741          |
+| 0.0946        | 2.3514 | 574  | 0.2828          |
+| 0.0968        | 2.6864 | 656  | 0.2783          |
 
 
 ### Framework versions

checkpoint-244/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-244/config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "Qwen/Qwen2.5-3B-Instruct",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 70,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.1",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151665
+}

checkpoint-244/generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.48.1"
+}

checkpoint-244/latest

@@ -0,0 +1 @@
+global_step244

checkpoint-244/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b177b17305dfd0a160a85100c35a38e3ac87b207f8f13f906dd114b62534a2d
+size 4956450288

checkpoint-244/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb9ce7028d6aa2ffffc37212cb54fb4d943af4c4dad356c80f1621510f4f6a21
+size 1835586736

checkpoint-244/model.safetensors.index.json

@@ -0,0 +1,442 @@
+{
+  "metadata": {
+    "total_size": 6791987200
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

checkpoint-244/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9affc1541e7e94c18354d5173bc55400c5f07faf3d080c6d453d48e7a8d6ac3
+size 14512

checkpoint-244/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4748c3ebf0e4c051c58b92e4a8c5b87cdb39d55cfdc2aec81a1baef0f02fc113
+size 14512

checkpoint-244/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5cb186d02e42c19d881269361281b0d1dc724284e39baf6809ced6fd93070319
+size 1064

checkpoint-244/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-244/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

checkpoint-244/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-244/trainer_state.json

@@ -0,0 +1,1765 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9969356486210419,
+  "eval_steps": 82,
+  "global_step": 244,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0040858018386108275,
+      "grad_norm": 4.75867223739624,
+      "learning_rate": 6.666666666666667e-07,
+      "loss": 1.3989,
+      "step": 1
+    },
+    {
+      "epoch": 0.0040858018386108275,
+      "eval_loss": 1.7111468315124512,
+      "eval_runtime": 5.4436,
+      "eval_samples_per_second": 14.512,
+      "eval_steps_per_second": 1.837,
+      "step": 1
+    },
+    {
+      "epoch": 0.008171603677221655,
+      "grad_norm": 4.975377559661865,
+      "learning_rate": 1.3333333333333334e-06,
+      "loss": 1.4837,
+      "step": 2
+    },
+    {
+      "epoch": 0.012257405515832482,
+      "grad_norm": 5.219729900360107,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.5181,
+      "step": 3
+    },
+    {
+      "epoch": 0.01634320735444331,
+      "grad_norm": 4.57335901260376,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 1.4106,
+      "step": 4
+    },
+    {
+      "epoch": 0.020429009193054137,
+      "grad_norm": 3.840559720993042,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 1.3763,
+      "step": 5
+    },
+    {
+      "epoch": 0.024514811031664963,
+      "grad_norm": 3.2056212425231934,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.1876,
+      "step": 6
+    },
+    {
+      "epoch": 0.028600612870275793,
+      "grad_norm": 2.6987595558166504,
+      "learning_rate": 4.666666666666667e-06,
+      "loss": 1.2154,
+      "step": 7
+    },
+    {
+      "epoch": 0.03268641470888662,
+      "grad_norm": 2.378502130508423,
+      "learning_rate": 5.333333333333334e-06,
+      "loss": 1.1594,
+      "step": 8
+    },
+    {
+      "epoch": 0.03677221654749745,
+      "grad_norm": 1.7688865661621094,
+      "learning_rate": 6e-06,
+      "loss": 0.8435,
+      "step": 9
+    },
+    {
+      "epoch": 0.04085801838610827,
+      "grad_norm": 1.3263744115829468,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 0.7219,
+      "step": 10
+    },
+    {
+      "epoch": 0.0449438202247191,
+      "grad_norm": 1.3509997129440308,
+      "learning_rate": 7.333333333333333e-06,
+      "loss": 0.8172,
+      "step": 11
+    },
+    {
+      "epoch": 0.049029622063329927,
+      "grad_norm": 1.4541417360305786,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.7393,
+      "step": 12
+    },
+    {
+      "epoch": 0.05311542390194075,
+      "grad_norm": 1.181699275970459,
+      "learning_rate": 8.666666666666668e-06,
+      "loss": 0.664,
+      "step": 13
+    },
+    {
+      "epoch": 0.05720122574055159,
+      "grad_norm": 0.9503294825553894,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.6222,
+      "step": 14
+    },
+    {
+      "epoch": 0.06128702757916241,
+      "grad_norm": 0.7614471316337585,
+      "learning_rate": 1e-05,
+      "loss": 0.56,
+      "step": 15
+    },
+    {
+      "epoch": 0.06537282941777324,
+      "grad_norm": 0.9878801107406616,
+      "learning_rate": 1.0666666666666667e-05,
+      "loss": 0.5548,
+      "step": 16
+    },
+    {
+      "epoch": 0.06945863125638406,
+      "grad_norm": 0.8131901025772095,
+      "learning_rate": 1.1333333333333334e-05,
+      "loss": 0.4878,
+      "step": 17
+    },
+    {
+      "epoch": 0.0735444330949949,
+      "grad_norm": 0.7322743535041809,
+      "learning_rate": 1.2e-05,
+      "loss": 0.5159,
+      "step": 18
+    },
+    {
+      "epoch": 0.07763023493360573,
+      "grad_norm": 0.6428759098052979,
+      "learning_rate": 1.2666666666666667e-05,
+      "loss": 0.4575,
+      "step": 19
+    },
+    {
+      "epoch": 0.08171603677221655,
+      "grad_norm": 0.562318742275238,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.4571,
+      "step": 20
+    },
+    {
+      "epoch": 0.08580183861082738,
+      "grad_norm": 0.5707699060440063,
+      "learning_rate": 1.4e-05,
+      "loss": 0.4592,
+      "step": 21
+    },
+    {
+      "epoch": 0.0898876404494382,
+      "grad_norm": 0.5272228717803955,
+      "learning_rate": 1.4666666666666666e-05,
+      "loss": 0.4457,
+      "step": 22
+    },
+    {
+      "epoch": 0.09397344228804903,
+      "grad_norm": 0.5120903253555298,
+      "learning_rate": 1.5333333333333334e-05,
+      "loss": 0.4034,
+      "step": 23
+    },
+    {
+      "epoch": 0.09805924412665985,
+      "grad_norm": 0.46359285712242126,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.4037,
+      "step": 24
+    },
+    {
+      "epoch": 0.10214504596527069,
+      "grad_norm": 0.49431198835372925,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.3875,
+      "step": 25
+    },
+    {
+      "epoch": 0.1062308478038815,
+      "grad_norm": 0.4450273811817169,
+      "learning_rate": 1.7333333333333336e-05,
+      "loss": 0.3797,
+      "step": 26
+    },
+    {
+      "epoch": 0.11031664964249234,
+      "grad_norm": 0.4551868140697479,
+      "learning_rate": 1.8e-05,
+      "loss": 0.3512,
+      "step": 27
+    },
+    {
+      "epoch": 0.11440245148110317,
+      "grad_norm": 0.5083736777305603,
+      "learning_rate": 1.866666666666667e-05,
+      "loss": 0.3906,
+      "step": 28
+    },
+    {
+      "epoch": 0.118488253319714,
+      "grad_norm": 0.47295963764190674,
+      "learning_rate": 1.9333333333333333e-05,
+      "loss": 0.3554,
+      "step": 29
+    },
+    {
+      "epoch": 0.12257405515832483,
+      "grad_norm": 0.4848616123199463,
+      "learning_rate": 2e-05,
+      "loss": 0.3712,
+      "step": 30
+    },
+    {
+      "epoch": 0.12665985699693566,
+      "grad_norm": 0.4398118555545807,
+      "learning_rate": 1.999989986294826e-05,
+      "loss": 0.3694,
+      "step": 31
+    },
+    {
+      "epoch": 0.13074565883554648,
+      "grad_norm": 0.41183602809906006,
+      "learning_rate": 1.9999599453798523e-05,
+      "loss": 0.3336,
+      "step": 32
+    },
+    {
+      "epoch": 0.1348314606741573,
+      "grad_norm": 0.492713987827301,
+      "learning_rate": 1.999909877856721e-05,
+      "loss": 0.3657,
+      "step": 33
+    },
+    {
+      "epoch": 0.13891726251276812,
+      "grad_norm": 0.4517015516757965,
+      "learning_rate": 1.9998397847281548e-05,
+      "loss": 0.367,
+      "step": 34
+    },
+    {
+      "epoch": 0.14300306435137897,
+      "grad_norm": 0.4641965627670288,
+      "learning_rate": 1.9997496673979375e-05,
+      "loss": 0.3565,
+      "step": 35
+    },
+    {
+      "epoch": 0.1470888661899898,
+      "grad_norm": 0.4812065064907074,
+      "learning_rate": 1.9996395276708856e-05,
+      "loss": 0.3773,
+      "step": 36
+    },
+    {
+      "epoch": 0.1511746680286006,
+      "grad_norm": 0.42300987243652344,
+      "learning_rate": 1.999509367752813e-05,
+      "loss": 0.3643,
+      "step": 37
+    },
+    {
+      "epoch": 0.15526046986721145,
+      "grad_norm": 0.4512963593006134,
+      "learning_rate": 1.9993591902504854e-05,
+      "loss": 0.3409,
+      "step": 38
+    },
+    {
+      "epoch": 0.15934627170582227,
+      "grad_norm": 0.41626426577568054,
+      "learning_rate": 1.9991889981715696e-05,
+      "loss": 0.3546,
+      "step": 39
+    },
+    {
+      "epoch": 0.1634320735444331,
+      "grad_norm": 0.43549367785453796,
+      "learning_rate": 1.9989987949245725e-05,
+      "loss": 0.3091,
+      "step": 40
+    },
+    {
+      "epoch": 0.1675178753830439,
+      "grad_norm": 0.4042600393295288,
+      "learning_rate": 1.9987885843187717e-05,
+      "loss": 0.3174,
+      "step": 41
+    },
+    {
+      "epoch": 0.17160367722165476,
+      "grad_norm": 0.4394363462924957,
+      "learning_rate": 1.9985583705641418e-05,
+      "loss": 0.3601,
+      "step": 42
+    },
+    {
+      "epoch": 0.17568947906026558,
+      "grad_norm": 0.4294170141220093,
+      "learning_rate": 1.9983081582712684e-05,
+      "loss": 0.3283,
+      "step": 43
+    },
+    {
+      "epoch": 0.1797752808988764,
+      "grad_norm": 0.44452300667762756,
+      "learning_rate": 1.998037952451255e-05,
+      "loss": 0.3367,
+      "step": 44
+    },
+    {
+      "epoch": 0.18386108273748722,
+      "grad_norm": 0.4113090932369232,
+      "learning_rate": 1.9977477585156252e-05,
+      "loss": 0.2986,
+      "step": 45
+    },
+    {
+      "epoch": 0.18794688457609807,
+      "grad_norm": 0.44443050026893616,
+      "learning_rate": 1.9974375822762117e-05,
+      "loss": 0.3463,
+      "step": 46
+    },
+    {
+      "epoch": 0.1920326864147089,
+      "grad_norm": 0.4303809106349945,
+      "learning_rate": 1.9971074299450414e-05,
+      "loss": 0.3281,
+      "step": 47
+    },
+    {
+      "epoch": 0.1961184882533197,
+      "grad_norm": 0.4178621470928192,
+      "learning_rate": 1.9967573081342103e-05,
+      "loss": 0.3629,
+      "step": 48
+    },
+    {
+      "epoch": 0.20020429009193055,
+      "grad_norm": 0.38657113909721375,
+      "learning_rate": 1.9963872238557516e-05,
+      "loss": 0.3225,
+      "step": 49
+    },
+    {
+      "epoch": 0.20429009193054137,
+      "grad_norm": 0.5300270915031433,
+      "learning_rate": 1.9959971845214953e-05,
+      "loss": 0.3279,
+      "step": 50
+    },
+    {
+      "epoch": 0.2083758937691522,
+      "grad_norm": 0.4061177968978882,
+      "learning_rate": 1.9955871979429188e-05,
+      "loss": 0.3278,
+      "step": 51
+    },
+    {
+      "epoch": 0.212461695607763,
+      "grad_norm": 0.41504785418510437,
+      "learning_rate": 1.9951572723309918e-05,
+      "loss": 0.3096,
+      "step": 52
+    },
+    {
+      "epoch": 0.21654749744637386,
+      "grad_norm": 0.4208971858024597,
+      "learning_rate": 1.9947074162960113e-05,
+      "loss": 0.3187,
+      "step": 53
+    },
+    {
+      "epoch": 0.22063329928498468,
+      "grad_norm": 0.36819201707839966,
+      "learning_rate": 1.9942376388474282e-05,
+      "loss": 0.3167,
+      "step": 54
+    },
+    {
+      "epoch": 0.2247191011235955,
+      "grad_norm": 0.43327596783638,
+      "learning_rate": 1.993747949393668e-05,
+      "loss": 0.3188,
+      "step": 55
+    },
+    {
+      "epoch": 0.22880490296220635,
+      "grad_norm": 0.4377865791320801,
+      "learning_rate": 1.9932383577419432e-05,
+      "loss": 0.3478,
+      "step": 56
+    },
+    {
+      "epoch": 0.23289070480081717,
+      "grad_norm": 0.43336397409439087,
+      "learning_rate": 1.992708874098054e-05,
+      "loss": 0.3025,
+      "step": 57
+    },
+    {
+      "epoch": 0.236976506639428,
+      "grad_norm": 0.4399135410785675,
+      "learning_rate": 1.9921595090661872e-05,
+      "loss": 0.3098,
+      "step": 58
+    },
+    {
+      "epoch": 0.2410623084780388,
+      "grad_norm": 0.4253901243209839,
+      "learning_rate": 1.991590273648702e-05,
+      "loss": 0.3303,
+      "step": 59
+    },
+    {
+      "epoch": 0.24514811031664965,
+      "grad_norm": 0.39254307746887207,
+      "learning_rate": 1.9910011792459086e-05,
+      "loss": 0.3018,
+      "step": 60
+    },
+    {
+      "epoch": 0.24923391215526047,
+      "grad_norm": 0.4217659831047058,
+      "learning_rate": 1.9903922376558432e-05,
+      "loss": 0.285,
+      "step": 61
+    },
+    {
+      "epoch": 0.2533197139938713,
+      "grad_norm": 0.48558109998703003,
+      "learning_rate": 1.989763461074029e-05,
+      "loss": 0.3221,
+      "step": 62
+    },
+    {
+      "epoch": 0.2574055158324821,
+      "grad_norm": 0.47454214096069336,
+      "learning_rate": 1.989114862093232e-05,
+      "loss": 0.3056,
+      "step": 63
+    },
+    {
+      "epoch": 0.26149131767109296,
+      "grad_norm": 0.4013993442058563,
+      "learning_rate": 1.9884464537032103e-05,
+      "loss": 0.3376,
+      "step": 64
+    },
+    {
+      "epoch": 0.26557711950970375,
+      "grad_norm": 0.4264606237411499,
+      "learning_rate": 1.9877582492904533e-05,
+      "loss": 0.3158,
+      "step": 65
+    },
+    {
+      "epoch": 0.2696629213483146,
+      "grad_norm": 0.5440453886985779,
+      "learning_rate": 1.9870502626379127e-05,
+      "loss": 0.3056,
+      "step": 66
+    },
+    {
+      "epoch": 0.27374872318692545,
+      "grad_norm": 0.40003377199172974,
+      "learning_rate": 1.9863225079247286e-05,
+      "loss": 0.3357,
+      "step": 67
+    },
+    {
+      "epoch": 0.27783452502553624,
+      "grad_norm": 0.39155763387680054,
+      "learning_rate": 1.985574999725943e-05,
+      "loss": 0.2819,
+      "step": 68
+    },
+    {
+      "epoch": 0.2819203268641471,
+      "grad_norm": 0.4461009204387665,
+      "learning_rate": 1.9848077530122083e-05,
+      "loss": 0.2732,
+      "step": 69
+    },
+    {
+      "epoch": 0.28600612870275793,
+      "grad_norm": 0.38970062136650085,
+      "learning_rate": 1.9840207831494903e-05,
+      "loss": 0.2957,
+      "step": 70
+    },
+    {
+      "epoch": 0.2900919305413687,
+      "grad_norm": 0.4369664788246155,
+      "learning_rate": 1.983214105898757e-05,
+      "loss": 0.3158,
+      "step": 71
+    },
+    {
+      "epoch": 0.2941777323799796,
+      "grad_norm": 0.4734659492969513,
+      "learning_rate": 1.9823877374156647e-05,
+      "loss": 0.3054,
+      "step": 72
+    },
+    {
+      "epoch": 0.2982635342185904,
+      "grad_norm": 0.3933468461036682,
+      "learning_rate": 1.9815416942502346e-05,
+      "loss": 0.286,
+      "step": 73
+    },
+    {
+      "epoch": 0.3023493360572012,
+      "grad_norm": 0.4472273290157318,
+      "learning_rate": 1.98067599334652e-05,
+      "loss": 0.3149,
+      "step": 74
+    },
+    {
+      "epoch": 0.30643513789581206,
+      "grad_norm": 0.43143752217292786,
+      "learning_rate": 1.979790652042268e-05,
+      "loss": 0.2792,
+      "step": 75
+    },
+    {
+      "epoch": 0.3105209397344229,
+      "grad_norm": 0.4325246512889862,
+      "learning_rate": 1.978885688068572e-05,
+      "loss": 0.3024,
+      "step": 76
+    },
+    {
+      "epoch": 0.3146067415730337,
+      "grad_norm": 0.48796600103378296,
+      "learning_rate": 1.9779611195495177e-05,
+      "loss": 0.3343,
+      "step": 77
+    },
+    {
+      "epoch": 0.31869254341164455,
+      "grad_norm": 0.40505748987197876,
+      "learning_rate": 1.977016965001817e-05,
+      "loss": 0.2753,
+      "step": 78
+    },
+    {
+      "epoch": 0.32277834525025534,
+      "grad_norm": 0.40753036737442017,
+      "learning_rate": 1.976053243334442e-05,
+      "loss": 0.3073,
+      "step": 79
+    },
+    {
+      "epoch": 0.3268641470888662,
+      "grad_norm": 0.4000149071216583,
+      "learning_rate": 1.9750699738482403e-05,
+      "loss": 0.284,
+      "step": 80
+    },
+    {
+      "epoch": 0.33094994892747703,
+      "grad_norm": 0.42099907994270325,
+      "learning_rate": 1.9740671762355548e-05,
+      "loss": 0.2881,
+      "step": 81
+    },
+    {
+      "epoch": 0.3350357507660878,
+      "grad_norm": 0.4155902564525604,
+      "learning_rate": 1.973044870579824e-05,
+      "loss": 0.2969,
+      "step": 82
+    },
+    {
+      "epoch": 0.3350357507660878,
+      "eval_loss": 0.31923907995224,
+      "eval_runtime": 5.81,
+      "eval_samples_per_second": 13.597,
+      "eval_steps_per_second": 1.721,
+      "step": 82
+    },
+    {
+      "epoch": 0.3391215526046987,
+      "grad_norm": 0.39282551407814026,
+      "learning_rate": 1.972003077355183e-05,
+      "loss": 0.2948,
+      "step": 83
+    },
+    {
+      "epoch": 0.3432073544433095,
+      "grad_norm": 0.4381943643093109,
+      "learning_rate": 1.9709418174260523e-05,
+      "loss": 0.3454,
+      "step": 84
+    },
+    {
+      "epoch": 0.3472931562819203,
+      "grad_norm": 0.4093382954597473,
+      "learning_rate": 1.9698611120467196e-05,
+      "loss": 0.2962,
+      "step": 85
+    },
+    {
+      "epoch": 0.35137895812053116,
+      "grad_norm": 0.450135737657547,
+      "learning_rate": 1.9687609828609156e-05,
+      "loss": 0.3243,
+      "step": 86
+    },
+    {
+      "epoch": 0.355464759959142,
+      "grad_norm": 0.4139018654823303,
+      "learning_rate": 1.9676414519013782e-05,
+      "loss": 0.2996,
+      "step": 87
+    },
+    {
+      "epoch": 0.3595505617977528,
+      "grad_norm": 0.40026575326919556,
+      "learning_rate": 1.966502541589414e-05,
+      "loss": 0.2788,
+      "step": 88
+    },
+    {
+      "epoch": 0.36363636363636365,
+      "grad_norm": 0.36627820134162903,
+      "learning_rate": 1.965344274734447e-05,
+      "loss": 0.2857,
+      "step": 89
+    },
+    {
+      "epoch": 0.36772216547497444,
+      "grad_norm": 0.42685478925704956,
+      "learning_rate": 1.9641666745335626e-05,
+      "loss": 0.2995,
+      "step": 90
+    },
+    {
+      "epoch": 0.3718079673135853,
+      "grad_norm": 0.374288946390152,
+      "learning_rate": 1.9629697645710432e-05,
+      "loss": 0.3056,
+      "step": 91
+    },
+    {
+      "epoch": 0.37589376915219613,
+      "grad_norm": 0.3649786114692688,
+      "learning_rate": 1.961753568817896e-05,
+      "loss": 0.2854,
+      "step": 92
+    },
+    {
+      "epoch": 0.3799795709908069,
+      "grad_norm": 0.38573023676872253,
+      "learning_rate": 1.9605181116313725e-05,
+      "loss": 0.2667,
+      "step": 93
+    },
+    {
+      "epoch": 0.3840653728294178,
+      "grad_norm": 0.37577807903289795,
+      "learning_rate": 1.9592634177544803e-05,
+      "loss": 0.2815,
+      "step": 94
+    },
+    {
+      "epoch": 0.3881511746680286,
+      "grad_norm": 0.4320047199726105,
+      "learning_rate": 1.957989512315489e-05,
+      "loss": 0.3094,
+      "step": 95
+    },
+    {
+      "epoch": 0.3922369765066394,
+      "grad_norm": 0.3816889524459839,
+      "learning_rate": 1.9566964208274254e-05,
+      "loss": 0.292,
+      "step": 96
+    },
+    {
+      "epoch": 0.39632277834525026,
+      "grad_norm": 0.3946669399738312,
+      "learning_rate": 1.9553841691875632e-05,
+      "loss": 0.3002,
+      "step": 97
+    },
+    {
+      "epoch": 0.4004085801838611,
+      "grad_norm": 0.36885613203048706,
+      "learning_rate": 1.9540527836769047e-05,
+      "loss": 0.2583,
+      "step": 98
+    },
+    {
+      "epoch": 0.4044943820224719,
+      "grad_norm": 0.37865176796913147,
+      "learning_rate": 1.9527022909596537e-05,
+      "loss": 0.2787,
+      "step": 99
+    },
+    {
+      "epoch": 0.40858018386108275,
+      "grad_norm": 0.4429585337638855,
+      "learning_rate": 1.951332718082682e-05,
+      "loss": 0.3226,
+      "step": 100
+    },
+    {
+      "epoch": 0.41266598569969354,
+      "grad_norm": 0.3926009237766266,
+      "learning_rate": 1.9499440924749878e-05,
+      "loss": 0.2914,
+      "step": 101
+    },
+    {
+      "epoch": 0.4167517875383044,
+      "grad_norm": 0.3467339277267456,
+      "learning_rate": 1.9485364419471454e-05,
+      "loss": 0.266,
+      "step": 102
+    },
+    {
+      "epoch": 0.42083758937691523,
+      "grad_norm": 0.4126642644405365,
+      "learning_rate": 1.9471097946907506e-05,
+      "loss": 0.2775,
+      "step": 103
+    },
+    {
+      "epoch": 0.424923391215526,
+      "grad_norm": 0.44586020708084106,
+      "learning_rate": 1.9456641792778527e-05,
+      "loss": 0.2884,
+      "step": 104
+    },
+    {
+      "epoch": 0.4290091930541369,
+      "grad_norm": 0.3969588279724121,
+      "learning_rate": 1.9441996246603848e-05,
+      "loss": 0.2835,
+      "step": 105
+    },
+    {
+      "epoch": 0.4330949948927477,
+      "grad_norm": 0.38928356766700745,
+      "learning_rate": 1.9427161601695833e-05,
+      "loss": 0.2826,
+      "step": 106
+    },
+    {
+      "epoch": 0.4371807967313585,
+      "grad_norm": 0.4089799225330353,
+      "learning_rate": 1.9412138155154e-05,
+      "loss": 0.2817,
+      "step": 107
+    },
+    {
+      "epoch": 0.44126659856996936,
+      "grad_norm": 0.375505656003952,
+      "learning_rate": 1.9396926207859085e-05,
+      "loss": 0.2882,
+      "step": 108
+    },
+    {
+      "epoch": 0.4453524004085802,
+      "grad_norm": 0.406118780374527,
+      "learning_rate": 1.9381526064466995e-05,
+      "loss": 0.2861,
+      "step": 109
+    },
+    {
+      "epoch": 0.449438202247191,
+      "grad_norm": 0.3882409334182739,
+      "learning_rate": 1.9365938033402715e-05,
+      "loss": 0.261,
+      "step": 110
+    },
+    {
+      "epoch": 0.45352400408580185,
+      "grad_norm": 0.4351583421230316,
+      "learning_rate": 1.9350162426854152e-05,
+      "loss": 0.3014,
+      "step": 111
+    },
+    {
+      "epoch": 0.4576098059244127,
+      "grad_norm": 0.3621097505092621,
+      "learning_rate": 1.933419956076584e-05,
+      "loss": 0.2728,
+      "step": 112
+    },
+    {
+      "epoch": 0.4616956077630235,
+      "grad_norm": 0.3881032466888428,
+      "learning_rate": 1.9318049754832656e-05,
+      "loss": 0.2736,
+      "step": 113
+    },
+    {
+      "epoch": 0.46578140960163433,
+      "grad_norm": 0.37627285718917847,
+      "learning_rate": 1.9301713332493386e-05,
+      "loss": 0.2707,
+      "step": 114
+    },
+    {
+      "epoch": 0.4698672114402451,
+      "grad_norm": 0.4285913109779358,
+      "learning_rate": 1.9285190620924267e-05,
+      "loss": 0.2815,
+      "step": 115
+    },
+    {
+      "epoch": 0.473953013278856,
+      "grad_norm": 0.35718926787376404,
+      "learning_rate": 1.926848195103242e-05,
+      "loss": 0.2621,
+      "step": 116
+    },
+    {
+      "epoch": 0.4780388151174668,
+      "grad_norm": 0.3852044641971588,
+      "learning_rate": 1.925158765744924e-05,
+      "loss": 0.283,
+      "step": 117
+    },
+    {
+      "epoch": 0.4821246169560776,
+      "grad_norm": 0.3884032368659973,
+      "learning_rate": 1.923450807852367e-05,
+      "loss": 0.2711,
+      "step": 118
+    },
+    {
+      "epoch": 0.48621041879468846,
+      "grad_norm": 0.4398249685764313,
+      "learning_rate": 1.9217243556315445e-05,
+      "loss": 0.2757,
+      "step": 119
+    },
+    {
+      "epoch": 0.4902962206332993,
+      "grad_norm": 0.36689624190330505,
+      "learning_rate": 1.9199794436588244e-05,
+      "loss": 0.2669,
+      "step": 120
+    },
+    {
+      "epoch": 0.4943820224719101,
+      "grad_norm": 0.46398666501045227,
+      "learning_rate": 1.9182161068802742e-05,
+      "loss": 0.2683,
+      "step": 121
+    },
+    {
+      "epoch": 0.49846782431052095,
+      "grad_norm": 0.40020987391471863,
+      "learning_rate": 1.916434380610963e-05,
+      "loss": 0.2927,
+      "step": 122
+    },
+    {
+      "epoch": 0.5025536261491318,
+      "grad_norm": 0.4032459259033203,
+      "learning_rate": 1.9146343005342546e-05,
+      "loss": 0.31,
+      "step": 123
+    },
+    {
+      "epoch": 0.5066394279877426,
+      "grad_norm": 0.44166550040245056,
+      "learning_rate": 1.912815902701091e-05,
+      "loss": 0.2842,
+      "step": 124
+    },
+    {
+      "epoch": 0.5107252298263534,
+      "grad_norm": 0.39895153045654297,
+      "learning_rate": 1.9109792235292715e-05,
+      "loss": 0.2766,
+      "step": 125
+    },
+    {
+      "epoch": 0.5148110316649642,
+      "grad_norm": 0.3415013253688812,
+      "learning_rate": 1.909124299802724e-05,
+      "loss": 0.2761,
+      "step": 126
+    },
+    {
+      "epoch": 0.5188968335035751,
+      "grad_norm": 0.3837663531303406,
+      "learning_rate": 1.9072511686707663e-05,
+      "loss": 0.2797,
+      "step": 127
+    },
+    {
+      "epoch": 0.5229826353421859,
+      "grad_norm": 0.4030819833278656,
+      "learning_rate": 1.9053598676473656e-05,
+      "loss": 0.2932,
+      "step": 128
+    },
+    {
+      "epoch": 0.5270684371807968,
+      "grad_norm": 0.40120938420295715,
+      "learning_rate": 1.9034504346103825e-05,
+      "loss": 0.2698,
+      "step": 129
+    },
+    {
+      "epoch": 0.5311542390194075,
+      "grad_norm": 0.3621327579021454,
+      "learning_rate": 1.9015229078008163e-05,
+      "loss": 0.298,
+      "step": 130
+    },
+    {
+      "epoch": 0.5352400408580184,
+      "grad_norm": 0.33476150035858154,
+      "learning_rate": 1.8995773258220374e-05,
+      "loss": 0.2612,
+      "step": 131
+    },
+    {
+      "epoch": 0.5393258426966292,
+      "grad_norm": 0.3523140549659729,
+      "learning_rate": 1.8976137276390145e-05,
+      "loss": 0.2671,
+      "step": 132
+    },
+    {
+      "epoch": 0.54341164453524,
+      "grad_norm": 0.3624558746814728,
+      "learning_rate": 1.8956321525775337e-05,
+      "loss": 0.2687,
+      "step": 133
+    },
+    {
+      "epoch": 0.5474974463738509,
+      "grad_norm": 0.35892072319984436,
+      "learning_rate": 1.8936326403234125e-05,
+      "loss": 0.2755,
+      "step": 134
+    },
+    {
+      "epoch": 0.5515832482124617,
+      "grad_norm": 0.3678256869316101,
+      "learning_rate": 1.891615230921703e-05,
+      "loss": 0.278,
+      "step": 135
+    },
+    {
+      "epoch": 0.5556690500510725,
+      "grad_norm": 0.38125160336494446,
+      "learning_rate": 1.8895799647758912e-05,
+      "loss": 0.2765,
+      "step": 136
+    },
+    {
+      "epoch": 0.5597548518896833,
+      "grad_norm": 0.40152257680892944,
+      "learning_rate": 1.8875268826470875e-05,
+      "loss": 0.3239,
+      "step": 137
+    },
+    {
+      "epoch": 0.5638406537282942,
+      "grad_norm": 0.3935178816318512,
+      "learning_rate": 1.8854560256532098e-05,
+      "loss": 0.2956,
+      "step": 138
+    },
+    {
+      "epoch": 0.567926455566905,
+      "grad_norm": 0.4389478266239166,
+      "learning_rate": 1.8833674352681613e-05,
+      "loss": 0.2968,
+      "step": 139
+    },
+    {
+      "epoch": 0.5720122574055159,
+      "grad_norm": 0.3884355127811432,
+      "learning_rate": 1.881261153320999e-05,
+      "loss": 0.3074,
+      "step": 140
+    },
+    {
+      "epoch": 0.5760980592441267,
+      "grad_norm": 0.4054373502731323,
+      "learning_rate": 1.879137221995095e-05,
+      "loss": 0.2996,
+      "step": 141
+    },
+    {
+      "epoch": 0.5801838610827375,
+      "grad_norm": 0.4423893690109253,
+      "learning_rate": 1.8769956838272937e-05,
+      "loss": 0.3082,
+      "step": 142
+    },
+    {
+      "epoch": 0.5842696629213483,
+      "grad_norm": 0.42978307604789734,
+      "learning_rate": 1.8748365817070586e-05,
+      "loss": 0.2878,
+      "step": 143
+    },
+    {
+      "epoch": 0.5883554647599591,
+      "grad_norm": 0.38182228803634644,
+      "learning_rate": 1.8726599588756144e-05,
+      "loss": 0.2649,
+      "step": 144
+    },
+    {
+      "epoch": 0.59244126659857,
+      "grad_norm": 0.43477413058280945,
+      "learning_rate": 1.8704658589250795e-05,
+      "loss": 0.271,
+      "step": 145
+    },
+    {
+      "epoch": 0.5965270684371808,
+      "grad_norm": 0.3876926898956299,
+      "learning_rate": 1.868254325797594e-05,
+      "loss": 0.2804,
+      "step": 146
+    },
+    {
+      "epoch": 0.6006128702757916,
+      "grad_norm": 0.39310601353645325,
+      "learning_rate": 1.866025403784439e-05,
+      "loss": 0.2767,
+      "step": 147
+    },
+    {
+      "epoch": 0.6046986721144024,
+      "grad_norm": 0.421290785074234,
+      "learning_rate": 1.8637791375251505e-05,
+      "loss": 0.2668,
+      "step": 148
+    },
+    {
+      "epoch": 0.6087844739530133,
+      "grad_norm": 0.450023353099823,
+      "learning_rate": 1.8615155720066247e-05,
+      "loss": 0.2888,
+      "step": 149
+    },
+    {
+      "epoch": 0.6128702757916241,
+      "grad_norm": 0.3645341396331787,
+      "learning_rate": 1.859234752562217e-05,
+      "loss": 0.2828,
+      "step": 150
+    },
+    {
+      "epoch": 0.616956077630235,
+      "grad_norm": 0.41853606700897217,
+      "learning_rate": 1.8569367248708343e-05,
+      "loss": 0.284,
+      "step": 151
+    },
+    {
+      "epoch": 0.6210418794688458,
+      "grad_norm": 0.3675737679004669,
+      "learning_rate": 1.8546215349560204e-05,
+      "loss": 0.2933,
+      "step": 152
+    },
+    {
+      "epoch": 0.6251276813074566,
+      "grad_norm": 0.3668256998062134,
+      "learning_rate": 1.8522892291850335e-05,
+      "loss": 0.2729,
+      "step": 153
+    },
+    {
+      "epoch": 0.6292134831460674,
+      "grad_norm": 0.34576019644737244,
+      "learning_rate": 1.849939854267919e-05,
+      "loss": 0.2612,
+      "step": 154
+    },
+    {
+      "epoch": 0.6332992849846782,
+      "grad_norm": 0.41370126605033875,
+      "learning_rate": 1.847573457256571e-05,
+      "loss": 0.2693,
+      "step": 155
+    },
+    {
+      "epoch": 0.6373850868232891,
+      "grad_norm": 0.4205566644668579,
+      "learning_rate": 1.845190085543795e-05,
+      "loss": 0.2746,
+      "step": 156
+    },
+    {
+      "epoch": 0.6414708886618999,
+      "grad_norm": 0.3997614085674286,
+      "learning_rate": 1.8427897868623535e-05,
+      "loss": 0.2813,
+      "step": 157
+    },
+    {
+      "epoch": 0.6455566905005107,
+      "grad_norm": 0.41005200147628784,
+      "learning_rate": 1.840372609284013e-05,
+      "loss": 0.2647,
+      "step": 158
+    },
+    {
+      "epoch": 0.6496424923391215,
+      "grad_norm": 0.4547550678253174,
+      "learning_rate": 1.8379386012185813e-05,
+      "loss": 0.2791,
+      "step": 159
+    },
+    {
+      "epoch": 0.6537282941777324,
+      "grad_norm": 0.4075047969818115,
+      "learning_rate": 1.8354878114129368e-05,
+      "loss": 0.2769,
+      "step": 160
+    },
+    {
+      "epoch": 0.6578140960163432,
+      "grad_norm": 0.37060046195983887,
+      "learning_rate": 1.8330202889500518e-05,
+      "loss": 0.3028,
+      "step": 161
+    },
+    {
+      "epoch": 0.6618998978549541,
+      "grad_norm": 0.35541340708732605,
+      "learning_rate": 1.8305360832480118e-05,
+      "loss": 0.2981,
+      "step": 162
+    },
+    {
+      "epoch": 0.6659856996935649,
+      "grad_norm": 0.3970625400543213,
+      "learning_rate": 1.8280352440590236e-05,
+      "loss": 0.2634,
+      "step": 163
+    },
+    {
+      "epoch": 0.6700715015321757,
+      "grad_norm": 0.4075865149497986,
+      "learning_rate": 1.82551782146842e-05,
+      "loss": 0.3027,
+      "step": 164
+    },
+    {
+      "epoch": 0.6700715015321757,
+      "eval_loss": 0.291363924741745,
+      "eval_runtime": 5.7936,
+      "eval_samples_per_second": 13.636,
+      "eval_steps_per_second": 1.726,
+      "step": 164
+    },
+    {
+      "epoch": 0.6741573033707865,
+      "grad_norm": 0.34390076994895935,
+      "learning_rate": 1.8229838658936566e-05,
+      "loss": 0.2536,
+      "step": 165
+    },
+    {
+      "epoch": 0.6782431052093973,
+      "grad_norm": 0.3729197084903717,
+      "learning_rate": 1.8204334280833005e-05,
+      "loss": 0.2739,
+      "step": 166
+    },
+    {
+      "epoch": 0.6823289070480082,
+      "grad_norm": 0.3974601924419403,
+      "learning_rate": 1.817866559116017e-05,
+      "loss": 0.2858,
+      "step": 167
+    },
+    {
+      "epoch": 0.686414708886619,
+      "grad_norm": 0.3424644470214844,
+      "learning_rate": 1.8152833103995443e-05,
+      "loss": 0.2305,
+      "step": 168
+    },
+    {
+      "epoch": 0.6905005107252298,
+      "grad_norm": 0.4293709397315979,
+      "learning_rate": 1.8126837336696645e-05,
+      "loss": 0.3179,
+      "step": 169
+    },
+    {
+      "epoch": 0.6945863125638406,
+      "grad_norm": 0.3259459435939789,
+      "learning_rate": 1.8100678809891668e-05,
+      "loss": 0.2589,
+      "step": 170
+    },
+    {
+      "epoch": 0.6986721144024515,
+      "grad_norm": 0.40771302580833435,
+      "learning_rate": 1.807435804746807e-05,
+      "loss": 0.2637,
+      "step": 171
+    },
+    {
+      "epoch": 0.7027579162410623,
+      "grad_norm": 0.3847212493419647,
+      "learning_rate": 1.8047875576562556e-05,
+      "loss": 0.2782,
+      "step": 172
+    },
+    {
+      "epoch": 0.7068437180796732,
+      "grad_norm": 0.35547974705696106,
+      "learning_rate": 1.802123192755044e-05,
+      "loss": 0.2695,
+      "step": 173
+    },
+    {
+      "epoch": 0.710929519918284,
+      "grad_norm": 0.3954298198223114,
+      "learning_rate": 1.7994427634035016e-05,
+      "loss": 0.3005,
+      "step": 174
+    },
+    {
+      "epoch": 0.7150153217568948,
+      "grad_norm": 0.3506409525871277,
+      "learning_rate": 1.796746323283686e-05,
+      "loss": 0.2716,
+      "step": 175
+    },
+    {
+      "epoch": 0.7191011235955056,
+      "grad_norm": 0.42227277159690857,
+      "learning_rate": 1.7940339263983112e-05,
+      "loss": 0.2915,
+      "step": 176
+    },
+    {
+      "epoch": 0.7231869254341164,
+      "grad_norm": 0.3948259949684143,
+      "learning_rate": 1.791305627069662e-05,
+      "loss": 0.2883,
+      "step": 177
+    },
+    {
+      "epoch": 0.7272727272727273,
+      "grad_norm": 0.3580792248249054,
+      "learning_rate": 1.7885614799385086e-05,
+      "loss": 0.2782,
+      "step": 178
+    },
+    {
+      "epoch": 0.7313585291113381,
+      "grad_norm": 0.39698660373687744,
+      "learning_rate": 1.785801539963012e-05,
+      "loss": 0.2657,
+      "step": 179
+    },
+    {
+      "epoch": 0.7354443309499489,
+      "grad_norm": 0.3663792610168457,
+      "learning_rate": 1.7830258624176224e-05,
+      "loss": 0.2686,
+      "step": 180
+    },
+    {
+      "epoch": 0.7395301327885597,
+      "grad_norm": 0.38216930627822876,
+      "learning_rate": 1.7802345028919728e-05,
+      "loss": 0.2706,
+      "step": 181
+    },
+    {
+      "epoch": 0.7436159346271706,
+      "grad_norm": 0.4187450706958771,
+      "learning_rate": 1.777427517289766e-05,
+      "loss": 0.2573,
+      "step": 182
+    },
+    {
+      "epoch": 0.7477017364657814,
+      "grad_norm": 0.34619036316871643,
+      "learning_rate": 1.7746049618276545e-05,
+      "loss": 0.269,
+      "step": 183
+    },
+    {
+      "epoch": 0.7517875383043923,
+      "grad_norm": 0.35370582342147827,
+      "learning_rate": 1.7717668930341152e-05,
+      "loss": 0.2552,
+      "step": 184
+    },
+    {
+      "epoch": 0.7558733401430031,
+      "grad_norm": 0.4264880418777466,
+      "learning_rate": 1.768913367748316e-05,
+      "loss": 0.2952,
+      "step": 185
+    },
+    {
+      "epoch": 0.7599591419816139,
+      "grad_norm": 0.39135676622390747,
+      "learning_rate": 1.766044443118978e-05,
+      "loss": 0.2661,
+      "step": 186
+    },
+    {
+      "epoch": 0.7640449438202247,
+      "grad_norm": 0.39061596989631653,
+      "learning_rate": 1.7631601766032337e-05,
+      "loss": 0.2737,
+      "step": 187
+    },
+    {
+      "epoch": 0.7681307456588355,
+      "grad_norm": 0.3799816966056824,
+      "learning_rate": 1.7602606259654704e-05,
+      "loss": 0.2767,
+      "step": 188
+    },
+    {
+      "epoch": 0.7722165474974464,
+      "grad_norm": 0.3592148721218109,
+      "learning_rate": 1.7573458492761802e-05,
+      "loss": 0.2448,
+      "step": 189
+    },
+    {
+      "epoch": 0.7763023493360572,
+      "grad_norm": 0.39084604382514954,
+      "learning_rate": 1.7544159049107902e-05,
+      "loss": 0.275,
+      "step": 190
+    },
+    {
+      "epoch": 0.780388151174668,
+      "grad_norm": 0.36443451046943665,
+      "learning_rate": 1.7514708515485002e-05,
+      "loss": 0.2645,
+      "step": 191
+    },
+    {
+      "epoch": 0.7844739530132788,
+      "grad_norm": 0.4001200497150421,
+      "learning_rate": 1.7485107481711014e-05,
+      "loss": 0.2724,
+      "step": 192
+    },
+    {
+      "epoch": 0.7885597548518897,
+      "grad_norm": 0.39093396067619324,
+      "learning_rate": 1.7455356540617988e-05,
+      "loss": 0.2712,
+      "step": 193
+    },
+    {
+      "epoch": 0.7926455566905005,
+      "grad_norm": 0.3430577218532562,
+      "learning_rate": 1.7425456288040236e-05,
+      "loss": 0.2489,
+      "step": 194
+    },
+    {
+      "epoch": 0.7967313585291114,
+      "grad_norm": 0.3573733866214752,
+      "learning_rate": 1.7395407322802374e-05,
+      "loss": 0.2696,
+      "step": 195
+    },
+    {
+      "epoch": 0.8008171603677222,
+      "grad_norm": 0.38158077001571655,
+      "learning_rate": 1.736521024670737e-05,
+      "loss": 0.2814,
+      "step": 196
+    },
+    {
+      "epoch": 0.804902962206333,
+      "grad_norm": 0.366470068693161,
+      "learning_rate": 1.733486566452446e-05,
+      "loss": 0.2529,
+      "step": 197
+    },
+    {
+      "epoch": 0.8089887640449438,
+      "grad_norm": 0.3718278408050537,
+      "learning_rate": 1.7304374183977032e-05,
+      "loss": 0.2747,
+      "step": 198
+    },
+    {
+      "epoch": 0.8130745658835546,
+      "grad_norm": 0.3395809233188629,
+      "learning_rate": 1.7273736415730488e-05,
+      "loss": 0.2693,
+      "step": 199
+    },
+    {
+      "epoch": 0.8171603677221655,
+      "grad_norm": 0.307731032371521,
+      "learning_rate": 1.7242952973379983e-05,
+      "loss": 0.2081,
+      "step": 200
+    },
+    {
+      "epoch": 0.8212461695607763,
+      "grad_norm": 0.3522433936595917,
+      "learning_rate": 1.7212024473438145e-05,
+      "loss": 0.2495,
+      "step": 201
+    },
+    {
+      "epoch": 0.8253319713993871,
+      "grad_norm": 0.35946980118751526,
+      "learning_rate": 1.7180951535322742e-05,
+      "loss": 0.2702,
+      "step": 202
+    },
+    {
+      "epoch": 0.8294177732379979,
+      "grad_norm": 0.3933047950267792,
+      "learning_rate": 1.7149734781344247e-05,
+      "loss": 0.2629,
+      "step": 203
+    },
+    {
+      "epoch": 0.8335035750766088,
+      "grad_norm": 0.3658384084701538,
+      "learning_rate": 1.7118374836693407e-05,
+      "loss": 0.2538,
+      "step": 204
+    },
+    {
+      "epoch": 0.8375893769152196,
+      "grad_norm": 0.3532220423221588,
+      "learning_rate": 1.7086872329428702e-05,
+      "loss": 0.2587,
+      "step": 205
+    },
+    {
+      "epoch": 0.8416751787538305,
+      "grad_norm": 0.3619686961174011,
+      "learning_rate": 1.705522789046377e-05,
+      "loss": 0.2658,
+      "step": 206
+    },
+    {
+      "epoch": 0.8457609805924413,
+      "grad_norm": 0.4083801209926605,
+      "learning_rate": 1.7023442153554776e-05,
+      "loss": 0.2614,
+      "step": 207
+    },
+    {
+      "epoch": 0.849846782431052,
+      "grad_norm": 0.3868924081325531,
+      "learning_rate": 1.6991515755287715e-05,
+      "loss": 0.2831,
+      "step": 208
+    },
+    {
+      "epoch": 0.8539325842696629,
+      "grad_norm": 0.38413897156715393,
+      "learning_rate": 1.695944933506567e-05,
+      "loss": 0.2596,
+      "step": 209
+    },
+    {
+      "epoch": 0.8580183861082737,
+      "grad_norm": 0.34999531507492065,
+      "learning_rate": 1.6927243535095995e-05,
+      "loss": 0.2842,
+      "step": 210
+    },
+    {
+      "epoch": 0.8621041879468846,
+      "grad_norm": 0.328204482793808,
+      "learning_rate": 1.6894899000377462e-05,
+      "loss": 0.2332,
+      "step": 211
+    },
+    {
+      "epoch": 0.8661899897854954,
+      "grad_norm": 0.3802552819252014,
+      "learning_rate": 1.686241637868734e-05,
+      "loss": 0.2709,
+      "step": 212
+    },
+    {
+      "epoch": 0.8702757916241062,
+      "grad_norm": 0.35758858919143677,
+      "learning_rate": 1.6829796320568416e-05,
+      "loss": 0.279,
+      "step": 213
+    },
+    {
+      "epoch": 0.874361593462717,
+      "grad_norm": 0.3561984896659851,
+      "learning_rate": 1.6797039479315994e-05,
+      "loss": 0.2868,
+      "step": 214
+    },
+    {
+      "epoch": 0.8784473953013279,
+      "grad_norm": 0.32591065764427185,
+      "learning_rate": 1.6764146510964762e-05,
+      "loss": 0.2485,
+      "step": 215
+    },
+    {
+      "epoch": 0.8825331971399387,
+      "grad_norm": 0.36409640312194824,
+      "learning_rate": 1.67311180742757e-05,
+      "loss": 0.2577,
+      "step": 216
+    },
+    {
+      "epoch": 0.8866189989785496,
+      "grad_norm": 0.34685492515563965,
+      "learning_rate": 1.669795483072287e-05,
+      "loss": 0.247,
+      "step": 217
+    },
+    {
+      "epoch": 0.8907048008171604,
+      "grad_norm": 0.3445712625980377,
+      "learning_rate": 1.6664657444480145e-05,
+      "loss": 0.2565,
+      "step": 218
+    },
+    {
+      "epoch": 0.8947906026557712,
+      "grad_norm": 0.34710460901260376,
+      "learning_rate": 1.6631226582407954e-05,
+      "loss": 0.2363,
+      "step": 219
+    },
+    {
+      "epoch": 0.898876404494382,
+      "grad_norm": 0.33726766705513,
+      "learning_rate": 1.6597662914039885e-05,
+      "loss": 0.2483,
+      "step": 220
+    },
+    {
+      "epoch": 0.9029622063329928,
+      "grad_norm": 0.34024032950401306,
+      "learning_rate": 1.65639671115693e-05,
+      "loss": 0.2474,
+      "step": 221
+    },
+    {
+      "epoch": 0.9070480081716037,
+      "grad_norm": 0.38807395100593567,
+      "learning_rate": 1.653013984983585e-05,
+      "loss": 0.2726,
+      "step": 222
+    },
+    {
+      "epoch": 0.9111338100102145,
+      "grad_norm": 0.36375290155410767,
+      "learning_rate": 1.6496181806312005e-05,
+      "loss": 0.2726,
+      "step": 223
+    },
+    {
+      "epoch": 0.9152196118488254,
+      "grad_norm": 0.36927178502082825,
+      "learning_rate": 1.6462093661089432e-05,
+      "loss": 0.2518,
+      "step": 224
+    },
+    {
+      "epoch": 0.9193054136874361,
+      "grad_norm": 0.3809269070625305,
+      "learning_rate": 1.6427876096865394e-05,
+      "loss": 0.2449,
+      "step": 225
+    },
+    {
+      "epoch": 0.923391215526047,
+      "grad_norm": 0.34634968638420105,
+      "learning_rate": 1.6393529798929103e-05,
+      "loss": 0.2575,
+      "step": 226
+    },
+    {
+      "epoch": 0.9274770173646578,
+      "grad_norm": 0.33054831624031067,
+      "learning_rate": 1.635905545514795e-05,
+      "loss": 0.2639,
+      "step": 227
+    },
+    {
+      "epoch": 0.9315628192032687,
+      "grad_norm": 0.35482174158096313,
+      "learning_rate": 1.6324453755953772e-05,
+      "loss": 0.2667,
+      "step": 228
+    },
+    {
+      "epoch": 0.9356486210418795,
+      "grad_norm": 0.3657509684562683,
+      "learning_rate": 1.6289725394328998e-05,
+      "loss": 0.255,
+      "step": 229
+    },
+    {
+      "epoch": 0.9397344228804902,
+      "grad_norm": 0.3343275785446167,
+      "learning_rate": 1.6254871065792776e-05,
+      "loss": 0.2336,
+      "step": 230
+    },
+    {
+      "epoch": 0.9438202247191011,
+      "grad_norm": 0.3493170142173767,
+      "learning_rate": 1.621989146838704e-05,
+      "loss": 0.2649,
+      "step": 231
+    },
+    {
+      "epoch": 0.947906026557712,
+      "grad_norm": 0.3305867612361908,
+      "learning_rate": 1.618478730266255e-05,
+      "loss": 0.2767,
+      "step": 232
+    },
+    {
+      "epoch": 0.9519918283963228,
+      "grad_norm": 0.35817259550094604,
+      "learning_rate": 1.6149559271664835e-05,
+      "loss": 0.2817,
+      "step": 233
+    },
+    {
+      "epoch": 0.9560776302349336,
+      "grad_norm": 0.37733370065689087,
+      "learning_rate": 1.6114208080920125e-05,
+      "loss": 0.2809,
+      "step": 234
+    },
+    {
+      "epoch": 0.9601634320735445,
+      "grad_norm": 0.3227766156196594,
+      "learning_rate": 1.607873443842122e-05,
+      "loss": 0.2545,
+      "step": 235
+    },
+    {
+      "epoch": 0.9642492339121552,
+      "grad_norm": 0.3445710241794586,
+      "learning_rate": 1.6043139054613326e-05,
+      "loss": 0.2476,
+      "step": 236
+    },
+    {
+      "epoch": 0.9683350357507661,
+      "grad_norm": 0.3375508785247803,
+      "learning_rate": 1.600742264237979e-05,
+      "loss": 0.2502,
+      "step": 237
+    },
+    {
+      "epoch": 0.9724208375893769,
+      "grad_norm": 0.356039434671402,
+      "learning_rate": 1.5971585917027864e-05,
+      "loss": 0.268,
+      "step": 238
+    },
+    {
+      "epoch": 0.9765066394279878,
+      "grad_norm": 0.34852373600006104,
+      "learning_rate": 1.5935629596274345e-05,
+      "loss": 0.2605,
+      "step": 239
+    },
+    {
+      "epoch": 0.9805924412665986,
+      "grad_norm": 0.3376101851463318,
+      "learning_rate": 1.5899554400231233e-05,
+      "loss": 0.2567,
+      "step": 240
+    },
+    {
+      "epoch": 0.9846782431052093,
+      "grad_norm": 0.32361170649528503,
+      "learning_rate": 1.586336105139127e-05,
+      "loss": 0.2481,
+      "step": 241
+    },
+    {
+      "epoch": 0.9887640449438202,
+      "grad_norm": 0.35558903217315674,
+      "learning_rate": 1.5827050274613512e-05,
+      "loss": 0.2514,
+      "step": 242
+    },
+    {
+      "epoch": 0.992849846782431,
+      "grad_norm": 0.31636619567871094,
+      "learning_rate": 1.579062279710879e-05,
+      "loss": 0.2237,
+      "step": 243
+    },
+    {
+      "epoch": 0.9969356486210419,
+      "grad_norm": 0.3540779948234558,
+      "learning_rate": 1.5754079348425137e-05,
+      "loss": 0.2381,
+      "step": 244
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 732,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 244,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.127797840707584e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-244/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8a51e619db41bfecd4e2978f86e8cb848022d32d79a042203708d80062927ea
+size 10744

checkpoint-244/zero_to_fp32.py

@@ -0,0 +1,760 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import gc
+import json
+import numpy as np
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device, weights_only=False)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in tqdm(files, desc='Loading checkpoint shards'):
+        state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+class GatheredTensor:
+    """
+    A pseudo tensor that collects partitioned weights.
+    It is more memory efficient when there are multiple groups.
+    """
+
+    def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
+        self.flat_groups = flat_groups
+        self.flat_groups_offset = flat_groups_offset
+        self.offset = offset
+        self.partitioned_numel = partitioned_numel
+        self.shape = shape
+        self.dtype = self.flat_groups[0][0].dtype
+
+    def contiguous(self):
+        """
+        Merge partitioned weights from flat_groups into a single tensor.
+        """
+        end_idx = self.offset + self.partitioned_numel
+        world_size = len(self.flat_groups)
+        pad_flat_param_chunks = []
+
+        for rank_i in range(world_size):
+            # for each rank, we need to collect weights from related group/groups
+            flat_groups_at_rank_i = self.flat_groups[rank_i]
+            start_group_id = None
+            end_group_id = None
+            for group_id in range(len(self.flat_groups_offset)):
+                if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
+                    start_group_id = group_id
+                if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
+                    end_group_id = group_id
+                    break
+            # collect weights from related group/groups
+            for group_id in range(start_group_id, end_group_id + 1):
+                flat_tensor = flat_groups_at_rank_i[group_id]
+                start_offset = self.offset - self.flat_groups_offset[group_id]
+                end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
+                pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
+
+        # collect weights from all ranks
+        pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
+        param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
+        return param
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
+
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # memory efficient tensor
+        tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
+        state_dict[name] = tensor
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def to_torch_tensor(state_dict, return_empty_tensor=False):
+    """
+    Convert state_dict of GatheredTensor to torch tensor
+    """
+    torch_state_dict = {}
+    converted_tensors = {}
+    for name, tensor in state_dict.items():
+        tensor_id = id(tensor)
+        if tensor_id in converted_tensors:  # shared tensors
+            shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
+            torch_state_dict[name] = shared_tensor
+        else:
+            converted_tensors[tensor_id] = name
+            if return_empty_tensor:
+                torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
+            else:
+                torch_state_dict[name] = tensor.contiguous()
+    return torch_state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                             tag=None,
+                                             exclude_frozen_parameters=False,
+                                             lazy_mode=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+        - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
+          Convert the pesduo tensor to torch tensor by ``.contiguous()``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
+    You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint. Or you can load state_dict in lazy mode ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
+        for name, lazy_tensor in state_dict.item():
+            tensor = lazy_tensor.contiguous()  # to cpu
+            print(name, tensor)
+            # del tensor to release memory if it no longer in use
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+    if lazy_mode:
+        return state_dict
+    else:
+        return to_torch_tensor(state_dict)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                                          tag,
+                                                          exclude_frozen_parameters,
+                                                          lazy_mode=True)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        # an memory-efficient approach for sharding
+        empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
+        state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model by shard
+    os.makedirs(output_dir, exist_ok=True)
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
+        shard_state_dict = to_torch_tensor(shard_state_dict)
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard_state_dict, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard_state_dict, output_path)
+        # release the memory of current shard
+        for tensor_name in list(shard_state_dict.keys()):
+            del state_dict[tensor_name]
+            del shard_state_dict[tensor_name]
+        del shard_state_dict
+        gc.collect()
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)

checkpoint-488/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-488/config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "Qwen/Qwen2.5-3B-Instruct",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 70,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.1",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151665
+}

checkpoint-488/generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.48.1"
+}

checkpoint-488/latest

@@ -0,0 +1 @@
+global_step488

checkpoint-488/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7bc974c51afa91050753be9509aad253632e3f54e8ef7abefff0fd407e809321
+size 4956450288

checkpoint-488/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f57fb6f644c16010eadbc4ff90a14eb769cef04dcea00d2116b60e65fb8db3f
+size 1835586736

checkpoint-488/model.safetensors.index.json

@@ -0,0 +1,442 @@
+{
+  "metadata": {
+    "total_size": 6791987200
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

checkpoint-488/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3dcb161b22b2558dbf7e3f8c871050cec383d11a40423fab11f18d5e630639bf
+size 14512

checkpoint-488/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d50af6aef769414a6f28fa1b1bc51ce707dc8ecd15474e03f99a2f10fde086be
+size 14512

checkpoint-488/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d8b2a59c30f5e09b1d7ce944fea889fdfc7000e147a68a8ad08ea9263213eb2
+size 1064

checkpoint-488/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-488/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

checkpoint-488/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-488/trainer_state.json

@@ -0,0 +1,3497 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.996935648621042,
+  "eval_steps": 82,
+  "global_step": 488,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0040858018386108275,
+      "grad_norm": 4.75867223739624,
+      "learning_rate": 6.666666666666667e-07,
+      "loss": 1.3989,
+      "step": 1
+    },
+    {
+      "epoch": 0.0040858018386108275,
+      "eval_loss": 1.7111468315124512,
+      "eval_runtime": 5.4436,
+      "eval_samples_per_second": 14.512,
+      "eval_steps_per_second": 1.837,
+      "step": 1
+    },
+    {
+      "epoch": 0.008171603677221655,
+      "grad_norm": 4.975377559661865,
+      "learning_rate": 1.3333333333333334e-06,
+      "loss": 1.4837,
+      "step": 2
+    },
+    {
+      "epoch": 0.012257405515832482,
+      "grad_norm": 5.219729900360107,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.5181,
+      "step": 3
+    },
+    {
+      "epoch": 0.01634320735444331,
+      "grad_norm": 4.57335901260376,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 1.4106,
+      "step": 4
+    },
+    {
+      "epoch": 0.020429009193054137,
+      "grad_norm": 3.840559720993042,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 1.3763,
+      "step": 5
+    },
+    {
+      "epoch": 0.024514811031664963,
+      "grad_norm": 3.2056212425231934,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.1876,
+      "step": 6
+    },
+    {
+      "epoch": 0.028600612870275793,
+      "grad_norm": 2.6987595558166504,
+      "learning_rate": 4.666666666666667e-06,
+      "loss": 1.2154,
+      "step": 7
+    },
+    {
+      "epoch": 0.03268641470888662,
+      "grad_norm": 2.378502130508423,
+      "learning_rate": 5.333333333333334e-06,
+      "loss": 1.1594,
+      "step": 8
+    },
+    {
+      "epoch": 0.03677221654749745,
+      "grad_norm": 1.7688865661621094,
+      "learning_rate": 6e-06,
+      "loss": 0.8435,
+      "step": 9
+    },
+    {
+      "epoch": 0.04085801838610827,
+      "grad_norm": 1.3263744115829468,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 0.7219,
+      "step": 10
+    },
+    {
+      "epoch": 0.0449438202247191,
+      "grad_norm": 1.3509997129440308,
+      "learning_rate": 7.333333333333333e-06,
+      "loss": 0.8172,
+      "step": 11
+    },
+    {
+      "epoch": 0.049029622063329927,
+      "grad_norm": 1.4541417360305786,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.7393,
+      "step": 12
+    },
+    {
+      "epoch": 0.05311542390194075,
+      "grad_norm": 1.181699275970459,
+      "learning_rate": 8.666666666666668e-06,
+      "loss": 0.664,
+      "step": 13
+    },
+    {
+      "epoch": 0.05720122574055159,
+      "grad_norm": 0.9503294825553894,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.6222,
+      "step": 14
+    },
+    {
+      "epoch": 0.06128702757916241,
+      "grad_norm": 0.7614471316337585,
+      "learning_rate": 1e-05,
+      "loss": 0.56,
+      "step": 15
+    },
+    {
+      "epoch": 0.06537282941777324,
+      "grad_norm": 0.9878801107406616,
+      "learning_rate": 1.0666666666666667e-05,
+      "loss": 0.5548,
+      "step": 16
+    },
+    {
+      "epoch": 0.06945863125638406,
+      "grad_norm": 0.8131901025772095,
+      "learning_rate": 1.1333333333333334e-05,
+      "loss": 0.4878,
+      "step": 17
+    },
+    {
+      "epoch": 0.0735444330949949,
+      "grad_norm": 0.7322743535041809,
+      "learning_rate": 1.2e-05,
+      "loss": 0.5159,
+      "step": 18
+    },
+    {
+      "epoch": 0.07763023493360573,
+      "grad_norm": 0.6428759098052979,
+      "learning_rate": 1.2666666666666667e-05,
+      "loss": 0.4575,
+      "step": 19
+    },
+    {
+      "epoch": 0.08171603677221655,
+      "grad_norm": 0.562318742275238,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.4571,
+      "step": 20
+    },
+    {
+      "epoch": 0.08580183861082738,
+      "grad_norm": 0.5707699060440063,
+      "learning_rate": 1.4e-05,
+      "loss": 0.4592,
+      "step": 21
+    },
+    {
+      "epoch": 0.0898876404494382,
+      "grad_norm": 0.5272228717803955,
+      "learning_rate": 1.4666666666666666e-05,
+      "loss": 0.4457,
+      "step": 22
+    },
+    {
+      "epoch": 0.09397344228804903,
+      "grad_norm": 0.5120903253555298,
+      "learning_rate": 1.5333333333333334e-05,
+      "loss": 0.4034,
+      "step": 23
+    },
+    {
+      "epoch": 0.09805924412665985,
+      "grad_norm": 0.46359285712242126,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.4037,
+      "step": 24
+    },
+    {
+      "epoch": 0.10214504596527069,
+      "grad_norm": 0.49431198835372925,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.3875,
+      "step": 25
+    },
+    {
+      "epoch": 0.1062308478038815,
+      "grad_norm": 0.4450273811817169,
+      "learning_rate": 1.7333333333333336e-05,
+      "loss": 0.3797,
+      "step": 26
+    },
+    {
+      "epoch": 0.11031664964249234,
+      "grad_norm": 0.4551868140697479,
+      "learning_rate": 1.8e-05,
+      "loss": 0.3512,
+      "step": 27
+    },
+    {
+      "epoch": 0.11440245148110317,
+      "grad_norm": 0.5083736777305603,
+      "learning_rate": 1.866666666666667e-05,
+      "loss": 0.3906,
+      "step": 28
+    },
+    {
+      "epoch": 0.118488253319714,
+      "grad_norm": 0.47295963764190674,
+      "learning_rate": 1.9333333333333333e-05,
+      "loss": 0.3554,
+      "step": 29
+    },
+    {
+      "epoch": 0.12257405515832483,
+      "grad_norm": 0.4848616123199463,
+      "learning_rate": 2e-05,
+      "loss": 0.3712,
+      "step": 30
+    },
+    {
+      "epoch": 0.12665985699693566,
+      "grad_norm": 0.4398118555545807,
+      "learning_rate": 1.999989986294826e-05,
+      "loss": 0.3694,
+      "step": 31
+    },
+    {
+      "epoch": 0.13074565883554648,
+      "grad_norm": 0.41183602809906006,
+      "learning_rate": 1.9999599453798523e-05,
+      "loss": 0.3336,
+      "step": 32
+    },
+    {
+      "epoch": 0.1348314606741573,
+      "grad_norm": 0.492713987827301,
+      "learning_rate": 1.999909877856721e-05,
+      "loss": 0.3657,
+      "step": 33
+    },
+    {
+      "epoch": 0.13891726251276812,
+      "grad_norm": 0.4517015516757965,
+      "learning_rate": 1.9998397847281548e-05,
+      "loss": 0.367,
+      "step": 34
+    },
+    {
+      "epoch": 0.14300306435137897,
+      "grad_norm": 0.4641965627670288,
+      "learning_rate": 1.9997496673979375e-05,
+      "loss": 0.3565,
+      "step": 35
+    },
+    {
+      "epoch": 0.1470888661899898,
+      "grad_norm": 0.4812065064907074,
+      "learning_rate": 1.9996395276708856e-05,
+      "loss": 0.3773,
+      "step": 36
+    },
+    {
+      "epoch": 0.1511746680286006,
+      "grad_norm": 0.42300987243652344,
+      "learning_rate": 1.999509367752813e-05,
+      "loss": 0.3643,
+      "step": 37
+    },
+    {
+      "epoch": 0.15526046986721145,
+      "grad_norm": 0.4512963593006134,
+      "learning_rate": 1.9993591902504854e-05,
+      "loss": 0.3409,
+      "step": 38
+    },
+    {
+      "epoch": 0.15934627170582227,
+      "grad_norm": 0.41626426577568054,
+      "learning_rate": 1.9991889981715696e-05,
+      "loss": 0.3546,
+      "step": 39
+    },
+    {
+      "epoch": 0.1634320735444331,
+      "grad_norm": 0.43549367785453796,
+      "learning_rate": 1.9989987949245725e-05,
+      "loss": 0.3091,
+      "step": 40
+    },
+    {
+      "epoch": 0.1675178753830439,
+      "grad_norm": 0.4042600393295288,
+      "learning_rate": 1.9987885843187717e-05,
+      "loss": 0.3174,
+      "step": 41
+    },
+    {
+      "epoch": 0.17160367722165476,
+      "grad_norm": 0.4394363462924957,
+      "learning_rate": 1.9985583705641418e-05,
+      "loss": 0.3601,
+      "step": 42
+    },
+    {
+      "epoch": 0.17568947906026558,
+      "grad_norm": 0.4294170141220093,
+      "learning_rate": 1.9983081582712684e-05,
+      "loss": 0.3283,
+      "step": 43
+    },
+    {
+      "epoch": 0.1797752808988764,
+      "grad_norm": 0.44452300667762756,
+      "learning_rate": 1.998037952451255e-05,
+      "loss": 0.3367,
+      "step": 44
+    },
+    {
+      "epoch": 0.18386108273748722,
+      "grad_norm": 0.4113090932369232,
+      "learning_rate": 1.9977477585156252e-05,
+      "loss": 0.2986,
+      "step": 45
+    },
+    {
+      "epoch": 0.18794688457609807,
+      "grad_norm": 0.44443050026893616,
+      "learning_rate": 1.9974375822762117e-05,
+      "loss": 0.3463,
+      "step": 46
+    },
+    {
+      "epoch": 0.1920326864147089,
+      "grad_norm": 0.4303809106349945,
+      "learning_rate": 1.9971074299450414e-05,
+      "loss": 0.3281,
+      "step": 47
+    },
+    {
+      "epoch": 0.1961184882533197,
+      "grad_norm": 0.4178621470928192,
+      "learning_rate": 1.9967573081342103e-05,
+      "loss": 0.3629,
+      "step": 48
+    },
+    {
+      "epoch": 0.20020429009193055,
+      "grad_norm": 0.38657113909721375,
+      "learning_rate": 1.9963872238557516e-05,
+      "loss": 0.3225,
+      "step": 49
+    },
+    {
+      "epoch": 0.20429009193054137,
+      "grad_norm": 0.5300270915031433,
+      "learning_rate": 1.9959971845214953e-05,
+      "loss": 0.3279,
+      "step": 50
+    },
+    {
+      "epoch": 0.2083758937691522,
+      "grad_norm": 0.4061177968978882,
+      "learning_rate": 1.9955871979429188e-05,
+      "loss": 0.3278,
+      "step": 51
+    },
+    {
+      "epoch": 0.212461695607763,
+      "grad_norm": 0.41504785418510437,
+      "learning_rate": 1.9951572723309918e-05,
+      "loss": 0.3096,
+      "step": 52
+    },
+    {
+      "epoch": 0.21654749744637386,
+      "grad_norm": 0.4208971858024597,
+      "learning_rate": 1.9947074162960113e-05,
+      "loss": 0.3187,
+      "step": 53
+    },
+    {
+      "epoch": 0.22063329928498468,
+      "grad_norm": 0.36819201707839966,
+      "learning_rate": 1.9942376388474282e-05,
+      "loss": 0.3167,
+      "step": 54
+    },
+    {
+      "epoch": 0.2247191011235955,
+      "grad_norm": 0.43327596783638,
+      "learning_rate": 1.993747949393668e-05,
+      "loss": 0.3188,
+      "step": 55
+    },
+    {
+      "epoch": 0.22880490296220635,
+      "grad_norm": 0.4377865791320801,
+      "learning_rate": 1.9932383577419432e-05,
+      "loss": 0.3478,
+      "step": 56
+    },
+    {
+      "epoch": 0.23289070480081717,
+      "grad_norm": 0.43336397409439087,
+      "learning_rate": 1.992708874098054e-05,
+      "loss": 0.3025,
+      "step": 57
+    },
+    {
+      "epoch": 0.236976506639428,
+      "grad_norm": 0.4399135410785675,
+      "learning_rate": 1.9921595090661872e-05,
+      "loss": 0.3098,
+      "step": 58
+    },
+    {
+      "epoch": 0.2410623084780388,
+      "grad_norm": 0.4253901243209839,
+      "learning_rate": 1.991590273648702e-05,
+      "loss": 0.3303,
+      "step": 59
+    },
+    {
+      "epoch": 0.24514811031664965,
+      "grad_norm": 0.39254307746887207,
+      "learning_rate": 1.9910011792459086e-05,
+      "loss": 0.3018,
+      "step": 60
+    },
+    {
+      "epoch": 0.24923391215526047,
+      "grad_norm": 0.4217659831047058,
+      "learning_rate": 1.9903922376558432e-05,
+      "loss": 0.285,
+      "step": 61
+    },
+    {
+      "epoch": 0.2533197139938713,
+      "grad_norm": 0.48558109998703003,
+      "learning_rate": 1.989763461074029e-05,
+      "loss": 0.3221,
+      "step": 62
+    },
+    {
+      "epoch": 0.2574055158324821,
+      "grad_norm": 0.47454214096069336,
+      "learning_rate": 1.989114862093232e-05,
+      "loss": 0.3056,
+      "step": 63
+    },
+    {
+      "epoch": 0.26149131767109296,
+      "grad_norm": 0.4013993442058563,
+      "learning_rate": 1.9884464537032103e-05,
+      "loss": 0.3376,
+      "step": 64
+    },
+    {
+      "epoch": 0.26557711950970375,
+      "grad_norm": 0.4264606237411499,
+      "learning_rate": 1.9877582492904533e-05,
+      "loss": 0.3158,
+      "step": 65
+    },
+    {
+      "epoch": 0.2696629213483146,
+      "grad_norm": 0.5440453886985779,
+      "learning_rate": 1.9870502626379127e-05,
+      "loss": 0.3056,
+      "step": 66
+    },
+    {
+      "epoch": 0.27374872318692545,
+      "grad_norm": 0.40003377199172974,
+      "learning_rate": 1.9863225079247286e-05,
+      "loss": 0.3357,
+      "step": 67
+    },
+    {
+      "epoch": 0.27783452502553624,
+      "grad_norm": 0.39155763387680054,
+      "learning_rate": 1.985574999725943e-05,
+      "loss": 0.2819,
+      "step": 68
+    },
+    {
+      "epoch": 0.2819203268641471,
+      "grad_norm": 0.4461009204387665,
+      "learning_rate": 1.9848077530122083e-05,
+      "loss": 0.2732,
+      "step": 69
+    },
+    {
+      "epoch": 0.28600612870275793,
+      "grad_norm": 0.38970062136650085,
+      "learning_rate": 1.9840207831494903e-05,
+      "loss": 0.2957,
+      "step": 70
+    },
+    {
+      "epoch": 0.2900919305413687,
+      "grad_norm": 0.4369664788246155,
+      "learning_rate": 1.983214105898757e-05,
+      "loss": 0.3158,
+      "step": 71
+    },
+    {
+      "epoch": 0.2941777323799796,
+      "grad_norm": 0.4734659492969513,
+      "learning_rate": 1.9823877374156647e-05,
+      "loss": 0.3054,
+      "step": 72
+    },
+    {
+      "epoch": 0.2982635342185904,
+      "grad_norm": 0.3933468461036682,
+      "learning_rate": 1.9815416942502346e-05,
+      "loss": 0.286,
+      "step": 73
+    },
+    {
+      "epoch": 0.3023493360572012,
+      "grad_norm": 0.4472273290157318,
+      "learning_rate": 1.98067599334652e-05,
+      "loss": 0.3149,
+      "step": 74
+    },
+    {
+      "epoch": 0.30643513789581206,
+      "grad_norm": 0.43143752217292786,
+      "learning_rate": 1.979790652042268e-05,
+      "loss": 0.2792,
+      "step": 75
+    },
+    {
+      "epoch": 0.3105209397344229,
+      "grad_norm": 0.4325246512889862,
+      "learning_rate": 1.978885688068572e-05,
+      "loss": 0.3024,
+      "step": 76
+    },
+    {
+      "epoch": 0.3146067415730337,
+      "grad_norm": 0.48796600103378296,
+      "learning_rate": 1.9779611195495177e-05,
+      "loss": 0.3343,
+      "step": 77
+    },
+    {
+      "epoch": 0.31869254341164455,
+      "grad_norm": 0.40505748987197876,
+      "learning_rate": 1.977016965001817e-05,
+      "loss": 0.2753,
+      "step": 78
+    },
+    {
+      "epoch": 0.32277834525025534,
+      "grad_norm": 0.40753036737442017,
+      "learning_rate": 1.976053243334442e-05,
+      "loss": 0.3073,
+      "step": 79
+    },
+    {
+      "epoch": 0.3268641470888662,
+      "grad_norm": 0.4000149071216583,
+      "learning_rate": 1.9750699738482403e-05,
+      "loss": 0.284,
+      "step": 80
+    },
+    {
+      "epoch": 0.33094994892747703,
+      "grad_norm": 0.42099907994270325,
+      "learning_rate": 1.9740671762355548e-05,
+      "loss": 0.2881,
+      "step": 81
+    },
+    {
+      "epoch": 0.3350357507660878,
+      "grad_norm": 0.4155902564525604,
+      "learning_rate": 1.973044870579824e-05,
+      "loss": 0.2969,
+      "step": 82
+    },
+    {
+      "epoch": 0.3350357507660878,
+      "eval_loss": 0.31923907995224,
+      "eval_runtime": 5.81,
+      "eval_samples_per_second": 13.597,
+      "eval_steps_per_second": 1.721,
+      "step": 82
+    },
+    {
+      "epoch": 0.3391215526046987,
+      "grad_norm": 0.39282551407814026,
+      "learning_rate": 1.972003077355183e-05,
+      "loss": 0.2948,
+      "step": 83
+    },
+    {
+      "epoch": 0.3432073544433095,
+      "grad_norm": 0.4381943643093109,
+      "learning_rate": 1.9709418174260523e-05,
+      "loss": 0.3454,
+      "step": 84
+    },
+    {
+      "epoch": 0.3472931562819203,
+      "grad_norm": 0.4093382954597473,
+      "learning_rate": 1.9698611120467196e-05,
+      "loss": 0.2962,
+      "step": 85
+    },
+    {
+      "epoch": 0.35137895812053116,
+      "grad_norm": 0.450135737657547,
+      "learning_rate": 1.9687609828609156e-05,
+      "loss": 0.3243,
+      "step": 86
+    },
+    {
+      "epoch": 0.355464759959142,
+      "grad_norm": 0.4139018654823303,
+      "learning_rate": 1.9676414519013782e-05,
+      "loss": 0.2996,
+      "step": 87
+    },
+    {
+      "epoch": 0.3595505617977528,
+      "grad_norm": 0.40026575326919556,
+      "learning_rate": 1.966502541589414e-05,
+      "loss": 0.2788,
+      "step": 88
+    },
+    {
+      "epoch": 0.36363636363636365,
+      "grad_norm": 0.36627820134162903,
+      "learning_rate": 1.965344274734447e-05,
+      "loss": 0.2857,
+      "step": 89
+    },
+    {
+      "epoch": 0.36772216547497444,
+      "grad_norm": 0.42685478925704956,
+      "learning_rate": 1.9641666745335626e-05,
+      "loss": 0.2995,
+      "step": 90
+    },
+    {
+      "epoch": 0.3718079673135853,
+      "grad_norm": 0.374288946390152,
+      "learning_rate": 1.9629697645710432e-05,
+      "loss": 0.3056,
+      "step": 91
+    },
+    {
+      "epoch": 0.37589376915219613,
+      "grad_norm": 0.3649786114692688,
+      "learning_rate": 1.961753568817896e-05,
+      "loss": 0.2854,
+      "step": 92
+    },
+    {
+      "epoch": 0.3799795709908069,
+      "grad_norm": 0.38573023676872253,
+      "learning_rate": 1.9605181116313725e-05,
+      "loss": 0.2667,
+      "step": 93
+    },
+    {
+      "epoch": 0.3840653728294178,
+      "grad_norm": 0.37577807903289795,
+      "learning_rate": 1.9592634177544803e-05,
+      "loss": 0.2815,
+      "step": 94
+    },
+    {
+      "epoch": 0.3881511746680286,
+      "grad_norm": 0.4320047199726105,
+      "learning_rate": 1.957989512315489e-05,
+      "loss": 0.3094,
+      "step": 95
+    },
+    {
+      "epoch": 0.3922369765066394,
+      "grad_norm": 0.3816889524459839,
+      "learning_rate": 1.9566964208274254e-05,
+      "loss": 0.292,
+      "step": 96
+    },
+    {
+      "epoch": 0.39632277834525026,
+      "grad_norm": 0.3946669399738312,
+      "learning_rate": 1.9553841691875632e-05,
+      "loss": 0.3002,
+      "step": 97
+    },
+    {
+      "epoch": 0.4004085801838611,
+      "grad_norm": 0.36885613203048706,
+      "learning_rate": 1.9540527836769047e-05,
+      "loss": 0.2583,
+      "step": 98
+    },
+    {
+      "epoch": 0.4044943820224719,
+      "grad_norm": 0.37865176796913147,
+      "learning_rate": 1.9527022909596537e-05,
+      "loss": 0.2787,
+      "step": 99
+    },
+    {
+      "epoch": 0.40858018386108275,
+      "grad_norm": 0.4429585337638855,
+      "learning_rate": 1.951332718082682e-05,
+      "loss": 0.3226,
+      "step": 100
+    },
+    {
+      "epoch": 0.41266598569969354,
+      "grad_norm": 0.3926009237766266,
+      "learning_rate": 1.9499440924749878e-05,
+      "loss": 0.2914,
+      "step": 101
+    },
+    {
+      "epoch": 0.4167517875383044,
+      "grad_norm": 0.3467339277267456,
+      "learning_rate": 1.9485364419471454e-05,
+      "loss": 0.266,
+      "step": 102
+    },
+    {
+      "epoch": 0.42083758937691523,
+      "grad_norm": 0.4126642644405365,
+      "learning_rate": 1.9471097946907506e-05,
+      "loss": 0.2775,
+      "step": 103
+    },
+    {
+      "epoch": 0.424923391215526,
+      "grad_norm": 0.44586020708084106,
+      "learning_rate": 1.9456641792778527e-05,
+      "loss": 0.2884,
+      "step": 104
+    },
+    {
+      "epoch": 0.4290091930541369,
+      "grad_norm": 0.3969588279724121,
+      "learning_rate": 1.9441996246603848e-05,
+      "loss": 0.2835,
+      "step": 105
+    },
+    {
+      "epoch": 0.4330949948927477,
+      "grad_norm": 0.38928356766700745,
+      "learning_rate": 1.9427161601695833e-05,
+      "loss": 0.2826,
+      "step": 106
+    },
+    {
+      "epoch": 0.4371807967313585,
+      "grad_norm": 0.4089799225330353,
+      "learning_rate": 1.9412138155154e-05,
+      "loss": 0.2817,
+      "step": 107
+    },
+    {
+      "epoch": 0.44126659856996936,
+      "grad_norm": 0.375505656003952,
+      "learning_rate": 1.9396926207859085e-05,
+      "loss": 0.2882,
+      "step": 108
+    },
+    {
+      "epoch": 0.4453524004085802,
+      "grad_norm": 0.406118780374527,
+      "learning_rate": 1.9381526064466995e-05,
+      "loss": 0.2861,
+      "step": 109
+    },
+    {
+      "epoch": 0.449438202247191,
+      "grad_norm": 0.3882409334182739,
+      "learning_rate": 1.9365938033402715e-05,
+      "loss": 0.261,
+      "step": 110
+    },
+    {
+      "epoch": 0.45352400408580185,
+      "grad_norm": 0.4351583421230316,
+      "learning_rate": 1.9350162426854152e-05,
+      "loss": 0.3014,
+      "step": 111
+    },
+    {
+      "epoch": 0.4576098059244127,
+      "grad_norm": 0.3621097505092621,
+      "learning_rate": 1.933419956076584e-05,
+      "loss": 0.2728,
+      "step": 112
+    },
+    {
+      "epoch": 0.4616956077630235,
+      "grad_norm": 0.3881032466888428,
+      "learning_rate": 1.9318049754832656e-05,
+      "loss": 0.2736,
+      "step": 113
+    },
+    {
+      "epoch": 0.46578140960163433,
+      "grad_norm": 0.37627285718917847,
+      "learning_rate": 1.9301713332493386e-05,
+      "loss": 0.2707,
+      "step": 114
+    },
+    {
+      "epoch": 0.4698672114402451,
+      "grad_norm": 0.4285913109779358,
+      "learning_rate": 1.9285190620924267e-05,
+      "loss": 0.2815,
+      "step": 115
+    },
+    {
+      "epoch": 0.473953013278856,
+      "grad_norm": 0.35718926787376404,
+      "learning_rate": 1.926848195103242e-05,
+      "loss": 0.2621,
+      "step": 116
+    },
+    {
+      "epoch": 0.4780388151174668,
+      "grad_norm": 0.3852044641971588,
+      "learning_rate": 1.925158765744924e-05,
+      "loss": 0.283,
+      "step": 117
+    },
+    {
+      "epoch": 0.4821246169560776,
+      "grad_norm": 0.3884032368659973,
+      "learning_rate": 1.923450807852367e-05,
+      "loss": 0.2711,
+      "step": 118
+    },
+    {
+      "epoch": 0.48621041879468846,
+      "grad_norm": 0.4398249685764313,
+      "learning_rate": 1.9217243556315445e-05,
+      "loss": 0.2757,
+      "step": 119
+    },
+    {
+      "epoch": 0.4902962206332993,
+      "grad_norm": 0.36689624190330505,
+      "learning_rate": 1.9199794436588244e-05,
+      "loss": 0.2669,
+      "step": 120
+    },
+    {
+      "epoch": 0.4943820224719101,
+      "grad_norm": 0.46398666501045227,
+      "learning_rate": 1.9182161068802742e-05,
+      "loss": 0.2683,
+      "step": 121
+    },
+    {
+      "epoch": 0.49846782431052095,
+      "grad_norm": 0.40020987391471863,
+      "learning_rate": 1.916434380610963e-05,
+      "loss": 0.2927,
+      "step": 122
+    },
+    {
+      "epoch": 0.5025536261491318,
+      "grad_norm": 0.4032459259033203,
+      "learning_rate": 1.9146343005342546e-05,
+      "loss": 0.31,
+      "step": 123
+    },
+    {
+      "epoch": 0.5066394279877426,
+      "grad_norm": 0.44166550040245056,
+      "learning_rate": 1.912815902701091e-05,
+      "loss": 0.2842,
+      "step": 124
+    },
+    {
+      "epoch": 0.5107252298263534,
+      "grad_norm": 0.39895153045654297,
+      "learning_rate": 1.9109792235292715e-05,
+      "loss": 0.2766,
+      "step": 125
+    },
+    {
+      "epoch": 0.5148110316649642,
+      "grad_norm": 0.3415013253688812,
+      "learning_rate": 1.909124299802724e-05,
+      "loss": 0.2761,
+      "step": 126
+    },
+    {
+      "epoch": 0.5188968335035751,
+      "grad_norm": 0.3837663531303406,
+      "learning_rate": 1.9072511686707663e-05,
+      "loss": 0.2797,
+      "step": 127
+    },
+    {
+      "epoch": 0.5229826353421859,
+      "grad_norm": 0.4030819833278656,
+      "learning_rate": 1.9053598676473656e-05,
+      "loss": 0.2932,
+      "step": 128
+    },
+    {
+      "epoch": 0.5270684371807968,
+      "grad_norm": 0.40120938420295715,
+      "learning_rate": 1.9034504346103825e-05,
+      "loss": 0.2698,
+      "step": 129
+    },
+    {
+      "epoch": 0.5311542390194075,
+      "grad_norm": 0.3621327579021454,
+      "learning_rate": 1.9015229078008163e-05,
+      "loss": 0.298,
+      "step": 130
+    },
+    {
+      "epoch": 0.5352400408580184,
+      "grad_norm": 0.33476150035858154,
+      "learning_rate": 1.8995773258220374e-05,
+      "loss": 0.2612,
+      "step": 131
+    },
+    {
+      "epoch": 0.5393258426966292,
+      "grad_norm": 0.3523140549659729,
+      "learning_rate": 1.8976137276390145e-05,
+      "loss": 0.2671,
+      "step": 132
+    },
+    {
+      "epoch": 0.54341164453524,
+      "grad_norm": 0.3624558746814728,
+      "learning_rate": 1.8956321525775337e-05,
+      "loss": 0.2687,
+      "step": 133
+    },
+    {
+      "epoch": 0.5474974463738509,
+      "grad_norm": 0.35892072319984436,
+      "learning_rate": 1.8936326403234125e-05,
+      "loss": 0.2755,
+      "step": 134
+    },
+    {
+      "epoch": 0.5515832482124617,
+      "grad_norm": 0.3678256869316101,
+      "learning_rate": 1.891615230921703e-05,
+      "loss": 0.278,
+      "step": 135
+    },
+    {
+      "epoch": 0.5556690500510725,
+      "grad_norm": 0.38125160336494446,
+      "learning_rate": 1.8895799647758912e-05,
+      "loss": 0.2765,
+      "step": 136
+    },
+    {
+      "epoch": 0.5597548518896833,
+      "grad_norm": 0.40152257680892944,
+      "learning_rate": 1.8875268826470875e-05,
+      "loss": 0.3239,
+      "step": 137
+    },
+    {
+      "epoch": 0.5638406537282942,
+      "grad_norm": 0.3935178816318512,
+      "learning_rate": 1.8854560256532098e-05,
+      "loss": 0.2956,
+      "step": 138
+    },
+    {
+      "epoch": 0.567926455566905,
+      "grad_norm": 0.4389478266239166,
+      "learning_rate": 1.8833674352681613e-05,
+      "loss": 0.2968,
+      "step": 139
+    },
+    {
+      "epoch": 0.5720122574055159,
+      "grad_norm": 0.3884355127811432,
+      "learning_rate": 1.881261153320999e-05,
+      "loss": 0.3074,
+      "step": 140
+    },
+    {
+      "epoch": 0.5760980592441267,
+      "grad_norm": 0.4054373502731323,
+      "learning_rate": 1.879137221995095e-05,
+      "loss": 0.2996,
+      "step": 141
+    },
+    {
+      "epoch": 0.5801838610827375,
+      "grad_norm": 0.4423893690109253,
+      "learning_rate": 1.8769956838272937e-05,
+      "loss": 0.3082,
+      "step": 142
+    },
+    {
+      "epoch": 0.5842696629213483,
+      "grad_norm": 0.42978307604789734,
+      "learning_rate": 1.8748365817070586e-05,
+      "loss": 0.2878,
+      "step": 143
+    },
+    {
+      "epoch": 0.5883554647599591,
+      "grad_norm": 0.38182228803634644,
+      "learning_rate": 1.8726599588756144e-05,
+      "loss": 0.2649,
+      "step": 144
+    },
+    {
+      "epoch": 0.59244126659857,
+      "grad_norm": 0.43477413058280945,
+      "learning_rate": 1.8704658589250795e-05,
+      "loss": 0.271,
+      "step": 145
+    },
+    {
+      "epoch": 0.5965270684371808,
+      "grad_norm": 0.3876926898956299,
+      "learning_rate": 1.868254325797594e-05,
+      "loss": 0.2804,
+      "step": 146
+    },
+    {
+      "epoch": 0.6006128702757916,
+      "grad_norm": 0.39310601353645325,
+      "learning_rate": 1.866025403784439e-05,
+      "loss": 0.2767,
+      "step": 147
+    },
+    {
+      "epoch": 0.6046986721144024,
+      "grad_norm": 0.421290785074234,
+      "learning_rate": 1.8637791375251505e-05,
+      "loss": 0.2668,
+      "step": 148
+    },
+    {
+      "epoch": 0.6087844739530133,
+      "grad_norm": 0.450023353099823,
+      "learning_rate": 1.8615155720066247e-05,
+      "loss": 0.2888,
+      "step": 149
+    },
+    {
+      "epoch": 0.6128702757916241,
+      "grad_norm": 0.3645341396331787,
+      "learning_rate": 1.859234752562217e-05,
+      "loss": 0.2828,
+      "step": 150
+    },
+    {
+      "epoch": 0.616956077630235,
+      "grad_norm": 0.41853606700897217,
+      "learning_rate": 1.8569367248708343e-05,
+      "loss": 0.284,
+      "step": 151
+    },
+    {
+      "epoch": 0.6210418794688458,
+      "grad_norm": 0.3675737679004669,
+      "learning_rate": 1.8546215349560204e-05,
+      "loss": 0.2933,
+      "step": 152
+    },
+    {
+      "epoch": 0.6251276813074566,
+      "grad_norm": 0.3668256998062134,
+      "learning_rate": 1.8522892291850335e-05,
+      "loss": 0.2729,
+      "step": 153
+    },
+    {
+      "epoch": 0.6292134831460674,
+      "grad_norm": 0.34576019644737244,
+      "learning_rate": 1.849939854267919e-05,
+      "loss": 0.2612,
+      "step": 154
+    },
+    {
+      "epoch": 0.6332992849846782,
+      "grad_norm": 0.41370126605033875,
+      "learning_rate": 1.847573457256571e-05,
+      "loss": 0.2693,
+      "step": 155
+    },
+    {
+      "epoch": 0.6373850868232891,
+      "grad_norm": 0.4205566644668579,
+      "learning_rate": 1.845190085543795e-05,
+      "loss": 0.2746,
+      "step": 156
+    },
+    {
+      "epoch": 0.6414708886618999,
+      "grad_norm": 0.3997614085674286,
+      "learning_rate": 1.8427897868623535e-05,
+      "loss": 0.2813,
+      "step": 157
+    },
+    {
+      "epoch": 0.6455566905005107,
+      "grad_norm": 0.41005200147628784,
+      "learning_rate": 1.840372609284013e-05,
+      "loss": 0.2647,
+      "step": 158
+    },
+    {
+      "epoch": 0.6496424923391215,
+      "grad_norm": 0.4547550678253174,
+      "learning_rate": 1.8379386012185813e-05,
+      "loss": 0.2791,
+      "step": 159
+    },
+    {
+      "epoch": 0.6537282941777324,
+      "grad_norm": 0.4075047969818115,
+      "learning_rate": 1.8354878114129368e-05,
+      "loss": 0.2769,
+      "step": 160
+    },
+    {
+      "epoch": 0.6578140960163432,
+      "grad_norm": 0.37060046195983887,
+      "learning_rate": 1.8330202889500518e-05,
+      "loss": 0.3028,
+      "step": 161
+    },
+    {
+      "epoch": 0.6618998978549541,
+      "grad_norm": 0.35541340708732605,
+      "learning_rate": 1.8305360832480118e-05,
+      "loss": 0.2981,
+      "step": 162
+    },
+    {
+      "epoch": 0.6659856996935649,
+      "grad_norm": 0.3970625400543213,
+      "learning_rate": 1.8280352440590236e-05,
+      "loss": 0.2634,
+      "step": 163
+    },
+    {
+      "epoch": 0.6700715015321757,
+      "grad_norm": 0.4075865149497986,
+      "learning_rate": 1.82551782146842e-05,
+      "loss": 0.3027,
+      "step": 164
+    },
+    {
+      "epoch": 0.6700715015321757,
+      "eval_loss": 0.291363924741745,
+      "eval_runtime": 5.7936,
+      "eval_samples_per_second": 13.636,
+      "eval_steps_per_second": 1.726,
+      "step": 164
+    },
+    {
+      "epoch": 0.6741573033707865,
+      "grad_norm": 0.34390076994895935,
+      "learning_rate": 1.8229838658936566e-05,
+      "loss": 0.2536,
+      "step": 165
+    },
+    {
+      "epoch": 0.6782431052093973,
+      "grad_norm": 0.3729197084903717,
+      "learning_rate": 1.8204334280833005e-05,
+      "loss": 0.2739,
+      "step": 166
+    },
+    {
+      "epoch": 0.6823289070480082,
+      "grad_norm": 0.3974601924419403,
+      "learning_rate": 1.817866559116017e-05,
+      "loss": 0.2858,
+      "step": 167
+    },
+    {
+      "epoch": 0.686414708886619,
+      "grad_norm": 0.3424644470214844,
+      "learning_rate": 1.8152833103995443e-05,
+      "loss": 0.2305,
+      "step": 168
+    },
+    {
+      "epoch": 0.6905005107252298,
+      "grad_norm": 0.4293709397315979,
+      "learning_rate": 1.8126837336696645e-05,
+      "loss": 0.3179,
+      "step": 169
+    },
+    {
+      "epoch": 0.6945863125638406,
+      "grad_norm": 0.3259459435939789,
+      "learning_rate": 1.8100678809891668e-05,
+      "loss": 0.2589,
+      "step": 170
+    },
+    {
+      "epoch": 0.6986721144024515,
+      "grad_norm": 0.40771302580833435,
+      "learning_rate": 1.807435804746807e-05,
+      "loss": 0.2637,
+      "step": 171
+    },
+    {
+      "epoch": 0.7027579162410623,
+      "grad_norm": 0.3847212493419647,
+      "learning_rate": 1.8047875576562556e-05,
+      "loss": 0.2782,
+      "step": 172
+    },
+    {
+      "epoch": 0.7068437180796732,
+      "grad_norm": 0.35547974705696106,
+      "learning_rate": 1.802123192755044e-05,
+      "loss": 0.2695,
+      "step": 173
+    },
+    {
+      "epoch": 0.710929519918284,
+      "grad_norm": 0.3954298198223114,
+      "learning_rate": 1.7994427634035016e-05,
+      "loss": 0.3005,
+      "step": 174
+    },
+    {
+      "epoch": 0.7150153217568948,
+      "grad_norm": 0.3506409525871277,
+      "learning_rate": 1.796746323283686e-05,
+      "loss": 0.2716,
+      "step": 175
+    },
+    {
+      "epoch": 0.7191011235955056,
+      "grad_norm": 0.42227277159690857,
+      "learning_rate": 1.7940339263983112e-05,
+      "loss": 0.2915,
+      "step": 176
+    },
+    {
+      "epoch": 0.7231869254341164,
+      "grad_norm": 0.3948259949684143,
+      "learning_rate": 1.791305627069662e-05,
+      "loss": 0.2883,
+      "step": 177
+    },
+    {
+      "epoch": 0.7272727272727273,
+      "grad_norm": 0.3580792248249054,
+      "learning_rate": 1.7885614799385086e-05,
+      "loss": 0.2782,
+      "step": 178
+    },
+    {
+      "epoch": 0.7313585291113381,
+      "grad_norm": 0.39698660373687744,
+      "learning_rate": 1.785801539963012e-05,
+      "loss": 0.2657,
+      "step": 179
+    },
+    {
+      "epoch": 0.7354443309499489,
+      "grad_norm": 0.3663792610168457,
+      "learning_rate": 1.7830258624176224e-05,
+      "loss": 0.2686,
+      "step": 180
+    },
+    {
+      "epoch": 0.7395301327885597,
+      "grad_norm": 0.38216930627822876,
+      "learning_rate": 1.7802345028919728e-05,
+      "loss": 0.2706,
+      "step": 181
+    },
+    {
+      "epoch": 0.7436159346271706,
+      "grad_norm": 0.4187450706958771,
+      "learning_rate": 1.777427517289766e-05,
+      "loss": 0.2573,
+      "step": 182
+    },
+    {
+      "epoch": 0.7477017364657814,
+      "grad_norm": 0.34619036316871643,
+      "learning_rate": 1.7746049618276545e-05,
+      "loss": 0.269,
+      "step": 183
+    },
+    {
+      "epoch": 0.7517875383043923,
+      "grad_norm": 0.35370582342147827,
+      "learning_rate": 1.7717668930341152e-05,
+      "loss": 0.2552,
+      "step": 184
+    },
+    {
+      "epoch": 0.7558733401430031,
+      "grad_norm": 0.4264880418777466,
+      "learning_rate": 1.768913367748316e-05,
+      "loss": 0.2952,
+      "step": 185
+    },
+    {
+      "epoch": 0.7599591419816139,
+      "grad_norm": 0.39135676622390747,
+      "learning_rate": 1.766044443118978e-05,
+      "loss": 0.2661,
+      "step": 186
+    },
+    {
+      "epoch": 0.7640449438202247,
+      "grad_norm": 0.39061596989631653,
+      "learning_rate": 1.7631601766032337e-05,
+      "loss": 0.2737,
+      "step": 187
+    },
+    {
+      "epoch": 0.7681307456588355,
+      "grad_norm": 0.3799816966056824,
+      "learning_rate": 1.7602606259654704e-05,
+      "loss": 0.2767,
+      "step": 188
+    },
+    {
+      "epoch": 0.7722165474974464,
+      "grad_norm": 0.3592148721218109,
+      "learning_rate": 1.7573458492761802e-05,
+      "loss": 0.2448,
+      "step": 189
+    },
+    {
+      "epoch": 0.7763023493360572,
+      "grad_norm": 0.39084604382514954,
+      "learning_rate": 1.7544159049107902e-05,
+      "loss": 0.275,
+      "step": 190
+    },
+    {
+      "epoch": 0.780388151174668,
+      "grad_norm": 0.36443451046943665,
+      "learning_rate": 1.7514708515485002e-05,
+      "loss": 0.2645,
+      "step": 191
+    },
+    {
+      "epoch": 0.7844739530132788,
+      "grad_norm": 0.4001200497150421,
+      "learning_rate": 1.7485107481711014e-05,
+      "loss": 0.2724,
+      "step": 192
+    },
+    {
+      "epoch": 0.7885597548518897,
+      "grad_norm": 0.39093396067619324,
+      "learning_rate": 1.7455356540617988e-05,
+      "loss": 0.2712,
+      "step": 193
+    },
+    {
+      "epoch": 0.7926455566905005,
+      "grad_norm": 0.3430577218532562,
+      "learning_rate": 1.7425456288040236e-05,
+      "loss": 0.2489,
+      "step": 194
+    },
+    {
+      "epoch": 0.7967313585291114,
+      "grad_norm": 0.3573733866214752,
+      "learning_rate": 1.7395407322802374e-05,
+      "loss": 0.2696,
+      "step": 195
+    },
+    {
+      "epoch": 0.8008171603677222,
+      "grad_norm": 0.38158077001571655,
+      "learning_rate": 1.736521024670737e-05,
+      "loss": 0.2814,
+      "step": 196
+    },
+    {
+      "epoch": 0.804902962206333,
+      "grad_norm": 0.366470068693161,
+      "learning_rate": 1.733486566452446e-05,
+      "loss": 0.2529,
+      "step": 197
+    },
+    {
+      "epoch": 0.8089887640449438,
+      "grad_norm": 0.3718278408050537,
+      "learning_rate": 1.7304374183977032e-05,
+      "loss": 0.2747,
+      "step": 198
+    },
+    {
+      "epoch": 0.8130745658835546,
+      "grad_norm": 0.3395809233188629,
+      "learning_rate": 1.7273736415730488e-05,
+      "loss": 0.2693,
+      "step": 199
+    },
+    {
+      "epoch": 0.8171603677221655,
+      "grad_norm": 0.307731032371521,
+      "learning_rate": 1.7242952973379983e-05,
+      "loss": 0.2081,
+      "step": 200
+    },
+    {
+      "epoch": 0.8212461695607763,
+      "grad_norm": 0.3522433936595917,
+      "learning_rate": 1.7212024473438145e-05,
+      "loss": 0.2495,
+      "step": 201
+    },
+    {
+      "epoch": 0.8253319713993871,
+      "grad_norm": 0.35946980118751526,
+      "learning_rate": 1.7180951535322742e-05,
+      "loss": 0.2702,
+      "step": 202
+    },
+    {
+      "epoch": 0.8294177732379979,
+      "grad_norm": 0.3933047950267792,
+      "learning_rate": 1.7149734781344247e-05,
+      "loss": 0.2629,
+      "step": 203
+    },
+    {
+      "epoch": 0.8335035750766088,
+      "grad_norm": 0.3658384084701538,
+      "learning_rate": 1.7118374836693407e-05,
+      "loss": 0.2538,
+      "step": 204
+    },
+    {
+      "epoch": 0.8375893769152196,
+      "grad_norm": 0.3532220423221588,
+      "learning_rate": 1.7086872329428702e-05,
+      "loss": 0.2587,
+      "step": 205
+    },
+    {
+      "epoch": 0.8416751787538305,
+      "grad_norm": 0.3619686961174011,
+      "learning_rate": 1.705522789046377e-05,
+      "loss": 0.2658,
+      "step": 206
+    },
+    {
+      "epoch": 0.8457609805924413,
+      "grad_norm": 0.4083801209926605,
+      "learning_rate": 1.7023442153554776e-05,
+      "loss": 0.2614,
+      "step": 207
+    },
+    {
+      "epoch": 0.849846782431052,
+      "grad_norm": 0.3868924081325531,
+      "learning_rate": 1.6991515755287715e-05,
+      "loss": 0.2831,
+      "step": 208
+    },
+    {
+      "epoch": 0.8539325842696629,
+      "grad_norm": 0.38413897156715393,
+      "learning_rate": 1.695944933506567e-05,
+      "loss": 0.2596,
+      "step": 209
+    },
+    {
+      "epoch": 0.8580183861082737,
+      "grad_norm": 0.34999531507492065,
+      "learning_rate": 1.6927243535095995e-05,
+      "loss": 0.2842,
+      "step": 210
+    },
+    {
+      "epoch": 0.8621041879468846,
+      "grad_norm": 0.328204482793808,
+      "learning_rate": 1.6894899000377462e-05,
+      "loss": 0.2332,
+      "step": 211
+    },
+    {
+      "epoch": 0.8661899897854954,
+      "grad_norm": 0.3802552819252014,
+      "learning_rate": 1.686241637868734e-05,
+      "loss": 0.2709,
+      "step": 212
+    },
+    {
+      "epoch": 0.8702757916241062,
+      "grad_norm": 0.35758858919143677,
+      "learning_rate": 1.6829796320568416e-05,
+      "loss": 0.279,
+      "step": 213
+    },
+    {
+      "epoch": 0.874361593462717,
+      "grad_norm": 0.3561984896659851,
+      "learning_rate": 1.6797039479315994e-05,
+      "loss": 0.2868,
+      "step": 214
+    },
+    {
+      "epoch": 0.8784473953013279,
+      "grad_norm": 0.32591065764427185,
+      "learning_rate": 1.6764146510964762e-05,
+      "loss": 0.2485,
+      "step": 215
+    },
+    {
+      "epoch": 0.8825331971399387,
+      "grad_norm": 0.36409640312194824,
+      "learning_rate": 1.67311180742757e-05,
+      "loss": 0.2577,
+      "step": 216
+    },
+    {
+      "epoch": 0.8866189989785496,
+      "grad_norm": 0.34685492515563965,
+      "learning_rate": 1.669795483072287e-05,
+      "loss": 0.247,
+      "step": 217
+    },
+    {
+      "epoch": 0.8907048008171604,
+      "grad_norm": 0.3445712625980377,
+      "learning_rate": 1.6664657444480145e-05,
+      "loss": 0.2565,
+      "step": 218
+    },
+    {
+      "epoch": 0.8947906026557712,
+      "grad_norm": 0.34710460901260376,
+      "learning_rate": 1.6631226582407954e-05,
+      "loss": 0.2363,
+      "step": 219
+    },
+    {
+      "epoch": 0.898876404494382,
+      "grad_norm": 0.33726766705513,
+      "learning_rate": 1.6597662914039885e-05,
+      "loss": 0.2483,
+      "step": 220
+    },
+    {
+      "epoch": 0.9029622063329928,
+      "grad_norm": 0.34024032950401306,
+      "learning_rate": 1.65639671115693e-05,
+      "loss": 0.2474,
+      "step": 221
+    },
+    {
+      "epoch": 0.9070480081716037,
+      "grad_norm": 0.38807395100593567,
+      "learning_rate": 1.653013984983585e-05,
+      "loss": 0.2726,
+      "step": 222
+    },
+    {
+      "epoch": 0.9111338100102145,
+      "grad_norm": 0.36375290155410767,
+      "learning_rate": 1.6496181806312005e-05,
+      "loss": 0.2726,
+      "step": 223
+    },
+    {
+      "epoch": 0.9152196118488254,
+      "grad_norm": 0.36927178502082825,
+      "learning_rate": 1.6462093661089432e-05,
+      "loss": 0.2518,
+      "step": 224
+    },
+    {
+      "epoch": 0.9193054136874361,
+      "grad_norm": 0.3809269070625305,
+      "learning_rate": 1.6427876096865394e-05,
+      "loss": 0.2449,
+      "step": 225
+    },
+    {
+      "epoch": 0.923391215526047,
+      "grad_norm": 0.34634968638420105,
+      "learning_rate": 1.6393529798929103e-05,
+      "loss": 0.2575,
+      "step": 226
+    },
+    {
+      "epoch": 0.9274770173646578,
+      "grad_norm": 0.33054831624031067,
+      "learning_rate": 1.635905545514795e-05,
+      "loss": 0.2639,
+      "step": 227
+    },
+    {
+      "epoch": 0.9315628192032687,
+      "grad_norm": 0.35482174158096313,
+      "learning_rate": 1.6324453755953772e-05,
+      "loss": 0.2667,
+      "step": 228
+    },
+    {
+      "epoch": 0.9356486210418795,
+      "grad_norm": 0.3657509684562683,
+      "learning_rate": 1.6289725394328998e-05,
+      "loss": 0.255,
+      "step": 229
+    },
+    {
+      "epoch": 0.9397344228804902,
+      "grad_norm": 0.3343275785446167,
+      "learning_rate": 1.6254871065792776e-05,
+      "loss": 0.2336,
+      "step": 230
+    },
+    {
+      "epoch": 0.9438202247191011,
+      "grad_norm": 0.3493170142173767,
+      "learning_rate": 1.621989146838704e-05,
+      "loss": 0.2649,
+      "step": 231
+    },
+    {
+      "epoch": 0.947906026557712,
+      "grad_norm": 0.3305867612361908,
+      "learning_rate": 1.618478730266255e-05,
+      "loss": 0.2767,
+      "step": 232
+    },
+    {
+      "epoch": 0.9519918283963228,
+      "grad_norm": 0.35817259550094604,
+      "learning_rate": 1.6149559271664835e-05,
+      "loss": 0.2817,
+      "step": 233
+    },
+    {
+      "epoch": 0.9560776302349336,
+      "grad_norm": 0.37733370065689087,
+      "learning_rate": 1.6114208080920125e-05,
+      "loss": 0.2809,
+      "step": 234
+    },
+    {
+      "epoch": 0.9601634320735445,
+      "grad_norm": 0.3227766156196594,
+      "learning_rate": 1.607873443842122e-05,
+      "loss": 0.2545,
+      "step": 235
+    },
+    {
+      "epoch": 0.9642492339121552,
+      "grad_norm": 0.3445710241794586,
+      "learning_rate": 1.6043139054613326e-05,
+      "loss": 0.2476,
+      "step": 236
+    },
+    {
+      "epoch": 0.9683350357507661,
+      "grad_norm": 0.3375508785247803,
+      "learning_rate": 1.600742264237979e-05,
+      "loss": 0.2502,
+      "step": 237
+    },
+    {
+      "epoch": 0.9724208375893769,
+      "grad_norm": 0.356039434671402,
+      "learning_rate": 1.5971585917027864e-05,
+      "loss": 0.268,
+      "step": 238
+    },
+    {
+      "epoch": 0.9765066394279878,
+      "grad_norm": 0.34852373600006104,
+      "learning_rate": 1.5935629596274345e-05,
+      "loss": 0.2605,
+      "step": 239
+    },
+    {
+      "epoch": 0.9805924412665986,
+      "grad_norm": 0.3376101851463318,
+      "learning_rate": 1.5899554400231233e-05,
+      "loss": 0.2567,
+      "step": 240
+    },
+    {
+      "epoch": 0.9846782431052093,
+      "grad_norm": 0.32361170649528503,
+      "learning_rate": 1.586336105139127e-05,
+      "loss": 0.2481,
+      "step": 241
+    },
+    {
+      "epoch": 0.9887640449438202,
+      "grad_norm": 0.35558903217315674,
+      "learning_rate": 1.5827050274613512e-05,
+      "loss": 0.2514,
+      "step": 242
+    },
+    {
+      "epoch": 0.992849846782431,
+      "grad_norm": 0.31636619567871094,
+      "learning_rate": 1.579062279710879e-05,
+      "loss": 0.2237,
+      "step": 243
+    },
+    {
+      "epoch": 0.9969356486210419,
+      "grad_norm": 0.3540779948234558,
+      "learning_rate": 1.5754079348425137e-05,
+      "loss": 0.2381,
+      "step": 244
+    },
+    {
+      "epoch": 1.0040858018386107,
+      "grad_norm": 0.7127255201339722,
+      "learning_rate": 1.57174206604332e-05,
+      "loss": 0.4477,
+      "step": 245
+    },
+    {
+      "epoch": 1.0081716036772217,
+      "grad_norm": 0.21768411993980408,
+      "learning_rate": 1.568064746731156e-05,
+      "loss": 0.177,
+      "step": 246
+    },
+    {
+      "epoch": 1.0081716036772217,
+      "eval_loss": 0.2854033410549164,
+      "eval_runtime": 5.5756,
+      "eval_samples_per_second": 14.169,
+      "eval_steps_per_second": 1.794,
+      "step": 246
+    },
+    {
+      "epoch": 1.0122574055158324,
+      "grad_norm": 0.24506381154060364,
+      "learning_rate": 1.564376050553205e-05,
+      "loss": 0.1647,
+      "step": 247
+    },
+    {
+      "epoch": 1.0163432073544434,
+      "grad_norm": 0.24179627001285553,
+      "learning_rate": 1.560676051384499e-05,
+      "loss": 0.1908,
+      "step": 248
+    },
+    {
+      "epoch": 1.0204290091930541,
+      "grad_norm": 0.2527990937232971,
+      "learning_rate": 1.5569648233264395e-05,
+      "loss": 0.1728,
+      "step": 249
+    },
+    {
+      "epoch": 1.0245148110316649,
+      "grad_norm": 0.28597134351730347,
+      "learning_rate": 1.553242440705314e-05,
+      "loss": 0.1854,
+      "step": 250
+    },
+    {
+      "epoch": 1.0286006128702758,
+      "grad_norm": 0.2613103985786438,
+      "learning_rate": 1.5495089780708062e-05,
+      "loss": 0.1762,
+      "step": 251
+    },
+    {
+      "epoch": 1.0326864147088866,
+      "grad_norm": 0.2806336581707001,
+      "learning_rate": 1.5457645101945046e-05,
+      "loss": 0.1801,
+      "step": 252
+    },
+    {
+      "epoch": 1.0367722165474975,
+      "grad_norm": 0.29933255910873413,
+      "learning_rate": 1.5420091120684042e-05,
+      "loss": 0.1869,
+      "step": 253
+    },
+    {
+      "epoch": 1.0408580183861083,
+      "grad_norm": 0.2678683400154114,
+      "learning_rate": 1.538242858903404e-05,
+      "loss": 0.1684,
+      "step": 254
+    },
+    {
+      "epoch": 1.0449438202247192,
+      "grad_norm": 0.27515852451324463,
+      "learning_rate": 1.5344658261278013e-05,
+      "loss": 0.1859,
+      "step": 255
+    },
+    {
+      "epoch": 1.04902962206333,
+      "grad_norm": 0.2876634895801544,
+      "learning_rate": 1.530678089385782e-05,
+      "loss": 0.1705,
+      "step": 256
+    },
+    {
+      "epoch": 1.0531154239019407,
+      "grad_norm": 0.2911262810230255,
+      "learning_rate": 1.5268797245359035e-05,
+      "loss": 0.1937,
+      "step": 257
+    },
+    {
+      "epoch": 1.0572012257405516,
+      "grad_norm": 0.3048553466796875,
+      "learning_rate": 1.5230708076495777e-05,
+      "loss": 0.1882,
+      "step": 258
+    },
+    {
+      "epoch": 1.0612870275791624,
+      "grad_norm": 0.28508955240249634,
+      "learning_rate": 1.519251415009546e-05,
+      "loss": 0.1767,
+      "step": 259
+    },
+    {
+      "epoch": 1.0653728294177733,
+      "grad_norm": 0.266313374042511,
+      "learning_rate": 1.5154216231083522e-05,
+      "loss": 0.1647,
+      "step": 260
+    },
+    {
+      "epoch": 1.069458631256384,
+      "grad_norm": 0.2724918723106384,
+      "learning_rate": 1.5115815086468103e-05,
+      "loss": 0.1685,
+      "step": 261
+    },
+    {
+      "epoch": 1.0735444330949948,
+      "grad_norm": 0.2324502021074295,
+      "learning_rate": 1.507731148532468e-05,
+      "loss": 0.1632,
+      "step": 262
+    },
+    {
+      "epoch": 1.0776302349336058,
+      "grad_norm": 0.26865899562835693,
+      "learning_rate": 1.5038706198780673e-05,
+      "loss": 0.1802,
+      "step": 263
+    },
+    {
+      "epoch": 1.0817160367722165,
+      "grad_norm": 0.29491883516311646,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.1803,
+      "step": 264
+    },
+    {
+      "epoch": 1.0858018386108275,
+      "grad_norm": 0.28987348079681396,
+      "learning_rate": 1.496119366416759e-05,
+      "loss": 0.1862,
+      "step": 265
+    },
+    {
+      "epoch": 1.0898876404494382,
+      "grad_norm": 0.27755048871040344,
+      "learning_rate": 1.492228796847385e-05,
+      "loss": 0.1741,
+      "step": 266
+    },
+    {
+      "epoch": 1.093973442288049,
+      "grad_norm": 0.2608552873134613,
+      "learning_rate": 1.4883283692099114e-05,
+      "loss": 0.1693,
+      "step": 267
+    },
+    {
+      "epoch": 1.09805924412666,
+      "grad_norm": 0.27284783124923706,
+      "learning_rate": 1.4844181616198028e-05,
+      "loss": 0.1878,
+      "step": 268
+    },
+    {
+      "epoch": 1.1021450459652706,
+      "grad_norm": 0.24481667578220367,
+      "learning_rate": 1.4804982523883915e-05,
+      "loss": 0.1589,
+      "step": 269
+    },
+    {
+      "epoch": 1.1062308478038816,
+      "grad_norm": 0.2996629774570465,
+      "learning_rate": 1.4765687200213079e-05,
+      "loss": 0.1823,
+      "step": 270
+    },
+    {
+      "epoch": 1.1103166496424923,
+      "grad_norm": 0.2922385632991791,
+      "learning_rate": 1.4726296432169095e-05,
+      "loss": 0.1769,
+      "step": 271
+    },
+    {
+      "epoch": 1.1144024514811033,
+      "grad_norm": 0.3046974241733551,
+      "learning_rate": 1.4686811008647037e-05,
+      "loss": 0.1823,
+      "step": 272
+    },
+    {
+      "epoch": 1.118488253319714,
+      "grad_norm": 0.2792796790599823,
+      "learning_rate": 1.4647231720437687e-05,
+      "loss": 0.1717,
+      "step": 273
+    },
+    {
+      "epoch": 1.1225740551583248,
+      "grad_norm": 0.27251774072647095,
+      "learning_rate": 1.4607559360211688e-05,
+      "loss": 0.1652,
+      "step": 274
+    },
+    {
+      "epoch": 1.1266598569969357,
+      "grad_norm": 0.2751109302043915,
+      "learning_rate": 1.456779472250368e-05,
+      "loss": 0.1713,
+      "step": 275
+    },
+    {
+      "epoch": 1.1307456588355465,
+      "grad_norm": 0.2737586796283722,
+      "learning_rate": 1.4527938603696376e-05,
+      "loss": 0.162,
+      "step": 276
+    },
+    {
+      "epoch": 1.1348314606741572,
+      "grad_norm": 0.24653682112693787,
+      "learning_rate": 1.4487991802004625e-05,
+      "loss": 0.1626,
+      "step": 277
+    },
+    {
+      "epoch": 1.1389172625127681,
+      "grad_norm": 0.46106576919555664,
+      "learning_rate": 1.4447955117459414e-05,
+      "loss": 0.1609,
+      "step": 278
+    },
+    {
+      "epoch": 1.1430030643513789,
+      "grad_norm": 0.27714091539382935,
+      "learning_rate": 1.4407829351891858e-05,
+      "loss": 0.1759,
+      "step": 279
+    },
+    {
+      "epoch": 1.1470888661899898,
+      "grad_norm": 0.2678029537200928,
+      "learning_rate": 1.436761530891713e-05,
+      "loss": 0.1753,
+      "step": 280
+    },
+    {
+      "epoch": 1.1511746680286006,
+      "grad_norm": 0.2559642791748047,
+      "learning_rate": 1.4327313793918362e-05,
+      "loss": 0.1778,
+      "step": 281
+    },
+    {
+      "epoch": 1.1552604698672115,
+      "grad_norm": 0.3033258616924286,
+      "learning_rate": 1.4286925614030542e-05,
+      "loss": 0.1871,
+      "step": 282
+    },
+    {
+      "epoch": 1.1593462717058223,
+      "grad_norm": 0.2658158540725708,
+      "learning_rate": 1.4246451578124321e-05,
+      "loss": 0.1782,
+      "step": 283
+    },
+    {
+      "epoch": 1.163432073544433,
+      "grad_norm": 0.2901168465614319,
+      "learning_rate": 1.4205892496789816e-05,
+      "loss": 0.174,
+      "step": 284
+    },
+    {
+      "epoch": 1.167517875383044,
+      "grad_norm": 0.23054322600364685,
+      "learning_rate": 1.4165249182320401e-05,
+      "loss": 0.1553,
+      "step": 285
+    },
+    {
+      "epoch": 1.1716036772216547,
+      "grad_norm": 0.267805278301239,
+      "learning_rate": 1.4124522448696407e-05,
+      "loss": 0.168,
+      "step": 286
+    },
+    {
+      "epoch": 1.1756894790602657,
+      "grad_norm": 0.26580214500427246,
+      "learning_rate": 1.4083713111568841e-05,
+      "loss": 0.167,
+      "step": 287
+    },
+    {
+      "epoch": 1.1797752808988764,
+      "grad_norm": 0.2736794948577881,
+      "learning_rate": 1.404282198824305e-05,
+      "loss": 0.1623,
+      "step": 288
+    },
+    {
+      "epoch": 1.1838610827374871,
+      "grad_norm": 0.25851017236709595,
+      "learning_rate": 1.4001849897662337e-05,
+      "loss": 0.1646,
+      "step": 289
+    },
+    {
+      "epoch": 1.187946884576098,
+      "grad_norm": 0.26858997344970703,
+      "learning_rate": 1.396079766039157e-05,
+      "loss": 0.1768,
+      "step": 290
+    },
+    {
+      "epoch": 1.1920326864147088,
+      "grad_norm": 0.2878361940383911,
+      "learning_rate": 1.3919666098600753e-05,
+      "loss": 0.1712,
+      "step": 291
+    },
+    {
+      "epoch": 1.1961184882533198,
+      "grad_norm": 0.23014627397060394,
+      "learning_rate": 1.387845603604855e-05,
+      "loss": 0.1595,
+      "step": 292
+    },
+    {
+      "epoch": 1.2002042900919305,
+      "grad_norm": 0.27550917863845825,
+      "learning_rate": 1.3837168298065798e-05,
+      "loss": 0.1639,
+      "step": 293
+    },
+    {
+      "epoch": 1.2042900919305413,
+      "grad_norm": 0.2697204053401947,
+      "learning_rate": 1.3795803711538966e-05,
+      "loss": 0.1619,
+      "step": 294
+    },
+    {
+      "epoch": 1.2083758937691522,
+      "grad_norm": 0.29666051268577576,
+      "learning_rate": 1.37543631048936e-05,
+      "loss": 0.1815,
+      "step": 295
+    },
+    {
+      "epoch": 1.212461695607763,
+      "grad_norm": 0.25596365332603455,
+      "learning_rate": 1.3712847308077737e-05,
+      "loss": 0.1629,
+      "step": 296
+    },
+    {
+      "epoch": 1.216547497446374,
+      "grad_norm": 0.25550931692123413,
+      "learning_rate": 1.3671257152545277e-05,
+      "loss": 0.1635,
+      "step": 297
+    },
+    {
+      "epoch": 1.2206332992849847,
+      "grad_norm": 0.2615107297897339,
+      "learning_rate": 1.3629593471239328e-05,
+      "loss": 0.1547,
+      "step": 298
+    },
+    {
+      "epoch": 1.2247191011235956,
+      "grad_norm": 0.2814185917377472,
+      "learning_rate": 1.3587857098575534e-05,
+      "loss": 0.1713,
+      "step": 299
+    },
+    {
+      "epoch": 1.2288049029622063,
+      "grad_norm": 0.2644117772579193,
+      "learning_rate": 1.3546048870425356e-05,
+      "loss": 0.1703,
+      "step": 300
+    },
+    {
+      "epoch": 1.232890704800817,
+      "grad_norm": 0.2645355463027954,
+      "learning_rate": 1.350416962409934e-05,
+      "loss": 0.159,
+      "step": 301
+    },
+    {
+      "epoch": 1.236976506639428,
+      "grad_norm": 0.2637065351009369,
+      "learning_rate": 1.346222019833033e-05,
+      "loss": 0.1647,
+      "step": 302
+    },
+    {
+      "epoch": 1.2410623084780388,
+      "grad_norm": 0.24007368087768555,
+      "learning_rate": 1.342020143325669e-05,
+      "loss": 0.1569,
+      "step": 303
+    },
+    {
+      "epoch": 1.2451481103166497,
+      "grad_norm": 0.2273741364479065,
+      "learning_rate": 1.3378114170405473e-05,
+      "loss": 0.1645,
+      "step": 304
+    },
+    {
+      "epoch": 1.2492339121552605,
+      "grad_norm": 0.2602927088737488,
+      "learning_rate": 1.3335959252675566e-05,
+      "loss": 0.1723,
+      "step": 305
+    },
+    {
+      "epoch": 1.2533197139938714,
+      "grad_norm": 0.28329333662986755,
+      "learning_rate": 1.3293737524320798e-05,
+      "loss": 0.1704,
+      "step": 306
+    },
+    {
+      "epoch": 1.2574055158324822,
+      "grad_norm": 0.270916610956192,
+      "learning_rate": 1.3251449830933052e-05,
+      "loss": 0.1621,
+      "step": 307
+    },
+    {
+      "epoch": 1.261491317671093,
+      "grad_norm": 0.268443763256073,
+      "learning_rate": 1.3209097019425317e-05,
+      "loss": 0.177,
+      "step": 308
+    },
+    {
+      "epoch": 1.2655771195097039,
+      "grad_norm": 0.2811964750289917,
+      "learning_rate": 1.3166679938014728e-05,
+      "loss": 0.1581,
+      "step": 309
+    },
+    {
+      "epoch": 1.2696629213483146,
+      "grad_norm": 0.2809509038925171,
+      "learning_rate": 1.3124199436205575e-05,
+      "loss": 0.1625,
+      "step": 310
+    },
+    {
+      "epoch": 1.2737487231869253,
+      "grad_norm": 0.27429160475730896,
+      "learning_rate": 1.3081656364772308e-05,
+      "loss": 0.1796,
+      "step": 311
+    },
+    {
+      "epoch": 1.2778345250255363,
+      "grad_norm": 0.2557787299156189,
+      "learning_rate": 1.303905157574247e-05,
+      "loss": 0.1664,
+      "step": 312
+    },
+    {
+      "epoch": 1.281920326864147,
+      "grad_norm": 0.3070502281188965,
+      "learning_rate": 1.2996385922379657e-05,
+      "loss": 0.1884,
+      "step": 313
+    },
+    {
+      "epoch": 1.286006128702758,
+      "grad_norm": 0.2685239315032959,
+      "learning_rate": 1.2953660259166413e-05,
+      "loss": 0.1728,
+      "step": 314
+    },
+    {
+      "epoch": 1.2900919305413687,
+      "grad_norm": 0.2761296331882477,
+      "learning_rate": 1.291087544178713e-05,
+      "loss": 0.1754,
+      "step": 315
+    },
+    {
+      "epoch": 1.2941777323799797,
+      "grad_norm": 0.29421859979629517,
+      "learning_rate": 1.2868032327110904e-05,
+      "loss": 0.1566,
+      "step": 316
+    },
+    {
+      "epoch": 1.2982635342185904,
+      "grad_norm": 0.2753983736038208,
+      "learning_rate": 1.2825131773174371e-05,
+      "loss": 0.1722,
+      "step": 317
+    },
+    {
+      "epoch": 1.3023493360572012,
+      "grad_norm": 0.280300498008728,
+      "learning_rate": 1.2782174639164528e-05,
+      "loss": 0.1743,
+      "step": 318
+    },
+    {
+      "epoch": 1.3064351378958121,
+      "grad_norm": 0.28724053502082825,
+      "learning_rate": 1.2739161785401525e-05,
+      "loss": 0.1727,
+      "step": 319
+    },
+    {
+      "epoch": 1.3105209397344229,
+      "grad_norm": 0.24978399276733398,
+      "learning_rate": 1.269609407332144e-05,
+      "loss": 0.1654,
+      "step": 320
+    },
+    {
+      "epoch": 1.3146067415730336,
+      "grad_norm": 0.2458401620388031,
+      "learning_rate": 1.2652972365459008e-05,
+      "loss": 0.1558,
+      "step": 321
+    },
+    {
+      "epoch": 1.3186925434116445,
+      "grad_norm": 0.29217007756233215,
+      "learning_rate": 1.2609797525430374e-05,
+      "loss": 0.1749,
+      "step": 322
+    },
+    {
+      "epoch": 1.3227783452502553,
+      "grad_norm": 0.2738885283470154,
+      "learning_rate": 1.2566570417915769e-05,
+      "loss": 0.1598,
+      "step": 323
+    },
+    {
+      "epoch": 1.3268641470888662,
+      "grad_norm": 0.23460422456264496,
+      "learning_rate": 1.2523291908642219e-05,
+      "loss": 0.1586,
+      "step": 324
+    },
+    {
+      "epoch": 1.330949948927477,
+      "grad_norm": 0.2899508476257324,
+      "learning_rate": 1.2479962864366186e-05,
+      "loss": 0.1698,
+      "step": 325
+    },
+    {
+      "epoch": 1.335035750766088,
+      "grad_norm": 0.2744244933128357,
+      "learning_rate": 1.243658415285622e-05,
+      "loss": 0.167,
+      "step": 326
+    },
+    {
+      "epoch": 1.3391215526046987,
+      "grad_norm": 0.3147677183151245,
+      "learning_rate": 1.2393156642875579e-05,
+      "loss": 0.1592,
+      "step": 327
+    },
+    {
+      "epoch": 1.3432073544433094,
+      "grad_norm": 0.26883426308631897,
+      "learning_rate": 1.2349681204164823e-05,
+      "loss": 0.1735,
+      "step": 328
+    },
+    {
+      "epoch": 1.3432073544433094,
+      "eval_loss": 0.2857210040092468,
+      "eval_runtime": 5.8046,
+      "eval_samples_per_second": 13.61,
+      "eval_steps_per_second": 1.723,
+      "step": 328
+    },
+    {
+      "epoch": 1.3472931562819204,
+      "grad_norm": 0.26572638750076294,
+      "learning_rate": 1.2306158707424402e-05,
+      "loss": 0.172,
+      "step": 329
+    },
+    {
+      "epoch": 1.351378958120531,
+      "grad_norm": 0.3158324062824249,
+      "learning_rate": 1.2262590024297226e-05,
+      "loss": 0.184,
+      "step": 330
+    },
+    {
+      "epoch": 1.355464759959142,
+      "grad_norm": 0.2606561779975891,
+      "learning_rate": 1.2218976027351177e-05,
+      "loss": 0.1681,
+      "step": 331
+    },
+    {
+      "epoch": 1.3595505617977528,
+      "grad_norm": 0.2860865592956543,
+      "learning_rate": 1.2175317590061676e-05,
+      "loss": 0.1768,
+      "step": 332
+    },
+    {
+      "epoch": 1.3636363636363638,
+      "grad_norm": 0.2928154766559601,
+      "learning_rate": 1.2131615586794162e-05,
+      "loss": 0.1654,
+      "step": 333
+    },
+    {
+      "epoch": 1.3677221654749745,
+      "grad_norm": 0.2754892110824585,
+      "learning_rate": 1.2087870892786588e-05,
+      "loss": 0.1679,
+      "step": 334
+    },
+    {
+      "epoch": 1.3718079673135852,
+      "grad_norm": 0.25418567657470703,
+      "learning_rate": 1.2044084384131891e-05,
+      "loss": 0.1692,
+      "step": 335
+    },
+    {
+      "epoch": 1.3758937691521962,
+      "grad_norm": 0.29680415987968445,
+      "learning_rate": 1.2000256937760446e-05,
+      "loss": 0.1835,
+      "step": 336
+    },
+    {
+      "epoch": 1.379979570990807,
+      "grad_norm": 0.25421565771102905,
+      "learning_rate": 1.1956389431422508e-05,
+      "loss": 0.1628,
+      "step": 337
+    },
+    {
+      "epoch": 1.3840653728294177,
+      "grad_norm": 0.26102015376091003,
+      "learning_rate": 1.1912482743670624e-05,
+      "loss": 0.1587,
+      "step": 338
+    },
+    {
+      "epoch": 1.3881511746680286,
+      "grad_norm": 0.2658519744873047,
+      "learning_rate": 1.1868537753842052e-05,
+      "loss": 0.1622,
+      "step": 339
+    },
+    {
+      "epoch": 1.3922369765066394,
+      "grad_norm": 0.25693395733833313,
+      "learning_rate": 1.1824555342041129e-05,
+      "loss": 0.1611,
+      "step": 340
+    },
+    {
+      "epoch": 1.3963227783452503,
+      "grad_norm": 0.24095548689365387,
+      "learning_rate": 1.1780536389121668e-05,
+      "loss": 0.1566,
+      "step": 341
+    },
+    {
+      "epoch": 1.400408580183861,
+      "grad_norm": 0.25440356135368347,
+      "learning_rate": 1.1736481776669307e-05,
+      "loss": 0.1646,
+      "step": 342
+    },
+    {
+      "epoch": 1.404494382022472,
+      "grad_norm": 0.23900751769542694,
+      "learning_rate": 1.1692392386983837e-05,
+      "loss": 0.1567,
+      "step": 343
+    },
+    {
+      "epoch": 1.4085801838610827,
+      "grad_norm": 0.2516697645187378,
+      "learning_rate": 1.1648269103061567e-05,
+      "loss": 0.1693,
+      "step": 344
+    },
+    {
+      "epoch": 1.4126659856996935,
+      "grad_norm": 0.23285552859306335,
+      "learning_rate": 1.1604112808577603e-05,
+      "loss": 0.1565,
+      "step": 345
+    },
+    {
+      "epoch": 1.4167517875383044,
+      "grad_norm": 0.22535811364650726,
+      "learning_rate": 1.155992438786818e-05,
+      "loss": 0.1519,
+      "step": 346
+    },
+    {
+      "epoch": 1.4208375893769152,
+      "grad_norm": 0.2757152020931244,
+      "learning_rate": 1.1515704725912926e-05,
+      "loss": 0.1824,
+      "step": 347
+    },
+    {
+      "epoch": 1.424923391215526,
+      "grad_norm": 0.25517934560775757,
+      "learning_rate": 1.1471454708317163e-05,
+      "loss": 0.1524,
+      "step": 348
+    },
+    {
+      "epoch": 1.4290091930541369,
+      "grad_norm": 0.26882752776145935,
+      "learning_rate": 1.1427175221294145e-05,
+      "loss": 0.1653,
+      "step": 349
+    },
+    {
+      "epoch": 1.4330949948927478,
+      "grad_norm": 0.2248525470495224,
+      "learning_rate": 1.1382867151647333e-05,
+      "loss": 0.1458,
+      "step": 350
+    },
+    {
+      "epoch": 1.4371807967313586,
+      "grad_norm": 0.2648623585700989,
+      "learning_rate": 1.1338531386752618e-05,
+      "loss": 0.1663,
+      "step": 351
+    },
+    {
+      "epoch": 1.4412665985699693,
+      "grad_norm": 0.2239081859588623,
+      "learning_rate": 1.1294168814540554e-05,
+      "loss": 0.1488,
+      "step": 352
+    },
+    {
+      "epoch": 1.4453524004085803,
+      "grad_norm": 0.2529364824295044,
+      "learning_rate": 1.1249780323478585e-05,
+      "loss": 0.1633,
+      "step": 353
+    },
+    {
+      "epoch": 1.449438202247191,
+      "grad_norm": 0.22921797633171082,
+      "learning_rate": 1.1205366802553231e-05,
+      "loss": 0.1648,
+      "step": 354
+    },
+    {
+      "epoch": 1.4535240040858017,
+      "grad_norm": 0.29341360926628113,
+      "learning_rate": 1.1160929141252303e-05,
+      "loss": 0.1657,
+      "step": 355
+    },
+    {
+      "epoch": 1.4576098059244127,
+      "grad_norm": 0.2699342966079712,
+      "learning_rate": 1.1116468229547079e-05,
+      "loss": 0.1726,
+      "step": 356
+    },
+    {
+      "epoch": 1.4616956077630234,
+      "grad_norm": 0.22347010672092438,
+      "learning_rate": 1.107198495787448e-05,
+      "loss": 0.1549,
+      "step": 357
+    },
+    {
+      "epoch": 1.4657814096016344,
+      "grad_norm": 0.2765299677848816,
+      "learning_rate": 1.1027480217119245e-05,
+      "loss": 0.1567,
+      "step": 358
+    },
+    {
+      "epoch": 1.4698672114402451,
+      "grad_norm": 0.2796229422092438,
+      "learning_rate": 1.0982954898596072e-05,
+      "loss": 0.1673,
+      "step": 359
+    },
+    {
+      "epoch": 1.473953013278856,
+      "grad_norm": 0.2708180546760559,
+      "learning_rate": 1.0938409894031793e-05,
+      "loss": 0.1608,
+      "step": 360
+    },
+    {
+      "epoch": 1.4780388151174668,
+      "grad_norm": 0.26708030700683594,
+      "learning_rate": 1.0893846095547493e-05,
+      "loss": 0.1672,
+      "step": 361
+    },
+    {
+      "epoch": 1.4821246169560776,
+      "grad_norm": 0.25234729051589966,
+      "learning_rate": 1.084926439564065e-05,
+      "loss": 0.1695,
+      "step": 362
+    },
+    {
+      "epoch": 1.4862104187946885,
+      "grad_norm": 0.23701204359531403,
+      "learning_rate": 1.0804665687167262e-05,
+      "loss": 0.1478,
+      "step": 363
+    },
+    {
+      "epoch": 1.4902962206332993,
+      "grad_norm": 0.23572878539562225,
+      "learning_rate": 1.0760050863323961e-05,
+      "loss": 0.1518,
+      "step": 364
+    },
+    {
+      "epoch": 1.49438202247191,
+      "grad_norm": 0.26712414622306824,
+      "learning_rate": 1.0715420817630137e-05,
+      "loss": 0.1641,
+      "step": 365
+    },
+    {
+      "epoch": 1.498467824310521,
+      "grad_norm": 0.2618795931339264,
+      "learning_rate": 1.0670776443910024e-05,
+      "loss": 0.1584,
+      "step": 366
+    },
+    {
+      "epoch": 1.502553626149132,
+      "grad_norm": 0.24355687201023102,
+      "learning_rate": 1.062611863627482e-05,
+      "loss": 0.155,
+      "step": 367
+    },
+    {
+      "epoch": 1.5066394279877426,
+      "grad_norm": 0.28303593397140503,
+      "learning_rate": 1.0581448289104759e-05,
+      "loss": 0.1699,
+      "step": 368
+    },
+    {
+      "epoch": 1.5107252298263534,
+      "grad_norm": 0.2682429254055023,
+      "learning_rate": 1.0536766297031216e-05,
+      "loss": 0.1638,
+      "step": 369
+    },
+    {
+      "epoch": 1.5148110316649643,
+      "grad_norm": 0.2611052095890045,
+      "learning_rate": 1.0492073554918782e-05,
+      "loss": 0.162,
+      "step": 370
+    },
+    {
+      "epoch": 1.518896833503575,
+      "grad_norm": 0.2545654773712158,
+      "learning_rate": 1.0447370957847343e-05,
+      "loss": 0.171,
+      "step": 371
+    },
+    {
+      "epoch": 1.5229826353421858,
+      "grad_norm": 0.2540684640407562,
+      "learning_rate": 1.0402659401094154e-05,
+      "loss": 0.1609,
+      "step": 372
+    },
+    {
+      "epoch": 1.5270684371807968,
+      "grad_norm": 0.29473230242729187,
+      "learning_rate": 1.0357939780115906e-05,
+      "loss": 0.1739,
+      "step": 373
+    },
+    {
+      "epoch": 1.5311542390194075,
+      "grad_norm": 0.23088738322257996,
+      "learning_rate": 1.0313212990530804e-05,
+      "loss": 0.1396,
+      "step": 374
+    },
+    {
+      "epoch": 1.5352400408580182,
+      "grad_norm": 0.2865520119667053,
+      "learning_rate": 1.0268479928100615e-05,
+      "loss": 0.1587,
+      "step": 375
+    },
+    {
+      "epoch": 1.5393258426966292,
+      "grad_norm": 0.26724815368652344,
+      "learning_rate": 1.0223741488712732e-05,
+      "loss": 0.1643,
+      "step": 376
+    },
+    {
+      "epoch": 1.5434116445352402,
+      "grad_norm": 0.2568652033805847,
+      "learning_rate": 1.0178998568362243e-05,
+      "loss": 0.1502,
+      "step": 377
+    },
+    {
+      "epoch": 1.547497446373851,
+      "grad_norm": 0.25489166378974915,
+      "learning_rate": 1.0134252063133976e-05,
+      "loss": 0.1551,
+      "step": 378
+    },
+    {
+      "epoch": 1.5515832482124616,
+      "grad_norm": 0.2938600480556488,
+      "learning_rate": 1.0089502869184549e-05,
+      "loss": 0.1721,
+      "step": 379
+    },
+    {
+      "epoch": 1.5556690500510726,
+      "grad_norm": 0.2571638822555542,
+      "learning_rate": 1.0044751882724436e-05,
+      "loss": 0.1596,
+      "step": 380
+    },
+    {
+      "epoch": 1.5597548518896833,
+      "grad_norm": 0.2504737079143524,
+      "learning_rate": 1e-05,
+      "loss": 0.1652,
+      "step": 381
+    },
+    {
+      "epoch": 1.563840653728294,
+      "grad_norm": 0.25643548369407654,
+      "learning_rate": 9.955248117275566e-06,
+      "loss": 0.1646,
+      "step": 382
+    },
+    {
+      "epoch": 1.567926455566905,
+      "grad_norm": 0.24690495431423187,
+      "learning_rate": 9.910497130815454e-06,
+      "loss": 0.1692,
+      "step": 383
+    },
+    {
+      "epoch": 1.572012257405516,
+      "grad_norm": 0.23503315448760986,
+      "learning_rate": 9.865747936866027e-06,
+      "loss": 0.1614,
+      "step": 384
+    },
+    {
+      "epoch": 1.5760980592441267,
+      "grad_norm": 0.2600212097167969,
+      "learning_rate": 9.821001431637759e-06,
+      "loss": 0.1843,
+      "step": 385
+    },
+    {
+      "epoch": 1.5801838610827375,
+      "grad_norm": 0.24049755930900574,
+      "learning_rate": 9.776258511287271e-06,
+      "loss": 0.1939,
+      "step": 386
+    },
+    {
+      "epoch": 1.5842696629213484,
+      "grad_norm": 0.26995447278022766,
+      "learning_rate": 9.73152007189939e-06,
+      "loss": 0.1608,
+      "step": 387
+    },
+    {
+      "epoch": 1.5883554647599591,
+      "grad_norm": 0.25705352425575256,
+      "learning_rate": 9.6867870094692e-06,
+      "loss": 0.1503,
+      "step": 388
+    },
+    {
+      "epoch": 1.5924412665985699,
+      "grad_norm": 0.2591187059879303,
+      "learning_rate": 9.642060219884096e-06,
+      "loss": 0.1601,
+      "step": 389
+    },
+    {
+      "epoch": 1.5965270684371808,
+      "grad_norm": 0.26638317108154297,
+      "learning_rate": 9.597340598905851e-06,
+      "loss": 0.1525,
+      "step": 390
+    },
+    {
+      "epoch": 1.6006128702757916,
+      "grad_norm": 0.27399975061416626,
+      "learning_rate": 9.55262904215266e-06,
+      "loss": 0.1571,
+      "step": 391
+    },
+    {
+      "epoch": 1.6046986721144023,
+      "grad_norm": 0.298513263463974,
+      "learning_rate": 9.50792644508122e-06,
+      "loss": 0.1734,
+      "step": 392
+    },
+    {
+      "epoch": 1.6087844739530133,
+      "grad_norm": 0.2932952344417572,
+      "learning_rate": 9.463233702968784e-06,
+      "loss": 0.1595,
+      "step": 393
+    },
+    {
+      "epoch": 1.6128702757916242,
+      "grad_norm": 0.2699350118637085,
+      "learning_rate": 9.418551710895243e-06,
+      "loss": 0.1513,
+      "step": 394
+    },
+    {
+      "epoch": 1.616956077630235,
+      "grad_norm": 0.2710689902305603,
+      "learning_rate": 9.373881363725182e-06,
+      "loss": 0.1558,
+      "step": 395
+    },
+    {
+      "epoch": 1.6210418794688457,
+      "grad_norm": 0.26967060565948486,
+      "learning_rate": 9.329223556089976e-06,
+      "loss": 0.1532,
+      "step": 396
+    },
+    {
+      "epoch": 1.6251276813074567,
+      "grad_norm": 0.26783767342567444,
+      "learning_rate": 9.284579182369868e-06,
+      "loss": 0.167,
+      "step": 397
+    },
+    {
+      "epoch": 1.6292134831460674,
+      "grad_norm": 0.2573103606700897,
+      "learning_rate": 9.239949136676042e-06,
+      "loss": 0.1675,
+      "step": 398
+    },
+    {
+      "epoch": 1.6332992849846781,
+      "grad_norm": 0.2554529905319214,
+      "learning_rate": 9.195334312832742e-06,
+      "loss": 0.1653,
+      "step": 399
+    },
+    {
+      "epoch": 1.637385086823289,
+      "grad_norm": 0.2697620391845703,
+      "learning_rate": 9.15073560435935e-06,
+      "loss": 0.1754,
+      "step": 400
+    },
+    {
+      "epoch": 1.6414708886619,
+      "grad_norm": 0.2908802032470703,
+      "learning_rate": 9.10615390445251e-06,
+      "loss": 0.1694,
+      "step": 401
+    },
+    {
+      "epoch": 1.6455566905005106,
+      "grad_norm": 0.28988802433013916,
+      "learning_rate": 9.061590105968208e-06,
+      "loss": 0.1596,
+      "step": 402
+    },
+    {
+      "epoch": 1.6496424923391215,
+      "grad_norm": 0.27670571208000183,
+      "learning_rate": 9.01704510140393e-06,
+      "loss": 0.1486,
+      "step": 403
+    },
+    {
+      "epoch": 1.6537282941777325,
+      "grad_norm": 0.29919058084487915,
+      "learning_rate": 8.97251978288076e-06,
+      "loss": 0.1668,
+      "step": 404
+    },
+    {
+      "epoch": 1.6578140960163432,
+      "grad_norm": 0.2605692446231842,
+      "learning_rate": 8.928015042125523e-06,
+      "loss": 0.1533,
+      "step": 405
+    },
+    {
+      "epoch": 1.661899897854954,
+      "grad_norm": 0.27188801765441895,
+      "learning_rate": 8.883531770452924e-06,
+      "loss": 0.1591,
+      "step": 406
+    },
+    {
+      "epoch": 1.665985699693565,
+      "grad_norm": 0.2607693374156952,
+      "learning_rate": 8.839070858747697e-06,
+      "loss": 0.1631,
+      "step": 407
+    },
+    {
+      "epoch": 1.6700715015321757,
+      "grad_norm": 0.26251208782196045,
+      "learning_rate": 8.79463319744677e-06,
+      "loss": 0.1669,
+      "step": 408
+    },
+    {
+      "epoch": 1.6741573033707864,
+      "grad_norm": 0.27655109763145447,
+      "learning_rate": 8.750219676521417e-06,
+      "loss": 0.1797,
+      "step": 409
+    },
+    {
+      "epoch": 1.6782431052093973,
+      "grad_norm": 0.2489909827709198,
+      "learning_rate": 8.705831185459446e-06,
+      "loss": 0.1684,
+      "step": 410
+    },
+    {
+      "epoch": 1.6782431052093973,
+      "eval_loss": 0.2804652154445648,
+      "eval_runtime": 5.3248,
+      "eval_samples_per_second": 14.836,
+      "eval_steps_per_second": 1.878,
+      "step": 410
+    },
+    {
+      "epoch": 1.6823289070480083,
+      "grad_norm": 0.2541872560977936,
+      "learning_rate": 8.661468613247387e-06,
+      "loss": 0.1738,
+      "step": 411
+    },
+    {
+      "epoch": 1.686414708886619,
+      "grad_norm": 0.26432761549949646,
+      "learning_rate": 8.617132848352672e-06,
+      "loss": 0.1523,
+      "step": 412
+    },
+    {
+      "epoch": 1.6905005107252298,
+      "grad_norm": 0.24682320654392242,
+      "learning_rate": 8.572824778705858e-06,
+      "loss": 0.1685,
+      "step": 413
+    },
+    {
+      "epoch": 1.6945863125638407,
+      "grad_norm": 0.255575567483902,
+      "learning_rate": 8.528545291682839e-06,
+      "loss": 0.1603,
+      "step": 414
+    },
+    {
+      "epoch": 1.6986721144024515,
+      "grad_norm": 0.27255284786224365,
+      "learning_rate": 8.484295274087077e-06,
+      "loss": 0.1649,
+      "step": 415
+    },
+    {
+      "epoch": 1.7027579162410622,
+      "grad_norm": 0.2935710549354553,
+      "learning_rate": 8.440075612131823e-06,
+      "loss": 0.1824,
+      "step": 416
+    },
+    {
+      "epoch": 1.7068437180796732,
+      "grad_norm": 0.28145232796669006,
+      "learning_rate": 8.395887191422397e-06,
+      "loss": 0.1664,
+      "step": 417
+    },
+    {
+      "epoch": 1.7109295199182841,
+      "grad_norm": 0.2540966272354126,
+      "learning_rate": 8.351730896938438e-06,
+      "loss": 0.139,
+      "step": 418
+    },
+    {
+      "epoch": 1.7150153217568946,
+      "grad_norm": 0.2761797606945038,
+      "learning_rate": 8.307607613016166e-06,
+      "loss": 0.1468,
+      "step": 419
+    },
+    {
+      "epoch": 1.7191011235955056,
+      "grad_norm": 0.26004406809806824,
+      "learning_rate": 8.263518223330698e-06,
+      "loss": 0.1791,
+      "step": 420
+    },
+    {
+      "epoch": 1.7231869254341166,
+      "grad_norm": 0.26706498861312866,
+      "learning_rate": 8.219463610878336e-06,
+      "loss": 0.1767,
+      "step": 421
+    },
+    {
+      "epoch": 1.7272727272727273,
+      "grad_norm": 0.25433361530303955,
+      "learning_rate": 8.175444657958875e-06,
+      "loss": 0.1641,
+      "step": 422
+    },
+    {
+      "epoch": 1.731358529111338,
+      "grad_norm": 0.28011849522590637,
+      "learning_rate": 8.131462246157953e-06,
+      "loss": 0.1667,
+      "step": 423
+    },
+    {
+      "epoch": 1.735444330949949,
+      "grad_norm": 0.24411511421203613,
+      "learning_rate": 8.087517256329376e-06,
+      "loss": 0.1484,
+      "step": 424
+    },
+    {
+      "epoch": 1.7395301327885597,
+      "grad_norm": 0.2515384554862976,
+      "learning_rate": 8.043610568577497e-06,
+      "loss": 0.149,
+      "step": 425
+    },
+    {
+      "epoch": 1.7436159346271705,
+      "grad_norm": 0.28085580468177795,
+      "learning_rate": 7.999743062239557e-06,
+      "loss": 0.1758,
+      "step": 426
+    },
+    {
+      "epoch": 1.7477017364657814,
+      "grad_norm": 0.2542356848716736,
+      "learning_rate": 7.95591561586811e-06,
+      "loss": 0.1526,
+      "step": 427
+    },
+    {
+      "epoch": 1.7517875383043924,
+      "grad_norm": 0.2624610960483551,
+      "learning_rate": 7.912129107213417e-06,
+      "loss": 0.1669,
+      "step": 428
+    },
+    {
+      "epoch": 1.7558733401430031,
+      "grad_norm": 0.2531009316444397,
+      "learning_rate": 7.868384413205842e-06,
+      "loss": 0.1728,
+      "step": 429
+    },
+    {
+      "epoch": 1.7599591419816139,
+      "grad_norm": 0.26832813024520874,
+      "learning_rate": 7.824682409938328e-06,
+      "loss": 0.1689,
+      "step": 430
+    },
+    {
+      "epoch": 1.7640449438202248,
+      "grad_norm": 0.26647037267684937,
+      "learning_rate": 7.781023972648826e-06,
+      "loss": 0.1566,
+      "step": 431
+    },
+    {
+      "epoch": 1.7681307456588355,
+      "grad_norm": 0.2441844940185547,
+      "learning_rate": 7.73740997570278e-06,
+      "loss": 0.1475,
+      "step": 432
+    },
+    {
+      "epoch": 1.7722165474974463,
+      "grad_norm": 0.26222023367881775,
+      "learning_rate": 7.6938412925756e-06,
+      "loss": 0.1627,
+      "step": 433
+    },
+    {
+      "epoch": 1.7763023493360572,
+      "grad_norm": 0.27849847078323364,
+      "learning_rate": 7.650318795835179e-06,
+      "loss": 0.1692,
+      "step": 434
+    },
+    {
+      "epoch": 1.780388151174668,
+      "grad_norm": 0.23362480103969574,
+      "learning_rate": 7.606843357124426e-06,
+      "loss": 0.1486,
+      "step": 435
+    },
+    {
+      "epoch": 1.7844739530132787,
+      "grad_norm": 0.25098103284835815,
+      "learning_rate": 7.563415847143782e-06,
+      "loss": 0.1586,
+      "step": 436
+    },
+    {
+      "epoch": 1.7885597548518897,
+      "grad_norm": 0.2666711211204529,
+      "learning_rate": 7.520037135633817e-06,
+      "loss": 0.1631,
+      "step": 437
+    },
+    {
+      "epoch": 1.7926455566905006,
+      "grad_norm": 0.25154757499694824,
+      "learning_rate": 7.476708091357783e-06,
+      "loss": 0.1496,
+      "step": 438
+    },
+    {
+      "epoch": 1.7967313585291114,
+      "grad_norm": 0.2870493233203888,
+      "learning_rate": 7.433429582084233e-06,
+      "loss": 0.1718,
+      "step": 439
+    },
+    {
+      "epoch": 1.800817160367722,
+      "grad_norm": 0.2450946867465973,
+      "learning_rate": 7.39020247456963e-06,
+      "loss": 0.1551,
+      "step": 440
+    },
+    {
+      "epoch": 1.804902962206333,
+      "grad_norm": 0.2701391577720642,
+      "learning_rate": 7.347027634540993e-06,
+      "loss": 0.1611,
+      "step": 441
+    },
+    {
+      "epoch": 1.8089887640449438,
+      "grad_norm": 0.25652557611465454,
+      "learning_rate": 7.303905926678565e-06,
+      "loss": 0.1571,
+      "step": 442
+    },
+    {
+      "epoch": 1.8130745658835545,
+      "grad_norm": 0.24130114912986755,
+      "learning_rate": 7.260838214598475e-06,
+      "loss": 0.1525,
+      "step": 443
+    },
+    {
+      "epoch": 1.8171603677221655,
+      "grad_norm": 0.2391010969877243,
+      "learning_rate": 7.217825360835475e-06,
+      "loss": 0.1478,
+      "step": 444
+    },
+    {
+      "epoch": 1.8212461695607765,
+      "grad_norm": 0.24808183312416077,
+      "learning_rate": 7.174868226825631e-06,
+      "loss": 0.1449,
+      "step": 445
+    },
+    {
+      "epoch": 1.825331971399387,
+      "grad_norm": 0.24367845058441162,
+      "learning_rate": 7.131967672889101e-06,
+      "loss": 0.1527,
+      "step": 446
+    },
+    {
+      "epoch": 1.829417773237998,
+      "grad_norm": 0.24614740908145905,
+      "learning_rate": 7.089124558212872e-06,
+      "loss": 0.1473,
+      "step": 447
+    },
+    {
+      "epoch": 1.8335035750766089,
+      "grad_norm": 0.23732498288154602,
+      "learning_rate": 7.04633974083359e-06,
+      "loss": 0.1676,
+      "step": 448
+    },
+    {
+      "epoch": 1.8375893769152196,
+      "grad_norm": 0.26191797852516174,
+      "learning_rate": 7.003614077620348e-06,
+      "loss": 0.1625,
+      "step": 449
+    },
+    {
+      "epoch": 1.8416751787538304,
+      "grad_norm": 0.22175060212612152,
+      "learning_rate": 6.960948424257532e-06,
+      "loss": 0.1417,
+      "step": 450
+    },
+    {
+      "epoch": 1.8457609805924413,
+      "grad_norm": 0.2599637806415558,
+      "learning_rate": 6.918343635227694e-06,
+      "loss": 0.1542,
+      "step": 451
+    },
+    {
+      "epoch": 1.849846782431052,
+      "grad_norm": 0.2902531325817108,
+      "learning_rate": 6.8758005637944245e-06,
+      "loss": 0.1673,
+      "step": 452
+    },
+    {
+      "epoch": 1.8539325842696628,
+      "grad_norm": 0.26200827956199646,
+      "learning_rate": 6.833320061985278e-06,
+      "loss": 0.1507,
+      "step": 453
+    },
+    {
+      "epoch": 1.8580183861082737,
+      "grad_norm": 0.22496499121189117,
+      "learning_rate": 6.7909029805746855e-06,
+      "loss": 0.1563,
+      "step": 454
+    },
+    {
+      "epoch": 1.8621041879468847,
+      "grad_norm": 0.26499348878860474,
+      "learning_rate": 6.7485501690669495e-06,
+      "loss": 0.1588,
+      "step": 455
+    },
+    {
+      "epoch": 1.8661899897854954,
+      "grad_norm": 0.21678292751312256,
+      "learning_rate": 6.706262475679205e-06,
+      "loss": 0.1446,
+      "step": 456
+    },
+    {
+      "epoch": 1.8702757916241062,
+      "grad_norm": 0.249608114361763,
+      "learning_rate": 6.664040747324437e-06,
+      "loss": 0.1574,
+      "step": 457
+    },
+    {
+      "epoch": 1.8743615934627171,
+      "grad_norm": 0.27170929312705994,
+      "learning_rate": 6.62188582959453e-06,
+      "loss": 0.1714,
+      "step": 458
+    },
+    {
+      "epoch": 1.8784473953013279,
+      "grad_norm": 0.26091060042381287,
+      "learning_rate": 6.579798566743314e-06,
+      "loss": 0.153,
+      "step": 459
+    },
+    {
+      "epoch": 1.8825331971399386,
+      "grad_norm": 0.2784002125263214,
+      "learning_rate": 6.537779801669677e-06,
+      "loss": 0.1594,
+      "step": 460
+    },
+    {
+      "epoch": 1.8866189989785496,
+      "grad_norm": 0.2827843427658081,
+      "learning_rate": 6.495830375900665e-06,
+      "loss": 0.1713,
+      "step": 461
+    },
+    {
+      "epoch": 1.8907048008171605,
+      "grad_norm": 0.24465838074684143,
+      "learning_rate": 6.453951129574644e-06,
+      "loss": 0.1398,
+      "step": 462
+    },
+    {
+      "epoch": 1.894790602655771,
+      "grad_norm": 0.24695105850696564,
+      "learning_rate": 6.41214290142447e-06,
+      "loss": 0.1569,
+      "step": 463
+    },
+    {
+      "epoch": 1.898876404494382,
+      "grad_norm": 0.23522843420505524,
+      "learning_rate": 6.370406528760675e-06,
+      "loss": 0.1572,
+      "step": 464
+    },
+    {
+      "epoch": 1.902962206332993,
+      "grad_norm": 0.28958627581596375,
+      "learning_rate": 6.3287428474547256e-06,
+      "loss": 0.1576,
+      "step": 465
+    },
+    {
+      "epoch": 1.9070480081716037,
+      "grad_norm": 0.22417336702346802,
+      "learning_rate": 6.287152691922264e-06,
+      "loss": 0.151,
+      "step": 466
+    },
+    {
+      "epoch": 1.9111338100102144,
+      "grad_norm": 0.24010370671749115,
+      "learning_rate": 6.245636895106403e-06,
+      "loss": 0.1422,
+      "step": 467
+    },
+    {
+      "epoch": 1.9152196118488254,
+      "grad_norm": 0.257285475730896,
+      "learning_rate": 6.204196288461037e-06,
+      "loss": 0.1541,
+      "step": 468
+    },
+    {
+      "epoch": 1.9193054136874361,
+      "grad_norm": 0.2468208223581314,
+      "learning_rate": 6.162831701934203e-06,
+      "loss": 0.1618,
+      "step": 469
+    },
+    {
+      "epoch": 1.9233912155260469,
+      "grad_norm": 0.2693644165992737,
+      "learning_rate": 6.121543963951453e-06,
+      "loss": 0.1597,
+      "step": 470
+    },
+    {
+      "epoch": 1.9274770173646578,
+      "grad_norm": 0.22864265739917755,
+      "learning_rate": 6.080333901399252e-06,
+      "loss": 0.1447,
+      "step": 471
+    },
+    {
+      "epoch": 1.9315628192032688,
+      "grad_norm": 0.2744729518890381,
+      "learning_rate": 6.039202339608432e-06,
+      "loss": 0.1649,
+      "step": 472
+    },
+    {
+      "epoch": 1.9356486210418795,
+      "grad_norm": 0.2626800537109375,
+      "learning_rate": 5.998150102337665e-06,
+      "loss": 0.1465,
+      "step": 473
+    },
+    {
+      "epoch": 1.9397344228804902,
+      "grad_norm": 0.24998779594898224,
+      "learning_rate": 5.957178011756952e-06,
+      "loss": 0.1314,
+      "step": 474
+    },
+    {
+      "epoch": 1.9438202247191012,
+      "grad_norm": 0.25133228302001953,
+      "learning_rate": 5.9162868884311596e-06,
+      "loss": 0.1541,
+      "step": 475
+    },
+    {
+      "epoch": 1.947906026557712,
+      "grad_norm": 0.27924278378486633,
+      "learning_rate": 5.875477551303596e-06,
+      "loss": 0.1588,
+      "step": 476
+    },
+    {
+      "epoch": 1.9519918283963227,
+      "grad_norm": 0.23838290572166443,
+      "learning_rate": 5.834750817679606e-06,
+      "loss": 0.1559,
+      "step": 477
+    },
+    {
+      "epoch": 1.9560776302349336,
+      "grad_norm": 0.20889320969581604,
+      "learning_rate": 5.794107503210187e-06,
+      "loss": 0.1376,
+      "step": 478
+    },
+    {
+      "epoch": 1.9601634320735446,
+      "grad_norm": 0.24007071554660797,
+      "learning_rate": 5.753548421875686e-06,
+      "loss": 0.1641,
+      "step": 479
+    },
+    {
+      "epoch": 1.9642492339121551,
+      "grad_norm": 0.25776174664497375,
+      "learning_rate": 5.713074385969457e-06,
+      "loss": 0.1486,
+      "step": 480
+    },
+    {
+      "epoch": 1.968335035750766,
+      "grad_norm": 0.24709415435791016,
+      "learning_rate": 5.672686206081638e-06,
+      "loss": 0.1647,
+      "step": 481
+    },
+    {
+      "epoch": 1.972420837589377,
+      "grad_norm": 0.2545711398124695,
+      "learning_rate": 5.632384691082874e-06,
+      "loss": 0.1558,
+      "step": 482
+    },
+    {
+      "epoch": 1.9765066394279878,
+      "grad_norm": 0.25180289149284363,
+      "learning_rate": 5.5921706481081405e-06,
+      "loss": 0.1405,
+      "step": 483
+    },
+    {
+      "epoch": 1.9805924412665985,
+      "grad_norm": 0.2353358417749405,
+      "learning_rate": 5.55204488254059e-06,
+      "loss": 0.1496,
+      "step": 484
+    },
+    {
+      "epoch": 1.9846782431052095,
+      "grad_norm": 0.25672510266304016,
+      "learning_rate": 5.512008197995379e-06,
+      "loss": 0.1557,
+      "step": 485
+    },
+    {
+      "epoch": 1.9887640449438202,
+      "grad_norm": 0.24256597459316254,
+      "learning_rate": 5.47206139630363e-06,
+      "loss": 0.1366,
+      "step": 486
+    },
+    {
+      "epoch": 1.992849846782431,
+      "grad_norm": 0.2704496681690216,
+      "learning_rate": 5.432205277496327e-06,
+      "loss": 0.1492,
+      "step": 487
+    },
+    {
+      "epoch": 1.996935648621042,
+      "grad_norm": 0.24868719279766083,
+      "learning_rate": 5.3924406397883174e-06,
+      "loss": 0.1632,
+      "step": 488
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 732,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 244,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.264375507248742e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-488/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8a51e619db41bfecd4e2978f86e8cb848022d32d79a042203708d80062927ea
+size 10744

checkpoint-488/zero_to_fp32.py

@@ -0,0 +1,760 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import gc
+import json
+import numpy as np
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device, weights_only=False)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in tqdm(files, desc='Loading checkpoint shards'):
+        state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+class GatheredTensor:
+    """
+    A pseudo tensor that collects partitioned weights.
+    It is more memory efficient when there are multiple groups.
+    """
+
+    def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
+        self.flat_groups = flat_groups
+        self.flat_groups_offset = flat_groups_offset
+        self.offset = offset
+        self.partitioned_numel = partitioned_numel
+        self.shape = shape
+        self.dtype = self.flat_groups[0][0].dtype
+
+    def contiguous(self):
+        """
+        Merge partitioned weights from flat_groups into a single tensor.
+        """
+        end_idx = self.offset + self.partitioned_numel
+        world_size = len(self.flat_groups)
+        pad_flat_param_chunks = []
+
+        for rank_i in range(world_size):
+            # for each rank, we need to collect weights from related group/groups
+            flat_groups_at_rank_i = self.flat_groups[rank_i]
+            start_group_id = None
+            end_group_id = None
+            for group_id in range(len(self.flat_groups_offset)):
+                if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
+                    start_group_id = group_id
+                if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
+                    end_group_id = group_id
+                    break
+            # collect weights from related group/groups
+            for group_id in range(start_group_id, end_group_id + 1):
+                flat_tensor = flat_groups_at_rank_i[group_id]
+                start_offset = self.offset - self.flat_groups_offset[group_id]
+                end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
+                pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
+
+        # collect weights from all ranks
+        pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
+        param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
+        return param
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
+
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # memory efficient tensor
+        tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
+        state_dict[name] = tensor
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def to_torch_tensor(state_dict, return_empty_tensor=False):
+    """
+    Convert state_dict of GatheredTensor to torch tensor
+    """
+    torch_state_dict = {}
+    converted_tensors = {}
+    for name, tensor in state_dict.items():
+        tensor_id = id(tensor)
+        if tensor_id in converted_tensors:  # shared tensors
+            shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
+            torch_state_dict[name] = shared_tensor
+        else:
+            converted_tensors[tensor_id] = name
+            if return_empty_tensor:
+                torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
+            else:
+                torch_state_dict[name] = tensor.contiguous()
+    return torch_state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                             tag=None,
+                                             exclude_frozen_parameters=False,
+                                             lazy_mode=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+        - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
+          Convert the pesduo tensor to torch tensor by ``.contiguous()``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
+    You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint. Or you can load state_dict in lazy mode ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
+        for name, lazy_tensor in state_dict.item():
+            tensor = lazy_tensor.contiguous()  # to cpu
+            print(name, tensor)
+            # del tensor to release memory if it no longer in use
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+    if lazy_mode:
+        return state_dict
+    else:
+        return to_torch_tensor(state_dict)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                                          tag,
+                                                          exclude_frozen_parameters,
+                                                          lazy_mode=True)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        # an memory-efficient approach for sharding
+        empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
+        state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model by shard
+    os.makedirs(output_dir, exist_ok=True)
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
+        shard_state_dict = to_torch_tensor(shard_state_dict)
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard_state_dict, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard_state_dict, output_path)
+        # release the memory of current shard
+        for tensor_name in list(shard_state_dict.keys()):
+            del state_dict[tensor_name]
+            del shard_state_dict[tensor_name]
+        del shard_state_dict
+        gc.collect()
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)

checkpoint-732/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-732/config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "Qwen/Qwen2.5-3B-Instruct",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 70,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.1",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151665
+}

checkpoint-732/generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.48.1"
+}

checkpoint-732/latest

@@ -0,0 +1 @@
+global_step732

checkpoint-732/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94e77145585997322f3e99da879afe33d1189e28971acf742a4a46e57fb43e28
+size 4956450288

checkpoint-732/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4416d56b700f372991da3e4f93be86ca790c88aec666652aa0d55ec0bfa11ce
+size 1835586736

checkpoint-732/model.safetensors.index.json

@@ -0,0 +1,442 @@
+{
+  "metadata": {
+    "total_size": 6791987200
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

checkpoint-732/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f3803bff3f596c03b55881de967a825b5734e4a581739164f9cb9e7fd1aee89
+size 14512

checkpoint-732/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d768a04b798e2ca42effbe096b8e4481f32a402a9125a2ced390586dab8eb29e
+size 14512

checkpoint-732/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3674e6d322bb18fac688dee98de72d6d1e9649274ab1079046232a9da36c9b5
+size 1064

checkpoint-732/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-732/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896