lyunee commited on 18 days ago

Commit

0beafb2

verified ·

1 Parent(s): c7fa0df

Upload continuous evaluation checkpoints, batch 25

Browse files

Files changed (34) hide show

.gitattributes +2 -0
continuous/qwen3/webssl_mae1b_full2b_224/finetune/model-00001-of-00002.safetensors +3 -0
continuous/qwen3/webssl_mae1b_full2b_224/finetune/model-00002-of-00002.safetensors +3 -0
continuous/qwen3/webssl_mae1b_full2b_224/finetune/model.safetensors.index.json +0 -0
continuous/qwen3/webssl_mae1b_full2b_224/finetune/special_tokens_map.json +31 -0
continuous/qwen3/webssl_mae1b_full2b_224/finetune/tokenizer.json +3 -0
continuous/qwen3/webssl_mae1b_full2b_224/finetune/tokenizer_config.json +240 -0
continuous/qwen3/webssl_mae1b_full2b_224/finetune/trainer_state.json +1856 -0
continuous/qwen3/webssl_mae1b_full2b_224/finetune/training_args.bin +3 -0
continuous/qwen3/webssl_mae1b_full2b_224/finetune/vocab.json +0 -0
continuous/qwen3/webssl_mae1b_full2b_224/pretrain/config.json +89 -0
continuous/qwen3/webssl_mae1b_full2b_224/pretrain/mm_projector.bin +3 -0
continuous/qwen3/webssl_mae1b_full2b_224/pretrain/trainer_state.json +421 -0
continuous/qwen3/webssl_mae300m_full2b_224/finetune/added_tokens.json +28 -0
continuous/qwen3/webssl_mae300m_full2b_224/finetune/chat_template.jinja +89 -0
continuous/qwen3/webssl_mae300m_full2b_224/finetune/config.json +89 -0
continuous/qwen3/webssl_mae300m_full2b_224/finetune/generation_config.json +12 -0
continuous/qwen3/webssl_mae300m_full2b_224/finetune/merges.txt +0 -0
continuous/qwen3/webssl_mae300m_full2b_224/finetune/model.safetensors +3 -0
continuous/qwen3/webssl_mae300m_full2b_224/finetune/special_tokens_map.json +31 -0
continuous/qwen3/webssl_mae300m_full2b_224/finetune/tokenizer.json +3 -0
continuous/qwen3/webssl_mae300m_full2b_224/finetune/tokenizer_config.json +240 -0
continuous/qwen3/webssl_mae300m_full2b_224/finetune/trainer_state.json +1856 -0
continuous/qwen3/webssl_mae300m_full2b_224/finetune/training_args.bin +3 -0
continuous/qwen3/webssl_mae300m_full2b_224/finetune/vocab.json +0 -0
continuous/qwen3/webssl_mae300m_full2b_224/pretrain/config.json +89 -0
continuous/qwen3/webssl_mae300m_full2b_224/pretrain/mm_projector.bin +3 -0
continuous/qwen3/webssl_mae300m_full2b_224/pretrain/trainer_state.json +421 -0
continuous/qwen3/webssl_mae3b_full2b_224/finetune/added_tokens.json +28 -0
continuous/qwen3/webssl_mae3b_full2b_224/finetune/chat_template.jinja +89 -0
continuous/qwen3/webssl_mae3b_full2b_224/finetune/config.json +89 -0
continuous/qwen3/webssl_mae3b_full2b_224/finetune/generation_config.json +12 -0
continuous/qwen3/webssl_mae3b_full2b_224/finetune/merges.txt +0 -0
continuous/qwen3/webssl_mae3b_full2b_224/finetune/model-00001-of-00003.safetensors +3 -0

.gitattributes CHANGED Viewed

@@ -173,3 +173,5 @@ continuous/qwen3/siglip2_sm16_256/finetune/tokenizer.json filter=lfs diff=lfs me
 continuous/qwen3/siglip2_sm16_384/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 continuous/qwen3/siglip2_sm16_512/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 continuous/qwen3/webssl_dino1b_full2b_224/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text

 continuous/qwen3/siglip2_sm16_384/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 continuous/qwen3/siglip2_sm16_512/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 continuous/qwen3/webssl_dino1b_full2b_224/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen3/webssl_mae1b_full2b_224/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen3/webssl_mae300m_full2b_224/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text

continuous/qwen3/webssl_mae1b_full2b_224/finetune/model-00001-of-00002.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:7806731a95835bed31c147fee57d0dfcb429d930767c962e58008b2756a4b1b1
+size 4992642952

continuous/qwen3/webssl_mae1b_full2b_224/finetune/model-00002-of-00002.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:024512fe738c7d4098a55d6ca5604c2a5697d6e96a0bc13b91068706dd6d3607
+size 1374197888

continuous/qwen3/webssl_mae1b_full2b_224/finetune/model.safetensors.index.json ADDED Viewed

The diff for this file is too large to render. See raw diff

continuous/qwen3/webssl_mae1b_full2b_224/finetune/special_tokens_map.json ADDED Viewed

	@@ -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/qwen3/webssl_mae1b_full2b_224/finetune/tokenizer.json ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

continuous/qwen3/webssl_mae1b_full2b_224/finetune/tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,240 @@

+{
+  "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
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 8192,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

continuous/qwen3/webssl_mae1b_full2b_224/finetune/trainer_state.json ADDED Viewed

	@@ -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": 17.7299747467041,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 2.4029,
+      "step": 10
+    },
+    {
+      "epoch": 0.007700452401578592,
+      "grad_norm": 7.755227088928223,
+      "learning_rate": 2.435897435897436e-06,
+      "loss": 1.9679,
+      "step": 20
+    },
+    {
+      "epoch": 0.01155067860236789,
+      "grad_norm": 4.285267353057861,
+      "learning_rate": 3.7179487179487184e-06,
+      "loss": 1.5153,
+      "step": 30
+    },
+    {
+      "epoch": 0.015400904803157185,
+      "grad_norm": 1.5284597873687744,
+      "learning_rate": 5e-06,
+      "loss": 1.1966,
+      "step": 40
+    },
+    {
+      "epoch": 0.019251131003946483,
+      "grad_norm": 0.852796733379364,
+      "learning_rate": 6.282051282051282e-06,
+      "loss": 1.0873,
+      "step": 50
+    },
+    {
+      "epoch": 0.02310135720473578,
+      "grad_norm": 0.4446020722389221,
+      "learning_rate": 7.564102564102564e-06,
+      "loss": 0.9988,
+      "step": 60
+    },
+    {
+      "epoch": 0.026951583405525073,
+      "grad_norm": 0.3317302465438843,
+      "learning_rate": 8.846153846153847e-06,
+      "loss": 0.9467,
+      "step": 70
+    },
+    {
+      "epoch": 0.03080180960631437,
+      "grad_norm": 0.30421674251556396,
+      "learning_rate": 9.999996111488723e-06,
+      "loss": 0.9433,
+      "step": 80
+    },
+    {
+      "epoch": 0.034652035807103666,
+      "grad_norm": 0.301008939743042,
+      "learning_rate": 9.999529497453782e-06,
+      "loss": 0.9106,
+      "step": 90
+    },
+    {
+      "epoch": 0.038502262007892966,
+      "grad_norm": 0.2635122537612915,
+      "learning_rate": 9.998285264324193e-06,
+      "loss": 0.8926,
+      "step": 100
+    },
+    {
+      "epoch": 0.04235248820868226,
+      "grad_norm": 0.26037874817848206,
+      "learning_rate": 9.996263605626052e-06,
+      "loss": 0.874,
+      "step": 110
+    },
+    {
+      "epoch": 0.04620271440947156,
+      "grad_norm": 0.2635752856731415,
+      "learning_rate": 9.99346483580503e-06,
+      "loss": 0.8844,
+      "step": 120
+    },
+    {
+      "epoch": 0.05005294061026085,
+      "grad_norm": 0.2551456093788147,
+      "learning_rate": 9.989889390177467e-06,
+      "loss": 0.8804,
+      "step": 130
+    },
+    {
+      "epoch": 0.053903166811050146,
+      "grad_norm": 0.30797356367111206,
+      "learning_rate": 9.985537824862647e-06,
+      "loss": 0.8741,
+      "step": 140
+    },
+    {
+      "epoch": 0.057753393011839446,
+      "grad_norm": 0.3106326162815094,
+      "learning_rate": 9.980410816696317e-06,
+      "loss": 0.8596,
+      "step": 150
+    },
+    {
+      "epoch": 0.06160361921262874,
+      "grad_norm": 0.30111292004585266,
+      "learning_rate": 9.97450916312541e-06,
+      "loss": 0.8648,
+      "step": 160
+    },
+    {
+      "epoch": 0.06545384541341803,
+      "grad_norm": 0.2535618841648102,
+      "learning_rate": 9.967833782084004e-06,
+      "loss": 0.8649,
+      "step": 170
+    },
+    {
+      "epoch": 0.06930407161420733,
+      "grad_norm": 0.2660275995731354,
+      "learning_rate": 9.960385711850554e-06,
+      "loss": 0.8935,
+      "step": 180
+    },
+    {
+      "epoch": 0.07315429781499663,
+      "grad_norm": 0.2638377547264099,
+      "learning_rate": 9.952166110886394e-06,
+      "loss": 0.8825,
+      "step": 190
+    },
+    {
+      "epoch": 0.07700452401578593,
+      "grad_norm": 0.2866048216819763,
+      "learning_rate": 9.943176257655567e-06,
+      "loss": 0.8731,
+      "step": 200
+    },
+    {
+      "epoch": 0.08085475021657522,
+      "grad_norm": 0.2608121931552887,
+      "learning_rate": 9.933417550425952e-06,
+      "loss": 0.8667,
+      "step": 210
+    },
+    {
+      "epoch": 0.08470497641736452,
+      "grad_norm": 0.27231666445732117,
+      "learning_rate": 9.9228915070518e-06,
+      "loss": 0.8556,
+      "step": 220
+    },
+    {
+      "epoch": 0.08855520261815382,
+      "grad_norm": 0.25046491622924805,
+      "learning_rate": 9.911599764737632e-06,
+      "loss": 0.8554,
+      "step": 230
+    },
+    {
+      "epoch": 0.09240542881894312,
+      "grad_norm": 0.255104660987854,
+      "learning_rate": 9.899544079783606e-06,
+      "loss": 0.8503,
+      "step": 240
+    },
+    {
+      "epoch": 0.0962556550197324,
+      "grad_norm": 0.30347946286201477,
+      "learning_rate": 9.886726327312338e-06,
+      "loss": 0.868,
+      "step": 250
+    },
+    {
+      "epoch": 0.1001058812205217,
+      "grad_norm": 0.25998759269714355,
+      "learning_rate": 9.87314850097724e-06,
+      "loss": 0.8648,
+      "step": 260
+    },
+    {
+      "epoch": 0.103956107421311,
+      "grad_norm": 0.2502237558364868,
+      "learning_rate": 9.858812712652452e-06,
+      "loss": 0.852,
+      "step": 270
+    },
+    {
+      "epoch": 0.10780633362210029,
+      "grad_norm": 0.26188385486602783,
+      "learning_rate": 9.843721192104341e-06,
+      "loss": 0.8521,
+      "step": 280
+    },
+    {
+      "epoch": 0.11165655982288959,
+      "grad_norm": 0.2584984004497528,
+      "learning_rate": 9.827876286644698e-06,
+      "loss": 0.8693,
+      "step": 290
+    },
+    {
+      "epoch": 0.11550678602367889,
+      "grad_norm": 0.23163898289203644,
+      "learning_rate": 9.811280460765643e-06,
+      "loss": 0.8288,
+      "step": 300
+    },
+    {
+      "epoch": 0.11935701222446819,
+      "grad_norm": 0.25573551654815674,
+      "learning_rate": 9.793936295756292e-06,
+      "loss": 0.844,
+      "step": 310
+    },
+    {
+      "epoch": 0.12320723842525748,
+      "grad_norm": 0.25987300276756287,
+      "learning_rate": 9.775846489301279e-06,
+      "loss": 0.8433,
+      "step": 320
+    },
+    {
+      "epoch": 0.12705746462604678,
+      "grad_norm": 0.25280123949050903,
+      "learning_rate": 9.757013855061157e-06,
+      "loss": 0.8434,
+      "step": 330
+    },
+    {
+      "epoch": 0.13090769082683607,
+      "grad_norm": 0.24767690896987915,
+      "learning_rate": 9.737441322234758e-06,
+      "loss": 0.8534,
+      "step": 340
+    },
+    {
+      "epoch": 0.13475791702762538,
+      "grad_norm": 0.27971237897872925,
+      "learning_rate": 9.717131935103605e-06,
+      "loss": 0.8539,
+      "step": 350
+    },
+    {
+      "epoch": 0.13860814322841467,
+      "grad_norm": 0.25154533982276917,
+      "learning_rate": 9.696088852558398e-06,
+      "loss": 0.8397,
+      "step": 360
+    },
+    {
+      "epoch": 0.14245836942920398,
+      "grad_norm": 0.24642755091190338,
+      "learning_rate": 9.674315347607686e-06,
+      "loss": 0.8673,
+      "step": 370
+    },
+    {
+      "epoch": 0.14630859562999327,
+      "grad_norm": 0.2610428035259247,
+      "learning_rate": 9.651814806868796e-06,
+      "loss": 0.8272,
+      "step": 380
+    },
+    {
+      "epoch": 0.15015882183078255,
+      "grad_norm": 0.25661370158195496,
+      "learning_rate": 9.62859073004107e-06,
+      "loss": 0.8345,
+      "step": 390
+    },
+    {
+      "epoch": 0.15400904803157187,
+      "grad_norm": 0.23866315186023712,
+      "learning_rate": 9.604646729361547e-06,
+      "loss": 0.8385,
+      "step": 400
+    },
+    {
+      "epoch": 0.15785927423236115,
+      "grad_norm": 0.23516882956027985,
+      "learning_rate": 9.579986529043099e-06,
+      "loss": 0.8266,
+      "step": 410
+    },
+    {
+      "epoch": 0.16170950043315044,
+      "grad_norm": 0.2487865537405014,
+      "learning_rate": 9.554613964695189e-06,
+      "loss": 0.8425,
+      "step": 420
+    },
+    {
+      "epoch": 0.16555972663393975,
+      "grad_norm": 0.25501614809036255,
+      "learning_rate": 9.528532982727281e-06,
+      "loss": 0.832,
+      "step": 430
+    },
+    {
+      "epoch": 0.16940995283472904,
+      "grad_norm": 0.2451484352350235,
+      "learning_rate": 9.501747639735018e-06,
+      "loss": 0.8311,
+      "step": 440
+    },
+    {
+      "epoch": 0.17326017903551832,
+      "grad_norm": 0.27020275592803955,
+      "learning_rate": 9.474262101869272e-06,
+      "loss": 0.8205,
+      "step": 450
+    },
+    {
+      "epoch": 0.17711040523630764,
+      "grad_norm": 0.2455163598060608,
+      "learning_rate": 9.446080644188141e-06,
+      "loss": 0.8255,
+      "step": 460
+    },
+    {
+      "epoch": 0.18096063143709692,
+      "grad_norm": 0.26695865392684937,
+      "learning_rate": 9.417207649992017e-06,
+      "loss": 0.8351,
+      "step": 470
+    },
+    {
+      "epoch": 0.18481085763788624,
+      "grad_norm": 0.24732646346092224,
+      "learning_rate": 9.387647610141816e-06,
+      "loss": 0.8334,
+      "step": 480
+    },
+    {
+      "epoch": 0.18866108383867553,
+      "grad_norm": 0.24370789527893066,
+      "learning_rate": 9.35740512236047e-06,
+      "loss": 0.8196,
+      "step": 490
+    },
+    {
+      "epoch": 0.1925113100394648,
+      "grad_norm": 0.23722778260707855,
+      "learning_rate": 9.326484890517809e-06,
+      "loss": 0.824,
+      "step": 500
+    },
+    {
+      "epoch": 0.19636153624025413,
+      "grad_norm": 0.24202680587768555,
+      "learning_rate": 9.294891723898926e-06,
+      "loss": 0.8202,
+      "step": 510
+    },
+    {
+      "epoch": 0.2002117624410434,
+      "grad_norm": 0.2452133297920227,
+      "learning_rate": 9.262630536456153e-06,
+      "loss": 0.8498,
+      "step": 520
+    },
+    {
+      "epoch": 0.2040619886418327,
+      "grad_norm": 0.2598040699958801,
+      "learning_rate": 9.229706346044749e-06,
+      "loss": 0.8212,
+      "step": 530
+    },
+    {
+      "epoch": 0.207912214842622,
+      "grad_norm": 0.26749783754348755,
+      "learning_rate": 9.196124273642422e-06,
+      "loss": 0.8353,
+      "step": 540
+    },
+    {
+      "epoch": 0.2117624410434113,
+      "grad_norm": 0.23363032937049866,
+      "learning_rate": 9.161889542552842e-06,
+      "loss": 0.8364,
+      "step": 550
+    },
+    {
+      "epoch": 0.21561266724420058,
+      "grad_norm": 0.22876934707164764,
+      "learning_rate": 9.127007477593193e-06,
+      "loss": 0.8423,
+      "step": 560
+    },
+    {
+      "epoch": 0.2194628934449899,
+      "grad_norm": 0.2691418528556824,
+      "learning_rate": 9.091483504265974e-06,
+      "loss": 0.8516,
+      "step": 570
+    },
+    {
+      "epoch": 0.22331311964577918,
+      "grad_norm": 0.26343607902526855,
+      "learning_rate": 9.055323147915116e-06,
+      "loss": 0.8303,
+      "step": 580
+    },
+    {
+      "epoch": 0.2271633458465685,
+      "grad_norm": 0.27397415041923523,
+      "learning_rate": 9.018532032866585e-06,
+      "loss": 0.8409,
+      "step": 590
+    },
+    {
+      "epoch": 0.23101357204735778,
+      "grad_norm": 0.2508753836154938,
+      "learning_rate": 8.981115881553579e-06,
+      "loss": 0.8328,
+      "step": 600
+    },
+    {
+      "epoch": 0.23486379824814707,
+      "grad_norm": 0.2785504162311554,
+      "learning_rate": 8.943080513626476e-06,
+      "loss": 0.8523,
+      "step": 610
+    },
+    {
+      "epoch": 0.23871402444893638,
+      "grad_norm": 0.24555262923240662,
+      "learning_rate": 8.904431845047641e-06,
+      "loss": 0.8356,
+      "step": 620
+    },
+    {
+      "epoch": 0.24256425064972567,
+      "grad_norm": 0.2479007989168167,
+      "learning_rate": 8.86517588717128e-06,
+      "loss": 0.853,
+      "step": 630
+    },
+    {
+      "epoch": 0.24641447685051496,
+      "grad_norm": 0.25234687328338623,
+      "learning_rate": 8.82531874580844e-06,
+      "loss": 0.841,
+      "step": 640
+    },
+    {
+      "epoch": 0.25026470305130427,
+      "grad_norm": 0.22737185657024384,
+      "learning_rate": 8.784866620277314e-06,
+      "loss": 0.7986,
+      "step": 650
+    },
+    {
+      "epoch": 0.25411492925209356,
+      "grad_norm": 0.26487383246421814,
+      "learning_rate": 8.743825802439018e-06,
+      "loss": 0.8357,
+      "step": 660
+    },
+    {
+      "epoch": 0.25796515545288284,
+      "grad_norm": 0.29560980200767517,
+      "learning_rate": 8.702202675718956e-06,
+      "loss": 0.8406,
+      "step": 670
+    },
+    {
+      "epoch": 0.26181538165367213,
+      "grad_norm": 0.24209658801555634,
+      "learning_rate": 8.660003714113959e-06,
+      "loss": 0.8111,
+      "step": 680
+    },
+    {
+      "epoch": 0.26566560785446147,
+      "grad_norm": 0.26490601897239685,
+      "learning_rate": 8.617235481185323e-06,
+      "loss": 0.8362,
+      "step": 690
+    },
+    {
+      "epoch": 0.26951583405525076,
+      "grad_norm": 0.24610784649848938,
+      "learning_rate": 8.573904629037931e-06,
+      "loss": 0.8403,
+      "step": 700
+    },
+    {
+      "epoch": 0.27336606025604004,
+      "grad_norm": 0.24308975040912628,
+      "learning_rate": 8.530017897285585e-06,
+      "loss": 0.8202,
+      "step": 710
+    },
+    {
+      "epoch": 0.27721628645682933,
+      "grad_norm": 0.22908444702625275,
+      "learning_rate": 8.485582112002746e-06,
+      "loss": 0.8249,
+      "step": 720
+    },
+    {
+      "epoch": 0.2810665126576186,
+      "grad_norm": 0.25292158126831055,
+      "learning_rate": 8.440604184662811e-06,
+      "loss": 0.8309,
+      "step": 730
+    },
+    {
+      "epoch": 0.28491673885840796,
+      "grad_norm": 0.23848000168800354,
+      "learning_rate": 8.395091111063118e-06,
+      "loss": 0.8195,
+      "step": 740
+    },
+    {
+      "epoch": 0.28876696505919724,
+      "grad_norm": 0.23105394840240479,
+      "learning_rate": 8.349049970236822e-06,
+      "loss": 0.818,
+      "step": 750
+    },
+    {
+      "epoch": 0.29261719125998653,
+      "grad_norm": 0.24539542198181152,
+      "learning_rate": 8.302487923351842e-06,
+      "loss": 0.8116,
+      "step": 760
+    },
+    {
+      "epoch": 0.2964674174607758,
+      "grad_norm": 0.24491305649280548,
+      "learning_rate": 8.255412212597021e-06,
+      "loss": 0.8288,
+      "step": 770
+    },
+    {
+      "epoch": 0.3003176436615651,
+      "grad_norm": 0.23482917249202728,
+      "learning_rate": 8.207830160055688e-06,
+      "loss": 0.8255,
+      "step": 780
+    },
+    {
+      "epoch": 0.3041678698623544,
+      "grad_norm": 0.24488867819309235,
+      "learning_rate": 8.159749166566786e-06,
+      "loss": 0.8108,
+      "step": 790
+    },
+    {
+      "epoch": 0.30801809606314373,
+      "grad_norm": 0.232904314994812,
+      "learning_rate": 8.11117671057377e-06,
+      "loss": 0.8383,
+      "step": 800
+    },
+    {
+      "epoch": 0.311868322263933,
+      "grad_norm": 0.24146021902561188,
+      "learning_rate": 8.062120346961404e-06,
+      "loss": 0.8311,
+      "step": 810
+    },
+    {
+      "epoch": 0.3157185484647223,
+      "grad_norm": 0.23775416612625122,
+      "learning_rate": 8.012587705880705e-06,
+      "loss": 0.8233,
+      "step": 820
+    },
+    {
+      "epoch": 0.3195687746655116,
+      "grad_norm": 0.24105529487133026,
+      "learning_rate": 7.962586491562146e-06,
+      "loss": 0.8095,
+      "step": 830
+    },
+    {
+      "epoch": 0.3234190008663009,
+      "grad_norm": 0.25394800305366516,
+      "learning_rate": 7.912124481117357e-06,
+      "loss": 0.8156,
+      "step": 840
+    },
+    {
+      "epoch": 0.3272692270670902,
+      "grad_norm": 0.2431480586528778,
+      "learning_rate": 7.861209523329494e-06,
+      "loss": 0.8254,
+      "step": 850
+    },
+    {
+      "epoch": 0.3311194532678795,
+      "grad_norm": 0.24034182727336884,
+      "learning_rate": 7.809849537432432e-06,
+      "loss": 0.8096,
+      "step": 860
+    },
+    {
+      "epoch": 0.3349696794686688,
+      "grad_norm": 0.2511649429798126,
+      "learning_rate": 7.758052511879035e-06,
+      "loss": 0.8202,
+      "step": 870
+    },
+    {
+      "epoch": 0.3388199056694581,
+      "grad_norm": 0.22481870651245117,
+      "learning_rate": 7.705826503098642e-06,
+      "loss": 0.8139,
+      "step": 880
+    },
+    {
+      "epoch": 0.34267013187024736,
+      "grad_norm": 0.22780050337314606,
+      "learning_rate": 7.653179634243979e-06,
+      "loss": 0.814,
+      "step": 890
+    },
+    {
+      "epoch": 0.34652035807103665,
+      "grad_norm": 0.22811852395534515,
+      "learning_rate": 7.600120093927696e-06,
+      "loss": 0.8212,
+      "step": 900
+    },
+    {
+      "epoch": 0.350370584271826,
+      "grad_norm": 0.2386000007390976,
+      "learning_rate": 7.546656134948729e-06,
+      "loss": 0.8106,
+      "step": 910
+    },
+    {
+      "epoch": 0.3542208104726153,
+      "grad_norm": 0.24598319828510284,
+      "learning_rate": 7.492796073008661e-06,
+      "loss": 0.8082,
+      "step": 920
+    },
+    {
+      "epoch": 0.35807103667340456,
+      "grad_norm": 0.25677958130836487,
+      "learning_rate": 7.438548285418326e-06,
+      "loss": 0.7941,
+      "step": 930
+    },
+    {
+      "epoch": 0.36192126287419385,
+      "grad_norm": 0.2536178231239319,
+      "learning_rate": 7.3839212097948e-06,
+      "loss": 0.8274,
+      "step": 940
+    },
+    {
+      "epoch": 0.36577148907498314,
+      "grad_norm": 0.24314261972904205,
+      "learning_rate": 7.328923342749039e-06,
+      "loss": 0.8436,
+      "step": 950
+    },
+    {
+      "epoch": 0.3696217152757725,
+      "grad_norm": 0.2545775771141052,
+      "learning_rate": 7.273563238564319e-06,
+      "loss": 0.8274,
+      "step": 960
+    },
+    {
+      "epoch": 0.37347194147656176,
+      "grad_norm": 0.24929113686084747,
+      "learning_rate": 7.217849507865724e-06,
+      "loss": 0.8092,
+      "step": 970
+    },
+    {
+      "epoch": 0.37732216767735105,
+      "grad_norm": 0.23766756057739258,
+      "learning_rate": 7.161790816280858e-06,
+      "loss": 0.8128,
+      "step": 980
+    },
+    {
+      "epoch": 0.38117239387814034,
+      "grad_norm": 0.2513536810874939,
+      "learning_rate": 7.1053958830920125e-06,
+      "loss": 0.8006,
+      "step": 990
+    },
+    {
+      "epoch": 0.3850226200789296,
+      "grad_norm": 0.24234692752361298,
+      "learning_rate": 7.0486734798799715e-06,
+      "loss": 0.8157,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3888728462797189,
+      "grad_norm": 0.23715092241764069,
+      "learning_rate": 6.991632429159708e-06,
+      "loss": 0.8384,
+      "step": 1010
+    },
+    {
+      "epoch": 0.39272307248050825,
+      "grad_norm": 0.27471646666526794,
+      "learning_rate": 6.934281603008132e-06,
+      "loss": 0.8315,
+      "step": 1020
+    },
+    {
+      "epoch": 0.39657329868129754,
+      "grad_norm": 0.24337616562843323,
+      "learning_rate": 6.876629921684143e-06,
+      "loss": 0.828,
+      "step": 1030
+    },
+    {
+      "epoch": 0.4004235248820868,
+      "grad_norm": 0.24312050640583038,
+      "learning_rate": 6.818686352241199e-06,
+      "loss": 0.8192,
+      "step": 1040
+    },
+    {
+      "epoch": 0.4042737510828761,
+      "grad_norm": 0.22770826518535614,
+      "learning_rate": 6.760459907132573e-06,
+      "loss": 0.8157,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4081239772836654,
+      "grad_norm": 0.2282528281211853,
+      "learning_rate": 6.701959642809589e-06,
+      "loss": 0.8222,
+      "step": 1060
+    },
+    {
+      "epoch": 0.41197420348445474,
+      "grad_norm": 0.23836731910705566,
+      "learning_rate": 6.643194658312987e-06,
+      "loss": 0.8148,
+      "step": 1070
+    },
+    {
+      "epoch": 0.415824429685244,
+      "grad_norm": 0.23913156986236572,
+      "learning_rate": 6.584174093857676e-06,
+      "loss": 0.8387,
+      "step": 1080
+    },
+    {
+      "epoch": 0.4196746558860333,
+      "grad_norm": 0.27235662937164307,
+      "learning_rate": 6.5249071294110735e-06,
+      "loss": 0.8175,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4235248820868226,
+      "grad_norm": 0.2524068057537079,
+      "learning_rate": 6.465402983265274e-06,
+      "loss": 0.8324,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4273751082876119,
+      "grad_norm": 0.24733808636665344,
+      "learning_rate": 6.405670910603243e-06,
+      "loss": 0.829,
+      "step": 1110
+    },
+    {
+      "epoch": 0.43122533448840117,
+      "grad_norm": 0.27174633741378784,
+      "learning_rate": 6.345720202059291e-06,
+      "loss": 0.8241,
+      "step": 1120
+    },
+    {
+      "epoch": 0.4350755606891905,
+      "grad_norm": 0.2472694367170334,
+      "learning_rate": 6.285560182274008e-06,
+      "loss": 0.8032,
+      "step": 1130
+    },
+    {
+      "epoch": 0.4389257868899798,
+      "grad_norm": 0.2384587973356247,
+      "learning_rate": 6.225200208443936e-06,
+      "loss": 0.825,
+      "step": 1140
+    },
+    {
+      "epoch": 0.4427760130907691,
+      "grad_norm": 0.22792819142341614,
+      "learning_rate": 6.16464966886616e-06,
+      "loss": 0.8159,
+      "step": 1150
+    },
+    {
+      "epoch": 0.44662623929155837,
+      "grad_norm": 0.2761196494102478,
+      "learning_rate": 6.103917981478065e-06,
+      "loss": 0.8382,
+      "step": 1160
+    },
+    {
+      "epoch": 0.45047646549234766,
+      "grad_norm": 0.24022290110588074,
+      "learning_rate": 6.043014592392489e-06,
+      "loss": 0.8505,
+      "step": 1170
+    },
+    {
+      "epoch": 0.454326691693137,
+      "grad_norm": 0.2300158143043518,
+      "learning_rate": 5.981948974428485e-06,
+      "loss": 0.8238,
+      "step": 1180
+    },
+    {
+      "epoch": 0.4581769178939263,
+      "grad_norm": 0.25429990887641907,
+      "learning_rate": 5.920730625637934e-06,
+      "loss": 0.805,
+      "step": 1190
+    },
+    {
+      "epoch": 0.46202714409471557,
+      "grad_norm": 0.2524241507053375,
+      "learning_rate": 5.8593690678282355e-06,
+      "loss": 0.8138,
+      "step": 1200
+    },
+    {
+      "epoch": 0.46587737029550486,
+      "grad_norm": 0.25416943430900574,
+      "learning_rate": 5.797873845081301e-06,
+      "loss": 0.8177,
+      "step": 1210
+    },
+    {
+      "epoch": 0.46972759649629414,
+      "grad_norm": 0.2439238578081131,
+      "learning_rate": 5.736254522269075e-06,
+      "loss": 0.8348,
+      "step": 1220
+    },
+    {
+      "epoch": 0.47357782269708343,
+      "grad_norm": 0.23820504546165466,
+      "learning_rate": 5.6745206835658386e-06,
+      "loss": 0.8353,
+      "step": 1230
+    },
+    {
+      "epoch": 0.47742804889787277,
+      "grad_norm": 0.2371950000524521,
+      "learning_rate": 5.612681930957495e-06,
+      "loss": 0.8065,
+      "step": 1240
+    },
+    {
+      "epoch": 0.48127827509866206,
+      "grad_norm": 0.2348092496395111,
+      "learning_rate": 5.550747882748095e-06,
+      "loss": 0.8063,
+      "step": 1250
+    },
+    {
+      "epoch": 0.48512850129945134,
+      "grad_norm": 0.25810694694519043,
+      "learning_rate": 5.4887281720638195e-06,
+      "loss": 0.811,
+      "step": 1260
+    },
+    {
+      "epoch": 0.48897872750024063,
+      "grad_norm": 0.2651573419570923,
+      "learning_rate": 5.42663244535466e-06,
+      "loss": 0.7998,
+      "step": 1270
+    },
+    {
+      "epoch": 0.4928289537010299,
+      "grad_norm": 0.26699432730674744,
+      "learning_rate": 5.3644703608940205e-06,
+      "loss": 0.8052,
+      "step": 1280
+    },
+    {
+      "epoch": 0.49667917990181926,
+      "grad_norm": 0.2504430413246155,
+      "learning_rate": 5.302251587276491e-06,
+      "loss": 0.817,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5005294061026085,
+      "grad_norm": 0.229568213224411,
+      "learning_rate": 5.2399858019140005e-06,
+      "loss": 0.8165,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5043796323033978,
+      "grad_norm": 0.23843039572238922,
+      "learning_rate": 5.177682689530615e-06,
+      "loss": 0.8312,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5082298585041871,
+      "grad_norm": 0.24708589911460876,
+      "learning_rate": 5.11535194065619e-06,
+      "loss": 0.799,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5120800847049765,
+      "grad_norm": 0.25791046023368835,
+      "learning_rate": 5.053003250119113e-06,
+      "loss": 0.8111,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5159303109057657,
+      "grad_norm": 0.24517269432544708,
+      "learning_rate": 4.9906463155383925e-06,
+      "loss": 0.817,
+      "step": 1340
+    },
+    {
+      "epoch": 0.519780537106555,
+      "grad_norm": 0.23399914801120758,
+      "learning_rate": 4.9282908358153046e-06,
+      "loss": 0.8074,
+      "step": 1350
+    },
+    {
+      "epoch": 0.5236307633073443,
+      "grad_norm": 0.2689371705055237,
+      "learning_rate": 4.8659465096248345e-06,
+      "loss": 0.8266,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5274809895081336,
+      "grad_norm": 0.26617351174354553,
+      "learning_rate": 4.803623033907169e-06,
+      "loss": 0.8269,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5313312157089229,
+      "grad_norm": 0.23417368531227112,
+      "learning_rate": 4.741330102359441e-06,
+      "loss": 0.8305,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5351814419097122,
+      "grad_norm": 0.2509380877017975,
+      "learning_rate": 4.679077403927996e-06,
+      "loss": 0.8114,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5390316681105015,
+      "grad_norm": 0.2372235357761383,
+      "learning_rate": 4.616874621301373e-06,
+      "loss": 0.8098,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5428818943112907,
+      "grad_norm": 0.2904515564441681,
+      "learning_rate": 4.554731429404293e-06,
+      "loss": 0.8254,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5467321205120801,
+      "grad_norm": 0.24489352107048035,
+      "learning_rate": 4.492657493892822e-06,
+      "loss": 0.8112,
+      "step": 1420
+    },
+    {
+      "epoch": 0.5505823467128694,
+      "grad_norm": 0.26507899165153503,
+      "learning_rate": 4.430662469650999e-06,
+      "loss": 0.8261,
+      "step": 1430
+    },
+    {
+      "epoch": 0.5544325729136587,
+      "grad_norm": 0.28443101048469543,
+      "learning_rate": 4.368755999289118e-06,
+      "loss": 0.8274,
+      "step": 1440
+    },
+    {
+      "epoch": 0.558282799114448,
+      "grad_norm": 0.27544644474983215,
+      "learning_rate": 4.306947711643949e-06,
+      "loss": 0.8144,
+      "step": 1450
+    },
+    {
+      "epoch": 0.5621330253152372,
+      "grad_norm": 0.24430562555789948,
+      "learning_rate": 4.245247220281068e-06,
+      "loss": 0.8158,
+      "step": 1460
+    },
+    {
+      "epoch": 0.5659832515160266,
+      "grad_norm": 0.23883377015590668,
+      "learning_rate": 4.183664121999587e-06,
+      "loss": 0.8147,
+      "step": 1470
+    },
+    {
+      "epoch": 0.5698334777168159,
+      "grad_norm": 0.2632313370704651,
+      "learning_rate": 4.122207995339474e-06,
+      "loss": 0.8089,
+      "step": 1480
+    },
+    {
+      "epoch": 0.5736837039176051,
+      "grad_norm": 0.2622612416744232,
+      "learning_rate": 4.060888399091734e-06,
+      "loss": 0.8165,
+      "step": 1490
+    },
+    {
+      "epoch": 0.5775339301183945,
+      "grad_norm": 0.23774294555187225,
+      "learning_rate": 3.999714870811639e-06,
+      "loss": 0.8222,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5813841563191837,
+      "grad_norm": 0.23753729462623596,
+      "learning_rate": 3.938696925335275e-06,
+      "loss": 0.8171,
+      "step": 1510
+    },
+    {
+      "epoch": 0.5852343825199731,
+      "grad_norm": 0.24273641407489777,
+      "learning_rate": 3.87784405329962e-06,
+      "loss": 0.8249,
+      "step": 1520
+    },
+    {
+      "epoch": 0.5890846087207623,
+      "grad_norm": 0.24283872544765472,
+      "learning_rate": 3.817165719666382e-06,
+      "loss": 0.8274,
+      "step": 1530
+    },
+    {
+      "epoch": 0.5929348349215516,
+      "grad_norm": 0.23352229595184326,
+      "learning_rate": 3.756671362249841e-06,
+      "loss": 0.8304,
+      "step": 1540
+    },
+    {
+      "epoch": 0.596785061122341,
+      "grad_norm": 0.23812121152877808,
+      "learning_rate": 3.6963703902488907e-06,
+      "loss": 0.8172,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6006352873231302,
+      "grad_norm": 0.23463571071624756,
+      "learning_rate": 3.6362721827835655e-06,
+      "loss": 0.8375,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6044855135239195,
+      "grad_norm": 0.2516980469226837,
+      "learning_rate": 3.576386087436212e-06,
+      "loss": 0.8301,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6083357397247088,
+      "grad_norm": 0.26430025696754456,
+      "learning_rate": 3.5167214187975913e-06,
+      "loss": 0.819,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6121859659254981,
+      "grad_norm": 0.21820878982543945,
+      "learning_rate": 3.457287457018086e-06,
+      "loss": 0.8218,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6160361921262875,
+      "grad_norm": 0.23315395414829254,
+      "learning_rate": 3.3980934463643003e-06,
+      "loss": 0.8095,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6198864183270767,
+      "grad_norm": 0.2232154905796051,
+      "learning_rate": 3.3391485937812053e-06,
+      "loss": 0.7918,
+      "step": 1610
+    },
+    {
+      "epoch": 0.623736644527866,
+      "grad_norm": 0.23934496939182281,
+      "learning_rate": 3.280462067460115e-06,
+      "loss": 0.791,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6275868707286553,
+      "grad_norm": 0.2464590072631836,
+      "learning_rate": 3.222042995412669e-06,
+      "loss": 0.8214,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6314370969294446,
+      "grad_norm": 0.22308507561683655,
+      "learning_rate": 3.163900464051089e-06,
+      "loss": 0.8111,
+      "step": 1640
+    },
+    {
+      "epoch": 0.635287323130234,
+      "grad_norm": 0.25184035301208496,
+      "learning_rate": 3.1060435167748826e-06,
+      "loss": 0.8202,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6391375493310232,
+      "grad_norm": 0.23813298344612122,
+      "learning_rate": 3.048481152564253e-06,
+      "loss": 0.8005,
+      "step": 1660
+    },
+    {
+      "epoch": 0.6429877755318125,
+      "grad_norm": 0.24741585552692413,
+      "learning_rate": 2.9912223245804017e-06,
+      "loss": 0.8222,
+      "step": 1670
+    },
+    {
+      "epoch": 0.6468380017326018,
+      "grad_norm": 0.25420695543289185,
+      "learning_rate": 2.934275938772981e-06,
+      "loss": 0.8209,
+      "step": 1680
+    },
+    {
+      "epoch": 0.6506882279333911,
+      "grad_norm": 0.28242963552474976,
+      "learning_rate": 2.8776508524948676e-06,
+      "loss": 0.8092,
+      "step": 1690
+    },
+    {
+      "epoch": 0.6545384541341804,
+      "grad_norm": 0.22491756081581116,
+      "learning_rate": 2.821355873124503e-06,
+      "loss": 0.8088,
+      "step": 1700
+    },
+    {
+      "epoch": 0.6583886803349697,
+      "grad_norm": 0.25639036297798157,
+      "learning_rate": 2.7653997566960144e-06,
+      "loss": 0.8223,
+      "step": 1710
+    },
+    {
+      "epoch": 0.662238906535759,
+      "grad_norm": 0.2528401017189026,
+      "learning_rate": 2.709791206537309e-06,
+      "loss": 0.8115,
+      "step": 1720
+    },
+    {
+      "epoch": 0.6660891327365482,
+      "grad_norm": 0.24089139699935913,
+      "learning_rate": 2.6545388719163747e-06,
+      "loss": 0.8201,
+      "step": 1730
+    },
+    {
+      "epoch": 0.6699393589373376,
+      "grad_norm": 0.23755860328674316,
+      "learning_rate": 2.599651346695979e-06,
+      "loss": 0.8164,
+      "step": 1740
+    },
+    {
+      "epoch": 0.6737895851381268,
+      "grad_norm": 0.24957744777202606,
+      "learning_rate": 2.545137167997007e-06,
+      "loss": 0.8123,
+      "step": 1750
+    },
+    {
+      "epoch": 0.6776398113389162,
+      "grad_norm": 0.23540356755256653,
+      "learning_rate": 2.4910048148705972e-06,
+      "loss": 0.8013,
+      "step": 1760
+    },
+    {
+      "epoch": 0.6814900375397055,
+      "grad_norm": 0.25638389587402344,
+      "learning_rate": 2.4372627069793327e-06,
+      "loss": 0.8433,
+      "step": 1770
+    },
+    {
+      "epoch": 0.6853402637404947,
+      "grad_norm": 0.23718877136707306,
+      "learning_rate": 2.3839192032876486e-06,
+      "loss": 0.8306,
+      "step": 1780
+    },
+    {
+      "epoch": 0.6891904899412841,
+      "grad_norm": 0.24170996248722076,
+      "learning_rate": 2.3309826007617047e-06,
+      "loss": 0.8411,
+      "step": 1790
+    },
+    {
+      "epoch": 0.6930407161420733,
+      "grad_norm": 0.23866517841815948,
+      "learning_rate": 2.278461133078881e-06,
+      "loss": 0.8044,
+      "step": 1800
+    },
+    {
+      "epoch": 0.6968909423428626,
+      "grad_norm": 0.24224691092967987,
+      "learning_rate": 2.2263629693471258e-06,
+      "loss": 0.796,
+      "step": 1810
+    },
+    {
+      "epoch": 0.700741168543652,
+      "grad_norm": 0.23219500482082367,
+      "learning_rate": 2.1746962128343414e-06,
+      "loss": 0.8211,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7045913947444412,
+      "grad_norm": 0.2389829009771347,
+      "learning_rate": 2.1234688997080183e-06,
+      "loss": 0.8087,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7084416209452306,
+      "grad_norm": 0.2574862837791443,
+      "learning_rate": 2.0726889977853053e-06,
+      "loss": 0.8129,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7122918471460198,
+      "grad_norm": 0.255316823720932,
+      "learning_rate": 2.022364405293703e-06,
+      "loss": 0.8258,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7161420733468091,
+      "grad_norm": 0.23428303003311157,
+      "learning_rate": 1.9725029496425814e-06,
+      "loss": 0.8179,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7199922995475985,
+      "grad_norm": 0.2455034852027893,
+      "learning_rate": 1.923112386205727e-06,
+      "loss": 0.8157,
+      "step": 1870
+    },
+    {
+      "epoch": 0.7238425257483877,
+      "grad_norm": 0.22841224074363708,
+      "learning_rate": 1.874200397115079e-06,
+      "loss": 0.8222,
+      "step": 1880
+    },
+    {
+      "epoch": 0.727692751949177,
+      "grad_norm": 0.25485020875930786,
+      "learning_rate": 1.825774590065857e-06,
+      "loss": 0.8073,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7315429781499663,
+      "grad_norm": 0.2465735524892807,
+      "learning_rate": 1.7778424971332813e-06,
+      "loss": 0.8138,
+      "step": 1900
+    },
+    {
+      "epoch": 0.7353932043507556,
+      "grad_norm": 0.238421231508255,
+      "learning_rate": 1.7304115736010407e-06,
+      "loss": 0.8132,
+      "step": 1910
+    },
+    {
+      "epoch": 0.739243430551545,
+      "grad_norm": 0.24235007166862488,
+      "learning_rate": 1.6834891968017108e-06,
+      "loss": 0.8047,
+      "step": 1920
+    },
+    {
+      "epoch": 0.7430936567523342,
+      "grad_norm": 0.24931269884109497,
+      "learning_rate": 1.6370826649692884e-06,
+      "loss": 0.8271,
+      "step": 1930
+    },
+    {
+      "epoch": 0.7469438829531235,
+      "grad_norm": 0.23852121829986572,
+      "learning_rate": 1.5911991961040463e-06,
+      "loss": 0.8316,
+      "step": 1940
+    },
+    {
+      "epoch": 0.7507941091539128,
+      "grad_norm": 0.25796014070510864,
+      "learning_rate": 1.545845926849851e-06,
+      "loss": 0.8162,
+      "step": 1950
+    },
+    {
+      "epoch": 0.7546443353547021,
+      "grad_norm": 0.26111653447151184,
+      "learning_rate": 1.5010299113841397e-06,
+      "loss": 0.8359,
+      "step": 1960
+    },
+    {
+      "epoch": 0.7584945615554913,
+      "grad_norm": 0.24633993208408356,
+      "learning_rate": 1.4567581203207238e-06,
+      "loss": 0.8233,
+      "step": 1970
+    },
+    {
+      "epoch": 0.7623447877562807,
+      "grad_norm": 0.2571364939212799,
+      "learning_rate": 1.413037439625593e-06,
+      "loss": 0.8239,
+      "step": 1980
+    },
+    {
+      "epoch": 0.76619501395707,
+      "grad_norm": 0.2614874243736267,
+      "learning_rate": 1.3698746695458847e-06,
+      "loss": 0.8158,
+      "step": 1990
+    },
+    {
+      "epoch": 0.7700452401578592,
+      "grad_norm": 0.2391175925731659,
+      "learning_rate": 1.3272765235521745e-06,
+      "loss": 0.8087,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7738954663586486,
+      "grad_norm": 0.224213644862175,
+      "learning_rate": 1.2852496272942877e-06,
+      "loss": 0.8148,
+      "step": 2010
+    },
+    {
+      "epoch": 0.7777456925594378,
+      "grad_norm": 0.22919270396232605,
+      "learning_rate": 1.2438005175707396e-06,
+      "loss": 0.8043,
+      "step": 2020
+    },
+    {
+      "epoch": 0.7815959187602272,
+      "grad_norm": 0.238815575838089,
+      "learning_rate": 1.2029356413120297e-06,
+      "loss": 0.7912,
+      "step": 2030
+    },
+    {
+      "epoch": 0.7854461449610165,
+      "grad_norm": 0.23838470876216888,
+      "learning_rate": 1.1626613545778786e-06,
+      "loss": 0.8078,
+      "step": 2040
+    },
+    {
+      "epoch": 0.7892963711618057,
+      "grad_norm": 0.22569675743579865,
+      "learning_rate": 1.1229839215686294e-06,
+      "loss": 0.812,
+      "step": 2050
+    },
+    {
+      "epoch": 0.7931465973625951,
+      "grad_norm": 0.22560414671897888,
+      "learning_rate": 1.0839095136509188e-06,
+      "loss": 0.7992,
+      "step": 2060
+    },
+    {
+      "epoch": 0.7969968235633843,
+      "grad_norm": 0.25999343395233154,
+      "learning_rate": 1.045444208397791e-06,
+      "loss": 0.8211,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8008470497641736,
+      "grad_norm": 0.23496925830841064,
+      "learning_rate": 1.007593988643402e-06,
+      "loss": 0.799,
+      "step": 2080
+    },
+    {
+      "epoch": 0.804697275964963,
+      "grad_norm": 0.23723725974559784,
+      "learning_rate": 9.703647415524613e-07,
+      "loss": 0.8215,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8085475021657522,
+      "grad_norm": 0.23337087035179138,
+      "learning_rate": 9.337622577045485e-07,
+      "loss": 0.8108,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8123977283665416,
+      "grad_norm": 0.23663261532783508,
+      "learning_rate": 8.977922301934588e-07,
+      "loss": 0.8173,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8162479545673308,
+      "grad_norm": 0.23756198585033417,
+      "learning_rate": 8.624602537417009e-07,
+      "loss": 0.82,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8200981807681201,
+      "grad_norm": 0.2486618310213089,
+      "learning_rate": 8.277718238303123e-07,
+      "loss": 0.8093,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8239484069689095,
+      "grad_norm": 0.24207133054733276,
+      "learning_rate": 7.937323358440935e-07,
+      "loss": 0.8118,
+      "step": 2140
+    },
+    {
+      "epoch": 0.8277986331696987,
+      "grad_norm": 0.25453847646713257,
+      "learning_rate": 7.603470842324223e-07,
+      "loss": 0.8054,
+      "step": 2150
+    },
+    {
+      "epoch": 0.831648859370488,
+      "grad_norm": 0.22955237329006195,
+      "learning_rate": 7.276212616857575e-07,
+      "loss": 0.8213,
+      "step": 2160
+    },
+    {
+      "epoch": 0.8354990855712773,
+      "grad_norm": 0.2256641536951065,
+      "learning_rate": 6.955599583279837e-07,
+      "loss": 0.8236,
+      "step": 2170
+    },
+    {
+      "epoch": 0.8393493117720666,
+      "grad_norm": 0.2519168555736542,
+      "learning_rate": 6.641681609246981e-07,
+      "loss": 0.8121,
+      "step": 2180
+    },
+    {
+      "epoch": 0.8431995379728559,
+      "grad_norm": 0.23200909793376923,
+      "learning_rate": 6.334507521075711e-07,
+      "loss": 0.823,
+      "step": 2190
+    },
+    {
+      "epoch": 0.8470497641736452,
+      "grad_norm": 0.2588474452495575,
+      "learning_rate": 6.034125096149179e-07,
+      "loss": 0.8137,
+      "step": 2200
+    },
+    {
+      "epoch": 0.8508999903744345,
+      "grad_norm": 0.2450336515903473,
+      "learning_rate": 5.740581055485722e-07,
+      "loss": 0.7858,
+      "step": 2210
+    },
+    {
+      "epoch": 0.8547502165752238,
+      "grad_norm": 0.2300826758146286,
+      "learning_rate": 5.453921056471961e-07,
+      "loss": 0.8208,
+      "step": 2220
+    },
+    {
+      "epoch": 0.8586004427760131,
+      "grad_norm": 0.2448449432849884,
+      "learning_rate": 5.174189685761288e-07,
+      "loss": 0.8125,
+      "step": 2230
+    },
+    {
+      "epoch": 0.8624506689768023,
+      "grad_norm": 0.24262696504592896,
+      "learning_rate": 4.901430452338979e-07,
+      "loss": 0.8241,
+      "step": 2240
+    },
+    {
+      "epoch": 0.8663008951775917,
+      "grad_norm": 0.23302185535430908,
+      "learning_rate": 4.6356857807548286e-07,
+      "loss": 0.8094,
+      "step": 2250
+    },
+    {
+      "epoch": 0.870151121378381,
+      "grad_norm": 0.24784019589424133,
+      "learning_rate": 4.376997004524541e-07,
+      "loss": 0.7982,
+      "step": 2260
+    },
+    {
+      "epoch": 0.8740013475791703,
+      "grad_norm": 0.2363031804561615,
+      "learning_rate": 4.12540435970073e-07,
+      "loss": 0.8108,
+      "step": 2270
+    },
+    {
+      "epoch": 0.8778515737799596,
+      "grad_norm": 0.23191504180431366,
+      "learning_rate": 3.880946978614703e-07,
+      "loss": 0.8165,
+      "step": 2280
+    },
+    {
+      "epoch": 0.8817017999807488,
+      "grad_norm": 0.2358606904745102,
+      "learning_rate": 3.643662883789878e-07,
+      "loss": 0.8249,
+      "step": 2290
+    },
+    {
+      "epoch": 0.8855520261815382,
+      "grad_norm": 0.22109848260879517,
+      "learning_rate": 3.41358898202781e-07,
+      "loss": 0.8003,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8894022523823275,
+      "grad_norm": 0.22310829162597656,
+      "learning_rate": 3.190761058667741e-07,
+      "loss": 0.8111,
+      "step": 2310
+    },
+    {
+      "epoch": 0.8932524785831167,
+      "grad_norm": 0.23296517133712769,
+      "learning_rate": 2.9752137720206463e-07,
+      "loss": 0.8191,
+      "step": 2320
+    },
+    {
+      "epoch": 0.8971027047839061,
+      "grad_norm": 0.23626050353050232,
+      "learning_rate": 2.7669806479785067e-07,
+      "loss": 0.8226,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9009529309846953,
+      "grad_norm": 0.23994319140911102,
+      "learning_rate": 2.566094074799763e-07,
+      "loss": 0.8181,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9048031571854847,
+      "grad_norm": 0.23575924336910248,
+      "learning_rate": 2.3725852980716557e-07,
+      "loss": 0.797,
+      "step": 2350
+    },
+    {
+      "epoch": 0.908653383386274,
+      "grad_norm": 0.24005688726902008,
+      "learning_rate": 2.1864844158503983e-07,
+      "loss": 0.8104,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9125036095870632,
+      "grad_norm": 0.2682107388973236,
+      "learning_rate": 2.007820373979702e-07,
+      "loss": 0.8031,
+      "step": 2370
+    },
+    {
+      "epoch": 0.9163538357878526,
+      "grad_norm": 0.24264788627624512,
+      "learning_rate": 1.8366209615886532e-07,
+      "loss": 0.8115,
+      "step": 2380
+    },
+    {
+      "epoch": 0.9202040619886418,
+      "grad_norm": 0.23757819831371307,
+      "learning_rate": 1.6729128067693613e-07,
+      "loss": 0.8201,
+      "step": 2390
+    },
+    {
+      "epoch": 0.9240542881894311,
+      "grad_norm": 0.2540438771247864,
+      "learning_rate": 1.5167213724353426e-07,
+      "loss": 0.8218,
+      "step": 2400
+    },
+    {
+      "epoch": 0.9279045143902205,
+      "grad_norm": 0.23444344103336334,
+      "learning_rate": 1.3680709523610058e-07,
+      "loss": 0.8049,
+      "step": 2410
+    },
+    {
+      "epoch": 0.9317547405910097,
+      "grad_norm": 0.23324888944625854,
+      "learning_rate": 1.2269846674030418e-07,
+      "loss": 0.8277,
+      "step": 2420
+    },
+    {
+      "epoch": 0.935604966791799,
+      "grad_norm": 0.23424877226352692,
+      "learning_rate": 1.0934844619042595e-07,
+      "loss": 0.8245,
+      "step": 2430
+    },
+    {
+      "epoch": 0.9394551929925883,
+      "grad_norm": 0.24835887551307678,
+      "learning_rate": 9.675911002803961e-08,
+      "loss": 0.8136,
+      "step": 2440
+    },
+    {
+      "epoch": 0.9433054191933776,
+      "grad_norm": 0.2521759569644928,
+      "learning_rate": 8.493241637904381e-08,
+      "loss": 0.8117,
+      "step": 2450
+    },
+    {
+      "epoch": 0.9471556453941669,
+      "grad_norm": 0.24677708745002747,
+      "learning_rate": 7.387020474909689e-08,
+      "loss": 0.8303,
+      "step": 2460
+    },
+    {
+      "epoch": 0.9510058715949562,
+      "grad_norm": 0.2518545389175415,
+      "learning_rate": 6.357419573750856e-08,
+      "loss": 0.8252,
+      "step": 2470
+    },
+    {
+      "epoch": 0.9548560977957455,
+      "grad_norm": 0.2372206151485443,
+      "learning_rate": 5.4045990769613365e-08,
+      "loss": 0.8081,
+      "step": 2480
+    },
+    {
+      "epoch": 0.9587063239965348,
+      "grad_norm": 0.28188490867614746,
+      "learning_rate": 4.5287071847694434e-08,
+      "loss": 0.8139,
+      "step": 2490
+    },
+    {
+      "epoch": 0.9625565501973241,
+      "grad_norm": 0.23625126481056213,
+      "learning_rate": 3.7298801320467817e-08,
+      "loss": 0.8065,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9664067763981133,
+      "grad_norm": 0.23558108508586884,
+      "learning_rate": 3.0082421671192576e-08,
+      "loss": 0.7996,
+      "step": 2510
+    },
+    {
+      "epoch": 0.9702570025989027,
+      "grad_norm": 0.24065488576889038,
+      "learning_rate": 2.3639055324410355e-08,
+      "loss": 0.8209,
+      "step": 2520
+    },
+    {
+      "epoch": 0.974107228799692,
+      "grad_norm": 0.23794448375701904,
+      "learning_rate": 1.7969704471369477e-08,
+      "loss": 0.8062,
+      "step": 2530
+    },
+    {
+      "epoch": 0.9779574550004813,
+      "grad_norm": 0.22986574470996857,
+      "learning_rate": 1.3075250914142413e-08,
+      "loss": 0.8104,
+      "step": 2540
+    },
+    {
+      "epoch": 0.9818076812012706,
+      "grad_norm": 0.23236393928527832,
+      "learning_rate": 8.956455928476049e-09,
+      "loss": 0.8202,
+      "step": 2550
+    },
+    {
+      "epoch": 0.9856579074020598,
+      "grad_norm": 0.23714137077331543,
+      "learning_rate": 5.6139601453786275e-09,
+      "loss": 0.8126,
+      "step": 2560
+    },
+    {
+      "epoch": 0.9895081336028492,
+      "grad_norm": 0.2448953092098236,
+      "learning_rate": 3.0482834514805602e-09,
+      "loss": 0.7992,
+      "step": 2570
+    },
+    {
+      "epoch": 0.9933583598036385,
+      "grad_norm": 0.2194613665342331,
+      "learning_rate": 1.2598249081713366e-09,
+      "loss": 0.795,
+      "step": 2580
+    },
+    {
+      "epoch": 0.9972085860044277,
+      "grad_norm": 0.2442457675933838,
+      "learning_rate": 2.488626895302826e-10,
+      "loss": 0.824,
+      "step": 2590
+    },
+    {
+      "epoch": 0.9999037443449803,
+      "step": 2597,
+      "total_flos": 6.41773487286911e+18,
+      "train_loss": 0.8413772296575018,
+      "train_runtime": 13191.169,
+      "train_samples_per_second": 50.405,
+      "train_steps_per_second": 0.197
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2597,
+  "num_input_tokens_seen": 418855352,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.41773487286911e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen3/webssl_mae1b_full2b_224/finetune/training_args.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:f95bd349407cdeee1c6d935c3e677ec434017ee87dd4a0cf7891b57dfa3c66dc
+size 6264

continuous/qwen3/webssl_mae1b_full2b_224/finetune/vocab.json ADDED Viewed

The diff for this file is too large to render. See raw diff

continuous/qwen3/webssl_mae1b_full2b_224/pretrain/config.json ADDED Viewed

	@@ -0,0 +1,89 @@

+{
+  "add_faster_video": false,
+  "add_time_instruction": false,
+  "architectures": [
+    "Qwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "faster_token_stride": 10,
+  "force_sample": false,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "image_aspect_ratio": "square",
+  "image_crop_resolution": null,
+  "image_grid_pinpoints": null,
+  "image_split_resolution": null,
+  "initializer_range": 0.02,
+  "intermediate_size": 6144,
+  "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": 40960,
+  "max_window_layers": 28,
+  "mm_hidden_size": 1536,
+  "mm_newline_position": "grid",
+  "mm_patch_merge_type": "flat",
+  "mm_projector_lr": null,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_resampler_type": null,
+  "mm_spatial_pool_mode": "bilinear",
+  "mm_spatial_pool_stride": null,
+  "mm_tunable_parts": "mm_mlp_adapter",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -1,
+  "mm_vision_tower": "hf:/cache/ckpt/download/tokenizer/continuous/webssl/webssl-mae1b-full2b-224",
+  "mm_vision_tower_lr": null,
+  "model_type": "qwen3",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 8,
+  "pad_token_id": 151643,
+  "pos_skipping_range": 4096,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000,
+  "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/qwen3/webssl_mae1b_full2b_224/pretrain/mm_projector.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:3bfa3bfa8816176abe0ca6367d350941797d23a58e16c63150b894e28f2482db
+size 29378616

continuous/qwen3/webssl_mae1b_full2b_224/pretrain/trainer_state.json ADDED Viewed

	@@ -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": 3.3422510623931885,
+      "learning_rate": 0.00032142857142857147,
+      "loss": 8.7035,
+      "step": 10
+    },
+    {
+      "epoch": 0.03669724770642202,
+      "grad_norm": 0.6650636792182922,
+      "learning_rate": 0.0006785714285714287,
+      "loss": 6.1488,
+      "step": 20
+    },
+    {
+      "epoch": 0.05504587155963303,
+      "grad_norm": 0.09326735883951187,
+      "learning_rate": 0.0009999907688176173,
+      "loss": 4.9215,
+      "step": 30
+    },
+    {
+      "epoch": 0.07339449541284404,
+      "grad_norm": 0.08658801764249802,
+      "learning_rate": 0.0009988834393115766,
+      "loss": 4.7427,
+      "step": 40
+    },
+    {
+      "epoch": 0.09174311926605505,
+      "grad_norm": 0.06587230414152145,
+      "learning_rate": 0.0009959345572940772,
+      "loss": 4.683,
+      "step": 50
+    },
+    {
+      "epoch": 0.11009174311926606,
+      "grad_norm": 0.07734886556863785,
+      "learning_rate": 0.0009911550081155983,
+      "loss": 4.6491,
+      "step": 60
+    },
+    {
+      "epoch": 0.12844036697247707,
+      "grad_norm": 0.08801127225160599,
+      "learning_rate": 0.0009845624347566433,
+      "loss": 4.6216,
+      "step": 70
+    },
+    {
+      "epoch": 0.14678899082568808,
+      "grad_norm": 0.06859865039587021,
+      "learning_rate": 0.0009761811727013547,
+      "loss": 4.6097,
+      "step": 80
+    },
+    {
+      "epoch": 0.1651376146788991,
+      "grad_norm": 0.04957215115427971,
+      "learning_rate": 0.0009660421601067666,
+      "loss": 4.5853,
+      "step": 90
+    },
+    {
+      "epoch": 0.1834862385321101,
+      "grad_norm": 0.10741186887025833,
+      "learning_rate": 0.0009541828235992926,
+      "loss": 4.5792,
+      "step": 100
+    },
+    {
+      "epoch": 0.2018348623853211,
+      "grad_norm": 0.07273948192596436,
+      "learning_rate": 0.0009406469401200151,
+      "loss": 4.5644,
+      "step": 110
+    },
+    {
+      "epoch": 0.22018348623853212,
+      "grad_norm": 0.08069972693920135,
+      "learning_rate": 0.0009254844753287492,
+      "loss": 4.5568,
+      "step": 120
+    },
+    {
+      "epoch": 0.23853211009174313,
+      "grad_norm": 0.11244538426399231,
+      "learning_rate": 0.0009087513991633924,
+      "loss": 4.5364,
+      "step": 130
+    },
+    {
+      "epoch": 0.25688073394495414,
+      "grad_norm": 0.13437429070472717,
+      "learning_rate": 0.0008905094792353917,
+      "loss": 4.5285,
+      "step": 140
+    },
+    {
+      "epoch": 0.27522935779816515,
+      "grad_norm": 0.09372588992118835,
+      "learning_rate": 0.0008708260528239789,
+      "loss": 4.5265,
+      "step": 150
+    },
+    {
+      "epoch": 0.29357798165137616,
+      "grad_norm": 0.15485930442810059,
+      "learning_rate": 0.0008497737783108237,
+      "loss": 4.5148,
+      "step": 160
+    },
+    {
+      "epoch": 0.3119266055045872,
+      "grad_norm": 0.10262388736009598,
+      "learning_rate": 0.0008274303669726426,
+      "loss": 4.5038,
+      "step": 170
+    },
+    {
+      "epoch": 0.3302752293577982,
+      "grad_norm": 0.2217174470424652,
+      "learning_rate": 0.0008038782961218137,
+      "loss": 4.4943,
+      "step": 180
+    },
+    {
+      "epoch": 0.3486238532110092,
+      "grad_norm": 0.10247748345136642,
+      "learning_rate": 0.000779204504653894,
+      "loss": 4.4934,
+      "step": 190
+    },
+    {
+      "epoch": 0.3669724770642202,
+      "grad_norm": 0.1312999129295349,
+      "learning_rate": 0.0007535000721258791,
+      "loss": 4.4725,
+      "step": 200
+    },
+    {
+      "epoch": 0.3853211009174312,
+      "grad_norm": 0.1405658721923828,
+      "learning_rate": 0.00072685988254984,
+      "loss": 4.4634,
+      "step": 210
+    },
+    {
+      "epoch": 0.4036697247706422,
+      "grad_norm": 0.10087669640779495,
+      "learning_rate": 0.0006993822741429906,
+      "loss": 4.4762,
+      "step": 220
+    },
+    {
+      "epoch": 0.42201834862385323,
+      "grad_norm": 0.07646191865205765,
+      "learning_rate": 0.0006711686763270818,
+      "loss": 4.4752,
+      "step": 230
+    },
+    {
+      "epoch": 0.44036697247706424,
+      "grad_norm": 0.15447399020195007,
+      "learning_rate": 0.0006423232353170831,
+      "loss": 4.4592,
+      "step": 240
+    },
+    {
+      "epoch": 0.45871559633027525,
+      "grad_norm": 0.1397457718849182,
+      "learning_rate": 0.0006129524296812335,
+      "loss": 4.4396,
+      "step": 250
+    },
+    {
+      "epoch": 0.47706422018348627,
+      "grad_norm": 0.162072092294693,
+      "learning_rate": 0.0005831646772915651,
+      "loss": 4.4529,
+      "step": 260
+    },
+    {
+      "epoch": 0.4954128440366973,
+      "grad_norm": 0.15283122658729553,
+      "learning_rate": 0.0005530699351157809,
+      "loss": 4.4333,
+      "step": 270
+    },
+    {
+      "epoch": 0.5137614678899083,
+      "grad_norm": 0.1735011488199234,
+      "learning_rate": 0.0005227792933277943,
+      "loss": 4.4304,
+      "step": 280
+    },
+    {
+      "epoch": 0.5321100917431193,
+      "grad_norm": 0.0945705696940422,
+      "learning_rate": 0.0004924045652352092,
+      "loss": 4.4152,
+      "step": 290
+    },
+    {
+      "epoch": 0.5504587155963303,
+      "grad_norm": 0.07350242137908936,
+      "learning_rate": 0.0004620578745374633,
+      "loss": 4.4137,
+      "step": 300
+    },
+    {
+      "epoch": 0.5688073394495413,
+      "grad_norm": 0.12692071497440338,
+      "learning_rate": 0.0004318512414382059,
+      "loss": 4.4266,
+      "step": 310
+    },
+    {
+      "epoch": 0.5871559633027523,
+      "grad_norm": 0.14522163569927216,
+      "learning_rate": 0.0004018961691397148,
+      "loss": 4.4186,
+      "step": 320
+    },
+    {
+      "epoch": 0.6055045871559633,
+      "grad_norm": 0.11378515511751175,
+      "learning_rate": 0.0003723032322457458,
+      "loss": 4.4009,
+      "step": 330
+    },
+    {
+      "epoch": 0.6238532110091743,
+      "grad_norm": 0.1136389896273613,
+      "learning_rate": 0.0003431816685921625,
+      "loss": 4.4002,
+      "step": 340
+    },
+    {
+      "epoch": 0.6422018348623854,
+      "grad_norm": 0.07967250794172287,
+      "learning_rate": 0.0003146389760120416,
+      "loss": 4.4091,
+      "step": 350
+    },
+    {
+      "epoch": 0.6605504587155964,
+      "grad_norm": 0.1101626381278038,
+      "learning_rate": 0.00028678051552373485,
+      "loss": 4.4123,
+      "step": 360
+    },
+    {
+      "epoch": 0.6788990825688074,
+      "grad_norm": 0.1312437802553177,
+      "learning_rate": 0.0002597091224066581,
+      "loss": 4.3938,
+      "step": 370
+    },
+    {
+      "epoch": 0.6972477064220184,
+      "grad_norm": 0.12144605815410614,
+      "learning_rate": 0.0002335247266004629,
+      "loss": 4.4098,
+      "step": 380
+    },
+    {
+      "epoch": 0.7155963302752294,
+      "grad_norm": 0.09792275726795197,
+      "learning_rate": 0.000208323983828828,
+      "loss": 4.3921,
+      "step": 390
+    },
+    {
+      "epoch": 0.7339449541284404,
+      "grad_norm": 0.08326287567615509,
+      "learning_rate": 0.0001841999188095224,
+      "loss": 4.3945,
+      "step": 400
+    },
+    {
+      "epoch": 0.7522935779816514,
+      "grad_norm": 0.14577406644821167,
+      "learning_rate": 0.00016124158186777676,
+      "loss": 4.3944,
+      "step": 410
+    },
+    {
+      "epoch": 0.7706422018348624,
+      "grad_norm": 0.10524504631757736,
+      "learning_rate": 0.0001395337202205211,
+      "loss": 4.3854,
+      "step": 420
+    },
+    {
+      "epoch": 0.7889908256880734,
+      "grad_norm": 0.09286966174840927,
+      "learning_rate": 0.00011915646514489292,
+      "loss": 4.3926,
+      "step": 430
+    },
+    {
+      "epoch": 0.8073394495412844,
+      "grad_norm": 0.08626274764537811,
+      "learning_rate": 0.00010018503618578818,
+      "loss": 4.3974,
+      "step": 440
+    },
+    {
+      "epoch": 0.8256880733944955,
+      "grad_norm": 0.09492462128400803,
+      "learning_rate": 8.268946349432583e-05,
+      "loss": 4.3895,
+      "step": 450
+    },
+    {
+      "epoch": 0.8440366972477065,
+      "grad_norm": 0.11178889870643616,
+      "learning_rate": 6.673432932217117e-05,
+      "loss": 4.3819,
+      "step": 460
+    },
+    {
+      "epoch": 0.8623853211009175,
+      "grad_norm": 0.08762934058904648,
+      "learning_rate": 5.2378529625954684e-05,
+      "loss": 4.3675,
+      "step": 470
+    },
+    {
+      "epoch": 0.8807339449541285,
+      "grad_norm": 0.09271983057260513,
+      "learning_rate": 3.9675056661785556e-05,
+      "loss": 4.3773,
+      "step": 480
+    },
+    {
+      "epoch": 0.8990825688073395,
+      "grad_norm": 0.07573556900024414,
+      "learning_rate": 2.8670803372379538e-05,
+      "loss": 4.3826,
+      "step": 490
+    },
+    {
+      "epoch": 0.9174311926605505,
+      "grad_norm": 0.0685679093003273,
+      "learning_rate": 1.9406390288876586e-05,
+      "loss": 4.3918,
+      "step": 500
+    },
+    {
+      "epoch": 0.9357798165137615,
+      "grad_norm": 0.06937853991985321,
+      "learning_rate": 1.191601558631461e-05,
+      "loss": 4.3801,
+      "step": 510
+    },
+    {
+      "epoch": 0.9541284403669725,
+      "grad_norm": 0.06387509405612946,
+      "learning_rate": 6.2273288462559305e-06,
+      "loss": 4.3722,
+      "step": 520
+    },
+    {
+      "epoch": 0.9724770642201835,
+      "grad_norm": 0.07696132361888885,
+      "learning_rate": 2.361328992552314e-06,
+      "loss": 4.3776,
+      "step": 530
+    },
+    {
+      "epoch": 0.9908256880733946,
+      "grad_norm": 0.07861525565385818,
+      "learning_rate": 3.322867770044202e-07,
+      "loss": 4.3817,
+      "step": 540
+    },
+    {
+      "epoch": 1.0,
+      "step": 545,
+      "total_flos": 4.8872137313746944e+17,
+      "train_loss": 4.578439607532746,
+      "train_runtime": 5083.4524,
+      "train_samples_per_second": 109.793,
+      "train_steps_per_second": 0.107
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 545,
+  "num_input_tokens_seen": 31896544,
+  "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": 4.8872137313746944e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen3/webssl_mae300m_full2b_224/finetune/added_tokens.json ADDED Viewed

	@@ -0,0 +1,28 @@

+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|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/qwen3/webssl_mae300m_full2b_224/finetune/chat_template.jinja ADDED Viewed

	@@ -0,0 +1,89 @@

+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# 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' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- 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' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

continuous/qwen3/webssl_mae300m_full2b_224/finetune/config.json ADDED Viewed

	@@ -0,0 +1,89 @@

+{
+  "add_faster_video": false,
+  "add_time_instruction": false,
+  "architectures": [
+    "LlavaQwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "faster_token_stride": 10,
+  "force_sample": false,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "image_aspect_ratio": "square",
+  "image_crop_resolution": null,
+  "image_grid_pinpoints": null,
+  "image_split_resolution": null,
+  "initializer_range": 0.02,
+  "intermediate_size": 6144,
+  "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": 40960,
+  "max_window_layers": 28,
+  "mm_hidden_size": 1024,
+  "mm_newline_position": "grid",
+  "mm_patch_merge_type": "flat",
+  "mm_projector_lr": 0.0001,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_resampler_type": null,
+  "mm_spatial_pool_mode": "bilinear",
+  "mm_spatial_pool_stride": null,
+  "mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -1,
+  "mm_vision_tower": "hf:/cache/ckpt/download/tokenizer/continuous/webssl/webssl-mae300m-full2b-224",
+  "mm_vision_tower_lr": null,
+  "model_type": "qwen3",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 8,
+  "pad_token_id": 151643,
+  "pos_skipping_range": 4096,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000,
+  "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/qwen3/webssl_mae300m_full2b_224/finetune/generation_config.json ADDED Viewed

	@@ -0,0 +1,12 @@

+{
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "temperature": 0.6,
+  "top_k": 20,
+  "top_p": 0.95,
+  "transformers_version": "4.57.6"
+}

continuous/qwen3/webssl_mae300m_full2b_224/finetune/merges.txt ADDED Viewed

The diff for this file is too large to render. See raw diff

continuous/qwen3/webssl_mae300m_full2b_224/finetune/model.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:e65ca3b62d60aa06a9dea7716cd2013720e8685f531de056b14913f199c92607
+size 4697459344

continuous/qwen3/webssl_mae300m_full2b_224/finetune/special_tokens_map.json ADDED Viewed

	@@ -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/qwen3/webssl_mae300m_full2b_224/finetune/tokenizer.json ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

continuous/qwen3/webssl_mae300m_full2b_224/finetune/tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,240 @@

+{
+  "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
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 8192,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

continuous/qwen3/webssl_mae300m_full2b_224/finetune/trainer_state.json ADDED Viewed

	@@ -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": 16.526716232299805,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 2.2993,
+      "step": 10
+    },
+    {
+      "epoch": 0.007700452401578592,
+      "grad_norm": 6.697127342224121,
+      "learning_rate": 2.435897435897436e-06,
+      "loss": 1.8983,
+      "step": 20
+    },
+    {
+      "epoch": 0.01155067860236789,
+      "grad_norm": 4.405467510223389,
+      "learning_rate": 3.7179487179487184e-06,
+      "loss": 1.5348,
+      "step": 30
+    },
+    {
+      "epoch": 0.015400904803157185,
+      "grad_norm": 1.2974697351455688,
+      "learning_rate": 5e-06,
+      "loss": 1.1978,
+      "step": 40
+    },
+    {
+      "epoch": 0.019251131003946483,
+      "grad_norm": 0.8192834854125977,
+      "learning_rate": 6.282051282051282e-06,
+      "loss": 1.0825,
+      "step": 50
+    },
+    {
+      "epoch": 0.02310135720473578,
+      "grad_norm": 0.43573758006095886,
+      "learning_rate": 7.564102564102564e-06,
+      "loss": 0.9945,
+      "step": 60
+    },
+    {
+      "epoch": 0.026951583405525073,
+      "grad_norm": 0.32553863525390625,
+      "learning_rate": 8.846153846153847e-06,
+      "loss": 0.945,
+      "step": 70
+    },
+    {
+      "epoch": 0.03080180960631437,
+      "grad_norm": 0.3107316792011261,
+      "learning_rate": 9.999996111488723e-06,
+      "loss": 0.9417,
+      "step": 80
+    },
+    {
+      "epoch": 0.034652035807103666,
+      "grad_norm": 0.28649044036865234,
+      "learning_rate": 9.999529497453782e-06,
+      "loss": 0.9112,
+      "step": 90
+    },
+    {
+      "epoch": 0.038502262007892966,
+      "grad_norm": 0.27347123622894287,
+      "learning_rate": 9.998285264324193e-06,
+      "loss": 0.8934,
+      "step": 100
+    },
+    {
+      "epoch": 0.04235248820868226,
+      "grad_norm": 0.2736946642398834,
+      "learning_rate": 9.996263605626052e-06,
+      "loss": 0.8745,
+      "step": 110
+    },
+    {
+      "epoch": 0.04620271440947156,
+      "grad_norm": 0.2656616270542145,
+      "learning_rate": 9.99346483580503e-06,
+      "loss": 0.8854,
+      "step": 120
+    },
+    {
+      "epoch": 0.05005294061026085,
+      "grad_norm": 0.26054418087005615,
+      "learning_rate": 9.989889390177467e-06,
+      "loss": 0.8808,
+      "step": 130
+    },
+    {
+      "epoch": 0.053903166811050146,
+      "grad_norm": 0.29992103576660156,
+      "learning_rate": 9.985537824862647e-06,
+      "loss": 0.8754,
+      "step": 140
+    },
+    {
+      "epoch": 0.057753393011839446,
+      "grad_norm": 0.356355756521225,
+      "learning_rate": 9.980410816696317e-06,
+      "loss": 0.861,
+      "step": 150
+    },
+    {
+      "epoch": 0.06160361921262874,
+      "grad_norm": 0.2781095802783966,
+      "learning_rate": 9.97450916312541e-06,
+      "loss": 0.8653,
+      "step": 160
+    },
+    {
+      "epoch": 0.06545384541341803,
+      "grad_norm": 0.23471404612064362,
+      "learning_rate": 9.967833782084004e-06,
+      "loss": 0.8658,
+      "step": 170
+    },
+    {
+      "epoch": 0.06930407161420733,
+      "grad_norm": 0.2657712996006012,
+      "learning_rate": 9.960385711850554e-06,
+      "loss": 0.8937,
+      "step": 180
+    },
+    {
+      "epoch": 0.07315429781499663,
+      "grad_norm": 0.2762538194656372,
+      "learning_rate": 9.952166110886394e-06,
+      "loss": 0.883,
+      "step": 190
+    },
+    {
+      "epoch": 0.07700452401578593,
+      "grad_norm": 0.28883278369903564,
+      "learning_rate": 9.943176257655567e-06,
+      "loss": 0.8735,
+      "step": 200
+    },
+    {
+      "epoch": 0.08085475021657522,
+      "grad_norm": 0.27570345997810364,
+      "learning_rate": 9.933417550425952e-06,
+      "loss": 0.8673,
+      "step": 210
+    },
+    {
+      "epoch": 0.08470497641736452,
+      "grad_norm": 0.26865601539611816,
+      "learning_rate": 9.9228915070518e-06,
+      "loss": 0.8567,
+      "step": 220
+    },
+    {
+      "epoch": 0.08855520261815382,
+      "grad_norm": 0.2584511339664459,
+      "learning_rate": 9.911599764737632e-06,
+      "loss": 0.8558,
+      "step": 230
+    },
+    {
+      "epoch": 0.09240542881894312,
+      "grad_norm": 0.25940266251564026,
+      "learning_rate": 9.899544079783606e-06,
+      "loss": 0.8501,
+      "step": 240
+    },
+    {
+      "epoch": 0.0962556550197324,
+      "grad_norm": 0.30833226442337036,
+      "learning_rate": 9.886726327312338e-06,
+      "loss": 0.8681,
+      "step": 250
+    },
+    {
+      "epoch": 0.1001058812205217,
+      "grad_norm": 0.26951369643211365,
+      "learning_rate": 9.87314850097724e-06,
+      "loss": 0.8653,
+      "step": 260
+    },
+    {
+      "epoch": 0.103956107421311,
+      "grad_norm": 0.2643968462944031,
+      "learning_rate": 9.858812712652452e-06,
+      "loss": 0.8525,
+      "step": 270
+    },
+    {
+      "epoch": 0.10780633362210029,
+      "grad_norm": 0.2660777270793915,
+      "learning_rate": 9.843721192104341e-06,
+      "loss": 0.8528,
+      "step": 280
+    },
+    {
+      "epoch": 0.11165655982288959,
+      "grad_norm": 0.26574066281318665,
+      "learning_rate": 9.827876286644698e-06,
+      "loss": 0.8693,
+      "step": 290
+    },
+    {
+      "epoch": 0.11550678602367889,
+      "grad_norm": 0.2351258546113968,
+      "learning_rate": 9.811280460765643e-06,
+      "loss": 0.8292,
+      "step": 300
+    },
+    {
+      "epoch": 0.11935701222446819,
+      "grad_norm": 0.24751019477844238,
+      "learning_rate": 9.793936295756292e-06,
+      "loss": 0.8441,
+      "step": 310
+    },
+    {
+      "epoch": 0.12320723842525748,
+      "grad_norm": 0.2670865058898926,
+      "learning_rate": 9.775846489301279e-06,
+      "loss": 0.844,
+      "step": 320
+    },
+    {
+      "epoch": 0.12705746462604678,
+      "grad_norm": 0.2891390323638916,
+      "learning_rate": 9.757013855061157e-06,
+      "loss": 0.844,
+      "step": 330
+    },
+    {
+      "epoch": 0.13090769082683607,
+      "grad_norm": 0.2591327726840973,
+      "learning_rate": 9.737441322234758e-06,
+      "loss": 0.854,
+      "step": 340
+    },
+    {
+      "epoch": 0.13475791702762538,
+      "grad_norm": 0.28134819865226746,
+      "learning_rate": 9.717131935103605e-06,
+      "loss": 0.8546,
+      "step": 350
+    },
+    {
+      "epoch": 0.13860814322841467,
+      "grad_norm": 0.25483229756355286,
+      "learning_rate": 9.696088852558398e-06,
+      "loss": 0.8405,
+      "step": 360
+    },
+    {
+      "epoch": 0.14245836942920398,
+      "grad_norm": 0.2702053189277649,
+      "learning_rate": 9.674315347607686e-06,
+      "loss": 0.8675,
+      "step": 370
+    },
+    {
+      "epoch": 0.14630859562999327,
+      "grad_norm": 0.26903215050697327,
+      "learning_rate": 9.651814806868796e-06,
+      "loss": 0.8282,
+      "step": 380
+    },
+    {
+      "epoch": 0.15015882183078255,
+      "grad_norm": 0.2678152024745941,
+      "learning_rate": 9.62859073004107e-06,
+      "loss": 0.8354,
+      "step": 390
+    },
+    {
+      "epoch": 0.15400904803157187,
+      "grad_norm": 0.24652773141860962,
+      "learning_rate": 9.604646729361547e-06,
+      "loss": 0.8401,
+      "step": 400
+    },
+    {
+      "epoch": 0.15785927423236115,
+      "grad_norm": 0.2463948130607605,
+      "learning_rate": 9.579986529043099e-06,
+      "loss": 0.8278,
+      "step": 410
+    },
+    {
+      "epoch": 0.16170950043315044,
+      "grad_norm": 0.2636524438858032,
+      "learning_rate": 9.554613964695189e-06,
+      "loss": 0.8439,
+      "step": 420
+    },
+    {
+      "epoch": 0.16555972663393975,
+      "grad_norm": 0.25433456897735596,
+      "learning_rate": 9.528532982727281e-06,
+      "loss": 0.8333,
+      "step": 430
+    },
+    {
+      "epoch": 0.16940995283472904,
+      "grad_norm": 0.2628912627696991,
+      "learning_rate": 9.501747639735018e-06,
+      "loss": 0.8321,
+      "step": 440
+    },
+    {
+      "epoch": 0.17326017903551832,
+      "grad_norm": 0.2806665599346161,
+      "learning_rate": 9.474262101869272e-06,
+      "loss": 0.8211,
+      "step": 450
+    },
+    {
+      "epoch": 0.17711040523630764,
+      "grad_norm": 0.25942814350128174,
+      "learning_rate": 9.446080644188141e-06,
+      "loss": 0.8264,
+      "step": 460
+    },
+    {
+      "epoch": 0.18096063143709692,
+      "grad_norm": 0.2718483507633209,
+      "learning_rate": 9.417207649992017e-06,
+      "loss": 0.836,
+      "step": 470
+    },
+    {
+      "epoch": 0.18481085763788624,
+      "grad_norm": 0.2521530091762543,
+      "learning_rate": 9.387647610141816e-06,
+      "loss": 0.834,
+      "step": 480
+    },
+    {
+      "epoch": 0.18866108383867553,
+      "grad_norm": 0.2406543344259262,
+      "learning_rate": 9.35740512236047e-06,
+      "loss": 0.8199,
+      "step": 490
+    },
+    {
+      "epoch": 0.1925113100394648,
+      "grad_norm": 0.2391744703054428,
+      "learning_rate": 9.326484890517809e-06,
+      "loss": 0.8243,
+      "step": 500
+    },
+    {
+      "epoch": 0.19636153624025413,
+      "grad_norm": 0.2550431191921234,
+      "learning_rate": 9.294891723898926e-06,
+      "loss": 0.821,
+      "step": 510
+    },
+    {
+      "epoch": 0.2002117624410434,
+      "grad_norm": 0.258022278547287,
+      "learning_rate": 9.262630536456153e-06,
+      "loss": 0.8508,
+      "step": 520
+    },
+    {
+      "epoch": 0.2040619886418327,
+      "grad_norm": 0.2654220759868622,
+      "learning_rate": 9.229706346044749e-06,
+      "loss": 0.8211,
+      "step": 530
+    },
+    {
+      "epoch": 0.207912214842622,
+      "grad_norm": 0.2596435546875,
+      "learning_rate": 9.196124273642422e-06,
+      "loss": 0.8357,
+      "step": 540
+    },
+    {
+      "epoch": 0.2117624410434113,
+      "grad_norm": 0.24092335999011993,
+      "learning_rate": 9.161889542552842e-06,
+      "loss": 0.8367,
+      "step": 550
+    },
+    {
+      "epoch": 0.21561266724420058,
+      "grad_norm": 0.2313719540834427,
+      "learning_rate": 9.127007477593193e-06,
+      "loss": 0.8424,
+      "step": 560
+    },
+    {
+      "epoch": 0.2194628934449899,
+      "grad_norm": 0.2714857757091522,
+      "learning_rate": 9.091483504265974e-06,
+      "loss": 0.8519,
+      "step": 570
+    },
+    {
+      "epoch": 0.22331311964577918,
+      "grad_norm": 0.265025794506073,
+      "learning_rate": 9.055323147915116e-06,
+      "loss": 0.8306,
+      "step": 580
+    },
+    {
+      "epoch": 0.2271633458465685,
+      "grad_norm": 0.27817314863204956,
+      "learning_rate": 9.018532032866585e-06,
+      "loss": 0.8415,
+      "step": 590
+    },
+    {
+      "epoch": 0.23101357204735778,
+      "grad_norm": 0.2591384947299957,
+      "learning_rate": 8.981115881553579e-06,
+      "loss": 0.8331,
+      "step": 600
+    },
+    {
+      "epoch": 0.23486379824814707,
+      "grad_norm": 0.27948033809661865,
+      "learning_rate": 8.943080513626476e-06,
+      "loss": 0.8529,
+      "step": 610
+    },
+    {
+      "epoch": 0.23871402444893638,
+      "grad_norm": 0.25202226638793945,
+      "learning_rate": 8.904431845047641e-06,
+      "loss": 0.8363,
+      "step": 620
+    },
+    {
+      "epoch": 0.24256425064972567,
+      "grad_norm": 0.2579434812068939,
+      "learning_rate": 8.86517588717128e-06,
+      "loss": 0.8535,
+      "step": 630
+    },
+    {
+      "epoch": 0.24641447685051496,
+      "grad_norm": 0.24299272894859314,
+      "learning_rate": 8.82531874580844e-06,
+      "loss": 0.8413,
+      "step": 640
+    },
+    {
+      "epoch": 0.25026470305130427,
+      "grad_norm": 0.23357634246349335,
+      "learning_rate": 8.784866620277314e-06,
+      "loss": 0.7986,
+      "step": 650
+    },
+    {
+      "epoch": 0.25411492925209356,
+      "grad_norm": 0.2770780324935913,
+      "learning_rate": 8.743825802439018e-06,
+      "loss": 0.8366,
+      "step": 660
+    },
+    {
+      "epoch": 0.25796515545288284,
+      "grad_norm": 0.29918918013572693,
+      "learning_rate": 8.702202675718956e-06,
+      "loss": 0.8412,
+      "step": 670
+    },
+    {
+      "epoch": 0.26181538165367213,
+      "grad_norm": 0.2673822045326233,
+      "learning_rate": 8.660003714113959e-06,
+      "loss": 0.8114,
+      "step": 680
+    },
+    {
+      "epoch": 0.26566560785446147,
+      "grad_norm": 0.2473033368587494,
+      "learning_rate": 8.617235481185323e-06,
+      "loss": 0.8366,
+      "step": 690
+    },
+    {
+      "epoch": 0.26951583405525076,
+      "grad_norm": 0.25063085556030273,
+      "learning_rate": 8.573904629037931e-06,
+      "loss": 0.8404,
+      "step": 700
+    },
+    {
+      "epoch": 0.27336606025604004,
+      "grad_norm": 0.2579231858253479,
+      "learning_rate": 8.530017897285585e-06,
+      "loss": 0.8204,
+      "step": 710
+    },
+    {
+      "epoch": 0.27721628645682933,
+      "grad_norm": 0.2530323565006256,
+      "learning_rate": 8.485582112002746e-06,
+      "loss": 0.8253,
+      "step": 720
+    },
+    {
+      "epoch": 0.2810665126576186,
+      "grad_norm": 0.2643014192581177,
+      "learning_rate": 8.440604184662811e-06,
+      "loss": 0.8315,
+      "step": 730
+    },
+    {
+      "epoch": 0.28491673885840796,
+      "grad_norm": 0.2573903799057007,
+      "learning_rate": 8.395091111063118e-06,
+      "loss": 0.8208,
+      "step": 740
+    },
+    {
+      "epoch": 0.28876696505919724,
+      "grad_norm": 0.23633959889411926,
+      "learning_rate": 8.349049970236822e-06,
+      "loss": 0.8191,
+      "step": 750
+    },
+    {
+      "epoch": 0.29261719125998653,
+      "grad_norm": 0.25756216049194336,
+      "learning_rate": 8.302487923351842e-06,
+      "loss": 0.8122,
+      "step": 760
+    },
+    {
+      "epoch": 0.2964674174607758,
+      "grad_norm": 0.2501530647277832,
+      "learning_rate": 8.255412212597021e-06,
+      "loss": 0.8295,
+      "step": 770
+    },
+    {
+      "epoch": 0.3003176436615651,
+      "grad_norm": 0.24190166592597961,
+      "learning_rate": 8.207830160055688e-06,
+      "loss": 0.8262,
+      "step": 780
+    },
+    {
+      "epoch": 0.3041678698623544,
+      "grad_norm": 0.26249244809150696,
+      "learning_rate": 8.159749166566786e-06,
+      "loss": 0.8116,
+      "step": 790
+    },
+    {
+      "epoch": 0.30801809606314373,
+      "grad_norm": 0.26733630895614624,
+      "learning_rate": 8.11117671057377e-06,
+      "loss": 0.839,
+      "step": 800
+    },
+    {
+      "epoch": 0.311868322263933,
+      "grad_norm": 0.25793781876564026,
+      "learning_rate": 8.062120346961404e-06,
+      "loss": 0.8313,
+      "step": 810
+    },
+    {
+      "epoch": 0.3157185484647223,
+      "grad_norm": 0.25373291969299316,
+      "learning_rate": 8.012587705880705e-06,
+      "loss": 0.8242,
+      "step": 820
+    },
+    {
+      "epoch": 0.3195687746655116,
+      "grad_norm": 0.25895601511001587,
+      "learning_rate": 7.962586491562146e-06,
+      "loss": 0.8102,
+      "step": 830
+    },
+    {
+      "epoch": 0.3234190008663009,
+      "grad_norm": 0.261120468378067,
+      "learning_rate": 7.912124481117357e-06,
+      "loss": 0.8162,
+      "step": 840
+    },
+    {
+      "epoch": 0.3272692270670902,
+      "grad_norm": 0.2601974904537201,
+      "learning_rate": 7.861209523329494e-06,
+      "loss": 0.8259,
+      "step": 850
+    },
+    {
+      "epoch": 0.3311194532678795,
+      "grad_norm": 0.24367204308509827,
+      "learning_rate": 7.809849537432432e-06,
+      "loss": 0.8097,
+      "step": 860
+    },
+    {
+      "epoch": 0.3349696794686688,
+      "grad_norm": 0.25117015838623047,
+      "learning_rate": 7.758052511879035e-06,
+      "loss": 0.8204,
+      "step": 870
+    },
+    {
+      "epoch": 0.3388199056694581,
+      "grad_norm": 0.23988498747348785,
+      "learning_rate": 7.705826503098642e-06,
+      "loss": 0.8148,
+      "step": 880
+    },
+    {
+      "epoch": 0.34267013187024736,
+      "grad_norm": 0.23775863647460938,
+      "learning_rate": 7.653179634243979e-06,
+      "loss": 0.8145,
+      "step": 890
+    },
+    {
+      "epoch": 0.34652035807103665,
+      "grad_norm": 0.23163731396198273,
+      "learning_rate": 7.600120093927696e-06,
+      "loss": 0.8217,
+      "step": 900
+    },
+    {
+      "epoch": 0.350370584271826,
+      "grad_norm": 0.24480977654457092,
+      "learning_rate": 7.546656134948729e-06,
+      "loss": 0.811,
+      "step": 910
+    },
+    {
+      "epoch": 0.3542208104726153,
+      "grad_norm": 0.262422114610672,
+      "learning_rate": 7.492796073008661e-06,
+      "loss": 0.8087,
+      "step": 920
+    },
+    {
+      "epoch": 0.35807103667340456,
+      "grad_norm": 0.2700716555118561,
+      "learning_rate": 7.438548285418326e-06,
+      "loss": 0.7953,
+      "step": 930
+    },
+    {
+      "epoch": 0.36192126287419385,
+      "grad_norm": 0.2739006280899048,
+      "learning_rate": 7.3839212097948e-06,
+      "loss": 0.8287,
+      "step": 940
+    },
+    {
+      "epoch": 0.36577148907498314,
+      "grad_norm": 0.25563907623291016,
+      "learning_rate": 7.328923342749039e-06,
+      "loss": 0.844,
+      "step": 950
+    },
+    {
+      "epoch": 0.3696217152757725,
+      "grad_norm": 0.26924723386764526,
+      "learning_rate": 7.273563238564319e-06,
+      "loss": 0.8277,
+      "step": 960
+    },
+    {
+      "epoch": 0.37347194147656176,
+      "grad_norm": 0.2707979083061218,
+      "learning_rate": 7.217849507865724e-06,
+      "loss": 0.8096,
+      "step": 970
+    },
+    {
+      "epoch": 0.37732216767735105,
+      "grad_norm": 0.2582569122314453,
+      "learning_rate": 7.161790816280858e-06,
+      "loss": 0.8133,
+      "step": 980
+    },
+    {
+      "epoch": 0.38117239387814034,
+      "grad_norm": 0.2604237496852875,
+      "learning_rate": 7.1053958830920125e-06,
+      "loss": 0.8015,
+      "step": 990
+    },
+    {
+      "epoch": 0.3850226200789296,
+      "grad_norm": 0.2445145845413208,
+      "learning_rate": 7.0486734798799715e-06,
+      "loss": 0.8157,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3888728462797189,
+      "grad_norm": 0.25000885128974915,
+      "learning_rate": 6.991632429159708e-06,
+      "loss": 0.839,
+      "step": 1010
+    },
+    {
+      "epoch": 0.39272307248050825,
+      "grad_norm": 0.2646527886390686,
+      "learning_rate": 6.934281603008132e-06,
+      "loss": 0.8324,
+      "step": 1020
+    },
+    {
+      "epoch": 0.39657329868129754,
+      "grad_norm": 0.2521289587020874,
+      "learning_rate": 6.876629921684143e-06,
+      "loss": 0.8291,
+      "step": 1030
+    },
+    {
+      "epoch": 0.4004235248820868,
+      "grad_norm": 0.26221325993537903,
+      "learning_rate": 6.818686352241199e-06,
+      "loss": 0.8201,
+      "step": 1040
+    },
+    {
+      "epoch": 0.4042737510828761,
+      "grad_norm": 0.22739070653915405,
+      "learning_rate": 6.760459907132573e-06,
+      "loss": 0.8161,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4081239772836654,
+      "grad_norm": 0.24100252985954285,
+      "learning_rate": 6.701959642809589e-06,
+      "loss": 0.8228,
+      "step": 1060
+    },
+    {
+      "epoch": 0.41197420348445474,
+      "grad_norm": 0.2527819871902466,
+      "learning_rate": 6.643194658312987e-06,
+      "loss": 0.815,
+      "step": 1070
+    },
+    {
+      "epoch": 0.415824429685244,
+      "grad_norm": 0.25091230869293213,
+      "learning_rate": 6.584174093857676e-06,
+      "loss": 0.8389,
+      "step": 1080
+    },
+    {
+      "epoch": 0.4196746558860333,
+      "grad_norm": 0.2762874960899353,
+      "learning_rate": 6.5249071294110735e-06,
+      "loss": 0.8182,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4235248820868226,
+      "grad_norm": 0.259245365858078,
+      "learning_rate": 6.465402983265274e-06,
+      "loss": 0.833,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4273751082876119,
+      "grad_norm": 0.2502368986606598,
+      "learning_rate": 6.405670910603243e-06,
+      "loss": 0.8294,
+      "step": 1110
+    },
+    {
+      "epoch": 0.43122533448840117,
+      "grad_norm": 0.2710523307323456,
+      "learning_rate": 6.345720202059291e-06,
+      "loss": 0.825,
+      "step": 1120
+    },
+    {
+      "epoch": 0.4350755606891905,
+      "grad_norm": 0.26031383872032166,
+      "learning_rate": 6.285560182274008e-06,
+      "loss": 0.8044,
+      "step": 1130
+    },
+    {
+      "epoch": 0.4389257868899798,
+      "grad_norm": 0.24232666194438934,
+      "learning_rate": 6.225200208443936e-06,
+      "loss": 0.826,
+      "step": 1140
+    },
+    {
+      "epoch": 0.4427760130907691,
+      "grad_norm": 0.23385505378246307,
+      "learning_rate": 6.16464966886616e-06,
+      "loss": 0.8165,
+      "step": 1150
+    },
+    {
+      "epoch": 0.44662623929155837,
+      "grad_norm": 0.28117090463638306,
+      "learning_rate": 6.103917981478065e-06,
+      "loss": 0.8389,
+      "step": 1160
+    },
+    {
+      "epoch": 0.45047646549234766,
+      "grad_norm": 0.252493679523468,
+      "learning_rate": 6.043014592392489e-06,
+      "loss": 0.851,
+      "step": 1170
+    },
+    {
+      "epoch": 0.454326691693137,
+      "grad_norm": 0.24117374420166016,
+      "learning_rate": 5.981948974428485e-06,
+      "loss": 0.8247,
+      "step": 1180
+    },
+    {
+      "epoch": 0.4581769178939263,
+      "grad_norm": 0.2584195137023926,
+      "learning_rate": 5.920730625637934e-06,
+      "loss": 0.8057,
+      "step": 1190
+    },
+    {
+      "epoch": 0.46202714409471557,
+      "grad_norm": 0.2587830126285553,
+      "learning_rate": 5.8593690678282355e-06,
+      "loss": 0.8145,
+      "step": 1200
+    },
+    {
+      "epoch": 0.46587737029550486,
+      "grad_norm": 0.26448705792427063,
+      "learning_rate": 5.797873845081301e-06,
+      "loss": 0.818,
+      "step": 1210
+    },
+    {
+      "epoch": 0.46972759649629414,
+      "grad_norm": 0.2465999573469162,
+      "learning_rate": 5.736254522269075e-06,
+      "loss": 0.8357,
+      "step": 1220
+    },
+    {
+      "epoch": 0.47357782269708343,
+      "grad_norm": 0.25537392497062683,
+      "learning_rate": 5.6745206835658386e-06,
+      "loss": 0.8361,
+      "step": 1230
+    },
+    {
+      "epoch": 0.47742804889787277,
+      "grad_norm": 0.2449575811624527,
+      "learning_rate": 5.612681930957495e-06,
+      "loss": 0.8076,
+      "step": 1240
+    },
+    {
+      "epoch": 0.48127827509866206,
+      "grad_norm": 0.23686227202415466,
+      "learning_rate": 5.550747882748095e-06,
+      "loss": 0.8074,
+      "step": 1250
+    },
+    {
+      "epoch": 0.48512850129945134,
+      "grad_norm": 0.2772762179374695,
+      "learning_rate": 5.4887281720638195e-06,
+      "loss": 0.8116,
+      "step": 1260
+    },
+    {
+      "epoch": 0.48897872750024063,
+      "grad_norm": 0.275837242603302,
+      "learning_rate": 5.42663244535466e-06,
+      "loss": 0.8003,
+      "step": 1270
+    },
+    {
+      "epoch": 0.4928289537010299,
+      "grad_norm": 0.26124870777130127,
+      "learning_rate": 5.3644703608940205e-06,
+      "loss": 0.8055,
+      "step": 1280
+    },
+    {
+      "epoch": 0.49667917990181926,
+      "grad_norm": 0.2568417489528656,
+      "learning_rate": 5.302251587276491e-06,
+      "loss": 0.8175,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5005294061026085,
+      "grad_norm": 0.24265603721141815,
+      "learning_rate": 5.2399858019140005e-06,
+      "loss": 0.8172,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5043796323033978,
+      "grad_norm": 0.2459190934896469,
+      "learning_rate": 5.177682689530615e-06,
+      "loss": 0.8318,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5082298585041871,
+      "grad_norm": 0.2577625513076782,
+      "learning_rate": 5.11535194065619e-06,
+      "loss": 0.7997,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5120800847049765,
+      "grad_norm": 0.27089083194732666,
+      "learning_rate": 5.053003250119113e-06,
+      "loss": 0.8123,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5159303109057657,
+      "grad_norm": 0.263537734746933,
+      "learning_rate": 4.9906463155383925e-06,
+      "loss": 0.8176,
+      "step": 1340
+    },
+    {
+      "epoch": 0.519780537106555,
+      "grad_norm": 0.24327850341796875,
+      "learning_rate": 4.9282908358153046e-06,
+      "loss": 0.808,
+      "step": 1350
+    },
+    {
+      "epoch": 0.5236307633073443,
+      "grad_norm": 0.2747257351875305,
+      "learning_rate": 4.8659465096248345e-06,
+      "loss": 0.8283,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5274809895081336,
+      "grad_norm": 0.27198493480682373,
+      "learning_rate": 4.803623033907169e-06,
+      "loss": 0.8276,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5313312157089229,
+      "grad_norm": 0.24149709939956665,
+      "learning_rate": 4.741330102359441e-06,
+      "loss": 0.8313,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5351814419097122,
+      "grad_norm": 0.26828083395957947,
+      "learning_rate": 4.679077403927996e-06,
+      "loss": 0.8125,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5390316681105015,
+      "grad_norm": 0.23802612721920013,
+      "learning_rate": 4.616874621301373e-06,
+      "loss": 0.8105,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5428818943112907,
+      "grad_norm": 0.26935386657714844,
+      "learning_rate": 4.554731429404293e-06,
+      "loss": 0.8263,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5467321205120801,
+      "grad_norm": 0.27729910612106323,
+      "learning_rate": 4.492657493892822e-06,
+      "loss": 0.8123,
+      "step": 1420
+    },
+    {
+      "epoch": 0.5505823467128694,
+      "grad_norm": 0.2870791554450989,
+      "learning_rate": 4.430662469650999e-06,
+      "loss": 0.8265,
+      "step": 1430
+    },
+    {
+      "epoch": 0.5544325729136587,
+      "grad_norm": 0.26641643047332764,
+      "learning_rate": 4.368755999289118e-06,
+      "loss": 0.828,
+      "step": 1440
+    },
+    {
+      "epoch": 0.558282799114448,
+      "grad_norm": 0.3034954369068146,
+      "learning_rate": 4.306947711643949e-06,
+      "loss": 0.8155,
+      "step": 1450
+    },
+    {
+      "epoch": 0.5621330253152372,
+      "grad_norm": 0.2527589499950409,
+      "learning_rate": 4.245247220281068e-06,
+      "loss": 0.8163,
+      "step": 1460
+    },
+    {
+      "epoch": 0.5659832515160266,
+      "grad_norm": 0.2456417828798294,
+      "learning_rate": 4.183664121999587e-06,
+      "loss": 0.8158,
+      "step": 1470
+    },
+    {
+      "epoch": 0.5698334777168159,
+      "grad_norm": 0.2610345184803009,
+      "learning_rate": 4.122207995339474e-06,
+      "loss": 0.8098,
+      "step": 1480
+    },
+    {
+      "epoch": 0.5736837039176051,
+      "grad_norm": 0.24188220500946045,
+      "learning_rate": 4.060888399091734e-06,
+      "loss": 0.8179,
+      "step": 1490
+    },
+    {
+      "epoch": 0.5775339301183945,
+      "grad_norm": 0.2502032518386841,
+      "learning_rate": 3.999714870811639e-06,
+      "loss": 0.8237,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5813841563191837,
+      "grad_norm": 0.2518673837184906,
+      "learning_rate": 3.938696925335275e-06,
+      "loss": 0.8177,
+      "step": 1510
+    },
+    {
+      "epoch": 0.5852343825199731,
+      "grad_norm": 0.25666046142578125,
+      "learning_rate": 3.87784405329962e-06,
+      "loss": 0.826,
+      "step": 1520
+    },
+    {
+      "epoch": 0.5890846087207623,
+      "grad_norm": 0.2596080005168915,
+      "learning_rate": 3.817165719666382e-06,
+      "loss": 0.8288,
+      "step": 1530
+    },
+    {
+      "epoch": 0.5929348349215516,
+      "grad_norm": 0.23970122635364532,
+      "learning_rate": 3.756671362249841e-06,
+      "loss": 0.8315,
+      "step": 1540
+    },
+    {
+      "epoch": 0.596785061122341,
+      "grad_norm": 0.25825774669647217,
+      "learning_rate": 3.6963703902488907e-06,
+      "loss": 0.8178,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6006352873231302,
+      "grad_norm": 0.25253039598464966,
+      "learning_rate": 3.6362721827835655e-06,
+      "loss": 0.8384,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6044855135239195,
+      "grad_norm": 0.2610830068588257,
+      "learning_rate": 3.576386087436212e-06,
+      "loss": 0.831,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6083357397247088,
+      "grad_norm": 0.2735256254673004,
+      "learning_rate": 3.5167214187975913e-06,
+      "loss": 0.8205,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6121859659254981,
+      "grad_norm": 0.21917392313480377,
+      "learning_rate": 3.457287457018086e-06,
+      "loss": 0.8225,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6160361921262875,
+      "grad_norm": 0.24496084451675415,
+      "learning_rate": 3.3980934463643003e-06,
+      "loss": 0.81,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6198864183270767,
+      "grad_norm": 0.22461284697055817,
+      "learning_rate": 3.3391485937812053e-06,
+      "loss": 0.7928,
+      "step": 1610
+    },
+    {
+      "epoch": 0.623736644527866,
+      "grad_norm": 0.258798211812973,
+      "learning_rate": 3.280462067460115e-06,
+      "loss": 0.7924,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6275868707286553,
+      "grad_norm": 0.2496568262577057,
+      "learning_rate": 3.222042995412669e-06,
+      "loss": 0.8219,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6314370969294446,
+      "grad_norm": 0.2331293523311615,
+      "learning_rate": 3.163900464051089e-06,
+      "loss": 0.8123,
+      "step": 1640
+    },
+    {
+      "epoch": 0.635287323130234,
+      "grad_norm": 0.25422975420951843,
+      "learning_rate": 3.1060435167748826e-06,
+      "loss": 0.8205,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6391375493310232,
+      "grad_norm": 0.2525100111961365,
+      "learning_rate": 3.048481152564253e-06,
+      "loss": 0.8012,
+      "step": 1660
+    },
+    {
+      "epoch": 0.6429877755318125,
+      "grad_norm": 0.26372870802879333,
+      "learning_rate": 2.9912223245804017e-06,
+      "loss": 0.8227,
+      "step": 1670
+    },
+    {
+      "epoch": 0.6468380017326018,
+      "grad_norm": 0.2545525133609772,
+      "learning_rate": 2.934275938772981e-06,
+      "loss": 0.8212,
+      "step": 1680
+    },
+    {
+      "epoch": 0.6506882279333911,
+      "grad_norm": 0.2672939896583557,
+      "learning_rate": 2.8776508524948676e-06,
+      "loss": 0.8105,
+      "step": 1690
+    },
+    {
+      "epoch": 0.6545384541341804,
+      "grad_norm": 0.23621158301830292,
+      "learning_rate": 2.821355873124503e-06,
+      "loss": 0.8096,
+      "step": 1700
+    },
+    {
+      "epoch": 0.6583886803349697,
+      "grad_norm": 0.2594734728336334,
+      "learning_rate": 2.7653997566960144e-06,
+      "loss": 0.8231,
+      "step": 1710
+    },
+    {
+      "epoch": 0.662238906535759,
+      "grad_norm": 0.26584935188293457,
+      "learning_rate": 2.709791206537309e-06,
+      "loss": 0.8128,
+      "step": 1720
+    },
+    {
+      "epoch": 0.6660891327365482,
+      "grad_norm": 0.25663405656814575,
+      "learning_rate": 2.6545388719163747e-06,
+      "loss": 0.8205,
+      "step": 1730
+    },
+    {
+      "epoch": 0.6699393589373376,
+      "grad_norm": 0.25418025255203247,
+      "learning_rate": 2.599651346695979e-06,
+      "loss": 0.8178,
+      "step": 1740
+    },
+    {
+      "epoch": 0.6737895851381268,
+      "grad_norm": 0.26431748270988464,
+      "learning_rate": 2.545137167997007e-06,
+      "loss": 0.8129,
+      "step": 1750
+    },
+    {
+      "epoch": 0.6776398113389162,
+      "grad_norm": 0.24246641993522644,
+      "learning_rate": 2.4910048148705972e-06,
+      "loss": 0.8022,
+      "step": 1760
+    },
+    {
+      "epoch": 0.6814900375397055,
+      "grad_norm": 0.2710860073566437,
+      "learning_rate": 2.4372627069793327e-06,
+      "loss": 0.8441,
+      "step": 1770
+    },
+    {
+      "epoch": 0.6853402637404947,
+      "grad_norm": 0.2491195648908615,
+      "learning_rate": 2.3839192032876486e-06,
+      "loss": 0.8319,
+      "step": 1780
+    },
+    {
+      "epoch": 0.6891904899412841,
+      "grad_norm": 0.24903680384159088,
+      "learning_rate": 2.3309826007617047e-06,
+      "loss": 0.8415,
+      "step": 1790
+    },
+    {
+      "epoch": 0.6930407161420733,
+      "grad_norm": 0.2570270001888275,
+      "learning_rate": 2.278461133078881e-06,
+      "loss": 0.8053,
+      "step": 1800
+    },
+    {
+      "epoch": 0.6968909423428626,
+      "grad_norm": 0.26018625497817993,
+      "learning_rate": 2.2263629693471258e-06,
+      "loss": 0.7967,
+      "step": 1810
+    },
+    {
+      "epoch": 0.700741168543652,
+      "grad_norm": 0.24607177078723907,
+      "learning_rate": 2.1746962128343414e-06,
+      "loss": 0.8223,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7045913947444412,
+      "grad_norm": 0.23676884174346924,
+      "learning_rate": 2.1234688997080183e-06,
+      "loss": 0.8095,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7084416209452306,
+      "grad_norm": 0.26461610198020935,
+      "learning_rate": 2.0726889977853053e-06,
+      "loss": 0.8139,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7122918471460198,
+      "grad_norm": 0.266209214925766,
+      "learning_rate": 2.022364405293703e-06,
+      "loss": 0.8266,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7161420733468091,
+      "grad_norm": 0.24904181063175201,
+      "learning_rate": 1.9725029496425814e-06,
+      "loss": 0.8191,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7199922995475985,
+      "grad_norm": 0.25554358959198,
+      "learning_rate": 1.923112386205727e-06,
+      "loss": 0.8169,
+      "step": 1870
+    },
+    {
+      "epoch": 0.7238425257483877,
+      "grad_norm": 0.2325001060962677,
+      "learning_rate": 1.874200397115079e-06,
+      "loss": 0.8233,
+      "step": 1880
+    },
+    {
+      "epoch": 0.727692751949177,
+      "grad_norm": 0.2647624611854553,
+      "learning_rate": 1.825774590065857e-06,
+      "loss": 0.8077,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7315429781499663,
+      "grad_norm": 0.26150187849998474,
+      "learning_rate": 1.7778424971332813e-06,
+      "loss": 0.8144,
+      "step": 1900
+    },
+    {
+      "epoch": 0.7353932043507556,
+      "grad_norm": 0.2502162456512451,
+      "learning_rate": 1.7304115736010407e-06,
+      "loss": 0.8142,
+      "step": 1910
+    },
+    {
+      "epoch": 0.739243430551545,
+      "grad_norm": 0.25100913643836975,
+      "learning_rate": 1.6834891968017108e-06,
+      "loss": 0.8053,
+      "step": 1920
+    },
+    {
+      "epoch": 0.7430936567523342,
+      "grad_norm": 0.2561488449573517,
+      "learning_rate": 1.6370826649692884e-06,
+      "loss": 0.8274,
+      "step": 1930
+    },
+    {
+      "epoch": 0.7469438829531235,
+      "grad_norm": 0.2497335523366928,
+      "learning_rate": 1.5911991961040463e-06,
+      "loss": 0.8322,
+      "step": 1940
+    },
+    {
+      "epoch": 0.7507941091539128,
+      "grad_norm": 0.25837206840515137,
+      "learning_rate": 1.545845926849851e-06,
+      "loss": 0.8178,
+      "step": 1950
+    },
+    {
+      "epoch": 0.7546443353547021,
+      "grad_norm": 0.2647581696510315,
+      "learning_rate": 1.5010299113841397e-06,
+      "loss": 0.8373,
+      "step": 1960
+    },
+    {
+      "epoch": 0.7584945615554913,
+      "grad_norm": 0.25618892908096313,
+      "learning_rate": 1.4567581203207238e-06,
+      "loss": 0.8246,
+      "step": 1970
+    },
+    {
+      "epoch": 0.7623447877562807,
+      "grad_norm": 0.26886770129203796,
+      "learning_rate": 1.413037439625593e-06,
+      "loss": 0.8243,
+      "step": 1980
+    },
+    {
+      "epoch": 0.76619501395707,
+      "grad_norm": 0.26743367314338684,
+      "learning_rate": 1.3698746695458847e-06,
+      "loss": 0.817,
+      "step": 1990
+    },
+    {
+      "epoch": 0.7700452401578592,
+      "grad_norm": 0.24756284058094025,
+      "learning_rate": 1.3272765235521745e-06,
+      "loss": 0.8095,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7738954663586486,
+      "grad_norm": 0.2359456568956375,
+      "learning_rate": 1.2852496272942877e-06,
+      "loss": 0.8161,
+      "step": 2010
+    },
+    {
+      "epoch": 0.7777456925594378,
+      "grad_norm": 0.2389555126428604,
+      "learning_rate": 1.2438005175707396e-06,
+      "loss": 0.8059,
+      "step": 2020
+    },
+    {
+      "epoch": 0.7815959187602272,
+      "grad_norm": 0.24247314035892487,
+      "learning_rate": 1.2029356413120297e-06,
+      "loss": 0.7926,
+      "step": 2030
+    },
+    {
+      "epoch": 0.7854461449610165,
+      "grad_norm": 0.25712308287620544,
+      "learning_rate": 1.1626613545778786e-06,
+      "loss": 0.8091,
+      "step": 2040
+    },
+    {
+      "epoch": 0.7892963711618057,
+      "grad_norm": 0.24122852087020874,
+      "learning_rate": 1.1229839215686294e-06,
+      "loss": 0.8133,
+      "step": 2050
+    },
+    {
+      "epoch": 0.7931465973625951,
+      "grad_norm": 0.22937846183776855,
+      "learning_rate": 1.0839095136509188e-06,
+      "loss": 0.7997,
+      "step": 2060
+    },
+    {
+      "epoch": 0.7969968235633843,
+      "grad_norm": 0.2698941230773926,
+      "learning_rate": 1.045444208397791e-06,
+      "loss": 0.822,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8008470497641736,
+      "grad_norm": 0.24451066553592682,
+      "learning_rate": 1.007593988643402e-06,
+      "loss": 0.8001,
+      "step": 2080
+    },
+    {
+      "epoch": 0.804697275964963,
+      "grad_norm": 0.24701625108718872,
+      "learning_rate": 9.703647415524613e-07,
+      "loss": 0.8223,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8085475021657522,
+      "grad_norm": 0.25038644671440125,
+      "learning_rate": 9.337622577045485e-07,
+      "loss": 0.8119,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8123977283665416,
+      "grad_norm": 0.2451173961162567,
+      "learning_rate": 8.977922301934588e-07,
+      "loss": 0.8187,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8162479545673308,
+      "grad_norm": 0.2495400309562683,
+      "learning_rate": 8.624602537417009e-07,
+      "loss": 0.8208,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8200981807681201,
+      "grad_norm": 0.2677256464958191,
+      "learning_rate": 8.277718238303123e-07,
+      "loss": 0.811,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8239484069689095,
+      "grad_norm": 0.2496872842311859,
+      "learning_rate": 7.937323358440935e-07,
+      "loss": 0.8132,
+      "step": 2140
+    },
+    {
+      "epoch": 0.8277986331696987,
+      "grad_norm": 0.2585522532463074,
+      "learning_rate": 7.603470842324223e-07,
+      "loss": 0.806,
+      "step": 2150
+    },
+    {
+      "epoch": 0.831648859370488,
+      "grad_norm": 0.2475339025259018,
+      "learning_rate": 7.276212616857575e-07,
+      "loss": 0.822,
+      "step": 2160
+    },
+    {
+      "epoch": 0.8354990855712773,
+      "grad_norm": 0.23316556215286255,
+      "learning_rate": 6.955599583279837e-07,
+      "loss": 0.8245,
+      "step": 2170
+    },
+    {
+      "epoch": 0.8393493117720666,
+      "grad_norm": 0.2505607306957245,
+      "learning_rate": 6.641681609246981e-07,
+      "loss": 0.8129,
+      "step": 2180
+    },
+    {
+      "epoch": 0.8431995379728559,
+      "grad_norm": 0.25390321016311646,
+      "learning_rate": 6.334507521075711e-07,
+      "loss": 0.8241,
+      "step": 2190
+    },
+    {
+      "epoch": 0.8470497641736452,
+      "grad_norm": 0.2795664668083191,
+      "learning_rate": 6.034125096149179e-07,
+      "loss": 0.8145,
+      "step": 2200
+    },
+    {
+      "epoch": 0.8508999903744345,
+      "grad_norm": 0.24296064674854279,
+      "learning_rate": 5.740581055485722e-07,
+      "loss": 0.7869,
+      "step": 2210
+    },
+    {
+      "epoch": 0.8547502165752238,
+      "grad_norm": 0.23110376298427582,
+      "learning_rate": 5.453921056471961e-07,
+      "loss": 0.8218,
+      "step": 2220
+    },
+    {
+      "epoch": 0.8586004427760131,
+      "grad_norm": 0.25192028284072876,
+      "learning_rate": 5.174189685761288e-07,
+      "loss": 0.8138,
+      "step": 2230
+    },
+    {
+      "epoch": 0.8624506689768023,
+      "grad_norm": 0.24650907516479492,
+      "learning_rate": 4.901430452338979e-07,
+      "loss": 0.8249,
+      "step": 2240
+    },
+    {
+      "epoch": 0.8663008951775917,
+      "grad_norm": 0.24833950400352478,
+      "learning_rate": 4.6356857807548286e-07,
+      "loss": 0.8103,
+      "step": 2250
+    },
+    {
+      "epoch": 0.870151121378381,
+      "grad_norm": 0.25431501865386963,
+      "learning_rate": 4.376997004524541e-07,
+      "loss": 0.7997,
+      "step": 2260
+    },
+    {
+      "epoch": 0.8740013475791703,
+      "grad_norm": 0.26715120673179626,
+      "learning_rate": 4.12540435970073e-07,
+      "loss": 0.8112,
+      "step": 2270
+    },
+    {
+      "epoch": 0.8778515737799596,
+      "grad_norm": 0.2367173135280609,
+      "learning_rate": 3.880946978614703e-07,
+      "loss": 0.8176,
+      "step": 2280
+    },
+    {
+      "epoch": 0.8817017999807488,
+      "grad_norm": 0.24484531581401825,
+      "learning_rate": 3.643662883789878e-07,
+      "loss": 0.8261,
+      "step": 2290
+    },
+    {
+      "epoch": 0.8855520261815382,
+      "grad_norm": 0.2280900925397873,
+      "learning_rate": 3.41358898202781e-07,
+      "loss": 0.8017,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8894022523823275,
+      "grad_norm": 0.23489460349082947,
+      "learning_rate": 3.190761058667741e-07,
+      "loss": 0.8126,
+      "step": 2310
+    },
+    {
+      "epoch": 0.8932524785831167,
+      "grad_norm": 0.245416060090065,
+      "learning_rate": 2.9752137720206463e-07,
+      "loss": 0.8202,
+      "step": 2320
+    },
+    {
+      "epoch": 0.8971027047839061,
+      "grad_norm": 0.24198678135871887,
+      "learning_rate": 2.7669806479785067e-07,
+      "loss": 0.8234,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9009529309846953,
+      "grad_norm": 0.24904951453208923,
+      "learning_rate": 2.566094074799763e-07,
+      "loss": 0.8198,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9048031571854847,
+      "grad_norm": 0.24282513558864594,
+      "learning_rate": 2.3725852980716557e-07,
+      "loss": 0.7983,
+      "step": 2350
+    },
+    {
+      "epoch": 0.908653383386274,
+      "grad_norm": 0.2402620017528534,
+      "learning_rate": 2.1864844158503983e-07,
+      "loss": 0.8119,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9125036095870632,
+      "grad_norm": 0.28123319149017334,
+      "learning_rate": 2.007820373979702e-07,
+      "loss": 0.8043,
+      "step": 2370
+    },
+    {
+      "epoch": 0.9163538357878526,
+      "grad_norm": 0.2445562779903412,
+      "learning_rate": 1.8366209615886532e-07,
+      "loss": 0.8121,
+      "step": 2380
+    },
+    {
+      "epoch": 0.9202040619886418,
+      "grad_norm": 0.24946707487106323,
+      "learning_rate": 1.6729128067693613e-07,
+      "loss": 0.821,
+      "step": 2390
+    },
+    {
+      "epoch": 0.9240542881894311,
+      "grad_norm": 0.2577192485332489,
+      "learning_rate": 1.5167213724353426e-07,
+      "loss": 0.8225,
+      "step": 2400
+    },
+    {
+      "epoch": 0.9279045143902205,
+      "grad_norm": 0.251924991607666,
+      "learning_rate": 1.3680709523610058e-07,
+      "loss": 0.8066,
+      "step": 2410
+    },
+    {
+      "epoch": 0.9317547405910097,
+      "grad_norm": 0.24295902252197266,
+      "learning_rate": 1.2269846674030418e-07,
+      "loss": 0.8287,
+      "step": 2420
+    },
+    {
+      "epoch": 0.935604966791799,
+      "grad_norm": 0.24939405918121338,
+      "learning_rate": 1.0934844619042595e-07,
+      "loss": 0.8251,
+      "step": 2430
+    },
+    {
+      "epoch": 0.9394551929925883,
+      "grad_norm": 0.25605061650276184,
+      "learning_rate": 9.675911002803961e-08,
+      "loss": 0.8151,
+      "step": 2440
+    },
+    {
+      "epoch": 0.9433054191933776,
+      "grad_norm": 0.2690197825431824,
+      "learning_rate": 8.493241637904381e-08,
+      "loss": 0.8124,
+      "step": 2450
+    },
+    {
+      "epoch": 0.9471556453941669,
+      "grad_norm": 0.25505584478378296,
+      "learning_rate": 7.387020474909689e-08,
+      "loss": 0.8317,
+      "step": 2460
+    },
+    {
+      "epoch": 0.9510058715949562,
+      "grad_norm": 0.26435327529907227,
+      "learning_rate": 6.357419573750856e-08,
+      "loss": 0.8264,
+      "step": 2470
+    },
+    {
+      "epoch": 0.9548560977957455,
+      "grad_norm": 0.24673746526241302,
+      "learning_rate": 5.4045990769613365e-08,
+      "loss": 0.8091,
+      "step": 2480
+    },
+    {
+      "epoch": 0.9587063239965348,
+      "grad_norm": 0.3056897521018982,
+      "learning_rate": 4.5287071847694434e-08,
+      "loss": 0.8152,
+      "step": 2490
+    },
+    {
+      "epoch": 0.9625565501973241,
+      "grad_norm": 0.23500357568264008,
+      "learning_rate": 3.7298801320467817e-08,
+      "loss": 0.8076,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9664067763981133,
+      "grad_norm": 0.24219276010990143,
+      "learning_rate": 3.0082421671192576e-08,
+      "loss": 0.8007,
+      "step": 2510
+    },
+    {
+      "epoch": 0.9702570025989027,
+      "grad_norm": 0.2531115710735321,
+      "learning_rate": 2.3639055324410355e-08,
+      "loss": 0.8223,
+      "step": 2520
+    },
+    {
+      "epoch": 0.974107228799692,
+      "grad_norm": 0.24507629871368408,
+      "learning_rate": 1.7969704471369477e-08,
+      "loss": 0.8074,
+      "step": 2530
+    },
+    {
+      "epoch": 0.9779574550004813,
+      "grad_norm": 0.23288261890411377,
+      "learning_rate": 1.3075250914142413e-08,
+      "loss": 0.8107,
+      "step": 2540
+    },
+    {
+      "epoch": 0.9818076812012706,
+      "grad_norm": 0.2483183890581131,
+      "learning_rate": 8.956455928476049e-09,
+      "loss": 0.8214,
+      "step": 2550
+    },
+    {
+      "epoch": 0.9856579074020598,
+      "grad_norm": 0.25237807631492615,
+      "learning_rate": 5.6139601453786275e-09,
+      "loss": 0.8133,
+      "step": 2560
+    },
+    {
+      "epoch": 0.9895081336028492,
+      "grad_norm": 0.2491920441389084,
+      "learning_rate": 3.0482834514805602e-09,
+      "loss": 0.8007,
+      "step": 2570
+    },
+    {
+      "epoch": 0.9933583598036385,
+      "grad_norm": 0.28852370381355286,
+      "learning_rate": 1.2598249081713366e-09,
+      "loss": 0.7962,
+      "step": 2580
+    },
+    {
+      "epoch": 0.9972085860044277,
+      "grad_norm": 0.26582247018814087,
+      "learning_rate": 2.488626895302826e-10,
+      "loss": 0.8244,
+      "step": 2590
+    },
+    {
+      "epoch": 0.9999037443449803,
+      "step": 2597,
+      "total_flos": 4.3227300419107553e+18,
+      "train_loss": 0.8415219830603704,
+      "train_runtime": 12782.9959,
+      "train_samples_per_second": 52.015,
+      "train_steps_per_second": 0.203
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2597,
+  "num_input_tokens_seen": 418855352,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.3227300419107553e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen3/webssl_mae300m_full2b_224/finetune/training_args.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:8f4d894a9fe9c23e46ba08449fd57d1df2ef01efb2e9d5c290c8022ea7fe2e89
+size 6264

continuous/qwen3/webssl_mae300m_full2b_224/finetune/vocab.json ADDED Viewed

The diff for this file is too large to render. See raw diff

continuous/qwen3/webssl_mae300m_full2b_224/pretrain/config.json ADDED Viewed

	@@ -0,0 +1,89 @@

+{
+  "add_faster_video": false,
+  "add_time_instruction": false,
+  "architectures": [
+    "Qwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "faster_token_stride": 10,
+  "force_sample": false,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "image_aspect_ratio": "square",
+  "image_crop_resolution": null,
+  "image_grid_pinpoints": null,
+  "image_split_resolution": null,
+  "initializer_range": 0.02,
+  "intermediate_size": 6144,
+  "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": 40960,
+  "max_window_layers": 28,
+  "mm_hidden_size": 1024,
+  "mm_newline_position": "grid",
+  "mm_patch_merge_type": "flat",
+  "mm_projector_lr": null,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_resampler_type": null,
+  "mm_spatial_pool_mode": "bilinear",
+  "mm_spatial_pool_stride": null,
+  "mm_tunable_parts": "mm_mlp_adapter",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -1,
+  "mm_vision_tower": "hf:/cache/ckpt/download/tokenizer/continuous/webssl/webssl-mae300m-full2b-224",
+  "mm_vision_tower_lr": null,
+  "model_type": "qwen3",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 8,
+  "pad_token_id": 151643,
+  "pos_skipping_range": 4096,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000,
+  "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/qwen3/webssl_mae300m_full2b_224/pretrain/mm_projector.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:4ecb2bc2325cd921fa2424fc7d2dc001794cf6ee827f2b5c842d993f9dc558e9
+size 25184312

continuous/qwen3/webssl_mae300m_full2b_224/pretrain/trainer_state.json ADDED Viewed

	@@ -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": 20.19011688232422,
+      "learning_rate": 0.00032142857142857147,
+      "loss": 7.5539,
+      "step": 10
+    },
+    {
+      "epoch": 0.03669724770642202,
+      "grad_norm": 0.93219393491745,
+      "learning_rate": 0.0006785714285714287,
+      "loss": 5.2029,
+      "step": 20
+    },
+    {
+      "epoch": 0.05504587155963303,
+      "grad_norm": 0.22062698006629944,
+      "learning_rate": 0.0009999907688176173,
+      "loss": 4.7917,
+      "step": 30
+    },
+    {
+      "epoch": 0.07339449541284404,
+      "grad_norm": 0.1880311816930771,
+      "learning_rate": 0.0009988834393115766,
+      "loss": 4.6865,
+      "step": 40
+    },
+    {
+      "epoch": 0.09174311926605505,
+      "grad_norm": 0.08846257627010345,
+      "learning_rate": 0.0009959345572940772,
+      "loss": 4.6426,
+      "step": 50
+    },
+    {
+      "epoch": 0.11009174311926606,
+      "grad_norm": 0.25353196263313293,
+      "learning_rate": 0.0009911550081155983,
+      "loss": 4.6201,
+      "step": 60
+    },
+    {
+      "epoch": 0.12844036697247707,
+      "grad_norm": 0.15428447723388672,
+      "learning_rate": 0.0009845624347566433,
+      "loss": 4.5943,
+      "step": 70
+    },
+    {
+      "epoch": 0.14678899082568808,
+      "grad_norm": 0.1874399334192276,
+      "learning_rate": 0.0009761811727013547,
+      "loss": 4.5878,
+      "step": 80
+    },
+    {
+      "epoch": 0.1651376146788991,
+      "grad_norm": 0.12632235884666443,
+      "learning_rate": 0.0009660421601067666,
+      "loss": 4.5632,
+      "step": 90
+    },
+    {
+      "epoch": 0.1834862385321101,
+      "grad_norm": 0.1696982979774475,
+      "learning_rate": 0.0009541828235992926,
+      "loss": 4.5522,
+      "step": 100
+    },
+    {
+      "epoch": 0.2018348623853211,
+      "grad_norm": 0.15228746831417084,
+      "learning_rate": 0.0009406469401200151,
+      "loss": 4.5314,
+      "step": 110
+    },
+    {
+      "epoch": 0.22018348623853212,
+      "grad_norm": 0.2892380952835083,
+      "learning_rate": 0.0009254844753287492,
+      "loss": 4.5382,
+      "step": 120
+    },
+    {
+      "epoch": 0.23853211009174313,
+      "grad_norm": 0.22636963427066803,
+      "learning_rate": 0.0009087513991633924,
+      "loss": 4.5139,
+      "step": 130
+    },
+    {
+      "epoch": 0.25688073394495414,
+      "grad_norm": 0.20649690926074982,
+      "learning_rate": 0.0008905094792353917,
+      "loss": 4.4975,
+      "step": 140
+    },
+    {
+      "epoch": 0.27522935779816515,
+      "grad_norm": 0.18735134601593018,
+      "learning_rate": 0.0008708260528239789,
+      "loss": 4.4944,
+      "step": 150
+    },
+    {
+      "epoch": 0.29357798165137616,
+      "grad_norm": 0.3019122779369354,
+      "learning_rate": 0.0008497737783108237,
+      "loss": 4.4861,
+      "step": 160
+    },
+    {
+      "epoch": 0.3119266055045872,
+      "grad_norm": 0.18039239943027496,
+      "learning_rate": 0.0008274303669726426,
+      "loss": 4.4719,
+      "step": 170
+    },
+    {
+      "epoch": 0.3302752293577982,
+      "grad_norm": 0.21757178008556366,
+      "learning_rate": 0.0008038782961218137,
+      "loss": 4.4571,
+      "step": 180
+    },
+    {
+      "epoch": 0.3486238532110092,
+      "grad_norm": 0.24669215083122253,
+      "learning_rate": 0.000779204504653894,
+      "loss": 4.4587,
+      "step": 190
+    },
+    {
+      "epoch": 0.3669724770642202,
+      "grad_norm": 0.21597912907600403,
+      "learning_rate": 0.0007535000721258791,
+      "loss": 4.4369,
+      "step": 200
+    },
+    {
+      "epoch": 0.3853211009174312,
+      "grad_norm": 0.35777223110198975,
+      "learning_rate": 0.00072685988254984,
+      "loss": 4.4319,
+      "step": 210
+    },
+    {
+      "epoch": 0.4036697247706422,
+      "grad_norm": 0.29724183678627014,
+      "learning_rate": 0.0006993822741429906,
+      "loss": 4.4452,
+      "step": 220
+    },
+    {
+      "epoch": 0.42201834862385323,
+      "grad_norm": 0.2329730987548828,
+      "learning_rate": 0.0006711686763270818,
+      "loss": 4.4442,
+      "step": 230
+    },
+    {
+      "epoch": 0.44036697247706424,
+      "grad_norm": 0.14722830057144165,
+      "learning_rate": 0.0006423232353170831,
+      "loss": 4.4289,
+      "step": 240
+    },
+    {
+      "epoch": 0.45871559633027525,
+      "grad_norm": 0.20723925530910492,
+      "learning_rate": 0.0006129524296812335,
+      "loss": 4.4115,
+      "step": 250
+    },
+    {
+      "epoch": 0.47706422018348627,
+      "grad_norm": 0.21002638339996338,
+      "learning_rate": 0.0005831646772915651,
+      "loss": 4.4209,
+      "step": 260
+    },
+    {
+      "epoch": 0.4954128440366973,
+      "grad_norm": 0.23184272646903992,
+      "learning_rate": 0.0005530699351157809,
+      "loss": 4.4019,
+      "step": 270
+    },
+    {
+      "epoch": 0.5137614678899083,
+      "grad_norm": 0.2037023901939392,
+      "learning_rate": 0.0005227792933277943,
+      "loss": 4.4018,
+      "step": 280
+    },
+    {
+      "epoch": 0.5321100917431193,
+      "grad_norm": 0.21283107995986938,
+      "learning_rate": 0.0004924045652352092,
+      "loss": 4.3885,
+      "step": 290
+    },
+    {
+      "epoch": 0.5504587155963303,
+      "grad_norm": 0.19201147556304932,
+      "learning_rate": 0.0004620578745374633,
+      "loss": 4.3887,
+      "step": 300
+    },
+    {
+      "epoch": 0.5688073394495413,
+      "grad_norm": 0.21232424676418304,
+      "learning_rate": 0.0004318512414382059,
+      "loss": 4.4018,
+      "step": 310
+    },
+    {
+      "epoch": 0.5871559633027523,
+      "grad_norm": 0.25659358501434326,
+      "learning_rate": 0.0004018961691397148,
+      "loss": 4.3923,
+      "step": 320
+    },
+    {
+      "epoch": 0.6055045871559633,
+      "grad_norm": 0.17947867512702942,
+      "learning_rate": 0.0003723032322457458,
+      "loss": 4.3753,
+      "step": 330
+    },
+    {
+      "epoch": 0.6238532110091743,
+      "grad_norm": 0.23180800676345825,
+      "learning_rate": 0.0003431816685921625,
+      "loss": 4.3733,
+      "step": 340
+    },
+    {
+      "epoch": 0.6422018348623854,
+      "grad_norm": 0.1882370114326477,
+      "learning_rate": 0.0003146389760120416,
+      "loss": 4.3855,
+      "step": 350
+    },
+    {
+      "epoch": 0.6605504587155964,
+      "grad_norm": 0.2762691080570221,
+      "learning_rate": 0.00028678051552373485,
+      "loss": 4.3879,
+      "step": 360
+    },
+    {
+      "epoch": 0.6788990825688074,
+      "grad_norm": 0.16338391602039337,
+      "learning_rate": 0.0002597091224066581,
+      "loss": 4.3718,
+      "step": 370
+    },
+    {
+      "epoch": 0.6972477064220184,
+      "grad_norm": 0.1669435054063797,
+      "learning_rate": 0.0002335247266004629,
+      "loss": 4.389,
+      "step": 380
+    },
+    {
+      "epoch": 0.7155963302752294,
+      "grad_norm": 0.21548154950141907,
+      "learning_rate": 0.000208323983828828,
+      "loss": 4.3697,
+      "step": 390
+    },
+    {
+      "epoch": 0.7339449541284404,
+      "grad_norm": 0.23686401546001434,
+      "learning_rate": 0.0001841999188095224,
+      "loss": 4.371,
+      "step": 400
+    },
+    {
+      "epoch": 0.7522935779816514,
+      "grad_norm": 0.1486746370792389,
+      "learning_rate": 0.00016124158186777676,
+      "loss": 4.3708,
+      "step": 410
+    },
+    {
+      "epoch": 0.7706422018348624,
+      "grad_norm": 0.17956212162971497,
+      "learning_rate": 0.0001395337202205211,
+      "loss": 4.3629,
+      "step": 420
+    },
+    {
+      "epoch": 0.7889908256880734,
+      "grad_norm": 0.17735496163368225,
+      "learning_rate": 0.00011915646514489292,
+      "loss": 4.3706,
+      "step": 430
+    },
+    {
+      "epoch": 0.8073394495412844,
+      "grad_norm": 0.18571214377880096,
+      "learning_rate": 0.00010018503618578818,
+      "loss": 4.3766,
+      "step": 440
+    },
+    {
+      "epoch": 0.8256880733944955,
+      "grad_norm": 0.14848405122756958,
+      "learning_rate": 8.268946349432583e-05,
+      "loss": 4.3667,
+      "step": 450
+    },
+    {
+      "epoch": 0.8440366972477065,
+      "grad_norm": 0.14829348027706146,
+      "learning_rate": 6.673432932217117e-05,
+      "loss": 4.3597,
+      "step": 460
+    },
+    {
+      "epoch": 0.8623853211009175,
+      "grad_norm": 0.17006702721118927,
+      "learning_rate": 5.2378529625954684e-05,
+      "loss": 4.3445,
+      "step": 470
+    },
+    {
+      "epoch": 0.8807339449541285,
+      "grad_norm": 0.12240241467952728,
+      "learning_rate": 3.9675056661785556e-05,
+      "loss": 4.3566,
+      "step": 480
+    },
+    {
+      "epoch": 0.8990825688073395,
+      "grad_norm": 0.13030721247196198,
+      "learning_rate": 2.8670803372379538e-05,
+      "loss": 4.3605,
+      "step": 490
+    },
+    {
+      "epoch": 0.9174311926605505,
+      "grad_norm": 0.1439724564552307,
+      "learning_rate": 1.9406390288876586e-05,
+      "loss": 4.3696,
+      "step": 500
+    },
+    {
+      "epoch": 0.9357798165137615,
+      "grad_norm": 0.12027649581432343,
+      "learning_rate": 1.191601558631461e-05,
+      "loss": 4.3599,
+      "step": 510
+    },
+    {
+      "epoch": 0.9541284403669725,
+      "grad_norm": 0.11720609664916992,
+      "learning_rate": 6.2273288462559305e-06,
+      "loss": 4.3494,
+      "step": 520
+    },
+    {
+      "epoch": 0.9724770642201835,
+      "grad_norm": 0.11378282308578491,
+      "learning_rate": 2.361328992552314e-06,
+      "loss": 4.3562,
+      "step": 530
+    },
+    {
+      "epoch": 0.9908256880733946,
+      "grad_norm": 0.12794020771980286,
+      "learning_rate": 3.322867770044202e-07,
+      "loss": 4.3618,
+      "step": 540
+    },
+    {
+      "epoch": 1.0,
+      "step": 545,
+      "total_flos": 3.420461531618345e+17,
+      "train_loss": 4.512503237243092,
+      "train_runtime": 5089.194,
+      "train_samples_per_second": 109.669,
+      "train_steps_per_second": 0.107
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 545,
+  "num_input_tokens_seen": 33142912,
+  "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": 3.420461531618345e+17,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen3/webssl_mae3b_full2b_224/finetune/added_tokens.json ADDED Viewed

	@@ -0,0 +1,28 @@

+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|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/qwen3/webssl_mae3b_full2b_224/finetune/chat_template.jinja ADDED Viewed

	@@ -0,0 +1,89 @@

+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# 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' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- 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' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

continuous/qwen3/webssl_mae3b_full2b_224/finetune/config.json ADDED Viewed

	@@ -0,0 +1,89 @@

+{
+  "add_faster_video": false,
+  "add_time_instruction": false,
+  "architectures": [
+    "LlavaQwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "faster_token_stride": 10,
+  "force_sample": false,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "image_aspect_ratio": "square",
+  "image_crop_resolution": null,
+  "image_grid_pinpoints": null,
+  "image_split_resolution": null,
+  "initializer_range": 0.02,
+  "intermediate_size": 6144,
+  "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": 40960,
+  "max_window_layers": 28,
+  "mm_hidden_size": 3072,
+  "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-mae3b-full2b-224",
+  "mm_vision_tower_lr": null,
+  "model_type": "qwen3",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 8,
+  "pad_token_id": 151643,
+  "pos_skipping_range": 4096,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000,
+  "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/qwen3/webssl_mae3b_full2b_224/finetune/generation_config.json ADDED Viewed

	@@ -0,0 +1,12 @@

+{
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "temperature": 0.6,
+  "top_k": 20,
+  "top_p": 0.95,
+  "transformers_version": "4.57.6"
+}

continuous/qwen3/webssl_mae3b_full2b_224/finetune/merges.txt ADDED Viewed

The diff for this file is too large to render. See raw diff

continuous/qwen3/webssl_mae3b_full2b_224/finetune/model-00001-of-00003.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:b53d3d9beefcaf5e2a4122766084d1ef8f61fff3d95ac3dd5225940e183fc2bd
+size 4956883872