Upload complete continuous vision-encoder evaluation checkpoints

.gitattributes

@@ -33,3 +33,18 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/clip_meta__l14/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/clip_openai__l14/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/dino_vitb16/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/dino_vitb8/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/dino_vits16/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/dino_vits8/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/dinov2_base/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/dinov2_giant/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/dinov2_large/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/dinov2_small/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/dinov3_vitl16/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/eupe_convnext_b/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/eupe_convnext_s/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/eupe_convnext_t/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/eupe_vit_b/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text

continuous/qwen25/clip_meta__l14/finetune/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
+}

continuous/qwen25/clip_meta__l14/finetune/config.json

@@ -0,0 +1,57 @@
+{
+  "add_faster_video": false,
+  "add_time_instruction": false,
+  "architectures": [
+    "LlavaQwenForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "faster_token_stride": 10,
+  "force_sample": false,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "image_aspect_ratio": "square",
+  "image_crop_resolution": null,
+  "image_grid_pinpoints": null,
+  "image_split_resolution": null,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "mm_hidden_size": 1024,
+  "mm_newline_position": "grid",
+  "mm_patch_merge_type": "flat",
+  "mm_projector_lr": 0.0001,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_resampler_type": null,
+  "mm_spatial_pool_mode": "bilinear",
+  "mm_spatial_pool_stride": null,
+  "mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -2,
+  "mm_vision_tower": "open_clip_hub:ViT-L-14",
+  "mm_vision_tower_lr": null,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "pos_skipping_range": 4096,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "tokenizer_model_max_length": 8192,
+  "tokenizer_padding_side": "right",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "use_mm_proj": true,
+  "use_pos_skipping": false,
+  "use_sliding_window": false,
+  "vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/vit_l14_metaclip.safetensors",
+  "vocab_size": 151936
+}

continuous/qwen25/clip_meta__l14/finetune/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": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.51.3"
+}

continuous/qwen25/clip_meta__l14/finetune/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a8f6d5a9c0e1bc9b2d84992562a3b9ab79406dac9a325da9d3d9e839b85550a
+size 4177931704

continuous/qwen25/clip_meta__l14/finetune/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
+  }
+}

continuous/qwen25/clip_meta__l14/finetune/tokenizer.json

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

continuous/qwen25/clip_meta__l14/finetune/tokenizer_config.json

@@ -0,0 +1,209 @@
+{
+  "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": 8192,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

continuous/qwen25/clip_meta__l14/finetune/trainer_state.json

@@ -0,0 +1,1856 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9999037443449803,
+  "eval_steps": 500,
+  "global_step": 2597,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.003850226200789296,
+      "grad_norm": 13.328228950500488,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 1.915,
+      "step": 10
+    },
+    {
+      "epoch": 0.007700452401578592,
+      "grad_norm": 11.559072494506836,
+      "learning_rate": 2.435897435897436e-06,
+      "loss": 1.5487,
+      "step": 20
+    },
+    {
+      "epoch": 0.01155067860236789,
+      "grad_norm": 3.560225009918213,
+      "learning_rate": 3.7179487179487184e-06,
+      "loss": 1.0948,
+      "step": 30
+    },
+    {
+      "epoch": 0.015400904803157185,
+      "grad_norm": 0.7330309748649597,
+      "learning_rate": 5e-06,
+      "loss": 0.9315,
+      "step": 40
+    },
+    {
+      "epoch": 0.019251131003946483,
+      "grad_norm": 0.3289840817451477,
+      "learning_rate": 6.282051282051282e-06,
+      "loss": 0.8979,
+      "step": 50
+    },
+    {
+      "epoch": 0.02310135720473578,
+      "grad_norm": 0.2926309108734131,
+      "learning_rate": 7.564102564102564e-06,
+      "loss": 0.8514,
+      "step": 60
+    },
+    {
+      "epoch": 0.026951583405525073,
+      "grad_norm": 0.232636958360672,
+      "learning_rate": 8.846153846153847e-06,
+      "loss": 0.8262,
+      "step": 70
+    },
+    {
+      "epoch": 0.03080180960631437,
+      "grad_norm": 0.2551595866680145,
+      "learning_rate": 9.999996111488723e-06,
+      "loss": 0.8353,
+      "step": 80
+    },
+    {
+      "epoch": 0.034652035807103666,
+      "grad_norm": 0.2186211794614792,
+      "learning_rate": 9.999529497453782e-06,
+      "loss": 0.8161,
+      "step": 90
+    },
+    {
+      "epoch": 0.038502262007892966,
+      "grad_norm": 0.2360212802886963,
+      "learning_rate": 9.998285264324193e-06,
+      "loss": 0.7946,
+      "step": 100
+    },
+    {
+      "epoch": 0.04235248820868226,
+      "grad_norm": 0.20676834881305695,
+      "learning_rate": 9.996263605626052e-06,
+      "loss": 0.7862,
+      "step": 110
+    },
+    {
+      "epoch": 0.04620271440947156,
+      "grad_norm": 0.23274122178554535,
+      "learning_rate": 9.99346483580503e-06,
+      "loss": 0.7989,
+      "step": 120
+    },
+    {
+      "epoch": 0.05005294061026085,
+      "grad_norm": 0.21537458896636963,
+      "learning_rate": 9.989889390177467e-06,
+      "loss": 0.7862,
+      "step": 130
+    },
+    {
+      "epoch": 0.053903166811050146,
+      "grad_norm": 0.2462105005979538,
+      "learning_rate": 9.985537824862647e-06,
+      "loss": 0.7901,
+      "step": 140
+    },
+    {
+      "epoch": 0.057753393011839446,
+      "grad_norm": 0.2393636703491211,
+      "learning_rate": 9.980410816696317e-06,
+      "loss": 0.7706,
+      "step": 150
+    },
+    {
+      "epoch": 0.06160361921262874,
+      "grad_norm": 0.22327250242233276,
+      "learning_rate": 9.97450916312541e-06,
+      "loss": 0.7804,
+      "step": 160
+    },
+    {
+      "epoch": 0.06545384541341803,
+      "grad_norm": 0.19368113577365875,
+      "learning_rate": 9.967833782084004e-06,
+      "loss": 0.7813,
+      "step": 170
+    },
+    {
+      "epoch": 0.06930407161420733,
+      "grad_norm": 0.23856046795845032,
+      "learning_rate": 9.960385711850554e-06,
+      "loss": 0.8064,
+      "step": 180
+    },
+    {
+      "epoch": 0.07315429781499663,
+      "grad_norm": 0.24432745575904846,
+      "learning_rate": 9.952166110886394e-06,
+      "loss": 0.7989,
+      "step": 190
+    },
+    {
+      "epoch": 0.07700452401578593,
+      "grad_norm": 0.2434455305337906,
+      "learning_rate": 9.943176257655567e-06,
+      "loss": 0.7894,
+      "step": 200
+    },
+    {
+      "epoch": 0.08085475021657522,
+      "grad_norm": 0.22042745351791382,
+      "learning_rate": 9.933417550425952e-06,
+      "loss": 0.7846,
+      "step": 210
+    },
+    {
+      "epoch": 0.08470497641736452,
+      "grad_norm": 0.22528015077114105,
+      "learning_rate": 9.9228915070518e-06,
+      "loss": 0.769,
+      "step": 220
+    },
+    {
+      "epoch": 0.08855520261815382,
+      "grad_norm": 0.2191612720489502,
+      "learning_rate": 9.911599764737632e-06,
+      "loss": 0.772,
+      "step": 230
+    },
+    {
+      "epoch": 0.09240542881894312,
+      "grad_norm": 0.22885620594024658,
+      "learning_rate": 9.899544079783606e-06,
+      "loss": 0.7665,
+      "step": 240
+    },
+    {
+      "epoch": 0.0962556550197324,
+      "grad_norm": 0.22552402317523956,
+      "learning_rate": 9.886726327312338e-06,
+      "loss": 0.7824,
+      "step": 250
+    },
+    {
+      "epoch": 0.1001058812205217,
+      "grad_norm": 0.22139668464660645,
+      "learning_rate": 9.87314850097724e-06,
+      "loss": 0.7795,
+      "step": 260
+    },
+    {
+      "epoch": 0.103956107421311,
+      "grad_norm": 0.21742850542068481,
+      "learning_rate": 9.858812712652452e-06,
+      "loss": 0.7692,
+      "step": 270
+    },
+    {
+      "epoch": 0.10780633362210029,
+      "grad_norm": 0.21192149817943573,
+      "learning_rate": 9.843721192104341e-06,
+      "loss": 0.767,
+      "step": 280
+    },
+    {
+      "epoch": 0.11165655982288959,
+      "grad_norm": 0.21844960749149323,
+      "learning_rate": 9.827876286644698e-06,
+      "loss": 0.7888,
+      "step": 290
+    },
+    {
+      "epoch": 0.11550678602367889,
+      "grad_norm": 0.2103869765996933,
+      "learning_rate": 9.811280460765643e-06,
+      "loss": 0.7513,
+      "step": 300
+    },
+    {
+      "epoch": 0.11935701222446819,
+      "grad_norm": 0.21498507261276245,
+      "learning_rate": 9.793936295756292e-06,
+      "loss": 0.7601,
+      "step": 310
+    },
+    {
+      "epoch": 0.12320723842525748,
+      "grad_norm": 0.2215348482131958,
+      "learning_rate": 9.775846489301279e-06,
+      "loss": 0.7637,
+      "step": 320
+    },
+    {
+      "epoch": 0.12705746462604678,
+      "grad_norm": 0.22566714882850647,
+      "learning_rate": 9.757013855061157e-06,
+      "loss": 0.7613,
+      "step": 330
+    },
+    {
+      "epoch": 0.13090769082683607,
+      "grad_norm": 0.22558389604091644,
+      "learning_rate": 9.737441322234758e-06,
+      "loss": 0.7756,
+      "step": 340
+    },
+    {
+      "epoch": 0.13475791702762538,
+      "grad_norm": 0.22736240923404694,
+      "learning_rate": 9.717131935103605e-06,
+      "loss": 0.7752,
+      "step": 350
+    },
+    {
+      "epoch": 0.13860814322841467,
+      "grad_norm": 0.22556090354919434,
+      "learning_rate": 9.696088852558398e-06,
+      "loss": 0.7608,
+      "step": 360
+    },
+    {
+      "epoch": 0.14245836942920398,
+      "grad_norm": 0.22164835035800934,
+      "learning_rate": 9.674315347607686e-06,
+      "loss": 0.7872,
+      "step": 370
+    },
+    {
+      "epoch": 0.14630859562999327,
+      "grad_norm": 0.2192438691854477,
+      "learning_rate": 9.651814806868796e-06,
+      "loss": 0.7522,
+      "step": 380
+    },
+    {
+      "epoch": 0.15015882183078255,
+      "grad_norm": 0.23910200595855713,
+      "learning_rate": 9.62859073004107e-06,
+      "loss": 0.7541,
+      "step": 390
+    },
+    {
+      "epoch": 0.15400904803157187,
+      "grad_norm": 0.20879314839839935,
+      "learning_rate": 9.604646729361547e-06,
+      "loss": 0.7597,
+      "step": 400
+    },
+    {
+      "epoch": 0.15785927423236115,
+      "grad_norm": 0.20812919735908508,
+      "learning_rate": 9.579986529043099e-06,
+      "loss": 0.7497,
+      "step": 410
+    },
+    {
+      "epoch": 0.16170950043315044,
+      "grad_norm": 0.21980519592761993,
+      "learning_rate": 9.554613964695189e-06,
+      "loss": 0.7669,
+      "step": 420
+    },
+    {
+      "epoch": 0.16555972663393975,
+      "grad_norm": 0.2162604033946991,
+      "learning_rate": 9.528532982727281e-06,
+      "loss": 0.7591,
+      "step": 430
+    },
+    {
+      "epoch": 0.16940995283472904,
+      "grad_norm": 0.20994342863559723,
+      "learning_rate": 9.501747639735018e-06,
+      "loss": 0.7552,
+      "step": 440
+    },
+    {
+      "epoch": 0.17326017903551832,
+      "grad_norm": 0.21485087275505066,
+      "learning_rate": 9.474262101869272e-06,
+      "loss": 0.7444,
+      "step": 450
+    },
+    {
+      "epoch": 0.17711040523630764,
+      "grad_norm": 0.2211788445711136,
+      "learning_rate": 9.446080644188141e-06,
+      "loss": 0.7521,
+      "step": 460
+    },
+    {
+      "epoch": 0.18096063143709692,
+      "grad_norm": 0.21153376996517181,
+      "learning_rate": 9.417207649992017e-06,
+      "loss": 0.7564,
+      "step": 470
+    },
+    {
+      "epoch": 0.18481085763788624,
+      "grad_norm": 0.2248362898826599,
+      "learning_rate": 9.387647610141816e-06,
+      "loss": 0.754,
+      "step": 480
+    },
+    {
+      "epoch": 0.18866108383867553,
+      "grad_norm": 0.20200952887535095,
+      "learning_rate": 9.35740512236047e-06,
+      "loss": 0.7418,
+      "step": 490
+    },
+    {
+      "epoch": 0.1925113100394648,
+      "grad_norm": 0.2008904218673706,
+      "learning_rate": 9.326484890517809e-06,
+      "loss": 0.7545,
+      "step": 500
+    },
+    {
+      "epoch": 0.19636153624025413,
+      "grad_norm": 0.22293621301651,
+      "learning_rate": 9.294891723898926e-06,
+      "loss": 0.7424,
+      "step": 510
+    },
+    {
+      "epoch": 0.2002117624410434,
+      "grad_norm": 0.23695115745067596,
+      "learning_rate": 9.262630536456153e-06,
+      "loss": 0.7731,
+      "step": 520
+    },
+    {
+      "epoch": 0.2040619886418327,
+      "grad_norm": 0.22950884699821472,
+      "learning_rate": 9.229706346044749e-06,
+      "loss": 0.7473,
+      "step": 530
+    },
+    {
+      "epoch": 0.207912214842622,
+      "grad_norm": 0.23465582728385925,
+      "learning_rate": 9.196124273642422e-06,
+      "loss": 0.7585,
+      "step": 540
+    },
+    {
+      "epoch": 0.2117624410434113,
+      "grad_norm": 0.2156888246536255,
+      "learning_rate": 9.161889542552842e-06,
+      "loss": 0.7639,
+      "step": 550
+    },
+    {
+      "epoch": 0.21561266724420058,
+      "grad_norm": 0.2076529562473297,
+      "learning_rate": 9.127007477593193e-06,
+      "loss": 0.7652,
+      "step": 560
+    },
+    {
+      "epoch": 0.2194628934449899,
+      "grad_norm": 0.24058207869529724,
+      "learning_rate": 9.091483504265974e-06,
+      "loss": 0.7725,
+      "step": 570
+    },
+    {
+      "epoch": 0.22331311964577918,
+      "grad_norm": 0.22434863448143005,
+      "learning_rate": 9.055323147915116e-06,
+      "loss": 0.753,
+      "step": 580
+    },
+    {
+      "epoch": 0.2271633458465685,
+      "grad_norm": 0.22181716561317444,
+      "learning_rate": 9.018532032866585e-06,
+      "loss": 0.7626,
+      "step": 590
+    },
+    {
+      "epoch": 0.23101357204735778,
+      "grad_norm": 0.2190052717924118,
+      "learning_rate": 8.981115881553579e-06,
+      "loss": 0.7555,
+      "step": 600
+    },
+    {
+      "epoch": 0.23486379824814707,
+      "grad_norm": 0.22813725471496582,
+      "learning_rate": 8.943080513626476e-06,
+      "loss": 0.7758,
+      "step": 610
+    },
+    {
+      "epoch": 0.23871402444893638,
+      "grad_norm": 0.21975436806678772,
+      "learning_rate": 8.904431845047641e-06,
+      "loss": 0.7605,
+      "step": 620
+    },
+    {
+      "epoch": 0.24256425064972567,
+      "grad_norm": 0.2270445078611374,
+      "learning_rate": 8.86517588717128e-06,
+      "loss": 0.7762,
+      "step": 630
+    },
+    {
+      "epoch": 0.24641447685051496,
+      "grad_norm": 0.21056774258613586,
+      "learning_rate": 8.82531874580844e-06,
+      "loss": 0.7693,
+      "step": 640
+    },
+    {
+      "epoch": 0.25026470305130427,
+      "grad_norm": 0.1958078145980835,
+      "learning_rate": 8.784866620277314e-06,
+      "loss": 0.7266,
+      "step": 650
+    },
+    {
+      "epoch": 0.25411492925209356,
+      "grad_norm": 0.23990647494792938,
+      "learning_rate": 8.743825802439018e-06,
+      "loss": 0.7572,
+      "step": 660
+    },
+    {
+      "epoch": 0.25796515545288284,
+      "grad_norm": 0.23645712435245514,
+      "learning_rate": 8.702202675718956e-06,
+      "loss": 0.7673,
+      "step": 670
+    },
+    {
+      "epoch": 0.26181538165367213,
+      "grad_norm": 0.2085576206445694,
+      "learning_rate": 8.660003714113959e-06,
+      "loss": 0.7346,
+      "step": 680
+    },
+    {
+      "epoch": 0.26566560785446147,
+      "grad_norm": 0.22540390491485596,
+      "learning_rate": 8.617235481185323e-06,
+      "loss": 0.7559,
+      "step": 690
+    },
+    {
+      "epoch": 0.26951583405525076,
+      "grad_norm": 0.22542905807495117,
+      "learning_rate": 8.573904629037931e-06,
+      "loss": 0.7626,
+      "step": 700
+    },
+    {
+      "epoch": 0.27336606025604004,
+      "grad_norm": 0.21824517846107483,
+      "learning_rate": 8.530017897285585e-06,
+      "loss": 0.7458,
+      "step": 710
+    },
+    {
+      "epoch": 0.27721628645682933,
+      "grad_norm": 0.1985330879688263,
+      "learning_rate": 8.485582112002746e-06,
+      "loss": 0.7504,
+      "step": 720
+    },
+    {
+      "epoch": 0.2810665126576186,
+      "grad_norm": 0.2157362997531891,
+      "learning_rate": 8.440604184662811e-06,
+      "loss": 0.7514,
+      "step": 730
+    },
+    {
+      "epoch": 0.28491673885840796,
+      "grad_norm": 0.22309021651744843,
+      "learning_rate": 8.395091111063118e-06,
+      "loss": 0.7425,
+      "step": 740
+    },
+    {
+      "epoch": 0.28876696505919724,
+      "grad_norm": 0.20909708738327026,
+      "learning_rate": 8.349049970236822e-06,
+      "loss": 0.7436,
+      "step": 750
+    },
+    {
+      "epoch": 0.29261719125998653,
+      "grad_norm": 0.22002920508384705,
+      "learning_rate": 8.302487923351842e-06,
+      "loss": 0.7341,
+      "step": 760
+    },
+    {
+      "epoch": 0.2964674174607758,
+      "grad_norm": 0.21200227737426758,
+      "learning_rate": 8.255412212597021e-06,
+      "loss": 0.7525,
+      "step": 770
+    },
+    {
+      "epoch": 0.3003176436615651,
+      "grad_norm": 0.21074150502681732,
+      "learning_rate": 8.207830160055688e-06,
+      "loss": 0.7492,
+      "step": 780
+    },
+    {
+      "epoch": 0.3041678698623544,
+      "grad_norm": 0.2208990901708603,
+      "learning_rate": 8.159749166566786e-06,
+      "loss": 0.7351,
+      "step": 790
+    },
+    {
+      "epoch": 0.30801809606314373,
+      "grad_norm": 0.21119946241378784,
+      "learning_rate": 8.11117671057377e-06,
+      "loss": 0.7616,
+      "step": 800
+    },
+    {
+      "epoch": 0.311868322263933,
+      "grad_norm": 0.22767092287540436,
+      "learning_rate": 8.062120346961404e-06,
+      "loss": 0.7535,
+      "step": 810
+    },
+    {
+      "epoch": 0.3157185484647223,
+      "grad_norm": 0.2025594562292099,
+      "learning_rate": 8.012587705880705e-06,
+      "loss": 0.7485,
+      "step": 820
+    },
+    {
+      "epoch": 0.3195687746655116,
+      "grad_norm": 0.21677997708320618,
+      "learning_rate": 7.962586491562146e-06,
+      "loss": 0.734,
+      "step": 830
+    },
+    {
+      "epoch": 0.3234190008663009,
+      "grad_norm": 0.22862166166305542,
+      "learning_rate": 7.912124481117357e-06,
+      "loss": 0.7397,
+      "step": 840
+    },
+    {
+      "epoch": 0.3272692270670902,
+      "grad_norm": 0.21503989398479462,
+      "learning_rate": 7.861209523329494e-06,
+      "loss": 0.753,
+      "step": 850
+    },
+    {
+      "epoch": 0.3311194532678795,
+      "grad_norm": 0.21931888163089752,
+      "learning_rate": 7.809849537432432e-06,
+      "loss": 0.7351,
+      "step": 860
+    },
+    {
+      "epoch": 0.3349696794686688,
+      "grad_norm": 0.21136625111103058,
+      "learning_rate": 7.758052511879035e-06,
+      "loss": 0.7453,
+      "step": 870
+    },
+    {
+      "epoch": 0.3388199056694581,
+      "grad_norm": 0.20572823286056519,
+      "learning_rate": 7.705826503098642e-06,
+      "loss": 0.7417,
+      "step": 880
+    },
+    {
+      "epoch": 0.34267013187024736,
+      "grad_norm": 0.2112719714641571,
+      "learning_rate": 7.653179634243979e-06,
+      "loss": 0.7407,
+      "step": 890
+    },
+    {
+      "epoch": 0.34652035807103665,
+      "grad_norm": 0.19393602013587952,
+      "learning_rate": 7.600120093927696e-06,
+      "loss": 0.7423,
+      "step": 900
+    },
+    {
+      "epoch": 0.350370584271826,
+      "grad_norm": 0.22093500196933746,
+      "learning_rate": 7.546656134948729e-06,
+      "loss": 0.7285,
+      "step": 910
+    },
+    {
+      "epoch": 0.3542208104726153,
+      "grad_norm": 0.2171906679868698,
+      "learning_rate": 7.492796073008661e-06,
+      "loss": 0.7333,
+      "step": 920
+    },
+    {
+      "epoch": 0.35807103667340456,
+      "grad_norm": 0.2338457852602005,
+      "learning_rate": 7.438548285418326e-06,
+      "loss": 0.7211,
+      "step": 930
+    },
+    {
+      "epoch": 0.36192126287419385,
+      "grad_norm": 0.23658554255962372,
+      "learning_rate": 7.3839212097948e-06,
+      "loss": 0.7549,
+      "step": 940
+    },
+    {
+      "epoch": 0.36577148907498314,
+      "grad_norm": 0.2158682644367218,
+      "learning_rate": 7.328923342749039e-06,
+      "loss": 0.7691,
+      "step": 950
+    },
+    {
+      "epoch": 0.3696217152757725,
+      "grad_norm": 0.22416085004806519,
+      "learning_rate": 7.273563238564319e-06,
+      "loss": 0.7546,
+      "step": 960
+    },
+    {
+      "epoch": 0.37347194147656176,
+      "grad_norm": 0.22308750450611115,
+      "learning_rate": 7.217849507865724e-06,
+      "loss": 0.7332,
+      "step": 970
+    },
+    {
+      "epoch": 0.37732216767735105,
+      "grad_norm": 0.21231094002723694,
+      "learning_rate": 7.161790816280858e-06,
+      "loss": 0.7408,
+      "step": 980
+    },
+    {
+      "epoch": 0.38117239387814034,
+      "grad_norm": 0.22008801996707916,
+      "learning_rate": 7.1053958830920125e-06,
+      "loss": 0.7214,
+      "step": 990
+    },
+    {
+      "epoch": 0.3850226200789296,
+      "grad_norm": 0.2120552361011505,
+      "learning_rate": 7.0486734798799715e-06,
+      "loss": 0.7408,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3888728462797189,
+      "grad_norm": 0.20816121995449066,
+      "learning_rate": 6.991632429159708e-06,
+      "loss": 0.7647,
+      "step": 1010
+    },
+    {
+      "epoch": 0.39272307248050825,
+      "grad_norm": 0.23089586198329926,
+      "learning_rate": 6.934281603008132e-06,
+      "loss": 0.7554,
+      "step": 1020
+    },
+    {
+      "epoch": 0.39657329868129754,
+      "grad_norm": 0.20898233354091644,
+      "learning_rate": 6.876629921684143e-06,
+      "loss": 0.7529,
+      "step": 1030
+    },
+    {
+      "epoch": 0.4004235248820868,
+      "grad_norm": 0.2259947806596756,
+      "learning_rate": 6.818686352241199e-06,
+      "loss": 0.7433,
+      "step": 1040
+    },
+    {
+      "epoch": 0.4042737510828761,
+      "grad_norm": 0.19634276628494263,
+      "learning_rate": 6.760459907132573e-06,
+      "loss": 0.7395,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4081239772836654,
+      "grad_norm": 0.19797876477241516,
+      "learning_rate": 6.701959642809589e-06,
+      "loss": 0.7451,
+      "step": 1060
+    },
+    {
+      "epoch": 0.41197420348445474,
+      "grad_norm": 0.21752998232841492,
+      "learning_rate": 6.643194658312987e-06,
+      "loss": 0.7351,
+      "step": 1070
+    },
+    {
+      "epoch": 0.415824429685244,
+      "grad_norm": 0.21043641865253448,
+      "learning_rate": 6.584174093857676e-06,
+      "loss": 0.7596,
+      "step": 1080
+    },
+    {
+      "epoch": 0.4196746558860333,
+      "grad_norm": 0.2385396957397461,
+      "learning_rate": 6.5249071294110735e-06,
+      "loss": 0.7448,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4235248820868226,
+      "grad_norm": 0.22310703992843628,
+      "learning_rate": 6.465402983265274e-06,
+      "loss": 0.7583,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4273751082876119,
+      "grad_norm": 0.22086073458194733,
+      "learning_rate": 6.405670910603243e-06,
+      "loss": 0.7529,
+      "step": 1110
+    },
+    {
+      "epoch": 0.43122533448840117,
+      "grad_norm": 0.2317456752061844,
+      "learning_rate": 6.345720202059291e-06,
+      "loss": 0.7493,
+      "step": 1120
+    },
+    {
+      "epoch": 0.4350755606891905,
+      "grad_norm": 0.21624058485031128,
+      "learning_rate": 6.285560182274008e-06,
+      "loss": 0.7273,
+      "step": 1130
+    },
+    {
+      "epoch": 0.4389257868899798,
+      "grad_norm": 0.20788072049617767,
+      "learning_rate": 6.225200208443936e-06,
+      "loss": 0.7492,
+      "step": 1140
+    },
+    {
+      "epoch": 0.4427760130907691,
+      "grad_norm": 0.21032989025115967,
+      "learning_rate": 6.16464966886616e-06,
+      "loss": 0.7385,
+      "step": 1150
+    },
+    {
+      "epoch": 0.44662623929155837,
+      "grad_norm": 0.24712848663330078,
+      "learning_rate": 6.103917981478065e-06,
+      "loss": 0.7633,
+      "step": 1160
+    },
+    {
+      "epoch": 0.45047646549234766,
+      "grad_norm": 0.21161438524723053,
+      "learning_rate": 6.043014592392489e-06,
+      "loss": 0.7743,
+      "step": 1170
+    },
+    {
+      "epoch": 0.454326691693137,
+      "grad_norm": 0.19967734813690186,
+      "learning_rate": 5.981948974428485e-06,
+      "loss": 0.7481,
+      "step": 1180
+    },
+    {
+      "epoch": 0.4581769178939263,
+      "grad_norm": 0.2185441255569458,
+      "learning_rate": 5.920730625637934e-06,
+      "loss": 0.7263,
+      "step": 1190
+    },
+    {
+      "epoch": 0.46202714409471557,
+      "grad_norm": 0.22365616261959076,
+      "learning_rate": 5.8593690678282355e-06,
+      "loss": 0.7393,
+      "step": 1200
+    },
+    {
+      "epoch": 0.46587737029550486,
+      "grad_norm": 0.2252226322889328,
+      "learning_rate": 5.797873845081301e-06,
+      "loss": 0.7434,
+      "step": 1210
+    },
+    {
+      "epoch": 0.46972759649629414,
+      "grad_norm": 0.21944968402385712,
+      "learning_rate": 5.736254522269075e-06,
+      "loss": 0.7581,
+      "step": 1220
+    },
+    {
+      "epoch": 0.47357782269708343,
+      "grad_norm": 0.21004584431648254,
+      "learning_rate": 5.6745206835658386e-06,
+      "loss": 0.7634,
+      "step": 1230
+    },
+    {
+      "epoch": 0.47742804889787277,
+      "grad_norm": 0.20573219656944275,
+      "learning_rate": 5.612681930957495e-06,
+      "loss": 0.7321,
+      "step": 1240
+    },
+    {
+      "epoch": 0.48127827509866206,
+      "grad_norm": 0.20263981819152832,
+      "learning_rate": 5.550747882748095e-06,
+      "loss": 0.7289,
+      "step": 1250
+    },
+    {
+      "epoch": 0.48512850129945134,
+      "grad_norm": 0.22885310649871826,
+      "learning_rate": 5.4887281720638195e-06,
+      "loss": 0.7391,
+      "step": 1260
+    },
+    {
+      "epoch": 0.48897872750024063,
+      "grad_norm": 0.24098259210586548,
+      "learning_rate": 5.42663244535466e-06,
+      "loss": 0.7199,
+      "step": 1270
+    },
+    {
+      "epoch": 0.4928289537010299,
+      "grad_norm": 0.21779608726501465,
+      "learning_rate": 5.3644703608940205e-06,
+      "loss": 0.7316,
+      "step": 1280
+    },
+    {
+      "epoch": 0.49667917990181926,
+      "grad_norm": 0.22561374306678772,
+      "learning_rate": 5.302251587276491e-06,
+      "loss": 0.741,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5005294061026085,
+      "grad_norm": 0.21318982541561127,
+      "learning_rate": 5.2399858019140005e-06,
+      "loss": 0.7426,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5043796323033978,
+      "grad_norm": 0.20850186049938202,
+      "learning_rate": 5.177682689530615e-06,
+      "loss": 0.7559,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5082298585041871,
+      "grad_norm": 0.22151678800582886,
+      "learning_rate": 5.11535194065619e-06,
+      "loss": 0.7257,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5120800847049765,
+      "grad_norm": 0.23383907973766327,
+      "learning_rate": 5.053003250119113e-06,
+      "loss": 0.7332,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5159303109057657,
+      "grad_norm": 0.227830171585083,
+      "learning_rate": 4.9906463155383925e-06,
+      "loss": 0.7409,
+      "step": 1340
+    },
+    {
+      "epoch": 0.519780537106555,
+      "grad_norm": 0.21640034019947052,
+      "learning_rate": 4.9282908358153046e-06,
+      "loss": 0.7367,
+      "step": 1350
+    },
+    {
+      "epoch": 0.5236307633073443,
+      "grad_norm": 0.24438755214214325,
+      "learning_rate": 4.8659465096248345e-06,
+      "loss": 0.7513,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5274809895081336,
+      "grad_norm": 0.22123686969280243,
+      "learning_rate": 4.803623033907169e-06,
+      "loss": 0.747,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5313312157089229,
+      "grad_norm": 0.20811326801776886,
+      "learning_rate": 4.741330102359441e-06,
+      "loss": 0.7563,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5351814419097122,
+      "grad_norm": 0.22682732343673706,
+      "learning_rate": 4.679077403927996e-06,
+      "loss": 0.7346,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5390316681105015,
+      "grad_norm": 0.20908166468143463,
+      "learning_rate": 4.616874621301373e-06,
+      "loss": 0.7362,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5428818943112907,
+      "grad_norm": 0.22428293526172638,
+      "learning_rate": 4.554731429404293e-06,
+      "loss": 0.7495,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5467321205120801,
+      "grad_norm": 0.22708803415298462,
+      "learning_rate": 4.492657493892822e-06,
+      "loss": 0.733,
+      "step": 1420
+    },
+    {
+      "epoch": 0.5505823467128694,
+      "grad_norm": 0.2316155880689621,
+      "learning_rate": 4.430662469650999e-06,
+      "loss": 0.75,
+      "step": 1430
+    },
+    {
+      "epoch": 0.5544325729136587,
+      "grad_norm": 0.22217819094657898,
+      "learning_rate": 4.368755999289118e-06,
+      "loss": 0.7549,
+      "step": 1440
+    },
+    {
+      "epoch": 0.558282799114448,
+      "grad_norm": 0.22791409492492676,
+      "learning_rate": 4.306947711643949e-06,
+      "loss": 0.7343,
+      "step": 1450
+    },
+    {
+      "epoch": 0.5621330253152372,
+      "grad_norm": 0.22647956013679504,
+      "learning_rate": 4.245247220281068e-06,
+      "loss": 0.741,
+      "step": 1460
+    },
+    {
+      "epoch": 0.5659832515160266,
+      "grad_norm": 0.2165859192609787,
+      "learning_rate": 4.183664121999587e-06,
+      "loss": 0.735,
+      "step": 1470
+    },
+    {
+      "epoch": 0.5698334777168159,
+      "grad_norm": 0.22653508186340332,
+      "learning_rate": 4.122207995339474e-06,
+      "loss": 0.7346,
+      "step": 1480
+    },
+    {
+      "epoch": 0.5736837039176051,
+      "grad_norm": 0.21507251262664795,
+      "learning_rate": 4.060888399091734e-06,
+      "loss": 0.7389,
+      "step": 1490
+    },
+    {
+      "epoch": 0.5775339301183945,
+      "grad_norm": 0.20572637021541595,
+      "learning_rate": 3.999714870811639e-06,
+      "loss": 0.7508,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5813841563191837,
+      "grad_norm": 0.21468766033649445,
+      "learning_rate": 3.938696925335275e-06,
+      "loss": 0.7428,
+      "step": 1510
+    },
+    {
+      "epoch": 0.5852343825199731,
+      "grad_norm": 0.21237066388130188,
+      "learning_rate": 3.87784405329962e-06,
+      "loss": 0.7505,
+      "step": 1520
+    },
+    {
+      "epoch": 0.5890846087207623,
+      "grad_norm": 0.21394744515419006,
+      "learning_rate": 3.817165719666382e-06,
+      "loss": 0.7506,
+      "step": 1530
+    },
+    {
+      "epoch": 0.5929348349215516,
+      "grad_norm": 0.209946408867836,
+      "learning_rate": 3.756671362249841e-06,
+      "loss": 0.758,
+      "step": 1540
+    },
+    {
+      "epoch": 0.596785061122341,
+      "grad_norm": 0.2123248428106308,
+      "learning_rate": 3.6963703902488907e-06,
+      "loss": 0.743,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6006352873231302,
+      "grad_norm": 0.21147572994232178,
+      "learning_rate": 3.6362721827835655e-06,
+      "loss": 0.7558,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6044855135239195,
+      "grad_norm": 0.22717633843421936,
+      "learning_rate": 3.576386087436212e-06,
+      "loss": 0.7546,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6083357397247088,
+      "grad_norm": 0.2297956645488739,
+      "learning_rate": 3.5167214187975913e-06,
+      "loss": 0.7441,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6121859659254981,
+      "grad_norm": 0.18952436745166779,
+      "learning_rate": 3.457287457018086e-06,
+      "loss": 0.7472,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6160361921262875,
+      "grad_norm": 0.2052864134311676,
+      "learning_rate": 3.3980934463643003e-06,
+      "loss": 0.7349,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6198864183270767,
+      "grad_norm": 0.20157356560230255,
+      "learning_rate": 3.3391485937812053e-06,
+      "loss": 0.7163,
+      "step": 1610
+    },
+    {
+      "epoch": 0.623736644527866,
+      "grad_norm": 0.21618697047233582,
+      "learning_rate": 3.280462067460115e-06,
+      "loss": 0.7158,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6275868707286553,
+      "grad_norm": 0.21041062474250793,
+      "learning_rate": 3.222042995412669e-06,
+      "loss": 0.7482,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6314370969294446,
+      "grad_norm": 0.19329747557640076,
+      "learning_rate": 3.163900464051089e-06,
+      "loss": 0.7365,
+      "step": 1640
+    },
+    {
+      "epoch": 0.635287323130234,
+      "grad_norm": 0.2148531675338745,
+      "learning_rate": 3.1060435167748826e-06,
+      "loss": 0.7477,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6391375493310232,
+      "grad_norm": 0.21839700639247894,
+      "learning_rate": 3.048481152564253e-06,
+      "loss": 0.7275,
+      "step": 1660
+    },
+    {
+      "epoch": 0.6429877755318125,
+      "grad_norm": 0.23094922304153442,
+      "learning_rate": 2.9912223245804017e-06,
+      "loss": 0.7494,
+      "step": 1670
+    },
+    {
+      "epoch": 0.6468380017326018,
+      "grad_norm": 0.22274698317050934,
+      "learning_rate": 2.934275938772981e-06,
+      "loss": 0.7423,
+      "step": 1680
+    },
+    {
+      "epoch": 0.6506882279333911,
+      "grad_norm": 0.21643158793449402,
+      "learning_rate": 2.8776508524948676e-06,
+      "loss": 0.7333,
+      "step": 1690
+    },
+    {
+      "epoch": 0.6545384541341804,
+      "grad_norm": 0.19300071895122528,
+      "learning_rate": 2.821355873124503e-06,
+      "loss": 0.7322,
+      "step": 1700
+    },
+    {
+      "epoch": 0.6583886803349697,
+      "grad_norm": 0.22333557903766632,
+      "learning_rate": 2.7653997566960144e-06,
+      "loss": 0.7459,
+      "step": 1710
+    },
+    {
+      "epoch": 0.662238906535759,
+      "grad_norm": 0.22105351090431213,
+      "learning_rate": 2.709791206537309e-06,
+      "loss": 0.7354,
+      "step": 1720
+    },
+    {
+      "epoch": 0.6660891327365482,
+      "grad_norm": 0.21868261694908142,
+      "learning_rate": 2.6545388719163747e-06,
+      "loss": 0.7372,
+      "step": 1730
+    },
+    {
+      "epoch": 0.6699393589373376,
+      "grad_norm": 0.21677537262439728,
+      "learning_rate": 2.599651346695979e-06,
+      "loss": 0.7394,
+      "step": 1740
+    },
+    {
+      "epoch": 0.6737895851381268,
+      "grad_norm": 0.20651423931121826,
+      "learning_rate": 2.545137167997007e-06,
+      "loss": 0.7385,
+      "step": 1750
+    },
+    {
+      "epoch": 0.6776398113389162,
+      "grad_norm": 0.20399147272109985,
+      "learning_rate": 2.4910048148705972e-06,
+      "loss": 0.7256,
+      "step": 1760
+    },
+    {
+      "epoch": 0.6814900375397055,
+      "grad_norm": 0.22342067956924438,
+      "learning_rate": 2.4372627069793327e-06,
+      "loss": 0.7715,
+      "step": 1770
+    },
+    {
+      "epoch": 0.6853402637404947,
+      "grad_norm": 0.22846345603466034,
+      "learning_rate": 2.3839192032876486e-06,
+      "loss": 0.7543,
+      "step": 1780
+    },
+    {
+      "epoch": 0.6891904899412841,
+      "grad_norm": 0.21247322857379913,
+      "learning_rate": 2.3309826007617047e-06,
+      "loss": 0.7642,
+      "step": 1790
+    },
+    {
+      "epoch": 0.6930407161420733,
+      "grad_norm": 0.21544276177883148,
+      "learning_rate": 2.278461133078881e-06,
+      "loss": 0.7288,
+      "step": 1800
+    },
+    {
+      "epoch": 0.6968909423428626,
+      "grad_norm": 0.23202063143253326,
+      "learning_rate": 2.2263629693471258e-06,
+      "loss": 0.7211,
+      "step": 1810
+    },
+    {
+      "epoch": 0.700741168543652,
+      "grad_norm": 0.2046472728252411,
+      "learning_rate": 2.1746962128343414e-06,
+      "loss": 0.7424,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7045913947444412,
+      "grad_norm": 0.20659509301185608,
+      "learning_rate": 2.1234688997080183e-06,
+      "loss": 0.7336,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7084416209452306,
+      "grad_norm": 0.22934786975383759,
+      "learning_rate": 2.0726889977853053e-06,
+      "loss": 0.7392,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7122918471460198,
+      "grad_norm": 0.24095727503299713,
+      "learning_rate": 2.022364405293703e-06,
+      "loss": 0.7502,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7161420733468091,
+      "grad_norm": 0.21410596370697021,
+      "learning_rate": 1.9725029496425814e-06,
+      "loss": 0.7447,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7199922995475985,
+      "grad_norm": 0.1991892158985138,
+      "learning_rate": 1.923112386205727e-06,
+      "loss": 0.7397,
+      "step": 1870
+    },
+    {
+      "epoch": 0.7238425257483877,
+      "grad_norm": 0.21255438029766083,
+      "learning_rate": 1.874200397115079e-06,
+      "loss": 0.7489,
+      "step": 1880
+    },
+    {
+      "epoch": 0.727692751949177,
+      "grad_norm": 0.23104898631572723,
+      "learning_rate": 1.825774590065857e-06,
+      "loss": 0.7311,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7315429781499663,
+      "grad_norm": 0.21840150654315948,
+      "learning_rate": 1.7778424971332813e-06,
+      "loss": 0.7421,
+      "step": 1900
+    },
+    {
+      "epoch": 0.7353932043507556,
+      "grad_norm": 0.2116774171590805,
+      "learning_rate": 1.7304115736010407e-06,
+      "loss": 0.738,
+      "step": 1910
+    },
+    {
+      "epoch": 0.739243430551545,
+      "grad_norm": 0.21480920910835266,
+      "learning_rate": 1.6834891968017108e-06,
+      "loss": 0.7337,
+      "step": 1920
+    },
+    {
+      "epoch": 0.7430936567523342,
+      "grad_norm": 0.22207403182983398,
+      "learning_rate": 1.6370826649692884e-06,
+      "loss": 0.7471,
+      "step": 1930
+    },
+    {
+      "epoch": 0.7469438829531235,
+      "grad_norm": 0.227326899766922,
+      "learning_rate": 1.5911991961040463e-06,
+      "loss": 0.7554,
+      "step": 1940
+    },
+    {
+      "epoch": 0.7507941091539128,
+      "grad_norm": 0.22266922891139984,
+      "learning_rate": 1.545845926849851e-06,
+      "loss": 0.7442,
+      "step": 1950
+    },
+    {
+      "epoch": 0.7546443353547021,
+      "grad_norm": 0.23701070249080658,
+      "learning_rate": 1.5010299113841397e-06,
+      "loss": 0.7604,
+      "step": 1960
+    },
+    {
+      "epoch": 0.7584945615554913,
+      "grad_norm": 0.21560271084308624,
+      "learning_rate": 1.4567581203207238e-06,
+      "loss": 0.7506,
+      "step": 1970
+    },
+    {
+      "epoch": 0.7623447877562807,
+      "grad_norm": 0.22547630965709686,
+      "learning_rate": 1.413037439625593e-06,
+      "loss": 0.7488,
+      "step": 1980
+    },
+    {
+      "epoch": 0.76619501395707,
+      "grad_norm": 0.22960172593593597,
+      "learning_rate": 1.3698746695458847e-06,
+      "loss": 0.7429,
+      "step": 1990
+    },
+    {
+      "epoch": 0.7700452401578592,
+      "grad_norm": 0.21283532679080963,
+      "learning_rate": 1.3272765235521745e-06,
+      "loss": 0.735,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7738954663586486,
+      "grad_norm": 0.21392519772052765,
+      "learning_rate": 1.2852496272942877e-06,
+      "loss": 0.7452,
+      "step": 2010
+    },
+    {
+      "epoch": 0.7777456925594378,
+      "grad_norm": 0.20342864096164703,
+      "learning_rate": 1.2438005175707396e-06,
+      "loss": 0.7309,
+      "step": 2020
+    },
+    {
+      "epoch": 0.7815959187602272,
+      "grad_norm": 0.20789526402950287,
+      "learning_rate": 1.2029356413120297e-06,
+      "loss": 0.7152,
+      "step": 2030
+    },
+    {
+      "epoch": 0.7854461449610165,
+      "grad_norm": 0.21169370412826538,
+      "learning_rate": 1.1626613545778786e-06,
+      "loss": 0.732,
+      "step": 2040
+    },
+    {
+      "epoch": 0.7892963711618057,
+      "grad_norm": 0.206139475107193,
+      "learning_rate": 1.1229839215686294e-06,
+      "loss": 0.7365,
+      "step": 2050
+    },
+    {
+      "epoch": 0.7931465973625951,
+      "grad_norm": 0.20360983908176422,
+      "learning_rate": 1.0839095136509188e-06,
+      "loss": 0.7198,
+      "step": 2060
+    },
+    {
+      "epoch": 0.7969968235633843,
+      "grad_norm": 0.2288307100534439,
+      "learning_rate": 1.045444208397791e-06,
+      "loss": 0.7419,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8008470497641736,
+      "grad_norm": 0.20068387687206268,
+      "learning_rate": 1.007593988643402e-06,
+      "loss": 0.7262,
+      "step": 2080
+    },
+    {
+      "epoch": 0.804697275964963,
+      "grad_norm": 0.20640259981155396,
+      "learning_rate": 9.703647415524613e-07,
+      "loss": 0.7479,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8085475021657522,
+      "grad_norm": 0.21271160244941711,
+      "learning_rate": 9.337622577045485e-07,
+      "loss": 0.7361,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8123977283665416,
+      "grad_norm": 0.2196018099784851,
+      "learning_rate": 8.977922301934588e-07,
+      "loss": 0.7462,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8162479545673308,
+      "grad_norm": 0.22012674808502197,
+      "learning_rate": 8.624602537417009e-07,
+      "loss": 0.7467,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8200981807681201,
+      "grad_norm": 0.23485329747200012,
+      "learning_rate": 8.277718238303123e-07,
+      "loss": 0.7329,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8239484069689095,
+      "grad_norm": 0.21476028859615326,
+      "learning_rate": 7.937323358440935e-07,
+      "loss": 0.7368,
+      "step": 2140
+    },
+    {
+      "epoch": 0.8277986331696987,
+      "grad_norm": 0.23252347111701965,
+      "learning_rate": 7.603470842324223e-07,
+      "loss": 0.728,
+      "step": 2150
+    },
+    {
+      "epoch": 0.831648859370488,
+      "grad_norm": 0.21281921863555908,
+      "learning_rate": 7.276212616857575e-07,
+      "loss": 0.7474,
+      "step": 2160
+    },
+    {
+      "epoch": 0.8354990855712773,
+      "grad_norm": 0.21281103789806366,
+      "learning_rate": 6.955599583279837e-07,
+      "loss": 0.7456,
+      "step": 2170
+    },
+    {
+      "epoch": 0.8393493117720666,
+      "grad_norm": 0.2096223682165146,
+      "learning_rate": 6.641681609246981e-07,
+      "loss": 0.7382,
+      "step": 2180
+    },
+    {
+      "epoch": 0.8431995379728559,
+      "grad_norm": 0.20383666455745697,
+      "learning_rate": 6.334507521075711e-07,
+      "loss": 0.7442,
+      "step": 2190
+    },
+    {
+      "epoch": 0.8470497641736452,
+      "grad_norm": 0.2195318192243576,
+      "learning_rate": 6.034125096149179e-07,
+      "loss": 0.74,
+      "step": 2200
+    },
+    {
+      "epoch": 0.8508999903744345,
+      "grad_norm": 0.2180963158607483,
+      "learning_rate": 5.740581055485722e-07,
+      "loss": 0.711,
+      "step": 2210
+    },
+    {
+      "epoch": 0.8547502165752238,
+      "grad_norm": 0.20202787220478058,
+      "learning_rate": 5.453921056471961e-07,
+      "loss": 0.7426,
+      "step": 2220
+    },
+    {
+      "epoch": 0.8586004427760131,
+      "grad_norm": 0.21528249979019165,
+      "learning_rate": 5.174189685761288e-07,
+      "loss": 0.7402,
+      "step": 2230
+    },
+    {
+      "epoch": 0.8624506689768023,
+      "grad_norm": 0.1991991102695465,
+      "learning_rate": 4.901430452338979e-07,
+      "loss": 0.7508,
+      "step": 2240
+    },
+    {
+      "epoch": 0.8663008951775917,
+      "grad_norm": 0.20598487555980682,
+      "learning_rate": 4.6356857807548286e-07,
+      "loss": 0.734,
+      "step": 2250
+    },
+    {
+      "epoch": 0.870151121378381,
+      "grad_norm": 0.2149941623210907,
+      "learning_rate": 4.376997004524541e-07,
+      "loss": 0.7241,
+      "step": 2260
+    },
+    {
+      "epoch": 0.8740013475791703,
+      "grad_norm": 0.21323302388191223,
+      "learning_rate": 4.12540435970073e-07,
+      "loss": 0.7307,
+      "step": 2270
+    },
+    {
+      "epoch": 0.8778515737799596,
+      "grad_norm": 0.2169979065656662,
+      "learning_rate": 3.880946978614703e-07,
+      "loss": 0.7441,
+      "step": 2280
+    },
+    {
+      "epoch": 0.8817017999807488,
+      "grad_norm": 0.21354342997074127,
+      "learning_rate": 3.643662883789878e-07,
+      "loss": 0.7456,
+      "step": 2290
+    },
+    {
+      "epoch": 0.8855520261815382,
+      "grad_norm": 0.20236074924468994,
+      "learning_rate": 3.41358898202781e-07,
+      "loss": 0.726,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8894022523823275,
+      "grad_norm": 0.19989879429340363,
+      "learning_rate": 3.190761058667741e-07,
+      "loss": 0.7366,
+      "step": 2310
+    },
+    {
+      "epoch": 0.8932524785831167,
+      "grad_norm": 0.2090791016817093,
+      "learning_rate": 2.9752137720206463e-07,
+      "loss": 0.7406,
+      "step": 2320
+    },
+    {
+      "epoch": 0.8971027047839061,
+      "grad_norm": 0.21063074469566345,
+      "learning_rate": 2.7669806479785067e-07,
+      "loss": 0.7463,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9009529309846953,
+      "grad_norm": 0.2124110609292984,
+      "learning_rate": 2.566094074799763e-07,
+      "loss": 0.7386,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9048031571854847,
+      "grad_norm": 0.2137504369020462,
+      "learning_rate": 2.3725852980716557e-07,
+      "loss": 0.7252,
+      "step": 2350
+    },
+    {
+      "epoch": 0.908653383386274,
+      "grad_norm": 0.2108803689479828,
+      "learning_rate": 2.1864844158503983e-07,
+      "loss": 0.7354,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9125036095870632,
+      "grad_norm": 0.22819434106349945,
+      "learning_rate": 2.007820373979702e-07,
+      "loss": 0.7297,
+      "step": 2370
+    },
+    {
+      "epoch": 0.9163538357878526,
+      "grad_norm": 0.2116864174604416,
+      "learning_rate": 1.8366209615886532e-07,
+      "loss": 0.7381,
+      "step": 2380
+    },
+    {
+      "epoch": 0.9202040619886418,
+      "grad_norm": 0.2119535207748413,
+      "learning_rate": 1.6729128067693613e-07,
+      "loss": 0.7502,
+      "step": 2390
+    },
+    {
+      "epoch": 0.9240542881894311,
+      "grad_norm": 0.22711175680160522,
+      "learning_rate": 1.5167213724353426e-07,
+      "loss": 0.7498,
+      "step": 2400
+    },
+    {
+      "epoch": 0.9279045143902205,
+      "grad_norm": 0.21131488680839539,
+      "learning_rate": 1.3680709523610058e-07,
+      "loss": 0.7309,
+      "step": 2410
+    },
+    {
+      "epoch": 0.9317547405910097,
+      "grad_norm": 0.21398335695266724,
+      "learning_rate": 1.2269846674030418e-07,
+      "loss": 0.7543,
+      "step": 2420
+    },
+    {
+      "epoch": 0.935604966791799,
+      "grad_norm": 0.22851786017417908,
+      "learning_rate": 1.0934844619042595e-07,
+      "loss": 0.7488,
+      "step": 2430
+    },
+    {
+      "epoch": 0.9394551929925883,
+      "grad_norm": 0.22577407956123352,
+      "learning_rate": 9.675911002803961e-08,
+      "loss": 0.7374,
+      "step": 2440
+    },
+    {
+      "epoch": 0.9433054191933776,
+      "grad_norm": 0.23182269930839539,
+      "learning_rate": 8.493241637904381e-08,
+      "loss": 0.7346,
+      "step": 2450
+    },
+    {
+      "epoch": 0.9471556453941669,
+      "grad_norm": 0.21514001488685608,
+      "learning_rate": 7.387020474909689e-08,
+      "loss": 0.7538,
+      "step": 2460
+    },
+    {
+      "epoch": 0.9510058715949562,
+      "grad_norm": 0.23438365757465363,
+      "learning_rate": 6.357419573750856e-08,
+      "loss": 0.7499,
+      "step": 2470
+    },
+    {
+      "epoch": 0.9548560977957455,
+      "grad_norm": 0.216669499874115,
+      "learning_rate": 5.4045990769613365e-08,
+      "loss": 0.7307,
+      "step": 2480
+    },
+    {
+      "epoch": 0.9587063239965348,
+      "grad_norm": 0.21037673950195312,
+      "learning_rate": 4.5287071847694434e-08,
+      "loss": 0.7377,
+      "step": 2490
+    },
+    {
+      "epoch": 0.9625565501973241,
+      "grad_norm": 0.20731136202812195,
+      "learning_rate": 3.7298801320467817e-08,
+      "loss": 0.7354,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9664067763981133,
+      "grad_norm": 0.21320414543151855,
+      "learning_rate": 3.0082421671192576e-08,
+      "loss": 0.7289,
+      "step": 2510
+    },
+    {
+      "epoch": 0.9702570025989027,
+      "grad_norm": 0.20740681886672974,
+      "learning_rate": 2.3639055324410355e-08,
+      "loss": 0.7494,
+      "step": 2520
+    },
+    {
+      "epoch": 0.974107228799692,
+      "grad_norm": 0.2134709507226944,
+      "learning_rate": 1.7969704471369477e-08,
+      "loss": 0.7334,
+      "step": 2530
+    },
+    {
+      "epoch": 0.9779574550004813,
+      "grad_norm": 0.2059420943260193,
+      "learning_rate": 1.3075250914142413e-08,
+      "loss": 0.7351,
+      "step": 2540
+    },
+    {
+      "epoch": 0.9818076812012706,
+      "grad_norm": 0.2084321528673172,
+      "learning_rate": 8.956455928476049e-09,
+      "loss": 0.7438,
+      "step": 2550
+    },
+    {
+      "epoch": 0.9856579074020598,
+      "grad_norm": 0.20996084809303284,
+      "learning_rate": 5.6139601453786275e-09,
+      "loss": 0.7364,
+      "step": 2560
+    },
+    {
+      "epoch": 0.9895081336028492,
+      "grad_norm": 0.21059970557689667,
+      "learning_rate": 3.0482834514805602e-09,
+      "loss": 0.7266,
+      "step": 2570
+    },
+    {
+      "epoch": 0.9933583598036385,
+      "grad_norm": 0.20549727976322174,
+      "learning_rate": 1.2598249081713366e-09,
+      "loss": 0.7177,
+      "step": 2580
+    },
+    {
+      "epoch": 0.9972085860044277,
+      "grad_norm": 0.2230440378189087,
+      "learning_rate": 2.488626895302826e-10,
+      "loss": 0.7523,
+      "step": 2590
+    },
+    {
+      "epoch": 0.9999037443449803,
+      "step": 2597,
+      "total_flos": 4.066856054212788e+18,
+      "train_loss": 0.7590124202224443,
+      "train_runtime": 8620.4238,
+      "train_samples_per_second": 77.131,
+      "train_steps_per_second": 0.301
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2597,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.066856054212788e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen25/clip_meta__l14/finetune/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db0e26cb4b7c2031e6451d267827894fe1dc57165e5d3f818d5c4671ec835827
+size 6136

continuous/qwen25/clip_meta__l14/pretrain/config.json

@@ -0,0 +1,58 @@
+{
+  "_attn_implementation_autoset": true,
+  "add_faster_video": false,
+  "add_time_instruction": false,
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "faster_token_stride": 10,
+  "force_sample": false,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "image_aspect_ratio": "square",
+  "image_crop_resolution": null,
+  "image_grid_pinpoints": null,
+  "image_split_resolution": null,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "mm_hidden_size": 1024,
+  "mm_newline_position": "grid",
+  "mm_patch_merge_type": "flat",
+  "mm_projector_lr": null,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_resampler_type": null,
+  "mm_spatial_pool_mode": "bilinear",
+  "mm_spatial_pool_stride": null,
+  "mm_tunable_parts": "mm_mlp_adapter",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -2,
+  "mm_vision_tower": "open_clip_hub:ViT-L-14",
+  "mm_vision_tower_lr": null,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "pos_skipping_range": 4096,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "tokenizer_model_max_length": 8192,
+  "tokenizer_padding_side": "right",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "use_mm_proj": true,
+  "use_pos_skipping": false,
+  "use_sliding_window": false,
+  "vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/vit_l14_metaclip.safetensors",
+  "vocab_size": 151936
+}

continuous/qwen25/clip_meta__l14/pretrain/mm_projector.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d817d08dc32a6e7ff833c9ff526d70e008cf73c7f99b3d50e94ca59f1576291
+size 15743032

continuous/qwen25/clip_meta__l14/pretrain/trainer_state.json

@@ -0,0 +1,421 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 545,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01834862385321101,
+      "grad_norm": 4.802586555480957,
+      "learning_rate": 0.00032142857142857147,
+      "loss": 4.8043,
+      "step": 10
+    },
+    {
+      "epoch": 0.03669724770642202,
+      "grad_norm": 1.4347777366638184,
+      "learning_rate": 0.0006785714285714287,
+      "loss": 3.9991,
+      "step": 20
+    },
+    {
+      "epoch": 0.05504587155963303,
+      "grad_norm": 0.26819685101509094,
+      "learning_rate": 0.0009999907688176173,
+      "loss": 3.4423,
+      "step": 30
+    },
+    {
+      "epoch": 0.07339449541284404,
+      "grad_norm": 0.15500269830226898,
+      "learning_rate": 0.0009988834393115766,
+      "loss": 3.2685,
+      "step": 40
+    },
+    {
+      "epoch": 0.09174311926605505,
+      "grad_norm": 0.23330330848693848,
+      "learning_rate": 0.0009959345572940772,
+      "loss": 3.2286,
+      "step": 50
+    },
+    {
+      "epoch": 0.11009174311926606,
+      "grad_norm": 0.14809183776378632,
+      "learning_rate": 0.0009911550081155983,
+      "loss": 3.1993,
+      "step": 60
+    },
+    {
+      "epoch": 0.12844036697247707,
+      "grad_norm": 0.11319894343614578,
+      "learning_rate": 0.0009845624347566433,
+      "loss": 3.1795,
+      "step": 70
+    },
+    {
+      "epoch": 0.14678899082568808,
+      "grad_norm": 0.1879793256521225,
+      "learning_rate": 0.0009761811727013547,
+      "loss": 3.1635,
+      "step": 80
+    },
+    {
+      "epoch": 0.1651376146788991,
+      "grad_norm": 0.11530029773712158,
+      "learning_rate": 0.0009660421601067666,
+      "loss": 3.1353,
+      "step": 90
+    },
+    {
+      "epoch": 0.1834862385321101,
+      "grad_norm": 0.3522357940673828,
+      "learning_rate": 0.0009541828235992926,
+      "loss": 3.1092,
+      "step": 100
+    },
+    {
+      "epoch": 0.2018348623853211,
+      "grad_norm": 0.23432326316833496,
+      "learning_rate": 0.0009406469401200151,
+      "loss": 3.0616,
+      "step": 110
+    },
+    {
+      "epoch": 0.22018348623853212,
+      "grad_norm": 0.17483241856098175,
+      "learning_rate": 0.0009254844753287492,
+      "loss": 3.0303,
+      "step": 120
+    },
+    {
+      "epoch": 0.23853211009174313,
+      "grad_norm": 0.17836016416549683,
+      "learning_rate": 0.0009087513991633924,
+      "loss": 2.9721,
+      "step": 130
+    },
+    {
+      "epoch": 0.25688073394495414,
+      "grad_norm": 0.16639737784862518,
+      "learning_rate": 0.0008905094792353917,
+      "loss": 2.9394,
+      "step": 140
+    },
+    {
+      "epoch": 0.27522935779816515,
+      "grad_norm": 0.15990173816680908,
+      "learning_rate": 0.0008708260528239789,
+      "loss": 2.9136,
+      "step": 150
+    },
+    {
+      "epoch": 0.29357798165137616,
+      "grad_norm": 0.19283396005630493,
+      "learning_rate": 0.0008497737783108237,
+      "loss": 2.8736,
+      "step": 160
+    },
+    {
+      "epoch": 0.3119266055045872,
+      "grad_norm": 0.16985860466957092,
+      "learning_rate": 0.0008274303669726426,
+      "loss": 2.8547,
+      "step": 170
+    },
+    {
+      "epoch": 0.3302752293577982,
+      "grad_norm": 0.17568475008010864,
+      "learning_rate": 0.0008038782961218137,
+      "loss": 2.8233,
+      "step": 180
+    },
+    {
+      "epoch": 0.3486238532110092,
+      "grad_norm": 0.1817346215248108,
+      "learning_rate": 0.000779204504653894,
+      "loss": 2.8071,
+      "step": 190
+    },
+    {
+      "epoch": 0.3669724770642202,
+      "grad_norm": 0.15953803062438965,
+      "learning_rate": 0.0007535000721258791,
+      "loss": 2.7746,
+      "step": 200
+    },
+    {
+      "epoch": 0.3853211009174312,
+      "grad_norm": 0.21847215294837952,
+      "learning_rate": 0.00072685988254984,
+      "loss": 2.752,
+      "step": 210
+    },
+    {
+      "epoch": 0.4036697247706422,
+      "grad_norm": 0.17570634186267853,
+      "learning_rate": 0.0006993822741429906,
+      "loss": 2.7397,
+      "step": 220
+    },
+    {
+      "epoch": 0.42201834862385323,
+      "grad_norm": 0.15951746702194214,
+      "learning_rate": 0.0006711686763270818,
+      "loss": 2.7069,
+      "step": 230
+    },
+    {
+      "epoch": 0.44036697247706424,
+      "grad_norm": 0.1350003331899643,
+      "learning_rate": 0.0006423232353170831,
+      "loss": 2.6742,
+      "step": 240
+    },
+    {
+      "epoch": 0.45871559633027525,
+      "grad_norm": 0.15698154270648956,
+      "learning_rate": 0.0006129524296812335,
+      "loss": 2.6582,
+      "step": 250
+    },
+    {
+      "epoch": 0.47706422018348627,
+      "grad_norm": 0.15252625942230225,
+      "learning_rate": 0.0005831646772915651,
+      "loss": 2.6523,
+      "step": 260
+    },
+    {
+      "epoch": 0.4954128440366973,
+      "grad_norm": 0.13270673155784607,
+      "learning_rate": 0.0005530699351157809,
+      "loss": 2.6259,
+      "step": 270
+    },
+    {
+      "epoch": 0.5137614678899083,
+      "grad_norm": 0.1841563731431961,
+      "learning_rate": 0.0005227792933277943,
+      "loss": 2.6275,
+      "step": 280
+    },
+    {
+      "epoch": 0.5321100917431193,
+      "grad_norm": 0.24035830795764923,
+      "learning_rate": 0.0004924045652352092,
+      "loss": 2.6051,
+      "step": 290
+    },
+    {
+      "epoch": 0.5504587155963303,
+      "grad_norm": 0.17008879780769348,
+      "learning_rate": 0.0004620578745374633,
+      "loss": 2.6042,
+      "step": 300
+    },
+    {
+      "epoch": 0.5688073394495413,
+      "grad_norm": 0.16404852271080017,
+      "learning_rate": 0.0004318512414382059,
+      "loss": 2.6053,
+      "step": 310
+    },
+    {
+      "epoch": 0.5871559633027523,
+      "grad_norm": 0.15697196125984192,
+      "learning_rate": 0.0004018961691397148,
+      "loss": 2.6004,
+      "step": 320
+    },
+    {
+      "epoch": 0.6055045871559633,
+      "grad_norm": 0.20189659297466278,
+      "learning_rate": 0.0003723032322457458,
+      "loss": 2.5841,
+      "step": 330
+    },
+    {
+      "epoch": 0.6238532110091743,
+      "grad_norm": 0.14686717092990875,
+      "learning_rate": 0.0003431816685921625,
+      "loss": 2.5857,
+      "step": 340
+    },
+    {
+      "epoch": 0.6422018348623854,
+      "grad_norm": 0.13427212834358215,
+      "learning_rate": 0.0003146389760120416,
+      "loss": 2.5819,
+      "step": 350
+    },
+    {
+      "epoch": 0.6605504587155964,
+      "grad_norm": 0.14426584541797638,
+      "learning_rate": 0.00028678051552373485,
+      "loss": 2.5834,
+      "step": 360
+    },
+    {
+      "epoch": 0.6788990825688074,
+      "grad_norm": 0.11759243905544281,
+      "learning_rate": 0.0002597091224066581,
+      "loss": 2.5726,
+      "step": 370
+    },
+    {
+      "epoch": 0.6972477064220184,
+      "grad_norm": 0.12667301297187805,
+      "learning_rate": 0.0002335247266004629,
+      "loss": 2.5794,
+      "step": 380
+    },
+    {
+      "epoch": 0.7155963302752294,
+      "grad_norm": 0.16403713822364807,
+      "learning_rate": 0.000208323983828828,
+      "loss": 2.5709,
+      "step": 390
+    },
+    {
+      "epoch": 0.7339449541284404,
+      "grad_norm": 0.12070130556821823,
+      "learning_rate": 0.0001841999188095224,
+      "loss": 2.5624,
+      "step": 400
+    },
+    {
+      "epoch": 0.7522935779816514,
+      "grad_norm": 0.14710962772369385,
+      "learning_rate": 0.00016124158186777676,
+      "loss": 2.5648,
+      "step": 410
+    },
+    {
+      "epoch": 0.7706422018348624,
+      "grad_norm": 0.13482585549354553,
+      "learning_rate": 0.0001395337202205211,
+      "loss": 2.5614,
+      "step": 420
+    },
+    {
+      "epoch": 0.7889908256880734,
+      "grad_norm": 0.1419251412153244,
+      "learning_rate": 0.00011915646514489292,
+      "loss": 2.5572,
+      "step": 430
+    },
+    {
+      "epoch": 0.8073394495412844,
+      "grad_norm": 0.12650972604751587,
+      "learning_rate": 0.00010018503618578818,
+      "loss": 2.5689,
+      "step": 440
+    },
+    {
+      "epoch": 0.8256880733944955,
+      "grad_norm": 0.1585581749677658,
+      "learning_rate": 8.268946349432583e-05,
+      "loss": 2.5623,
+      "step": 450
+    },
+    {
+      "epoch": 0.8440366972477065,
+      "grad_norm": 0.11538032442331314,
+      "learning_rate": 6.673432932217117e-05,
+      "loss": 2.5521,
+      "step": 460
+    },
+    {
+      "epoch": 0.8623853211009175,
+      "grad_norm": 0.12863601744174957,
+      "learning_rate": 5.2378529625954684e-05,
+      "loss": 2.5438,
+      "step": 470
+    },
+    {
+      "epoch": 0.8807339449541285,
+      "grad_norm": 0.1361401379108429,
+      "learning_rate": 3.9675056661785556e-05,
+      "loss": 2.5485,
+      "step": 480
+    },
+    {
+      "epoch": 0.8990825688073395,
+      "grad_norm": 0.12866006791591644,
+      "learning_rate": 2.8670803372379538e-05,
+      "loss": 2.5583,
+      "step": 490
+    },
+    {
+      "epoch": 0.9174311926605505,
+      "grad_norm": 0.11287451535463333,
+      "learning_rate": 1.9406390288876586e-05,
+      "loss": 2.5596,
+      "step": 500
+    },
+    {
+      "epoch": 0.9357798165137615,
+      "grad_norm": 0.12047653645277023,
+      "learning_rate": 1.191601558631461e-05,
+      "loss": 2.5535,
+      "step": 510
+    },
+    {
+      "epoch": 0.9541284403669725,
+      "grad_norm": 0.1164703443646431,
+      "learning_rate": 6.2273288462559305e-06,
+      "loss": 2.5495,
+      "step": 520
+    },
+    {
+      "epoch": 0.9724770642201835,
+      "grad_norm": 0.10380826890468597,
+      "learning_rate": 2.361328992552314e-06,
+      "loss": 2.5512,
+      "step": 530
+    },
+    {
+      "epoch": 0.9908256880733946,
+      "grad_norm": 0.12064791470766068,
+      "learning_rate": 3.322867770044202e-07,
+      "loss": 2.5503,
+      "step": 540
+    },
+    {
+      "epoch": 1.0,
+      "step": 545,
+      "total_flos": 3.2179952270966784e+17,
+      "train_loss": 2.8096533521599727,
+      "train_runtime": 5291.3094,
+      "train_samples_per_second": 105.48,
+      "train_steps_per_second": 0.103
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 545,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.2179952270966784e+17,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen25/clip_openai__l14/finetune/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
+}

continuous/qwen25/clip_openai__l14/finetune/config.json

@@ -0,0 +1,57 @@
+{
+  "add_faster_video": false,
+  "add_time_instruction": false,
+  "architectures": [
+    "LlavaQwenForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "faster_token_stride": 10,
+  "force_sample": false,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "image_aspect_ratio": "square",
+  "image_crop_resolution": null,
+  "image_grid_pinpoints": null,
+  "image_split_resolution": null,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "mm_hidden_size": 1024,
+  "mm_newline_position": "grid",
+  "mm_patch_merge_type": "flat",
+  "mm_projector_lr": 0.0001,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_resampler_type": null,
+  "mm_spatial_pool_mode": "bilinear",
+  "mm_spatial_pool_stride": null,
+  "mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -2,
+  "mm_vision_tower": "/cache/ckpt/download/tokenizer/continuous/clip-vit-large-patch14",
+  "mm_vision_tower_lr": null,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "pos_skipping_range": 4096,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "tokenizer_model_max_length": 8192,
+  "tokenizer_padding_side": "right",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "use_mm_proj": true,
+  "use_pos_skipping": false,
+  "use_sliding_window": false,
+  "vision_tower_pretrained": null,
+  "vocab_size": 151936
+}

continuous/qwen25/clip_openai__l14/finetune/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": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.51.3"
+}

continuous/qwen25/clip_openai__l14/finetune/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2c11136fd6e2173345f1de3e3e79868b71173bd632b4a53e383a700c423c31f
+size 4176377000

continuous/qwen25/clip_openai__l14/finetune/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
+  }
+}

continuous/qwen25/clip_openai__l14/finetune/tokenizer.json

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

continuous/qwen25/clip_openai__l14/finetune/tokenizer_config.json

@@ -0,0 +1,209 @@
+{
+  "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": 8192,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

continuous/qwen25/clip_openai__l14/finetune/trainer_state.json

@@ -0,0 +1,1856 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9999037443449803,
+  "eval_steps": 500,
+  "global_step": 2597,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.003850226200789296,
+      "grad_norm": 15.072680473327637,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 1.9475,
+      "step": 10
+    },
+    {
+      "epoch": 0.007700452401578592,
+      "grad_norm": 6.510941505432129,
+      "learning_rate": 2.435897435897436e-06,
+      "loss": 1.5907,
+      "step": 20
+    },
+    {
+      "epoch": 0.01155067860236789,
+      "grad_norm": 3.687537431716919,
+      "learning_rate": 3.7179487179487184e-06,
+      "loss": 1.1308,
+      "step": 30
+    },
+    {
+      "epoch": 0.015400904803157185,
+      "grad_norm": 0.7150320410728455,
+      "learning_rate": 5e-06,
+      "loss": 0.9375,
+      "step": 40
+    },
+    {
+      "epoch": 0.019251131003946483,
+      "grad_norm": 0.36461079120635986,
+      "learning_rate": 6.282051282051282e-06,
+      "loss": 0.9014,
+      "step": 50
+    },
+    {
+      "epoch": 0.02310135720473578,
+      "grad_norm": 0.27750125527381897,
+      "learning_rate": 7.564102564102564e-06,
+      "loss": 0.853,
+      "step": 60
+    },
+    {
+      "epoch": 0.026951583405525073,
+      "grad_norm": 0.23068949580192566,
+      "learning_rate": 8.846153846153847e-06,
+      "loss": 0.8267,
+      "step": 70
+    },
+    {
+      "epoch": 0.03080180960631437,
+      "grad_norm": 0.23514236509799957,
+      "learning_rate": 9.999996111488723e-06,
+      "loss": 0.8341,
+      "step": 80
+    },
+    {
+      "epoch": 0.034652035807103666,
+      "grad_norm": 0.21340784430503845,
+      "learning_rate": 9.999529497453782e-06,
+      "loss": 0.8158,
+      "step": 90
+    },
+    {
+      "epoch": 0.038502262007892966,
+      "grad_norm": 0.2282683402299881,
+      "learning_rate": 9.998285264324193e-06,
+      "loss": 0.7936,
+      "step": 100
+    },
+    {
+      "epoch": 0.04235248820868226,
+      "grad_norm": 0.20050622522830963,
+      "learning_rate": 9.996263605626052e-06,
+      "loss": 0.7848,
+      "step": 110
+    },
+    {
+      "epoch": 0.04620271440947156,
+      "grad_norm": 0.2283468395471573,
+      "learning_rate": 9.99346483580503e-06,
+      "loss": 0.797,
+      "step": 120
+    },
+    {
+      "epoch": 0.05005294061026085,
+      "grad_norm": 0.20367617905139923,
+      "learning_rate": 9.989889390177467e-06,
+      "loss": 0.7846,
+      "step": 130
+    },
+    {
+      "epoch": 0.053903166811050146,
+      "grad_norm": 0.23123222589492798,
+      "learning_rate": 9.985537824862647e-06,
+      "loss": 0.7887,
+      "step": 140
+    },
+    {
+      "epoch": 0.057753393011839446,
+      "grad_norm": 0.2338813692331314,
+      "learning_rate": 9.980410816696317e-06,
+      "loss": 0.7695,
+      "step": 150
+    },
+    {
+      "epoch": 0.06160361921262874,
+      "grad_norm": 0.22110392153263092,
+      "learning_rate": 9.97450916312541e-06,
+      "loss": 0.7793,
+      "step": 160
+    },
+    {
+      "epoch": 0.06545384541341803,
+      "grad_norm": 0.19178803265094757,
+      "learning_rate": 9.967833782084004e-06,
+      "loss": 0.7802,
+      "step": 170
+    },
+    {
+      "epoch": 0.06930407161420733,
+      "grad_norm": 0.22627422213554382,
+      "learning_rate": 9.960385711850554e-06,
+      "loss": 0.8047,
+      "step": 180
+    },
+    {
+      "epoch": 0.07315429781499663,
+      "grad_norm": 0.22268758714199066,
+      "learning_rate": 9.952166110886394e-06,
+      "loss": 0.7976,
+      "step": 190
+    },
+    {
+      "epoch": 0.07700452401578593,
+      "grad_norm": 0.23791377246379852,
+      "learning_rate": 9.943176257655567e-06,
+      "loss": 0.7883,
+      "step": 200
+    },
+    {
+      "epoch": 0.08085475021657522,
+      "grad_norm": 0.21042615175247192,
+      "learning_rate": 9.933417550425952e-06,
+      "loss": 0.7835,
+      "step": 210
+    },
+    {
+      "epoch": 0.08470497641736452,
+      "grad_norm": 0.21256029605865479,
+      "learning_rate": 9.9228915070518e-06,
+      "loss": 0.7674,
+      "step": 220
+    },
+    {
+      "epoch": 0.08855520261815382,
+      "grad_norm": 0.21929651498794556,
+      "learning_rate": 9.911599764737632e-06,
+      "loss": 0.7713,
+      "step": 230
+    },
+    {
+      "epoch": 0.09240542881894312,
+      "grad_norm": 0.2209857851266861,
+      "learning_rate": 9.899544079783606e-06,
+      "loss": 0.7646,
+      "step": 240
+    },
+    {
+      "epoch": 0.0962556550197324,
+      "grad_norm": 0.2210368663072586,
+      "learning_rate": 9.886726327312338e-06,
+      "loss": 0.7813,
+      "step": 250
+    },
+    {
+      "epoch": 0.1001058812205217,
+      "grad_norm": 0.21067795157432556,
+      "learning_rate": 9.87314850097724e-06,
+      "loss": 0.7789,
+      "step": 260
+    },
+    {
+      "epoch": 0.103956107421311,
+      "grad_norm": 0.20645610988140106,
+      "learning_rate": 9.858812712652452e-06,
+      "loss": 0.7684,
+      "step": 270
+    },
+    {
+      "epoch": 0.10780633362210029,
+      "grad_norm": 0.20018301904201508,
+      "learning_rate": 9.843721192104341e-06,
+      "loss": 0.7655,
+      "step": 280
+    },
+    {
+      "epoch": 0.11165655982288959,
+      "grad_norm": 0.20920097827911377,
+      "learning_rate": 9.827876286644698e-06,
+      "loss": 0.7876,
+      "step": 290
+    },
+    {
+      "epoch": 0.11550678602367889,
+      "grad_norm": 0.20037592947483063,
+      "learning_rate": 9.811280460765643e-06,
+      "loss": 0.75,
+      "step": 300
+    },
+    {
+      "epoch": 0.11935701222446819,
+      "grad_norm": 0.21144743263721466,
+      "learning_rate": 9.793936295756292e-06,
+      "loss": 0.7581,
+      "step": 310
+    },
+    {
+      "epoch": 0.12320723842525748,
+      "grad_norm": 0.21567706763744354,
+      "learning_rate": 9.775846489301279e-06,
+      "loss": 0.7627,
+      "step": 320
+    },
+    {
+      "epoch": 0.12705746462604678,
+      "grad_norm": 0.22188110649585724,
+      "learning_rate": 9.757013855061157e-06,
+      "loss": 0.7604,
+      "step": 330
+    },
+    {
+      "epoch": 0.13090769082683607,
+      "grad_norm": 0.21025799214839935,
+      "learning_rate": 9.737441322234758e-06,
+      "loss": 0.7741,
+      "step": 340
+    },
+    {
+      "epoch": 0.13475791702762538,
+      "grad_norm": 0.2189192771911621,
+      "learning_rate": 9.717131935103605e-06,
+      "loss": 0.7735,
+      "step": 350
+    },
+    {
+      "epoch": 0.13860814322841467,
+      "grad_norm": 0.21292485296726227,
+      "learning_rate": 9.696088852558398e-06,
+      "loss": 0.76,
+      "step": 360
+    },
+    {
+      "epoch": 0.14245836942920398,
+      "grad_norm": 0.21283544600009918,
+      "learning_rate": 9.674315347607686e-06,
+      "loss": 0.7861,
+      "step": 370
+    },
+    {
+      "epoch": 0.14630859562999327,
+      "grad_norm": 0.21176132559776306,
+      "learning_rate": 9.651814806868796e-06,
+      "loss": 0.7515,
+      "step": 380
+    },
+    {
+      "epoch": 0.15015882183078255,
+      "grad_norm": 0.22220370173454285,
+      "learning_rate": 9.62859073004107e-06,
+      "loss": 0.7529,
+      "step": 390
+    },
+    {
+      "epoch": 0.15400904803157187,
+      "grad_norm": 0.1980036199092865,
+      "learning_rate": 9.604646729361547e-06,
+      "loss": 0.7577,
+      "step": 400
+    },
+    {
+      "epoch": 0.15785927423236115,
+      "grad_norm": 0.20259426534175873,
+      "learning_rate": 9.579986529043099e-06,
+      "loss": 0.7477,
+      "step": 410
+    },
+    {
+      "epoch": 0.16170950043315044,
+      "grad_norm": 0.21437054872512817,
+      "learning_rate": 9.554613964695189e-06,
+      "loss": 0.7651,
+      "step": 420
+    },
+    {
+      "epoch": 0.16555972663393975,
+      "grad_norm": 0.21797312796115875,
+      "learning_rate": 9.528532982727281e-06,
+      "loss": 0.7584,
+      "step": 430
+    },
+    {
+      "epoch": 0.16940995283472904,
+      "grad_norm": 0.20009411871433258,
+      "learning_rate": 9.501747639735018e-06,
+      "loss": 0.7544,
+      "step": 440
+    },
+    {
+      "epoch": 0.17326017903551832,
+      "grad_norm": 0.21109560132026672,
+      "learning_rate": 9.474262101869272e-06,
+      "loss": 0.744,
+      "step": 450
+    },
+    {
+      "epoch": 0.17711040523630764,
+      "grad_norm": 0.2137918472290039,
+      "learning_rate": 9.446080644188141e-06,
+      "loss": 0.7501,
+      "step": 460
+    },
+    {
+      "epoch": 0.18096063143709692,
+      "grad_norm": 0.2067296802997589,
+      "learning_rate": 9.417207649992017e-06,
+      "loss": 0.7554,
+      "step": 470
+    },
+    {
+      "epoch": 0.18481085763788624,
+      "grad_norm": 0.21451012790203094,
+      "learning_rate": 9.387647610141816e-06,
+      "loss": 0.7528,
+      "step": 480
+    },
+    {
+      "epoch": 0.18866108383867553,
+      "grad_norm": 0.19431039690971375,
+      "learning_rate": 9.35740512236047e-06,
+      "loss": 0.7411,
+      "step": 490
+    },
+    {
+      "epoch": 0.1925113100394648,
+      "grad_norm": 0.19407013058662415,
+      "learning_rate": 9.326484890517809e-06,
+      "loss": 0.7527,
+      "step": 500
+    },
+    {
+      "epoch": 0.19636153624025413,
+      "grad_norm": 0.21351060271263123,
+      "learning_rate": 9.294891723898926e-06,
+      "loss": 0.742,
+      "step": 510
+    },
+    {
+      "epoch": 0.2002117624410434,
+      "grad_norm": 0.2252502143383026,
+      "learning_rate": 9.262630536456153e-06,
+      "loss": 0.7714,
+      "step": 520
+    },
+    {
+      "epoch": 0.2040619886418327,
+      "grad_norm": 0.22209516167640686,
+      "learning_rate": 9.229706346044749e-06,
+      "loss": 0.7462,
+      "step": 530
+    },
+    {
+      "epoch": 0.207912214842622,
+      "grad_norm": 0.2265491783618927,
+      "learning_rate": 9.196124273642422e-06,
+      "loss": 0.7573,
+      "step": 540
+    },
+    {
+      "epoch": 0.2117624410434113,
+      "grad_norm": 0.20315907895565033,
+      "learning_rate": 9.161889542552842e-06,
+      "loss": 0.7632,
+      "step": 550
+    },
+    {
+      "epoch": 0.21561266724420058,
+      "grad_norm": 0.1986212581396103,
+      "learning_rate": 9.127007477593193e-06,
+      "loss": 0.7629,
+      "step": 560
+    },
+    {
+      "epoch": 0.2194628934449899,
+      "grad_norm": 0.22537949681282043,
+      "learning_rate": 9.091483504265974e-06,
+      "loss": 0.7714,
+      "step": 570
+    },
+    {
+      "epoch": 0.22331311964577918,
+      "grad_norm": 0.22120918333530426,
+      "learning_rate": 9.055323147915116e-06,
+      "loss": 0.7527,
+      "step": 580
+    },
+    {
+      "epoch": 0.2271633458465685,
+      "grad_norm": 0.211568683385849,
+      "learning_rate": 9.018532032866585e-06,
+      "loss": 0.7619,
+      "step": 590
+    },
+    {
+      "epoch": 0.23101357204735778,
+      "grad_norm": 0.20857249200344086,
+      "learning_rate": 8.981115881553579e-06,
+      "loss": 0.7545,
+      "step": 600
+    },
+    {
+      "epoch": 0.23486379824814707,
+      "grad_norm": 0.22424684464931488,
+      "learning_rate": 8.943080513626476e-06,
+      "loss": 0.7748,
+      "step": 610
+    },
+    {
+      "epoch": 0.23871402444893638,
+      "grad_norm": 0.2085992395877838,
+      "learning_rate": 8.904431845047641e-06,
+      "loss": 0.7593,
+      "step": 620
+    },
+    {
+      "epoch": 0.24256425064972567,
+      "grad_norm": 0.21925665438175201,
+      "learning_rate": 8.86517588717128e-06,
+      "loss": 0.7747,
+      "step": 630
+    },
+    {
+      "epoch": 0.24641447685051496,
+      "grad_norm": 0.20848813652992249,
+      "learning_rate": 8.82531874580844e-06,
+      "loss": 0.7685,
+      "step": 640
+    },
+    {
+      "epoch": 0.25026470305130427,
+      "grad_norm": 0.19311203062534332,
+      "learning_rate": 8.784866620277314e-06,
+      "loss": 0.7254,
+      "step": 650
+    },
+    {
+      "epoch": 0.25411492925209356,
+      "grad_norm": 0.22552074491977692,
+      "learning_rate": 8.743825802439018e-06,
+      "loss": 0.7553,
+      "step": 660
+    },
+    {
+      "epoch": 0.25796515545288284,
+      "grad_norm": 0.23058241605758667,
+      "learning_rate": 8.702202675718956e-06,
+      "loss": 0.7665,
+      "step": 670
+    },
+    {
+      "epoch": 0.26181538165367213,
+      "grad_norm": 0.20083573460578918,
+      "learning_rate": 8.660003714113959e-06,
+      "loss": 0.7331,
+      "step": 680
+    },
+    {
+      "epoch": 0.26566560785446147,
+      "grad_norm": 0.20828908681869507,
+      "learning_rate": 8.617235481185323e-06,
+      "loss": 0.7544,
+      "step": 690
+    },
+    {
+      "epoch": 0.26951583405525076,
+      "grad_norm": 0.21839794516563416,
+      "learning_rate": 8.573904629037931e-06,
+      "loss": 0.7618,
+      "step": 700
+    },
+    {
+      "epoch": 0.27336606025604004,
+      "grad_norm": 0.21304471790790558,
+      "learning_rate": 8.530017897285585e-06,
+      "loss": 0.7441,
+      "step": 710
+    },
+    {
+      "epoch": 0.27721628645682933,
+      "grad_norm": 0.1948530673980713,
+      "learning_rate": 8.485582112002746e-06,
+      "loss": 0.7499,
+      "step": 720
+    },
+    {
+      "epoch": 0.2810665126576186,
+      "grad_norm": 0.20722390711307526,
+      "learning_rate": 8.440604184662811e-06,
+      "loss": 0.7509,
+      "step": 730
+    },
+    {
+      "epoch": 0.28491673885840796,
+      "grad_norm": 0.2130240648984909,
+      "learning_rate": 8.395091111063118e-06,
+      "loss": 0.7414,
+      "step": 740
+    },
+    {
+      "epoch": 0.28876696505919724,
+      "grad_norm": 0.1983703076839447,
+      "learning_rate": 8.349049970236822e-06,
+      "loss": 0.7429,
+      "step": 750
+    },
+    {
+      "epoch": 0.29261719125998653,
+      "grad_norm": 0.2101503610610962,
+      "learning_rate": 8.302487923351842e-06,
+      "loss": 0.7333,
+      "step": 760
+    },
+    {
+      "epoch": 0.2964674174607758,
+      "grad_norm": 0.20597362518310547,
+      "learning_rate": 8.255412212597021e-06,
+      "loss": 0.7513,
+      "step": 770
+    },
+    {
+      "epoch": 0.3003176436615651,
+      "grad_norm": 0.19798938930034637,
+      "learning_rate": 8.207830160055688e-06,
+      "loss": 0.7482,
+      "step": 780
+    },
+    {
+      "epoch": 0.3041678698623544,
+      "grad_norm": 0.21805530786514282,
+      "learning_rate": 8.159749166566786e-06,
+      "loss": 0.7341,
+      "step": 790
+    },
+    {
+      "epoch": 0.30801809606314373,
+      "grad_norm": 0.2040349245071411,
+      "learning_rate": 8.11117671057377e-06,
+      "loss": 0.7605,
+      "step": 800
+    },
+    {
+      "epoch": 0.311868322263933,
+      "grad_norm": 0.222140371799469,
+      "learning_rate": 8.062120346961404e-06,
+      "loss": 0.7529,
+      "step": 810
+    },
+    {
+      "epoch": 0.3157185484647223,
+      "grad_norm": 0.19525088369846344,
+      "learning_rate": 8.012587705880705e-06,
+      "loss": 0.747,
+      "step": 820
+    },
+    {
+      "epoch": 0.3195687746655116,
+      "grad_norm": 0.20610684156417847,
+      "learning_rate": 7.962586491562146e-06,
+      "loss": 0.7326,
+      "step": 830
+    },
+    {
+      "epoch": 0.3234190008663009,
+      "grad_norm": 0.2197851538658142,
+      "learning_rate": 7.912124481117357e-06,
+      "loss": 0.7386,
+      "step": 840
+    },
+    {
+      "epoch": 0.3272692270670902,
+      "grad_norm": 0.20672369003295898,
+      "learning_rate": 7.861209523329494e-06,
+      "loss": 0.7517,
+      "step": 850
+    },
+    {
+      "epoch": 0.3311194532678795,
+      "grad_norm": 0.2138279229402542,
+      "learning_rate": 7.809849537432432e-06,
+      "loss": 0.7336,
+      "step": 860
+    },
+    {
+      "epoch": 0.3349696794686688,
+      "grad_norm": 0.20638605952262878,
+      "learning_rate": 7.758052511879035e-06,
+      "loss": 0.7445,
+      "step": 870
+    },
+    {
+      "epoch": 0.3388199056694581,
+      "grad_norm": 0.19783490896224976,
+      "learning_rate": 7.705826503098642e-06,
+      "loss": 0.7403,
+      "step": 880
+    },
+    {
+      "epoch": 0.34267013187024736,
+      "grad_norm": 0.2015966773033142,
+      "learning_rate": 7.653179634243979e-06,
+      "loss": 0.7396,
+      "step": 890
+    },
+    {
+      "epoch": 0.34652035807103665,
+      "grad_norm": 0.19495956599712372,
+      "learning_rate": 7.600120093927696e-06,
+      "loss": 0.742,
+      "step": 900
+    },
+    {
+      "epoch": 0.350370584271826,
+      "grad_norm": 0.21907925605773926,
+      "learning_rate": 7.546656134948729e-06,
+      "loss": 0.7268,
+      "step": 910
+    },
+    {
+      "epoch": 0.3542208104726153,
+      "grad_norm": 0.22091691195964813,
+      "learning_rate": 7.492796073008661e-06,
+      "loss": 0.7322,
+      "step": 920
+    },
+    {
+      "epoch": 0.35807103667340456,
+      "grad_norm": 0.2214176505804062,
+      "learning_rate": 7.438548285418326e-06,
+      "loss": 0.72,
+      "step": 930
+    },
+    {
+      "epoch": 0.36192126287419385,
+      "grad_norm": 0.23300231993198395,
+      "learning_rate": 7.3839212097948e-06,
+      "loss": 0.7538,
+      "step": 940
+    },
+    {
+      "epoch": 0.36577148907498314,
+      "grad_norm": 0.2123703807592392,
+      "learning_rate": 7.328923342749039e-06,
+      "loss": 0.767,
+      "step": 950
+    },
+    {
+      "epoch": 0.3696217152757725,
+      "grad_norm": 0.22403818368911743,
+      "learning_rate": 7.273563238564319e-06,
+      "loss": 0.7542,
+      "step": 960
+    },
+    {
+      "epoch": 0.37347194147656176,
+      "grad_norm": 0.20945429801940918,
+      "learning_rate": 7.217849507865724e-06,
+      "loss": 0.7319,
+      "step": 970
+    },
+    {
+      "epoch": 0.37732216767735105,
+      "grad_norm": 0.20953916013240814,
+      "learning_rate": 7.161790816280858e-06,
+      "loss": 0.7401,
+      "step": 980
+    },
+    {
+      "epoch": 0.38117239387814034,
+      "grad_norm": 0.21496032178401947,
+      "learning_rate": 7.1053958830920125e-06,
+      "loss": 0.72,
+      "step": 990
+    },
+    {
+      "epoch": 0.3850226200789296,
+      "grad_norm": 0.21071797609329224,
+      "learning_rate": 7.0486734798799715e-06,
+      "loss": 0.74,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3888728462797189,
+      "grad_norm": 0.2062385529279709,
+      "learning_rate": 6.991632429159708e-06,
+      "loss": 0.7636,
+      "step": 1010
+    },
+    {
+      "epoch": 0.39272307248050825,
+      "grad_norm": 0.22598320245742798,
+      "learning_rate": 6.934281603008132e-06,
+      "loss": 0.7541,
+      "step": 1020
+    },
+    {
+      "epoch": 0.39657329868129754,
+      "grad_norm": 0.2025587260723114,
+      "learning_rate": 6.876629921684143e-06,
+      "loss": 0.7526,
+      "step": 1030
+    },
+    {
+      "epoch": 0.4004235248820868,
+      "grad_norm": 0.2127130776643753,
+      "learning_rate": 6.818686352241199e-06,
+      "loss": 0.7417,
+      "step": 1040
+    },
+    {
+      "epoch": 0.4042737510828761,
+      "grad_norm": 0.1935843825340271,
+      "learning_rate": 6.760459907132573e-06,
+      "loss": 0.7389,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4081239772836654,
+      "grad_norm": 0.192708358168602,
+      "learning_rate": 6.701959642809589e-06,
+      "loss": 0.744,
+      "step": 1060
+    },
+    {
+      "epoch": 0.41197420348445474,
+      "grad_norm": 0.20989014208316803,
+      "learning_rate": 6.643194658312987e-06,
+      "loss": 0.7342,
+      "step": 1070
+    },
+    {
+      "epoch": 0.415824429685244,
+      "grad_norm": 0.20252841711044312,
+      "learning_rate": 6.584174093857676e-06,
+      "loss": 0.7579,
+      "step": 1080
+    },
+    {
+      "epoch": 0.4196746558860333,
+      "grad_norm": 0.23374849557876587,
+      "learning_rate": 6.5249071294110735e-06,
+      "loss": 0.7434,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4235248820868226,
+      "grad_norm": 0.2124534696340561,
+      "learning_rate": 6.465402983265274e-06,
+      "loss": 0.7577,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4273751082876119,
+      "grad_norm": 0.2143521010875702,
+      "learning_rate": 6.405670910603243e-06,
+      "loss": 0.7514,
+      "step": 1110
+    },
+    {
+      "epoch": 0.43122533448840117,
+      "grad_norm": 0.22323313355445862,
+      "learning_rate": 6.345720202059291e-06,
+      "loss": 0.7482,
+      "step": 1120
+    },
+    {
+      "epoch": 0.4350755606891905,
+      "grad_norm": 0.21044665575027466,
+      "learning_rate": 6.285560182274008e-06,
+      "loss": 0.7258,
+      "step": 1130
+    },
+    {
+      "epoch": 0.4389257868899798,
+      "grad_norm": 0.2094753235578537,
+      "learning_rate": 6.225200208443936e-06,
+      "loss": 0.7483,
+      "step": 1140
+    },
+    {
+      "epoch": 0.4427760130907691,
+      "grad_norm": 0.20136062800884247,
+      "learning_rate": 6.16464966886616e-06,
+      "loss": 0.7367,
+      "step": 1150
+    },
+    {
+      "epoch": 0.44662623929155837,
+      "grad_norm": 0.24400964379310608,
+      "learning_rate": 6.103917981478065e-06,
+      "loss": 0.7627,
+      "step": 1160
+    },
+    {
+      "epoch": 0.45047646549234766,
+      "grad_norm": 0.2108762413263321,
+      "learning_rate": 6.043014592392489e-06,
+      "loss": 0.7733,
+      "step": 1170
+    },
+    {
+      "epoch": 0.454326691693137,
+      "grad_norm": 0.19752711057662964,
+      "learning_rate": 5.981948974428485e-06,
+      "loss": 0.7466,
+      "step": 1180
+    },
+    {
+      "epoch": 0.4581769178939263,
+      "grad_norm": 0.21444246172904968,
+      "learning_rate": 5.920730625637934e-06,
+      "loss": 0.7249,
+      "step": 1190
+    },
+    {
+      "epoch": 0.46202714409471557,
+      "grad_norm": 0.2196703404188156,
+      "learning_rate": 5.8593690678282355e-06,
+      "loss": 0.7392,
+      "step": 1200
+    },
+    {
+      "epoch": 0.46587737029550486,
+      "grad_norm": 0.2186385691165924,
+      "learning_rate": 5.797873845081301e-06,
+      "loss": 0.7424,
+      "step": 1210
+    },
+    {
+      "epoch": 0.46972759649629414,
+      "grad_norm": 0.21084672212600708,
+      "learning_rate": 5.736254522269075e-06,
+      "loss": 0.7574,
+      "step": 1220
+    },
+    {
+      "epoch": 0.47357782269708343,
+      "grad_norm": 0.20639672875404358,
+      "learning_rate": 5.6745206835658386e-06,
+      "loss": 0.7626,
+      "step": 1230
+    },
+    {
+      "epoch": 0.47742804889787277,
+      "grad_norm": 0.19815504550933838,
+      "learning_rate": 5.612681930957495e-06,
+      "loss": 0.7309,
+      "step": 1240
+    },
+    {
+      "epoch": 0.48127827509866206,
+      "grad_norm": 0.19687172770500183,
+      "learning_rate": 5.550747882748095e-06,
+      "loss": 0.7279,
+      "step": 1250
+    },
+    {
+      "epoch": 0.48512850129945134,
+      "grad_norm": 0.22359219193458557,
+      "learning_rate": 5.4887281720638195e-06,
+      "loss": 0.7382,
+      "step": 1260
+    },
+    {
+      "epoch": 0.48897872750024063,
+      "grad_norm": 0.2364218682050705,
+      "learning_rate": 5.42663244535466e-06,
+      "loss": 0.719,
+      "step": 1270
+    },
+    {
+      "epoch": 0.4928289537010299,
+      "grad_norm": 0.21197214722633362,
+      "learning_rate": 5.3644703608940205e-06,
+      "loss": 0.7303,
+      "step": 1280
+    },
+    {
+      "epoch": 0.49667917990181926,
+      "grad_norm": 0.21293418109416962,
+      "learning_rate": 5.302251587276491e-06,
+      "loss": 0.7393,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5005294061026085,
+      "grad_norm": 0.20284439623355865,
+      "learning_rate": 5.2399858019140005e-06,
+      "loss": 0.7415,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5043796323033978,
+      "grad_norm": 0.2055528312921524,
+      "learning_rate": 5.177682689530615e-06,
+      "loss": 0.7545,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5082298585041871,
+      "grad_norm": 0.21756330132484436,
+      "learning_rate": 5.11535194065619e-06,
+      "loss": 0.7241,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5120800847049765,
+      "grad_norm": 0.23702985048294067,
+      "learning_rate": 5.053003250119113e-06,
+      "loss": 0.7326,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5159303109057657,
+      "grad_norm": 0.21646200120449066,
+      "learning_rate": 4.9906463155383925e-06,
+      "loss": 0.7397,
+      "step": 1340
+    },
+    {
+      "epoch": 0.519780537106555,
+      "grad_norm": 0.20811885595321655,
+      "learning_rate": 4.9282908358153046e-06,
+      "loss": 0.7357,
+      "step": 1350
+    },
+    {
+      "epoch": 0.5236307633073443,
+      "grad_norm": 0.23617668449878693,
+      "learning_rate": 4.8659465096248345e-06,
+      "loss": 0.7503,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5274809895081336,
+      "grad_norm": 0.21591898798942566,
+      "learning_rate": 4.803623033907169e-06,
+      "loss": 0.7463,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5313312157089229,
+      "grad_norm": 0.19966216385364532,
+      "learning_rate": 4.741330102359441e-06,
+      "loss": 0.7547,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5351814419097122,
+      "grad_norm": 0.22445836663246155,
+      "learning_rate": 4.679077403927996e-06,
+      "loss": 0.7335,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5390316681105015,
+      "grad_norm": 0.2048346847295761,
+      "learning_rate": 4.616874621301373e-06,
+      "loss": 0.7351,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5428818943112907,
+      "grad_norm": 0.21623772382736206,
+      "learning_rate": 4.554731429404293e-06,
+      "loss": 0.7483,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5467321205120801,
+      "grad_norm": 0.22417451441287994,
+      "learning_rate": 4.492657493892822e-06,
+      "loss": 0.7317,
+      "step": 1420
+    },
+    {
+      "epoch": 0.5505823467128694,
+      "grad_norm": 0.22367490828037262,
+      "learning_rate": 4.430662469650999e-06,
+      "loss": 0.7484,
+      "step": 1430
+    },
+    {
+      "epoch": 0.5544325729136587,
+      "grad_norm": 0.22007058560848236,
+      "learning_rate": 4.368755999289118e-06,
+      "loss": 0.7539,
+      "step": 1440
+    },
+    {
+      "epoch": 0.558282799114448,
+      "grad_norm": 0.2181667536497116,
+      "learning_rate": 4.306947711643949e-06,
+      "loss": 0.7335,
+      "step": 1450
+    },
+    {
+      "epoch": 0.5621330253152372,
+      "grad_norm": 0.2234165370464325,
+      "learning_rate": 4.245247220281068e-06,
+      "loss": 0.7393,
+      "step": 1460
+    },
+    {
+      "epoch": 0.5659832515160266,
+      "grad_norm": 0.20719125866889954,
+      "learning_rate": 4.183664121999587e-06,
+      "loss": 0.7339,
+      "step": 1470
+    },
+    {
+      "epoch": 0.5698334777168159,
+      "grad_norm": 0.21672560274600983,
+      "learning_rate": 4.122207995339474e-06,
+      "loss": 0.7337,
+      "step": 1480
+    },
+    {
+      "epoch": 0.5736837039176051,
+      "grad_norm": 0.20906181633472443,
+      "learning_rate": 4.060888399091734e-06,
+      "loss": 0.7368,
+      "step": 1490
+    },
+    {
+      "epoch": 0.5775339301183945,
+      "grad_norm": 0.20378564298152924,
+      "learning_rate": 3.999714870811639e-06,
+      "loss": 0.7491,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5813841563191837,
+      "grad_norm": 0.21033139526844025,
+      "learning_rate": 3.938696925335275e-06,
+      "loss": 0.7421,
+      "step": 1510
+    },
+    {
+      "epoch": 0.5852343825199731,
+      "grad_norm": 0.2087218314409256,
+      "learning_rate": 3.87784405329962e-06,
+      "loss": 0.7488,
+      "step": 1520
+    },
+    {
+      "epoch": 0.5890846087207623,
+      "grad_norm": 0.20862263441085815,
+      "learning_rate": 3.817165719666382e-06,
+      "loss": 0.7494,
+      "step": 1530
+    },
+    {
+      "epoch": 0.5929348349215516,
+      "grad_norm": 0.2069379836320877,
+      "learning_rate": 3.756671362249841e-06,
+      "loss": 0.7562,
+      "step": 1540
+    },
+    {
+      "epoch": 0.596785061122341,
+      "grad_norm": 0.2072131633758545,
+      "learning_rate": 3.6963703902488907e-06,
+      "loss": 0.7423,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6006352873231302,
+      "grad_norm": 0.20903153717517853,
+      "learning_rate": 3.6362721827835655e-06,
+      "loss": 0.7543,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6044855135239195,
+      "grad_norm": 0.22152350842952728,
+      "learning_rate": 3.576386087436212e-06,
+      "loss": 0.7533,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6083357397247088,
+      "grad_norm": 0.2263679802417755,
+      "learning_rate": 3.5167214187975913e-06,
+      "loss": 0.7427,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6121859659254981,
+      "grad_norm": 0.18581515550613403,
+      "learning_rate": 3.457287457018086e-06,
+      "loss": 0.7448,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6160361921262875,
+      "grad_norm": 0.19600284099578857,
+      "learning_rate": 3.3980934463643003e-06,
+      "loss": 0.734,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6198864183270767,
+      "grad_norm": 0.19718419015407562,
+      "learning_rate": 3.3391485937812053e-06,
+      "loss": 0.7152,
+      "step": 1610
+    },
+    {
+      "epoch": 0.623736644527866,
+      "grad_norm": 0.21032793819904327,
+      "learning_rate": 3.280462067460115e-06,
+      "loss": 0.7143,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6275868707286553,
+      "grad_norm": 0.2074533849954605,
+      "learning_rate": 3.222042995412669e-06,
+      "loss": 0.7472,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6314370969294446,
+      "grad_norm": 0.18720458447933197,
+      "learning_rate": 3.163900464051089e-06,
+      "loss": 0.736,
+      "step": 1640
+    },
+    {
+      "epoch": 0.635287323130234,
+      "grad_norm": 0.21133342385292053,
+      "learning_rate": 3.1060435167748826e-06,
+      "loss": 0.7467,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6391375493310232,
+      "grad_norm": 0.20777830481529236,
+      "learning_rate": 3.048481152564253e-06,
+      "loss": 0.7267,
+      "step": 1660
+    },
+    {
+      "epoch": 0.6429877755318125,
+      "grad_norm": 0.22091570496559143,
+      "learning_rate": 2.9912223245804017e-06,
+      "loss": 0.7479,
+      "step": 1670
+    },
+    {
+      "epoch": 0.6468380017326018,
+      "grad_norm": 0.21918714046478271,
+      "learning_rate": 2.934275938772981e-06,
+      "loss": 0.7412,
+      "step": 1680
+    },
+    {
+      "epoch": 0.6506882279333911,
+      "grad_norm": 0.2169800102710724,
+      "learning_rate": 2.8776508524948676e-06,
+      "loss": 0.7322,
+      "step": 1690
+    },
+    {
+      "epoch": 0.6545384541341804,
+      "grad_norm": 0.18885155022144318,
+      "learning_rate": 2.821355873124503e-06,
+      "loss": 0.7308,
+      "step": 1700
+    },
+    {
+      "epoch": 0.6583886803349697,
+      "grad_norm": 0.21148116886615753,
+      "learning_rate": 2.7653997566960144e-06,
+      "loss": 0.7445,
+      "step": 1710
+    },
+    {
+      "epoch": 0.662238906535759,
+      "grad_norm": 0.2149786502122879,
+      "learning_rate": 2.709791206537309e-06,
+      "loss": 0.7345,
+      "step": 1720
+    },
+    {
+      "epoch": 0.6660891327365482,
+      "grad_norm": 0.21566441655158997,
+      "learning_rate": 2.6545388719163747e-06,
+      "loss": 0.7353,
+      "step": 1730
+    },
+    {
+      "epoch": 0.6699393589373376,
+      "grad_norm": 0.21236012876033783,
+      "learning_rate": 2.599651346695979e-06,
+      "loss": 0.7384,
+      "step": 1740
+    },
+    {
+      "epoch": 0.6737895851381268,
+      "grad_norm": 0.20321287214756012,
+      "learning_rate": 2.545137167997007e-06,
+      "loss": 0.7369,
+      "step": 1750
+    },
+    {
+      "epoch": 0.6776398113389162,
+      "grad_norm": 0.1989978849887848,
+      "learning_rate": 2.4910048148705972e-06,
+      "loss": 0.7248,
+      "step": 1760
+    },
+    {
+      "epoch": 0.6814900375397055,
+      "grad_norm": 0.2179379016160965,
+      "learning_rate": 2.4372627069793327e-06,
+      "loss": 0.7704,
+      "step": 1770
+    },
+    {
+      "epoch": 0.6853402637404947,
+      "grad_norm": 0.22134222090244293,
+      "learning_rate": 2.3839192032876486e-06,
+      "loss": 0.7532,
+      "step": 1780
+    },
+    {
+      "epoch": 0.6891904899412841,
+      "grad_norm": 0.20678532123565674,
+      "learning_rate": 2.3309826007617047e-06,
+      "loss": 0.7631,
+      "step": 1790
+    },
+    {
+      "epoch": 0.6930407161420733,
+      "grad_norm": 0.2110583633184433,
+      "learning_rate": 2.278461133078881e-06,
+      "loss": 0.728,
+      "step": 1800
+    },
+    {
+      "epoch": 0.6968909423428626,
+      "grad_norm": 0.2225019633769989,
+      "learning_rate": 2.2263629693471258e-06,
+      "loss": 0.7199,
+      "step": 1810
+    },
+    {
+      "epoch": 0.700741168543652,
+      "grad_norm": 0.19816574454307556,
+      "learning_rate": 2.1746962128343414e-06,
+      "loss": 0.7405,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7045913947444412,
+      "grad_norm": 0.202466681599617,
+      "learning_rate": 2.1234688997080183e-06,
+      "loss": 0.7318,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7084416209452306,
+      "grad_norm": 0.2380831390619278,
+      "learning_rate": 2.0726889977853053e-06,
+      "loss": 0.7379,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7122918471460198,
+      "grad_norm": 0.23160475492477417,
+      "learning_rate": 2.022364405293703e-06,
+      "loss": 0.7494,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7161420733468091,
+      "grad_norm": 0.208930104970932,
+      "learning_rate": 1.9725029496425814e-06,
+      "loss": 0.7442,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7199922995475985,
+      "grad_norm": 0.19125483930110931,
+      "learning_rate": 1.923112386205727e-06,
+      "loss": 0.7389,
+      "step": 1870
+    },
+    {
+      "epoch": 0.7238425257483877,
+      "grad_norm": 0.2023487687110901,
+      "learning_rate": 1.874200397115079e-06,
+      "loss": 0.7479,
+      "step": 1880
+    },
+    {
+      "epoch": 0.727692751949177,
+      "grad_norm": 0.22046050429344177,
+      "learning_rate": 1.825774590065857e-06,
+      "loss": 0.7299,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7315429781499663,
+      "grad_norm": 0.21409069001674652,
+      "learning_rate": 1.7778424971332813e-06,
+      "loss": 0.7412,
+      "step": 1900
+    },
+    {
+      "epoch": 0.7353932043507556,
+      "grad_norm": 0.2044023871421814,
+      "learning_rate": 1.7304115736010407e-06,
+      "loss": 0.7367,
+      "step": 1910
+    },
+    {
+      "epoch": 0.739243430551545,
+      "grad_norm": 0.20931850373744965,
+      "learning_rate": 1.6834891968017108e-06,
+      "loss": 0.7326,
+      "step": 1920
+    },
+    {
+      "epoch": 0.7430936567523342,
+      "grad_norm": 0.2151743322610855,
+      "learning_rate": 1.6370826649692884e-06,
+      "loss": 0.7456,
+      "step": 1930
+    },
+    {
+      "epoch": 0.7469438829531235,
+      "grad_norm": 0.2232065498828888,
+      "learning_rate": 1.5911991961040463e-06,
+      "loss": 0.7543,
+      "step": 1940
+    },
+    {
+      "epoch": 0.7507941091539128,
+      "grad_norm": 0.21889033913612366,
+      "learning_rate": 1.545845926849851e-06,
+      "loss": 0.7427,
+      "step": 1950
+    },
+    {
+      "epoch": 0.7546443353547021,
+      "grad_norm": 0.23165398836135864,
+      "learning_rate": 1.5010299113841397e-06,
+      "loss": 0.7591,
+      "step": 1960
+    },
+    {
+      "epoch": 0.7584945615554913,
+      "grad_norm": 0.208988219499588,
+      "learning_rate": 1.4567581203207238e-06,
+      "loss": 0.749,
+      "step": 1970
+    },
+    {
+      "epoch": 0.7623447877562807,
+      "grad_norm": 0.21974849700927734,
+      "learning_rate": 1.413037439625593e-06,
+      "loss": 0.7472,
+      "step": 1980
+    },
+    {
+      "epoch": 0.76619501395707,
+      "grad_norm": 0.2204838991165161,
+      "learning_rate": 1.3698746695458847e-06,
+      "loss": 0.7415,
+      "step": 1990
+    },
+    {
+      "epoch": 0.7700452401578592,
+      "grad_norm": 0.20803312957286835,
+      "learning_rate": 1.3272765235521745e-06,
+      "loss": 0.7338,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7738954663586486,
+      "grad_norm": 0.20757119357585907,
+      "learning_rate": 1.2852496272942877e-06,
+      "loss": 0.7441,
+      "step": 2010
+    },
+    {
+      "epoch": 0.7777456925594378,
+      "grad_norm": 0.2014625519514084,
+      "learning_rate": 1.2438005175707396e-06,
+      "loss": 0.7299,
+      "step": 2020
+    },
+    {
+      "epoch": 0.7815959187602272,
+      "grad_norm": 0.20603476464748383,
+      "learning_rate": 1.2029356413120297e-06,
+      "loss": 0.7147,
+      "step": 2030
+    },
+    {
+      "epoch": 0.7854461449610165,
+      "grad_norm": 0.20986656844615936,
+      "learning_rate": 1.1626613545778786e-06,
+      "loss": 0.7313,
+      "step": 2040
+    },
+    {
+      "epoch": 0.7892963711618057,
+      "grad_norm": 0.2034718543291092,
+      "learning_rate": 1.1229839215686294e-06,
+      "loss": 0.7351,
+      "step": 2050
+    },
+    {
+      "epoch": 0.7931465973625951,
+      "grad_norm": 0.1971336156129837,
+      "learning_rate": 1.0839095136509188e-06,
+      "loss": 0.7185,
+      "step": 2060
+    },
+    {
+      "epoch": 0.7969968235633843,
+      "grad_norm": 0.22867900133132935,
+      "learning_rate": 1.045444208397791e-06,
+      "loss": 0.7405,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8008470497641736,
+      "grad_norm": 0.20251429080963135,
+      "learning_rate": 1.007593988643402e-06,
+      "loss": 0.7246,
+      "step": 2080
+    },
+    {
+      "epoch": 0.804697275964963,
+      "grad_norm": 0.20122471451759338,
+      "learning_rate": 9.703647415524613e-07,
+      "loss": 0.7469,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8085475021657522,
+      "grad_norm": 0.20914405584335327,
+      "learning_rate": 9.337622577045485e-07,
+      "loss": 0.7345,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8123977283665416,
+      "grad_norm": 0.2090124487876892,
+      "learning_rate": 8.977922301934588e-07,
+      "loss": 0.7454,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8162479545673308,
+      "grad_norm": 0.2120453417301178,
+      "learning_rate": 8.624602537417009e-07,
+      "loss": 0.7458,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8200981807681201,
+      "grad_norm": 0.22438132762908936,
+      "learning_rate": 8.277718238303123e-07,
+      "loss": 0.7318,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8239484069689095,
+      "grad_norm": 0.2156093418598175,
+      "learning_rate": 7.937323358440935e-07,
+      "loss": 0.736,
+      "step": 2140
+    },
+    {
+      "epoch": 0.8277986331696987,
+      "grad_norm": 0.22631944715976715,
+      "learning_rate": 7.603470842324223e-07,
+      "loss": 0.7272,
+      "step": 2150
+    },
+    {
+      "epoch": 0.831648859370488,
+      "grad_norm": 0.2018340528011322,
+      "learning_rate": 7.276212616857575e-07,
+      "loss": 0.7463,
+      "step": 2160
+    },
+    {
+      "epoch": 0.8354990855712773,
+      "grad_norm": 0.21235322952270508,
+      "learning_rate": 6.955599583279837e-07,
+      "loss": 0.7446,
+      "step": 2170
+    },
+    {
+      "epoch": 0.8393493117720666,
+      "grad_norm": 0.20948825776576996,
+      "learning_rate": 6.641681609246981e-07,
+      "loss": 0.7377,
+      "step": 2180
+    },
+    {
+      "epoch": 0.8431995379728559,
+      "grad_norm": 0.19640007615089417,
+      "learning_rate": 6.334507521075711e-07,
+      "loss": 0.7427,
+      "step": 2190
+    },
+    {
+      "epoch": 0.8470497641736452,
+      "grad_norm": 0.21754327416419983,
+      "learning_rate": 6.034125096149179e-07,
+      "loss": 0.7392,
+      "step": 2200
+    },
+    {
+      "epoch": 0.8508999903744345,
+      "grad_norm": 0.21332880854606628,
+      "learning_rate": 5.740581055485722e-07,
+      "loss": 0.7094,
+      "step": 2210
+    },
+    {
+      "epoch": 0.8547502165752238,
+      "grad_norm": 0.19794656336307526,
+      "learning_rate": 5.453921056471961e-07,
+      "loss": 0.741,
+      "step": 2220
+    },
+    {
+      "epoch": 0.8586004427760131,
+      "grad_norm": 0.21449586749076843,
+      "learning_rate": 5.174189685761288e-07,
+      "loss": 0.7396,
+      "step": 2230
+    },
+    {
+      "epoch": 0.8624506689768023,
+      "grad_norm": 0.19703209400177002,
+      "learning_rate": 4.901430452338979e-07,
+      "loss": 0.7499,
+      "step": 2240
+    },
+    {
+      "epoch": 0.8663008951775917,
+      "grad_norm": 0.19983692467212677,
+      "learning_rate": 4.6356857807548286e-07,
+      "loss": 0.7325,
+      "step": 2250
+    },
+    {
+      "epoch": 0.870151121378381,
+      "grad_norm": 0.20859761536121368,
+      "learning_rate": 4.376997004524541e-07,
+      "loss": 0.7242,
+      "step": 2260
+    },
+    {
+      "epoch": 0.8740013475791703,
+      "grad_norm": 0.20711976289749146,
+      "learning_rate": 4.12540435970073e-07,
+      "loss": 0.7294,
+      "step": 2270
+    },
+    {
+      "epoch": 0.8778515737799596,
+      "grad_norm": 0.21211639046669006,
+      "learning_rate": 3.880946978614703e-07,
+      "loss": 0.7436,
+      "step": 2280
+    },
+    {
+      "epoch": 0.8817017999807488,
+      "grad_norm": 0.2080458551645279,
+      "learning_rate": 3.643662883789878e-07,
+      "loss": 0.7443,
+      "step": 2290
+    },
+    {
+      "epoch": 0.8855520261815382,
+      "grad_norm": 0.2000010907649994,
+      "learning_rate": 3.41358898202781e-07,
+      "loss": 0.725,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8894022523823275,
+      "grad_norm": 0.1971263438463211,
+      "learning_rate": 3.190761058667741e-07,
+      "loss": 0.7362,
+      "step": 2310
+    },
+    {
+      "epoch": 0.8932524785831167,
+      "grad_norm": 0.20479591190814972,
+      "learning_rate": 2.9752137720206463e-07,
+      "loss": 0.7394,
+      "step": 2320
+    },
+    {
+      "epoch": 0.8971027047839061,
+      "grad_norm": 0.20584715902805328,
+      "learning_rate": 2.7669806479785067e-07,
+      "loss": 0.746,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9009529309846953,
+      "grad_norm": 0.20519720017910004,
+      "learning_rate": 2.566094074799763e-07,
+      "loss": 0.7376,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9048031571854847,
+      "grad_norm": 0.20791608095169067,
+      "learning_rate": 2.3725852980716557e-07,
+      "loss": 0.7244,
+      "step": 2350
+    },
+    {
+      "epoch": 0.908653383386274,
+      "grad_norm": 0.2051246464252472,
+      "learning_rate": 2.1864844158503983e-07,
+      "loss": 0.7333,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9125036095870632,
+      "grad_norm": 0.22404487431049347,
+      "learning_rate": 2.007820373979702e-07,
+      "loss": 0.7277,
+      "step": 2370
+    },
+    {
+      "epoch": 0.9163538357878526,
+      "grad_norm": 0.20663705468177795,
+      "learning_rate": 1.8366209615886532e-07,
+      "loss": 0.737,
+      "step": 2380
+    },
+    {
+      "epoch": 0.9202040619886418,
+      "grad_norm": 0.21071551740169525,
+      "learning_rate": 1.6729128067693613e-07,
+      "loss": 0.7502,
+      "step": 2390
+    },
+    {
+      "epoch": 0.9240542881894311,
+      "grad_norm": 0.22272652387619019,
+      "learning_rate": 1.5167213724353426e-07,
+      "loss": 0.7487,
+      "step": 2400
+    },
+    {
+      "epoch": 0.9279045143902205,
+      "grad_norm": 0.20834046602249146,
+      "learning_rate": 1.3680709523610058e-07,
+      "loss": 0.7294,
+      "step": 2410
+    },
+    {
+      "epoch": 0.9317547405910097,
+      "grad_norm": 0.20752699673175812,
+      "learning_rate": 1.2269846674030418e-07,
+      "loss": 0.7531,
+      "step": 2420
+    },
+    {
+      "epoch": 0.935604966791799,
+      "grad_norm": 0.22016888856887817,
+      "learning_rate": 1.0934844619042595e-07,
+      "loss": 0.7481,
+      "step": 2430
+    },
+    {
+      "epoch": 0.9394551929925883,
+      "grad_norm": 0.21831564605236053,
+      "learning_rate": 9.675911002803961e-08,
+      "loss": 0.7365,
+      "step": 2440
+    },
+    {
+      "epoch": 0.9433054191933776,
+      "grad_norm": 0.22567403316497803,
+      "learning_rate": 8.493241637904381e-08,
+      "loss": 0.7334,
+      "step": 2450
+    },
+    {
+      "epoch": 0.9471556453941669,
+      "grad_norm": 0.20750615000724792,
+      "learning_rate": 7.387020474909689e-08,
+      "loss": 0.7528,
+      "step": 2460
+    },
+    {
+      "epoch": 0.9510058715949562,
+      "grad_norm": 0.22462479770183563,
+      "learning_rate": 6.357419573750856e-08,
+      "loss": 0.7482,
+      "step": 2470
+    },
+    {
+      "epoch": 0.9548560977957455,
+      "grad_norm": 0.21179573237895966,
+      "learning_rate": 5.4045990769613365e-08,
+      "loss": 0.7294,
+      "step": 2480
+    },
+    {
+      "epoch": 0.9587063239965348,
+      "grad_norm": 0.20741049945354462,
+      "learning_rate": 4.5287071847694434e-08,
+      "loss": 0.7369,
+      "step": 2490
+    },
+    {
+      "epoch": 0.9625565501973241,
+      "grad_norm": 0.2064109593629837,
+      "learning_rate": 3.7298801320467817e-08,
+      "loss": 0.7345,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9664067763981133,
+      "grad_norm": 0.21065875887870789,
+      "learning_rate": 3.0082421671192576e-08,
+      "loss": 0.7282,
+      "step": 2510
+    },
+    {
+      "epoch": 0.9702570025989027,
+      "grad_norm": 0.20871451497077942,
+      "learning_rate": 2.3639055324410355e-08,
+      "loss": 0.7481,
+      "step": 2520
+    },
+    {
+      "epoch": 0.974107228799692,
+      "grad_norm": 0.20505525171756744,
+      "learning_rate": 1.7969704471369477e-08,
+      "loss": 0.7322,
+      "step": 2530
+    },
+    {
+      "epoch": 0.9779574550004813,
+      "grad_norm": 0.19834499061107635,
+      "learning_rate": 1.3075250914142413e-08,
+      "loss": 0.7342,
+      "step": 2540
+    },
+    {
+      "epoch": 0.9818076812012706,
+      "grad_norm": 0.20289583504199982,
+      "learning_rate": 8.956455928476049e-09,
+      "loss": 0.7427,
+      "step": 2550
+    },
+    {
+      "epoch": 0.9856579074020598,
+      "grad_norm": 0.20495815575122833,
+      "learning_rate": 5.6139601453786275e-09,
+      "loss": 0.7355,
+      "step": 2560
+    },
+    {
+      "epoch": 0.9895081336028492,
+      "grad_norm": 0.20706807076931,
+      "learning_rate": 3.0482834514805602e-09,
+      "loss": 0.7258,
+      "step": 2570
+    },
+    {
+      "epoch": 0.9933583598036385,
+      "grad_norm": 0.20245809853076935,
+      "learning_rate": 1.2598249081713366e-09,
+      "loss": 0.7162,
+      "step": 2580
+    },
+    {
+      "epoch": 0.9972085860044277,
+      "grad_norm": 0.2140657752752304,
+      "learning_rate": 2.488626895302826e-10,
+      "loss": 0.7511,
+      "step": 2590
+    },
+    {
+      "epoch": 0.9999037443449803,
+      "step": 2597,
+      "total_flos": 4.064218269983179e+18,
+      "train_loss": 0.7583706730551384,
+      "train_runtime": 8563.2617,
+      "train_samples_per_second": 77.646,
+      "train_steps_per_second": 0.303
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2597,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.064218269983179e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen25/clip_openai__l14/finetune/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cd75e1ba9fe673b1a2ff24f573cb86d17e830814225cea0c0b79ed2a90a8da7
+size 6136

continuous/qwen25/clip_openai__l14/pretrain/config.json

@@ -0,0 +1,58 @@
+{
+  "_attn_implementation_autoset": true,
+  "add_faster_video": false,
+  "add_time_instruction": false,
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "faster_token_stride": 10,
+  "force_sample": false,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "image_aspect_ratio": "square",
+  "image_crop_resolution": null,
+  "image_grid_pinpoints": null,
+  "image_split_resolution": null,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "mm_hidden_size": 1024,
+  "mm_newline_position": "grid",
+  "mm_patch_merge_type": "flat",
+  "mm_projector_lr": null,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_resampler_type": null,
+  "mm_spatial_pool_mode": "bilinear",
+  "mm_spatial_pool_stride": null,
+  "mm_tunable_parts": "mm_mlp_adapter",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -2,
+  "mm_vision_tower": "/cache/ckpt/download/tokenizer/continuous/clip-vit-large-patch14",
+  "mm_vision_tower_lr": null,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "pos_skipping_range": 4096,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "tokenizer_model_max_length": 8192,
+  "tokenizer_padding_side": "right",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "use_mm_proj": true,
+  "use_pos_skipping": false,
+  "use_sliding_window": false,
+  "vision_tower_pretrained": null,
+  "vocab_size": 151936
+}

continuous/qwen25/clip_openai__l14/pretrain/mm_projector.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e471e25c0cfbcd839541c9cf2818d3937804d90589860ceccfbfaa1843bb75c
+size 15743032

continuous/qwen25/clip_openai__l14/pretrain/trainer_state.json

@@ -0,0 +1,421 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 545,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01834862385321101,
+      "grad_norm": 1.0560815334320068,
+      "learning_rate": 0.00032142857142857147,
+      "loss": 4.8777,
+      "step": 10
+    },
+    {
+      "epoch": 0.03669724770642202,
+      "grad_norm": 0.8729127049446106,
+      "learning_rate": 0.0006785714285714287,
+      "loss": 3.8661,
+      "step": 20
+    },
+    {
+      "epoch": 0.05504587155963303,
+      "grad_norm": 0.12406693398952484,
+      "learning_rate": 0.0009999907688176173,
+      "loss": 3.3469,
+      "step": 30
+    },
+    {
+      "epoch": 0.07339449541284404,
+      "grad_norm": 0.0805206149816513,
+      "learning_rate": 0.0009988834393115766,
+      "loss": 3.2417,
+      "step": 40
+    },
+    {
+      "epoch": 0.09174311926605505,
+      "grad_norm": 0.10036973655223846,
+      "learning_rate": 0.0009959345572940772,
+      "loss": 3.2156,
+      "step": 50
+    },
+    {
+      "epoch": 0.11009174311926606,
+      "grad_norm": 0.08534352481365204,
+      "learning_rate": 0.0009911550081155983,
+      "loss": 3.1881,
+      "step": 60
+    },
+    {
+      "epoch": 0.12844036697247707,
+      "grad_norm": 0.08022696524858475,
+      "learning_rate": 0.0009845624347566433,
+      "loss": 3.1619,
+      "step": 70
+    },
+    {
+      "epoch": 0.14678899082568808,
+      "grad_norm": 0.1522204577922821,
+      "learning_rate": 0.0009761811727013547,
+      "loss": 3.1409,
+      "step": 80
+    },
+    {
+      "epoch": 0.1651376146788991,
+      "grad_norm": 0.09141265600919724,
+      "learning_rate": 0.0009660421601067666,
+      "loss": 3.1009,
+      "step": 90
+    },
+    {
+      "epoch": 0.1834862385321101,
+      "grad_norm": 0.10249502956867218,
+      "learning_rate": 0.0009541828235992926,
+      "loss": 3.0585,
+      "step": 100
+    },
+    {
+      "epoch": 0.2018348623853211,
+      "grad_norm": 0.11859428137540817,
+      "learning_rate": 0.0009406469401200151,
+      "loss": 2.9974,
+      "step": 110
+    },
+    {
+      "epoch": 0.22018348623853212,
+      "grad_norm": 0.15691904723644257,
+      "learning_rate": 0.0009254844753287492,
+      "loss": 2.9514,
+      "step": 120
+    },
+    {
+      "epoch": 0.23853211009174313,
+      "grad_norm": 0.14870306849479675,
+      "learning_rate": 0.0009087513991633924,
+      "loss": 2.8855,
+      "step": 130
+    },
+    {
+      "epoch": 0.25688073394495414,
+      "grad_norm": 0.13546976447105408,
+      "learning_rate": 0.0008905094792353917,
+      "loss": 2.8492,
+      "step": 140
+    },
+    {
+      "epoch": 0.27522935779816515,
+      "grad_norm": 0.13613446056842804,
+      "learning_rate": 0.0008708260528239789,
+      "loss": 2.8195,
+      "step": 150
+    },
+    {
+      "epoch": 0.29357798165137616,
+      "grad_norm": 0.12859301269054413,
+      "learning_rate": 0.0008497737783108237,
+      "loss": 2.7751,
+      "step": 160
+    },
+    {
+      "epoch": 0.3119266055045872,
+      "grad_norm": 0.12455487251281738,
+      "learning_rate": 0.0008274303669726426,
+      "loss": 2.7456,
+      "step": 170
+    },
+    {
+      "epoch": 0.3302752293577982,
+      "grad_norm": 0.19460447132587433,
+      "learning_rate": 0.0008038782961218137,
+      "loss": 2.7147,
+      "step": 180
+    },
+    {
+      "epoch": 0.3486238532110092,
+      "grad_norm": 0.15065814554691315,
+      "learning_rate": 0.000779204504653894,
+      "loss": 2.6957,
+      "step": 190
+    },
+    {
+      "epoch": 0.3669724770642202,
+      "grad_norm": 0.11610583961009979,
+      "learning_rate": 0.0007535000721258791,
+      "loss": 2.6606,
+      "step": 200
+    },
+    {
+      "epoch": 0.3853211009174312,
+      "grad_norm": 0.1328841596841812,
+      "learning_rate": 0.00072685988254984,
+      "loss": 2.6465,
+      "step": 210
+    },
+    {
+      "epoch": 0.4036697247706422,
+      "grad_norm": 0.18471264839172363,
+      "learning_rate": 0.0006993822741429906,
+      "loss": 2.649,
+      "step": 220
+    },
+    {
+      "epoch": 0.42201834862385323,
+      "grad_norm": 0.12255788594484329,
+      "learning_rate": 0.0006711686763270818,
+      "loss": 2.6327,
+      "step": 230
+    },
+    {
+      "epoch": 0.44036697247706424,
+      "grad_norm": 0.14685308933258057,
+      "learning_rate": 0.0006423232353170831,
+      "loss": 2.619,
+      "step": 240
+    },
+    {
+      "epoch": 0.45871559633027525,
+      "grad_norm": 0.10123993456363678,
+      "learning_rate": 0.0006129524296812335,
+      "loss": 2.6138,
+      "step": 250
+    },
+    {
+      "epoch": 0.47706422018348627,
+      "grad_norm": 0.10980258136987686,
+      "learning_rate": 0.0005831646772915651,
+      "loss": 2.6184,
+      "step": 260
+    },
+    {
+      "epoch": 0.4954128440366973,
+      "grad_norm": 0.10687059164047241,
+      "learning_rate": 0.0005530699351157809,
+      "loss": 2.594,
+      "step": 270
+    },
+    {
+      "epoch": 0.5137614678899083,
+      "grad_norm": 0.1282455027103424,
+      "learning_rate": 0.0005227792933277943,
+      "loss": 2.597,
+      "step": 280
+    },
+    {
+      "epoch": 0.5321100917431193,
+      "grad_norm": 0.1193169504404068,
+      "learning_rate": 0.0004924045652352092,
+      "loss": 2.582,
+      "step": 290
+    },
+    {
+      "epoch": 0.5504587155963303,
+      "grad_norm": 0.10008552670478821,
+      "learning_rate": 0.0004620578745374633,
+      "loss": 2.58,
+      "step": 300
+    },
+    {
+      "epoch": 0.5688073394495413,
+      "grad_norm": 0.1447674036026001,
+      "learning_rate": 0.0004318512414382059,
+      "loss": 2.5881,
+      "step": 310
+    },
+    {
+      "epoch": 0.5871559633027523,
+      "grad_norm": 0.11730887740850449,
+      "learning_rate": 0.0004018961691397148,
+      "loss": 2.5833,
+      "step": 320
+    },
+    {
+      "epoch": 0.6055045871559633,
+      "grad_norm": 0.12038342654705048,
+      "learning_rate": 0.0003723032322457458,
+      "loss": 2.5664,
+      "step": 330
+    },
+    {
+      "epoch": 0.6238532110091743,
+      "grad_norm": 0.13851997256278992,
+      "learning_rate": 0.0003431816685921625,
+      "loss": 2.5698,
+      "step": 340
+    },
+    {
+      "epoch": 0.6422018348623854,
+      "grad_norm": 0.14255763590335846,
+      "learning_rate": 0.0003146389760120416,
+      "loss": 2.5693,
+      "step": 350
+    },
+    {
+      "epoch": 0.6605504587155964,
+      "grad_norm": 0.10141333192586899,
+      "learning_rate": 0.00028678051552373485,
+      "loss": 2.5722,
+      "step": 360
+    },
+    {
+      "epoch": 0.6788990825688074,
+      "grad_norm": 0.11664506793022156,
+      "learning_rate": 0.0002597091224066581,
+      "loss": 2.5638,
+      "step": 370
+    },
+    {
+      "epoch": 0.6972477064220184,
+      "grad_norm": 0.1107507199048996,
+      "learning_rate": 0.0002335247266004629,
+      "loss": 2.573,
+      "step": 380
+    },
+    {
+      "epoch": 0.7155963302752294,
+      "grad_norm": 0.11118254065513611,
+      "learning_rate": 0.000208323983828828,
+      "loss": 2.5598,
+      "step": 390
+    },
+    {
+      "epoch": 0.7339449541284404,
+      "grad_norm": 0.09765281528234482,
+      "learning_rate": 0.0001841999188095224,
+      "loss": 2.5525,
+      "step": 400
+    },
+    {
+      "epoch": 0.7522935779816514,
+      "grad_norm": 0.10624087601900101,
+      "learning_rate": 0.00016124158186777676,
+      "loss": 2.5584,
+      "step": 410
+    },
+    {
+      "epoch": 0.7706422018348624,
+      "grad_norm": 0.11790075153112411,
+      "learning_rate": 0.0001395337202205211,
+      "loss": 2.5564,
+      "step": 420
+    },
+    {
+      "epoch": 0.7889908256880734,
+      "grad_norm": 0.10939404368400574,
+      "learning_rate": 0.00011915646514489292,
+      "loss": 2.5505,
+      "step": 430
+    },
+    {
+      "epoch": 0.8073394495412844,
+      "grad_norm": 0.10540983080863953,
+      "learning_rate": 0.00010018503618578818,
+      "loss": 2.5603,
+      "step": 440
+    },
+    {
+      "epoch": 0.8256880733944955,
+      "grad_norm": 0.10599078983068466,
+      "learning_rate": 8.268946349432583e-05,
+      "loss": 2.5539,
+      "step": 450
+    },
+    {
+      "epoch": 0.8440366972477065,
+      "grad_norm": 0.09850326925516129,
+      "learning_rate": 6.673432932217117e-05,
+      "loss": 2.5458,
+      "step": 460
+    },
+    {
+      "epoch": 0.8623853211009175,
+      "grad_norm": 0.09631116688251495,
+      "learning_rate": 5.2378529625954684e-05,
+      "loss": 2.5382,
+      "step": 470
+    },
+    {
+      "epoch": 0.8807339449541285,
+      "grad_norm": 0.1180977150797844,
+      "learning_rate": 3.9675056661785556e-05,
+      "loss": 2.5411,
+      "step": 480
+    },
+    {
+      "epoch": 0.8990825688073395,
+      "grad_norm": 0.10064341872930527,
+      "learning_rate": 2.8670803372379538e-05,
+      "loss": 2.5504,
+      "step": 490
+    },
+    {
+      "epoch": 0.9174311926605505,
+      "grad_norm": 0.09566023200750351,
+      "learning_rate": 1.9406390288876586e-05,
+      "loss": 2.5551,
+      "step": 500
+    },
+    {
+      "epoch": 0.9357798165137615,
+      "grad_norm": 0.10022998601198196,
+      "learning_rate": 1.191601558631461e-05,
+      "loss": 2.5454,
+      "step": 510
+    },
+    {
+      "epoch": 0.9541284403669725,
+      "grad_norm": 0.09621183574199677,
+      "learning_rate": 6.2273288462559305e-06,
+      "loss": 2.5433,
+      "step": 520
+    },
+    {
+      "epoch": 0.9724770642201835,
+      "grad_norm": 0.0961390882730484,
+      "learning_rate": 2.361328992552314e-06,
+      "loss": 2.5454,
+      "step": 530
+    },
+    {
+      "epoch": 0.9908256880733946,
+      "grad_norm": 0.09193284809589386,
+      "learning_rate": 3.322867770044202e-07,
+      "loss": 2.5431,
+      "step": 540
+    },
+    {
+      "epoch": 1.0,
+      "step": 545,
+      "total_flos": 3.2159080125772595e+17,
+      "train_loss": 2.7725335339887427,
+      "train_runtime": 5268.5957,
+      "train_samples_per_second": 105.935,
+      "train_steps_per_second": 0.103
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 545,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.2159080125772595e+17,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen25/dino_vitb16/finetune/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
+}

continuous/qwen25/dino_vitb16/finetune/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

continuous/qwen25/dino_vitb16/finetune/config.json

@@ -0,0 +1,87 @@
+{
+  "add_faster_video": false,
+  "add_time_instruction": false,
+  "architectures": [
+    "LlavaQwenForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "faster_token_stride": 10,
+  "force_sample": false,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "image_aspect_ratio": "square",
+  "image_crop_resolution": null,
+  "image_grid_pinpoints": null,
+  "image_split_resolution": null,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "mm_hidden_size": 768,
+  "mm_newline_position": "grid",
+  "mm_patch_merge_type": "flat",
+  "mm_projector_lr": 0.0001,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_resampler_type": null,
+  "mm_spatial_pool_mode": "bilinear",
+  "mm_spatial_pool_stride": null,
+  "mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -1,
+  "mm_vision_tower": "hf:/cache/ckpt/download/tokenizer/continuous/dino/dino-vitb16",
+  "mm_vision_tower_lr": null,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "pad_token_id": 151643,
+  "pos_skipping_range": 4096,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "tokenizer_model_max_length": 8192,
+  "tokenizer_padding_side": "right",
+  "transformers_version": "4.57.6",
+  "use_cache": true,
+  "use_mm_proj": true,
+  "use_pos_skipping": false,
+  "use_sliding_window": false,
+  "vision_tower_pretrained": null,
+  "vocab_size": 151936
+}

continuous/qwen25/dino_vitb16/finetune/generation_config.json

@@ -0,0 +1,13 @@
+{
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.57.6"
+}

continuous/qwen25/dino_vitb16/finetune/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a041e968797f738791fa99ca0095a7e627d78045ab4a539a596da6de29443b8a
+size 3741191240

continuous/qwen25/dino_vitb16/finetune/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
+  }
+}

continuous/qwen25/dino_vitb16/finetune/tokenizer.json

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

continuous/qwen25/dino_vitb16/finetune/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,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 8192,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

continuous/qwen25/dino_vitb16/finetune/trainer_state.json

@@ -0,0 +1,1856 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9999037443449803,
+  "eval_steps": 500,
+  "global_step": 2597,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.003850226200789296,
+      "grad_norm": 15.886174201965332,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 1.861,
+      "step": 10
+    },
+    {
+      "epoch": 0.007700452401578592,
+      "grad_norm": 14.544974327087402,
+      "learning_rate": 2.435897435897436e-06,
+      "loss": 1.6202,
+      "step": 20
+    },
+    {
+      "epoch": 0.01155067860236789,
+      "grad_norm": 3.6004083156585693,
+      "learning_rate": 3.7179487179487184e-06,
+      "loss": 1.1466,
+      "step": 30
+    },
+    {
+      "epoch": 0.015400904803157185,
+      "grad_norm": 0.813385009765625,
+      "learning_rate": 5e-06,
+      "loss": 0.9778,
+      "step": 40
+    },
+    {
+      "epoch": 0.019251131003946483,
+      "grad_norm": 0.40260574221611023,
+      "learning_rate": 6.282051282051282e-06,
+      "loss": 0.9416,
+      "step": 50
+    },
+    {
+      "epoch": 0.02310135720473578,
+      "grad_norm": 0.2990524470806122,
+      "learning_rate": 7.564102564102564e-06,
+      "loss": 0.8946,
+      "step": 60
+    },
+    {
+      "epoch": 0.026951583405525073,
+      "grad_norm": 0.24172627925872803,
+      "learning_rate": 8.846153846153847e-06,
+      "loss": 0.8688,
+      "step": 70
+    },
+    {
+      "epoch": 0.03080180960631437,
+      "grad_norm": 0.26063546538352966,
+      "learning_rate": 9.999996111488723e-06,
+      "loss": 0.8778,
+      "step": 80
+    },
+    {
+      "epoch": 0.034652035807103666,
+      "grad_norm": 0.2366272509098053,
+      "learning_rate": 9.999529497453782e-06,
+      "loss": 0.858,
+      "step": 90
+    },
+    {
+      "epoch": 0.038502262007892966,
+      "grad_norm": 0.24803021550178528,
+      "learning_rate": 9.998285264324193e-06,
+      "loss": 0.8393,
+      "step": 100
+    },
+    {
+      "epoch": 0.04235248820868226,
+      "grad_norm": 0.21908074617385864,
+      "learning_rate": 9.996263605626052e-06,
+      "loss": 0.8279,
+      "step": 110
+    },
+    {
+      "epoch": 0.04620271440947156,
+      "grad_norm": 0.2403685599565506,
+      "learning_rate": 9.99346483580503e-06,
+      "loss": 0.8405,
+      "step": 120
+    },
+    {
+      "epoch": 0.05005294061026085,
+      "grad_norm": 0.21946322917938232,
+      "learning_rate": 9.989889390177467e-06,
+      "loss": 0.8325,
+      "step": 130
+    },
+    {
+      "epoch": 0.053903166811050146,
+      "grad_norm": 0.24591295421123505,
+      "learning_rate": 9.985537824862647e-06,
+      "loss": 0.8324,
+      "step": 140
+    },
+    {
+      "epoch": 0.057753393011839446,
+      "grad_norm": 0.25028693675994873,
+      "learning_rate": 9.980410816696317e-06,
+      "loss": 0.8149,
+      "step": 150
+    },
+    {
+      "epoch": 0.06160361921262874,
+      "grad_norm": 0.23148101568222046,
+      "learning_rate": 9.97450916312541e-06,
+      "loss": 0.822,
+      "step": 160
+    },
+    {
+      "epoch": 0.06545384541341803,
+      "grad_norm": 0.21118290722370148,
+      "learning_rate": 9.967833782084004e-06,
+      "loss": 0.8228,
+      "step": 170
+    },
+    {
+      "epoch": 0.06930407161420733,
+      "grad_norm": 0.2397470474243164,
+      "learning_rate": 9.960385711850554e-06,
+      "loss": 0.8499,
+      "step": 180
+    },
+    {
+      "epoch": 0.07315429781499663,
+      "grad_norm": 0.24370163679122925,
+      "learning_rate": 9.952166110886394e-06,
+      "loss": 0.8408,
+      "step": 190
+    },
+    {
+      "epoch": 0.07700452401578593,
+      "grad_norm": 0.26398253440856934,
+      "learning_rate": 9.943176257655567e-06,
+      "loss": 0.8353,
+      "step": 200
+    },
+    {
+      "epoch": 0.08085475021657522,
+      "grad_norm": 0.23155631124973297,
+      "learning_rate": 9.933417550425952e-06,
+      "loss": 0.8238,
+      "step": 210
+    },
+    {
+      "epoch": 0.08470497641736452,
+      "grad_norm": 0.23766176402568817,
+      "learning_rate": 9.9228915070518e-06,
+      "loss": 0.8142,
+      "step": 220
+    },
+    {
+      "epoch": 0.08855520261815382,
+      "grad_norm": 0.23766803741455078,
+      "learning_rate": 9.911599764737632e-06,
+      "loss": 0.8167,
+      "step": 230
+    },
+    {
+      "epoch": 0.09240542881894312,
+      "grad_norm": 0.23552413284778595,
+      "learning_rate": 9.899544079783606e-06,
+      "loss": 0.81,
+      "step": 240
+    },
+    {
+      "epoch": 0.0962556550197324,
+      "grad_norm": 0.23163732886314392,
+      "learning_rate": 9.886726327312338e-06,
+      "loss": 0.8256,
+      "step": 250
+    },
+    {
+      "epoch": 0.1001058812205217,
+      "grad_norm": 0.2270461916923523,
+      "learning_rate": 9.87314850097724e-06,
+      "loss": 0.8245,
+      "step": 260
+    },
+    {
+      "epoch": 0.103956107421311,
+      "grad_norm": 0.23508179187774658,
+      "learning_rate": 9.858812712652452e-06,
+      "loss": 0.8127,
+      "step": 270
+    },
+    {
+      "epoch": 0.10780633362210029,
+      "grad_norm": 0.21588268876075745,
+      "learning_rate": 9.843721192104341e-06,
+      "loss": 0.8145,
+      "step": 280
+    },
+    {
+      "epoch": 0.11165655982288959,
+      "grad_norm": 0.22882364690303802,
+      "learning_rate": 9.827876286644698e-06,
+      "loss": 0.8331,
+      "step": 290
+    },
+    {
+      "epoch": 0.11550678602367889,
+      "grad_norm": 0.2110714465379715,
+      "learning_rate": 9.811280460765643e-06,
+      "loss": 0.7931,
+      "step": 300
+    },
+    {
+      "epoch": 0.11935701222446819,
+      "grad_norm": 0.22629065811634064,
+      "learning_rate": 9.793936295756292e-06,
+      "loss": 0.8037,
+      "step": 310
+    },
+    {
+      "epoch": 0.12320723842525748,
+      "grad_norm": 0.238492950797081,
+      "learning_rate": 9.775846489301279e-06,
+      "loss": 0.8077,
+      "step": 320
+    },
+    {
+      "epoch": 0.12705746462604678,
+      "grad_norm": 0.23632951080799103,
+      "learning_rate": 9.757013855061157e-06,
+      "loss": 0.8062,
+      "step": 330
+    },
+    {
+      "epoch": 0.13090769082683607,
+      "grad_norm": 0.2525606155395508,
+      "learning_rate": 9.737441322234758e-06,
+      "loss": 0.8179,
+      "step": 340
+    },
+    {
+      "epoch": 0.13475791702762538,
+      "grad_norm": 0.23464250564575195,
+      "learning_rate": 9.717131935103605e-06,
+      "loss": 0.8173,
+      "step": 350
+    },
+    {
+      "epoch": 0.13860814322841467,
+      "grad_norm": 0.23275303840637207,
+      "learning_rate": 9.696088852558398e-06,
+      "loss": 0.8027,
+      "step": 360
+    },
+    {
+      "epoch": 0.14245836942920398,
+      "grad_norm": 0.22598494589328766,
+      "learning_rate": 9.674315347607686e-06,
+      "loss": 0.8288,
+      "step": 370
+    },
+    {
+      "epoch": 0.14630859562999327,
+      "grad_norm": 0.22737692296504974,
+      "learning_rate": 9.651814806868796e-06,
+      "loss": 0.7951,
+      "step": 380
+    },
+    {
+      "epoch": 0.15015882183078255,
+      "grad_norm": 0.2566891014575958,
+      "learning_rate": 9.62859073004107e-06,
+      "loss": 0.7989,
+      "step": 390
+    },
+    {
+      "epoch": 0.15400904803157187,
+      "grad_norm": 0.20842015743255615,
+      "learning_rate": 9.604646729361547e-06,
+      "loss": 0.806,
+      "step": 400
+    },
+    {
+      "epoch": 0.15785927423236115,
+      "grad_norm": 0.2263842076063156,
+      "learning_rate": 9.579986529043099e-06,
+      "loss": 0.7976,
+      "step": 410
+    },
+    {
+      "epoch": 0.16170950043315044,
+      "grad_norm": 0.2343895584344864,
+      "learning_rate": 9.554613964695189e-06,
+      "loss": 0.81,
+      "step": 420
+    },
+    {
+      "epoch": 0.16555972663393975,
+      "grad_norm": 0.22741858661174774,
+      "learning_rate": 9.528532982727281e-06,
+      "loss": 0.8006,
+      "step": 430
+    },
+    {
+      "epoch": 0.16940995283472904,
+      "grad_norm": 0.2158607393503189,
+      "learning_rate": 9.501747639735018e-06,
+      "loss": 0.8004,
+      "step": 440
+    },
+    {
+      "epoch": 0.17326017903551832,
+      "grad_norm": 0.22586774826049805,
+      "learning_rate": 9.474262101869272e-06,
+      "loss": 0.7881,
+      "step": 450
+    },
+    {
+      "epoch": 0.17711040523630764,
+      "grad_norm": 0.2275198996067047,
+      "learning_rate": 9.446080644188141e-06,
+      "loss": 0.7938,
+      "step": 460
+    },
+    {
+      "epoch": 0.18096063143709692,
+      "grad_norm": 0.2191532403230667,
+      "learning_rate": 9.417207649992017e-06,
+      "loss": 0.8021,
+      "step": 470
+    },
+    {
+      "epoch": 0.18481085763788624,
+      "grad_norm": 0.228336364030838,
+      "learning_rate": 9.387647610141816e-06,
+      "loss": 0.8018,
+      "step": 480
+    },
+    {
+      "epoch": 0.18866108383867553,
+      "grad_norm": 0.20619072020053864,
+      "learning_rate": 9.35740512236047e-06,
+      "loss": 0.7868,
+      "step": 490
+    },
+    {
+      "epoch": 0.1925113100394648,
+      "grad_norm": 0.2025853544473648,
+      "learning_rate": 9.326484890517809e-06,
+      "loss": 0.7951,
+      "step": 500
+    },
+    {
+      "epoch": 0.19636153624025413,
+      "grad_norm": 0.22733475267887115,
+      "learning_rate": 9.294891723898926e-06,
+      "loss": 0.7879,
+      "step": 510
+    },
+    {
+      "epoch": 0.2002117624410434,
+      "grad_norm": 0.24535058438777924,
+      "learning_rate": 9.262630536456153e-06,
+      "loss": 0.8168,
+      "step": 520
+    },
+    {
+      "epoch": 0.2040619886418327,
+      "grad_norm": 0.23272159695625305,
+      "learning_rate": 9.229706346044749e-06,
+      "loss": 0.7924,
+      "step": 530
+    },
+    {
+      "epoch": 0.207912214842622,
+      "grad_norm": 0.23702380061149597,
+      "learning_rate": 9.196124273642422e-06,
+      "loss": 0.8025,
+      "step": 540
+    },
+    {
+      "epoch": 0.2117624410434113,
+      "grad_norm": 0.21898318827152252,
+      "learning_rate": 9.161889542552842e-06,
+      "loss": 0.8053,
+      "step": 550
+    },
+    {
+      "epoch": 0.21561266724420058,
+      "grad_norm": 0.2141326665878296,
+      "learning_rate": 9.127007477593193e-06,
+      "loss": 0.8097,
+      "step": 560
+    },
+    {
+      "epoch": 0.2194628934449899,
+      "grad_norm": 0.2385537475347519,
+      "learning_rate": 9.091483504265974e-06,
+      "loss": 0.8208,
+      "step": 570
+    },
+    {
+      "epoch": 0.22331311964577918,
+      "grad_norm": 0.2331015020608902,
+      "learning_rate": 9.055323147915116e-06,
+      "loss": 0.7972,
+      "step": 580
+    },
+    {
+      "epoch": 0.2271633458465685,
+      "grad_norm": 0.22859793901443481,
+      "learning_rate": 9.018532032866585e-06,
+      "loss": 0.8097,
+      "step": 590
+    },
+    {
+      "epoch": 0.23101357204735778,
+      "grad_norm": 0.21381455659866333,
+      "learning_rate": 8.981115881553579e-06,
+      "loss": 0.8009,
+      "step": 600
+    },
+    {
+      "epoch": 0.23486379824814707,
+      "grad_norm": 0.2393539398908615,
+      "learning_rate": 8.943080513626476e-06,
+      "loss": 0.8212,
+      "step": 610
+    },
+    {
+      "epoch": 0.23871402444893638,
+      "grad_norm": 0.22123350203037262,
+      "learning_rate": 8.904431845047641e-06,
+      "loss": 0.8043,
+      "step": 620
+    },
+    {
+      "epoch": 0.24256425064972567,
+      "grad_norm": 0.24215556681156158,
+      "learning_rate": 8.86517588717128e-06,
+      "loss": 0.8216,
+      "step": 630
+    },
+    {
+      "epoch": 0.24641447685051496,
+      "grad_norm": 0.21687035262584686,
+      "learning_rate": 8.82531874580844e-06,
+      "loss": 0.8099,
+      "step": 640
+    },
+    {
+      "epoch": 0.25026470305130427,
+      "grad_norm": 0.20918869972229004,
+      "learning_rate": 8.784866620277314e-06,
+      "loss": 0.7682,
+      "step": 650
+    },
+    {
+      "epoch": 0.25411492925209356,
+      "grad_norm": 0.2463165670633316,
+      "learning_rate": 8.743825802439018e-06,
+      "loss": 0.8049,
+      "step": 660
+    },
+    {
+      "epoch": 0.25796515545288284,
+      "grad_norm": 0.2630014419555664,
+      "learning_rate": 8.702202675718956e-06,
+      "loss": 0.8091,
+      "step": 670
+    },
+    {
+      "epoch": 0.26181538165367213,
+      "grad_norm": 0.215959370136261,
+      "learning_rate": 8.660003714113959e-06,
+      "loss": 0.78,
+      "step": 680
+    },
+    {
+      "epoch": 0.26566560785446147,
+      "grad_norm": 0.2245296835899353,
+      "learning_rate": 8.617235481185323e-06,
+      "loss": 0.8038,
+      "step": 690
+    },
+    {
+      "epoch": 0.26951583405525076,
+      "grad_norm": 0.22034761309623718,
+      "learning_rate": 8.573904629037931e-06,
+      "loss": 0.8078,
+      "step": 700
+    },
+    {
+      "epoch": 0.27336606025604004,
+      "grad_norm": 0.22144842147827148,
+      "learning_rate": 8.530017897285585e-06,
+      "loss": 0.7886,
+      "step": 710
+    },
+    {
+      "epoch": 0.27721628645682933,
+      "grad_norm": 0.21133451163768768,
+      "learning_rate": 8.485582112002746e-06,
+      "loss": 0.7938,
+      "step": 720
+    },
+    {
+      "epoch": 0.2810665126576186,
+      "grad_norm": 0.2201513648033142,
+      "learning_rate": 8.440604184662811e-06,
+      "loss": 0.7982,
+      "step": 730
+    },
+    {
+      "epoch": 0.28491673885840796,
+      "grad_norm": 0.22036927938461304,
+      "learning_rate": 8.395091111063118e-06,
+      "loss": 0.7893,
+      "step": 740
+    },
+    {
+      "epoch": 0.28876696505919724,
+      "grad_norm": 0.21067984402179718,
+      "learning_rate": 8.349049970236822e-06,
+      "loss": 0.7862,
+      "step": 750
+    },
+    {
+      "epoch": 0.29261719125998653,
+      "grad_norm": 0.22430507838726044,
+      "learning_rate": 8.302487923351842e-06,
+      "loss": 0.78,
+      "step": 760
+    },
+    {
+      "epoch": 0.2964674174607758,
+      "grad_norm": 0.2219175398349762,
+      "learning_rate": 8.255412212597021e-06,
+      "loss": 0.7974,
+      "step": 770
+    },
+    {
+      "epoch": 0.3003176436615651,
+      "grad_norm": 0.2043299674987793,
+      "learning_rate": 8.207830160055688e-06,
+      "loss": 0.7934,
+      "step": 780
+    },
+    {
+      "epoch": 0.3041678698623544,
+      "grad_norm": 0.22788502275943756,
+      "learning_rate": 8.159749166566786e-06,
+      "loss": 0.7803,
+      "step": 790
+    },
+    {
+      "epoch": 0.30801809606314373,
+      "grad_norm": 0.21497862040996552,
+      "learning_rate": 8.11117671057377e-06,
+      "loss": 0.8076,
+      "step": 800
+    },
+    {
+      "epoch": 0.311868322263933,
+      "grad_norm": 0.22472186386585236,
+      "learning_rate": 8.062120346961404e-06,
+      "loss": 0.8008,
+      "step": 810
+    },
+    {
+      "epoch": 0.3157185484647223,
+      "grad_norm": 0.21011999249458313,
+      "learning_rate": 8.012587705880705e-06,
+      "loss": 0.7954,
+      "step": 820
+    },
+    {
+      "epoch": 0.3195687746655116,
+      "grad_norm": 0.2189938724040985,
+      "learning_rate": 7.962586491562146e-06,
+      "loss": 0.7806,
+      "step": 830
+    },
+    {
+      "epoch": 0.3234190008663009,
+      "grad_norm": 0.2391739785671234,
+      "learning_rate": 7.912124481117357e-06,
+      "loss": 0.7851,
+      "step": 840
+    },
+    {
+      "epoch": 0.3272692270670902,
+      "grad_norm": 0.21755380928516388,
+      "learning_rate": 7.861209523329494e-06,
+      "loss": 0.7945,
+      "step": 850
+    },
+    {
+      "epoch": 0.3311194532678795,
+      "grad_norm": 0.22036243975162506,
+      "learning_rate": 7.809849537432432e-06,
+      "loss": 0.7823,
+      "step": 860
+    },
+    {
+      "epoch": 0.3349696794686688,
+      "grad_norm": 0.21523839235305786,
+      "learning_rate": 7.758052511879035e-06,
+      "loss": 0.7904,
+      "step": 870
+    },
+    {
+      "epoch": 0.3388199056694581,
+      "grad_norm": 0.20617268979549408,
+      "learning_rate": 7.705826503098642e-06,
+      "loss": 0.7848,
+      "step": 880
+    },
+    {
+      "epoch": 0.34267013187024736,
+      "grad_norm": 0.20794928073883057,
+      "learning_rate": 7.653179634243979e-06,
+      "loss": 0.7843,
+      "step": 890
+    },
+    {
+      "epoch": 0.34652035807103665,
+      "grad_norm": 0.19777055084705353,
+      "learning_rate": 7.600120093927696e-06,
+      "loss": 0.7912,
+      "step": 900
+    },
+    {
+      "epoch": 0.350370584271826,
+      "grad_norm": 0.2251940816640854,
+      "learning_rate": 7.546656134948729e-06,
+      "loss": 0.7761,
+      "step": 910
+    },
+    {
+      "epoch": 0.3542208104726153,
+      "grad_norm": 0.2215731143951416,
+      "learning_rate": 7.492796073008661e-06,
+      "loss": 0.7769,
+      "step": 920
+    },
+    {
+      "epoch": 0.35807103667340456,
+      "grad_norm": 0.23822146654129028,
+      "learning_rate": 7.438548285418326e-06,
+      "loss": 0.766,
+      "step": 930
+    },
+    {
+      "epoch": 0.36192126287419385,
+      "grad_norm": 0.25262001156806946,
+      "learning_rate": 7.3839212097948e-06,
+      "loss": 0.7978,
+      "step": 940
+    },
+    {
+      "epoch": 0.36577148907498314,
+      "grad_norm": 0.2275586575269699,
+      "learning_rate": 7.328923342749039e-06,
+      "loss": 0.8121,
+      "step": 950
+    },
+    {
+      "epoch": 0.3696217152757725,
+      "grad_norm": 0.23380136489868164,
+      "learning_rate": 7.273563238564319e-06,
+      "loss": 0.7979,
+      "step": 960
+    },
+    {
+      "epoch": 0.37347194147656176,
+      "grad_norm": 0.22159229218959808,
+      "learning_rate": 7.217849507865724e-06,
+      "loss": 0.7791,
+      "step": 970
+    },
+    {
+      "epoch": 0.37732216767735105,
+      "grad_norm": 0.2144671529531479,
+      "learning_rate": 7.161790816280858e-06,
+      "loss": 0.7832,
+      "step": 980
+    },
+    {
+      "epoch": 0.38117239387814034,
+      "grad_norm": 0.2322656363248825,
+      "learning_rate": 7.1053958830920125e-06,
+      "loss": 0.7696,
+      "step": 990
+    },
+    {
+      "epoch": 0.3850226200789296,
+      "grad_norm": 0.2163584679365158,
+      "learning_rate": 7.0486734798799715e-06,
+      "loss": 0.7827,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3888728462797189,
+      "grad_norm": 0.21091416478157043,
+      "learning_rate": 6.991632429159708e-06,
+      "loss": 0.8072,
+      "step": 1010
+    },
+    {
+      "epoch": 0.39272307248050825,
+      "grad_norm": 0.23351222276687622,
+      "learning_rate": 6.934281603008132e-06,
+      "loss": 0.8017,
+      "step": 1020
+    },
+    {
+      "epoch": 0.39657329868129754,
+      "grad_norm": 0.22083984315395355,
+      "learning_rate": 6.876629921684143e-06,
+      "loss": 0.7987,
+      "step": 1030
+    },
+    {
+      "epoch": 0.4004235248820868,
+      "grad_norm": 0.22236652672290802,
+      "learning_rate": 6.818686352241199e-06,
+      "loss": 0.7897,
+      "step": 1040
+    },
+    {
+      "epoch": 0.4042737510828761,
+      "grad_norm": 0.20624788105487823,
+      "learning_rate": 6.760459907132573e-06,
+      "loss": 0.7862,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4081239772836654,
+      "grad_norm": 0.20265156030654907,
+      "learning_rate": 6.701959642809589e-06,
+      "loss": 0.7922,
+      "step": 1060
+    },
+    {
+      "epoch": 0.41197420348445474,
+      "grad_norm": 0.22034257650375366,
+      "learning_rate": 6.643194658312987e-06,
+      "loss": 0.7817,
+      "step": 1070
+    },
+    {
+      "epoch": 0.415824429685244,
+      "grad_norm": 0.21443665027618408,
+      "learning_rate": 6.584174093857676e-06,
+      "loss": 0.8068,
+      "step": 1080
+    },
+    {
+      "epoch": 0.4196746558860333,
+      "grad_norm": 0.2525565028190613,
+      "learning_rate": 6.5249071294110735e-06,
+      "loss": 0.7878,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4235248820868226,
+      "grad_norm": 0.22210122644901276,
+      "learning_rate": 6.465402983265274e-06,
+      "loss": 0.8029,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4273751082876119,
+      "grad_norm": 0.2285618782043457,
+      "learning_rate": 6.405670910603243e-06,
+      "loss": 0.8001,
+      "step": 1110
+    },
+    {
+      "epoch": 0.43122533448840117,
+      "grad_norm": 0.241199791431427,
+      "learning_rate": 6.345720202059291e-06,
+      "loss": 0.7953,
+      "step": 1120
+    },
+    {
+      "epoch": 0.4350755606891905,
+      "grad_norm": 0.21832212805747986,
+      "learning_rate": 6.285560182274008e-06,
+      "loss": 0.7723,
+      "step": 1130
+    },
+    {
+      "epoch": 0.4389257868899798,
+      "grad_norm": 0.2120783030986786,
+      "learning_rate": 6.225200208443936e-06,
+      "loss": 0.7935,
+      "step": 1140
+    },
+    {
+      "epoch": 0.4427760130907691,
+      "grad_norm": 0.225476935505867,
+      "learning_rate": 6.16464966886616e-06,
+      "loss": 0.7869,
+      "step": 1150
+    },
+    {
+      "epoch": 0.44662623929155837,
+      "grad_norm": 0.2547451853752136,
+      "learning_rate": 6.103917981478065e-06,
+      "loss": 0.8089,
+      "step": 1160
+    },
+    {
+      "epoch": 0.45047646549234766,
+      "grad_norm": 0.2202947586774826,
+      "learning_rate": 6.043014592392489e-06,
+      "loss": 0.8218,
+      "step": 1170
+    },
+    {
+      "epoch": 0.454326691693137,
+      "grad_norm": 0.2097046822309494,
+      "learning_rate": 5.981948974428485e-06,
+      "loss": 0.7923,
+      "step": 1180
+    },
+    {
+      "epoch": 0.4581769178939263,
+      "grad_norm": 0.2259206771850586,
+      "learning_rate": 5.920730625637934e-06,
+      "loss": 0.7749,
+      "step": 1190
+    },
+    {
+      "epoch": 0.46202714409471557,
+      "grad_norm": 0.23157721757888794,
+      "learning_rate": 5.8593690678282355e-06,
+      "loss": 0.7848,
+      "step": 1200
+    },
+    {
+      "epoch": 0.46587737029550486,
+      "grad_norm": 0.23050157725811005,
+      "learning_rate": 5.797873845081301e-06,
+      "loss": 0.7874,
+      "step": 1210
+    },
+    {
+      "epoch": 0.46972759649629414,
+      "grad_norm": 0.2354915738105774,
+      "learning_rate": 5.736254522269075e-06,
+      "loss": 0.8043,
+      "step": 1220
+    },
+    {
+      "epoch": 0.47357782269708343,
+      "grad_norm": 0.21703866124153137,
+      "learning_rate": 5.6745206835658386e-06,
+      "loss": 0.8073,
+      "step": 1230
+    },
+    {
+      "epoch": 0.47742804889787277,
+      "grad_norm": 0.21643134951591492,
+      "learning_rate": 5.612681930957495e-06,
+      "loss": 0.7757,
+      "step": 1240
+    },
+    {
+      "epoch": 0.48127827509866206,
+      "grad_norm": 0.2111319899559021,
+      "learning_rate": 5.550747882748095e-06,
+      "loss": 0.7757,
+      "step": 1250
+    },
+    {
+      "epoch": 0.48512850129945134,
+      "grad_norm": 0.23320414125919342,
+      "learning_rate": 5.4887281720638195e-06,
+      "loss": 0.7833,
+      "step": 1260
+    },
+    {
+      "epoch": 0.48897872750024063,
+      "grad_norm": 0.2499268651008606,
+      "learning_rate": 5.42663244535466e-06,
+      "loss": 0.7689,
+      "step": 1270
+    },
+    {
+      "epoch": 0.4928289537010299,
+      "grad_norm": 0.2196178138256073,
+      "learning_rate": 5.3644703608940205e-06,
+      "loss": 0.7757,
+      "step": 1280
+    },
+    {
+      "epoch": 0.49667917990181926,
+      "grad_norm": 0.22487597167491913,
+      "learning_rate": 5.302251587276491e-06,
+      "loss": 0.7875,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5005294061026085,
+      "grad_norm": 0.21756483614444733,
+      "learning_rate": 5.2399858019140005e-06,
+      "loss": 0.7886,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5043796323033978,
+      "grad_norm": 0.2100469470024109,
+      "learning_rate": 5.177682689530615e-06,
+      "loss": 0.7987,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5082298585041871,
+      "grad_norm": 0.22758349776268005,
+      "learning_rate": 5.11535194065619e-06,
+      "loss": 0.7718,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5120800847049765,
+      "grad_norm": 0.24539288878440857,
+      "learning_rate": 5.053003250119113e-06,
+      "loss": 0.7817,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5159303109057657,
+      "grad_norm": 0.22653166949748993,
+      "learning_rate": 4.9906463155383925e-06,
+      "loss": 0.7859,
+      "step": 1340
+    },
+    {
+      "epoch": 0.519780537106555,
+      "grad_norm": 0.21743519604206085,
+      "learning_rate": 4.9282908358153046e-06,
+      "loss": 0.7777,
+      "step": 1350
+    },
+    {
+      "epoch": 0.5236307633073443,
+      "grad_norm": 0.24924598634243011,
+      "learning_rate": 4.8659465096248345e-06,
+      "loss": 0.7977,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5274809895081336,
+      "grad_norm": 0.22528785467147827,
+      "learning_rate": 4.803623033907169e-06,
+      "loss": 0.7966,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5313312157089229,
+      "grad_norm": 0.2076415866613388,
+      "learning_rate": 4.741330102359441e-06,
+      "loss": 0.8013,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5351814419097122,
+      "grad_norm": 0.23262561857700348,
+      "learning_rate": 4.679077403927996e-06,
+      "loss": 0.782,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5390316681105015,
+      "grad_norm": 0.2168881893157959,
+      "learning_rate": 4.616874621301373e-06,
+      "loss": 0.7813,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5428818943112907,
+      "grad_norm": 0.23193693161010742,
+      "learning_rate": 4.554731429404293e-06,
+      "loss": 0.795,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5467321205120801,
+      "grad_norm": 0.23597458004951477,
+      "learning_rate": 4.492657493892822e-06,
+      "loss": 0.7823,
+      "step": 1420
+    },
+    {
+      "epoch": 0.5505823467128694,
+      "grad_norm": 0.23424269258975983,
+      "learning_rate": 4.430662469650999e-06,
+      "loss": 0.7968,
+      "step": 1430
+    },
+    {
+      "epoch": 0.5544325729136587,
+      "grad_norm": 0.22645926475524902,
+      "learning_rate": 4.368755999289118e-06,
+      "loss": 0.7986,
+      "step": 1440
+    },
+    {
+      "epoch": 0.558282799114448,
+      "grad_norm": 0.23223719000816345,
+      "learning_rate": 4.306947711643949e-06,
+      "loss": 0.7815,
+      "step": 1450
+    },
+    {
+      "epoch": 0.5621330253152372,
+      "grad_norm": 0.229193776845932,
+      "learning_rate": 4.245247220281068e-06,
+      "loss": 0.7863,
+      "step": 1460
+    },
+    {
+      "epoch": 0.5659832515160266,
+      "grad_norm": 0.21918119490146637,
+      "learning_rate": 4.183664121999587e-06,
+      "loss": 0.7835,
+      "step": 1470
+    },
+    {
+      "epoch": 0.5698334777168159,
+      "grad_norm": 0.23694296181201935,
+      "learning_rate": 4.122207995339474e-06,
+      "loss": 0.7785,
+      "step": 1480
+    },
+    {
+      "epoch": 0.5736837039176051,
+      "grad_norm": 0.2214203029870987,
+      "learning_rate": 4.060888399091734e-06,
+      "loss": 0.7854,
+      "step": 1490
+    },
+    {
+      "epoch": 0.5775339301183945,
+      "grad_norm": 0.21025076508522034,
+      "learning_rate": 3.999714870811639e-06,
+      "loss": 0.7935,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5813841563191837,
+      "grad_norm": 0.22226454317569733,
+      "learning_rate": 3.938696925335275e-06,
+      "loss": 0.788,
+      "step": 1510
+    },
+    {
+      "epoch": 0.5852343825199731,
+      "grad_norm": 0.21747390925884247,
+      "learning_rate": 3.87784405329962e-06,
+      "loss": 0.7965,
+      "step": 1520
+    },
+    {
+      "epoch": 0.5890846087207623,
+      "grad_norm": 0.2199631631374359,
+      "learning_rate": 3.817165719666382e-06,
+      "loss": 0.7989,
+      "step": 1530
+    },
+    {
+      "epoch": 0.5929348349215516,
+      "grad_norm": 0.21142235398292542,
+      "learning_rate": 3.756671362249841e-06,
+      "loss": 0.8021,
+      "step": 1540
+    },
+    {
+      "epoch": 0.596785061122341,
+      "grad_norm": 0.22255395352840424,
+      "learning_rate": 3.6963703902488907e-06,
+      "loss": 0.7889,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6006352873231302,
+      "grad_norm": 0.21671995520591736,
+      "learning_rate": 3.6362721827835655e-06,
+      "loss": 0.8067,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6044855135239195,
+      "grad_norm": 0.23468241095542908,
+      "learning_rate": 3.576386087436212e-06,
+      "loss": 0.8007,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6083357397247088,
+      "grad_norm": 0.24499160051345825,
+      "learning_rate": 3.5167214187975913e-06,
+      "loss": 0.7904,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6121859659254981,
+      "grad_norm": 0.19319795072078705,
+      "learning_rate": 3.457287457018086e-06,
+      "loss": 0.7914,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6160361921262875,
+      "grad_norm": 0.21402058005332947,
+      "learning_rate": 3.3980934463643003e-06,
+      "loss": 0.7807,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6198864183270767,
+      "grad_norm": 0.20608900487422943,
+      "learning_rate": 3.3391485937812053e-06,
+      "loss": 0.7618,
+      "step": 1610
+    },
+    {
+      "epoch": 0.623736644527866,
+      "grad_norm": 0.22335533797740936,
+      "learning_rate": 3.280462067460115e-06,
+      "loss": 0.7619,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6275868707286553,
+      "grad_norm": 0.21809925138950348,
+      "learning_rate": 3.222042995412669e-06,
+      "loss": 0.7934,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6314370969294446,
+      "grad_norm": 0.1946660578250885,
+      "learning_rate": 3.163900464051089e-06,
+      "loss": 0.7825,
+      "step": 1640
+    },
+    {
+      "epoch": 0.635287323130234,
+      "grad_norm": 0.22199198603630066,
+      "learning_rate": 3.1060435167748826e-06,
+      "loss": 0.7917,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6391375493310232,
+      "grad_norm": 0.2225673794746399,
+      "learning_rate": 3.048481152564253e-06,
+      "loss": 0.7721,
+      "step": 1660
+    },
+    {
+      "epoch": 0.6429877755318125,
+      "grad_norm": 0.23183125257492065,
+      "learning_rate": 2.9912223245804017e-06,
+      "loss": 0.7908,
+      "step": 1670
+    },
+    {
+      "epoch": 0.6468380017326018,
+      "grad_norm": 0.23210756480693817,
+      "learning_rate": 2.934275938772981e-06,
+      "loss": 0.7897,
+      "step": 1680
+    },
+    {
+      "epoch": 0.6506882279333911,
+      "grad_norm": 0.22565940022468567,
+      "learning_rate": 2.8776508524948676e-06,
+      "loss": 0.7778,
+      "step": 1690
+    },
+    {
+      "epoch": 0.6545384541341804,
+      "grad_norm": 0.1991378515958786,
+      "learning_rate": 2.821355873124503e-06,
+      "loss": 0.7804,
+      "step": 1700
+    },
+    {
+      "epoch": 0.6583886803349697,
+      "grad_norm": 0.2253904640674591,
+      "learning_rate": 2.7653997566960144e-06,
+      "loss": 0.7902,
+      "step": 1710
+    },
+    {
+      "epoch": 0.662238906535759,
+      "grad_norm": 0.22690114378929138,
+      "learning_rate": 2.709791206537309e-06,
+      "loss": 0.7802,
+      "step": 1720
+    },
+    {
+      "epoch": 0.6660891327365482,
+      "grad_norm": 0.22200503945350647,
+      "learning_rate": 2.6545388719163747e-06,
+      "loss": 0.7872,
+      "step": 1730
+    },
+    {
+      "epoch": 0.6699393589373376,
+      "grad_norm": 0.2250942587852478,
+      "learning_rate": 2.599651346695979e-06,
+      "loss": 0.786,
+      "step": 1740
+    },
+    {
+      "epoch": 0.6737895851381268,
+      "grad_norm": 0.2151332050561905,
+      "learning_rate": 2.545137167997007e-06,
+      "loss": 0.7836,
+      "step": 1750
+    },
+    {
+      "epoch": 0.6776398113389162,
+      "grad_norm": 0.20609252154827118,
+      "learning_rate": 2.4910048148705972e-06,
+      "loss": 0.7737,
+      "step": 1760
+    },
+    {
+      "epoch": 0.6814900375397055,
+      "grad_norm": 0.22625099122524261,
+      "learning_rate": 2.4372627069793327e-06,
+      "loss": 0.8146,
+      "step": 1770
+    },
+    {
+      "epoch": 0.6853402637404947,
+      "grad_norm": 0.22999650239944458,
+      "learning_rate": 2.3839192032876486e-06,
+      "loss": 0.801,
+      "step": 1780
+    },
+    {
+      "epoch": 0.6891904899412841,
+      "grad_norm": 0.21538399159908295,
+      "learning_rate": 2.3309826007617047e-06,
+      "loss": 0.8112,
+      "step": 1790
+    },
+    {
+      "epoch": 0.6930407161420733,
+      "grad_norm": 0.2159758061170578,
+      "learning_rate": 2.278461133078881e-06,
+      "loss": 0.7741,
+      "step": 1800
+    },
+    {
+      "epoch": 0.6968909423428626,
+      "grad_norm": 0.23252451419830322,
+      "learning_rate": 2.2263629693471258e-06,
+      "loss": 0.7671,
+      "step": 1810
+    },
+    {
+      "epoch": 0.700741168543652,
+      "grad_norm": 0.2094578593969345,
+      "learning_rate": 2.1746962128343414e-06,
+      "loss": 0.7894,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7045913947444412,
+      "grad_norm": 0.21040736138820648,
+      "learning_rate": 2.1234688997080183e-06,
+      "loss": 0.7784,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7084416209452306,
+      "grad_norm": 0.2255791574716568,
+      "learning_rate": 2.0726889977853053e-06,
+      "loss": 0.7853,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7122918471460198,
+      "grad_norm": 0.2307324856519699,
+      "learning_rate": 2.022364405293703e-06,
+      "loss": 0.794,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7161420733468091,
+      "grad_norm": 0.21707436442375183,
+      "learning_rate": 1.9725029496425814e-06,
+      "loss": 0.7899,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7199922995475985,
+      "grad_norm": 0.20414993166923523,
+      "learning_rate": 1.923112386205727e-06,
+      "loss": 0.7856,
+      "step": 1870
+    },
+    {
+      "epoch": 0.7238425257483877,
+      "grad_norm": 0.20887282490730286,
+      "learning_rate": 1.874200397115079e-06,
+      "loss": 0.7934,
+      "step": 1880
+    },
+    {
+      "epoch": 0.727692751949177,
+      "grad_norm": 0.23055794835090637,
+      "learning_rate": 1.825774590065857e-06,
+      "loss": 0.7768,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7315429781499663,
+      "grad_norm": 0.22657276690006256,
+      "learning_rate": 1.7778424971332813e-06,
+      "loss": 0.7867,
+      "step": 1900
+    },
+    {
+      "epoch": 0.7353932043507556,
+      "grad_norm": 0.2123354971408844,
+      "learning_rate": 1.7304115736010407e-06,
+      "loss": 0.7851,
+      "step": 1910
+    },
+    {
+      "epoch": 0.739243430551545,
+      "grad_norm": 0.22400633990764618,
+      "learning_rate": 1.6834891968017108e-06,
+      "loss": 0.7777,
+      "step": 1920
+    },
+    {
+      "epoch": 0.7430936567523342,
+      "grad_norm": 0.22626149654388428,
+      "learning_rate": 1.6370826649692884e-06,
+      "loss": 0.7963,
+      "step": 1930
+    },
+    {
+      "epoch": 0.7469438829531235,
+      "grad_norm": 0.23092301189899445,
+      "learning_rate": 1.5911991961040463e-06,
+      "loss": 0.8006,
+      "step": 1940
+    },
+    {
+      "epoch": 0.7507941091539128,
+      "grad_norm": 0.22290658950805664,
+      "learning_rate": 1.545845926849851e-06,
+      "loss": 0.7859,
+      "step": 1950
+    },
+    {
+      "epoch": 0.7546443353547021,
+      "grad_norm": 0.23918414115905762,
+      "learning_rate": 1.5010299113841397e-06,
+      "loss": 0.8068,
+      "step": 1960
+    },
+    {
+      "epoch": 0.7584945615554913,
+      "grad_norm": 0.22336329519748688,
+      "learning_rate": 1.4567581203207238e-06,
+      "loss": 0.7938,
+      "step": 1970
+    },
+    {
+      "epoch": 0.7623447877562807,
+      "grad_norm": 0.22779512405395508,
+      "learning_rate": 1.413037439625593e-06,
+      "loss": 0.7959,
+      "step": 1980
+    },
+    {
+      "epoch": 0.76619501395707,
+      "grad_norm": 0.23040682077407837,
+      "learning_rate": 1.3698746695458847e-06,
+      "loss": 0.7852,
+      "step": 1990
+    },
+    {
+      "epoch": 0.7700452401578592,
+      "grad_norm": 0.21686115860939026,
+      "learning_rate": 1.3272765235521745e-06,
+      "loss": 0.7788,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7738954663586486,
+      "grad_norm": 0.2231774628162384,
+      "learning_rate": 1.2852496272942877e-06,
+      "loss": 0.7871,
+      "step": 2010
+    },
+    {
+      "epoch": 0.7777456925594378,
+      "grad_norm": 0.20669811964035034,
+      "learning_rate": 1.2438005175707396e-06,
+      "loss": 0.7744,
+      "step": 2020
+    },
+    {
+      "epoch": 0.7815959187602272,
+      "grad_norm": 0.21545907855033875,
+      "learning_rate": 1.2029356413120297e-06,
+      "loss": 0.7619,
+      "step": 2030
+    },
+    {
+      "epoch": 0.7854461449610165,
+      "grad_norm": 0.21647612750530243,
+      "learning_rate": 1.1626613545778786e-06,
+      "loss": 0.7777,
+      "step": 2040
+    },
+    {
+      "epoch": 0.7892963711618057,
+      "grad_norm": 0.2070571482181549,
+      "learning_rate": 1.1229839215686294e-06,
+      "loss": 0.7814,
+      "step": 2050
+    },
+    {
+      "epoch": 0.7931465973625951,
+      "grad_norm": 0.20127008855342865,
+      "learning_rate": 1.0839095136509188e-06,
+      "loss": 0.7677,
+      "step": 2060
+    },
+    {
+      "epoch": 0.7969968235633843,
+      "grad_norm": 0.23721298575401306,
+      "learning_rate": 1.045444208397791e-06,
+      "loss": 0.7906,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8008470497641736,
+      "grad_norm": 0.2113662213087082,
+      "learning_rate": 1.007593988643402e-06,
+      "loss": 0.772,
+      "step": 2080
+    },
+    {
+      "epoch": 0.804697275964963,
+      "grad_norm": 0.20856694877147675,
+      "learning_rate": 9.703647415524613e-07,
+      "loss": 0.7925,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8085475021657522,
+      "grad_norm": 0.21733498573303223,
+      "learning_rate": 9.337622577045485e-07,
+      "loss": 0.7823,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8123977283665416,
+      "grad_norm": 0.21811848878860474,
+      "learning_rate": 8.977922301934588e-07,
+      "loss": 0.7891,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8162479545673308,
+      "grad_norm": 0.2184433937072754,
+      "learning_rate": 8.624602537417009e-07,
+      "loss": 0.7906,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8200981807681201,
+      "grad_norm": 0.2347899079322815,
+      "learning_rate": 8.277718238303123e-07,
+      "loss": 0.7772,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8239484069689095,
+      "grad_norm": 0.22499045729637146,
+      "learning_rate": 7.937323358440935e-07,
+      "loss": 0.7813,
+      "step": 2140
+    },
+    {
+      "epoch": 0.8277986331696987,
+      "grad_norm": 0.23316697776317596,
+      "learning_rate": 7.603470842324223e-07,
+      "loss": 0.7738,
+      "step": 2150
+    },
+    {
+      "epoch": 0.831648859370488,
+      "grad_norm": 0.21508166193962097,
+      "learning_rate": 7.276212616857575e-07,
+      "loss": 0.7911,
+      "step": 2160
+    },
+    {
+      "epoch": 0.8354990855712773,
+      "grad_norm": 0.2173163890838623,
+      "learning_rate": 6.955599583279837e-07,
+      "loss": 0.7935,
+      "step": 2170
+    },
+    {
+      "epoch": 0.8393493117720666,
+      "grad_norm": 0.21352429687976837,
+      "learning_rate": 6.641681609246981e-07,
+      "loss": 0.7823,
+      "step": 2180
+    },
+    {
+      "epoch": 0.8431995379728559,
+      "grad_norm": 0.2120044231414795,
+      "learning_rate": 6.334507521075711e-07,
+      "loss": 0.7934,
+      "step": 2190
+    },
+    {
+      "epoch": 0.8470497641736452,
+      "grad_norm": 0.23786482214927673,
+      "learning_rate": 6.034125096149179e-07,
+      "loss": 0.7854,
+      "step": 2200
+    },
+    {
+      "epoch": 0.8508999903744345,
+      "grad_norm": 0.21740296483039856,
+      "learning_rate": 5.740581055485722e-07,
+      "loss": 0.7552,
+      "step": 2210
+    },
+    {
+      "epoch": 0.8547502165752238,
+      "grad_norm": 0.20594458281993866,
+      "learning_rate": 5.453921056471961e-07,
+      "loss": 0.7892,
+      "step": 2220
+    },
+    {
+      "epoch": 0.8586004427760131,
+      "grad_norm": 0.2221790850162506,
+      "learning_rate": 5.174189685761288e-07,
+      "loss": 0.7844,
+      "step": 2230
+    },
+    {
+      "epoch": 0.8624506689768023,
+      "grad_norm": 0.20403246581554413,
+      "learning_rate": 4.901430452338979e-07,
+      "loss": 0.794,
+      "step": 2240
+    },
+    {
+      "epoch": 0.8663008951775917,
+      "grad_norm": 0.21037837862968445,
+      "learning_rate": 4.6356857807548286e-07,
+      "loss": 0.7811,
+      "step": 2250
+    },
+    {
+      "epoch": 0.870151121378381,
+      "grad_norm": 0.21810811758041382,
+      "learning_rate": 4.376997004524541e-07,
+      "loss": 0.7699,
+      "step": 2260
+    },
+    {
+      "epoch": 0.8740013475791703,
+      "grad_norm": 0.21736755967140198,
+      "learning_rate": 4.12540435970073e-07,
+      "loss": 0.779,
+      "step": 2270
+    },
+    {
+      "epoch": 0.8778515737799596,
+      "grad_norm": 0.21637940406799316,
+      "learning_rate": 3.880946978614703e-07,
+      "loss": 0.7865,
+      "step": 2280
+    },
+    {
+      "epoch": 0.8817017999807488,
+      "grad_norm": 0.21787485480308533,
+      "learning_rate": 3.643662883789878e-07,
+      "loss": 0.7936,
+      "step": 2290
+    },
+    {
+      "epoch": 0.8855520261815382,
+      "grad_norm": 0.20262400805950165,
+      "learning_rate": 3.41358898202781e-07,
+      "loss": 0.7715,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8894022523823275,
+      "grad_norm": 0.2038927674293518,
+      "learning_rate": 3.190761058667741e-07,
+      "loss": 0.7844,
+      "step": 2310
+    },
+    {
+      "epoch": 0.8932524785831167,
+      "grad_norm": 0.21181254088878632,
+      "learning_rate": 2.9752137720206463e-07,
+      "loss": 0.7883,
+      "step": 2320
+    },
+    {
+      "epoch": 0.8971027047839061,
+      "grad_norm": 0.21606294810771942,
+      "learning_rate": 2.7669806479785067e-07,
+      "loss": 0.7926,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9009529309846953,
+      "grad_norm": 0.21430377662181854,
+      "learning_rate": 2.566094074799763e-07,
+      "loss": 0.7879,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9048031571854847,
+      "grad_norm": 0.21501865983009338,
+      "learning_rate": 2.3725852980716557e-07,
+      "loss": 0.7701,
+      "step": 2350
+    },
+    {
+      "epoch": 0.908653383386274,
+      "grad_norm": 0.21181419491767883,
+      "learning_rate": 2.1864844158503983e-07,
+      "loss": 0.7819,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9125036095870632,
+      "grad_norm": 0.23571734130382538,
+      "learning_rate": 2.007820373979702e-07,
+      "loss": 0.7743,
+      "step": 2370
+    },
+    {
+      "epoch": 0.9163538357878526,
+      "grad_norm": 0.22296272218227386,
+      "learning_rate": 1.8366209615886532e-07,
+      "loss": 0.7844,
+      "step": 2380
+    },
+    {
+      "epoch": 0.9202040619886418,
+      "grad_norm": 0.21221713721752167,
+      "learning_rate": 1.6729128067693613e-07,
+      "loss": 0.7917,
+      "step": 2390
+    },
+    {
+      "epoch": 0.9240542881894311,
+      "grad_norm": 0.23268206417560577,
+      "learning_rate": 1.5167213724353426e-07,
+      "loss": 0.7929,
+      "step": 2400
+    },
+    {
+      "epoch": 0.9279045143902205,
+      "grad_norm": 0.21401500701904297,
+      "learning_rate": 1.3680709523610058e-07,
+      "loss": 0.7729,
+      "step": 2410
+    },
+    {
+      "epoch": 0.9317547405910097,
+      "grad_norm": 0.21829456090927124,
+      "learning_rate": 1.2269846674030418e-07,
+      "loss": 0.7998,
+      "step": 2420
+    },
+    {
+      "epoch": 0.935604966791799,
+      "grad_norm": 0.2288154810667038,
+      "learning_rate": 1.0934844619042595e-07,
+      "loss": 0.7927,
+      "step": 2430
+    },
+    {
+      "epoch": 0.9394551929925883,
+      "grad_norm": 0.22978942096233368,
+      "learning_rate": 9.675911002803961e-08,
+      "loss": 0.7836,
+      "step": 2440
+    },
+    {
+      "epoch": 0.9433054191933776,
+      "grad_norm": 0.2328425645828247,
+      "learning_rate": 8.493241637904381e-08,
+      "loss": 0.7818,
+      "step": 2450
+    },
+    {
+      "epoch": 0.9471556453941669,
+      "grad_norm": 0.2185315191745758,
+      "learning_rate": 7.387020474909689e-08,
+      "loss": 0.7997,
+      "step": 2460
+    },
+    {
+      "epoch": 0.9510058715949562,
+      "grad_norm": 0.22810150682926178,
+      "learning_rate": 6.357419573750856e-08,
+      "loss": 0.7957,
+      "step": 2470
+    },
+    {
+      "epoch": 0.9548560977957455,
+      "grad_norm": 0.21997477114200592,
+      "learning_rate": 5.4045990769613365e-08,
+      "loss": 0.777,
+      "step": 2480
+    },
+    {
+      "epoch": 0.9587063239965348,
+      "grad_norm": 0.21755611896514893,
+      "learning_rate": 4.5287071847694434e-08,
+      "loss": 0.7843,
+      "step": 2490
+    },
+    {
+      "epoch": 0.9625565501973241,
+      "grad_norm": 0.2165653109550476,
+      "learning_rate": 3.7298801320467817e-08,
+      "loss": 0.7792,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9664067763981133,
+      "grad_norm": 0.21925577521324158,
+      "learning_rate": 3.0082421671192576e-08,
+      "loss": 0.7716,
+      "step": 2510
+    },
+    {
+      "epoch": 0.9702570025989027,
+      "grad_norm": 0.2154638171195984,
+      "learning_rate": 2.3639055324410355e-08,
+      "loss": 0.7909,
+      "step": 2520
+    },
+    {
+      "epoch": 0.974107228799692,
+      "grad_norm": 0.2190164476633072,
+      "learning_rate": 1.7969704471369477e-08,
+      "loss": 0.7786,
+      "step": 2530
+    },
+    {
+      "epoch": 0.9779574550004813,
+      "grad_norm": 0.2086085081100464,
+      "learning_rate": 1.3075250914142413e-08,
+      "loss": 0.7789,
+      "step": 2540
+    },
+    {
+      "epoch": 0.9818076812012706,
+      "grad_norm": 0.21022354066371918,
+      "learning_rate": 8.956455928476049e-09,
+      "loss": 0.79,
+      "step": 2550
+    },
+    {
+      "epoch": 0.9856579074020598,
+      "grad_norm": 0.21410445868968964,
+      "learning_rate": 5.6139601453786275e-09,
+      "loss": 0.7835,
+      "step": 2560
+    },
+    {
+      "epoch": 0.9895081336028492,
+      "grad_norm": 0.21238112449645996,
+      "learning_rate": 3.0482834514805602e-09,
+      "loss": 0.7718,
+      "step": 2570
+    },
+    {
+      "epoch": 0.9933583598036385,
+      "grad_norm": 0.20928148925304413,
+      "learning_rate": 1.2598249081713366e-09,
+      "loss": 0.7644,
+      "step": 2580
+    },
+    {
+      "epoch": 0.9972085860044277,
+      "grad_norm": 0.22562502324581146,
+      "learning_rate": 2.488626895302826e-10,
+      "loss": 0.7949,
+      "step": 2590
+    },
+    {
+      "epoch": 0.9999037443449803,
+      "step": 2597,
+      "total_flos": 3.5190682007822664e+18,
+      "train_loss": 0.8038170997757704,
+      "train_runtime": 9865.4598,
+      "train_samples_per_second": 67.397,
+      "train_steps_per_second": 0.263
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2597,
+  "num_input_tokens_seen": 418855352,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.5190682007822664e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen25/dino_vitb16/finetune/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c578b37175cf95fbbd2baa07ef66dc64ce89d7eb53461d296fbb5392afb93ad5
+size 6264

continuous/qwen25/dino_vitb16/pretrain/config.json

@@ -0,0 +1,87 @@
+{
+  "add_faster_video": false,
+  "add_time_instruction": false,
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "faster_token_stride": 10,
+  "force_sample": false,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "image_aspect_ratio": "square",
+  "image_crop_resolution": null,
+  "image_grid_pinpoints": null,
+  "image_split_resolution": null,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "mm_hidden_size": 768,
+  "mm_newline_position": "grid",
+  "mm_patch_merge_type": "flat",
+  "mm_projector_lr": null,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_resampler_type": null,
+  "mm_spatial_pool_mode": "bilinear",
+  "mm_spatial_pool_stride": null,
+  "mm_tunable_parts": "mm_mlp_adapter",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -1,
+  "mm_vision_tower": "hf:/cache/ckpt/download/tokenizer/continuous/dino/dino-vitb16",
+  "mm_vision_tower_lr": null,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "pad_token_id": 151643,
+  "pos_skipping_range": 4096,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "tokenizer_model_max_length": 8192,
+  "tokenizer_padding_side": "right",
+  "transformers_version": "4.57.6",
+  "use_cache": true,
+  "use_mm_proj": true,
+  "use_pos_skipping": false,
+  "use_sliding_window": false,
+  "vision_tower_pretrained": null,
+  "vocab_size": 151936
+}

continuous/qwen25/dino_vitb16/pretrain/mm_projector.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d877751e22d61f47afb0fb8a8dbd906e039907f611bcb6328e4d335bafea8c0
+size 14170168

continuous/qwen25/dino_vitb16/pretrain/trainer_state.json

@@ -0,0 +1,421 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 545,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01834862385321101,
+      "grad_norm": 2.3113088607788086,
+      "learning_rate": 0.00032142857142857147,
+      "loss": 4.7231,
+      "step": 10
+    },
+    {
+      "epoch": 0.03669724770642202,
+      "grad_norm": 0.6537547707557678,
+      "learning_rate": 0.0006785714285714287,
+      "loss": 3.735,
+      "step": 20
+    },
+    {
+      "epoch": 0.05504587155963303,
+      "grad_norm": 0.893591046333313,
+      "learning_rate": 0.0009999907688176173,
+      "loss": 3.4082,
+      "step": 30
+    },
+    {
+      "epoch": 0.07339449541284404,
+      "grad_norm": 0.2732776403427124,
+      "learning_rate": 0.0009988834393115766,
+      "loss": 3.3059,
+      "step": 40
+    },
+    {
+      "epoch": 0.09174311926605505,
+      "grad_norm": 0.39956939220428467,
+      "learning_rate": 0.0009959345572940772,
+      "loss": 3.2609,
+      "step": 50
+    },
+    {
+      "epoch": 0.11009174311926606,
+      "grad_norm": 0.368915319442749,
+      "learning_rate": 0.0009911550081155983,
+      "loss": 3.2311,
+      "step": 60
+    },
+    {
+      "epoch": 0.12844036697247707,
+      "grad_norm": 0.305540531873703,
+      "learning_rate": 0.0009845624347566433,
+      "loss": 3.2038,
+      "step": 70
+    },
+    {
+      "epoch": 0.14678899082568808,
+      "grad_norm": 0.24530617892742157,
+      "learning_rate": 0.0009761811727013547,
+      "loss": 3.1844,
+      "step": 80
+    },
+    {
+      "epoch": 0.1651376146788991,
+      "grad_norm": 0.32060831785202026,
+      "learning_rate": 0.0009660421601067666,
+      "loss": 3.1615,
+      "step": 90
+    },
+    {
+      "epoch": 0.1834862385321101,
+      "grad_norm": 0.40295523405075073,
+      "learning_rate": 0.0009541828235992926,
+      "loss": 3.1554,
+      "step": 100
+    },
+    {
+      "epoch": 0.2018348623853211,
+      "grad_norm": 0.3805027902126312,
+      "learning_rate": 0.0009406469401200151,
+      "loss": 3.1437,
+      "step": 110
+    },
+    {
+      "epoch": 0.22018348623853212,
+      "grad_norm": 0.20889247953891754,
+      "learning_rate": 0.0009254844753287492,
+      "loss": 3.1368,
+      "step": 120
+    },
+    {
+      "epoch": 0.23853211009174313,
+      "grad_norm": 0.3483445644378662,
+      "learning_rate": 0.0009087513991633924,
+      "loss": 3.1116,
+      "step": 130
+    },
+    {
+      "epoch": 0.25688073394495414,
+      "grad_norm": 0.3215317130088806,
+      "learning_rate": 0.0008905094792353917,
+      "loss": 3.1051,
+      "step": 140
+    },
+    {
+      "epoch": 0.27522935779816515,
+      "grad_norm": 0.4045512080192566,
+      "learning_rate": 0.0008708260528239789,
+      "loss": 3.1048,
+      "step": 150
+    },
+    {
+      "epoch": 0.29357798165137616,
+      "grad_norm": 0.23699481785297394,
+      "learning_rate": 0.0008497737783108237,
+      "loss": 3.0932,
+      "step": 160
+    },
+    {
+      "epoch": 0.3119266055045872,
+      "grad_norm": 0.2774140238761902,
+      "learning_rate": 0.0008274303669726426,
+      "loss": 3.091,
+      "step": 170
+    },
+    {
+      "epoch": 0.3302752293577982,
+      "grad_norm": 0.3296770751476288,
+      "learning_rate": 0.0008038782961218137,
+      "loss": 3.0774,
+      "step": 180
+    },
+    {
+      "epoch": 0.3486238532110092,
+      "grad_norm": 0.23089993000030518,
+      "learning_rate": 0.000779204504653894,
+      "loss": 3.0773,
+      "step": 190
+    },
+    {
+      "epoch": 0.3669724770642202,
+      "grad_norm": 0.22251616418361664,
+      "learning_rate": 0.0007535000721258791,
+      "loss": 3.0568,
+      "step": 200
+    },
+    {
+      "epoch": 0.3853211009174312,
+      "grad_norm": 0.3530291020870209,
+      "learning_rate": 0.00072685988254984,
+      "loss": 3.0555,
+      "step": 210
+    },
+    {
+      "epoch": 0.4036697247706422,
+      "grad_norm": 0.25301140546798706,
+      "learning_rate": 0.0006993822741429906,
+      "loss": 3.0642,
+      "step": 220
+    },
+    {
+      "epoch": 0.42201834862385323,
+      "grad_norm": 0.22952724993228912,
+      "learning_rate": 0.0006711686763270818,
+      "loss": 3.0655,
+      "step": 230
+    },
+    {
+      "epoch": 0.44036697247706424,
+      "grad_norm": 0.23224326968193054,
+      "learning_rate": 0.0006423232353170831,
+      "loss": 3.0489,
+      "step": 240
+    },
+    {
+      "epoch": 0.45871559633027525,
+      "grad_norm": 0.22438764572143555,
+      "learning_rate": 0.0006129524296812335,
+      "loss": 3.0391,
+      "step": 250
+    },
+    {
+      "epoch": 0.47706422018348627,
+      "grad_norm": 0.19759856164455414,
+      "learning_rate": 0.0005831646772915651,
+      "loss": 3.0502,
+      "step": 260
+    },
+    {
+      "epoch": 0.4954128440366973,
+      "grad_norm": 0.29346680641174316,
+      "learning_rate": 0.0005530699351157809,
+      "loss": 3.0324,
+      "step": 270
+    },
+    {
+      "epoch": 0.5137614678899083,
+      "grad_norm": 0.23926813900470734,
+      "learning_rate": 0.0005227792933277943,
+      "loss": 3.0358,
+      "step": 280
+    },
+    {
+      "epoch": 0.5321100917431193,
+      "grad_norm": 0.22763842344284058,
+      "learning_rate": 0.0004924045652352092,
+      "loss": 3.0181,
+      "step": 290
+    },
+    {
+      "epoch": 0.5504587155963303,
+      "grad_norm": 0.2015685886144638,
+      "learning_rate": 0.0004620578745374633,
+      "loss": 3.0192,
+      "step": 300
+    },
+    {
+      "epoch": 0.5688073394495413,
+      "grad_norm": 0.2214621603488922,
+      "learning_rate": 0.0004318512414382059,
+      "loss": 3.0261,
+      "step": 310
+    },
+    {
+      "epoch": 0.5871559633027523,
+      "grad_norm": 0.21981632709503174,
+      "learning_rate": 0.0004018961691397148,
+      "loss": 3.0231,
+      "step": 320
+    },
+    {
+      "epoch": 0.6055045871559633,
+      "grad_norm": 0.17365193367004395,
+      "learning_rate": 0.0003723032322457458,
+      "loss": 3.0074,
+      "step": 330
+    },
+    {
+      "epoch": 0.6238532110091743,
+      "grad_norm": 0.24682600796222687,
+      "learning_rate": 0.0003431816685921625,
+      "loss": 3.0104,
+      "step": 340
+    },
+    {
+      "epoch": 0.6422018348623854,
+      "grad_norm": 0.20161883533000946,
+      "learning_rate": 0.0003146389760120416,
+      "loss": 3.0153,
+      "step": 350
+    },
+    {
+      "epoch": 0.6605504587155964,
+      "grad_norm": 0.21748259663581848,
+      "learning_rate": 0.00028678051552373485,
+      "loss": 3.015,
+      "step": 360
+    },
+    {
+      "epoch": 0.6788990825688074,
+      "grad_norm": 0.19612346589565277,
+      "learning_rate": 0.0002597091224066581,
+      "loss": 3.0047,
+      "step": 370
+    },
+    {
+      "epoch": 0.6972477064220184,
+      "grad_norm": 0.24143937230110168,
+      "learning_rate": 0.0002335247266004629,
+      "loss": 3.0171,
+      "step": 380
+    },
+    {
+      "epoch": 0.7155963302752294,
+      "grad_norm": 0.21020065248012543,
+      "learning_rate": 0.000208323983828828,
+      "loss": 3.0062,
+      "step": 390
+    },
+    {
+      "epoch": 0.7339449541284404,
+      "grad_norm": 0.17595268785953522,
+      "learning_rate": 0.0001841999188095224,
+      "loss": 2.999,
+      "step": 400
+    },
+    {
+      "epoch": 0.7522935779816514,
+      "grad_norm": 0.17472228407859802,
+      "learning_rate": 0.00016124158186777676,
+      "loss": 2.997,
+      "step": 410
+    },
+    {
+      "epoch": 0.7706422018348624,
+      "grad_norm": 0.19418755173683167,
+      "learning_rate": 0.0001395337202205211,
+      "loss": 2.9948,
+      "step": 420
+    },
+    {
+      "epoch": 0.7889908256880734,
+      "grad_norm": 0.16787466406822205,
+      "learning_rate": 0.00011915646514489292,
+      "loss": 2.9974,
+      "step": 430
+    },
+    {
+      "epoch": 0.8073394495412844,
+      "grad_norm": 0.21235014498233795,
+      "learning_rate": 0.00010018503618578818,
+      "loss": 3.0044,
+      "step": 440
+    },
+    {
+      "epoch": 0.8256880733944955,
+      "grad_norm": 0.17561322450637817,
+      "learning_rate": 8.268946349432583e-05,
+      "loss": 3.0018,
+      "step": 450
+    },
+    {
+      "epoch": 0.8440366972477065,
+      "grad_norm": 0.18145301938056946,
+      "learning_rate": 6.673432932217117e-05,
+      "loss": 2.991,
+      "step": 460
+    },
+    {
+      "epoch": 0.8623853211009175,
+      "grad_norm": 0.1743510514497757,
+      "learning_rate": 5.2378529625954684e-05,
+      "loss": 2.98,
+      "step": 470
+    },
+    {
+      "epoch": 0.8807339449541285,
+      "grad_norm": 0.17581529915332794,
+      "learning_rate": 3.9675056661785556e-05,
+      "loss": 2.9886,
+      "step": 480
+    },
+    {
+      "epoch": 0.8990825688073395,
+      "grad_norm": 0.18438376486301422,
+      "learning_rate": 2.8670803372379538e-05,
+      "loss": 2.9934,
+      "step": 490
+    },
+    {
+      "epoch": 0.9174311926605505,
+      "grad_norm": 0.1702006757259369,
+      "learning_rate": 1.9406390288876586e-05,
+      "loss": 3.0018,
+      "step": 500
+    },
+    {
+      "epoch": 0.9357798165137615,
+      "grad_norm": 0.16569823026657104,
+      "learning_rate": 1.191601558631461e-05,
+      "loss": 2.9908,
+      "step": 510
+    },
+    {
+      "epoch": 0.9541284403669725,
+      "grad_norm": 0.15600372850894928,
+      "learning_rate": 6.2273288462559305e-06,
+      "loss": 2.9849,
+      "step": 520
+    },
+    {
+      "epoch": 0.9724770642201835,
+      "grad_norm": 0.16866841912269592,
+      "learning_rate": 2.361328992552314e-06,
+      "loss": 2.9881,
+      "step": 530
+    },
+    {
+      "epoch": 0.9908256880733946,
+      "grad_norm": 0.1598730981349945,
+      "learning_rate": 3.322867770044202e-07,
+      "loss": 2.9893,
+      "step": 540
+    },
+    {
+      "epoch": 1.0,
+      "step": 545,
+      "total_flos": 2.8892470700723405e+17,
+      "train_loss": 3.106896718926386,
+      "train_runtime": 5358.9968,
+      "train_samples_per_second": 104.148,
+      "train_steps_per_second": 0.102
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 545,
+  "num_input_tokens_seen": 34389120,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.8892470700723405e+17,
+  "train_batch_size": 128,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen25/dino_vitb8/finetune/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
+}

continuous/qwen25/dino_vitb8/finetune/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

continuous/qwen25/dino_vitb8/finetune/config.json

@@ -0,0 +1,87 @@
+{
+  "add_faster_video": false,
+  "add_time_instruction": false,
+  "architectures": [
+    "LlavaQwenForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "faster_token_stride": 10,
+  "force_sample": false,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "image_aspect_ratio": "square",
+  "image_crop_resolution": null,
+  "image_grid_pinpoints": null,
+  "image_split_resolution": null,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "mm_hidden_size": 768,
+  "mm_newline_position": "grid",
+  "mm_patch_merge_type": "flat",
+  "mm_projector_lr": 0.0001,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_resampler_type": null,
+  "mm_spatial_pool_mode": "bilinear",
+  "mm_spatial_pool_stride": null,
+  "mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -1,
+  "mm_vision_tower": "hf:/cache/ckpt/download/tokenizer/continuous/dino/dino-vitb8",
+  "mm_vision_tower_lr": null,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "pad_token_id": 151643,
+  "pos_skipping_range": 4096,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "tokenizer_model_max_length": 8192,
+  "tokenizer_padding_side": "right",
+  "transformers_version": "4.57.6",
+  "use_cache": true,
+  "use_mm_proj": true,
+  "use_pos_skipping": false,
+  "use_sliding_window": false,
+  "vision_tower_pretrained": null,
+  "vocab_size": 151936
+}

continuous/qwen25/dino_vitb8/finetune/generation_config.json

@@ -0,0 +1,13 @@
+{
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.57.6"
+}

continuous/qwen25/dino_vitb8/finetune/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82e33e43ef9a88cb80424f5cce7922d489172226541a00115306c480c60bbed3
+size 3741209672