Upload continuous evaluation checkpoints, batch 1

.gitattributes

@@ -101,3 +101,7 @@ continuous/qwen25/siglip2_l16_384/finetune/tokenizer.json filter=lfs diff=lfs me
 continuous/qwen25/siglip2_l16_512/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 continuous/qwen25/siglip2_sm14_224/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 continuous/qwen25/siglip2_sm14_384/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/siglip2_sm16_256/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/siglip2_sm16_384/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/siglip2_sm16_512/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen25/webssl_dino1b_full2b_224/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text

continuous/qwen25/siglip2_sm16_256/finetune/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a585663629ba2bc0c602f21730f18f8e75e9a394f8b62394d69b83ce83b2525
+size 4426565120

continuous/qwen25/siglip2_sm16_256/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/siglip2_sm16_256/finetune/tokenizer.json

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

continuous/qwen25/siglip2_sm16_256/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/siglip2_sm16_256/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": 16.5223388671875,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 1.8588,
+      "step": 10
+    },
+    {
+      "epoch": 0.007700452401578592,
+      "grad_norm": 10.279077529907227,
+      "learning_rate": 2.435897435897436e-06,
+      "loss": 1.4833,
+      "step": 20
+    },
+    {
+      "epoch": 0.01155067860236789,
+      "grad_norm": 3.5310592651367188,
+      "learning_rate": 3.7179487179487184e-06,
+      "loss": 1.0649,
+      "step": 30
+    },
+    {
+      "epoch": 0.015400904803157185,
+      "grad_norm": 0.6137400269508362,
+      "learning_rate": 5e-06,
+      "loss": 0.9026,
+      "step": 40
+    },
+    {
+      "epoch": 0.019251131003946483,
+      "grad_norm": 0.36786022782325745,
+      "learning_rate": 6.282051282051282e-06,
+      "loss": 0.871,
+      "step": 50
+    },
+    {
+      "epoch": 0.02310135720473578,
+      "grad_norm": 0.27316445112228394,
+      "learning_rate": 7.564102564102564e-06,
+      "loss": 0.8227,
+      "step": 60
+    },
+    {
+      "epoch": 0.026951583405525073,
+      "grad_norm": 0.229481503367424,
+      "learning_rate": 8.846153846153847e-06,
+      "loss": 0.7994,
+      "step": 70
+    },
+    {
+      "epoch": 0.03080180960631437,
+      "grad_norm": 0.22887854278087616,
+      "learning_rate": 9.999996111488723e-06,
+      "loss": 0.8093,
+      "step": 80
+    },
+    {
+      "epoch": 0.034652035807103666,
+      "grad_norm": 0.21463118493556976,
+      "learning_rate": 9.999529497453782e-06,
+      "loss": 0.793,
+      "step": 90
+    },
+    {
+      "epoch": 0.038502262007892966,
+      "grad_norm": 0.21482983231544495,
+      "learning_rate": 9.998285264324193e-06,
+      "loss": 0.7673,
+      "step": 100
+    },
+    {
+      "epoch": 0.04235248820868226,
+      "grad_norm": 0.2140115350484848,
+      "learning_rate": 9.996263605626052e-06,
+      "loss": 0.7633,
+      "step": 110
+    },
+    {
+      "epoch": 0.04620271440947156,
+      "grad_norm": 0.23018792271614075,
+      "learning_rate": 9.99346483580503e-06,
+      "loss": 0.7753,
+      "step": 120
+    },
+    {
+      "epoch": 0.05005294061026085,
+      "grad_norm": 0.2062118947505951,
+      "learning_rate": 9.989889390177467e-06,
+      "loss": 0.7642,
+      "step": 130
+    },
+    {
+      "epoch": 0.053903166811050146,
+      "grad_norm": 0.2358466535806656,
+      "learning_rate": 9.985537824862647e-06,
+      "loss": 0.7682,
+      "step": 140
+    },
+    {
+      "epoch": 0.057753393011839446,
+      "grad_norm": 0.22833313047885895,
+      "learning_rate": 9.980410816696317e-06,
+      "loss": 0.7462,
+      "step": 150
+    },
+    {
+      "epoch": 0.06160361921262874,
+      "grad_norm": 0.21916185319423676,
+      "learning_rate": 9.97450916312541e-06,
+      "loss": 0.7587,
+      "step": 160
+    },
+    {
+      "epoch": 0.06545384541341803,
+      "grad_norm": 0.19528739154338837,
+      "learning_rate": 9.967833782084004e-06,
+      "loss": 0.7613,
+      "step": 170
+    },
+    {
+      "epoch": 0.06930407161420733,
+      "grad_norm": 0.23004452884197235,
+      "learning_rate": 9.960385711850554e-06,
+      "loss": 0.7832,
+      "step": 180
+    },
+    {
+      "epoch": 0.07315429781499663,
+      "grad_norm": 0.22251315414905548,
+      "learning_rate": 9.952166110886394e-06,
+      "loss": 0.7759,
+      "step": 190
+    },
+    {
+      "epoch": 0.07700452401578593,
+      "grad_norm": 0.22837677597999573,
+      "learning_rate": 9.943176257655567e-06,
+      "loss": 0.7669,
+      "step": 200
+    },
+    {
+      "epoch": 0.08085475021657522,
+      "grad_norm": 0.21194984018802643,
+      "learning_rate": 9.933417550425952e-06,
+      "loss": 0.764,
+      "step": 210
+    },
+    {
+      "epoch": 0.08470497641736452,
+      "grad_norm": 0.2097940742969513,
+      "learning_rate": 9.9228915070518e-06,
+      "loss": 0.7459,
+      "step": 220
+    },
+    {
+      "epoch": 0.08855520261815382,
+      "grad_norm": 0.2098655253648758,
+      "learning_rate": 9.911599764737632e-06,
+      "loss": 0.7491,
+      "step": 230
+    },
+    {
+      "epoch": 0.09240542881894312,
+      "grad_norm": 0.23441480100154877,
+      "learning_rate": 9.899544079783606e-06,
+      "loss": 0.7449,
+      "step": 240
+    },
+    {
+      "epoch": 0.0962556550197324,
+      "grad_norm": 0.22449661791324615,
+      "learning_rate": 9.886726327312338e-06,
+      "loss": 0.7615,
+      "step": 250
+    },
+    {
+      "epoch": 0.1001058812205217,
+      "grad_norm": 0.215620219707489,
+      "learning_rate": 9.87314850097724e-06,
+      "loss": 0.7574,
+      "step": 260
+    },
+    {
+      "epoch": 0.103956107421311,
+      "grad_norm": 0.21416860818862915,
+      "learning_rate": 9.858812712652452e-06,
+      "loss": 0.7464,
+      "step": 270
+    },
+    {
+      "epoch": 0.10780633362210029,
+      "grad_norm": 0.20462992787361145,
+      "learning_rate": 9.843721192104341e-06,
+      "loss": 0.7445,
+      "step": 280
+    },
+    {
+      "epoch": 0.11165655982288959,
+      "grad_norm": 0.20790264010429382,
+      "learning_rate": 9.827876286644698e-06,
+      "loss": 0.7666,
+      "step": 290
+    },
+    {
+      "epoch": 0.11550678602367889,
+      "grad_norm": 0.19666379690170288,
+      "learning_rate": 9.811280460765643e-06,
+      "loss": 0.729,
+      "step": 300
+    },
+    {
+      "epoch": 0.11935701222446819,
+      "grad_norm": 0.20709699392318726,
+      "learning_rate": 9.793936295756292e-06,
+      "loss": 0.7354,
+      "step": 310
+    },
+    {
+      "epoch": 0.12320723842525748,
+      "grad_norm": 0.22377778589725494,
+      "learning_rate": 9.775846489301279e-06,
+      "loss": 0.7449,
+      "step": 320
+    },
+    {
+      "epoch": 0.12705746462604678,
+      "grad_norm": 0.22159449756145477,
+      "learning_rate": 9.757013855061157e-06,
+      "loss": 0.7407,
+      "step": 330
+    },
+    {
+      "epoch": 0.13090769082683607,
+      "grad_norm": 0.2289222925901413,
+      "learning_rate": 9.737441322234758e-06,
+      "loss": 0.7529,
+      "step": 340
+    },
+    {
+      "epoch": 0.13475791702762538,
+      "grad_norm": 0.22146202623844147,
+      "learning_rate": 9.717131935103605e-06,
+      "loss": 0.7539,
+      "step": 350
+    },
+    {
+      "epoch": 0.13860814322841467,
+      "grad_norm": 0.21488253772258759,
+      "learning_rate": 9.696088852558398e-06,
+      "loss": 0.7394,
+      "step": 360
+    },
+    {
+      "epoch": 0.14245836942920398,
+      "grad_norm": 0.21021930873394012,
+      "learning_rate": 9.674315347607686e-06,
+      "loss": 0.7677,
+      "step": 370
+    },
+    {
+      "epoch": 0.14630859562999327,
+      "grad_norm": 0.21442870795726776,
+      "learning_rate": 9.651814806868796e-06,
+      "loss": 0.7322,
+      "step": 380
+    },
+    {
+      "epoch": 0.15015882183078255,
+      "grad_norm": 0.22260090708732605,
+      "learning_rate": 9.62859073004107e-06,
+      "loss": 0.7317,
+      "step": 390
+    },
+    {
+      "epoch": 0.15400904803157187,
+      "grad_norm": 0.1911792755126953,
+      "learning_rate": 9.604646729361547e-06,
+      "loss": 0.735,
+      "step": 400
+    },
+    {
+      "epoch": 0.15785927423236115,
+      "grad_norm": 0.21483303606510162,
+      "learning_rate": 9.579986529043099e-06,
+      "loss": 0.7261,
+      "step": 410
+    },
+    {
+      "epoch": 0.16170950043315044,
+      "grad_norm": 0.21139773726463318,
+      "learning_rate": 9.554613964695189e-06,
+      "loss": 0.7453,
+      "step": 420
+    },
+    {
+      "epoch": 0.16555972663393975,
+      "grad_norm": 0.21646830439567566,
+      "learning_rate": 9.528532982727281e-06,
+      "loss": 0.7387,
+      "step": 430
+    },
+    {
+      "epoch": 0.16940995283472904,
+      "grad_norm": 0.20398741960525513,
+      "learning_rate": 9.501747639735018e-06,
+      "loss": 0.7327,
+      "step": 440
+    },
+    {
+      "epoch": 0.17326017903551832,
+      "grad_norm": 0.21464811265468597,
+      "learning_rate": 9.474262101869272e-06,
+      "loss": 0.7228,
+      "step": 450
+    },
+    {
+      "epoch": 0.17711040523630764,
+      "grad_norm": 0.2256910502910614,
+      "learning_rate": 9.446080644188141e-06,
+      "loss": 0.7303,
+      "step": 460
+    },
+    {
+      "epoch": 0.18096063143709692,
+      "grad_norm": 0.20019513368606567,
+      "learning_rate": 9.417207649992017e-06,
+      "loss": 0.7335,
+      "step": 470
+    },
+    {
+      "epoch": 0.18481085763788624,
+      "grad_norm": 0.21618060767650604,
+      "learning_rate": 9.387647610141816e-06,
+      "loss": 0.7321,
+      "step": 480
+    },
+    {
+      "epoch": 0.18866108383867553,
+      "grad_norm": 0.19863753020763397,
+      "learning_rate": 9.35740512236047e-06,
+      "loss": 0.7211,
+      "step": 490
+    },
+    {
+      "epoch": 0.1925113100394648,
+      "grad_norm": 0.19954964518547058,
+      "learning_rate": 9.326484890517809e-06,
+      "loss": 0.7337,
+      "step": 500
+    },
+    {
+      "epoch": 0.19636153624025413,
+      "grad_norm": 0.2133297175168991,
+      "learning_rate": 9.294891723898926e-06,
+      "loss": 0.7213,
+      "step": 510
+    },
+    {
+      "epoch": 0.2002117624410434,
+      "grad_norm": 0.2278682142496109,
+      "learning_rate": 9.262630536456153e-06,
+      "loss": 0.7509,
+      "step": 520
+    },
+    {
+      "epoch": 0.2040619886418327,
+      "grad_norm": 0.22576934099197388,
+      "learning_rate": 9.229706346044749e-06,
+      "loss": 0.7283,
+      "step": 530
+    },
+    {
+      "epoch": 0.207912214842622,
+      "grad_norm": 0.22960007190704346,
+      "learning_rate": 9.196124273642422e-06,
+      "loss": 0.7367,
+      "step": 540
+    },
+    {
+      "epoch": 0.2117624410434113,
+      "grad_norm": 0.21174666285514832,
+      "learning_rate": 9.161889542552842e-06,
+      "loss": 0.7452,
+      "step": 550
+    },
+    {
+      "epoch": 0.21561266724420058,
+      "grad_norm": 0.20648834109306335,
+      "learning_rate": 9.127007477593193e-06,
+      "loss": 0.7432,
+      "step": 560
+    },
+    {
+      "epoch": 0.2194628934449899,
+      "grad_norm": 0.2264903485774994,
+      "learning_rate": 9.091483504265974e-06,
+      "loss": 0.7487,
+      "step": 570
+    },
+    {
+      "epoch": 0.22331311964577918,
+      "grad_norm": 0.22076357901096344,
+      "learning_rate": 9.055323147915116e-06,
+      "loss": 0.731,
+      "step": 580
+    },
+    {
+      "epoch": 0.2271633458465685,
+      "grad_norm": 0.21919870376586914,
+      "learning_rate": 9.018532032866585e-06,
+      "loss": 0.7383,
+      "step": 590
+    },
+    {
+      "epoch": 0.23101357204735778,
+      "grad_norm": 0.1975955069065094,
+      "learning_rate": 8.981115881553579e-06,
+      "loss": 0.7328,
+      "step": 600
+    },
+    {
+      "epoch": 0.23486379824814707,
+      "grad_norm": 0.21961162984371185,
+      "learning_rate": 8.943080513626476e-06,
+      "loss": 0.7527,
+      "step": 610
+    },
+    {
+      "epoch": 0.23871402444893638,
+      "grad_norm": 0.20877903699874878,
+      "learning_rate": 8.904431845047641e-06,
+      "loss": 0.7384,
+      "step": 620
+    },
+    {
+      "epoch": 0.24256425064972567,
+      "grad_norm": 0.2213476300239563,
+      "learning_rate": 8.86517588717128e-06,
+      "loss": 0.7537,
+      "step": 630
+    },
+    {
+      "epoch": 0.24641447685051496,
+      "grad_norm": 0.20463763177394867,
+      "learning_rate": 8.82531874580844e-06,
+      "loss": 0.7499,
+      "step": 640
+    },
+    {
+      "epoch": 0.25026470305130427,
+      "grad_norm": 0.19397540390491486,
+      "learning_rate": 8.784866620277314e-06,
+      "loss": 0.7056,
+      "step": 650
+    },
+    {
+      "epoch": 0.25411492925209356,
+      "grad_norm": 0.22309426963329315,
+      "learning_rate": 8.743825802439018e-06,
+      "loss": 0.7333,
+      "step": 660
+    },
+    {
+      "epoch": 0.25796515545288284,
+      "grad_norm": 0.2459244728088379,
+      "learning_rate": 8.702202675718956e-06,
+      "loss": 0.746,
+      "step": 670
+    },
+    {
+      "epoch": 0.26181538165367213,
+      "grad_norm": 0.20446452498435974,
+      "learning_rate": 8.660003714113959e-06,
+      "loss": 0.7116,
+      "step": 680
+    },
+    {
+      "epoch": 0.26566560785446147,
+      "grad_norm": 0.2165312021970749,
+      "learning_rate": 8.617235481185323e-06,
+      "loss": 0.7334,
+      "step": 690
+    },
+    {
+      "epoch": 0.26951583405525076,
+      "grad_norm": 0.22003930807113647,
+      "learning_rate": 8.573904629037931e-06,
+      "loss": 0.7403,
+      "step": 700
+    },
+    {
+      "epoch": 0.27336606025604004,
+      "grad_norm": 0.20852847397327423,
+      "learning_rate": 8.530017897285585e-06,
+      "loss": 0.7247,
+      "step": 710
+    },
+    {
+      "epoch": 0.27721628645682933,
+      "grad_norm": 0.19812658429145813,
+      "learning_rate": 8.485582112002746e-06,
+      "loss": 0.7317,
+      "step": 720
+    },
+    {
+      "epoch": 0.2810665126576186,
+      "grad_norm": 0.20769116282463074,
+      "learning_rate": 8.440604184662811e-06,
+      "loss": 0.7297,
+      "step": 730
+    },
+    {
+      "epoch": 0.28491673885840796,
+      "grad_norm": 0.2131969928741455,
+      "learning_rate": 8.395091111063118e-06,
+      "loss": 0.7211,
+      "step": 740
+    },
+    {
+      "epoch": 0.28876696505919724,
+      "grad_norm": 0.19743511080741882,
+      "learning_rate": 8.349049970236822e-06,
+      "loss": 0.722,
+      "step": 750
+    },
+    {
+      "epoch": 0.29261719125998653,
+      "grad_norm": 0.21159623563289642,
+      "learning_rate": 8.302487923351842e-06,
+      "loss": 0.7125,
+      "step": 760
+    },
+    {
+      "epoch": 0.2964674174607758,
+      "grad_norm": 0.20609627664089203,
+      "learning_rate": 8.255412212597021e-06,
+      "loss": 0.7322,
+      "step": 770
+    },
+    {
+      "epoch": 0.3003176436615651,
+      "grad_norm": 0.20166881382465363,
+      "learning_rate": 8.207830160055688e-06,
+      "loss": 0.7285,
+      "step": 780
+    },
+    {
+      "epoch": 0.3041678698623544,
+      "grad_norm": 0.20951834321022034,
+      "learning_rate": 8.159749166566786e-06,
+      "loss": 0.7144,
+      "step": 790
+    },
+    {
+      "epoch": 0.30801809606314373,
+      "grad_norm": 0.20974215865135193,
+      "learning_rate": 8.11117671057377e-06,
+      "loss": 0.7379,
+      "step": 800
+    },
+    {
+      "epoch": 0.311868322263933,
+      "grad_norm": 0.22068065404891968,
+      "learning_rate": 8.062120346961404e-06,
+      "loss": 0.7291,
+      "step": 810
+    },
+    {
+      "epoch": 0.3157185484647223,
+      "grad_norm": 0.19726143777370453,
+      "learning_rate": 8.012587705880705e-06,
+      "loss": 0.7251,
+      "step": 820
+    },
+    {
+      "epoch": 0.3195687746655116,
+      "grad_norm": 0.2066507637500763,
+      "learning_rate": 7.962586491562146e-06,
+      "loss": 0.7112,
+      "step": 830
+    },
+    {
+      "epoch": 0.3234190008663009,
+      "grad_norm": 0.2240840345621109,
+      "learning_rate": 7.912124481117357e-06,
+      "loss": 0.7172,
+      "step": 840
+    },
+    {
+      "epoch": 0.3272692270670902,
+      "grad_norm": 0.2060500979423523,
+      "learning_rate": 7.861209523329494e-06,
+      "loss": 0.7326,
+      "step": 850
+    },
+    {
+      "epoch": 0.3311194532678795,
+      "grad_norm": 0.21054407954216003,
+      "learning_rate": 7.809849537432432e-06,
+      "loss": 0.7127,
+      "step": 860
+    },
+    {
+      "epoch": 0.3349696794686688,
+      "grad_norm": 0.20849989354610443,
+      "learning_rate": 7.758052511879035e-06,
+      "loss": 0.7244,
+      "step": 870
+    },
+    {
+      "epoch": 0.3388199056694581,
+      "grad_norm": 0.2084037810564041,
+      "learning_rate": 7.705826503098642e-06,
+      "loss": 0.7173,
+      "step": 880
+    },
+    {
+      "epoch": 0.34267013187024736,
+      "grad_norm": 0.20228733122348785,
+      "learning_rate": 7.653179634243979e-06,
+      "loss": 0.7207,
+      "step": 890
+    },
+    {
+      "epoch": 0.34652035807103665,
+      "grad_norm": 0.18771784007549286,
+      "learning_rate": 7.600120093927696e-06,
+      "loss": 0.7192,
+      "step": 900
+    },
+    {
+      "epoch": 0.350370584271826,
+      "grad_norm": 0.22171728312969208,
+      "learning_rate": 7.546656134948729e-06,
+      "loss": 0.7058,
+      "step": 910
+    },
+    {
+      "epoch": 0.3542208104726153,
+      "grad_norm": 0.21345941722393036,
+      "learning_rate": 7.492796073008661e-06,
+      "loss": 0.7105,
+      "step": 920
+    },
+    {
+      "epoch": 0.35807103667340456,
+      "grad_norm": 0.22487321496009827,
+      "learning_rate": 7.438548285418326e-06,
+      "loss": 0.6996,
+      "step": 930
+    },
+    {
+      "epoch": 0.36192126287419385,
+      "grad_norm": 0.2343212515115738,
+      "learning_rate": 7.3839212097948e-06,
+      "loss": 0.7328,
+      "step": 940
+    },
+    {
+      "epoch": 0.36577148907498314,
+      "grad_norm": 0.21486330032348633,
+      "learning_rate": 7.328923342749039e-06,
+      "loss": 0.7496,
+      "step": 950
+    },
+    {
+      "epoch": 0.3696217152757725,
+      "grad_norm": 0.21829958260059357,
+      "learning_rate": 7.273563238564319e-06,
+      "loss": 0.7331,
+      "step": 960
+    },
+    {
+      "epoch": 0.37347194147656176,
+      "grad_norm": 0.2102513164281845,
+      "learning_rate": 7.217849507865724e-06,
+      "loss": 0.7082,
+      "step": 970
+    },
+    {
+      "epoch": 0.37732216767735105,
+      "grad_norm": 0.21385601162910461,
+      "learning_rate": 7.161790816280858e-06,
+      "loss": 0.721,
+      "step": 980
+    },
+    {
+      "epoch": 0.38117239387814034,
+      "grad_norm": 0.2102343887090683,
+      "learning_rate": 7.1053958830920125e-06,
+      "loss": 0.6965,
+      "step": 990
+    },
+    {
+      "epoch": 0.3850226200789296,
+      "grad_norm": 0.20738691091537476,
+      "learning_rate": 7.0486734798799715e-06,
+      "loss": 0.719,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3888728462797189,
+      "grad_norm": 0.20244348049163818,
+      "learning_rate": 6.991632429159708e-06,
+      "loss": 0.7426,
+      "step": 1010
+    },
+    {
+      "epoch": 0.39272307248050825,
+      "grad_norm": 0.22459867596626282,
+      "learning_rate": 6.934281603008132e-06,
+      "loss": 0.7316,
+      "step": 1020
+    },
+    {
+      "epoch": 0.39657329868129754,
+      "grad_norm": 0.20640070736408234,
+      "learning_rate": 6.876629921684143e-06,
+      "loss": 0.73,
+      "step": 1030
+    },
+    {
+      "epoch": 0.4004235248820868,
+      "grad_norm": 0.21491914987564087,
+      "learning_rate": 6.818686352241199e-06,
+      "loss": 0.7194,
+      "step": 1040
+    },
+    {
+      "epoch": 0.4042737510828761,
+      "grad_norm": 0.19549460709095,
+      "learning_rate": 6.760459907132573e-06,
+      "loss": 0.7165,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4081239772836654,
+      "grad_norm": 0.19286374747753143,
+      "learning_rate": 6.701959642809589e-06,
+      "loss": 0.7218,
+      "step": 1060
+    },
+    {
+      "epoch": 0.41197420348445474,
+      "grad_norm": 0.214615598320961,
+      "learning_rate": 6.643194658312987e-06,
+      "loss": 0.71,
+      "step": 1070
+    },
+    {
+      "epoch": 0.415824429685244,
+      "grad_norm": 0.19974036514759064,
+      "learning_rate": 6.584174093857676e-06,
+      "loss": 0.738,
+      "step": 1080
+    },
+    {
+      "epoch": 0.4196746558860333,
+      "grad_norm": 0.2292102426290512,
+      "learning_rate": 6.5249071294110735e-06,
+      "loss": 0.7204,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4235248820868226,
+      "grad_norm": 0.21654601395130157,
+      "learning_rate": 6.465402983265274e-06,
+      "loss": 0.7352,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4273751082876119,
+      "grad_norm": 0.215592160820961,
+      "learning_rate": 6.405670910603243e-06,
+      "loss": 0.7295,
+      "step": 1110
+    },
+    {
+      "epoch": 0.43122533448840117,
+      "grad_norm": 0.2242317497730255,
+      "learning_rate": 6.345720202059291e-06,
+      "loss": 0.7256,
+      "step": 1120
+    },
+    {
+      "epoch": 0.4350755606891905,
+      "grad_norm": 0.20697486400604248,
+      "learning_rate": 6.285560182274008e-06,
+      "loss": 0.7036,
+      "step": 1130
+    },
+    {
+      "epoch": 0.4389257868899798,
+      "grad_norm": 0.20553885400295258,
+      "learning_rate": 6.225200208443936e-06,
+      "loss": 0.7275,
+      "step": 1140
+    },
+    {
+      "epoch": 0.4427760130907691,
+      "grad_norm": 0.20471136271953583,
+      "learning_rate": 6.16464966886616e-06,
+      "loss": 0.7147,
+      "step": 1150
+    },
+    {
+      "epoch": 0.44662623929155837,
+      "grad_norm": 0.24267861247062683,
+      "learning_rate": 6.103917981478065e-06,
+      "loss": 0.742,
+      "step": 1160
+    },
+    {
+      "epoch": 0.45047646549234766,
+      "grad_norm": 0.20877651870250702,
+      "learning_rate": 6.043014592392489e-06,
+      "loss": 0.7516,
+      "step": 1170
+    },
+    {
+      "epoch": 0.454326691693137,
+      "grad_norm": 0.19957910478115082,
+      "learning_rate": 5.981948974428485e-06,
+      "loss": 0.7255,
+      "step": 1180
+    },
+    {
+      "epoch": 0.4581769178939263,
+      "grad_norm": 0.21823523938655853,
+      "learning_rate": 5.920730625637934e-06,
+      "loss": 0.702,
+      "step": 1190
+    },
+    {
+      "epoch": 0.46202714409471557,
+      "grad_norm": 0.21917138993740082,
+      "learning_rate": 5.8593690678282355e-06,
+      "loss": 0.7174,
+      "step": 1200
+    },
+    {
+      "epoch": 0.46587737029550486,
+      "grad_norm": 0.22885772585868835,
+      "learning_rate": 5.797873845081301e-06,
+      "loss": 0.7212,
+      "step": 1210
+    },
+    {
+      "epoch": 0.46972759649629414,
+      "grad_norm": 0.21561771631240845,
+      "learning_rate": 5.736254522269075e-06,
+      "loss": 0.7352,
+      "step": 1220
+    },
+    {
+      "epoch": 0.47357782269708343,
+      "grad_norm": 0.20762450993061066,
+      "learning_rate": 5.6745206835658386e-06,
+      "loss": 0.7402,
+      "step": 1230
+    },
+    {
+      "epoch": 0.47742804889787277,
+      "grad_norm": 0.19863781332969666,
+      "learning_rate": 5.612681930957495e-06,
+      "loss": 0.7092,
+      "step": 1240
+    },
+    {
+      "epoch": 0.48127827509866206,
+      "grad_norm": 0.1985444277524948,
+      "learning_rate": 5.550747882748095e-06,
+      "loss": 0.7042,
+      "step": 1250
+    },
+    {
+      "epoch": 0.48512850129945134,
+      "grad_norm": 0.21716754138469696,
+      "learning_rate": 5.4887281720638195e-06,
+      "loss": 0.7183,
+      "step": 1260
+    },
+    {
+      "epoch": 0.48897872750024063,
+      "grad_norm": 0.23859506845474243,
+      "learning_rate": 5.42663244535466e-06,
+      "loss": 0.6974,
+      "step": 1270
+    },
+    {
+      "epoch": 0.4928289537010299,
+      "grad_norm": 0.21331992745399475,
+      "learning_rate": 5.3644703608940205e-06,
+      "loss": 0.7089,
+      "step": 1280
+    },
+    {
+      "epoch": 0.49667917990181926,
+      "grad_norm": 0.21410170197486877,
+      "learning_rate": 5.302251587276491e-06,
+      "loss": 0.7172,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5005294061026085,
+      "grad_norm": 0.1990998089313507,
+      "learning_rate": 5.2399858019140005e-06,
+      "loss": 0.7193,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5043796323033978,
+      "grad_norm": 0.196112722158432,
+      "learning_rate": 5.177682689530615e-06,
+      "loss": 0.7356,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5082298585041871,
+      "grad_norm": 0.2142736315727234,
+      "learning_rate": 5.11535194065619e-06,
+      "loss": 0.7015,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5120800847049765,
+      "grad_norm": 0.22944317758083344,
+      "learning_rate": 5.053003250119113e-06,
+      "loss": 0.7086,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5159303109057657,
+      "grad_norm": 0.21348921954631805,
+      "learning_rate": 4.9906463155383925e-06,
+      "loss": 0.7185,
+      "step": 1340
+    },
+    {
+      "epoch": 0.519780537106555,
+      "grad_norm": 0.2071591168642044,
+      "learning_rate": 4.9282908358153046e-06,
+      "loss": 0.715,
+      "step": 1350
+    },
+    {
+      "epoch": 0.5236307633073443,
+      "grad_norm": 0.23308095335960388,
+      "learning_rate": 4.8659465096248345e-06,
+      "loss": 0.7272,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5274809895081336,
+      "grad_norm": 0.21591894328594208,
+      "learning_rate": 4.803623033907169e-06,
+      "loss": 0.7216,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5313312157089229,
+      "grad_norm": 0.20193378627300262,
+      "learning_rate": 4.741330102359441e-06,
+      "loss": 0.7332,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5351814419097122,
+      "grad_norm": 0.2236698418855667,
+      "learning_rate": 4.679077403927996e-06,
+      "loss": 0.7117,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5390316681105015,
+      "grad_norm": 0.20412880182266235,
+      "learning_rate": 4.616874621301373e-06,
+      "loss": 0.7135,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5428818943112907,
+      "grad_norm": 0.2142334133386612,
+      "learning_rate": 4.554731429404293e-06,
+      "loss": 0.728,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5467321205120801,
+      "grad_norm": 0.21829096972942352,
+      "learning_rate": 4.492657493892822e-06,
+      "loss": 0.7089,
+      "step": 1420
+    },
+    {
+      "epoch": 0.5505823467128694,
+      "grad_norm": 0.22476932406425476,
+      "learning_rate": 4.430662469650999e-06,
+      "loss": 0.7262,
+      "step": 1430
+    },
+    {
+      "epoch": 0.5544325729136587,
+      "grad_norm": 0.2143750786781311,
+      "learning_rate": 4.368755999289118e-06,
+      "loss": 0.7331,
+      "step": 1440
+    },
+    {
+      "epoch": 0.558282799114448,
+      "grad_norm": 0.21672391891479492,
+      "learning_rate": 4.306947711643949e-06,
+      "loss": 0.7108,
+      "step": 1450
+    },
+    {
+      "epoch": 0.5621330253152372,
+      "grad_norm": 0.22160565853118896,
+      "learning_rate": 4.245247220281068e-06,
+      "loss": 0.7174,
+      "step": 1460
+    },
+    {
+      "epoch": 0.5659832515160266,
+      "grad_norm": 0.20425084233283997,
+      "learning_rate": 4.183664121999587e-06,
+      "loss": 0.7114,
+      "step": 1470
+    },
+    {
+      "epoch": 0.5698334777168159,
+      "grad_norm": 0.2185916155576706,
+      "learning_rate": 4.122207995339474e-06,
+      "loss": 0.7107,
+      "step": 1480
+    },
+    {
+      "epoch": 0.5736837039176051,
+      "grad_norm": 0.2048671841621399,
+      "learning_rate": 4.060888399091734e-06,
+      "loss": 0.7158,
+      "step": 1490
+    },
+    {
+      "epoch": 0.5775339301183945,
+      "grad_norm": 0.20187291502952576,
+      "learning_rate": 3.999714870811639e-06,
+      "loss": 0.7293,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5813841563191837,
+      "grad_norm": 0.209339901804924,
+      "learning_rate": 3.938696925335275e-06,
+      "loss": 0.7204,
+      "step": 1510
+    },
+    {
+      "epoch": 0.5852343825199731,
+      "grad_norm": 0.20689433813095093,
+      "learning_rate": 3.87784405329962e-06,
+      "loss": 0.728,
+      "step": 1520
+    },
+    {
+      "epoch": 0.5890846087207623,
+      "grad_norm": 0.20608074963092804,
+      "learning_rate": 3.817165719666382e-06,
+      "loss": 0.7264,
+      "step": 1530
+    },
+    {
+      "epoch": 0.5929348349215516,
+      "grad_norm": 0.20672789216041565,
+      "learning_rate": 3.756671362249841e-06,
+      "loss": 0.7359,
+      "step": 1540
+    },
+    {
+      "epoch": 0.596785061122341,
+      "grad_norm": 0.20468318462371826,
+      "learning_rate": 3.6963703902488907e-06,
+      "loss": 0.7201,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6006352873231302,
+      "grad_norm": 0.20581862330436707,
+      "learning_rate": 3.6362721827835655e-06,
+      "loss": 0.7303,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6044855135239195,
+      "grad_norm": 0.21857067942619324,
+      "learning_rate": 3.576386087436212e-06,
+      "loss": 0.731,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6083357397247088,
+      "grad_norm": 0.2229854017496109,
+      "learning_rate": 3.5167214187975913e-06,
+      "loss": 0.7213,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6121859659254981,
+      "grad_norm": 0.18574266135692596,
+      "learning_rate": 3.457287457018086e-06,
+      "loss": 0.7224,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6160361921262875,
+      "grad_norm": 0.19852448999881744,
+      "learning_rate": 3.3980934463643003e-06,
+      "loss": 0.7133,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6198864183270767,
+      "grad_norm": 0.20007064938545227,
+      "learning_rate": 3.3391485937812053e-06,
+      "loss": 0.6933,
+      "step": 1610
+    },
+    {
+      "epoch": 0.623736644527866,
+      "grad_norm": 0.2134966105222702,
+      "learning_rate": 3.280462067460115e-06,
+      "loss": 0.6931,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6275868707286553,
+      "grad_norm": 0.20651835203170776,
+      "learning_rate": 3.222042995412669e-06,
+      "loss": 0.725,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6314370969294446,
+      "grad_norm": 0.18892312049865723,
+      "learning_rate": 3.163900464051089e-06,
+      "loss": 0.7112,
+      "step": 1640
+    },
+    {
+      "epoch": 0.635287323130234,
+      "grad_norm": 0.20917019248008728,
+      "learning_rate": 3.1060435167748826e-06,
+      "loss": 0.7241,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6391375493310232,
+      "grad_norm": 0.2079373598098755,
+      "learning_rate": 3.048481152564253e-06,
+      "loss": 0.7051,
+      "step": 1660
+    },
+    {
+      "epoch": 0.6429877755318125,
+      "grad_norm": 0.22217437624931335,
+      "learning_rate": 2.9912223245804017e-06,
+      "loss": 0.7282,
+      "step": 1670
+    },
+    {
+      "epoch": 0.6468380017326018,
+      "grad_norm": 0.22039827704429626,
+      "learning_rate": 2.934275938772981e-06,
+      "loss": 0.718,
+      "step": 1680
+    },
+    {
+      "epoch": 0.6506882279333911,
+      "grad_norm": 0.20900562405586243,
+      "learning_rate": 2.8776508524948676e-06,
+      "loss": 0.7095,
+      "step": 1690
+    },
+    {
+      "epoch": 0.6545384541341804,
+      "grad_norm": 0.19117359817028046,
+      "learning_rate": 2.821355873124503e-06,
+      "loss": 0.7072,
+      "step": 1700
+    },
+    {
+      "epoch": 0.6583886803349697,
+      "grad_norm": 0.2131463885307312,
+      "learning_rate": 2.7653997566960144e-06,
+      "loss": 0.7232,
+      "step": 1710
+    },
+    {
+      "epoch": 0.662238906535759,
+      "grad_norm": 0.21731935441493988,
+      "learning_rate": 2.709791206537309e-06,
+      "loss": 0.7133,
+      "step": 1720
+    },
+    {
+      "epoch": 0.6660891327365482,
+      "grad_norm": 0.21039685606956482,
+      "learning_rate": 2.6545388719163747e-06,
+      "loss": 0.7111,
+      "step": 1730
+    },
+    {
+      "epoch": 0.6699393589373376,
+      "grad_norm": 0.2095271199941635,
+      "learning_rate": 2.599651346695979e-06,
+      "loss": 0.7161,
+      "step": 1740
+    },
+    {
+      "epoch": 0.6737895851381268,
+      "grad_norm": 0.20354920625686646,
+      "learning_rate": 2.545137167997007e-06,
+      "loss": 0.7161,
+      "step": 1750
+    },
+    {
+      "epoch": 0.6776398113389162,
+      "grad_norm": 0.20993053913116455,
+      "learning_rate": 2.4910048148705972e-06,
+      "loss": 0.7017,
+      "step": 1760
+    },
+    {
+      "epoch": 0.6814900375397055,
+      "grad_norm": 0.2231331467628479,
+      "learning_rate": 2.4372627069793327e-06,
+      "loss": 0.7469,
+      "step": 1770
+    },
+    {
+      "epoch": 0.6853402637404947,
+      "grad_norm": 0.22422011196613312,
+      "learning_rate": 2.3839192032876486e-06,
+      "loss": 0.7311,
+      "step": 1780
+    },
+    {
+      "epoch": 0.6891904899412841,
+      "grad_norm": 0.2099415510892868,
+      "learning_rate": 2.3309826007617047e-06,
+      "loss": 0.7413,
+      "step": 1790
+    },
+    {
+      "epoch": 0.6930407161420733,
+      "grad_norm": 0.2139127254486084,
+      "learning_rate": 2.278461133078881e-06,
+      "loss": 0.7048,
+      "step": 1800
+    },
+    {
+      "epoch": 0.6968909423428626,
+      "grad_norm": 0.22147376835346222,
+      "learning_rate": 2.2263629693471258e-06,
+      "loss": 0.6969,
+      "step": 1810
+    },
+    {
+      "epoch": 0.700741168543652,
+      "grad_norm": 0.2021714299917221,
+      "learning_rate": 2.1746962128343414e-06,
+      "loss": 0.7173,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7045913947444412,
+      "grad_norm": 0.20828241109848022,
+      "learning_rate": 2.1234688997080183e-06,
+      "loss": 0.7094,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7084416209452306,
+      "grad_norm": 0.22232882678508759,
+      "learning_rate": 2.0726889977853053e-06,
+      "loss": 0.7153,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7122918471460198,
+      "grad_norm": 0.22181066870689392,
+      "learning_rate": 2.022364405293703e-06,
+      "loss": 0.727,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7161420733468091,
+      "grad_norm": 0.21218717098236084,
+      "learning_rate": 1.9725029496425814e-06,
+      "loss": 0.7212,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7199922995475985,
+      "grad_norm": 0.193947896361351,
+      "learning_rate": 1.923112386205727e-06,
+      "loss": 0.7167,
+      "step": 1870
+    },
+    {
+      "epoch": 0.7238425257483877,
+      "grad_norm": 0.20337000489234924,
+      "learning_rate": 1.874200397115079e-06,
+      "loss": 0.7258,
+      "step": 1880
+    },
+    {
+      "epoch": 0.727692751949177,
+      "grad_norm": 0.21859799325466156,
+      "learning_rate": 1.825774590065857e-06,
+      "loss": 0.7075,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7315429781499663,
+      "grad_norm": 0.21706527471542358,
+      "learning_rate": 1.7778424971332813e-06,
+      "loss": 0.718,
+      "step": 1900
+    },
+    {
+      "epoch": 0.7353932043507556,
+      "grad_norm": 0.20317557454109192,
+      "learning_rate": 1.7304115736010407e-06,
+      "loss": 0.7143,
+      "step": 1910
+    },
+    {
+      "epoch": 0.739243430551545,
+      "grad_norm": 0.211965411901474,
+      "learning_rate": 1.6834891968017108e-06,
+      "loss": 0.7108,
+      "step": 1920
+    },
+    {
+      "epoch": 0.7430936567523342,
+      "grad_norm": 0.2189830094575882,
+      "learning_rate": 1.6370826649692884e-06,
+      "loss": 0.7208,
+      "step": 1930
+    },
+    {
+      "epoch": 0.7469438829531235,
+      "grad_norm": 0.22744102776050568,
+      "learning_rate": 1.5911991961040463e-06,
+      "loss": 0.7334,
+      "step": 1940
+    },
+    {
+      "epoch": 0.7507941091539128,
+      "grad_norm": 0.21669457852840424,
+      "learning_rate": 1.545845926849851e-06,
+      "loss": 0.7217,
+      "step": 1950
+    },
+    {
+      "epoch": 0.7546443353547021,
+      "grad_norm": 0.22910113632678986,
+      "learning_rate": 1.5010299113841397e-06,
+      "loss": 0.7376,
+      "step": 1960
+    },
+    {
+      "epoch": 0.7584945615554913,
+      "grad_norm": 0.20612262189388275,
+      "learning_rate": 1.4567581203207238e-06,
+      "loss": 0.7268,
+      "step": 1970
+    },
+    {
+      "epoch": 0.7623447877562807,
+      "grad_norm": 0.21892288327217102,
+      "learning_rate": 1.413037439625593e-06,
+      "loss": 0.7239,
+      "step": 1980
+    },
+    {
+      "epoch": 0.76619501395707,
+      "grad_norm": 0.2234027087688446,
+      "learning_rate": 1.3698746695458847e-06,
+      "loss": 0.7195,
+      "step": 1990
+    },
+    {
+      "epoch": 0.7700452401578592,
+      "grad_norm": 0.20858773589134216,
+      "learning_rate": 1.3272765235521745e-06,
+      "loss": 0.7114,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7738954663586486,
+      "grad_norm": 0.2070193886756897,
+      "learning_rate": 1.2852496272942877e-06,
+      "loss": 0.7227,
+      "step": 2010
+    },
+    {
+      "epoch": 0.7777456925594378,
+      "grad_norm": 0.19900351762771606,
+      "learning_rate": 1.2438005175707396e-06,
+      "loss": 0.7046,
+      "step": 2020
+    },
+    {
+      "epoch": 0.7815959187602272,
+      "grad_norm": 0.20986634492874146,
+      "learning_rate": 1.2029356413120297e-06,
+      "loss": 0.691,
+      "step": 2030
+    },
+    {
+      "epoch": 0.7854461449610165,
+      "grad_norm": 0.20780734717845917,
+      "learning_rate": 1.1626613545778786e-06,
+      "loss": 0.7083,
+      "step": 2040
+    },
+    {
+      "epoch": 0.7892963711618057,
+      "grad_norm": 0.2026374489068985,
+      "learning_rate": 1.1229839215686294e-06,
+      "loss": 0.7124,
+      "step": 2050
+    },
+    {
+      "epoch": 0.7931465973625951,
+      "grad_norm": 0.1927707940340042,
+      "learning_rate": 1.0839095136509188e-06,
+      "loss": 0.6939,
+      "step": 2060
+    },
+    {
+      "epoch": 0.7969968235633843,
+      "grad_norm": 0.22303506731987,
+      "learning_rate": 1.045444208397791e-06,
+      "loss": 0.7163,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8008470497641736,
+      "grad_norm": 0.20913976430892944,
+      "learning_rate": 1.007593988643402e-06,
+      "loss": 0.7021,
+      "step": 2080
+    },
+    {
+      "epoch": 0.804697275964963,
+      "grad_norm": 0.20027056336402893,
+      "learning_rate": 9.703647415524613e-07,
+      "loss": 0.7251,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8085475021657522,
+      "grad_norm": 0.20703306794166565,
+      "learning_rate": 9.337622577045485e-07,
+      "loss": 0.7096,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8123977283665416,
+      "grad_norm": 0.2127198427915573,
+      "learning_rate": 8.977922301934588e-07,
+      "loss": 0.7248,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8162479545673308,
+      "grad_norm": 0.21170416474342346,
+      "learning_rate": 8.624602537417009e-07,
+      "loss": 0.7252,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8200981807681201,
+      "grad_norm": 0.2224230319261551,
+      "learning_rate": 8.277718238303123e-07,
+      "loss": 0.7085,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8239484069689095,
+      "grad_norm": 0.21198385953903198,
+      "learning_rate": 7.937323358440935e-07,
+      "loss": 0.7136,
+      "step": 2140
+    },
+    {
+      "epoch": 0.8277986331696987,
+      "grad_norm": 0.22860290110111237,
+      "learning_rate": 7.603470842324223e-07,
+      "loss": 0.7026,
+      "step": 2150
+    },
+    {
+      "epoch": 0.831648859370488,
+      "grad_norm": 0.20368261635303497,
+      "learning_rate": 7.276212616857575e-07,
+      "loss": 0.7241,
+      "step": 2160
+    },
+    {
+      "epoch": 0.8354990855712773,
+      "grad_norm": 0.20911520719528198,
+      "learning_rate": 6.955599583279837e-07,
+      "loss": 0.7217,
+      "step": 2170
+    },
+    {
+      "epoch": 0.8393493117720666,
+      "grad_norm": 0.20555202662944794,
+      "learning_rate": 6.641681609246981e-07,
+      "loss": 0.716,
+      "step": 2180
+    },
+    {
+      "epoch": 0.8431995379728559,
+      "grad_norm": 0.20811577141284943,
+      "learning_rate": 6.334507521075711e-07,
+      "loss": 0.7193,
+      "step": 2190
+    },
+    {
+      "epoch": 0.8470497641736452,
+      "grad_norm": 0.21851827204227448,
+      "learning_rate": 6.034125096149179e-07,
+      "loss": 0.7181,
+      "step": 2200
+    },
+    {
+      "epoch": 0.8508999903744345,
+      "grad_norm": 0.20766447484493256,
+      "learning_rate": 5.740581055485722e-07,
+      "loss": 0.6859,
+      "step": 2210
+    },
+    {
+      "epoch": 0.8547502165752238,
+      "grad_norm": 0.1962718367576599,
+      "learning_rate": 5.453921056471961e-07,
+      "loss": 0.7193,
+      "step": 2220
+    },
+    {
+      "epoch": 0.8586004427760131,
+      "grad_norm": 0.2107202708721161,
+      "learning_rate": 5.174189685761288e-07,
+      "loss": 0.7164,
+      "step": 2230
+    },
+    {
+      "epoch": 0.8624506689768023,
+      "grad_norm": 0.19262035191059113,
+      "learning_rate": 4.901430452338979e-07,
+      "loss": 0.728,
+      "step": 2240
+    },
+    {
+      "epoch": 0.8663008951775917,
+      "grad_norm": 0.2017219066619873,
+      "learning_rate": 4.6356857807548286e-07,
+      "loss": 0.7092,
+      "step": 2250
+    },
+    {
+      "epoch": 0.870151121378381,
+      "grad_norm": 0.20879675447940826,
+      "learning_rate": 4.376997004524541e-07,
+      "loss": 0.6998,
+      "step": 2260
+    },
+    {
+      "epoch": 0.8740013475791703,
+      "grad_norm": 0.2043522298336029,
+      "learning_rate": 4.12540435970073e-07,
+      "loss": 0.7053,
+      "step": 2270
+    },
+    {
+      "epoch": 0.8778515737799596,
+      "grad_norm": 0.20711331069469452,
+      "learning_rate": 3.880946978614703e-07,
+      "loss": 0.7229,
+      "step": 2280
+    },
+    {
+      "epoch": 0.8817017999807488,
+      "grad_norm": 0.20641683042049408,
+      "learning_rate": 3.643662883789878e-07,
+      "loss": 0.719,
+      "step": 2290
+    },
+    {
+      "epoch": 0.8855520261815382,
+      "grad_norm": 0.1957588791847229,
+      "learning_rate": 3.41358898202781e-07,
+      "loss": 0.7013,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8894022523823275,
+      "grad_norm": 0.19569769501686096,
+      "learning_rate": 3.190761058667741e-07,
+      "loss": 0.7122,
+      "step": 2310
+    },
+    {
+      "epoch": 0.8932524785831167,
+      "grad_norm": 0.20647300779819489,
+      "learning_rate": 2.9752137720206463e-07,
+      "loss": 0.7154,
+      "step": 2320
+    },
+    {
+      "epoch": 0.8971027047839061,
+      "grad_norm": 0.20414847135543823,
+      "learning_rate": 2.7669806479785067e-07,
+      "loss": 0.7222,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9009529309846953,
+      "grad_norm": 0.2061527669429779,
+      "learning_rate": 2.566094074799763e-07,
+      "loss": 0.7127,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9048031571854847,
+      "grad_norm": 0.21131685376167297,
+      "learning_rate": 2.3725852980716557e-07,
+      "loss": 0.7022,
+      "step": 2350
+    },
+    {
+      "epoch": 0.908653383386274,
+      "grad_norm": 0.20414315164089203,
+      "learning_rate": 2.1864844158503983e-07,
+      "loss": 0.7093,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9125036095870632,
+      "grad_norm": 0.2215423732995987,
+      "learning_rate": 2.007820373979702e-07,
+      "loss": 0.7032,
+      "step": 2370
+    },
+    {
+      "epoch": 0.9163538357878526,
+      "grad_norm": 0.2057316154241562,
+      "learning_rate": 1.8366209615886532e-07,
+      "loss": 0.7145,
+      "step": 2380
+    },
+    {
+      "epoch": 0.9202040619886418,
+      "grad_norm": 0.20643140375614166,
+      "learning_rate": 1.6729128067693613e-07,
+      "loss": 0.7284,
+      "step": 2390
+    },
+    {
+      "epoch": 0.9240542881894311,
+      "grad_norm": 0.22213076055049896,
+      "learning_rate": 1.5167213724353426e-07,
+      "loss": 0.7265,
+      "step": 2400
+    },
+    {
+      "epoch": 0.9279045143902205,
+      "grad_norm": 0.20755769312381744,
+      "learning_rate": 1.3680709523610058e-07,
+      "loss": 0.7056,
+      "step": 2410
+    },
+    {
+      "epoch": 0.9317547405910097,
+      "grad_norm": 0.20528154075145721,
+      "learning_rate": 1.2269846674030418e-07,
+      "loss": 0.729,
+      "step": 2420
+    },
+    {
+      "epoch": 0.935604966791799,
+      "grad_norm": 0.21954205632209778,
+      "learning_rate": 1.0934844619042595e-07,
+      "loss": 0.7261,
+      "step": 2430
+    },
+    {
+      "epoch": 0.9394551929925883,
+      "grad_norm": 0.21796636283397675,
+      "learning_rate": 9.675911002803961e-08,
+      "loss": 0.7125,
+      "step": 2440
+    },
+    {
+      "epoch": 0.9433054191933776,
+      "grad_norm": 0.22204765677452087,
+      "learning_rate": 8.493241637904381e-08,
+      "loss": 0.7098,
+      "step": 2450
+    },
+    {
+      "epoch": 0.9471556453941669,
+      "grad_norm": 0.20998162031173706,
+      "learning_rate": 7.387020474909689e-08,
+      "loss": 0.7312,
+      "step": 2460
+    },
+    {
+      "epoch": 0.9510058715949562,
+      "grad_norm": 0.2268427461385727,
+      "learning_rate": 6.357419573750856e-08,
+      "loss": 0.7266,
+      "step": 2470
+    },
+    {
+      "epoch": 0.9548560977957455,
+      "grad_norm": 0.20226220786571503,
+      "learning_rate": 5.4045990769613365e-08,
+      "loss": 0.7061,
+      "step": 2480
+    },
+    {
+      "epoch": 0.9587063239965348,
+      "grad_norm": 0.2090945690870285,
+      "learning_rate": 4.5287071847694434e-08,
+      "loss": 0.7129,
+      "step": 2490
+    },
+    {
+      "epoch": 0.9625565501973241,
+      "grad_norm": 0.20961636304855347,
+      "learning_rate": 3.7298801320467817e-08,
+      "loss": 0.712,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9664067763981133,
+      "grad_norm": 0.20712393522262573,
+      "learning_rate": 3.0082421671192576e-08,
+      "loss": 0.7054,
+      "step": 2510
+    },
+    {
+      "epoch": 0.9702570025989027,
+      "grad_norm": 0.20745469629764557,
+      "learning_rate": 2.3639055324410355e-08,
+      "loss": 0.726,
+      "step": 2520
+    },
+    {
+      "epoch": 0.974107228799692,
+      "grad_norm": 0.20707735419273376,
+      "learning_rate": 1.7969704471369477e-08,
+      "loss": 0.7098,
+      "step": 2530
+    },
+    {
+      "epoch": 0.9779574550004813,
+      "grad_norm": 0.19838108122348785,
+      "learning_rate": 1.3075250914142413e-08,
+      "loss": 0.713,
+      "step": 2540
+    },
+    {
+      "epoch": 0.9818076812012706,
+      "grad_norm": 0.19777509570121765,
+      "learning_rate": 8.956455928476049e-09,
+      "loss": 0.7204,
+      "step": 2550
+    },
+    {
+      "epoch": 0.9856579074020598,
+      "grad_norm": 0.20410962402820587,
+      "learning_rate": 5.6139601453786275e-09,
+      "loss": 0.7112,
+      "step": 2560
+    },
+    {
+      "epoch": 0.9895081336028492,
+      "grad_norm": 0.20628327131271362,
+      "learning_rate": 3.0482834514805602e-09,
+      "loss": 0.703,
+      "step": 2570
+    },
+    {
+      "epoch": 0.9933583598036385,
+      "grad_norm": 0.20474329590797424,
+      "learning_rate": 1.2598249081713366e-09,
+      "loss": 0.6917,
+      "step": 2580
+    },
+    {
+      "epoch": 0.9972085860044277,
+      "grad_norm": 0.21514707803726196,
+      "learning_rate": 2.488626895302826e-10,
+      "loss": 0.7288,
+      "step": 2590
+    },
+    {
+      "epoch": 0.9999037443449803,
+      "step": 2597,
+      "total_flos": 4.3787821407461704e+18,
+      "train_loss": 0.7356037460109753,
+      "train_runtime": 8565.6387,
+      "train_samples_per_second": 77.624,
+      "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": 4.3787821407461704e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen25/siglip2_sm16_256/finetune/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e062922aae0aaa76a7668e54ff7a5fc2337c1f1dc54c55c5d52ecccdc798dbe
+size 6136

continuous/qwen25/siglip2_sm16_256/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": 1152,
+  "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-SO400M-16-SigLIP2-256",
+  "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/siglip2/siglip2_sm16_256.npz",
+  "vocab_size": 151936
+}

continuous/qwen25/siglip2_sm16_256/pretrain/mm_projector.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f27d33266f22624ddd0783d68b1e4ce1545997d878ee470ba7951fe54f06fd42
+size 16529464

continuous/qwen25/siglip2_sm16_256/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": 2.6759276390075684,
+      "learning_rate": 0.00032142857142857147,
+      "loss": 4.5663,
+      "step": 10
+    },
+    {
+      "epoch": 0.03669724770642202,
+      "grad_norm": 0.1978318989276886,
+      "learning_rate": 0.0006785714285714287,
+      "loss": 3.4192,
+      "step": 20
+    },
+    {
+      "epoch": 0.05504587155963303,
+      "grad_norm": 0.16996213793754578,
+      "learning_rate": 0.0009999907688176173,
+      "loss": 3.2561,
+      "step": 30
+    },
+    {
+      "epoch": 0.07339449541284404,
+      "grad_norm": 0.14298075437545776,
+      "learning_rate": 0.0009988834393115766,
+      "loss": 3.1693,
+      "step": 40
+    },
+    {
+      "epoch": 0.09174311926605505,
+      "grad_norm": 0.15948356688022614,
+      "learning_rate": 0.0009959345572940772,
+      "loss": 3.0803,
+      "step": 50
+    },
+    {
+      "epoch": 0.11009174311926606,
+      "grad_norm": 0.1624080091714859,
+      "learning_rate": 0.0009911550081155983,
+      "loss": 2.9826,
+      "step": 60
+    },
+    {
+      "epoch": 0.12844036697247707,
+      "grad_norm": 0.141656756401062,
+      "learning_rate": 0.0009845624347566433,
+      "loss": 2.9169,
+      "step": 70
+    },
+    {
+      "epoch": 0.14678899082568808,
+      "grad_norm": 0.1767512857913971,
+      "learning_rate": 0.0009761811727013547,
+      "loss": 2.8651,
+      "step": 80
+    },
+    {
+      "epoch": 0.1651376146788991,
+      "grad_norm": 0.1317259520292282,
+      "learning_rate": 0.0009660421601067666,
+      "loss": 2.8167,
+      "step": 90
+    },
+    {
+      "epoch": 0.1834862385321101,
+      "grad_norm": 0.1974857747554779,
+      "learning_rate": 0.0009541828235992926,
+      "loss": 2.7944,
+      "step": 100
+    },
+    {
+      "epoch": 0.2018348623853211,
+      "grad_norm": 0.14653581380844116,
+      "learning_rate": 0.0009406469401200151,
+      "loss": 2.76,
+      "step": 110
+    },
+    {
+      "epoch": 0.22018348623853212,
+      "grad_norm": 0.19846349954605103,
+      "learning_rate": 0.0009254844753287492,
+      "loss": 2.7427,
+      "step": 120
+    },
+    {
+      "epoch": 0.23853211009174313,
+      "grad_norm": 0.19003595411777496,
+      "learning_rate": 0.0009087513991633924,
+      "loss": 2.7115,
+      "step": 130
+    },
+    {
+      "epoch": 0.25688073394495414,
+      "grad_norm": 0.14613375067710876,
+      "learning_rate": 0.0008905094792353917,
+      "loss": 2.6962,
+      "step": 140
+    },
+    {
+      "epoch": 0.27522935779816515,
+      "grad_norm": 0.1403244584798813,
+      "learning_rate": 0.0008708260528239789,
+      "loss": 2.6871,
+      "step": 150
+    },
+    {
+      "epoch": 0.29357798165137616,
+      "grad_norm": 0.13765309751033783,
+      "learning_rate": 0.0008497737783108237,
+      "loss": 2.6617,
+      "step": 160
+    },
+    {
+      "epoch": 0.3119266055045872,
+      "grad_norm": 0.14537347853183746,
+      "learning_rate": 0.0008274303669726426,
+      "loss": 2.652,
+      "step": 170
+    },
+    {
+      "epoch": 0.3302752293577982,
+      "grad_norm": 0.1508922278881073,
+      "learning_rate": 0.0008038782961218137,
+      "loss": 2.6315,
+      "step": 180
+    },
+    {
+      "epoch": 0.3486238532110092,
+      "grad_norm": 0.16825449466705322,
+      "learning_rate": 0.000779204504653894,
+      "loss": 2.6074,
+      "step": 190
+    },
+    {
+      "epoch": 0.3669724770642202,
+      "grad_norm": 0.14644593000411987,
+      "learning_rate": 0.0007535000721258791,
+      "loss": 2.569,
+      "step": 200
+    },
+    {
+      "epoch": 0.3853211009174312,
+      "grad_norm": 0.15500357747077942,
+      "learning_rate": 0.00072685988254984,
+      "loss": 2.5505,
+      "step": 210
+    },
+    {
+      "epoch": 0.4036697247706422,
+      "grad_norm": 0.13833682239055634,
+      "learning_rate": 0.0006993822741429906,
+      "loss": 2.5468,
+      "step": 220
+    },
+    {
+      "epoch": 0.42201834862385323,
+      "grad_norm": 0.14456671476364136,
+      "learning_rate": 0.0006711686763270818,
+      "loss": 2.5245,
+      "step": 230
+    },
+    {
+      "epoch": 0.44036697247706424,
+      "grad_norm": 0.1353011131286621,
+      "learning_rate": 0.0006423232353170831,
+      "loss": 2.5056,
+      "step": 240
+    },
+    {
+      "epoch": 0.45871559633027525,
+      "grad_norm": 0.14815090596675873,
+      "learning_rate": 0.0006129524296812335,
+      "loss": 2.4953,
+      "step": 250
+    },
+    {
+      "epoch": 0.47706422018348627,
+      "grad_norm": 0.1291130632162094,
+      "learning_rate": 0.0005831646772915651,
+      "loss": 2.4932,
+      "step": 260
+    },
+    {
+      "epoch": 0.4954128440366973,
+      "grad_norm": 0.11578568071126938,
+      "learning_rate": 0.0005530699351157809,
+      "loss": 2.4685,
+      "step": 270
+    },
+    {
+      "epoch": 0.5137614678899083,
+      "grad_norm": 0.1318187266588211,
+      "learning_rate": 0.0005227792933277943,
+      "loss": 2.4685,
+      "step": 280
+    },
+    {
+      "epoch": 0.5321100917431193,
+      "grad_norm": 0.12018845975399017,
+      "learning_rate": 0.0004924045652352092,
+      "loss": 2.4551,
+      "step": 290
+    },
+    {
+      "epoch": 0.5504587155963303,
+      "grad_norm": 0.12459403276443481,
+      "learning_rate": 0.0004620578745374633,
+      "loss": 2.4514,
+      "step": 300
+    },
+    {
+      "epoch": 0.5688073394495413,
+      "grad_norm": 0.13622161746025085,
+      "learning_rate": 0.0004318512414382059,
+      "loss": 2.4557,
+      "step": 310
+    },
+    {
+      "epoch": 0.5871559633027523,
+      "grad_norm": 0.1258372962474823,
+      "learning_rate": 0.0004018961691397148,
+      "loss": 2.4552,
+      "step": 320
+    },
+    {
+      "epoch": 0.6055045871559633,
+      "grad_norm": 0.13844631612300873,
+      "learning_rate": 0.0003723032322457458,
+      "loss": 2.4354,
+      "step": 330
+    },
+    {
+      "epoch": 0.6238532110091743,
+      "grad_norm": 0.12740880250930786,
+      "learning_rate": 0.0003431816685921625,
+      "loss": 2.4364,
+      "step": 340
+    },
+    {
+      "epoch": 0.6422018348623854,
+      "grad_norm": 0.11718268692493439,
+      "learning_rate": 0.0003146389760120416,
+      "loss": 2.4341,
+      "step": 350
+    },
+    {
+      "epoch": 0.6605504587155964,
+      "grad_norm": 0.12721788883209229,
+      "learning_rate": 0.00028678051552373485,
+      "loss": 2.4354,
+      "step": 360
+    },
+    {
+      "epoch": 0.6788990825688074,
+      "grad_norm": 0.11615406721830368,
+      "learning_rate": 0.0002597091224066581,
+      "loss": 2.4269,
+      "step": 370
+    },
+    {
+      "epoch": 0.6972477064220184,
+      "grad_norm": 0.12004566937685013,
+      "learning_rate": 0.0002335247266004629,
+      "loss": 2.4337,
+      "step": 380
+    },
+    {
+      "epoch": 0.7155963302752294,
+      "grad_norm": 0.12581804394721985,
+      "learning_rate": 0.000208323983828828,
+      "loss": 2.4231,
+      "step": 390
+    },
+    {
+      "epoch": 0.7339449541284404,
+      "grad_norm": 0.11600212007761002,
+      "learning_rate": 0.0001841999188095224,
+      "loss": 2.4136,
+      "step": 400
+    },
+    {
+      "epoch": 0.7522935779816514,
+      "grad_norm": 0.11806987226009369,
+      "learning_rate": 0.00016124158186777676,
+      "loss": 2.4203,
+      "step": 410
+    },
+    {
+      "epoch": 0.7706422018348624,
+      "grad_norm": 0.11500902473926544,
+      "learning_rate": 0.0001395337202205211,
+      "loss": 2.4146,
+      "step": 420
+    },
+    {
+      "epoch": 0.7889908256880734,
+      "grad_norm": 0.11047299206256866,
+      "learning_rate": 0.00011915646514489292,
+      "loss": 2.4139,
+      "step": 430
+    },
+    {
+      "epoch": 0.8073394495412844,
+      "grad_norm": 0.11657673120498657,
+      "learning_rate": 0.00010018503618578818,
+      "loss": 2.419,
+      "step": 440
+    },
+    {
+      "epoch": 0.8256880733944955,
+      "grad_norm": 0.11404619365930557,
+      "learning_rate": 8.268946349432583e-05,
+      "loss": 2.4112,
+      "step": 450
+    },
+    {
+      "epoch": 0.8440366972477065,
+      "grad_norm": 0.11190588027238846,
+      "learning_rate": 6.673432932217117e-05,
+      "loss": 2.4084,
+      "step": 460
+    },
+    {
+      "epoch": 0.8623853211009175,
+      "grad_norm": 0.11402174085378647,
+      "learning_rate": 5.2378529625954684e-05,
+      "loss": 2.398,
+      "step": 470
+    },
+    {
+      "epoch": 0.8807339449541285,
+      "grad_norm": 0.10369860380887985,
+      "learning_rate": 3.9675056661785556e-05,
+      "loss": 2.4029,
+      "step": 480
+    },
+    {
+      "epoch": 0.8990825688073395,
+      "grad_norm": 0.10978896915912628,
+      "learning_rate": 2.8670803372379538e-05,
+      "loss": 2.4093,
+      "step": 490
+    },
+    {
+      "epoch": 0.9174311926605505,
+      "grad_norm": 0.10856729000806808,
+      "learning_rate": 1.9406390288876586e-05,
+      "loss": 2.4159,
+      "step": 500
+    },
+    {
+      "epoch": 0.9357798165137615,
+      "grad_norm": 0.1097959578037262,
+      "learning_rate": 1.191601558631461e-05,
+      "loss": 2.4056,
+      "step": 510
+    },
+    {
+      "epoch": 0.9541284403669725,
+      "grad_norm": 0.11092640459537506,
+      "learning_rate": 6.2273288462559305e-06,
+      "loss": 2.4016,
+      "step": 520
+    },
+    {
+      "epoch": 0.9724770642201835,
+      "grad_norm": 0.10984369367361069,
+      "learning_rate": 2.361328992552314e-06,
+      "loss": 2.405,
+      "step": 530
+    },
+    {
+      "epoch": 0.9908256880733946,
+      "grad_norm": 0.11385040730237961,
+      "learning_rate": 3.322867770044202e-07,
+      "loss": 2.4024,
+      "step": 540
+    },
+    {
+      "epoch": 1.0,
+      "step": 545,
+      "total_flos": 3.464814042902692e+17,
+      "train_loss": 2.6209807964639924,
+      "train_runtime": 5263.8357,
+      "train_samples_per_second": 106.031,
+      "train_steps_per_second": 0.104
+    }
+  ],
+  "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": 3.464814042902692e+17,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen25/siglip2_sm16_384/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/siglip2_sm16_384/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": 1152,
+  "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-SO400M-16-SigLIP2-384",
+  "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/siglip2/siglip2_sm16_384.npz",
+  "vocab_size": 151936
+}

continuous/qwen25/siglip2_sm16_384/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/siglip2_sm16_384/finetune/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea894b0a8ce24ed26fca190fdb3670e68f01307789e00be77eed9d44959c40d4
+size 4427302400

continuous/qwen25/siglip2_sm16_384/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/siglip2_sm16_384/finetune/tokenizer.json

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

continuous/qwen25/siglip2_sm16_384/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/siglip2_sm16_384/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": 8.265983581542969,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 1.9383,
+      "step": 10
+    },
+    {
+      "epoch": 0.007700452401578592,
+      "grad_norm": 6.2031073570251465,
+      "learning_rate": 2.435897435897436e-06,
+      "loss": 1.5551,
+      "step": 20
+    },
+    {
+      "epoch": 0.01155067860236789,
+      "grad_norm": 2.0367112159729004,
+      "learning_rate": 3.7179487179487184e-06,
+      "loss": 1.0666,
+      "step": 30
+    },
+    {
+      "epoch": 0.015400904803157185,
+      "grad_norm": 0.3497946262359619,
+      "learning_rate": 5e-06,
+      "loss": 0.8899,
+      "step": 40
+    },
+    {
+      "epoch": 0.019251131003946483,
+      "grad_norm": 0.21375155448913574,
+      "learning_rate": 6.282051282051282e-06,
+      "loss": 0.8631,
+      "step": 50
+    },
+    {
+      "epoch": 0.02310135720473578,
+      "grad_norm": 0.15747852623462677,
+      "learning_rate": 7.564102564102564e-06,
+      "loss": 0.8175,
+      "step": 60
+    },
+    {
+      "epoch": 0.026951583405525073,
+      "grad_norm": 0.12693524360656738,
+      "learning_rate": 8.846153846153847e-06,
+      "loss": 0.7943,
+      "step": 70
+    },
+    {
+      "epoch": 0.03080180960631437,
+      "grad_norm": 0.1360969990491867,
+      "learning_rate": 9.999996111488723e-06,
+      "loss": 0.804,
+      "step": 80
+    },
+    {
+      "epoch": 0.034652035807103666,
+      "grad_norm": 0.143288716673851,
+      "learning_rate": 9.999529497453782e-06,
+      "loss": 0.7947,
+      "step": 90
+    },
+    {
+      "epoch": 0.038502262007892966,
+      "grad_norm": 0.12381410598754883,
+      "learning_rate": 9.998285264324193e-06,
+      "loss": 0.7618,
+      "step": 100
+    },
+    {
+      "epoch": 0.04235248820868226,
+      "grad_norm": 0.11233817040920258,
+      "learning_rate": 9.996263605626052e-06,
+      "loss": 0.756,
+      "step": 110
+    },
+    {
+      "epoch": 0.04620271440947156,
+      "grad_norm": 0.12002866715192795,
+      "learning_rate": 9.99346483580503e-06,
+      "loss": 0.7672,
+      "step": 120
+    },
+    {
+      "epoch": 0.05005294061026085,
+      "grad_norm": 0.11726276576519012,
+      "learning_rate": 9.989889390177467e-06,
+      "loss": 0.751,
+      "step": 130
+    },
+    {
+      "epoch": 0.053903166811050146,
+      "grad_norm": 0.12554366886615753,
+      "learning_rate": 9.985537824862647e-06,
+      "loss": 0.7595,
+      "step": 140
+    },
+    {
+      "epoch": 0.057753393011839446,
+      "grad_norm": 0.1385081559419632,
+      "learning_rate": 9.980410816696317e-06,
+      "loss": 0.7465,
+      "step": 150
+    },
+    {
+      "epoch": 0.06160361921262874,
+      "grad_norm": 0.12145545333623886,
+      "learning_rate": 9.97450916312541e-06,
+      "loss": 0.7482,
+      "step": 160
+    },
+    {
+      "epoch": 0.06545384541341803,
+      "grad_norm": 0.10640033334493637,
+      "learning_rate": 9.967833782084004e-06,
+      "loss": 0.7492,
+      "step": 170
+    },
+    {
+      "epoch": 0.06930407161420733,
+      "grad_norm": 0.12344648689031601,
+      "learning_rate": 9.960385711850554e-06,
+      "loss": 0.7679,
+      "step": 180
+    },
+    {
+      "epoch": 0.07315429781499663,
+      "grad_norm": 0.12483686208724976,
+      "learning_rate": 9.952166110886394e-06,
+      "loss": 0.7704,
+      "step": 190
+    },
+    {
+      "epoch": 0.07700452401578593,
+      "grad_norm": 0.128290593624115,
+      "learning_rate": 9.943176257655567e-06,
+      "loss": 0.7502,
+      "step": 200
+    },
+    {
+      "epoch": 0.08085475021657522,
+      "grad_norm": 0.116289421916008,
+      "learning_rate": 9.933417550425952e-06,
+      "loss": 0.7587,
+      "step": 210
+    },
+    {
+      "epoch": 0.08470497641736452,
+      "grad_norm": 0.1165238693356514,
+      "learning_rate": 9.9228915070518e-06,
+      "loss": 0.7328,
+      "step": 220
+    },
+    {
+      "epoch": 0.08855520261815382,
+      "grad_norm": 0.12276219576597214,
+      "learning_rate": 9.911599764737632e-06,
+      "loss": 0.7393,
+      "step": 230
+    },
+    {
+      "epoch": 0.09240542881894312,
+      "grad_norm": 0.1238882765173912,
+      "learning_rate": 9.899544079783606e-06,
+      "loss": 0.725,
+      "step": 240
+    },
+    {
+      "epoch": 0.0962556550197324,
+      "grad_norm": 0.11844491213560104,
+      "learning_rate": 9.886726327312338e-06,
+      "loss": 0.7443,
+      "step": 250
+    },
+    {
+      "epoch": 0.1001058812205217,
+      "grad_norm": 0.11988139897584915,
+      "learning_rate": 9.87314850097724e-06,
+      "loss": 0.7479,
+      "step": 260
+    },
+    {
+      "epoch": 0.103956107421311,
+      "grad_norm": 0.11356190592050552,
+      "learning_rate": 9.858812712652452e-06,
+      "loss": 0.7329,
+      "step": 270
+    },
+    {
+      "epoch": 0.10780633362210029,
+      "grad_norm": 0.11202036589384079,
+      "learning_rate": 9.843721192104341e-06,
+      "loss": 0.7349,
+      "step": 280
+    },
+    {
+      "epoch": 0.11165655982288959,
+      "grad_norm": 0.11301938444375992,
+      "learning_rate": 9.827876286644698e-06,
+      "loss": 0.7531,
+      "step": 290
+    },
+    {
+      "epoch": 0.11550678602367889,
+      "grad_norm": 0.11288069933652878,
+      "learning_rate": 9.811280460765643e-06,
+      "loss": 0.7146,
+      "step": 300
+    },
+    {
+      "epoch": 0.11935701222446819,
+      "grad_norm": 0.11385844647884369,
+      "learning_rate": 9.793936295756292e-06,
+      "loss": 0.7254,
+      "step": 310
+    },
+    {
+      "epoch": 0.12320723842525748,
+      "grad_norm": 0.11850383132696152,
+      "learning_rate": 9.775846489301279e-06,
+      "loss": 0.732,
+      "step": 320
+    },
+    {
+      "epoch": 0.12705746462604678,
+      "grad_norm": 0.11828074604272842,
+      "learning_rate": 9.757013855061157e-06,
+      "loss": 0.731,
+      "step": 330
+    },
+    {
+      "epoch": 0.13090769082683607,
+      "grad_norm": 0.12360090017318726,
+      "learning_rate": 9.737441322234758e-06,
+      "loss": 0.7383,
+      "step": 340
+    },
+    {
+      "epoch": 0.13475791702762538,
+      "grad_norm": 0.12391717731952667,
+      "learning_rate": 9.717131935103605e-06,
+      "loss": 0.7442,
+      "step": 350
+    },
+    {
+      "epoch": 0.13860814322841467,
+      "grad_norm": 0.11035119742155075,
+      "learning_rate": 9.696088852558398e-06,
+      "loss": 0.7277,
+      "step": 360
+    },
+    {
+      "epoch": 0.14245836942920398,
+      "grad_norm": 0.11479122191667557,
+      "learning_rate": 9.674315347607686e-06,
+      "loss": 0.7516,
+      "step": 370
+    },
+    {
+      "epoch": 0.14630859562999327,
+      "grad_norm": 0.1141548603773117,
+      "learning_rate": 9.651814806868796e-06,
+      "loss": 0.7266,
+      "step": 380
+    },
+    {
+      "epoch": 0.15015882183078255,
+      "grad_norm": 0.11828261613845825,
+      "learning_rate": 9.62859073004107e-06,
+      "loss": 0.7251,
+      "step": 390
+    },
+    {
+      "epoch": 0.15400904803157187,
+      "grad_norm": 0.10258225351572037,
+      "learning_rate": 9.604646729361547e-06,
+      "loss": 0.7272,
+      "step": 400
+    },
+    {
+      "epoch": 0.15785927423236115,
+      "grad_norm": 0.11326754093170166,
+      "learning_rate": 9.579986529043099e-06,
+      "loss": 0.7102,
+      "step": 410
+    },
+    {
+      "epoch": 0.16170950043315044,
+      "grad_norm": 0.12070559710264206,
+      "learning_rate": 9.554613964695189e-06,
+      "loss": 0.7279,
+      "step": 420
+    },
+    {
+      "epoch": 0.16555972663393975,
+      "grad_norm": 0.12030478566884995,
+      "learning_rate": 9.528532982727281e-06,
+      "loss": 0.7247,
+      "step": 430
+    },
+    {
+      "epoch": 0.16940995283472904,
+      "grad_norm": 0.11862793564796448,
+      "learning_rate": 9.501747639735018e-06,
+      "loss": 0.7188,
+      "step": 440
+    },
+    {
+      "epoch": 0.17326017903551832,
+      "grad_norm": 0.1128699854016304,
+      "learning_rate": 9.474262101869272e-06,
+      "loss": 0.7072,
+      "step": 450
+    },
+    {
+      "epoch": 0.17711040523630764,
+      "grad_norm": 0.11644523590803146,
+      "learning_rate": 9.446080644188141e-06,
+      "loss": 0.7231,
+      "step": 460
+    },
+    {
+      "epoch": 0.18096063143709692,
+      "grad_norm": 0.11756866425275803,
+      "learning_rate": 9.417207649992017e-06,
+      "loss": 0.7233,
+      "step": 470
+    },
+    {
+      "epoch": 0.18481085763788624,
+      "grad_norm": 0.11568261682987213,
+      "learning_rate": 9.387647610141816e-06,
+      "loss": 0.7158,
+      "step": 480
+    },
+    {
+      "epoch": 0.18866108383867553,
+      "grad_norm": 0.10407742112874985,
+      "learning_rate": 9.35740512236047e-06,
+      "loss": 0.7101,
+      "step": 490
+    },
+    {
+      "epoch": 0.1925113100394648,
+      "grad_norm": 0.10781878232955933,
+      "learning_rate": 9.326484890517809e-06,
+      "loss": 0.72,
+      "step": 500
+    },
+    {
+      "epoch": 0.19636153624025413,
+      "grad_norm": 0.11831319332122803,
+      "learning_rate": 9.294891723898926e-06,
+      "loss": 0.7001,
+      "step": 510
+    },
+    {
+      "epoch": 0.2002117624410434,
+      "grad_norm": 0.12265226989984512,
+      "learning_rate": 9.262630536456153e-06,
+      "loss": 0.7382,
+      "step": 520
+    },
+    {
+      "epoch": 0.2040619886418327,
+      "grad_norm": 0.12133713066577911,
+      "learning_rate": 9.229706346044749e-06,
+      "loss": 0.7112,
+      "step": 530
+    },
+    {
+      "epoch": 0.207912214842622,
+      "grad_norm": 0.12656784057617188,
+      "learning_rate": 9.196124273642422e-06,
+      "loss": 0.7265,
+      "step": 540
+    },
+    {
+      "epoch": 0.2117624410434113,
+      "grad_norm": 0.11293134838342667,
+      "learning_rate": 9.161889542552842e-06,
+      "loss": 0.7358,
+      "step": 550
+    },
+    {
+      "epoch": 0.21561266724420058,
+      "grad_norm": 0.11008590459823608,
+      "learning_rate": 9.127007477593193e-06,
+      "loss": 0.728,
+      "step": 560
+    },
+    {
+      "epoch": 0.2194628934449899,
+      "grad_norm": 0.12754128873348236,
+      "learning_rate": 9.091483504265974e-06,
+      "loss": 0.7383,
+      "step": 570
+    },
+    {
+      "epoch": 0.22331311964577918,
+      "grad_norm": 0.115738645195961,
+      "learning_rate": 9.055323147915116e-06,
+      "loss": 0.7129,
+      "step": 580
+    },
+    {
+      "epoch": 0.2271633458465685,
+      "grad_norm": 0.113907091319561,
+      "learning_rate": 9.018532032866585e-06,
+      "loss": 0.7282,
+      "step": 590
+    },
+    {
+      "epoch": 0.23101357204735778,
+      "grad_norm": 0.11580183357000351,
+      "learning_rate": 8.981115881553579e-06,
+      "loss": 0.7273,
+      "step": 600
+    },
+    {
+      "epoch": 0.23486379824814707,
+      "grad_norm": 0.11657311767339706,
+      "learning_rate": 8.943080513626476e-06,
+      "loss": 0.7385,
+      "step": 610
+    },
+    {
+      "epoch": 0.23871402444893638,
+      "grad_norm": 0.12347078323364258,
+      "learning_rate": 8.904431845047641e-06,
+      "loss": 0.7301,
+      "step": 620
+    },
+    {
+      "epoch": 0.24256425064972567,
+      "grad_norm": 0.1243547573685646,
+      "learning_rate": 8.86517588717128e-06,
+      "loss": 0.7349,
+      "step": 630
+    },
+    {
+      "epoch": 0.24641447685051496,
+      "grad_norm": 0.10942399501800537,
+      "learning_rate": 8.82531874580844e-06,
+      "loss": 0.7358,
+      "step": 640
+    },
+    {
+      "epoch": 0.25026470305130427,
+      "grad_norm": 0.10495594888925552,
+      "learning_rate": 8.784866620277314e-06,
+      "loss": 0.6926,
+      "step": 650
+    },
+    {
+      "epoch": 0.25411492925209356,
+      "grad_norm": 0.12046761065721512,
+      "learning_rate": 8.743825802439018e-06,
+      "loss": 0.7077,
+      "step": 660
+    },
+    {
+      "epoch": 0.25796515545288284,
+      "grad_norm": 0.12549445033073425,
+      "learning_rate": 8.702202675718956e-06,
+      "loss": 0.7325,
+      "step": 670
+    },
+    {
+      "epoch": 0.26181538165367213,
+      "grad_norm": 0.11115598678588867,
+      "learning_rate": 8.660003714113959e-06,
+      "loss": 0.7051,
+      "step": 680
+    },
+    {
+      "epoch": 0.26566560785446147,
+      "grad_norm": 0.11337855458259583,
+      "learning_rate": 8.617235481185323e-06,
+      "loss": 0.7153,
+      "step": 690
+    },
+    {
+      "epoch": 0.26951583405525076,
+      "grad_norm": 0.12993240356445312,
+      "learning_rate": 8.573904629037931e-06,
+      "loss": 0.7228,
+      "step": 700
+    },
+    {
+      "epoch": 0.27336606025604004,
+      "grad_norm": 0.11236017197370529,
+      "learning_rate": 8.530017897285585e-06,
+      "loss": 0.7088,
+      "step": 710
+    },
+    {
+      "epoch": 0.27721628645682933,
+      "grad_norm": 0.10832232981920242,
+      "learning_rate": 8.485582112002746e-06,
+      "loss": 0.7193,
+      "step": 720
+    },
+    {
+      "epoch": 0.2810665126576186,
+      "grad_norm": 0.11175823956727982,
+      "learning_rate": 8.440604184662811e-06,
+      "loss": 0.719,
+      "step": 730
+    },
+    {
+      "epoch": 0.28491673885840796,
+      "grad_norm": 0.11585818976163864,
+      "learning_rate": 8.395091111063118e-06,
+      "loss": 0.7066,
+      "step": 740
+    },
+    {
+      "epoch": 0.28876696505919724,
+      "grad_norm": 0.1034124419093132,
+      "learning_rate": 8.349049970236822e-06,
+      "loss": 0.7078,
+      "step": 750
+    },
+    {
+      "epoch": 0.29261719125998653,
+      "grad_norm": 0.10868314653635025,
+      "learning_rate": 8.302487923351842e-06,
+      "loss": 0.6963,
+      "step": 760
+    },
+    {
+      "epoch": 0.2964674174607758,
+      "grad_norm": 0.11769723147153854,
+      "learning_rate": 8.255412212597021e-06,
+      "loss": 0.7179,
+      "step": 770
+    },
+    {
+      "epoch": 0.3003176436615651,
+      "grad_norm": 0.12348273396492004,
+      "learning_rate": 8.207830160055688e-06,
+      "loss": 0.713,
+      "step": 780
+    },
+    {
+      "epoch": 0.3041678698623544,
+      "grad_norm": 0.11286687105894089,
+      "learning_rate": 8.159749166566786e-06,
+      "loss": 0.6998,
+      "step": 790
+    },
+    {
+      "epoch": 0.30801809606314373,
+      "grad_norm": 0.11574821919202805,
+      "learning_rate": 8.11117671057377e-06,
+      "loss": 0.7238,
+      "step": 800
+    },
+    {
+      "epoch": 0.311868322263933,
+      "grad_norm": 0.11833364516496658,
+      "learning_rate": 8.062120346961404e-06,
+      "loss": 0.709,
+      "step": 810
+    },
+    {
+      "epoch": 0.3157185484647223,
+      "grad_norm": 0.10440419614315033,
+      "learning_rate": 8.012587705880705e-06,
+      "loss": 0.7155,
+      "step": 820
+    },
+    {
+      "epoch": 0.3195687746655116,
+      "grad_norm": 0.11665399372577667,
+      "learning_rate": 7.962586491562146e-06,
+      "loss": 0.7035,
+      "step": 830
+    },
+    {
+      "epoch": 0.3234190008663009,
+      "grad_norm": 0.1175532191991806,
+      "learning_rate": 7.912124481117357e-06,
+      "loss": 0.7042,
+      "step": 840
+    },
+    {
+      "epoch": 0.3272692270670902,
+      "grad_norm": 0.10723235458135605,
+      "learning_rate": 7.861209523329494e-06,
+      "loss": 0.7122,
+      "step": 850
+    },
+    {
+      "epoch": 0.3311194532678795,
+      "grad_norm": 0.107881560921669,
+      "learning_rate": 7.809849537432432e-06,
+      "loss": 0.7022,
+      "step": 860
+    },
+    {
+      "epoch": 0.3349696794686688,
+      "grad_norm": 0.1054963618516922,
+      "learning_rate": 7.758052511879035e-06,
+      "loss": 0.7004,
+      "step": 870
+    },
+    {
+      "epoch": 0.3388199056694581,
+      "grad_norm": 0.10923632979393005,
+      "learning_rate": 7.705826503098642e-06,
+      "loss": 0.7064,
+      "step": 880
+    },
+    {
+      "epoch": 0.34267013187024736,
+      "grad_norm": 0.11138350516557693,
+      "learning_rate": 7.653179634243979e-06,
+      "loss": 0.7012,
+      "step": 890
+    },
+    {
+      "epoch": 0.34652035807103665,
+      "grad_norm": 0.09999624639749527,
+      "learning_rate": 7.600120093927696e-06,
+      "loss": 0.695,
+      "step": 900
+    },
+    {
+      "epoch": 0.350370584271826,
+      "grad_norm": 0.12062712758779526,
+      "learning_rate": 7.546656134948729e-06,
+      "loss": 0.6926,
+      "step": 910
+    },
+    {
+      "epoch": 0.3542208104726153,
+      "grad_norm": 0.10820627957582474,
+      "learning_rate": 7.492796073008661e-06,
+      "loss": 0.69,
+      "step": 920
+    },
+    {
+      "epoch": 0.35807103667340456,
+      "grad_norm": 0.11970065534114838,
+      "learning_rate": 7.438548285418326e-06,
+      "loss": 0.6858,
+      "step": 930
+    },
+    {
+      "epoch": 0.36192126287419385,
+      "grad_norm": 0.11925574392080307,
+      "learning_rate": 7.3839212097948e-06,
+      "loss": 0.7213,
+      "step": 940
+    },
+    {
+      "epoch": 0.36577148907498314,
+      "grad_norm": 0.11157175898551941,
+      "learning_rate": 7.328923342749039e-06,
+      "loss": 0.7343,
+      "step": 950
+    },
+    {
+      "epoch": 0.3696217152757725,
+      "grad_norm": 0.1209806278347969,
+      "learning_rate": 7.273563238564319e-06,
+      "loss": 0.7124,
+      "step": 960
+    },
+    {
+      "epoch": 0.37347194147656176,
+      "grad_norm": 0.10802151262760162,
+      "learning_rate": 7.217849507865724e-06,
+      "loss": 0.6886,
+      "step": 970
+    },
+    {
+      "epoch": 0.37732216767735105,
+      "grad_norm": 0.11031200736761093,
+      "learning_rate": 7.161790816280858e-06,
+      "loss": 0.7058,
+      "step": 980
+    },
+    {
+      "epoch": 0.38117239387814034,
+      "grad_norm": 0.11942834407091141,
+      "learning_rate": 7.1053958830920125e-06,
+      "loss": 0.6774,
+      "step": 990
+    },
+    {
+      "epoch": 0.3850226200789296,
+      "grad_norm": 0.10959523916244507,
+      "learning_rate": 7.0486734798799715e-06,
+      "loss": 0.7091,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3888728462797189,
+      "grad_norm": 0.11255992949008942,
+      "learning_rate": 6.991632429159708e-06,
+      "loss": 0.73,
+      "step": 1010
+    },
+    {
+      "epoch": 0.39272307248050825,
+      "grad_norm": 0.11681436747312546,
+      "learning_rate": 6.934281603008132e-06,
+      "loss": 0.7049,
+      "step": 1020
+    },
+    {
+      "epoch": 0.39657329868129754,
+      "grad_norm": 0.1113327294588089,
+      "learning_rate": 6.876629921684143e-06,
+      "loss": 0.7176,
+      "step": 1030
+    },
+    {
+      "epoch": 0.4004235248820868,
+      "grad_norm": 0.1152559146285057,
+      "learning_rate": 6.818686352241199e-06,
+      "loss": 0.7141,
+      "step": 1040
+    },
+    {
+      "epoch": 0.4042737510828761,
+      "grad_norm": 0.10966623574495316,
+      "learning_rate": 6.760459907132573e-06,
+      "loss": 0.7027,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4081239772836654,
+      "grad_norm": 0.1089562252163887,
+      "learning_rate": 6.701959642809589e-06,
+      "loss": 0.7063,
+      "step": 1060
+    },
+    {
+      "epoch": 0.41197420348445474,
+      "grad_norm": 0.11264059692621231,
+      "learning_rate": 6.643194658312987e-06,
+      "loss": 0.6882,
+      "step": 1070
+    },
+    {
+      "epoch": 0.415824429685244,
+      "grad_norm": 0.1071435809135437,
+      "learning_rate": 6.584174093857676e-06,
+      "loss": 0.7259,
+      "step": 1080
+    },
+    {
+      "epoch": 0.4196746558860333,
+      "grad_norm": 0.1257365494966507,
+      "learning_rate": 6.5249071294110735e-06,
+      "loss": 0.6995,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4235248820868226,
+      "grad_norm": 0.11635745316743851,
+      "learning_rate": 6.465402983265274e-06,
+      "loss": 0.7186,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4273751082876119,
+      "grad_norm": 0.11439164727926254,
+      "learning_rate": 6.405670910603243e-06,
+      "loss": 0.7158,
+      "step": 1110
+    },
+    {
+      "epoch": 0.43122533448840117,
+      "grad_norm": 0.12154637277126312,
+      "learning_rate": 6.345720202059291e-06,
+      "loss": 0.7062,
+      "step": 1120
+    },
+    {
+      "epoch": 0.4350755606891905,
+      "grad_norm": 0.1102345734834671,
+      "learning_rate": 6.285560182274008e-06,
+      "loss": 0.6944,
+      "step": 1130
+    },
+    {
+      "epoch": 0.4389257868899798,
+      "grad_norm": 0.1106484904885292,
+      "learning_rate": 6.225200208443936e-06,
+      "loss": 0.7026,
+      "step": 1140
+    },
+    {
+      "epoch": 0.4427760130907691,
+      "grad_norm": 0.10597451031208038,
+      "learning_rate": 6.16464966886616e-06,
+      "loss": 0.6967,
+      "step": 1150
+    },
+    {
+      "epoch": 0.44662623929155837,
+      "grad_norm": 0.12949316203594208,
+      "learning_rate": 6.103917981478065e-06,
+      "loss": 0.7258,
+      "step": 1160
+    },
+    {
+      "epoch": 0.45047646549234766,
+      "grad_norm": 0.10967691987752914,
+      "learning_rate": 6.043014592392489e-06,
+      "loss": 0.7317,
+      "step": 1170
+    },
+    {
+      "epoch": 0.454326691693137,
+      "grad_norm": 0.10290511697530746,
+      "learning_rate": 5.981948974428485e-06,
+      "loss": 0.7052,
+      "step": 1180
+    },
+    {
+      "epoch": 0.4581769178939263,
+      "grad_norm": 0.11438062787055969,
+      "learning_rate": 5.920730625637934e-06,
+      "loss": 0.6844,
+      "step": 1190
+    },
+    {
+      "epoch": 0.46202714409471557,
+      "grad_norm": 0.11923769116401672,
+      "learning_rate": 5.8593690678282355e-06,
+      "loss": 0.6977,
+      "step": 1200
+    },
+    {
+      "epoch": 0.46587737029550486,
+      "grad_norm": 0.11448656022548676,
+      "learning_rate": 5.797873845081301e-06,
+      "loss": 0.7,
+      "step": 1210
+    },
+    {
+      "epoch": 0.46972759649629414,
+      "grad_norm": 0.11611119657754898,
+      "learning_rate": 5.736254522269075e-06,
+      "loss": 0.7126,
+      "step": 1220
+    },
+    {
+      "epoch": 0.47357782269708343,
+      "grad_norm": 0.10960757732391357,
+      "learning_rate": 5.6745206835658386e-06,
+      "loss": 0.7194,
+      "step": 1230
+    },
+    {
+      "epoch": 0.47742804889787277,
+      "grad_norm": 0.10850224643945694,
+      "learning_rate": 5.612681930957495e-06,
+      "loss": 0.6952,
+      "step": 1240
+    },
+    {
+      "epoch": 0.48127827509866206,
+      "grad_norm": 0.10815806686878204,
+      "learning_rate": 5.550747882748095e-06,
+      "loss": 0.6894,
+      "step": 1250
+    },
+    {
+      "epoch": 0.48512850129945134,
+      "grad_norm": 0.11456377059221268,
+      "learning_rate": 5.4887281720638195e-06,
+      "loss": 0.6985,
+      "step": 1260
+    },
+    {
+      "epoch": 0.48897872750024063,
+      "grad_norm": 0.1271165907382965,
+      "learning_rate": 5.42663244535466e-06,
+      "loss": 0.6827,
+      "step": 1270
+    },
+    {
+      "epoch": 0.4928289537010299,
+      "grad_norm": 0.10642170161008835,
+      "learning_rate": 5.3644703608940205e-06,
+      "loss": 0.6902,
+      "step": 1280
+    },
+    {
+      "epoch": 0.49667917990181926,
+      "grad_norm": 0.11253950744867325,
+      "learning_rate": 5.302251587276491e-06,
+      "loss": 0.7089,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5005294061026085,
+      "grad_norm": 0.11363402754068375,
+      "learning_rate": 5.2399858019140005e-06,
+      "loss": 0.7045,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5043796323033978,
+      "grad_norm": 0.11016932874917984,
+      "learning_rate": 5.177682689530615e-06,
+      "loss": 0.7198,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5082298585041871,
+      "grad_norm": 0.12260254472494125,
+      "learning_rate": 5.11535194065619e-06,
+      "loss": 0.6915,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5120800847049765,
+      "grad_norm": 0.12205204367637634,
+      "learning_rate": 5.053003250119113e-06,
+      "loss": 0.6928,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5159303109057657,
+      "grad_norm": 0.11264455318450928,
+      "learning_rate": 4.9906463155383925e-06,
+      "loss": 0.7033,
+      "step": 1340
+    },
+    {
+      "epoch": 0.519780537106555,
+      "grad_norm": 0.11192315816879272,
+      "learning_rate": 4.9282908358153046e-06,
+      "loss": 0.6984,
+      "step": 1350
+    },
+    {
+      "epoch": 0.5236307633073443,
+      "grad_norm": 0.13072162866592407,
+      "learning_rate": 4.8659465096248345e-06,
+      "loss": 0.7034,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5274809895081336,
+      "grad_norm": 0.1100064143538475,
+      "learning_rate": 4.803623033907169e-06,
+      "loss": 0.7055,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5313312157089229,
+      "grad_norm": 0.10766682028770447,
+      "learning_rate": 4.741330102359441e-06,
+      "loss": 0.7171,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5351814419097122,
+      "grad_norm": 0.12846867740154266,
+      "learning_rate": 4.679077403927996e-06,
+      "loss": 0.6894,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5390316681105015,
+      "grad_norm": 0.11385906487703323,
+      "learning_rate": 4.616874621301373e-06,
+      "loss": 0.6985,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5428818943112907,
+      "grad_norm": 0.11475247144699097,
+      "learning_rate": 4.554731429404293e-06,
+      "loss": 0.7145,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5467321205120801,
+      "grad_norm": 0.1149231567978859,
+      "learning_rate": 4.492657493892822e-06,
+      "loss": 0.6939,
+      "step": 1420
+    },
+    {
+      "epoch": 0.5505823467128694,
+      "grad_norm": 0.11526394635438919,
+      "learning_rate": 4.430662469650999e-06,
+      "loss": 0.7089,
+      "step": 1430
+    },
+    {
+      "epoch": 0.5544325729136587,
+      "grad_norm": 0.1106191873550415,
+      "learning_rate": 4.368755999289118e-06,
+      "loss": 0.7129,
+      "step": 1440
+    },
+    {
+      "epoch": 0.558282799114448,
+      "grad_norm": 0.12260071933269501,
+      "learning_rate": 4.306947711643949e-06,
+      "loss": 0.6902,
+      "step": 1450
+    },
+    {
+      "epoch": 0.5621330253152372,
+      "grad_norm": 0.11018304526805878,
+      "learning_rate": 4.245247220281068e-06,
+      "loss": 0.7002,
+      "step": 1460
+    },
+    {
+      "epoch": 0.5659832515160266,
+      "grad_norm": 0.10749536007642746,
+      "learning_rate": 4.183664121999587e-06,
+      "loss": 0.6935,
+      "step": 1470
+    },
+    {
+      "epoch": 0.5698334777168159,
+      "grad_norm": 0.11514180153608322,
+      "learning_rate": 4.122207995339474e-06,
+      "loss": 0.6905,
+      "step": 1480
+    },
+    {
+      "epoch": 0.5736837039176051,
+      "grad_norm": 0.11017462611198425,
+      "learning_rate": 4.060888399091734e-06,
+      "loss": 0.7002,
+      "step": 1490
+    },
+    {
+      "epoch": 0.5775339301183945,
+      "grad_norm": 0.11048802733421326,
+      "learning_rate": 3.999714870811639e-06,
+      "loss": 0.7144,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5813841563191837,
+      "grad_norm": 0.10919424891471863,
+      "learning_rate": 3.938696925335275e-06,
+      "loss": 0.7028,
+      "step": 1510
+    },
+    {
+      "epoch": 0.5852343825199731,
+      "grad_norm": 0.11765211075544357,
+      "learning_rate": 3.87784405329962e-06,
+      "loss": 0.7087,
+      "step": 1520
+    },
+    {
+      "epoch": 0.5890846087207623,
+      "grad_norm": 0.11160047352313995,
+      "learning_rate": 3.817165719666382e-06,
+      "loss": 0.704,
+      "step": 1530
+    },
+    {
+      "epoch": 0.5929348349215516,
+      "grad_norm": 0.11005199700593948,
+      "learning_rate": 3.756671362249841e-06,
+      "loss": 0.7183,
+      "step": 1540
+    },
+    {
+      "epoch": 0.596785061122341,
+      "grad_norm": 0.11469442397356033,
+      "learning_rate": 3.6963703902488907e-06,
+      "loss": 0.7077,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6006352873231302,
+      "grad_norm": 0.12150226533412933,
+      "learning_rate": 3.6362721827835655e-06,
+      "loss": 0.7141,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6044855135239195,
+      "grad_norm": 0.11834833770990372,
+      "learning_rate": 3.576386087436212e-06,
+      "loss": 0.7074,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6083357397247088,
+      "grad_norm": 0.11631608009338379,
+      "learning_rate": 3.5167214187975913e-06,
+      "loss": 0.7113,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6121859659254981,
+      "grad_norm": 0.10068583488464355,
+      "learning_rate": 3.457287457018086e-06,
+      "loss": 0.7062,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6160361921262875,
+      "grad_norm": 0.1051119938492775,
+      "learning_rate": 3.3980934463643003e-06,
+      "loss": 0.6944,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6198864183270767,
+      "grad_norm": 0.10199523717164993,
+      "learning_rate": 3.3391485937812053e-06,
+      "loss": 0.6717,
+      "step": 1610
+    },
+    {
+      "epoch": 0.623736644527866,
+      "grad_norm": 0.11558383703231812,
+      "learning_rate": 3.280462067460115e-06,
+      "loss": 0.6759,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6275868707286553,
+      "grad_norm": 0.10855776071548462,
+      "learning_rate": 3.222042995412669e-06,
+      "loss": 0.714,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6314370969294446,
+      "grad_norm": 0.09911976009607315,
+      "learning_rate": 3.163900464051089e-06,
+      "loss": 0.6932,
+      "step": 1640
+    },
+    {
+      "epoch": 0.635287323130234,
+      "grad_norm": 0.11288108676671982,
+      "learning_rate": 3.1060435167748826e-06,
+      "loss": 0.7061,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6391375493310232,
+      "grad_norm": 0.11214844882488251,
+      "learning_rate": 3.048481152564253e-06,
+      "loss": 0.6934,
+      "step": 1660
+    },
+    {
+      "epoch": 0.6429877755318125,
+      "grad_norm": 0.11853953450918198,
+      "learning_rate": 2.9912223245804017e-06,
+      "loss": 0.7164,
+      "step": 1670
+    },
+    {
+      "epoch": 0.6468380017326018,
+      "grad_norm": 0.1129305437207222,
+      "learning_rate": 2.934275938772981e-06,
+      "loss": 0.6952,
+      "step": 1680
+    },
+    {
+      "epoch": 0.6506882279333911,
+      "grad_norm": 0.10780379176139832,
+      "learning_rate": 2.8776508524948676e-06,
+      "loss": 0.6846,
+      "step": 1690
+    },
+    {
+      "epoch": 0.6545384541341804,
+      "grad_norm": 0.10118725895881653,
+      "learning_rate": 2.821355873124503e-06,
+      "loss": 0.685,
+      "step": 1700
+    },
+    {
+      "epoch": 0.6583886803349697,
+      "grad_norm": 0.11421024054288864,
+      "learning_rate": 2.7653997566960144e-06,
+      "loss": 0.7037,
+      "step": 1710
+    },
+    {
+      "epoch": 0.662238906535759,
+      "grad_norm": 0.11296897381544113,
+      "learning_rate": 2.709791206537309e-06,
+      "loss": 0.694,
+      "step": 1720
+    },
+    {
+      "epoch": 0.6660891327365482,
+      "grad_norm": 0.11072462797164917,
+      "learning_rate": 2.6545388719163747e-06,
+      "loss": 0.6897,
+      "step": 1730
+    },
+    {
+      "epoch": 0.6699393589373376,
+      "grad_norm": 0.11128582060337067,
+      "learning_rate": 2.599651346695979e-06,
+      "loss": 0.7029,
+      "step": 1740
+    },
+    {
+      "epoch": 0.6737895851381268,
+      "grad_norm": 0.11217836290597916,
+      "learning_rate": 2.545137167997007e-06,
+      "loss": 0.6993,
+      "step": 1750
+    },
+    {
+      "epoch": 0.6776398113389162,
+      "grad_norm": 0.10789728909730911,
+      "learning_rate": 2.4910048148705972e-06,
+      "loss": 0.6843,
+      "step": 1760
+    },
+    {
+      "epoch": 0.6814900375397055,
+      "grad_norm": 0.11674684286117554,
+      "learning_rate": 2.4372627069793327e-06,
+      "loss": 0.7266,
+      "step": 1770
+    },
+    {
+      "epoch": 0.6853402637404947,
+      "grad_norm": 0.1190313771367073,
+      "learning_rate": 2.3839192032876486e-06,
+      "loss": 0.71,
+      "step": 1780
+    },
+    {
+      "epoch": 0.6891904899412841,
+      "grad_norm": 0.1104375571012497,
+      "learning_rate": 2.3309826007617047e-06,
+      "loss": 0.7212,
+      "step": 1790
+    },
+    {
+      "epoch": 0.6930407161420733,
+      "grad_norm": 0.10845618695020676,
+      "learning_rate": 2.278461133078881e-06,
+      "loss": 0.6848,
+      "step": 1800
+    },
+    {
+      "epoch": 0.6968909423428626,
+      "grad_norm": 0.1124584898352623,
+      "learning_rate": 2.2263629693471258e-06,
+      "loss": 0.6807,
+      "step": 1810
+    },
+    {
+      "epoch": 0.700741168543652,
+      "grad_norm": 0.10723639279603958,
+      "learning_rate": 2.1746962128343414e-06,
+      "loss": 0.7009,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7045913947444412,
+      "grad_norm": 0.11180757731199265,
+      "learning_rate": 2.1234688997080183e-06,
+      "loss": 0.7012,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7084416209452306,
+      "grad_norm": 0.1281813234090805,
+      "learning_rate": 2.0726889977853053e-06,
+      "loss": 0.6988,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7122918471460198,
+      "grad_norm": 0.12354621291160583,
+      "learning_rate": 2.022364405293703e-06,
+      "loss": 0.7105,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7161420733468091,
+      "grad_norm": 0.11235840618610382,
+      "learning_rate": 1.9725029496425814e-06,
+      "loss": 0.7065,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7199922995475985,
+      "grad_norm": 0.10087255388498306,
+      "learning_rate": 1.923112386205727e-06,
+      "loss": 0.7007,
+      "step": 1870
+    },
+    {
+      "epoch": 0.7238425257483877,
+      "grad_norm": 0.1126994714140892,
+      "learning_rate": 1.874200397115079e-06,
+      "loss": 0.709,
+      "step": 1880
+    },
+    {
+      "epoch": 0.727692751949177,
+      "grad_norm": 0.1166677251458168,
+      "learning_rate": 1.825774590065857e-06,
+      "loss": 0.693,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7315429781499663,
+      "grad_norm": 0.1166413426399231,
+      "learning_rate": 1.7778424971332813e-06,
+      "loss": 0.7026,
+      "step": 1900
+    },
+    {
+      "epoch": 0.7353932043507556,
+      "grad_norm": 0.10672884434461594,
+      "learning_rate": 1.7304115736010407e-06,
+      "loss": 0.7043,
+      "step": 1910
+    },
+    {
+      "epoch": 0.739243430551545,
+      "grad_norm": 0.11122455447912216,
+      "learning_rate": 1.6834891968017108e-06,
+      "loss": 0.6978,
+      "step": 1920
+    },
+    {
+      "epoch": 0.7430936567523342,
+      "grad_norm": 0.1161666139960289,
+      "learning_rate": 1.6370826649692884e-06,
+      "loss": 0.7039,
+      "step": 1930
+    },
+    {
+      "epoch": 0.7469438829531235,
+      "grad_norm": 0.1187717393040657,
+      "learning_rate": 1.5911991961040463e-06,
+      "loss": 0.7176,
+      "step": 1940
+    },
+    {
+      "epoch": 0.7507941091539128,
+      "grad_norm": 0.11241371184587479,
+      "learning_rate": 1.545845926849851e-06,
+      "loss": 0.7048,
+      "step": 1950
+    },
+    {
+      "epoch": 0.7546443353547021,
+      "grad_norm": 0.11637844145298004,
+      "learning_rate": 1.5010299113841397e-06,
+      "loss": 0.7197,
+      "step": 1960
+    },
+    {
+      "epoch": 0.7584945615554913,
+      "grad_norm": 0.11043717712163925,
+      "learning_rate": 1.4567581203207238e-06,
+      "loss": 0.7176,
+      "step": 1970
+    },
+    {
+      "epoch": 0.7623447877562807,
+      "grad_norm": 0.11378881335258484,
+      "learning_rate": 1.413037439625593e-06,
+      "loss": 0.7007,
+      "step": 1980
+    },
+    {
+      "epoch": 0.76619501395707,
+      "grad_norm": 0.12594033777713776,
+      "learning_rate": 1.3698746695458847e-06,
+      "loss": 0.702,
+      "step": 1990
+    },
+    {
+      "epoch": 0.7700452401578592,
+      "grad_norm": 0.11072988063097,
+      "learning_rate": 1.3272765235521745e-06,
+      "loss": 0.7035,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7738954663586486,
+      "grad_norm": 0.11150892823934555,
+      "learning_rate": 1.2852496272942877e-06,
+      "loss": 0.7044,
+      "step": 2010
+    },
+    {
+      "epoch": 0.7777456925594378,
+      "grad_norm": 0.10707996040582657,
+      "learning_rate": 1.2438005175707396e-06,
+      "loss": 0.687,
+      "step": 2020
+    },
+    {
+      "epoch": 0.7815959187602272,
+      "grad_norm": 0.11259715259075165,
+      "learning_rate": 1.2029356413120297e-06,
+      "loss": 0.6736,
+      "step": 2030
+    },
+    {
+      "epoch": 0.7854461449610165,
+      "grad_norm": 0.11276605725288391,
+      "learning_rate": 1.1626613545778786e-06,
+      "loss": 0.6909,
+      "step": 2040
+    },
+    {
+      "epoch": 0.7892963711618057,
+      "grad_norm": 0.10938549786806107,
+      "learning_rate": 1.1229839215686294e-06,
+      "loss": 0.6997,
+      "step": 2050
+    },
+    {
+      "epoch": 0.7931465973625951,
+      "grad_norm": 0.10552886873483658,
+      "learning_rate": 1.0839095136509188e-06,
+      "loss": 0.6777,
+      "step": 2060
+    },
+    {
+      "epoch": 0.7969968235633843,
+      "grad_norm": 0.11417779326438904,
+      "learning_rate": 1.045444208397791e-06,
+      "loss": 0.69,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8008470497641736,
+      "grad_norm": 0.10782705247402191,
+      "learning_rate": 1.007593988643402e-06,
+      "loss": 0.681,
+      "step": 2080
+    },
+    {
+      "epoch": 0.804697275964963,
+      "grad_norm": 0.104763463139534,
+      "learning_rate": 9.703647415524613e-07,
+      "loss": 0.7047,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8085475021657522,
+      "grad_norm": 0.11093178391456604,
+      "learning_rate": 9.337622577045485e-07,
+      "loss": 0.6983,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8123977283665416,
+      "grad_norm": 0.1129617989063263,
+      "learning_rate": 8.977922301934588e-07,
+      "loss": 0.7124,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8162479545673308,
+      "grad_norm": 0.11247554421424866,
+      "learning_rate": 8.624602537417009e-07,
+      "loss": 0.7092,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8200981807681201,
+      "grad_norm": 0.1139712780714035,
+      "learning_rate": 8.277718238303123e-07,
+      "loss": 0.6897,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8239484069689095,
+      "grad_norm": 0.1133948341012001,
+      "learning_rate": 7.937323358440935e-07,
+      "loss": 0.703,
+      "step": 2140
+    },
+    {
+      "epoch": 0.8277986331696987,
+      "grad_norm": 0.13035932183265686,
+      "learning_rate": 7.603470842324223e-07,
+      "loss": 0.6846,
+      "step": 2150
+    },
+    {
+      "epoch": 0.831648859370488,
+      "grad_norm": 0.11306580901145935,
+      "learning_rate": 7.276212616857575e-07,
+      "loss": 0.7059,
+      "step": 2160
+    },
+    {
+      "epoch": 0.8354990855712773,
+      "grad_norm": 0.11329696327447891,
+      "learning_rate": 6.955599583279837e-07,
+      "loss": 0.7057,
+      "step": 2170
+    },
+    {
+      "epoch": 0.8393493117720666,
+      "grad_norm": 0.10971033573150635,
+      "learning_rate": 6.641681609246981e-07,
+      "loss": 0.6975,
+      "step": 2180
+    },
+    {
+      "epoch": 0.8431995379728559,
+      "grad_norm": 0.10894191265106201,
+      "learning_rate": 6.334507521075711e-07,
+      "loss": 0.7052,
+      "step": 2190
+    },
+    {
+      "epoch": 0.8470497641736452,
+      "grad_norm": 0.12159916013479233,
+      "learning_rate": 6.034125096149179e-07,
+      "loss": 0.6933,
+      "step": 2200
+    },
+    {
+      "epoch": 0.8508999903744345,
+      "grad_norm": 0.126884326338768,
+      "learning_rate": 5.740581055485722e-07,
+      "loss": 0.6775,
+      "step": 2210
+    },
+    {
+      "epoch": 0.8547502165752238,
+      "grad_norm": 0.10263621062040329,
+      "learning_rate": 5.453921056471961e-07,
+      "loss": 0.6998,
+      "step": 2220
+    },
+    {
+      "epoch": 0.8586004427760131,
+      "grad_norm": 0.11732914298772812,
+      "learning_rate": 5.174189685761288e-07,
+      "loss": 0.6951,
+      "step": 2230
+    },
+    {
+      "epoch": 0.8624506689768023,
+      "grad_norm": 0.10687677562236786,
+      "learning_rate": 4.901430452338979e-07,
+      "loss": 0.713,
+      "step": 2240
+    },
+    {
+      "epoch": 0.8663008951775917,
+      "grad_norm": 0.11015113443136215,
+      "learning_rate": 4.6356857807548286e-07,
+      "loss": 0.6939,
+      "step": 2250
+    },
+    {
+      "epoch": 0.870151121378381,
+      "grad_norm": 0.11084848642349243,
+      "learning_rate": 4.376997004524541e-07,
+      "loss": 0.6827,
+      "step": 2260
+    },
+    {
+      "epoch": 0.8740013475791703,
+      "grad_norm": 0.10526624321937561,
+      "learning_rate": 4.12540435970073e-07,
+      "loss": 0.6828,
+      "step": 2270
+    },
+    {
+      "epoch": 0.8778515737799596,
+      "grad_norm": 0.1116974949836731,
+      "learning_rate": 3.880946978614703e-07,
+      "loss": 0.7068,
+      "step": 2280
+    },
+    {
+      "epoch": 0.8817017999807488,
+      "grad_norm": 0.1070835217833519,
+      "learning_rate": 3.643662883789878e-07,
+      "loss": 0.7019,
+      "step": 2290
+    },
+    {
+      "epoch": 0.8855520261815382,
+      "grad_norm": 0.10582892596721649,
+      "learning_rate": 3.41358898202781e-07,
+      "loss": 0.6873,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8894022523823275,
+      "grad_norm": 0.10296995937824249,
+      "learning_rate": 3.190761058667741e-07,
+      "loss": 0.6904,
+      "step": 2310
+    },
+    {
+      "epoch": 0.8932524785831167,
+      "grad_norm": 0.10488659143447876,
+      "learning_rate": 2.9752137720206463e-07,
+      "loss": 0.6968,
+      "step": 2320
+    },
+    {
+      "epoch": 0.8971027047839061,
+      "grad_norm": 0.11368091404438019,
+      "learning_rate": 2.7669806479785067e-07,
+      "loss": 0.7074,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9009529309846953,
+      "grad_norm": 0.1063510924577713,
+      "learning_rate": 2.566094074799763e-07,
+      "loss": 0.6964,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9048031571854847,
+      "grad_norm": 0.1139088049530983,
+      "learning_rate": 2.3725852980716557e-07,
+      "loss": 0.6872,
+      "step": 2350
+    },
+    {
+      "epoch": 0.908653383386274,
+      "grad_norm": 0.10771569609642029,
+      "learning_rate": 2.1864844158503983e-07,
+      "loss": 0.6978,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9125036095870632,
+      "grad_norm": 0.11983660608530045,
+      "learning_rate": 2.007820373979702e-07,
+      "loss": 0.6877,
+      "step": 2370
+    },
+    {
+      "epoch": 0.9163538357878526,
+      "grad_norm": 0.1092521995306015,
+      "learning_rate": 1.8366209615886532e-07,
+      "loss": 0.6932,
+      "step": 2380
+    },
+    {
+      "epoch": 0.9202040619886418,
+      "grad_norm": 0.10891921073198318,
+      "learning_rate": 1.6729128067693613e-07,
+      "loss": 0.7146,
+      "step": 2390
+    },
+    {
+      "epoch": 0.9240542881894311,
+      "grad_norm": 0.11188556253910065,
+      "learning_rate": 1.5167213724353426e-07,
+      "loss": 0.7035,
+      "step": 2400
+    },
+    {
+      "epoch": 0.9279045143902205,
+      "grad_norm": 0.10971692949533463,
+      "learning_rate": 1.3680709523610058e-07,
+      "loss": 0.6866,
+      "step": 2410
+    },
+    {
+      "epoch": 0.9317547405910097,
+      "grad_norm": 0.11035142093896866,
+      "learning_rate": 1.2269846674030418e-07,
+      "loss": 0.7139,
+      "step": 2420
+    },
+    {
+      "epoch": 0.935604966791799,
+      "grad_norm": 0.1190849021077156,
+      "learning_rate": 1.0934844619042595e-07,
+      "loss": 0.7072,
+      "step": 2430
+    },
+    {
+      "epoch": 0.9394551929925883,
+      "grad_norm": 0.11458474397659302,
+      "learning_rate": 9.675911002803961e-08,
+      "loss": 0.6945,
+      "step": 2440
+    },
+    {
+      "epoch": 0.9433054191933776,
+      "grad_norm": 0.11658325046300888,
+      "learning_rate": 8.493241637904381e-08,
+      "loss": 0.6924,
+      "step": 2450
+    },
+    {
+      "epoch": 0.9471556453941669,
+      "grad_norm": 0.11345871537923813,
+      "learning_rate": 7.387020474909689e-08,
+      "loss": 0.7142,
+      "step": 2460
+    },
+    {
+      "epoch": 0.9510058715949562,
+      "grad_norm": 0.12846674025058746,
+      "learning_rate": 6.357419573750856e-08,
+      "loss": 0.7019,
+      "step": 2470
+    },
+    {
+      "epoch": 0.9548560977957455,
+      "grad_norm": 0.12026581913232803,
+      "learning_rate": 5.4045990769613365e-08,
+      "loss": 0.6927,
+      "step": 2480
+    },
+    {
+      "epoch": 0.9587063239965348,
+      "grad_norm": 0.11361832916736603,
+      "learning_rate": 4.5287071847694434e-08,
+      "loss": 0.693,
+      "step": 2490
+    },
+    {
+      "epoch": 0.9625565501973241,
+      "grad_norm": 0.10997363179922104,
+      "learning_rate": 3.7298801320467817e-08,
+      "loss": 0.6949,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9664067763981133,
+      "grad_norm": 0.11490064859390259,
+      "learning_rate": 3.0082421671192576e-08,
+      "loss": 0.6882,
+      "step": 2510
+    },
+    {
+      "epoch": 0.9702570025989027,
+      "grad_norm": 0.10568021982908249,
+      "learning_rate": 2.3639055324410355e-08,
+      "loss": 0.705,
+      "step": 2520
+    },
+    {
+      "epoch": 0.974107228799692,
+      "grad_norm": 0.11061687022447586,
+      "learning_rate": 1.7969704471369477e-08,
+      "loss": 0.6921,
+      "step": 2530
+    },
+    {
+      "epoch": 0.9779574550004813,
+      "grad_norm": 0.10393280535936356,
+      "learning_rate": 1.3075250914142413e-08,
+      "loss": 0.6968,
+      "step": 2540
+    },
+    {
+      "epoch": 0.9818076812012706,
+      "grad_norm": 0.1011970266699791,
+      "learning_rate": 8.956455928476049e-09,
+      "loss": 0.7071,
+      "step": 2550
+    },
+    {
+      "epoch": 0.9856579074020598,
+      "grad_norm": 0.09775857627391815,
+      "learning_rate": 5.6139601453786275e-09,
+      "loss": 0.6861,
+      "step": 2560
+    },
+    {
+      "epoch": 0.9895081336028492,
+      "grad_norm": 0.11433733254671097,
+      "learning_rate": 3.0482834514805602e-09,
+      "loss": 0.6915,
+      "step": 2570
+    },
+    {
+      "epoch": 0.9933583598036385,
+      "grad_norm": 0.11014392971992493,
+      "learning_rate": 1.2598249081713366e-09,
+      "loss": 0.673,
+      "step": 2580
+    },
+    {
+      "epoch": 0.9972085860044277,
+      "grad_norm": 0.10732517391443253,
+      "learning_rate": 2.488626895302826e-10,
+      "loss": 0.7078,
+      "step": 2590
+    },
+    {
+      "epoch": 0.9999037443449803,
+      "step": 2597,
+      "total_flos": 3.473337592332681e+18,
+      "train_loss": 0.7211110119090707,
+      "train_runtime": 10590.5593,
+      "train_samples_per_second": 62.783,
+      "train_steps_per_second": 0.245
+    }
+  ],
+  "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": 3.473337592332681e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen25/siglip2_sm16_384/finetune/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e12785fd3f060bae8a9d8f3bf5e8ccf353d5b30956e90d75019c7234caaa8263
+size 6136

continuous/qwen25/siglip2_sm16_384/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": 1152,
+  "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-SO400M-16-SigLIP2-384",
+  "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/siglip2/siglip2_sm16_384.npz",
+  "vocab_size": 151936
+}

continuous/qwen25/siglip2_sm16_384/pretrain/mm_projector.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5590b614c76c78646f6a142c2526a9b51f2d1491fdd0bf76af004f714507405a
+size 16529464

continuous/qwen25/siglip2_sm16_384/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": 3.576890707015991,
+      "learning_rate": 0.00032142857142857147,
+      "loss": 4.5957,
+      "step": 10
+    },
+    {
+      "epoch": 0.03669724770642202,
+      "grad_norm": 0.17926253378391266,
+      "learning_rate": 0.0006785714285714287,
+      "loss": 3.545,
+      "step": 20
+    },
+    {
+      "epoch": 0.05504587155963303,
+      "grad_norm": 0.10073135793209076,
+      "learning_rate": 0.0009999907688176173,
+      "loss": 3.288,
+      "step": 30
+    },
+    {
+      "epoch": 0.07339449541284404,
+      "grad_norm": 0.06231551989912987,
+      "learning_rate": 0.0009988834393115766,
+      "loss": 3.2253,
+      "step": 40
+    },
+    {
+      "epoch": 0.09174311926605505,
+      "grad_norm": 0.06278876960277557,
+      "learning_rate": 0.0009959345572940772,
+      "loss": 3.2035,
+      "step": 50
+    },
+    {
+      "epoch": 0.11009174311926606,
+      "grad_norm": 0.0781743973493576,
+      "learning_rate": 0.0009911550081155983,
+      "loss": 3.1654,
+      "step": 60
+    },
+    {
+      "epoch": 0.12844036697247707,
+      "grad_norm": 0.09320346266031265,
+      "learning_rate": 0.0009845624347566433,
+      "loss": 3.0798,
+      "step": 70
+    },
+    {
+      "epoch": 0.14678899082568808,
+      "grad_norm": 0.09932755678892136,
+      "learning_rate": 0.0009761811727013547,
+      "loss": 3.0039,
+      "step": 80
+    },
+    {
+      "epoch": 0.1651376146788991,
+      "grad_norm": 0.09712540358304977,
+      "learning_rate": 0.0009660421601067666,
+      "loss": 2.9454,
+      "step": 90
+    },
+    {
+      "epoch": 0.1834862385321101,
+      "grad_norm": 0.07971840351819992,
+      "learning_rate": 0.0009541828235992926,
+      "loss": 2.9093,
+      "step": 100
+    },
+    {
+      "epoch": 0.2018348623853211,
+      "grad_norm": 0.09526852518320084,
+      "learning_rate": 0.0009406469401200151,
+      "loss": 2.8654,
+      "step": 110
+    },
+    {
+      "epoch": 0.22018348623853212,
+      "grad_norm": 0.07504100352525711,
+      "learning_rate": 0.0009254844753287492,
+      "loss": 2.8419,
+      "step": 120
+    },
+    {
+      "epoch": 0.23853211009174313,
+      "grad_norm": 0.07606084644794464,
+      "learning_rate": 0.0009087513991633924,
+      "loss": 2.8036,
+      "step": 130
+    },
+    {
+      "epoch": 0.25688073394495414,
+      "grad_norm": 0.09968791157007217,
+      "learning_rate": 0.0008905094792353917,
+      "loss": 2.7823,
+      "step": 140
+    },
+    {
+      "epoch": 0.27522935779816515,
+      "grad_norm": 0.09380267560482025,
+      "learning_rate": 0.0008708260528239789,
+      "loss": 2.7705,
+      "step": 150
+    },
+    {
+      "epoch": 0.29357798165137616,
+      "grad_norm": 0.09614922851324081,
+      "learning_rate": 0.0008497737783108237,
+      "loss": 2.7407,
+      "step": 160
+    },
+    {
+      "epoch": 0.3119266055045872,
+      "grad_norm": 0.09775327891111374,
+      "learning_rate": 0.0008274303669726426,
+      "loss": 2.7297,
+      "step": 170
+    },
+    {
+      "epoch": 0.3302752293577982,
+      "grad_norm": 0.09403182566165924,
+      "learning_rate": 0.0008038782961218137,
+      "loss": 2.7106,
+      "step": 180
+    },
+    {
+      "epoch": 0.3486238532110092,
+      "grad_norm": 0.08660382777452469,
+      "learning_rate": 0.000779204504653894,
+      "loss": 2.6955,
+      "step": 190
+    },
+    {
+      "epoch": 0.3669724770642202,
+      "grad_norm": 0.09103938192129135,
+      "learning_rate": 0.0007535000721258791,
+      "loss": 2.6709,
+      "step": 200
+    },
+    {
+      "epoch": 0.3853211009174312,
+      "grad_norm": 0.08734503388404846,
+      "learning_rate": 0.00072685988254984,
+      "loss": 2.6578,
+      "step": 210
+    },
+    {
+      "epoch": 0.4036697247706422,
+      "grad_norm": 0.09613344073295593,
+      "learning_rate": 0.0006993822741429906,
+      "loss": 2.6651,
+      "step": 220
+    },
+    {
+      "epoch": 0.42201834862385323,
+      "grad_norm": 0.09078177809715271,
+      "learning_rate": 0.0006711686763270818,
+      "loss": 2.6452,
+      "step": 230
+    },
+    {
+      "epoch": 0.44036697247706424,
+      "grad_norm": 0.08587699383497238,
+      "learning_rate": 0.0006423232353170831,
+      "loss": 2.6266,
+      "step": 240
+    },
+    {
+      "epoch": 0.45871559633027525,
+      "grad_norm": 0.08511137217283249,
+      "learning_rate": 0.0006129524296812335,
+      "loss": 2.6053,
+      "step": 250
+    },
+    {
+      "epoch": 0.47706422018348627,
+      "grad_norm": 0.0845821276307106,
+      "learning_rate": 0.0005831646772915651,
+      "loss": 2.5967,
+      "step": 260
+    },
+    {
+      "epoch": 0.4954128440366973,
+      "grad_norm": 0.08163905888795853,
+      "learning_rate": 0.0005530699351157809,
+      "loss": 2.5631,
+      "step": 270
+    },
+    {
+      "epoch": 0.5137614678899083,
+      "grad_norm": 0.08580261468887329,
+      "learning_rate": 0.0005227792933277943,
+      "loss": 2.5536,
+      "step": 280
+    },
+    {
+      "epoch": 0.5321100917431193,
+      "grad_norm": 0.08362813293933868,
+      "learning_rate": 0.0004924045652352092,
+      "loss": 2.5296,
+      "step": 290
+    },
+    {
+      "epoch": 0.5504587155963303,
+      "grad_norm": 0.07746115326881409,
+      "learning_rate": 0.0004620578745374633,
+      "loss": 2.5174,
+      "step": 300
+    },
+    {
+      "epoch": 0.5688073394495413,
+      "grad_norm": 0.09116074442863464,
+      "learning_rate": 0.0004318512414382059,
+      "loss": 2.5136,
+      "step": 310
+    },
+    {
+      "epoch": 0.5871559633027523,
+      "grad_norm": 0.0962422713637352,
+      "learning_rate": 0.0004018961691397148,
+      "loss": 2.5061,
+      "step": 320
+    },
+    {
+      "epoch": 0.6055045871559633,
+      "grad_norm": 0.10093940049409866,
+      "learning_rate": 0.0003723032322457458,
+      "loss": 2.4847,
+      "step": 330
+    },
+    {
+      "epoch": 0.6238532110091743,
+      "grad_norm": 0.08544225990772247,
+      "learning_rate": 0.0003431816685921625,
+      "loss": 2.4824,
+      "step": 340
+    },
+    {
+      "epoch": 0.6422018348623854,
+      "grad_norm": 0.1030346006155014,
+      "learning_rate": 0.0003146389760120416,
+      "loss": 2.4756,
+      "step": 350
+    },
+    {
+      "epoch": 0.6605504587155964,
+      "grad_norm": 0.07144945859909058,
+      "learning_rate": 0.00028678051552373485,
+      "loss": 2.4749,
+      "step": 360
+    },
+    {
+      "epoch": 0.6788990825688074,
+      "grad_norm": 0.07039240747690201,
+      "learning_rate": 0.0002597091224066581,
+      "loss": 2.464,
+      "step": 370
+    },
+    {
+      "epoch": 0.6972477064220184,
+      "grad_norm": 0.07905304431915283,
+      "learning_rate": 0.0002335247266004629,
+      "loss": 2.4711,
+      "step": 380
+    },
+    {
+      "epoch": 0.7155963302752294,
+      "grad_norm": 0.07283951342105865,
+      "learning_rate": 0.000208323983828828,
+      "loss": 2.46,
+      "step": 390
+    },
+    {
+      "epoch": 0.7339449541284404,
+      "grad_norm": 0.07436925917863846,
+      "learning_rate": 0.0001841999188095224,
+      "loss": 2.4491,
+      "step": 400
+    },
+    {
+      "epoch": 0.7522935779816514,
+      "grad_norm": 0.06956744194030762,
+      "learning_rate": 0.00016124158186777676,
+      "loss": 2.455,
+      "step": 410
+    },
+    {
+      "epoch": 0.7706422018348624,
+      "grad_norm": 0.07781730592250824,
+      "learning_rate": 0.0001395337202205211,
+      "loss": 2.4482,
+      "step": 420
+    },
+    {
+      "epoch": 0.7889908256880734,
+      "grad_norm": 0.07016037404537201,
+      "learning_rate": 0.00011915646514489292,
+      "loss": 2.4466,
+      "step": 430
+    },
+    {
+      "epoch": 0.8073394495412844,
+      "grad_norm": 0.07375302165746689,
+      "learning_rate": 0.00010018503618578818,
+      "loss": 2.4514,
+      "step": 440
+    },
+    {
+      "epoch": 0.8256880733944955,
+      "grad_norm": 0.07523083686828613,
+      "learning_rate": 8.268946349432583e-05,
+      "loss": 2.4446,
+      "step": 450
+    },
+    {
+      "epoch": 0.8440366972477065,
+      "grad_norm": 0.06791547685861588,
+      "learning_rate": 6.673432932217117e-05,
+      "loss": 2.4393,
+      "step": 460
+    },
+    {
+      "epoch": 0.8623853211009175,
+      "grad_norm": 0.06926409900188446,
+      "learning_rate": 5.2378529625954684e-05,
+      "loss": 2.4295,
+      "step": 470
+    },
+    {
+      "epoch": 0.8807339449541285,
+      "grad_norm": 0.06726662814617157,
+      "learning_rate": 3.9675056661785556e-05,
+      "loss": 2.4354,
+      "step": 480
+    },
+    {
+      "epoch": 0.8990825688073395,
+      "grad_norm": 0.06586063653230667,
+      "learning_rate": 2.8670803372379538e-05,
+      "loss": 2.4401,
+      "step": 490
+    },
+    {
+      "epoch": 0.9174311926605505,
+      "grad_norm": 0.06758908927440643,
+      "learning_rate": 1.9406390288876586e-05,
+      "loss": 2.4471,
+      "step": 500
+    },
+    {
+      "epoch": 0.9357798165137615,
+      "grad_norm": 0.06711822748184204,
+      "learning_rate": 1.191601558631461e-05,
+      "loss": 2.4357,
+      "step": 510
+    },
+    {
+      "epoch": 0.9541284403669725,
+      "grad_norm": 0.06686167418956757,
+      "learning_rate": 6.2273288462559305e-06,
+      "loss": 2.4329,
+      "step": 520
+    },
+    {
+      "epoch": 0.9724770642201835,
+      "grad_norm": 0.06549740582704544,
+      "learning_rate": 2.361328992552314e-06,
+      "loss": 2.4375,
+      "step": 530
+    },
+    {
+      "epoch": 0.9908256880733946,
+      "grad_norm": 0.06334860622882843,
+      "learning_rate": 3.322867770044202e-07,
+      "loss": 2.4322,
+      "step": 540
+    },
+    {
+      "epoch": 1.0,
+      "step": 545,
+      "total_flos": 3.335222257747231e+17,
+      "train_loss": 2.6920029158985943,
+      "train_runtime": 5352.04,
+      "train_samples_per_second": 104.283,
+      "train_steps_per_second": 0.102
+    }
+  ],
+  "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": 3.335222257747231e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen25/siglip2_sm16_512/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/siglip2_sm16_512/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": 1152,
+  "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-SO400M-16-SigLIP2-512",
+  "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/siglip2/siglip2_sm16_512.npz",
+  "vocab_size": 151936
+}

continuous/qwen25/siglip2_sm16_512/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/siglip2_sm16_512/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/siglip2_sm16_512/finetune/tokenizer.json

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

continuous/qwen25/siglip2_sm16_512/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/siglip2_sm16_512/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": 8.157779693603516,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 1.5444,
+      "step": 10
+    },
+    {
+      "epoch": 0.007700452401578592,
+      "grad_norm": 2.069674491882324,
+      "learning_rate": 2.435897435897436e-06,
+      "loss": 1.2177,
+      "step": 20
+    },
+    {
+      "epoch": 0.01155067860236789,
+      "grad_norm": 1.694888949394226,
+      "learning_rate": 3.7179487179487184e-06,
+      "loss": 1.0107,
+      "step": 30
+    },
+    {
+      "epoch": 0.015400904803157185,
+      "grad_norm": 0.33352357149124146,
+      "learning_rate": 5e-06,
+      "loss": 0.8708,
+      "step": 40
+    },
+    {
+      "epoch": 0.019251131003946483,
+      "grad_norm": 0.21481186151504517,
+      "learning_rate": 6.282051282051282e-06,
+      "loss": 0.8603,
+      "step": 50
+    },
+    {
+      "epoch": 0.02310135720473578,
+      "grad_norm": 0.16007263958454132,
+      "learning_rate": 7.564102564102564e-06,
+      "loss": 0.818,
+      "step": 60
+    },
+    {
+      "epoch": 0.026951583405525073,
+      "grad_norm": 0.12302619218826294,
+      "learning_rate": 8.846153846153847e-06,
+      "loss": 0.794,
+      "step": 70
+    },
+    {
+      "epoch": 0.03080180960631437,
+      "grad_norm": 0.1334974318742752,
+      "learning_rate": 9.999996111488723e-06,
+      "loss": 0.8047,
+      "step": 80
+    },
+    {
+      "epoch": 0.034652035807103666,
+      "grad_norm": 0.13376188278198242,
+      "learning_rate": 9.999529497453782e-06,
+      "loss": 0.7953,
+      "step": 90
+    },
+    {
+      "epoch": 0.038502262007892966,
+      "grad_norm": 0.12296710163354874,
+      "learning_rate": 9.998285264324193e-06,
+      "loss": 0.7628,
+      "step": 100
+    },
+    {
+      "epoch": 0.04235248820868226,
+      "grad_norm": 0.11967366188764572,
+      "learning_rate": 9.996263605626052e-06,
+      "loss": 0.7575,
+      "step": 110
+    },
+    {
+      "epoch": 0.04620271440947156,
+      "grad_norm": 0.12936440110206604,
+      "learning_rate": 9.99346483580503e-06,
+      "loss": 0.769,
+      "step": 120
+    },
+    {
+      "epoch": 0.05005294061026085,
+      "grad_norm": 0.12079093605279922,
+      "learning_rate": 9.989889390177467e-06,
+      "loss": 0.7523,
+      "step": 130
+    },
+    {
+      "epoch": 0.053903166811050146,
+      "grad_norm": 0.1317361295223236,
+      "learning_rate": 9.985537824862647e-06,
+      "loss": 0.76,
+      "step": 140
+    },
+    {
+      "epoch": 0.057753393011839446,
+      "grad_norm": 0.14114120602607727,
+      "learning_rate": 9.980410816696317e-06,
+      "loss": 0.7466,
+      "step": 150
+    },
+    {
+      "epoch": 0.06160361921262874,
+      "grad_norm": 0.1292302906513214,
+      "learning_rate": 9.97450916312541e-06,
+      "loss": 0.749,
+      "step": 160
+    },
+    {
+      "epoch": 0.06545384541341803,
+      "grad_norm": 0.11038309335708618,
+      "learning_rate": 9.967833782084004e-06,
+      "loss": 0.7493,
+      "step": 170
+    },
+    {
+      "epoch": 0.06930407161420733,
+      "grad_norm": 0.12500010430812836,
+      "learning_rate": 9.960385711850554e-06,
+      "loss": 0.7686,
+      "step": 180
+    },
+    {
+      "epoch": 0.07315429781499663,
+      "grad_norm": 0.12344563007354736,
+      "learning_rate": 9.952166110886394e-06,
+      "loss": 0.7717,
+      "step": 190
+    },
+    {
+      "epoch": 0.07700452401578593,
+      "grad_norm": 0.12659776210784912,
+      "learning_rate": 9.943176257655567e-06,
+      "loss": 0.7511,
+      "step": 200
+    },
+    {
+      "epoch": 0.08085475021657522,
+      "grad_norm": 0.11685220897197723,
+      "learning_rate": 9.933417550425952e-06,
+      "loss": 0.759,
+      "step": 210
+    },
+    {
+      "epoch": 0.08470497641736452,
+      "grad_norm": 0.11734801530838013,
+      "learning_rate": 9.9228915070518e-06,
+      "loss": 0.7331,
+      "step": 220
+    },
+    {
+      "epoch": 0.08855520261815382,
+      "grad_norm": 0.13113221526145935,
+      "learning_rate": 9.911599764737632e-06,
+      "loss": 0.7406,
+      "step": 230
+    },
+    {
+      "epoch": 0.09240542881894312,
+      "grad_norm": 0.12880147993564606,
+      "learning_rate": 9.899544079783606e-06,
+      "loss": 0.7254,
+      "step": 240
+    },
+    {
+      "epoch": 0.0962556550197324,
+      "grad_norm": 0.12021780014038086,
+      "learning_rate": 9.886726327312338e-06,
+      "loss": 0.745,
+      "step": 250
+    },
+    {
+      "epoch": 0.1001058812205217,
+      "grad_norm": 0.1293252408504486,
+      "learning_rate": 9.87314850097724e-06,
+      "loss": 0.7485,
+      "step": 260
+    },
+    {
+      "epoch": 0.103956107421311,
+      "grad_norm": 0.11995439976453781,
+      "learning_rate": 9.858812712652452e-06,
+      "loss": 0.7336,
+      "step": 270
+    },
+    {
+      "epoch": 0.10780633362210029,
+      "grad_norm": 0.11761198937892914,
+      "learning_rate": 9.843721192104341e-06,
+      "loss": 0.7352,
+      "step": 280
+    },
+    {
+      "epoch": 0.11165655982288959,
+      "grad_norm": 0.11756610125303268,
+      "learning_rate": 9.827876286644698e-06,
+      "loss": 0.7526,
+      "step": 290
+    },
+    {
+      "epoch": 0.11550678602367889,
+      "grad_norm": 0.11191927641630173,
+      "learning_rate": 9.811280460765643e-06,
+      "loss": 0.7154,
+      "step": 300
+    },
+    {
+      "epoch": 0.11935701222446819,
+      "grad_norm": 0.11439258605241776,
+      "learning_rate": 9.793936295756292e-06,
+      "loss": 0.7257,
+      "step": 310
+    },
+    {
+      "epoch": 0.12320723842525748,
+      "grad_norm": 0.11650232970714569,
+      "learning_rate": 9.775846489301279e-06,
+      "loss": 0.7312,
+      "step": 320
+    },
+    {
+      "epoch": 0.12705746462604678,
+      "grad_norm": 0.11637041717767715,
+      "learning_rate": 9.757013855061157e-06,
+      "loss": 0.7303,
+      "step": 330
+    },
+    {
+      "epoch": 0.13090769082683607,
+      "grad_norm": 0.11799060553312302,
+      "learning_rate": 9.737441322234758e-06,
+      "loss": 0.738,
+      "step": 340
+    },
+    {
+      "epoch": 0.13475791702762538,
+      "grad_norm": 0.1261454075574875,
+      "learning_rate": 9.717131935103605e-06,
+      "loss": 0.744,
+      "step": 350
+    },
+    {
+      "epoch": 0.13860814322841467,
+      "grad_norm": 0.1119656190276146,
+      "learning_rate": 9.696088852558398e-06,
+      "loss": 0.7277,
+      "step": 360
+    },
+    {
+      "epoch": 0.14245836942920398,
+      "grad_norm": 0.11379571259021759,
+      "learning_rate": 9.674315347607686e-06,
+      "loss": 0.7518,
+      "step": 370
+    },
+    {
+      "epoch": 0.14630859562999327,
+      "grad_norm": 0.11650161445140839,
+      "learning_rate": 9.651814806868796e-06,
+      "loss": 0.7278,
+      "step": 380
+    },
+    {
+      "epoch": 0.15015882183078255,
+      "grad_norm": 0.11736980080604553,
+      "learning_rate": 9.62859073004107e-06,
+      "loss": 0.7242,
+      "step": 390
+    },
+    {
+      "epoch": 0.15400904803157187,
+      "grad_norm": 0.10497795790433884,
+      "learning_rate": 9.604646729361547e-06,
+      "loss": 0.728,
+      "step": 400
+    },
+    {
+      "epoch": 0.15785927423236115,
+      "grad_norm": 0.12334537506103516,
+      "learning_rate": 9.579986529043099e-06,
+      "loss": 0.7098,
+      "step": 410
+    },
+    {
+      "epoch": 0.16170950043315044,
+      "grad_norm": 0.12592849135398865,
+      "learning_rate": 9.554613964695189e-06,
+      "loss": 0.7281,
+      "step": 420
+    },
+    {
+      "epoch": 0.16555972663393975,
+      "grad_norm": 0.11706776916980743,
+      "learning_rate": 9.528532982727281e-06,
+      "loss": 0.7239,
+      "step": 430
+    },
+    {
+      "epoch": 0.16940995283472904,
+      "grad_norm": 0.12140273302793503,
+      "learning_rate": 9.501747639735018e-06,
+      "loss": 0.7186,
+      "step": 440
+    },
+    {
+      "epoch": 0.17326017903551832,
+      "grad_norm": 0.1175365298986435,
+      "learning_rate": 9.474262101869272e-06,
+      "loss": 0.7074,
+      "step": 450
+    },
+    {
+      "epoch": 0.17711040523630764,
+      "grad_norm": 0.12024547904729843,
+      "learning_rate": 9.446080644188141e-06,
+      "loss": 0.7233,
+      "step": 460
+    },
+    {
+      "epoch": 0.18096063143709692,
+      "grad_norm": 0.12276304513216019,
+      "learning_rate": 9.417207649992017e-06,
+      "loss": 0.7232,
+      "step": 470
+    },
+    {
+      "epoch": 0.18481085763788624,
+      "grad_norm": 0.12062890082597733,
+      "learning_rate": 9.387647610141816e-06,
+      "loss": 0.7157,
+      "step": 480
+    },
+    {
+      "epoch": 0.18866108383867553,
+      "grad_norm": 0.10733546316623688,
+      "learning_rate": 9.35740512236047e-06,
+      "loss": 0.7099,
+      "step": 490
+    },
+    {
+      "epoch": 0.1925113100394648,
+      "grad_norm": 0.12261103093624115,
+      "learning_rate": 9.326484890517809e-06,
+      "loss": 0.7188,
+      "step": 500
+    },
+    {
+      "epoch": 0.19636153624025413,
+      "grad_norm": 0.12567655742168427,
+      "learning_rate": 9.294891723898926e-06,
+      "loss": 0.699,
+      "step": 510
+    },
+    {
+      "epoch": 0.2002117624410434,
+      "grad_norm": 0.1224989965558052,
+      "learning_rate": 9.262630536456153e-06,
+      "loss": 0.7376,
+      "step": 520
+    },
+    {
+      "epoch": 0.2040619886418327,
+      "grad_norm": 0.12366101890802383,
+      "learning_rate": 9.229706346044749e-06,
+      "loss": 0.7102,
+      "step": 530
+    },
+    {
+      "epoch": 0.207912214842622,
+      "grad_norm": 0.12583835422992706,
+      "learning_rate": 9.196124273642422e-06,
+      "loss": 0.7253,
+      "step": 540
+    },
+    {
+      "epoch": 0.2117624410434113,
+      "grad_norm": 0.11258824169635773,
+      "learning_rate": 9.161889542552842e-06,
+      "loss": 0.735,
+      "step": 550
+    },
+    {
+      "epoch": 0.21561266724420058,
+      "grad_norm": 0.11222588270902634,
+      "learning_rate": 9.127007477593193e-06,
+      "loss": 0.7286,
+      "step": 560
+    },
+    {
+      "epoch": 0.2194628934449899,
+      "grad_norm": 0.12222995609045029,
+      "learning_rate": 9.091483504265974e-06,
+      "loss": 0.7374,
+      "step": 570
+    },
+    {
+      "epoch": 0.22331311964577918,
+      "grad_norm": 0.12346022576093674,
+      "learning_rate": 9.055323147915116e-06,
+      "loss": 0.7131,
+      "step": 580
+    },
+    {
+      "epoch": 0.2271633458465685,
+      "grad_norm": 0.11889581382274628,
+      "learning_rate": 9.018532032866585e-06,
+      "loss": 0.7284,
+      "step": 590
+    },
+    {
+      "epoch": 0.23101357204735778,
+      "grad_norm": 0.12287424504756927,
+      "learning_rate": 8.981115881553579e-06,
+      "loss": 0.7265,
+      "step": 600
+    },
+    {
+      "epoch": 0.23486379824814707,
+      "grad_norm": 0.12596268951892853,
+      "learning_rate": 8.943080513626476e-06,
+      "loss": 0.7376,
+      "step": 610
+    },
+    {
+      "epoch": 0.23871402444893638,
+      "grad_norm": 0.11670832335948944,
+      "learning_rate": 8.904431845047641e-06,
+      "loss": 0.7293,
+      "step": 620
+    },
+    {
+      "epoch": 0.24256425064972567,
+      "grad_norm": 0.12210650742053986,
+      "learning_rate": 8.86517588717128e-06,
+      "loss": 0.7343,
+      "step": 630
+    },
+    {
+      "epoch": 0.24641447685051496,
+      "grad_norm": 0.10909974575042725,
+      "learning_rate": 8.82531874580844e-06,
+      "loss": 0.7355,
+      "step": 640
+    },
+    {
+      "epoch": 0.25026470305130427,
+      "grad_norm": 0.10707728564739227,
+      "learning_rate": 8.784866620277314e-06,
+      "loss": 0.691,
+      "step": 650
+    },
+    {
+      "epoch": 0.25411492925209356,
+      "grad_norm": 0.119625985622406,
+      "learning_rate": 8.743825802439018e-06,
+      "loss": 0.7062,
+      "step": 660
+    },
+    {
+      "epoch": 0.25796515545288284,
+      "grad_norm": 0.12268903851509094,
+      "learning_rate": 8.702202675718956e-06,
+      "loss": 0.7309,
+      "step": 670
+    },
+    {
+      "epoch": 0.26181538165367213,
+      "grad_norm": 0.11124296486377716,
+      "learning_rate": 8.660003714113959e-06,
+      "loss": 0.7043,
+      "step": 680
+    },
+    {
+      "epoch": 0.26566560785446147,
+      "grad_norm": 0.1167481318116188,
+      "learning_rate": 8.617235481185323e-06,
+      "loss": 0.7149,
+      "step": 690
+    },
+    {
+      "epoch": 0.26951583405525076,
+      "grad_norm": 0.12251907587051392,
+      "learning_rate": 8.573904629037931e-06,
+      "loss": 0.7218,
+      "step": 700
+    },
+    {
+      "epoch": 0.27336606025604004,
+      "grad_norm": 0.11618099361658096,
+      "learning_rate": 8.530017897285585e-06,
+      "loss": 0.7077,
+      "step": 710
+    },
+    {
+      "epoch": 0.27721628645682933,
+      "grad_norm": 0.107750304043293,
+      "learning_rate": 8.485582112002746e-06,
+      "loss": 0.7177,
+      "step": 720
+    },
+    {
+      "epoch": 0.2810665126576186,
+      "grad_norm": 0.11926370859146118,
+      "learning_rate": 8.440604184662811e-06,
+      "loss": 0.7185,
+      "step": 730
+    },
+    {
+      "epoch": 0.28491673885840796,
+      "grad_norm": 0.1172381266951561,
+      "learning_rate": 8.395091111063118e-06,
+      "loss": 0.7052,
+      "step": 740
+    },
+    {
+      "epoch": 0.28876696505919724,
+      "grad_norm": 0.10965097695589066,
+      "learning_rate": 8.349049970236822e-06,
+      "loss": 0.7062,
+      "step": 750
+    },
+    {
+      "epoch": 0.29261719125998653,
+      "grad_norm": 0.11074922978878021,
+      "learning_rate": 8.302487923351842e-06,
+      "loss": 0.6951,
+      "step": 760
+    },
+    {
+      "epoch": 0.2964674174607758,
+      "grad_norm": 0.11895854026079178,
+      "learning_rate": 8.255412212597021e-06,
+      "loss": 0.7163,
+      "step": 770
+    },
+    {
+      "epoch": 0.3003176436615651,
+      "grad_norm": 0.11898300051689148,
+      "learning_rate": 8.207830160055688e-06,
+      "loss": 0.7115,
+      "step": 780
+    },
+    {
+      "epoch": 0.3041678698623544,
+      "grad_norm": 0.11723551154136658,
+      "learning_rate": 8.159749166566786e-06,
+      "loss": 0.6985,
+      "step": 790
+    },
+    {
+      "epoch": 0.30801809606314373,
+      "grad_norm": 0.11621058732271194,
+      "learning_rate": 8.11117671057377e-06,
+      "loss": 0.7222,
+      "step": 800
+    },
+    {
+      "epoch": 0.311868322263933,
+      "grad_norm": 0.11543254554271698,
+      "learning_rate": 8.062120346961404e-06,
+      "loss": 0.7076,
+      "step": 810
+    },
+    {
+      "epoch": 0.3157185484647223,
+      "grad_norm": 0.10591620206832886,
+      "learning_rate": 8.012587705880705e-06,
+      "loss": 0.7145,
+      "step": 820
+    },
+    {
+      "epoch": 0.3195687746655116,
+      "grad_norm": 0.11738603562116623,
+      "learning_rate": 7.962586491562146e-06,
+      "loss": 0.7019,
+      "step": 830
+    },
+    {
+      "epoch": 0.3234190008663009,
+      "grad_norm": 0.11932270228862762,
+      "learning_rate": 7.912124481117357e-06,
+      "loss": 0.7032,
+      "step": 840
+    },
+    {
+      "epoch": 0.3272692270670902,
+      "grad_norm": 0.11175958067178726,
+      "learning_rate": 7.861209523329494e-06,
+      "loss": 0.7112,
+      "step": 850
+    },
+    {
+      "epoch": 0.3311194532678795,
+      "grad_norm": 0.10911305248737335,
+      "learning_rate": 7.809849537432432e-06,
+      "loss": 0.7005,
+      "step": 860
+    },
+    {
+      "epoch": 0.3349696794686688,
+      "grad_norm": 0.10896562039852142,
+      "learning_rate": 7.758052511879035e-06,
+      "loss": 0.699,
+      "step": 870
+    },
+    {
+      "epoch": 0.3388199056694581,
+      "grad_norm": 0.10926252603530884,
+      "learning_rate": 7.705826503098642e-06,
+      "loss": 0.7053,
+      "step": 880
+    },
+    {
+      "epoch": 0.34267013187024736,
+      "grad_norm": 0.11763233691453934,
+      "learning_rate": 7.653179634243979e-06,
+      "loss": 0.6986,
+      "step": 890
+    },
+    {
+      "epoch": 0.34652035807103665,
+      "grad_norm": 0.10085593163967133,
+      "learning_rate": 7.600120093927696e-06,
+      "loss": 0.6939,
+      "step": 900
+    },
+    {
+      "epoch": 0.350370584271826,
+      "grad_norm": 0.12219878286123276,
+      "learning_rate": 7.546656134948729e-06,
+      "loss": 0.6907,
+      "step": 910
+    },
+    {
+      "epoch": 0.3542208104726153,
+      "grad_norm": 0.11128900945186615,
+      "learning_rate": 7.492796073008661e-06,
+      "loss": 0.6884,
+      "step": 920
+    },
+    {
+      "epoch": 0.35807103667340456,
+      "grad_norm": 0.12422802299261093,
+      "learning_rate": 7.438548285418326e-06,
+      "loss": 0.6846,
+      "step": 930
+    },
+    {
+      "epoch": 0.36192126287419385,
+      "grad_norm": 0.1223989799618721,
+      "learning_rate": 7.3839212097948e-06,
+      "loss": 0.7199,
+      "step": 940
+    },
+    {
+      "epoch": 0.36577148907498314,
+      "grad_norm": 0.1059228852391243,
+      "learning_rate": 7.328923342749039e-06,
+      "loss": 0.7328,
+      "step": 950
+    },
+    {
+      "epoch": 0.3696217152757725,
+      "grad_norm": 0.11564399302005768,
+      "learning_rate": 7.273563238564319e-06,
+      "loss": 0.7106,
+      "step": 960
+    },
+    {
+      "epoch": 0.37347194147656176,
+      "grad_norm": 0.11198079586029053,
+      "learning_rate": 7.217849507865724e-06,
+      "loss": 0.687,
+      "step": 970
+    },
+    {
+      "epoch": 0.37732216767735105,
+      "grad_norm": 0.1099444255232811,
+      "learning_rate": 7.161790816280858e-06,
+      "loss": 0.7038,
+      "step": 980
+    },
+    {
+      "epoch": 0.38117239387814034,
+      "grad_norm": 0.11633051186800003,
+      "learning_rate": 7.1053958830920125e-06,
+      "loss": 0.6755,
+      "step": 990
+    },
+    {
+      "epoch": 0.3850226200789296,
+      "grad_norm": 0.11083025485277176,
+      "learning_rate": 7.0486734798799715e-06,
+      "loss": 0.7068,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3888728462797189,
+      "grad_norm": 0.11431955546140671,
+      "learning_rate": 6.991632429159708e-06,
+      "loss": 0.7285,
+      "step": 1010
+    },
+    {
+      "epoch": 0.39272307248050825,
+      "grad_norm": 0.11991675943136215,
+      "learning_rate": 6.934281603008132e-06,
+      "loss": 0.7038,
+      "step": 1020
+    },
+    {
+      "epoch": 0.39657329868129754,
+      "grad_norm": 0.11010961979627609,
+      "learning_rate": 6.876629921684143e-06,
+      "loss": 0.7154,
+      "step": 1030
+    },
+    {
+      "epoch": 0.4004235248820868,
+      "grad_norm": 0.11045470833778381,
+      "learning_rate": 6.818686352241199e-06,
+      "loss": 0.7118,
+      "step": 1040
+    },
+    {
+      "epoch": 0.4042737510828761,
+      "grad_norm": 0.10537164658308029,
+      "learning_rate": 6.760459907132573e-06,
+      "loss": 0.7014,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4081239772836654,
+      "grad_norm": 0.11091981083154678,
+      "learning_rate": 6.701959642809589e-06,
+      "loss": 0.7051,
+      "step": 1060
+    },
+    {
+      "epoch": 0.41197420348445474,
+      "grad_norm": 0.11931976675987244,
+      "learning_rate": 6.643194658312987e-06,
+      "loss": 0.6869,
+      "step": 1070
+    },
+    {
+      "epoch": 0.415824429685244,
+      "grad_norm": 0.10873254388570786,
+      "learning_rate": 6.584174093857676e-06,
+      "loss": 0.7249,
+      "step": 1080
+    },
+    {
+      "epoch": 0.4196746558860333,
+      "grad_norm": 0.12490089982748032,
+      "learning_rate": 6.5249071294110735e-06,
+      "loss": 0.6969,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4235248820868226,
+      "grad_norm": 0.12381531298160553,
+      "learning_rate": 6.465402983265274e-06,
+      "loss": 0.717,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4273751082876119,
+      "grad_norm": 0.11477857828140259,
+      "learning_rate": 6.405670910603243e-06,
+      "loss": 0.7137,
+      "step": 1110
+    },
+    {
+      "epoch": 0.43122533448840117,
+      "grad_norm": 0.11906164139509201,
+      "learning_rate": 6.345720202059291e-06,
+      "loss": 0.7045,
+      "step": 1120
+    },
+    {
+      "epoch": 0.4350755606891905,
+      "grad_norm": 0.11087071150541306,
+      "learning_rate": 6.285560182274008e-06,
+      "loss": 0.6925,
+      "step": 1130
+    },
+    {
+      "epoch": 0.4389257868899798,
+      "grad_norm": 0.11290790140628815,
+      "learning_rate": 6.225200208443936e-06,
+      "loss": 0.7004,
+      "step": 1140
+    },
+    {
+      "epoch": 0.4427760130907691,
+      "grad_norm": 0.10713221877813339,
+      "learning_rate": 6.16464966886616e-06,
+      "loss": 0.6945,
+      "step": 1150
+    },
+    {
+      "epoch": 0.44662623929155837,
+      "grad_norm": 0.13314682245254517,
+      "learning_rate": 6.103917981478065e-06,
+      "loss": 0.7246,
+      "step": 1160
+    },
+    {
+      "epoch": 0.45047646549234766,
+      "grad_norm": 0.11017236113548279,
+      "learning_rate": 6.043014592392489e-06,
+      "loss": 0.7303,
+      "step": 1170
+    },
+    {
+      "epoch": 0.454326691693137,
+      "grad_norm": 0.10324367135763168,
+      "learning_rate": 5.981948974428485e-06,
+      "loss": 0.7032,
+      "step": 1180
+    },
+    {
+      "epoch": 0.4581769178939263,
+      "grad_norm": 0.11601092666387558,
+      "learning_rate": 5.920730625637934e-06,
+      "loss": 0.6833,
+      "step": 1190
+    },
+    {
+      "epoch": 0.46202714409471557,
+      "grad_norm": 0.11853735148906708,
+      "learning_rate": 5.8593690678282355e-06,
+      "loss": 0.6961,
+      "step": 1200
+    },
+    {
+      "epoch": 0.46587737029550486,
+      "grad_norm": 0.11531855911016464,
+      "learning_rate": 5.797873845081301e-06,
+      "loss": 0.6983,
+      "step": 1210
+    },
+    {
+      "epoch": 0.46972759649629414,
+      "grad_norm": 0.11639226973056793,
+      "learning_rate": 5.736254522269075e-06,
+      "loss": 0.71,
+      "step": 1220
+    },
+    {
+      "epoch": 0.47357782269708343,
+      "grad_norm": 0.11076462268829346,
+      "learning_rate": 5.6745206835658386e-06,
+      "loss": 0.7176,
+      "step": 1230
+    },
+    {
+      "epoch": 0.47742804889787277,
+      "grad_norm": 0.10609985142946243,
+      "learning_rate": 5.612681930957495e-06,
+      "loss": 0.6932,
+      "step": 1240
+    },
+    {
+      "epoch": 0.48127827509866206,
+      "grad_norm": 0.10618698596954346,
+      "learning_rate": 5.550747882748095e-06,
+      "loss": 0.6874,
+      "step": 1250
+    },
+    {
+      "epoch": 0.48512850129945134,
+      "grad_norm": 0.11811337620019913,
+      "learning_rate": 5.4887281720638195e-06,
+      "loss": 0.6959,
+      "step": 1260
+    },
+    {
+      "epoch": 0.48897872750024063,
+      "grad_norm": 0.1287222057580948,
+      "learning_rate": 5.42663244535466e-06,
+      "loss": 0.681,
+      "step": 1270
+    },
+    {
+      "epoch": 0.4928289537010299,
+      "grad_norm": 0.10762614756822586,
+      "learning_rate": 5.3644703608940205e-06,
+      "loss": 0.6883,
+      "step": 1280
+    },
+    {
+      "epoch": 0.49667917990181926,
+      "grad_norm": 0.11591973155736923,
+      "learning_rate": 5.302251587276491e-06,
+      "loss": 0.7067,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5005294061026085,
+      "grad_norm": 0.11014992743730545,
+      "learning_rate": 5.2399858019140005e-06,
+      "loss": 0.702,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5043796323033978,
+      "grad_norm": 0.11259311437606812,
+      "learning_rate": 5.177682689530615e-06,
+      "loss": 0.7176,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5082298585041871,
+      "grad_norm": 0.12066808342933655,
+      "learning_rate": 5.11535194065619e-06,
+      "loss": 0.6896,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5120800847049765,
+      "grad_norm": 0.12152079492807388,
+      "learning_rate": 5.053003250119113e-06,
+      "loss": 0.6898,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5159303109057657,
+      "grad_norm": 0.11432269960641861,
+      "learning_rate": 4.9906463155383925e-06,
+      "loss": 0.7004,
+      "step": 1340
+    },
+    {
+      "epoch": 0.519780537106555,
+      "grad_norm": 0.11438273638486862,
+      "learning_rate": 4.9282908358153046e-06,
+      "loss": 0.6967,
+      "step": 1350
+    },
+    {
+      "epoch": 0.5236307633073443,
+      "grad_norm": 0.12920495867729187,
+      "learning_rate": 4.8659465096248345e-06,
+      "loss": 0.7013,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5274809895081336,
+      "grad_norm": 0.10822011530399323,
+      "learning_rate": 4.803623033907169e-06,
+      "loss": 0.7031,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5313312157089229,
+      "grad_norm": 0.10786820948123932,
+      "learning_rate": 4.741330102359441e-06,
+      "loss": 0.715,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5351814419097122,
+      "grad_norm": 0.12308599799871445,
+      "learning_rate": 4.679077403927996e-06,
+      "loss": 0.6876,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5390316681105015,
+      "grad_norm": 0.11544201523065567,
+      "learning_rate": 4.616874621301373e-06,
+      "loss": 0.6964,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5428818943112907,
+      "grad_norm": 0.11415854841470718,
+      "learning_rate": 4.554731429404293e-06,
+      "loss": 0.7122,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5467321205120801,
+      "grad_norm": 0.11675619333982468,
+      "learning_rate": 4.492657493892822e-06,
+      "loss": 0.692,
+      "step": 1420
+    },
+    {
+      "epoch": 0.5505823467128694,
+      "grad_norm": 0.11594927310943604,
+      "learning_rate": 4.430662469650999e-06,
+      "loss": 0.7063,
+      "step": 1430
+    },
+    {
+      "epoch": 0.5544325729136587,
+      "grad_norm": 0.11460361629724503,
+      "learning_rate": 4.368755999289118e-06,
+      "loss": 0.7111,
+      "step": 1440
+    },
+    {
+      "epoch": 0.558282799114448,
+      "grad_norm": 0.12353015691041946,
+      "learning_rate": 4.306947711643949e-06,
+      "loss": 0.6889,
+      "step": 1450
+    },
+    {
+      "epoch": 0.5621330253152372,
+      "grad_norm": 0.11003080010414124,
+      "learning_rate": 4.245247220281068e-06,
+      "loss": 0.698,
+      "step": 1460
+    },
+    {
+      "epoch": 0.5659832515160266,
+      "grad_norm": 0.11045505851507187,
+      "learning_rate": 4.183664121999587e-06,
+      "loss": 0.6913,
+      "step": 1470
+    },
+    {
+      "epoch": 0.5698334777168159,
+      "grad_norm": 0.1205449029803276,
+      "learning_rate": 4.122207995339474e-06,
+      "loss": 0.6892,
+      "step": 1480
+    },
+    {
+      "epoch": 0.5736837039176051,
+      "grad_norm": 0.1131453886628151,
+      "learning_rate": 4.060888399091734e-06,
+      "loss": 0.6978,
+      "step": 1490
+    },
+    {
+      "epoch": 0.5775339301183945,
+      "grad_norm": 0.11320823431015015,
+      "learning_rate": 3.999714870811639e-06,
+      "loss": 0.7123,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5813841563191837,
+      "grad_norm": 0.11280110478401184,
+      "learning_rate": 3.938696925335275e-06,
+      "loss": 0.6999,
+      "step": 1510
+    },
+    {
+      "epoch": 0.5852343825199731,
+      "grad_norm": 0.11832990497350693,
+      "learning_rate": 3.87784405329962e-06,
+      "loss": 0.7058,
+      "step": 1520
+    },
+    {
+      "epoch": 0.5890846087207623,
+      "grad_norm": 0.10939127206802368,
+      "learning_rate": 3.817165719666382e-06,
+      "loss": 0.7021,
+      "step": 1530
+    },
+    {
+      "epoch": 0.5929348349215516,
+      "grad_norm": 0.11118987947702408,
+      "learning_rate": 3.756671362249841e-06,
+      "loss": 0.7169,
+      "step": 1540
+    },
+    {
+      "epoch": 0.596785061122341,
+      "grad_norm": 0.11282653361558914,
+      "learning_rate": 3.6963703902488907e-06,
+      "loss": 0.7054,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6006352873231302,
+      "grad_norm": 0.1208108589053154,
+      "learning_rate": 3.6362721827835655e-06,
+      "loss": 0.7116,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6044855135239195,
+      "grad_norm": 0.11597804725170135,
+      "learning_rate": 3.576386087436212e-06,
+      "loss": 0.705,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6083357397247088,
+      "grad_norm": 0.11687645316123962,
+      "learning_rate": 3.5167214187975913e-06,
+      "loss": 0.7089,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6121859659254981,
+      "grad_norm": 0.10266745835542679,
+      "learning_rate": 3.457287457018086e-06,
+      "loss": 0.7034,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6160361921262875,
+      "grad_norm": 0.10696641355752945,
+      "learning_rate": 3.3980934463643003e-06,
+      "loss": 0.692,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6198864183270767,
+      "grad_norm": 0.10320296138525009,
+      "learning_rate": 3.3391485937812053e-06,
+      "loss": 0.6692,
+      "step": 1610
+    },
+    {
+      "epoch": 0.623736644527866,
+      "grad_norm": 0.11244715750217438,
+      "learning_rate": 3.280462067460115e-06,
+      "loss": 0.6735,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6275868707286553,
+      "grad_norm": 0.10968407243490219,
+      "learning_rate": 3.222042995412669e-06,
+      "loss": 0.7122,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6314370969294446,
+      "grad_norm": 0.1009286493062973,
+      "learning_rate": 3.163900464051089e-06,
+      "loss": 0.691,
+      "step": 1640
+    },
+    {
+      "epoch": 0.635287323130234,
+      "grad_norm": 0.11039683222770691,
+      "learning_rate": 3.1060435167748826e-06,
+      "loss": 0.7029,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6391375493310232,
+      "grad_norm": 0.11405289173126221,
+      "learning_rate": 3.048481152564253e-06,
+      "loss": 0.6918,
+      "step": 1660
+    },
+    {
+      "epoch": 0.6429877755318125,
+      "grad_norm": 0.12267174571752548,
+      "learning_rate": 2.9912223245804017e-06,
+      "loss": 0.7134,
+      "step": 1670
+    },
+    {
+      "epoch": 0.6468380017326018,
+      "grad_norm": 0.11549415439367294,
+      "learning_rate": 2.934275938772981e-06,
+      "loss": 0.6928,
+      "step": 1680
+    },
+    {
+      "epoch": 0.6506882279333911,
+      "grad_norm": 0.10781225562095642,
+      "learning_rate": 2.8776508524948676e-06,
+      "loss": 0.6814,
+      "step": 1690
+    },
+    {
+      "epoch": 0.6545384541341804,
+      "grad_norm": 0.10515108704566956,
+      "learning_rate": 2.821355873124503e-06,
+      "loss": 0.6825,
+      "step": 1700
+    },
+    {
+      "epoch": 0.6583886803349697,
+      "grad_norm": 0.1138271912932396,
+      "learning_rate": 2.7653997566960144e-06,
+      "loss": 0.7011,
+      "step": 1710
+    },
+    {
+      "epoch": 0.662238906535759,
+      "grad_norm": 0.11283262819051743,
+      "learning_rate": 2.709791206537309e-06,
+      "loss": 0.6916,
+      "step": 1720
+    },
+    {
+      "epoch": 0.6660891327365482,
+      "grad_norm": 0.11299089342355728,
+      "learning_rate": 2.6545388719163747e-06,
+      "loss": 0.6878,
+      "step": 1730
+    },
+    {
+      "epoch": 0.6699393589373376,
+      "grad_norm": 0.11119722574949265,
+      "learning_rate": 2.599651346695979e-06,
+      "loss": 0.701,
+      "step": 1740
+    },
+    {
+      "epoch": 0.6737895851381268,
+      "grad_norm": 0.11087042093276978,
+      "learning_rate": 2.545137167997007e-06,
+      "loss": 0.6973,
+      "step": 1750
+    },
+    {
+      "epoch": 0.6776398113389162,
+      "grad_norm": 0.10593321174383163,
+      "learning_rate": 2.4910048148705972e-06,
+      "loss": 0.682,
+      "step": 1760
+    },
+    {
+      "epoch": 0.6814900375397055,
+      "grad_norm": 0.11736290156841278,
+      "learning_rate": 2.4372627069793327e-06,
+      "loss": 0.725,
+      "step": 1770
+    },
+    {
+      "epoch": 0.6853402637404947,
+      "grad_norm": 0.11857771128416061,
+      "learning_rate": 2.3839192032876486e-06,
+      "loss": 0.7078,
+      "step": 1780
+    },
+    {
+      "epoch": 0.6891904899412841,
+      "grad_norm": 0.11438692361116409,
+      "learning_rate": 2.3309826007617047e-06,
+      "loss": 0.7186,
+      "step": 1790
+    },
+    {
+      "epoch": 0.6930407161420733,
+      "grad_norm": 0.10710203647613525,
+      "learning_rate": 2.278461133078881e-06,
+      "loss": 0.682,
+      "step": 1800
+    },
+    {
+      "epoch": 0.6968909423428626,
+      "grad_norm": 0.1150662750005722,
+      "learning_rate": 2.2263629693471258e-06,
+      "loss": 0.6793,
+      "step": 1810
+    },
+    {
+      "epoch": 0.700741168543652,
+      "grad_norm": 0.1067887470126152,
+      "learning_rate": 2.1746962128343414e-06,
+      "loss": 0.6985,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7045913947444412,
+      "grad_norm": 0.1094072088599205,
+      "learning_rate": 2.1234688997080183e-06,
+      "loss": 0.6986,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7084416209452306,
+      "grad_norm": 0.12142537534236908,
+      "learning_rate": 2.0726889977853053e-06,
+      "loss": 0.697,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7122918471460198,
+      "grad_norm": 0.11892116069793701,
+      "learning_rate": 2.022364405293703e-06,
+      "loss": 0.7078,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7161420733468091,
+      "grad_norm": 0.1125875934958458,
+      "learning_rate": 1.9725029496425814e-06,
+      "loss": 0.7046,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7199922995475985,
+      "grad_norm": 0.10260457545518875,
+      "learning_rate": 1.923112386205727e-06,
+      "loss": 0.6989,
+      "step": 1870
+    },
+    {
+      "epoch": 0.7238425257483877,
+      "grad_norm": 0.11379207670688629,
+      "learning_rate": 1.874200397115079e-06,
+      "loss": 0.706,
+      "step": 1880
+    },
+    {
+      "epoch": 0.727692751949177,
+      "grad_norm": 0.11633028090000153,
+      "learning_rate": 1.825774590065857e-06,
+      "loss": 0.6908,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7315429781499663,
+      "grad_norm": 0.12611044943332672,
+      "learning_rate": 1.7778424971332813e-06,
+      "loss": 0.7004,
+      "step": 1900
+    },
+    {
+      "epoch": 0.7353932043507556,
+      "grad_norm": 0.10737134516239166,
+      "learning_rate": 1.7304115736010407e-06,
+      "loss": 0.701,
+      "step": 1910
+    },
+    {
+      "epoch": 0.739243430551545,
+      "grad_norm": 0.10753585398197174,
+      "learning_rate": 1.6834891968017108e-06,
+      "loss": 0.6948,
+      "step": 1920
+    },
+    {
+      "epoch": 0.7430936567523342,
+      "grad_norm": 0.11780033260583878,
+      "learning_rate": 1.6370826649692884e-06,
+      "loss": 0.7014,
+      "step": 1930
+    },
+    {
+      "epoch": 0.7469438829531235,
+      "grad_norm": 0.1202617734670639,
+      "learning_rate": 1.5911991961040463e-06,
+      "loss": 0.7153,
+      "step": 1940
+    },
+    {
+      "epoch": 0.7507941091539128,
+      "grad_norm": 0.11155620217323303,
+      "learning_rate": 1.545845926849851e-06,
+      "loss": 0.7024,
+      "step": 1950
+    },
+    {
+      "epoch": 0.7546443353547021,
+      "grad_norm": 0.1169600561261177,
+      "learning_rate": 1.5010299113841397e-06,
+      "loss": 0.7174,
+      "step": 1960
+    },
+    {
+      "epoch": 0.7584945615554913,
+      "grad_norm": 0.11197064816951752,
+      "learning_rate": 1.4567581203207238e-06,
+      "loss": 0.7148,
+      "step": 1970
+    },
+    {
+      "epoch": 0.7623447877562807,
+      "grad_norm": 0.11338818073272705,
+      "learning_rate": 1.413037439625593e-06,
+      "loss": 0.6977,
+      "step": 1980
+    },
+    {
+      "epoch": 0.76619501395707,
+      "grad_norm": 0.12384513020515442,
+      "learning_rate": 1.3698746695458847e-06,
+      "loss": 0.6993,
+      "step": 1990
+    },
+    {
+      "epoch": 0.7700452401578592,
+      "grad_norm": 0.1113700345158577,
+      "learning_rate": 1.3272765235521745e-06,
+      "loss": 0.701,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7738954663586486,
+      "grad_norm": 0.11112021654844284,
+      "learning_rate": 1.2852496272942877e-06,
+      "loss": 0.7017,
+      "step": 2010
+    },
+    {
+      "epoch": 0.7777456925594378,
+      "grad_norm": 0.10407859832048416,
+      "learning_rate": 1.2438005175707396e-06,
+      "loss": 0.6845,
+      "step": 2020
+    },
+    {
+      "epoch": 0.7815959187602272,
+      "grad_norm": 0.12019067257642746,
+      "learning_rate": 1.2029356413120297e-06,
+      "loss": 0.6717,
+      "step": 2030
+    },
+    {
+      "epoch": 0.7854461449610165,
+      "grad_norm": 0.1137850433588028,
+      "learning_rate": 1.1626613545778786e-06,
+      "loss": 0.6891,
+      "step": 2040
+    },
+    {
+      "epoch": 0.7892963711618057,
+      "grad_norm": 0.11080484837293625,
+      "learning_rate": 1.1229839215686294e-06,
+      "loss": 0.6966,
+      "step": 2050
+    },
+    {
+      "epoch": 0.7931465973625951,
+      "grad_norm": 0.10433514416217804,
+      "learning_rate": 1.0839095136509188e-06,
+      "loss": 0.6756,
+      "step": 2060
+    },
+    {
+      "epoch": 0.7969968235633843,
+      "grad_norm": 0.11623077839612961,
+      "learning_rate": 1.045444208397791e-06,
+      "loss": 0.6874,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8008470497641736,
+      "grad_norm": 0.10453623533248901,
+      "learning_rate": 1.007593988643402e-06,
+      "loss": 0.6788,
+      "step": 2080
+    },
+    {
+      "epoch": 0.804697275964963,
+      "grad_norm": 0.10526397079229355,
+      "learning_rate": 9.703647415524613e-07,
+      "loss": 0.7019,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8085475021657522,
+      "grad_norm": 0.11028701812028885,
+      "learning_rate": 9.337622577045485e-07,
+      "loss": 0.6951,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8123977283665416,
+      "grad_norm": 0.11055753380060196,
+      "learning_rate": 8.977922301934588e-07,
+      "loss": 0.71,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8162479545673308,
+      "grad_norm": 0.11437880992889404,
+      "learning_rate": 8.624602537417009e-07,
+      "loss": 0.7066,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8200981807681201,
+      "grad_norm": 0.11723239719867706,
+      "learning_rate": 8.277718238303123e-07,
+      "loss": 0.6868,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8239484069689095,
+      "grad_norm": 0.11289215087890625,
+      "learning_rate": 7.937323358440935e-07,
+      "loss": 0.6999,
+      "step": 2140
+    },
+    {
+      "epoch": 0.8277986331696987,
+      "grad_norm": 0.13426369428634644,
+      "learning_rate": 7.603470842324223e-07,
+      "loss": 0.682,
+      "step": 2150
+    },
+    {
+      "epoch": 0.831648859370488,
+      "grad_norm": 0.11011021584272385,
+      "learning_rate": 7.276212616857575e-07,
+      "loss": 0.7031,
+      "step": 2160
+    },
+    {
+      "epoch": 0.8354990855712773,
+      "grad_norm": 0.11117232590913773,
+      "learning_rate": 6.955599583279837e-07,
+      "loss": 0.7037,
+      "step": 2170
+    },
+    {
+      "epoch": 0.8393493117720666,
+      "grad_norm": 0.10840250551700592,
+      "learning_rate": 6.641681609246981e-07,
+      "loss": 0.695,
+      "step": 2180
+    },
+    {
+      "epoch": 0.8431995379728559,
+      "grad_norm": 0.10846778750419617,
+      "learning_rate": 6.334507521075711e-07,
+      "loss": 0.7029,
+      "step": 2190
+    },
+    {
+      "epoch": 0.8470497641736452,
+      "grad_norm": 0.11735308170318604,
+      "learning_rate": 6.034125096149179e-07,
+      "loss": 0.6909,
+      "step": 2200
+    },
+    {
+      "epoch": 0.8508999903744345,
+      "grad_norm": 0.12722741067409515,
+      "learning_rate": 5.740581055485722e-07,
+      "loss": 0.6747,
+      "step": 2210
+    },
+    {
+      "epoch": 0.8547502165752238,
+      "grad_norm": 0.10288499295711517,
+      "learning_rate": 5.453921056471961e-07,
+      "loss": 0.697,
+      "step": 2220
+    },
+    {
+      "epoch": 0.8586004427760131,
+      "grad_norm": 0.1153920367360115,
+      "learning_rate": 5.174189685761288e-07,
+      "loss": 0.6918,
+      "step": 2230
+    },
+    {
+      "epoch": 0.8624506689768023,
+      "grad_norm": 0.10736320167779922,
+      "learning_rate": 4.901430452338979e-07,
+      "loss": 0.7108,
+      "step": 2240
+    },
+    {
+      "epoch": 0.8663008951775917,
+      "grad_norm": 0.11054076999425888,
+      "learning_rate": 4.6356857807548286e-07,
+      "loss": 0.691,
+      "step": 2250
+    },
+    {
+      "epoch": 0.870151121378381,
+      "grad_norm": 0.11257917433977127,
+      "learning_rate": 4.376997004524541e-07,
+      "loss": 0.6804,
+      "step": 2260
+    },
+    {
+      "epoch": 0.8740013475791703,
+      "grad_norm": 0.10684496909379959,
+      "learning_rate": 4.12540435970073e-07,
+      "loss": 0.68,
+      "step": 2270
+    },
+    {
+      "epoch": 0.8778515737799596,
+      "grad_norm": 0.11061599105596542,
+      "learning_rate": 3.880946978614703e-07,
+      "loss": 0.7037,
+      "step": 2280
+    },
+    {
+      "epoch": 0.8817017999807488,
+      "grad_norm": 0.10692539066076279,
+      "learning_rate": 3.643662883789878e-07,
+      "loss": 0.6991,
+      "step": 2290
+    },
+    {
+      "epoch": 0.8855520261815382,
+      "grad_norm": 0.10562099516391754,
+      "learning_rate": 3.41358898202781e-07,
+      "loss": 0.6848,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8894022523823275,
+      "grad_norm": 0.10146348923444748,
+      "learning_rate": 3.190761058667741e-07,
+      "loss": 0.6879,
+      "step": 2310
+    },
+    {
+      "epoch": 0.8932524785831167,
+      "grad_norm": 0.104509137570858,
+      "learning_rate": 2.9752137720206463e-07,
+      "loss": 0.6944,
+      "step": 2320
+    },
+    {
+      "epoch": 0.8971027047839061,
+      "grad_norm": 0.11515794694423676,
+      "learning_rate": 2.7669806479785067e-07,
+      "loss": 0.7051,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9009529309846953,
+      "grad_norm": 0.10906580835580826,
+      "learning_rate": 2.566094074799763e-07,
+      "loss": 0.6943,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9048031571854847,
+      "grad_norm": 0.10929170995950699,
+      "learning_rate": 2.3725852980716557e-07,
+      "loss": 0.6844,
+      "step": 2350
+    },
+    {
+      "epoch": 0.908653383386274,
+      "grad_norm": 0.10922319442033768,
+      "learning_rate": 2.1864844158503983e-07,
+      "loss": 0.6952,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9125036095870632,
+      "grad_norm": 0.11573971062898636,
+      "learning_rate": 2.007820373979702e-07,
+      "loss": 0.6851,
+      "step": 2370
+    },
+    {
+      "epoch": 0.9163538357878526,
+      "grad_norm": 0.10849408805370331,
+      "learning_rate": 1.8366209615886532e-07,
+      "loss": 0.6911,
+      "step": 2380
+    },
+    {
+      "epoch": 0.9202040619886418,
+      "grad_norm": 0.11006024479866028,
+      "learning_rate": 1.6729128067693613e-07,
+      "loss": 0.7123,
+      "step": 2390
+    },
+    {
+      "epoch": 0.9240542881894311,
+      "grad_norm": 0.11278330534696579,
+      "learning_rate": 1.5167213724353426e-07,
+      "loss": 0.7011,
+      "step": 2400
+    },
+    {
+      "epoch": 0.9279045143902205,
+      "grad_norm": 0.11009896546602249,
+      "learning_rate": 1.3680709523610058e-07,
+      "loss": 0.6839,
+      "step": 2410
+    },
+    {
+      "epoch": 0.9317547405910097,
+      "grad_norm": 0.11090218275785446,
+      "learning_rate": 1.2269846674030418e-07,
+      "loss": 0.7111,
+      "step": 2420
+    },
+    {
+      "epoch": 0.935604966791799,
+      "grad_norm": 0.11766745150089264,
+      "learning_rate": 1.0934844619042595e-07,
+      "loss": 0.7046,
+      "step": 2430
+    },
+    {
+      "epoch": 0.9394551929925883,
+      "grad_norm": 0.11908461153507233,
+      "learning_rate": 9.675911002803961e-08,
+      "loss": 0.6932,
+      "step": 2440
+    },
+    {
+      "epoch": 0.9433054191933776,
+      "grad_norm": 0.11983766406774521,
+      "learning_rate": 8.493241637904381e-08,
+      "loss": 0.6897,
+      "step": 2450
+    },
+    {
+      "epoch": 0.9471556453941669,
+      "grad_norm": 0.1142464429140091,
+      "learning_rate": 7.387020474909689e-08,
+      "loss": 0.7109,
+      "step": 2460
+    },
+    {
+      "epoch": 0.9510058715949562,
+      "grad_norm": 0.11587635427713394,
+      "learning_rate": 6.357419573750856e-08,
+      "loss": 0.6984,
+      "step": 2470
+    },
+    {
+      "epoch": 0.9548560977957455,
+      "grad_norm": 0.10844162106513977,
+      "learning_rate": 5.4045990769613365e-08,
+      "loss": 0.6891,
+      "step": 2480
+    },
+    {
+      "epoch": 0.9587063239965348,
+      "grad_norm": 0.11132313311100006,
+      "learning_rate": 4.5287071847694434e-08,
+      "loss": 0.6897,
+      "step": 2490
+    },
+    {
+      "epoch": 0.9625565501973241,
+      "grad_norm": 0.11132436990737915,
+      "learning_rate": 3.7298801320467817e-08,
+      "loss": 0.6924,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9664067763981133,
+      "grad_norm": 0.11074534058570862,
+      "learning_rate": 3.0082421671192576e-08,
+      "loss": 0.6854,
+      "step": 2510
+    },
+    {
+      "epoch": 0.9702570025989027,
+      "grad_norm": 0.10552199929952621,
+      "learning_rate": 2.3639055324410355e-08,
+      "loss": 0.702,
+      "step": 2520
+    },
+    {
+      "epoch": 0.974107228799692,
+      "grad_norm": 0.1128070130944252,
+      "learning_rate": 1.7969704471369477e-08,
+      "loss": 0.6896,
+      "step": 2530
+    },
+    {
+      "epoch": 0.9779574550004813,
+      "grad_norm": 0.10198631882667542,
+      "learning_rate": 1.3075250914142413e-08,
+      "loss": 0.6941,
+      "step": 2540
+    },
+    {
+      "epoch": 0.9818076812012706,
+      "grad_norm": 0.10689588636159897,
+      "learning_rate": 8.956455928476049e-09,
+      "loss": 0.7043,
+      "step": 2550
+    },
+    {
+      "epoch": 0.9856579074020598,
+      "grad_norm": 0.09838446974754333,
+      "learning_rate": 5.6139601453786275e-09,
+      "loss": 0.6835,
+      "step": 2560
+    },
+    {
+      "epoch": 0.9895081336028492,
+      "grad_norm": 0.12220260500907898,
+      "learning_rate": 3.0482834514805602e-09,
+      "loss": 0.6885,
+      "step": 2570
+    },
+    {
+      "epoch": 0.9933583598036385,
+      "grad_norm": 0.11062927544116974,
+      "learning_rate": 1.2598249081713366e-09,
+      "loss": 0.6699,
+      "step": 2580
+    },
+    {
+      "epoch": 0.9972085860044277,
+      "grad_norm": 0.11020120233297348,
+      "learning_rate": 2.488626895302826e-10,
+      "loss": 0.7056,
+      "step": 2590
+    },
+    {
+      "epoch": 0.9999037443449803,
+      "step": 2597,
+      "total_flos": 3.474366201566593e+18,
+      "train_loss": 0.7163843880904194,
+      "train_runtime": 14355.3235,
+      "train_samples_per_second": 46.318,
+      "train_steps_per_second": 0.181
+    }
+  ],
+  "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": 3.474366201566593e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen25/siglip2_sm16_512/finetune/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8ce34a7eb937ea17998f2609cf49c2c94e34a69ea2f20c929f1c599e762c90c
+size 6136

continuous/qwen25/siglip2_sm16_512/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": 1152,
+  "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-SO400M-16-SigLIP2-512",
+  "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/siglip2/siglip2_sm16_512.npz",
+  "vocab_size": 151936
+}

continuous/qwen25/siglip2_sm16_512/pretrain/mm_projector.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72f35671ab56a36295891cdea92df86291ba4f3bf4ce3abbef7a5ae86bbfacc8
+size 16529464

continuous/qwen25/siglip2_sm16_512/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": 0.45372238755226135,
+      "learning_rate": 0.00032142857142857147,
+      "loss": 4.4333,
+      "step": 10
+    },
+    {
+      "epoch": 0.03669724770642202,
+      "grad_norm": 0.03948935493826866,
+      "learning_rate": 0.0006785714285714287,
+      "loss": 3.4161,
+      "step": 20
+    },
+    {
+      "epoch": 0.05504587155963303,
+      "grad_norm": 0.025701947510242462,
+      "learning_rate": 0.0009999907688176173,
+      "loss": 3.3072,
+      "step": 30
+    },
+    {
+      "epoch": 0.07339449541284404,
+      "grad_norm": 0.017654959112405777,
+      "learning_rate": 0.0009988834393115766,
+      "loss": 3.2414,
+      "step": 40
+    },
+    {
+      "epoch": 0.09174311926605505,
+      "grad_norm": 0.023506812751293182,
+      "learning_rate": 0.0009959345572940772,
+      "loss": 3.2122,
+      "step": 50
+    },
+    {
+      "epoch": 0.11009174311926606,
+      "grad_norm": 0.020618710666894913,
+      "learning_rate": 0.0009911550081155983,
+      "loss": 3.1757,
+      "step": 60
+    },
+    {
+      "epoch": 0.12844036697247707,
+      "grad_norm": 0.03247332572937012,
+      "learning_rate": 0.0009845624347566433,
+      "loss": 3.1284,
+      "step": 70
+    },
+    {
+      "epoch": 0.14678899082568808,
+      "grad_norm": 0.03148689493536949,
+      "learning_rate": 0.0009761811727013547,
+      "loss": 3.0784,
+      "step": 80
+    },
+    {
+      "epoch": 0.1651376146788991,
+      "grad_norm": 0.030381716787815094,
+      "learning_rate": 0.0009660421601067666,
+      "loss": 3.0207,
+      "step": 90
+    },
+    {
+      "epoch": 0.1834862385321101,
+      "grad_norm": 0.02760123647749424,
+      "learning_rate": 0.0009541828235992926,
+      "loss": 2.974,
+      "step": 100
+    },
+    {
+      "epoch": 0.2018348623853211,
+      "grad_norm": 0.02657889388501644,
+      "learning_rate": 0.0009406469401200151,
+      "loss": 2.9175,
+      "step": 110
+    },
+    {
+      "epoch": 0.22018348623853212,
+      "grad_norm": 0.03149131312966347,
+      "learning_rate": 0.0009254844753287492,
+      "loss": 2.8867,
+      "step": 120
+    },
+    {
+      "epoch": 0.23853211009174313,
+      "grad_norm": 0.0311084296554327,
+      "learning_rate": 0.0009087513991633924,
+      "loss": 2.8439,
+      "step": 130
+    },
+    {
+      "epoch": 0.25688073394495414,
+      "grad_norm": 0.03508621081709862,
+      "learning_rate": 0.0008905094792353917,
+      "loss": 2.8239,
+      "step": 140
+    },
+    {
+      "epoch": 0.27522935779816515,
+      "grad_norm": 0.03206424042582512,
+      "learning_rate": 0.0008708260528239789,
+      "loss": 2.8064,
+      "step": 150
+    },
+    {
+      "epoch": 0.29357798165137616,
+      "grad_norm": 0.029012400656938553,
+      "learning_rate": 0.0008497737783108237,
+      "loss": 2.779,
+      "step": 160
+    },
+    {
+      "epoch": 0.3119266055045872,
+      "grad_norm": 0.030429657548666,
+      "learning_rate": 0.0008274303669726426,
+      "loss": 2.7658,
+      "step": 170
+    },
+    {
+      "epoch": 0.3302752293577982,
+      "grad_norm": 0.02562856115400791,
+      "learning_rate": 0.0008038782961218137,
+      "loss": 2.7479,
+      "step": 180
+    },
+    {
+      "epoch": 0.3486238532110092,
+      "grad_norm": 0.029610559344291687,
+      "learning_rate": 0.000779204504653894,
+      "loss": 2.7364,
+      "step": 190
+    },
+    {
+      "epoch": 0.3669724770642202,
+      "grad_norm": 0.024697095155715942,
+      "learning_rate": 0.0007535000721258791,
+      "loss": 2.7067,
+      "step": 200
+    },
+    {
+      "epoch": 0.3853211009174312,
+      "grad_norm": 0.02923327498137951,
+      "learning_rate": 0.00072685988254984,
+      "loss": 2.696,
+      "step": 210
+    },
+    {
+      "epoch": 0.4036697247706422,
+      "grad_norm": 0.025054287165403366,
+      "learning_rate": 0.0006993822741429906,
+      "loss": 2.7053,
+      "step": 220
+    },
+    {
+      "epoch": 0.42201834862385323,
+      "grad_norm": 0.023383552208542824,
+      "learning_rate": 0.0006711686763270818,
+      "loss": 2.6898,
+      "step": 230
+    },
+    {
+      "epoch": 0.44036697247706424,
+      "grad_norm": 0.022852502763271332,
+      "learning_rate": 0.0006423232353170831,
+      "loss": 2.6749,
+      "step": 240
+    },
+    {
+      "epoch": 0.45871559633027525,
+      "grad_norm": 0.028149953112006187,
+      "learning_rate": 0.0006129524296812335,
+      "loss": 2.6603,
+      "step": 250
+    },
+    {
+      "epoch": 0.47706422018348627,
+      "grad_norm": 0.02433369867503643,
+      "learning_rate": 0.0005831646772915651,
+      "loss": 2.664,
+      "step": 260
+    },
+    {
+      "epoch": 0.4954128440366973,
+      "grad_norm": 0.02245154045522213,
+      "learning_rate": 0.0005530699351157809,
+      "loss": 2.6374,
+      "step": 270
+    },
+    {
+      "epoch": 0.5137614678899083,
+      "grad_norm": 0.024011801928281784,
+      "learning_rate": 0.0005227792933277943,
+      "loss": 2.6319,
+      "step": 280
+    },
+    {
+      "epoch": 0.5321100917431193,
+      "grad_norm": 0.02584795095026493,
+      "learning_rate": 0.0004924045652352092,
+      "loss": 2.6106,
+      "step": 290
+    },
+    {
+      "epoch": 0.5504587155963303,
+      "grad_norm": 0.02432131953537464,
+      "learning_rate": 0.0004620578745374633,
+      "loss": 2.5986,
+      "step": 300
+    },
+    {
+      "epoch": 0.5688073394495413,
+      "grad_norm": 0.02875160425901413,
+      "learning_rate": 0.0004318512414382059,
+      "loss": 2.5949,
+      "step": 310
+    },
+    {
+      "epoch": 0.5871559633027523,
+      "grad_norm": 0.025280844420194626,
+      "learning_rate": 0.0004018961691397148,
+      "loss": 2.582,
+      "step": 320
+    },
+    {
+      "epoch": 0.6055045871559633,
+      "grad_norm": 0.027783803641796112,
+      "learning_rate": 0.0003723032322457458,
+      "loss": 2.5516,
+      "step": 330
+    },
+    {
+      "epoch": 0.6238532110091743,
+      "grad_norm": 0.024526894092559814,
+      "learning_rate": 0.0003431816685921625,
+      "loss": 2.5492,
+      "step": 340
+    },
+    {
+      "epoch": 0.6422018348623854,
+      "grad_norm": 0.027640661224722862,
+      "learning_rate": 0.0003146389760120416,
+      "loss": 2.5389,
+      "step": 350
+    },
+    {
+      "epoch": 0.6605504587155964,
+      "grad_norm": 0.026792721822857857,
+      "learning_rate": 0.00028678051552373485,
+      "loss": 2.5323,
+      "step": 360
+    },
+    {
+      "epoch": 0.6788990825688074,
+      "grad_norm": 0.022797593846917152,
+      "learning_rate": 0.0002597091224066581,
+      "loss": 2.5166,
+      "step": 370
+    },
+    {
+      "epoch": 0.6972477064220184,
+      "grad_norm": 0.026700230315327644,
+      "learning_rate": 0.0002335247266004629,
+      "loss": 2.5191,
+      "step": 380
+    },
+    {
+      "epoch": 0.7155963302752294,
+      "grad_norm": 0.026006123051047325,
+      "learning_rate": 0.000208323983828828,
+      "loss": 2.5053,
+      "step": 390
+    },
+    {
+      "epoch": 0.7339449541284404,
+      "grad_norm": 0.024035144597291946,
+      "learning_rate": 0.0001841999188095224,
+      "loss": 2.4915,
+      "step": 400
+    },
+    {
+      "epoch": 0.7522935779816514,
+      "grad_norm": 0.023278338834643364,
+      "learning_rate": 0.00016124158186777676,
+      "loss": 2.4917,
+      "step": 410
+    },
+    {
+      "epoch": 0.7706422018348624,
+      "grad_norm": 0.0227171890437603,
+      "learning_rate": 0.0001395337202205211,
+      "loss": 2.4845,
+      "step": 420
+    },
+    {
+      "epoch": 0.7889908256880734,
+      "grad_norm": 0.02192404307425022,
+      "learning_rate": 0.00011915646514489292,
+      "loss": 2.4819,
+      "step": 430
+    },
+    {
+      "epoch": 0.8073394495412844,
+      "grad_norm": 0.02318575046956539,
+      "learning_rate": 0.00010018503618578818,
+      "loss": 2.4863,
+      "step": 440
+    },
+    {
+      "epoch": 0.8256880733944955,
+      "grad_norm": 0.020790375769138336,
+      "learning_rate": 8.268946349432583e-05,
+      "loss": 2.4792,
+      "step": 450
+    },
+    {
+      "epoch": 0.8440366972477065,
+      "grad_norm": 0.019769182428717613,
+      "learning_rate": 6.673432932217117e-05,
+      "loss": 2.4711,
+      "step": 460
+    },
+    {
+      "epoch": 0.8623853211009175,
+      "grad_norm": 0.02165166847407818,
+      "learning_rate": 5.2378529625954684e-05,
+      "loss": 2.4618,
+      "step": 470
+    },
+    {
+      "epoch": 0.8807339449541285,
+      "grad_norm": 0.02298745885491371,
+      "learning_rate": 3.9675056661785556e-05,
+      "loss": 2.4684,
+      "step": 480
+    },
+    {
+      "epoch": 0.8990825688073395,
+      "grad_norm": 0.02049596607685089,
+      "learning_rate": 2.8670803372379538e-05,
+      "loss": 2.4744,
+      "step": 490
+    },
+    {
+      "epoch": 0.9174311926605505,
+      "grad_norm": 0.020036105066537857,
+      "learning_rate": 1.9406390288876586e-05,
+      "loss": 2.4769,
+      "step": 500
+    },
+    {
+      "epoch": 0.9357798165137615,
+      "grad_norm": 0.020878657698631287,
+      "learning_rate": 1.191601558631461e-05,
+      "loss": 2.4682,
+      "step": 510
+    },
+    {
+      "epoch": 0.9541284403669725,
+      "grad_norm": 0.02064594067633152,
+      "learning_rate": 6.2273288462559305e-06,
+      "loss": 2.464,
+      "step": 520
+    },
+    {
+      "epoch": 0.9724770642201835,
+      "grad_norm": 0.020197423174977303,
+      "learning_rate": 2.361328992552314e-06,
+      "loss": 2.4665,
+      "step": 530
+    },
+    {
+      "epoch": 0.9908256880733946,
+      "grad_norm": 0.019203364849090576,
+      "learning_rate": 3.322867770044202e-07,
+      "loss": 2.4648,
+      "step": 540
+    },
+    {
+      "epoch": 1.0,
+      "step": 545,
+      "total_flos": 3.0546476457682534e+17,
+      "train_loss": 2.730909860243491,
+      "train_runtime": 6492.2143,
+      "train_samples_per_second": 85.969,
+      "train_steps_per_second": 0.084
+    }
+  ],
+  "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": 3.0546476457682534e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen25/webssl_dino1b_full2b_224/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/webssl_dino1b_full2b_224/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/webssl_dino1b_full2b_224/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": 1536,
+  "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/webssl/webssl-dino1b-full2b-224",
+  "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/webssl_dino1b_full2b_224/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/webssl_dino1b_full2b_224/finetune/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4dd7fb4aaafea4f553fe6c3bd9f9281ac6e3c70c604da58f21ce33fb02b2f98e
+size 4979282296

continuous/qwen25/webssl_dino1b_full2b_224/finetune/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb3941d6b3c8ed79649f9a55bb9dddba48d06219df99af25864c3ffde9cb1d72
+size 863468272

continuous/qwen25/webssl_dino1b_full2b_224/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/webssl_dino1b_full2b_224/finetune/tokenizer.json

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

continuous/qwen25/webssl_dino1b_full2b_224/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
+}