lyunee commited on 17 days ago

Commit

ebd58d2

verified ·

1 Parent(s): 05793ab

Upload continuous evaluation checkpoints, batch 8

Browse files

Files changed (47) hide show

.gitattributes +3 -0
continuous/qwen3/eupe_vit_b/finetune/generation_config.json +12 -0
continuous/qwen3/eupe_vit_b/finetune/merges.txt +0 -0
continuous/qwen3/eupe_vit_b/finetune/model.safetensors +3 -0
continuous/qwen3/eupe_vit_b/finetune/special_tokens_map.json +31 -0
continuous/qwen3/eupe_vit_b/finetune/tokenizer.json +3 -0
continuous/qwen3/eupe_vit_b/finetune/tokenizer_config.json +240 -0
continuous/qwen3/eupe_vit_b/finetune/trainer_state.json +1856 -0
continuous/qwen3/eupe_vit_b/finetune/training_args.bin +3 -0
continuous/qwen3/eupe_vit_b/finetune/vocab.json +0 -0
continuous/qwen3/eupe_vit_b/pretrain/config.json +89 -0
continuous/qwen3/eupe_vit_b/pretrain/mm_projector.bin +3 -0
continuous/qwen3/eupe_vit_b/pretrain/trainer_state.json +421 -0
continuous/qwen3/eupe_vit_s/finetune/added_tokens.json +28 -0
continuous/qwen3/eupe_vit_s/finetune/chat_template.jinja +89 -0
continuous/qwen3/eupe_vit_s/finetune/config.json +89 -0
continuous/qwen3/eupe_vit_s/finetune/generation_config.json +12 -0
continuous/qwen3/eupe_vit_s/finetune/merges.txt +0 -0
continuous/qwen3/eupe_vit_s/finetune/model.safetensors +3 -0
continuous/qwen3/eupe_vit_s/finetune/special_tokens_map.json +31 -0
continuous/qwen3/eupe_vit_s/finetune/tokenizer.json +3 -0
continuous/qwen3/eupe_vit_s/finetune/tokenizer_config.json +240 -0
continuous/qwen3/eupe_vit_s/finetune/trainer_state.json +1856 -0
continuous/qwen3/eupe_vit_s/finetune/training_args.bin +3 -0
continuous/qwen3/eupe_vit_s/finetune/vocab.json +0 -0
continuous/qwen3/eupe_vit_s/pretrain/config.json +89 -0
continuous/qwen3/eupe_vit_s/pretrain/mm_projector.bin +3 -0
continuous/qwen3/eupe_vit_s/pretrain/trainer_state.json +421 -0
continuous/qwen3/eupe_vit_t/finetune/added_tokens.json +28 -0
continuous/qwen3/eupe_vit_t/finetune/chat_template.jinja +89 -0
continuous/qwen3/eupe_vit_t/finetune/config.json +89 -0
continuous/qwen3/eupe_vit_t/finetune/generation_config.json +12 -0
continuous/qwen3/eupe_vit_t/finetune/merges.txt +0 -0
continuous/qwen3/eupe_vit_t/finetune/model.safetensors +3 -0
continuous/qwen3/eupe_vit_t/finetune/special_tokens_map.json +31 -0
continuous/qwen3/eupe_vit_t/finetune/tokenizer.json +3 -0
continuous/qwen3/eupe_vit_t/finetune/tokenizer_config.json +240 -0
continuous/qwen3/eupe_vit_t/finetune/trainer_state.json +1856 -0
continuous/qwen3/eupe_vit_t/finetune/training_args.bin +3 -0
continuous/qwen3/eupe_vit_t/finetune/vocab.json +0 -0
continuous/qwen3/eupe_vit_t/pretrain/config.json +89 -0
continuous/qwen3/eupe_vit_t/pretrain/mm_projector.bin +3 -0
continuous/qwen3/eupe_vit_t/pretrain/trainer_state.json +421 -0
continuous/qwen3/ijepa_vith14/finetune/added_tokens.json +28 -0
continuous/qwen3/ijepa_vith14/finetune/config.json +59 -0
continuous/qwen3/ijepa_vith14/finetune/generation_config.json +13 -0
continuous/qwen3/ijepa_vith14/finetune/merges.txt +0 -0

.gitattributes CHANGED Viewed

@@ -122,3 +122,6 @@ continuous/qwen3/dinov2_large/finetune/tokenizer.json filter=lfs diff=lfs merge=
 continuous/qwen3/dinov2_small/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 continuous/qwen3/dinov3_vitl16/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 continuous/qwen3/eupe_convnext_b/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text

 continuous/qwen3/dinov2_small/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 continuous/qwen3/dinov3_vitl16/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 continuous/qwen3/eupe_convnext_b/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen3/eupe_vit_b/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen3/eupe_vit_s/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+continuous/qwen3/eupe_vit_t/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text

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

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

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

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:86002db5af4f2533cdde1036369ad8557d05b72a4026a3ee6fa6f5b27c023df6
+size 4258019728

continuous/qwen3/eupe_vit_b/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/eupe_vit_b/finetune/tokenizer.json ADDED Viewed

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

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

continuous/qwen3/eupe_vit_b/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/eupe_vit_b/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": 16.140653610229492,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 2.145,
+      "step": 10
+    },
+    {
+      "epoch": 0.007700452401578592,
+      "grad_norm": 6.641376972198486,
+      "learning_rate": 2.435897435897436e-06,
+      "loss": 1.841,
+      "step": 20
+    },
+    {
+      "epoch": 0.01155067860236789,
+      "grad_norm": 4.4970550537109375,
+      "learning_rate": 3.7179487179487184e-06,
+      "loss": 1.5246,
+      "step": 30
+    },
+    {
+      "epoch": 0.015400904803157185,
+      "grad_norm": 1.3125383853912354,
+      "learning_rate": 5e-06,
+      "loss": 1.201,
+      "step": 40
+    },
+    {
+      "epoch": 0.019251131003946483,
+      "grad_norm": 0.8419232368469238,
+      "learning_rate": 6.282051282051282e-06,
+      "loss": 1.0846,
+      "step": 50
+    },
+    {
+      "epoch": 0.02310135720473578,
+      "grad_norm": 0.4620702564716339,
+      "learning_rate": 7.564102564102564e-06,
+      "loss": 0.9976,
+      "step": 60
+    },
+    {
+      "epoch": 0.026951583405525073,
+      "grad_norm": 0.3452049791812897,
+      "learning_rate": 8.846153846153847e-06,
+      "loss": 0.9463,
+      "step": 70
+    },
+    {
+      "epoch": 0.03080180960631437,
+      "grad_norm": 0.3181668221950531,
+      "learning_rate": 9.999996111488723e-06,
+      "loss": 0.9439,
+      "step": 80
+    },
+    {
+      "epoch": 0.034652035807103666,
+      "grad_norm": 0.3164377212524414,
+      "learning_rate": 9.999529497453782e-06,
+      "loss": 0.9115,
+      "step": 90
+    },
+    {
+      "epoch": 0.038502262007892966,
+      "grad_norm": 0.2945152521133423,
+      "learning_rate": 9.998285264324193e-06,
+      "loss": 0.8931,
+      "step": 100
+    },
+    {
+      "epoch": 0.04235248820868226,
+      "grad_norm": 0.3400074541568756,
+      "learning_rate": 9.996263605626052e-06,
+      "loss": 0.8741,
+      "step": 110
+    },
+    {
+      "epoch": 0.04620271440947156,
+      "grad_norm": 0.2791471779346466,
+      "learning_rate": 9.99346483580503e-06,
+      "loss": 0.884,
+      "step": 120
+    },
+    {
+      "epoch": 0.05005294061026085,
+      "grad_norm": 0.27801376581192017,
+      "learning_rate": 9.989889390177467e-06,
+      "loss": 0.8788,
+      "step": 130
+    },
+    {
+      "epoch": 0.053903166811050146,
+      "grad_norm": 0.31570783257484436,
+      "learning_rate": 9.985537824862647e-06,
+      "loss": 0.8726,
+      "step": 140
+    },
+    {
+      "epoch": 0.057753393011839446,
+      "grad_norm": 0.35889339447021484,
+      "learning_rate": 9.980410816696317e-06,
+      "loss": 0.8577,
+      "step": 150
+    },
+    {
+      "epoch": 0.06160361921262874,
+      "grad_norm": 0.29259490966796875,
+      "learning_rate": 9.97450916312541e-06,
+      "loss": 0.8623,
+      "step": 160
+    },
+    {
+      "epoch": 0.06545384541341803,
+      "grad_norm": 0.26671141386032104,
+      "learning_rate": 9.967833782084004e-06,
+      "loss": 0.8624,
+      "step": 170
+    },
+    {
+      "epoch": 0.06930407161420733,
+      "grad_norm": 0.29952603578567505,
+      "learning_rate": 9.960385711850554e-06,
+      "loss": 0.8907,
+      "step": 180
+    },
+    {
+      "epoch": 0.07315429781499663,
+      "grad_norm": 0.3072148859500885,
+      "learning_rate": 9.952166110886394e-06,
+      "loss": 0.8794,
+      "step": 190
+    },
+    {
+      "epoch": 0.07700452401578593,
+      "grad_norm": 0.3318941295146942,
+      "learning_rate": 9.943176257655567e-06,
+      "loss": 0.869,
+      "step": 200
+    },
+    {
+      "epoch": 0.08085475021657522,
+      "grad_norm": 0.286873459815979,
+      "learning_rate": 9.933417550425952e-06,
+      "loss": 0.8627,
+      "step": 210
+    },
+    {
+      "epoch": 0.08470497641736452,
+      "grad_norm": 0.28734204173088074,
+      "learning_rate": 9.9228915070518e-06,
+      "loss": 0.8512,
+      "step": 220
+    },
+    {
+      "epoch": 0.08855520261815382,
+      "grad_norm": 0.29912155866622925,
+      "learning_rate": 9.911599764737632e-06,
+      "loss": 0.8512,
+      "step": 230
+    },
+    {
+      "epoch": 0.09240542881894312,
+      "grad_norm": 0.2960100769996643,
+      "learning_rate": 9.899544079783606e-06,
+      "loss": 0.846,
+      "step": 240
+    },
+    {
+      "epoch": 0.0962556550197324,
+      "grad_norm": 0.33301469683647156,
+      "learning_rate": 9.886726327312338e-06,
+      "loss": 0.8624,
+      "step": 250
+    },
+    {
+      "epoch": 0.1001058812205217,
+      "grad_norm": 0.294306218624115,
+      "learning_rate": 9.87314850097724e-06,
+      "loss": 0.8602,
+      "step": 260
+    },
+    {
+      "epoch": 0.103956107421311,
+      "grad_norm": 0.277153342962265,
+      "learning_rate": 9.858812712652452e-06,
+      "loss": 0.8469,
+      "step": 270
+    },
+    {
+      "epoch": 0.10780633362210029,
+      "grad_norm": 0.29342925548553467,
+      "learning_rate": 9.843721192104341e-06,
+      "loss": 0.8469,
+      "step": 280
+    },
+    {
+      "epoch": 0.11165655982288959,
+      "grad_norm": 0.2856411337852478,
+      "learning_rate": 9.827876286644698e-06,
+      "loss": 0.8637,
+      "step": 290
+    },
+    {
+      "epoch": 0.11550678602367889,
+      "grad_norm": 0.26524317264556885,
+      "learning_rate": 9.811280460765643e-06,
+      "loss": 0.824,
+      "step": 300
+    },
+    {
+      "epoch": 0.11935701222446819,
+      "grad_norm": 0.2696005403995514,
+      "learning_rate": 9.793936295756292e-06,
+      "loss": 0.8379,
+      "step": 310
+    },
+    {
+      "epoch": 0.12320723842525748,
+      "grad_norm": 0.28368768095970154,
+      "learning_rate": 9.775846489301279e-06,
+      "loss": 0.8375,
+      "step": 320
+    },
+    {
+      "epoch": 0.12705746462604678,
+      "grad_norm": 0.30544790625572205,
+      "learning_rate": 9.757013855061157e-06,
+      "loss": 0.8371,
+      "step": 330
+    },
+    {
+      "epoch": 0.13090769082683607,
+      "grad_norm": 0.290192186832428,
+      "learning_rate": 9.737441322234758e-06,
+      "loss": 0.8484,
+      "step": 340
+    },
+    {
+      "epoch": 0.13475791702762538,
+      "grad_norm": 0.28923872113227844,
+      "learning_rate": 9.717131935103605e-06,
+      "loss": 0.8475,
+      "step": 350
+    },
+    {
+      "epoch": 0.13860814322841467,
+      "grad_norm": 0.28200894594192505,
+      "learning_rate": 9.696088852558398e-06,
+      "loss": 0.8343,
+      "step": 360
+    },
+    {
+      "epoch": 0.14245836942920398,
+      "grad_norm": 0.2817835509777069,
+      "learning_rate": 9.674315347607686e-06,
+      "loss": 0.8611,
+      "step": 370
+    },
+    {
+      "epoch": 0.14630859562999327,
+      "grad_norm": 0.2825669050216675,
+      "learning_rate": 9.651814806868796e-06,
+      "loss": 0.8206,
+      "step": 380
+    },
+    {
+      "epoch": 0.15015882183078255,
+      "grad_norm": 0.2736653685569763,
+      "learning_rate": 9.62859073004107e-06,
+      "loss": 0.8287,
+      "step": 390
+    },
+    {
+      "epoch": 0.15400904803157187,
+      "grad_norm": 0.2865540385246277,
+      "learning_rate": 9.604646729361547e-06,
+      "loss": 0.8322,
+      "step": 400
+    },
+    {
+      "epoch": 0.15785927423236115,
+      "grad_norm": 0.2636553943157196,
+      "learning_rate": 9.579986529043099e-06,
+      "loss": 0.8205,
+      "step": 410
+    },
+    {
+      "epoch": 0.16170950043315044,
+      "grad_norm": 0.26667794585227966,
+      "learning_rate": 9.554613964695189e-06,
+      "loss": 0.8364,
+      "step": 420
+    },
+    {
+      "epoch": 0.16555972663393975,
+      "grad_norm": 0.30360180139541626,
+      "learning_rate": 9.528532982727281e-06,
+      "loss": 0.8257,
+      "step": 430
+    },
+    {
+      "epoch": 0.16940995283472904,
+      "grad_norm": 0.28824734687805176,
+      "learning_rate": 9.501747639735018e-06,
+      "loss": 0.8241,
+      "step": 440
+    },
+    {
+      "epoch": 0.17326017903551832,
+      "grad_norm": 0.279202938079834,
+      "learning_rate": 9.474262101869272e-06,
+      "loss": 0.8133,
+      "step": 450
+    },
+    {
+      "epoch": 0.17711040523630764,
+      "grad_norm": 0.2998035252094269,
+      "learning_rate": 9.446080644188141e-06,
+      "loss": 0.8188,
+      "step": 460
+    },
+    {
+      "epoch": 0.18096063143709692,
+      "grad_norm": 0.26863977313041687,
+      "learning_rate": 9.417207649992017e-06,
+      "loss": 0.8282,
+      "step": 470
+    },
+    {
+      "epoch": 0.18481085763788624,
+      "grad_norm": 0.27271294593811035,
+      "learning_rate": 9.387647610141816e-06,
+      "loss": 0.8263,
+      "step": 480
+    },
+    {
+      "epoch": 0.18866108383867553,
+      "grad_norm": 0.26688480377197266,
+      "learning_rate": 9.35740512236047e-06,
+      "loss": 0.8128,
+      "step": 490
+    },
+    {
+      "epoch": 0.1925113100394648,
+      "grad_norm": 0.25128689408302307,
+      "learning_rate": 9.326484890517809e-06,
+      "loss": 0.8175,
+      "step": 500
+    },
+    {
+      "epoch": 0.19636153624025413,
+      "grad_norm": 0.2827357053756714,
+      "learning_rate": 9.294891723898926e-06,
+      "loss": 0.8137,
+      "step": 510
+    },
+    {
+      "epoch": 0.2002117624410434,
+      "grad_norm": 0.29749804735183716,
+      "learning_rate": 9.262630536456153e-06,
+      "loss": 0.8431,
+      "step": 520
+    },
+    {
+      "epoch": 0.2040619886418327,
+      "grad_norm": 0.3041918873786926,
+      "learning_rate": 9.229706346044749e-06,
+      "loss": 0.8144,
+      "step": 530
+    },
+    {
+      "epoch": 0.207912214842622,
+      "grad_norm": 0.29985082149505615,
+      "learning_rate": 9.196124273642422e-06,
+      "loss": 0.8283,
+      "step": 540
+    },
+    {
+      "epoch": 0.2117624410434113,
+      "grad_norm": 0.2646193504333496,
+      "learning_rate": 9.161889542552842e-06,
+      "loss": 0.83,
+      "step": 550
+    },
+    {
+      "epoch": 0.21561266724420058,
+      "grad_norm": 0.24975819885730743,
+      "learning_rate": 9.127007477593193e-06,
+      "loss": 0.8344,
+      "step": 560
+    },
+    {
+      "epoch": 0.2194628934449899,
+      "grad_norm": 0.3002760112285614,
+      "learning_rate": 9.091483504265974e-06,
+      "loss": 0.8439,
+      "step": 570
+    },
+    {
+      "epoch": 0.22331311964577918,
+      "grad_norm": 0.28924721479415894,
+      "learning_rate": 9.055323147915116e-06,
+      "loss": 0.8232,
+      "step": 580
+    },
+    {
+      "epoch": 0.2271633458465685,
+      "grad_norm": 0.2965131103992462,
+      "learning_rate": 9.018532032866585e-06,
+      "loss": 0.8339,
+      "step": 590
+    },
+    {
+      "epoch": 0.23101357204735778,
+      "grad_norm": 0.2788935601711273,
+      "learning_rate": 8.981115881553579e-06,
+      "loss": 0.825,
+      "step": 600
+    },
+    {
+      "epoch": 0.23486379824814707,
+      "grad_norm": 0.29534125328063965,
+      "learning_rate": 8.943080513626476e-06,
+      "loss": 0.8442,
+      "step": 610
+    },
+    {
+      "epoch": 0.23871402444893638,
+      "grad_norm": 0.27585574984550476,
+      "learning_rate": 8.904431845047641e-06,
+      "loss": 0.8287,
+      "step": 620
+    },
+    {
+      "epoch": 0.24256425064972567,
+      "grad_norm": 0.28762170672416687,
+      "learning_rate": 8.86517588717128e-06,
+      "loss": 0.8455,
+      "step": 630
+    },
+    {
+      "epoch": 0.24641447685051496,
+      "grad_norm": 0.2551642954349518,
+      "learning_rate": 8.82531874580844e-06,
+      "loss": 0.8342,
+      "step": 640
+    },
+    {
+      "epoch": 0.25026470305130427,
+      "grad_norm": 0.2592095136642456,
+      "learning_rate": 8.784866620277314e-06,
+      "loss": 0.7908,
+      "step": 650
+    },
+    {
+      "epoch": 0.25411492925209356,
+      "grad_norm": 0.29251375794410706,
+      "learning_rate": 8.743825802439018e-06,
+      "loss": 0.8282,
+      "step": 660
+    },
+    {
+      "epoch": 0.25796515545288284,
+      "grad_norm": 0.30638477206230164,
+      "learning_rate": 8.702202675718956e-06,
+      "loss": 0.8335,
+      "step": 670
+    },
+    {
+      "epoch": 0.26181538165367213,
+      "grad_norm": 0.27329686284065247,
+      "learning_rate": 8.660003714113959e-06,
+      "loss": 0.8035,
+      "step": 680
+    },
+    {
+      "epoch": 0.26566560785446147,
+      "grad_norm": 0.26749634742736816,
+      "learning_rate": 8.617235481185323e-06,
+      "loss": 0.8284,
+      "step": 690
+    },
+    {
+      "epoch": 0.26951583405525076,
+      "grad_norm": 0.2716856896877289,
+      "learning_rate": 8.573904629037931e-06,
+      "loss": 0.8338,
+      "step": 700
+    },
+    {
+      "epoch": 0.27336606025604004,
+      "grad_norm": 0.2683008313179016,
+      "learning_rate": 8.530017897285585e-06,
+      "loss": 0.8134,
+      "step": 710
+    },
+    {
+      "epoch": 0.27721628645682933,
+      "grad_norm": 0.26128676533699036,
+      "learning_rate": 8.485582112002746e-06,
+      "loss": 0.8173,
+      "step": 720
+    },
+    {
+      "epoch": 0.2810665126576186,
+      "grad_norm": 0.2963482141494751,
+      "learning_rate": 8.440604184662811e-06,
+      "loss": 0.8236,
+      "step": 730
+    },
+    {
+      "epoch": 0.28491673885840796,
+      "grad_norm": 0.2846742868423462,
+      "learning_rate": 8.395091111063118e-06,
+      "loss": 0.8116,
+      "step": 740
+    },
+    {
+      "epoch": 0.28876696505919724,
+      "grad_norm": 0.2504591643810272,
+      "learning_rate": 8.349049970236822e-06,
+      "loss": 0.8107,
+      "step": 750
+    },
+    {
+      "epoch": 0.29261719125998653,
+      "grad_norm": 0.2750128507614136,
+      "learning_rate": 8.302487923351842e-06,
+      "loss": 0.8036,
+      "step": 760
+    },
+    {
+      "epoch": 0.2964674174607758,
+      "grad_norm": 0.26835498213768005,
+      "learning_rate": 8.255412212597021e-06,
+      "loss": 0.8215,
+      "step": 770
+    },
+    {
+      "epoch": 0.3003176436615651,
+      "grad_norm": 0.25934073328971863,
+      "learning_rate": 8.207830160055688e-06,
+      "loss": 0.818,
+      "step": 780
+    },
+    {
+      "epoch": 0.3041678698623544,
+      "grad_norm": 0.28439557552337646,
+      "learning_rate": 8.159749166566786e-06,
+      "loss": 0.8025,
+      "step": 790
+    },
+    {
+      "epoch": 0.30801809606314373,
+      "grad_norm": 0.27402567863464355,
+      "learning_rate": 8.11117671057377e-06,
+      "loss": 0.8303,
+      "step": 800
+    },
+    {
+      "epoch": 0.311868322263933,
+      "grad_norm": 0.27807551622390747,
+      "learning_rate": 8.062120346961404e-06,
+      "loss": 0.8224,
+      "step": 810
+    },
+    {
+      "epoch": 0.3157185484647223,
+      "grad_norm": 0.2809593379497528,
+      "learning_rate": 8.012587705880705e-06,
+      "loss": 0.8154,
+      "step": 820
+    },
+    {
+      "epoch": 0.3195687746655116,
+      "grad_norm": 0.25905266404151917,
+      "learning_rate": 7.962586491562146e-06,
+      "loss": 0.8015,
+      "step": 830
+    },
+    {
+      "epoch": 0.3234190008663009,
+      "grad_norm": 0.28983157873153687,
+      "learning_rate": 7.912124481117357e-06,
+      "loss": 0.8073,
+      "step": 840
+    },
+    {
+      "epoch": 0.3272692270670902,
+      "grad_norm": 0.2889046370983124,
+      "learning_rate": 7.861209523329494e-06,
+      "loss": 0.8176,
+      "step": 850
+    },
+    {
+      "epoch": 0.3311194532678795,
+      "grad_norm": 0.2745714485645294,
+      "learning_rate": 7.809849537432432e-06,
+      "loss": 0.8015,
+      "step": 860
+    },
+    {
+      "epoch": 0.3349696794686688,
+      "grad_norm": 0.26837554574012756,
+      "learning_rate": 7.758052511879035e-06,
+      "loss": 0.8118,
+      "step": 870
+    },
+    {
+      "epoch": 0.3388199056694581,
+      "grad_norm": 0.24795915186405182,
+      "learning_rate": 7.705826503098642e-06,
+      "loss": 0.8065,
+      "step": 880
+    },
+    {
+      "epoch": 0.34267013187024736,
+      "grad_norm": 0.2535915970802307,
+      "learning_rate": 7.653179634243979e-06,
+      "loss": 0.8066,
+      "step": 890
+    },
+    {
+      "epoch": 0.34652035807103665,
+      "grad_norm": 0.24822942912578583,
+      "learning_rate": 7.600120093927696e-06,
+      "loss": 0.8129,
+      "step": 900
+    },
+    {
+      "epoch": 0.350370584271826,
+      "grad_norm": 0.27163317799568176,
+      "learning_rate": 7.546656134948729e-06,
+      "loss": 0.8022,
+      "step": 910
+    },
+    {
+      "epoch": 0.3542208104726153,
+      "grad_norm": 0.2616989314556122,
+      "learning_rate": 7.492796073008661e-06,
+      "loss": 0.7997,
+      "step": 920
+    },
+    {
+      "epoch": 0.35807103667340456,
+      "grad_norm": 0.2880381643772125,
+      "learning_rate": 7.438548285418326e-06,
+      "loss": 0.7863,
+      "step": 930
+    },
+    {
+      "epoch": 0.36192126287419385,
+      "grad_norm": 0.2995399236679077,
+      "learning_rate": 7.3839212097948e-06,
+      "loss": 0.8206,
+      "step": 940
+    },
+    {
+      "epoch": 0.36577148907498314,
+      "grad_norm": 0.26554712653160095,
+      "learning_rate": 7.328923342749039e-06,
+      "loss": 0.8358,
+      "step": 950
+    },
+    {
+      "epoch": 0.3696217152757725,
+      "grad_norm": 0.2895301580429077,
+      "learning_rate": 7.273563238564319e-06,
+      "loss": 0.8193,
+      "step": 960
+    },
+    {
+      "epoch": 0.37347194147656176,
+      "grad_norm": 0.2676248252391815,
+      "learning_rate": 7.217849507865724e-06,
+      "loss": 0.8008,
+      "step": 970
+    },
+    {
+      "epoch": 0.37732216767735105,
+      "grad_norm": 0.2548772692680359,
+      "learning_rate": 7.161790816280858e-06,
+      "loss": 0.8057,
+      "step": 980
+    },
+    {
+      "epoch": 0.38117239387814034,
+      "grad_norm": 0.2842244803905487,
+      "learning_rate": 7.1053958830920125e-06,
+      "loss": 0.7925,
+      "step": 990
+    },
+    {
+      "epoch": 0.3850226200789296,
+      "grad_norm": 0.2657583951950073,
+      "learning_rate": 7.0486734798799715e-06,
+      "loss": 0.8078,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3888728462797189,
+      "grad_norm": 0.268906831741333,
+      "learning_rate": 6.991632429159708e-06,
+      "loss": 0.8314,
+      "step": 1010
+    },
+    {
+      "epoch": 0.39272307248050825,
+      "grad_norm": 0.2800804078578949,
+      "learning_rate": 6.934281603008132e-06,
+      "loss": 0.8228,
+      "step": 1020
+    },
+    {
+      "epoch": 0.39657329868129754,
+      "grad_norm": 0.2691592574119568,
+      "learning_rate": 6.876629921684143e-06,
+      "loss": 0.8199,
+      "step": 1030
+    },
+    {
+      "epoch": 0.4004235248820868,
+      "grad_norm": 0.29676997661590576,
+      "learning_rate": 6.818686352241199e-06,
+      "loss": 0.811,
+      "step": 1040
+    },
+    {
+      "epoch": 0.4042737510828761,
+      "grad_norm": 0.23876076936721802,
+      "learning_rate": 6.760459907132573e-06,
+      "loss": 0.807,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4081239772836654,
+      "grad_norm": 0.25224635004997253,
+      "learning_rate": 6.701959642809589e-06,
+      "loss": 0.8145,
+      "step": 1060
+    },
+    {
+      "epoch": 0.41197420348445474,
+      "grad_norm": 0.27437809109687805,
+      "learning_rate": 6.643194658312987e-06,
+      "loss": 0.8063,
+      "step": 1070
+    },
+    {
+      "epoch": 0.415824429685244,
+      "grad_norm": 0.26933613419532776,
+      "learning_rate": 6.584174093857676e-06,
+      "loss": 0.8294,
+      "step": 1080
+    },
+    {
+      "epoch": 0.4196746558860333,
+      "grad_norm": 0.3020111918449402,
+      "learning_rate": 6.5249071294110735e-06,
+      "loss": 0.8094,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4235248820868226,
+      "grad_norm": 0.29017141461372375,
+      "learning_rate": 6.465402983265274e-06,
+      "loss": 0.8248,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4273751082876119,
+      "grad_norm": 0.2730906307697296,
+      "learning_rate": 6.405670910603243e-06,
+      "loss": 0.8204,
+      "step": 1110
+    },
+    {
+      "epoch": 0.43122533448840117,
+      "grad_norm": 0.3021923899650574,
+      "learning_rate": 6.345720202059291e-06,
+      "loss": 0.8156,
+      "step": 1120
+    },
+    {
+      "epoch": 0.4350755606891905,
+      "grad_norm": 0.27398547530174255,
+      "learning_rate": 6.285560182274008e-06,
+      "loss": 0.7948,
+      "step": 1130
+    },
+    {
+      "epoch": 0.4389257868899798,
+      "grad_norm": 0.2716691493988037,
+      "learning_rate": 6.225200208443936e-06,
+      "loss": 0.8172,
+      "step": 1140
+    },
+    {
+      "epoch": 0.4427760130907691,
+      "grad_norm": 0.24841035902500153,
+      "learning_rate": 6.16464966886616e-06,
+      "loss": 0.8073,
+      "step": 1150
+    },
+    {
+      "epoch": 0.44662623929155837,
+      "grad_norm": 0.3016848862171173,
+      "learning_rate": 6.103917981478065e-06,
+      "loss": 0.8304,
+      "step": 1160
+    },
+    {
+      "epoch": 0.45047646549234766,
+      "grad_norm": 0.2592402398586273,
+      "learning_rate": 6.043014592392489e-06,
+      "loss": 0.8421,
+      "step": 1170
+    },
+    {
+      "epoch": 0.454326691693137,
+      "grad_norm": 0.25379678606987,
+      "learning_rate": 5.981948974428485e-06,
+      "loss": 0.8158,
+      "step": 1180
+    },
+    {
+      "epoch": 0.4581769178939263,
+      "grad_norm": 0.2769143879413605,
+      "learning_rate": 5.920730625637934e-06,
+      "loss": 0.7962,
+      "step": 1190
+    },
+    {
+      "epoch": 0.46202714409471557,
+      "grad_norm": 0.2884799540042877,
+      "learning_rate": 5.8593690678282355e-06,
+      "loss": 0.8049,
+      "step": 1200
+    },
+    {
+      "epoch": 0.46587737029550486,
+      "grad_norm": 0.2782297432422638,
+      "learning_rate": 5.797873845081301e-06,
+      "loss": 0.8101,
+      "step": 1210
+    },
+    {
+      "epoch": 0.46972759649629414,
+      "grad_norm": 0.351400226354599,
+      "learning_rate": 5.736254522269075e-06,
+      "loss": 0.8269,
+      "step": 1220
+    },
+    {
+      "epoch": 0.47357782269708343,
+      "grad_norm": 0.2810094356536865,
+      "learning_rate": 5.6745206835658386e-06,
+      "loss": 0.8274,
+      "step": 1230
+    },
+    {
+      "epoch": 0.47742804889787277,
+      "grad_norm": 0.2528964877128601,
+      "learning_rate": 5.612681930957495e-06,
+      "loss": 0.7987,
+      "step": 1240
+    },
+    {
+      "epoch": 0.48127827509866206,
+      "grad_norm": 0.2532021999359131,
+      "learning_rate": 5.550747882748095e-06,
+      "loss": 0.7982,
+      "step": 1250
+    },
+    {
+      "epoch": 0.48512850129945134,
+      "grad_norm": 0.29498738050460815,
+      "learning_rate": 5.4887281720638195e-06,
+      "loss": 0.8029,
+      "step": 1260
+    },
+    {
+      "epoch": 0.48897872750024063,
+      "grad_norm": 0.3026972711086273,
+      "learning_rate": 5.42663244535466e-06,
+      "loss": 0.7918,
+      "step": 1270
+    },
+    {
+      "epoch": 0.4928289537010299,
+      "grad_norm": 0.29117217659950256,
+      "learning_rate": 5.3644703608940205e-06,
+      "loss": 0.7974,
+      "step": 1280
+    },
+    {
+      "epoch": 0.49667917990181926,
+      "grad_norm": 0.28178730607032776,
+      "learning_rate": 5.302251587276491e-06,
+      "loss": 0.8087,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5005294061026085,
+      "grad_norm": 0.2756960988044739,
+      "learning_rate": 5.2399858019140005e-06,
+      "loss": 0.8082,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5043796323033978,
+      "grad_norm": 0.26126834750175476,
+      "learning_rate": 5.177682689530615e-06,
+      "loss": 0.823,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5082298585041871,
+      "grad_norm": 0.2859497368335724,
+      "learning_rate": 5.11535194065619e-06,
+      "loss": 0.7905,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5120800847049765,
+      "grad_norm": 0.2878323793411255,
+      "learning_rate": 5.053003250119113e-06,
+      "loss": 0.8023,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5159303109057657,
+      "grad_norm": 0.2754213809967041,
+      "learning_rate": 4.9906463155383925e-06,
+      "loss": 0.8084,
+      "step": 1340
+    },
+    {
+      "epoch": 0.519780537106555,
+      "grad_norm": 0.2719596326351166,
+      "learning_rate": 4.9282908358153046e-06,
+      "loss": 0.7993,
+      "step": 1350
+    },
+    {
+      "epoch": 0.5236307633073443,
+      "grad_norm": 0.30703166127204895,
+      "learning_rate": 4.8659465096248345e-06,
+      "loss": 0.8187,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5274809895081336,
+      "grad_norm": 0.2926926612854004,
+      "learning_rate": 4.803623033907169e-06,
+      "loss": 0.8183,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5313312157089229,
+      "grad_norm": 0.2608540952205658,
+      "learning_rate": 4.741330102359441e-06,
+      "loss": 0.8226,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5351814419097122,
+      "grad_norm": 0.2792118489742279,
+      "learning_rate": 4.679077403927996e-06,
+      "loss": 0.8037,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5390316681105015,
+      "grad_norm": 0.2632480561733246,
+      "learning_rate": 4.616874621301373e-06,
+      "loss": 0.8026,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5428818943112907,
+      "grad_norm": 0.28038543462753296,
+      "learning_rate": 4.554731429404293e-06,
+      "loss": 0.817,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5467321205120801,
+      "grad_norm": 0.2886001467704773,
+      "learning_rate": 4.492657493892822e-06,
+      "loss": 0.8032,
+      "step": 1420
+    },
+    {
+      "epoch": 0.5505823467128694,
+      "grad_norm": 0.33963891863822937,
+      "learning_rate": 4.430662469650999e-06,
+      "loss": 0.8177,
+      "step": 1430
+    },
+    {
+      "epoch": 0.5544325729136587,
+      "grad_norm": 0.29751506447792053,
+      "learning_rate": 4.368755999289118e-06,
+      "loss": 0.8189,
+      "step": 1440
+    },
+    {
+      "epoch": 0.558282799114448,
+      "grad_norm": 0.3224038779735565,
+      "learning_rate": 4.306947711643949e-06,
+      "loss": 0.8061,
+      "step": 1450
+    },
+    {
+      "epoch": 0.5621330253152372,
+      "grad_norm": 0.2635518014431,
+      "learning_rate": 4.245247220281068e-06,
+      "loss": 0.8079,
+      "step": 1460
+    },
+    {
+      "epoch": 0.5659832515160266,
+      "grad_norm": 0.2868427038192749,
+      "learning_rate": 4.183664121999587e-06,
+      "loss": 0.8058,
+      "step": 1470
+    },
+    {
+      "epoch": 0.5698334777168159,
+      "grad_norm": 0.351610392332077,
+      "learning_rate": 4.122207995339474e-06,
+      "loss": 0.8015,
+      "step": 1480
+    },
+    {
+      "epoch": 0.5736837039176051,
+      "grad_norm": 0.2827518880367279,
+      "learning_rate": 4.060888399091734e-06,
+      "loss": 0.8081,
+      "step": 1490
+    },
+    {
+      "epoch": 0.5775339301183945,
+      "grad_norm": 0.25765788555145264,
+      "learning_rate": 3.999714870811639e-06,
+      "loss": 0.8145,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5813841563191837,
+      "grad_norm": 0.26858046650886536,
+      "learning_rate": 3.938696925335275e-06,
+      "loss": 0.8081,
+      "step": 1510
+    },
+    {
+      "epoch": 0.5852343825199731,
+      "grad_norm": 0.2742035686969757,
+      "learning_rate": 3.87784405329962e-06,
+      "loss": 0.8168,
+      "step": 1520
+    },
+    {
+      "epoch": 0.5890846087207623,
+      "grad_norm": 0.2613092064857483,
+      "learning_rate": 3.817165719666382e-06,
+      "loss": 0.8192,
+      "step": 1530
+    },
+    {
+      "epoch": 0.5929348349215516,
+      "grad_norm": 0.25879526138305664,
+      "learning_rate": 3.756671362249841e-06,
+      "loss": 0.8222,
+      "step": 1540
+    },
+    {
+      "epoch": 0.596785061122341,
+      "grad_norm": 0.26721668243408203,
+      "learning_rate": 3.6963703902488907e-06,
+      "loss": 0.8092,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6006352873231302,
+      "grad_norm": 0.27115049958229065,
+      "learning_rate": 3.6362721827835655e-06,
+      "loss": 0.828,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6044855135239195,
+      "grad_norm": 0.2922312915325165,
+      "learning_rate": 3.576386087436212e-06,
+      "loss": 0.8212,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6083357397247088,
+      "grad_norm": 0.28088536858558655,
+      "learning_rate": 3.5167214187975913e-06,
+      "loss": 0.811,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6121859659254981,
+      "grad_norm": 0.22707894444465637,
+      "learning_rate": 3.457287457018086e-06,
+      "loss": 0.8137,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6160361921262875,
+      "grad_norm": 0.27203404903411865,
+      "learning_rate": 3.3980934463643003e-06,
+      "loss": 0.8001,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6198864183270767,
+      "grad_norm": 0.24759404361248016,
+      "learning_rate": 3.3391485937812053e-06,
+      "loss": 0.7838,
+      "step": 1610
+    },
+    {
+      "epoch": 0.623736644527866,
+      "grad_norm": 0.26493775844573975,
+      "learning_rate": 3.280462067460115e-06,
+      "loss": 0.7827,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6275868707286553,
+      "grad_norm": 0.2910003364086151,
+      "learning_rate": 3.222042995412669e-06,
+      "loss": 0.813,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6314370969294446,
+      "grad_norm": 0.24596963822841644,
+      "learning_rate": 3.163900464051089e-06,
+      "loss": 0.8028,
+      "step": 1640
+    },
+    {
+      "epoch": 0.635287323130234,
+      "grad_norm": 0.27420517802238464,
+      "learning_rate": 3.1060435167748826e-06,
+      "loss": 0.8114,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6391375493310232,
+      "grad_norm": 0.2714826762676239,
+      "learning_rate": 3.048481152564253e-06,
+      "loss": 0.7921,
+      "step": 1660
+    },
+    {
+      "epoch": 0.6429877755318125,
+      "grad_norm": 0.28418129682540894,
+      "learning_rate": 2.9912223245804017e-06,
+      "loss": 0.814,
+      "step": 1670
+    },
+    {
+      "epoch": 0.6468380017326018,
+      "grad_norm": 0.2776290774345398,
+      "learning_rate": 2.934275938772981e-06,
+      "loss": 0.8118,
+      "step": 1680
+    },
+    {
+      "epoch": 0.6506882279333911,
+      "grad_norm": 0.29460471868515015,
+      "learning_rate": 2.8776508524948676e-06,
+      "loss": 0.8011,
+      "step": 1690
+    },
+    {
+      "epoch": 0.6545384541341804,
+      "grad_norm": 0.25294366478919983,
+      "learning_rate": 2.821355873124503e-06,
+      "loss": 0.7999,
+      "step": 1700
+    },
+    {
+      "epoch": 0.6583886803349697,
+      "grad_norm": 0.27538567781448364,
+      "learning_rate": 2.7653997566960144e-06,
+      "loss": 0.8136,
+      "step": 1710
+    },
+    {
+      "epoch": 0.662238906535759,
+      "grad_norm": 0.2818944454193115,
+      "learning_rate": 2.709791206537309e-06,
+      "loss": 0.8032,
+      "step": 1720
+    },
+    {
+      "epoch": 0.6660891327365482,
+      "grad_norm": 0.2821090817451477,
+      "learning_rate": 2.6545388719163747e-06,
+      "loss": 0.8113,
+      "step": 1730
+    },
+    {
+      "epoch": 0.6699393589373376,
+      "grad_norm": 0.2641777992248535,
+      "learning_rate": 2.599651346695979e-06,
+      "loss": 0.8085,
+      "step": 1740
+    },
+    {
+      "epoch": 0.6737895851381268,
+      "grad_norm": 0.2814924120903015,
+      "learning_rate": 2.545137167997007e-06,
+      "loss": 0.8038,
+      "step": 1750
+    },
+    {
+      "epoch": 0.6776398113389162,
+      "grad_norm": 0.2578508257865906,
+      "learning_rate": 2.4910048148705972e-06,
+      "loss": 0.7925,
+      "step": 1760
+    },
+    {
+      "epoch": 0.6814900375397055,
+      "grad_norm": 0.28423959016799927,
+      "learning_rate": 2.4372627069793327e-06,
+      "loss": 0.8352,
+      "step": 1770
+    },
+    {
+      "epoch": 0.6853402637404947,
+      "grad_norm": 0.26789841055870056,
+      "learning_rate": 2.3839192032876486e-06,
+      "loss": 0.822,
+      "step": 1780
+    },
+    {
+      "epoch": 0.6891904899412841,
+      "grad_norm": 0.2688464820384979,
+      "learning_rate": 2.3309826007617047e-06,
+      "loss": 0.8325,
+      "step": 1790
+    },
+    {
+      "epoch": 0.6930407161420733,
+      "grad_norm": 0.2706427574157715,
+      "learning_rate": 2.278461133078881e-06,
+      "loss": 0.7962,
+      "step": 1800
+    },
+    {
+      "epoch": 0.6968909423428626,
+      "grad_norm": 0.279430091381073,
+      "learning_rate": 2.2263629693471258e-06,
+      "loss": 0.7868,
+      "step": 1810
+    },
+    {
+      "epoch": 0.700741168543652,
+      "grad_norm": 0.26170650124549866,
+      "learning_rate": 2.1746962128343414e-06,
+      "loss": 0.8126,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7045913947444412,
+      "grad_norm": 0.25984835624694824,
+      "learning_rate": 2.1234688997080183e-06,
+      "loss": 0.8,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7084416209452306,
+      "grad_norm": 0.2763589024543762,
+      "learning_rate": 2.0726889977853053e-06,
+      "loss": 0.8044,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7122918471460198,
+      "grad_norm": 0.27905601263046265,
+      "learning_rate": 2.022364405293703e-06,
+      "loss": 0.8173,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7161420733468091,
+      "grad_norm": 0.25517967343330383,
+      "learning_rate": 1.9725029496425814e-06,
+      "loss": 0.8095,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7199922995475985,
+      "grad_norm": 0.2711159288883209,
+      "learning_rate": 1.923112386205727e-06,
+      "loss": 0.8075,
+      "step": 1870
+    },
+    {
+      "epoch": 0.7238425257483877,
+      "grad_norm": 0.2559998631477356,
+      "learning_rate": 1.874200397115079e-06,
+      "loss": 0.813,
+      "step": 1880
+    },
+    {
+      "epoch": 0.727692751949177,
+      "grad_norm": 0.27188214659690857,
+      "learning_rate": 1.825774590065857e-06,
+      "loss": 0.7985,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7315429781499663,
+      "grad_norm": 0.28382328152656555,
+      "learning_rate": 1.7778424971332813e-06,
+      "loss": 0.804,
+      "step": 1900
+    },
+    {
+      "epoch": 0.7353932043507556,
+      "grad_norm": 0.2640560269355774,
+      "learning_rate": 1.7304115736010407e-06,
+      "loss": 0.8041,
+      "step": 1910
+    },
+    {
+      "epoch": 0.739243430551545,
+      "grad_norm": 0.273459255695343,
+      "learning_rate": 1.6834891968017108e-06,
+      "loss": 0.7962,
+      "step": 1920
+    },
+    {
+      "epoch": 0.7430936567523342,
+      "grad_norm": 0.2846871018409729,
+      "learning_rate": 1.6370826649692884e-06,
+      "loss": 0.8172,
+      "step": 1930
+    },
+    {
+      "epoch": 0.7469438829531235,
+      "grad_norm": 0.26902511715888977,
+      "learning_rate": 1.5911991961040463e-06,
+      "loss": 0.8236,
+      "step": 1940
+    },
+    {
+      "epoch": 0.7507941091539128,
+      "grad_norm": 0.2721902132034302,
+      "learning_rate": 1.545845926849851e-06,
+      "loss": 0.8086,
+      "step": 1950
+    },
+    {
+      "epoch": 0.7546443353547021,
+      "grad_norm": 0.2908647060394287,
+      "learning_rate": 1.5010299113841397e-06,
+      "loss": 0.8276,
+      "step": 1960
+    },
+    {
+      "epoch": 0.7584945615554913,
+      "grad_norm": 0.27422651648521423,
+      "learning_rate": 1.4567581203207238e-06,
+      "loss": 0.8154,
+      "step": 1970
+    },
+    {
+      "epoch": 0.7623447877562807,
+      "grad_norm": 0.2828887701034546,
+      "learning_rate": 1.413037439625593e-06,
+      "loss": 0.8148,
+      "step": 1980
+    },
+    {
+      "epoch": 0.76619501395707,
+      "grad_norm": 0.29020577669143677,
+      "learning_rate": 1.3698746695458847e-06,
+      "loss": 0.8082,
+      "step": 1990
+    },
+    {
+      "epoch": 0.7700452401578592,
+      "grad_norm": 0.26012346148490906,
+      "learning_rate": 1.3272765235521745e-06,
+      "loss": 0.7997,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7738954663586486,
+      "grad_norm": 0.2626100182533264,
+      "learning_rate": 1.2852496272942877e-06,
+      "loss": 0.8075,
+      "step": 2010
+    },
+    {
+      "epoch": 0.7777456925594378,
+      "grad_norm": 0.2562048137187958,
+      "learning_rate": 1.2438005175707396e-06,
+      "loss": 0.7961,
+      "step": 2020
+    },
+    {
+      "epoch": 0.7815959187602272,
+      "grad_norm": 0.26004430651664734,
+      "learning_rate": 1.2029356413120297e-06,
+      "loss": 0.7827,
+      "step": 2030
+    },
+    {
+      "epoch": 0.7854461449610165,
+      "grad_norm": 0.26805219054222107,
+      "learning_rate": 1.1626613545778786e-06,
+      "loss": 0.7995,
+      "step": 2040
+    },
+    {
+      "epoch": 0.7892963711618057,
+      "grad_norm": 0.25083449482917786,
+      "learning_rate": 1.1229839215686294e-06,
+      "loss": 0.8028,
+      "step": 2050
+    },
+    {
+      "epoch": 0.7931465973625951,
+      "grad_norm": 0.2506543695926666,
+      "learning_rate": 1.0839095136509188e-06,
+      "loss": 0.7905,
+      "step": 2060
+    },
+    {
+      "epoch": 0.7969968235633843,
+      "grad_norm": 0.2929697334766388,
+      "learning_rate": 1.045444208397791e-06,
+      "loss": 0.8126,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8008470497641736,
+      "grad_norm": 0.2615942358970642,
+      "learning_rate": 1.007593988643402e-06,
+      "loss": 0.7904,
+      "step": 2080
+    },
+    {
+      "epoch": 0.804697275964963,
+      "grad_norm": 0.2588602304458618,
+      "learning_rate": 9.703647415524613e-07,
+      "loss": 0.8132,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8085475021657522,
+      "grad_norm": 0.2689819931983948,
+      "learning_rate": 9.337622577045485e-07,
+      "loss": 0.8018,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8123977283665416,
+      "grad_norm": 0.27088138461112976,
+      "learning_rate": 8.977922301934588e-07,
+      "loss": 0.8092,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8162479545673308,
+      "grad_norm": 0.2661363482475281,
+      "learning_rate": 8.624602537417009e-07,
+      "loss": 0.8115,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8200981807681201,
+      "grad_norm": 0.2946307063102722,
+      "learning_rate": 8.277718238303123e-07,
+      "loss": 0.8013,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8239484069689095,
+      "grad_norm": 0.2682909071445465,
+      "learning_rate": 7.937323358440935e-07,
+      "loss": 0.8035,
+      "step": 2140
+    },
+    {
+      "epoch": 0.8277986331696987,
+      "grad_norm": 0.2739409804344177,
+      "learning_rate": 7.603470842324223e-07,
+      "loss": 0.7967,
+      "step": 2150
+    },
+    {
+      "epoch": 0.831648859370488,
+      "grad_norm": 0.2584935426712036,
+      "learning_rate": 7.276212616857575e-07,
+      "loss": 0.8127,
+      "step": 2160
+    },
+    {
+      "epoch": 0.8354990855712773,
+      "grad_norm": 0.2641301453113556,
+      "learning_rate": 6.955599583279837e-07,
+      "loss": 0.8146,
+      "step": 2170
+    },
+    {
+      "epoch": 0.8393493117720666,
+      "grad_norm": 0.2689859867095947,
+      "learning_rate": 6.641681609246981e-07,
+      "loss": 0.804,
+      "step": 2180
+    },
+    {
+      "epoch": 0.8431995379728559,
+      "grad_norm": 0.26793554425239563,
+      "learning_rate": 6.334507521075711e-07,
+      "loss": 0.8135,
+      "step": 2190
+    },
+    {
+      "epoch": 0.8470497641736452,
+      "grad_norm": 0.285454660654068,
+      "learning_rate": 6.034125096149179e-07,
+      "loss": 0.8044,
+      "step": 2200
+    },
+    {
+      "epoch": 0.8508999903744345,
+      "grad_norm": 0.26571914553642273,
+      "learning_rate": 5.740581055485722e-07,
+      "loss": 0.7778,
+      "step": 2210
+    },
+    {
+      "epoch": 0.8547502165752238,
+      "grad_norm": 0.25310105085372925,
+      "learning_rate": 5.453921056471961e-07,
+      "loss": 0.8126,
+      "step": 2220
+    },
+    {
+      "epoch": 0.8586004427760131,
+      "grad_norm": 0.28159284591674805,
+      "learning_rate": 5.174189685761288e-07,
+      "loss": 0.8043,
+      "step": 2230
+    },
+    {
+      "epoch": 0.8624506689768023,
+      "grad_norm": 0.2703046202659607,
+      "learning_rate": 4.901430452338979e-07,
+      "loss": 0.816,
+      "step": 2240
+    },
+    {
+      "epoch": 0.8663008951775917,
+      "grad_norm": 0.26512518525123596,
+      "learning_rate": 4.6356857807548286e-07,
+      "loss": 0.801,
+      "step": 2250
+    },
+    {
+      "epoch": 0.870151121378381,
+      "grad_norm": 0.26257163286209106,
+      "learning_rate": 4.376997004524541e-07,
+      "loss": 0.7906,
+      "step": 2260
+    },
+    {
+      "epoch": 0.8740013475791703,
+      "grad_norm": 0.28759631514549255,
+      "learning_rate": 4.12540435970073e-07,
+      "loss": 0.802,
+      "step": 2270
+    },
+    {
+      "epoch": 0.8778515737799596,
+      "grad_norm": 0.2538432478904724,
+      "learning_rate": 3.880946978614703e-07,
+      "loss": 0.809,
+      "step": 2280
+    },
+    {
+      "epoch": 0.8817017999807488,
+      "grad_norm": 0.265565425157547,
+      "learning_rate": 3.643662883789878e-07,
+      "loss": 0.8162,
+      "step": 2290
+    },
+    {
+      "epoch": 0.8855520261815382,
+      "grad_norm": 0.247894749045372,
+      "learning_rate": 3.41358898202781e-07,
+      "loss": 0.7913,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8894022523823275,
+      "grad_norm": 0.2515472173690796,
+      "learning_rate": 3.190761058667741e-07,
+      "loss": 0.8029,
+      "step": 2310
+    },
+    {
+      "epoch": 0.8932524785831167,
+      "grad_norm": 0.2541233003139496,
+      "learning_rate": 2.9752137720206463e-07,
+      "loss": 0.8105,
+      "step": 2320
+    },
+    {
+      "epoch": 0.8971027047839061,
+      "grad_norm": 0.2647615075111389,
+      "learning_rate": 2.7669806479785067e-07,
+      "loss": 0.814,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9009529309846953,
+      "grad_norm": 0.27337709069252014,
+      "learning_rate": 2.566094074799763e-07,
+      "loss": 0.8098,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9048031571854847,
+      "grad_norm": 0.2681642174720764,
+      "learning_rate": 2.3725852980716557e-07,
+      "loss": 0.7885,
+      "step": 2350
+    },
+    {
+      "epoch": 0.908653383386274,
+      "grad_norm": 0.26031583547592163,
+      "learning_rate": 2.1864844158503983e-07,
+      "loss": 0.8029,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9125036095870632,
+      "grad_norm": 0.28443485498428345,
+      "learning_rate": 2.007820373979702e-07,
+      "loss": 0.794,
+      "step": 2370
+    },
+    {
+      "epoch": 0.9163538357878526,
+      "grad_norm": 0.27149325609207153,
+      "learning_rate": 1.8366209615886532e-07,
+      "loss": 0.8027,
+      "step": 2380
+    },
+    {
+      "epoch": 0.9202040619886418,
+      "grad_norm": 0.2625353932380676,
+      "learning_rate": 1.6729128067693613e-07,
+      "loss": 0.8121,
+      "step": 2390
+    },
+    {
+      "epoch": 0.9240542881894311,
+      "grad_norm": 0.2859114408493042,
+      "learning_rate": 1.5167213724353426e-07,
+      "loss": 0.814,
+      "step": 2400
+    },
+    {
+      "epoch": 0.9279045143902205,
+      "grad_norm": 0.25346648693084717,
+      "learning_rate": 1.3680709523610058e-07,
+      "loss": 0.7965,
+      "step": 2410
+    },
+    {
+      "epoch": 0.9317547405910097,
+      "grad_norm": 0.2590552866458893,
+      "learning_rate": 1.2269846674030418e-07,
+      "loss": 0.8194,
+      "step": 2420
+    },
+    {
+      "epoch": 0.935604966791799,
+      "grad_norm": 0.2644140422344208,
+      "learning_rate": 1.0934844619042595e-07,
+      "loss": 0.8155,
+      "step": 2430
+    },
+    {
+      "epoch": 0.9394551929925883,
+      "grad_norm": 0.28044939041137695,
+      "learning_rate": 9.675911002803961e-08,
+      "loss": 0.8055,
+      "step": 2440
+    },
+    {
+      "epoch": 0.9433054191933776,
+      "grad_norm": 0.2930426597595215,
+      "learning_rate": 8.493241637904381e-08,
+      "loss": 0.802,
+      "step": 2450
+    },
+    {
+      "epoch": 0.9471556453941669,
+      "grad_norm": 0.28092557191848755,
+      "learning_rate": 7.387020474909689e-08,
+      "loss": 0.8221,
+      "step": 2460
+    },
+    {
+      "epoch": 0.9510058715949562,
+      "grad_norm": 0.2855808734893799,
+      "learning_rate": 6.357419573750856e-08,
+      "loss": 0.8169,
+      "step": 2470
+    },
+    {
+      "epoch": 0.9548560977957455,
+      "grad_norm": 0.2787171006202698,
+      "learning_rate": 5.4045990769613365e-08,
+      "loss": 0.7996,
+      "step": 2480
+    },
+    {
+      "epoch": 0.9587063239965348,
+      "grad_norm": 0.3188609778881073,
+      "learning_rate": 4.5287071847694434e-08,
+      "loss": 0.8054,
+      "step": 2490
+    },
+    {
+      "epoch": 0.9625565501973241,
+      "grad_norm": 0.2689768671989441,
+      "learning_rate": 3.7298801320467817e-08,
+      "loss": 0.7981,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9664067763981133,
+      "grad_norm": 0.2636423707008362,
+      "learning_rate": 3.0082421671192576e-08,
+      "loss": 0.7912,
+      "step": 2510
+    },
+    {
+      "epoch": 0.9702570025989027,
+      "grad_norm": 0.25990742444992065,
+      "learning_rate": 2.3639055324410355e-08,
+      "loss": 0.8132,
+      "step": 2520
+    },
+    {
+      "epoch": 0.974107228799692,
+      "grad_norm": 0.25681984424591064,
+      "learning_rate": 1.7969704471369477e-08,
+      "loss": 0.7974,
+      "step": 2530
+    },
+    {
+      "epoch": 0.9779574550004813,
+      "grad_norm": 0.2554507255554199,
+      "learning_rate": 1.3075250914142413e-08,
+      "loss": 0.801,
+      "step": 2540
+    },
+    {
+      "epoch": 0.9818076812012706,
+      "grad_norm": 0.26214444637298584,
+      "learning_rate": 8.956455928476049e-09,
+      "loss": 0.8117,
+      "step": 2550
+    },
+    {
+      "epoch": 0.9856579074020598,
+      "grad_norm": 0.2809244990348816,
+      "learning_rate": 5.6139601453786275e-09,
+      "loss": 0.803,
+      "step": 2560
+    },
+    {
+      "epoch": 0.9895081336028492,
+      "grad_norm": 0.25652220845222473,
+      "learning_rate": 3.0482834514805602e-09,
+      "loss": 0.7912,
+      "step": 2570
+    },
+    {
+      "epoch": 0.9933583598036385,
+      "grad_norm": 0.24525706470012665,
+      "learning_rate": 1.2598249081713366e-09,
+      "loss": 0.7871,
+      "step": 2580
+    },
+    {
+      "epoch": 0.9972085860044277,
+      "grad_norm": 0.2752741277217865,
+      "learning_rate": 2.488626895302826e-10,
+      "loss": 0.8157,
+      "step": 2590
+    },
+    {
+      "epoch": 0.9999037443449803,
+      "step": 2597,
+      "total_flos": 3.7718367080818934e+18,
+      "train_loss": 0.8325622657376341,
+      "train_runtime": 12946.7876,
+      "train_samples_per_second": 51.357,
+      "train_steps_per_second": 0.201
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2597,
+  "num_input_tokens_seen": 418855352,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.7718367080818934e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

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

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:8429007de700c5a2deb0bafcec55a10a63ccc216cca1d39d06906cb402211e06
+size 6264

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

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

continuous/qwen3/eupe_vit_b/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": 768,
+  "mm_newline_position": "grid",
+  "mm_patch_merge_type": "flat",
+  "mm_projector_lr": null,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_resampler_type": null,
+  "mm_spatial_pool_mode": "bilinear",
+  "mm_spatial_pool_stride": null,
+  "mm_tunable_parts": "mm_mlp_adapter",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -1,
+  "mm_vision_tower": "vtb_ssl:eupe",
+  "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": "/cache/ckpt/download/tokenizer/continuous/eupe/EUPE-ViT-B/EUPE-ViT-B.pt",
+  "vocab_size": 151936
+}

continuous/qwen3/eupe_vit_b/pretrain/mm_projector.bin ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:09618fafcdaa116623f0486f581f014d2d1e4ebefc193a39a686ba66c5e6b203
+size 23087160

continuous/qwen3/eupe_vit_b/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": 3.005220890045166,
+      "learning_rate": 0.00032142857142857147,
+      "loss": 10.6418,
+      "step": 10
+    },
+    {
+      "epoch": 0.03669724770642202,
+      "grad_norm": 1.4143332242965698,
+      "learning_rate": 0.0006785714285714287,
+      "loss": 6.2116,
+      "step": 20
+    },
+    {
+      "epoch": 0.05504587155963303,
+      "grad_norm": 0.13479675352573395,
+      "learning_rate": 0.0009999907688176173,
+      "loss": 5.0805,
+      "step": 30
+    },
+    {
+      "epoch": 0.07339449541284404,
+      "grad_norm": 0.07230617851018906,
+      "learning_rate": 0.0009988834393115766,
+      "loss": 4.8318,
+      "step": 40
+    },
+    {
+      "epoch": 0.09174311926605505,
+      "grad_norm": 0.05896417796611786,
+      "learning_rate": 0.0009959345572940772,
+      "loss": 4.7232,
+      "step": 50
+    },
+    {
+      "epoch": 0.11009174311926606,
+      "grad_norm": 0.06624633073806763,
+      "learning_rate": 0.0009911550081155983,
+      "loss": 4.6715,
+      "step": 60
+    },
+    {
+      "epoch": 0.12844036697247707,
+      "grad_norm": 0.076803058385849,
+      "learning_rate": 0.0009845624347566433,
+      "loss": 4.6233,
+      "step": 70
+    },
+    {
+      "epoch": 0.14678899082568808,
+      "grad_norm": 0.05275696516036987,
+      "learning_rate": 0.0009761811727013547,
+      "loss": 4.5923,
+      "step": 80
+    },
+    {
+      "epoch": 0.1651376146788991,
+      "grad_norm": 0.06038420647382736,
+      "learning_rate": 0.0009660421601067666,
+      "loss": 4.5567,
+      "step": 90
+    },
+    {
+      "epoch": 0.1834862385321101,
+      "grad_norm": 0.06986052542924881,
+      "learning_rate": 0.0009541828235992926,
+      "loss": 4.5344,
+      "step": 100
+    },
+    {
+      "epoch": 0.2018348623853211,
+      "grad_norm": 0.08537930995225906,
+      "learning_rate": 0.0009406469401200151,
+      "loss": 4.5042,
+      "step": 110
+    },
+    {
+      "epoch": 0.22018348623853212,
+      "grad_norm": 0.10681165754795074,
+      "learning_rate": 0.0009254844753287492,
+      "loss": 4.4927,
+      "step": 120
+    },
+    {
+      "epoch": 0.23853211009174313,
+      "grad_norm": 0.08336906135082245,
+      "learning_rate": 0.0009087513991633924,
+      "loss": 4.4626,
+      "step": 130
+    },
+    {
+      "epoch": 0.25688073394495414,
+      "grad_norm": 0.08041588217020035,
+      "learning_rate": 0.0008905094792353917,
+      "loss": 4.4419,
+      "step": 140
+    },
+    {
+      "epoch": 0.27522935779816515,
+      "grad_norm": 0.08322007954120636,
+      "learning_rate": 0.0008708260528239789,
+      "loss": 4.435,
+      "step": 150
+    },
+    {
+      "epoch": 0.29357798165137616,
+      "grad_norm": 0.08410292118787766,
+      "learning_rate": 0.0008497737783108237,
+      "loss": 4.4165,
+      "step": 160
+    },
+    {
+      "epoch": 0.3119266055045872,
+      "grad_norm": 0.09117364883422852,
+      "learning_rate": 0.0008274303669726426,
+      "loss": 4.4037,
+      "step": 170
+    },
+    {
+      "epoch": 0.3302752293577982,
+      "grad_norm": 0.10057204961776733,
+      "learning_rate": 0.0008038782961218137,
+      "loss": 4.3866,
+      "step": 180
+    },
+    {
+      "epoch": 0.3486238532110092,
+      "grad_norm": 0.11618110537528992,
+      "learning_rate": 0.000779204504653894,
+      "loss": 4.3824,
+      "step": 190
+    },
+    {
+      "epoch": 0.3669724770642202,
+      "grad_norm": 0.11787672340869904,
+      "learning_rate": 0.0007535000721258791,
+      "loss": 4.3631,
+      "step": 200
+    },
+    {
+      "epoch": 0.3853211009174312,
+      "grad_norm": 0.1052696630358696,
+      "learning_rate": 0.00072685988254984,
+      "loss": 4.3539,
+      "step": 210
+    },
+    {
+      "epoch": 0.4036697247706422,
+      "grad_norm": 0.09309299290180206,
+      "learning_rate": 0.0006993822741429906,
+      "loss": 4.3623,
+      "step": 220
+    },
+    {
+      "epoch": 0.42201834862385323,
+      "grad_norm": 0.09513290971517563,
+      "learning_rate": 0.0006711686763270818,
+      "loss": 4.3583,
+      "step": 230
+    },
+    {
+      "epoch": 0.44036697247706424,
+      "grad_norm": 0.17050080001354218,
+      "learning_rate": 0.0006423232353170831,
+      "loss": 4.3448,
+      "step": 240
+    },
+    {
+      "epoch": 0.45871559633027525,
+      "grad_norm": 0.0900091826915741,
+      "learning_rate": 0.0006129524296812335,
+      "loss": 4.3262,
+      "step": 250
+    },
+    {
+      "epoch": 0.47706422018348627,
+      "grad_norm": 0.09480103850364685,
+      "learning_rate": 0.0005831646772915651,
+      "loss": 4.3379,
+      "step": 260
+    },
+    {
+      "epoch": 0.4954128440366973,
+      "grad_norm": 0.09051118791103363,
+      "learning_rate": 0.0005530699351157809,
+      "loss": 4.3165,
+      "step": 270
+    },
+    {
+      "epoch": 0.5137614678899083,
+      "grad_norm": 0.09478103369474411,
+      "learning_rate": 0.0005227792933277943,
+      "loss": 4.3127,
+      "step": 280
+    },
+    {
+      "epoch": 0.5321100917431193,
+      "grad_norm": 0.1330287903547287,
+      "learning_rate": 0.0004924045652352092,
+      "loss": 4.2975,
+      "step": 290
+    },
+    {
+      "epoch": 0.5504587155963303,
+      "grad_norm": 0.0843457281589508,
+      "learning_rate": 0.0004620578745374633,
+      "loss": 4.2958,
+      "step": 300
+    },
+    {
+      "epoch": 0.5688073394495413,
+      "grad_norm": 0.10191740840673447,
+      "learning_rate": 0.0004318512414382059,
+      "loss": 4.3107,
+      "step": 310
+    },
+    {
+      "epoch": 0.5871559633027523,
+      "grad_norm": 0.08187822252511978,
+      "learning_rate": 0.0004018961691397148,
+      "loss": 4.303,
+      "step": 320
+    },
+    {
+      "epoch": 0.6055045871559633,
+      "grad_norm": 0.10961198061704636,
+      "learning_rate": 0.0003723032322457458,
+      "loss": 4.2839,
+      "step": 330
+    },
+    {
+      "epoch": 0.6238532110091743,
+      "grad_norm": 0.07864456623792648,
+      "learning_rate": 0.0003431816685921625,
+      "loss": 4.2852,
+      "step": 340
+    },
+    {
+      "epoch": 0.6422018348623854,
+      "grad_norm": 0.08187984675168991,
+      "learning_rate": 0.0003146389760120416,
+      "loss": 4.2893,
+      "step": 350
+    },
+    {
+      "epoch": 0.6605504587155964,
+      "grad_norm": 0.09183705598115921,
+      "learning_rate": 0.00028678051552373485,
+      "loss": 4.2926,
+      "step": 360
+    },
+    {
+      "epoch": 0.6788990825688074,
+      "grad_norm": 0.07713642716407776,
+      "learning_rate": 0.0002597091224066581,
+      "loss": 4.2764,
+      "step": 370
+    },
+    {
+      "epoch": 0.6972477064220184,
+      "grad_norm": 0.0668778344988823,
+      "learning_rate": 0.0002335247266004629,
+      "loss": 4.2932,
+      "step": 380
+    },
+    {
+      "epoch": 0.7155963302752294,
+      "grad_norm": 0.1051175519824028,
+      "learning_rate": 0.000208323983828828,
+      "loss": 4.2735,
+      "step": 390
+    },
+    {
+      "epoch": 0.7339449541284404,
+      "grad_norm": 0.07471863925457001,
+      "learning_rate": 0.0001841999188095224,
+      "loss": 4.2776,
+      "step": 400
+    },
+    {
+      "epoch": 0.7522935779816514,
+      "grad_norm": 0.0786602720618248,
+      "learning_rate": 0.00016124158186777676,
+      "loss": 4.2767,
+      "step": 410
+    },
+    {
+      "epoch": 0.7706422018348624,
+      "grad_norm": 0.08525223284959793,
+      "learning_rate": 0.0001395337202205211,
+      "loss": 4.2685,
+      "step": 420
+    },
+    {
+      "epoch": 0.7889908256880734,
+      "grad_norm": 0.0938214510679245,
+      "learning_rate": 0.00011915646514489292,
+      "loss": 4.2711,
+      "step": 430
+    },
+    {
+      "epoch": 0.8073394495412844,
+      "grad_norm": 0.07631278783082962,
+      "learning_rate": 0.00010018503618578818,
+      "loss": 4.2784,
+      "step": 440
+    },
+    {
+      "epoch": 0.8256880733944955,
+      "grad_norm": 0.07443192601203918,
+      "learning_rate": 8.268946349432583e-05,
+      "loss": 4.2698,
+      "step": 450
+    },
+    {
+      "epoch": 0.8440366972477065,
+      "grad_norm": 0.07218990474939346,
+      "learning_rate": 6.673432932217117e-05,
+      "loss": 4.2622,
+      "step": 460
+    },
+    {
+      "epoch": 0.8623853211009175,
+      "grad_norm": 0.06659010797739029,
+      "learning_rate": 5.2378529625954684e-05,
+      "loss": 4.2516,
+      "step": 470
+    },
+    {
+      "epoch": 0.8807339449541285,
+      "grad_norm": 0.06201259791851044,
+      "learning_rate": 3.9675056661785556e-05,
+      "loss": 4.2635,
+      "step": 480
+    },
+    {
+      "epoch": 0.8990825688073395,
+      "grad_norm": 0.07481935620307922,
+      "learning_rate": 2.8670803372379538e-05,
+      "loss": 4.2613,
+      "step": 490
+    },
+    {
+      "epoch": 0.9174311926605505,
+      "grad_norm": 0.06640496104955673,
+      "learning_rate": 1.9406390288876586e-05,
+      "loss": 4.2738,
+      "step": 500
+    },
+    {
+      "epoch": 0.9357798165137615,
+      "grad_norm": 0.06687294691801071,
+      "learning_rate": 1.191601558631461e-05,
+      "loss": 4.2646,
+      "step": 510
+    },
+    {
+      "epoch": 0.9541284403669725,
+      "grad_norm": 0.062066201120615005,
+      "learning_rate": 6.2273288462559305e-06,
+      "loss": 4.2555,
+      "step": 520
+    },
+    {
+      "epoch": 0.9724770642201835,
+      "grad_norm": 0.06396763771772385,
+      "learning_rate": 2.361328992552314e-06,
+      "loss": 4.2563,
+      "step": 530
+    },
+    {
+      "epoch": 0.9908256880733946,
+      "grad_norm": 0.0629340335726738,
+      "learning_rate": 3.322867770044202e-07,
+      "loss": 4.2624,
+      "step": 540
+    },
+    {
+      "epoch": 1.0,
+      "step": 545,
+      "total_flos": 2.8723174686326784e+17,
+      "train_loss": 4.528507586137964,
+      "train_runtime": 5086.7941,
+      "train_samples_per_second": 109.721,
+      "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": 2.8723174686326784e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen3/eupe_vit_s/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/eupe_vit_s/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/eupe_vit_s/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": 384,
+  "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": "vtb_ssl:eupe",
+  "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": "/cache/ckpt/download/tokenizer/continuous/eupe/EUPE-ViT-S/EUPE-ViT-S.pt",
+  "vocab_size": 151936
+}

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

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

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

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:1c770931f19a13598e6e4369fb0f6e922f899ad8880f3f0e3f2d743276d93cfa
+size 4126709392

continuous/qwen3/eupe_vit_s/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/eupe_vit_s/finetune/tokenizer.json ADDED Viewed

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

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

continuous/qwen3/eupe_vit_s/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/eupe_vit_s/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": 37.457523345947266,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 3.8345,
+      "step": 10
+    },
+    {
+      "epoch": 0.007700452401578592,
+      "grad_norm": 19.021774291992188,
+      "learning_rate": 2.435897435897436e-06,
+      "loss": 3.0515,
+      "step": 20
+    },
+    {
+      "epoch": 0.01155067860236789,
+      "grad_norm": 6.878671169281006,
+      "learning_rate": 3.7179487179487184e-06,
+      "loss": 1.9758,
+      "step": 30
+    },
+    {
+      "epoch": 0.015400904803157185,
+      "grad_norm": 3.7488157749176025,
+      "learning_rate": 5e-06,
+      "loss": 1.4157,
+      "step": 40
+    },
+    {
+      "epoch": 0.019251131003946483,
+      "grad_norm": 1.0679337978363037,
+      "learning_rate": 6.282051282051282e-06,
+      "loss": 1.1382,
+      "step": 50
+    },
+    {
+      "epoch": 0.02310135720473578,
+      "grad_norm": 0.512097954750061,
+      "learning_rate": 7.564102564102564e-06,
+      "loss": 1.0073,
+      "step": 60
+    },
+    {
+      "epoch": 0.026951583405525073,
+      "grad_norm": 0.38366472721099854,
+      "learning_rate": 8.846153846153847e-06,
+      "loss": 0.9479,
+      "step": 70
+    },
+    {
+      "epoch": 0.03080180960631437,
+      "grad_norm": 0.35376372933387756,
+      "learning_rate": 9.999996111488723e-06,
+      "loss": 0.9433,
+      "step": 80
+    },
+    {
+      "epoch": 0.034652035807103666,
+      "grad_norm": 0.32488545775413513,
+      "learning_rate": 9.999529497453782e-06,
+      "loss": 0.9077,
+      "step": 90
+    },
+    {
+      "epoch": 0.038502262007892966,
+      "grad_norm": 0.3343740701675415,
+      "learning_rate": 9.998285264324193e-06,
+      "loss": 0.887,
+      "step": 100
+    },
+    {
+      "epoch": 0.04235248820868226,
+      "grad_norm": 0.3116309940814972,
+      "learning_rate": 9.996263605626052e-06,
+      "loss": 0.8681,
+      "step": 110
+    },
+    {
+      "epoch": 0.04620271440947156,
+      "grad_norm": 0.324095755815506,
+      "learning_rate": 9.99346483580503e-06,
+      "loss": 0.8773,
+      "step": 120
+    },
+    {
+      "epoch": 0.05005294061026085,
+      "grad_norm": 0.2982901334762573,
+      "learning_rate": 9.989889390177467e-06,
+      "loss": 0.8742,
+      "step": 130
+    },
+    {
+      "epoch": 0.053903166811050146,
+      "grad_norm": 0.320011168718338,
+      "learning_rate": 9.985537824862647e-06,
+      "loss": 0.8676,
+      "step": 140
+    },
+    {
+      "epoch": 0.057753393011839446,
+      "grad_norm": 0.35751238465309143,
+      "learning_rate": 9.980410816696317e-06,
+      "loss": 0.8533,
+      "step": 150
+    },
+    {
+      "epoch": 0.06160361921262874,
+      "grad_norm": 0.33115360140800476,
+      "learning_rate": 9.97450916312541e-06,
+      "loss": 0.8577,
+      "step": 160
+    },
+    {
+      "epoch": 0.06545384541341803,
+      "grad_norm": 0.26115962862968445,
+      "learning_rate": 9.967833782084004e-06,
+      "loss": 0.8585,
+      "step": 170
+    },
+    {
+      "epoch": 0.06930407161420733,
+      "grad_norm": 0.31133976578712463,
+      "learning_rate": 9.960385711850554e-06,
+      "loss": 0.8861,
+      "step": 180
+    },
+    {
+      "epoch": 0.07315429781499663,
+      "grad_norm": 0.3005346357822418,
+      "learning_rate": 9.952166110886394e-06,
+      "loss": 0.876,
+      "step": 190
+    },
+    {
+      "epoch": 0.07700452401578593,
+      "grad_norm": 0.354958713054657,
+      "learning_rate": 9.943176257655567e-06,
+      "loss": 0.8655,
+      "step": 200
+    },
+    {
+      "epoch": 0.08085475021657522,
+      "grad_norm": 0.30377423763275146,
+      "learning_rate": 9.933417550425952e-06,
+      "loss": 0.8595,
+      "step": 210
+    },
+    {
+      "epoch": 0.08470497641736452,
+      "grad_norm": 0.293337881565094,
+      "learning_rate": 9.9228915070518e-06,
+      "loss": 0.8479,
+      "step": 220
+    },
+    {
+      "epoch": 0.08855520261815382,
+      "grad_norm": 0.2932908236980438,
+      "learning_rate": 9.911599764737632e-06,
+      "loss": 0.8485,
+      "step": 230
+    },
+    {
+      "epoch": 0.09240542881894312,
+      "grad_norm": 0.2997569441795349,
+      "learning_rate": 9.899544079783606e-06,
+      "loss": 0.8437,
+      "step": 240
+    },
+    {
+      "epoch": 0.0962556550197324,
+      "grad_norm": 0.3554689884185791,
+      "learning_rate": 9.886726327312338e-06,
+      "loss": 0.8591,
+      "step": 250
+    },
+    {
+      "epoch": 0.1001058812205217,
+      "grad_norm": 0.2766917049884796,
+      "learning_rate": 9.87314850097724e-06,
+      "loss": 0.856,
+      "step": 260
+    },
+    {
+      "epoch": 0.103956107421311,
+      "grad_norm": 0.31001904606819153,
+      "learning_rate": 9.858812712652452e-06,
+      "loss": 0.8438,
+      "step": 270
+    },
+    {
+      "epoch": 0.10780633362210029,
+      "grad_norm": 0.27615422010421753,
+      "learning_rate": 9.843721192104341e-06,
+      "loss": 0.8436,
+      "step": 280
+    },
+    {
+      "epoch": 0.11165655982288959,
+      "grad_norm": 0.29711732268333435,
+      "learning_rate": 9.827876286644698e-06,
+      "loss": 0.8602,
+      "step": 290
+    },
+    {
+      "epoch": 0.11550678602367889,
+      "grad_norm": 0.261282354593277,
+      "learning_rate": 9.811280460765643e-06,
+      "loss": 0.8207,
+      "step": 300
+    },
+    {
+      "epoch": 0.11935701222446819,
+      "grad_norm": 0.27261510491371155,
+      "learning_rate": 9.793936295756292e-06,
+      "loss": 0.8351,
+      "step": 310
+    },
+    {
+      "epoch": 0.12320723842525748,
+      "grad_norm": 0.2814849615097046,
+      "learning_rate": 9.775846489301279e-06,
+      "loss": 0.8342,
+      "step": 320
+    },
+    {
+      "epoch": 0.12705746462604678,
+      "grad_norm": 0.2970620393753052,
+      "learning_rate": 9.757013855061157e-06,
+      "loss": 0.8344,
+      "step": 330
+    },
+    {
+      "epoch": 0.13090769082683607,
+      "grad_norm": 0.27238816022872925,
+      "learning_rate": 9.737441322234758e-06,
+      "loss": 0.8452,
+      "step": 340
+    },
+    {
+      "epoch": 0.13475791702762538,
+      "grad_norm": 0.2974564731121063,
+      "learning_rate": 9.717131935103605e-06,
+      "loss": 0.8454,
+      "step": 350
+    },
+    {
+      "epoch": 0.13860814322841467,
+      "grad_norm": 0.27427807450294495,
+      "learning_rate": 9.696088852558398e-06,
+      "loss": 0.832,
+      "step": 360
+    },
+    {
+      "epoch": 0.14245836942920398,
+      "grad_norm": 0.29826340079307556,
+      "learning_rate": 9.674315347607686e-06,
+      "loss": 0.8581,
+      "step": 370
+    },
+    {
+      "epoch": 0.14630859562999327,
+      "grad_norm": 0.28688615560531616,
+      "learning_rate": 9.651814806868796e-06,
+      "loss": 0.8182,
+      "step": 380
+    },
+    {
+      "epoch": 0.15015882183078255,
+      "grad_norm": 0.29026779532432556,
+      "learning_rate": 9.62859073004107e-06,
+      "loss": 0.8267,
+      "step": 390
+    },
+    {
+      "epoch": 0.15400904803157187,
+      "grad_norm": 0.2825547456741333,
+      "learning_rate": 9.604646729361547e-06,
+      "loss": 0.8302,
+      "step": 400
+    },
+    {
+      "epoch": 0.15785927423236115,
+      "grad_norm": 0.26869380474090576,
+      "learning_rate": 9.579986529043099e-06,
+      "loss": 0.8187,
+      "step": 410
+    },
+    {
+      "epoch": 0.16170950043315044,
+      "grad_norm": 0.2695952355861664,
+      "learning_rate": 9.554613964695189e-06,
+      "loss": 0.8348,
+      "step": 420
+    },
+    {
+      "epoch": 0.16555972663393975,
+      "grad_norm": 0.3057861626148224,
+      "learning_rate": 9.528532982727281e-06,
+      "loss": 0.8251,
+      "step": 430
+    },
+    {
+      "epoch": 0.16940995283472904,
+      "grad_norm": 0.28000572323799133,
+      "learning_rate": 9.501747639735018e-06,
+      "loss": 0.8228,
+      "step": 440
+    },
+    {
+      "epoch": 0.17326017903551832,
+      "grad_norm": 0.2910799980163574,
+      "learning_rate": 9.474262101869272e-06,
+      "loss": 0.8125,
+      "step": 450
+    },
+    {
+      "epoch": 0.17711040523630764,
+      "grad_norm": 0.2715372145175934,
+      "learning_rate": 9.446080644188141e-06,
+      "loss": 0.8181,
+      "step": 460
+    },
+    {
+      "epoch": 0.18096063143709692,
+      "grad_norm": 0.27517038583755493,
+      "learning_rate": 9.417207649992017e-06,
+      "loss": 0.8267,
+      "step": 470
+    },
+    {
+      "epoch": 0.18481085763788624,
+      "grad_norm": 0.26840120553970337,
+      "learning_rate": 9.387647610141816e-06,
+      "loss": 0.825,
+      "step": 480
+    },
+    {
+      "epoch": 0.18866108383867553,
+      "grad_norm": 0.2754914164543152,
+      "learning_rate": 9.35740512236047e-06,
+      "loss": 0.8114,
+      "step": 490
+    },
+    {
+      "epoch": 0.1925113100394648,
+      "grad_norm": 0.2608506679534912,
+      "learning_rate": 9.326484890517809e-06,
+      "loss": 0.816,
+      "step": 500
+    },
+    {
+      "epoch": 0.19636153624025413,
+      "grad_norm": 0.2955158054828644,
+      "learning_rate": 9.294891723898926e-06,
+      "loss": 0.8125,
+      "step": 510
+    },
+    {
+      "epoch": 0.2002117624410434,
+      "grad_norm": 0.2819397449493408,
+      "learning_rate": 9.262630536456153e-06,
+      "loss": 0.8415,
+      "step": 520
+    },
+    {
+      "epoch": 0.2040619886418327,
+      "grad_norm": 0.30900993943214417,
+      "learning_rate": 9.229706346044749e-06,
+      "loss": 0.8133,
+      "step": 530
+    },
+    {
+      "epoch": 0.207912214842622,
+      "grad_norm": 0.2986776828765869,
+      "learning_rate": 9.196124273642422e-06,
+      "loss": 0.8274,
+      "step": 540
+    },
+    {
+      "epoch": 0.2117624410434113,
+      "grad_norm": 0.2652430236339569,
+      "learning_rate": 9.161889542552842e-06,
+      "loss": 0.8288,
+      "step": 550
+    },
+    {
+      "epoch": 0.21561266724420058,
+      "grad_norm": 0.24929198622703552,
+      "learning_rate": 9.127007477593193e-06,
+      "loss": 0.8338,
+      "step": 560
+    },
+    {
+      "epoch": 0.2194628934449899,
+      "grad_norm": 0.30876001715660095,
+      "learning_rate": 9.091483504265974e-06,
+      "loss": 0.8427,
+      "step": 570
+    },
+    {
+      "epoch": 0.22331311964577918,
+      "grad_norm": 0.30118411779403687,
+      "learning_rate": 9.055323147915116e-06,
+      "loss": 0.8224,
+      "step": 580
+    },
+    {
+      "epoch": 0.2271633458465685,
+      "grad_norm": 0.30162110924720764,
+      "learning_rate": 9.018532032866585e-06,
+      "loss": 0.8331,
+      "step": 590
+    },
+    {
+      "epoch": 0.23101357204735778,
+      "grad_norm": 0.26771101355552673,
+      "learning_rate": 8.981115881553579e-06,
+      "loss": 0.8237,
+      "step": 600
+    },
+    {
+      "epoch": 0.23486379824814707,
+      "grad_norm": 0.2961714267730713,
+      "learning_rate": 8.943080513626476e-06,
+      "loss": 0.8437,
+      "step": 610
+    },
+    {
+      "epoch": 0.23871402444893638,
+      "grad_norm": 0.2610303461551666,
+      "learning_rate": 8.904431845047641e-06,
+      "loss": 0.8277,
+      "step": 620
+    },
+    {
+      "epoch": 0.24256425064972567,
+      "grad_norm": 0.2921963632106781,
+      "learning_rate": 8.86517588717128e-06,
+      "loss": 0.8444,
+      "step": 630
+    },
+    {
+      "epoch": 0.24641447685051496,
+      "grad_norm": 0.2536105513572693,
+      "learning_rate": 8.82531874580844e-06,
+      "loss": 0.8325,
+      "step": 640
+    },
+    {
+      "epoch": 0.25026470305130427,
+      "grad_norm": 0.26467815041542053,
+      "learning_rate": 8.784866620277314e-06,
+      "loss": 0.7894,
+      "step": 650
+    },
+    {
+      "epoch": 0.25411492925209356,
+      "grad_norm": 0.3054841160774231,
+      "learning_rate": 8.743825802439018e-06,
+      "loss": 0.8275,
+      "step": 660
+    },
+    {
+      "epoch": 0.25796515545288284,
+      "grad_norm": 0.31488874554634094,
+      "learning_rate": 8.702202675718956e-06,
+      "loss": 0.8323,
+      "step": 670
+    },
+    {
+      "epoch": 0.26181538165367213,
+      "grad_norm": 0.2873603403568268,
+      "learning_rate": 8.660003714113959e-06,
+      "loss": 0.8025,
+      "step": 680
+    },
+    {
+      "epoch": 0.26566560785446147,
+      "grad_norm": 0.271373450756073,
+      "learning_rate": 8.617235481185323e-06,
+      "loss": 0.8279,
+      "step": 690
+    },
+    {
+      "epoch": 0.26951583405525076,
+      "grad_norm": 0.271182119846344,
+      "learning_rate": 8.573904629037931e-06,
+      "loss": 0.8321,
+      "step": 700
+    },
+    {
+      "epoch": 0.27336606025604004,
+      "grad_norm": 0.25431257486343384,
+      "learning_rate": 8.530017897285585e-06,
+      "loss": 0.8118,
+      "step": 710
+    },
+    {
+      "epoch": 0.27721628645682933,
+      "grad_norm": 0.2595232129096985,
+      "learning_rate": 8.485582112002746e-06,
+      "loss": 0.8162,
+      "step": 720
+    },
+    {
+      "epoch": 0.2810665126576186,
+      "grad_norm": 0.28155770897865295,
+      "learning_rate": 8.440604184662811e-06,
+      "loss": 0.8219,
+      "step": 730
+    },
+    {
+      "epoch": 0.28491673885840796,
+      "grad_norm": 0.2698063850402832,
+      "learning_rate": 8.395091111063118e-06,
+      "loss": 0.8115,
+      "step": 740
+    },
+    {
+      "epoch": 0.28876696505919724,
+      "grad_norm": 0.26184043288230896,
+      "learning_rate": 8.349049970236822e-06,
+      "loss": 0.8098,
+      "step": 750
+    },
+    {
+      "epoch": 0.29261719125998653,
+      "grad_norm": 0.28332093358039856,
+      "learning_rate": 8.302487923351842e-06,
+      "loss": 0.8028,
+      "step": 760
+    },
+    {
+      "epoch": 0.2964674174607758,
+      "grad_norm": 0.26574528217315674,
+      "learning_rate": 8.255412212597021e-06,
+      "loss": 0.8206,
+      "step": 770
+    },
+    {
+      "epoch": 0.3003176436615651,
+      "grad_norm": 0.26477858424186707,
+      "learning_rate": 8.207830160055688e-06,
+      "loss": 0.8168,
+      "step": 780
+    },
+    {
+      "epoch": 0.3041678698623544,
+      "grad_norm": 0.2909710109233856,
+      "learning_rate": 8.159749166566786e-06,
+      "loss": 0.802,
+      "step": 790
+    },
+    {
+      "epoch": 0.30801809606314373,
+      "grad_norm": 0.26872196793556213,
+      "learning_rate": 8.11117671057377e-06,
+      "loss": 0.8295,
+      "step": 800
+    },
+    {
+      "epoch": 0.311868322263933,
+      "grad_norm": 0.2955751121044159,
+      "learning_rate": 8.062120346961404e-06,
+      "loss": 0.8214,
+      "step": 810
+    },
+    {
+      "epoch": 0.3157185484647223,
+      "grad_norm": 0.2856772243976593,
+      "learning_rate": 8.012587705880705e-06,
+      "loss": 0.8143,
+      "step": 820
+    },
+    {
+      "epoch": 0.3195687746655116,
+      "grad_norm": 0.27354907989501953,
+      "learning_rate": 7.962586491562146e-06,
+      "loss": 0.8011,
+      "step": 830
+    },
+    {
+      "epoch": 0.3234190008663009,
+      "grad_norm": 0.2943623661994934,
+      "learning_rate": 7.912124481117357e-06,
+      "loss": 0.8064,
+      "step": 840
+    },
+    {
+      "epoch": 0.3272692270670902,
+      "grad_norm": 0.2794247269630432,
+      "learning_rate": 7.861209523329494e-06,
+      "loss": 0.8164,
+      "step": 850
+    },
+    {
+      "epoch": 0.3311194532678795,
+      "grad_norm": 0.2704644203186035,
+      "learning_rate": 7.809849537432432e-06,
+      "loss": 0.8008,
+      "step": 860
+    },
+    {
+      "epoch": 0.3349696794686688,
+      "grad_norm": 0.2663770914077759,
+      "learning_rate": 7.758052511879035e-06,
+      "loss": 0.8112,
+      "step": 870
+    },
+    {
+      "epoch": 0.3388199056694581,
+      "grad_norm": 0.2526009976863861,
+      "learning_rate": 7.705826503098642e-06,
+      "loss": 0.806,
+      "step": 880
+    },
+    {
+      "epoch": 0.34267013187024736,
+      "grad_norm": 0.25760725140571594,
+      "learning_rate": 7.653179634243979e-06,
+      "loss": 0.8055,
+      "step": 890
+    },
+    {
+      "epoch": 0.34652035807103665,
+      "grad_norm": 0.2461443543434143,
+      "learning_rate": 7.600120093927696e-06,
+      "loss": 0.8123,
+      "step": 900
+    },
+    {
+      "epoch": 0.350370584271826,
+      "grad_norm": 0.2686060965061188,
+      "learning_rate": 7.546656134948729e-06,
+      "loss": 0.8013,
+      "step": 910
+    },
+    {
+      "epoch": 0.3542208104726153,
+      "grad_norm": 0.27574703097343445,
+      "learning_rate": 7.492796073008661e-06,
+      "loss": 0.7994,
+      "step": 920
+    },
+    {
+      "epoch": 0.35807103667340456,
+      "grad_norm": 0.2963077425956726,
+      "learning_rate": 7.438548285418326e-06,
+      "loss": 0.785,
+      "step": 930
+    },
+    {
+      "epoch": 0.36192126287419385,
+      "grad_norm": 0.3159525990486145,
+      "learning_rate": 7.3839212097948e-06,
+      "loss": 0.8193,
+      "step": 940
+    },
+    {
+      "epoch": 0.36577148907498314,
+      "grad_norm": 0.26310208439826965,
+      "learning_rate": 7.328923342749039e-06,
+      "loss": 0.8344,
+      "step": 950
+    },
+    {
+      "epoch": 0.3696217152757725,
+      "grad_norm": 0.28071483969688416,
+      "learning_rate": 7.273563238564319e-06,
+      "loss": 0.8181,
+      "step": 960
+    },
+    {
+      "epoch": 0.37347194147656176,
+      "grad_norm": 0.27077096700668335,
+      "learning_rate": 7.217849507865724e-06,
+      "loss": 0.7991,
+      "step": 970
+    },
+    {
+      "epoch": 0.37732216767735105,
+      "grad_norm": 0.26021552085876465,
+      "learning_rate": 7.161790816280858e-06,
+      "loss": 0.8042,
+      "step": 980
+    },
+    {
+      "epoch": 0.38117239387814034,
+      "grad_norm": 0.29217037558555603,
+      "learning_rate": 7.1053958830920125e-06,
+      "loss": 0.7918,
+      "step": 990
+    },
+    {
+      "epoch": 0.3850226200789296,
+      "grad_norm": 0.26735565066337585,
+      "learning_rate": 7.0486734798799715e-06,
+      "loss": 0.8071,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3888728462797189,
+      "grad_norm": 0.27776283025741577,
+      "learning_rate": 6.991632429159708e-06,
+      "loss": 0.8301,
+      "step": 1010
+    },
+    {
+      "epoch": 0.39272307248050825,
+      "grad_norm": 0.28258854150772095,
+      "learning_rate": 6.934281603008132e-06,
+      "loss": 0.8217,
+      "step": 1020
+    },
+    {
+      "epoch": 0.39657329868129754,
+      "grad_norm": 0.27626365423202515,
+      "learning_rate": 6.876629921684143e-06,
+      "loss": 0.8192,
+      "step": 1030
+    },
+    {
+      "epoch": 0.4004235248820868,
+      "grad_norm": 0.2908881604671478,
+      "learning_rate": 6.818686352241199e-06,
+      "loss": 0.81,
+      "step": 1040
+    },
+    {
+      "epoch": 0.4042737510828761,
+      "grad_norm": 0.2527127265930176,
+      "learning_rate": 6.760459907132573e-06,
+      "loss": 0.8059,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4081239772836654,
+      "grad_norm": 0.2542715072631836,
+      "learning_rate": 6.701959642809589e-06,
+      "loss": 0.813,
+      "step": 1060
+    },
+    {
+      "epoch": 0.41197420348445474,
+      "grad_norm": 0.27520641684532166,
+      "learning_rate": 6.643194658312987e-06,
+      "loss": 0.8051,
+      "step": 1070
+    },
+    {
+      "epoch": 0.415824429685244,
+      "grad_norm": 0.2743719816207886,
+      "learning_rate": 6.584174093857676e-06,
+      "loss": 0.8285,
+      "step": 1080
+    },
+    {
+      "epoch": 0.4196746558860333,
+      "grad_norm": 0.2897748351097107,
+      "learning_rate": 6.5249071294110735e-06,
+      "loss": 0.8089,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4235248820868226,
+      "grad_norm": 0.28994622826576233,
+      "learning_rate": 6.465402983265274e-06,
+      "loss": 0.8231,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4273751082876119,
+      "grad_norm": 0.27983951568603516,
+      "learning_rate": 6.405670910603243e-06,
+      "loss": 0.819,
+      "step": 1110
+    },
+    {
+      "epoch": 0.43122533448840117,
+      "grad_norm": 0.30247461795806885,
+      "learning_rate": 6.345720202059291e-06,
+      "loss": 0.815,
+      "step": 1120
+    },
+    {
+      "epoch": 0.4350755606891905,
+      "grad_norm": 0.2754497826099396,
+      "learning_rate": 6.285560182274008e-06,
+      "loss": 0.7943,
+      "step": 1130
+    },
+    {
+      "epoch": 0.4389257868899798,
+      "grad_norm": 0.26850762963294983,
+      "learning_rate": 6.225200208443936e-06,
+      "loss": 0.8163,
+      "step": 1140
+    },
+    {
+      "epoch": 0.4427760130907691,
+      "grad_norm": 0.24763540923595428,
+      "learning_rate": 6.16464966886616e-06,
+      "loss": 0.8066,
+      "step": 1150
+    },
+    {
+      "epoch": 0.44662623929155837,
+      "grad_norm": 0.30838659405708313,
+      "learning_rate": 6.103917981478065e-06,
+      "loss": 0.8287,
+      "step": 1160
+    },
+    {
+      "epoch": 0.45047646549234766,
+      "grad_norm": 0.26564517617225647,
+      "learning_rate": 6.043014592392489e-06,
+      "loss": 0.84,
+      "step": 1170
+    },
+    {
+      "epoch": 0.454326691693137,
+      "grad_norm": 0.25592073798179626,
+      "learning_rate": 5.981948974428485e-06,
+      "loss": 0.8145,
+      "step": 1180
+    },
+    {
+      "epoch": 0.4581769178939263,
+      "grad_norm": 0.27388235926628113,
+      "learning_rate": 5.920730625637934e-06,
+      "loss": 0.7955,
+      "step": 1190
+    },
+    {
+      "epoch": 0.46202714409471557,
+      "grad_norm": 0.2692316770553589,
+      "learning_rate": 5.8593690678282355e-06,
+      "loss": 0.8038,
+      "step": 1200
+    },
+    {
+      "epoch": 0.46587737029550486,
+      "grad_norm": 0.28685471415519714,
+      "learning_rate": 5.797873845081301e-06,
+      "loss": 0.8085,
+      "step": 1210
+    },
+    {
+      "epoch": 0.46972759649629414,
+      "grad_norm": 0.2775149345397949,
+      "learning_rate": 5.736254522269075e-06,
+      "loss": 0.8256,
+      "step": 1220
+    },
+    {
+      "epoch": 0.47357782269708343,
+      "grad_norm": 0.27470898628234863,
+      "learning_rate": 5.6745206835658386e-06,
+      "loss": 0.8269,
+      "step": 1230
+    },
+    {
+      "epoch": 0.47742804889787277,
+      "grad_norm": 0.26433852314949036,
+      "learning_rate": 5.612681930957495e-06,
+      "loss": 0.7977,
+      "step": 1240
+    },
+    {
+      "epoch": 0.48127827509866206,
+      "grad_norm": 0.25470295548439026,
+      "learning_rate": 5.550747882748095e-06,
+      "loss": 0.7966,
+      "step": 1250
+    },
+    {
+      "epoch": 0.48512850129945134,
+      "grad_norm": 0.3010089695453644,
+      "learning_rate": 5.4887281720638195e-06,
+      "loss": 0.8014,
+      "step": 1260
+    },
+    {
+      "epoch": 0.48897872750024063,
+      "grad_norm": 0.3032512664794922,
+      "learning_rate": 5.42663244535466e-06,
+      "loss": 0.7901,
+      "step": 1270
+    },
+    {
+      "epoch": 0.4928289537010299,
+      "grad_norm": 0.3038042485713959,
+      "learning_rate": 5.3644703608940205e-06,
+      "loss": 0.7961,
+      "step": 1280
+    },
+    {
+      "epoch": 0.49667917990181926,
+      "grad_norm": 0.2762264311313629,
+      "learning_rate": 5.302251587276491e-06,
+      "loss": 0.8073,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5005294061026085,
+      "grad_norm": 0.25900888442993164,
+      "learning_rate": 5.2399858019140005e-06,
+      "loss": 0.8079,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5043796323033978,
+      "grad_norm": 0.2611720860004425,
+      "learning_rate": 5.177682689530615e-06,
+      "loss": 0.8224,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5082298585041871,
+      "grad_norm": 0.27345705032348633,
+      "learning_rate": 5.11535194065619e-06,
+      "loss": 0.7899,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5120800847049765,
+      "grad_norm": 0.3005550503730774,
+      "learning_rate": 5.053003250119113e-06,
+      "loss": 0.8014,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5159303109057657,
+      "grad_norm": 0.2734124958515167,
+      "learning_rate": 4.9906463155383925e-06,
+      "loss": 0.8073,
+      "step": 1340
+    },
+    {
+      "epoch": 0.519780537106555,
+      "grad_norm": 0.2689356207847595,
+      "learning_rate": 4.9282908358153046e-06,
+      "loss": 0.7984,
+      "step": 1350
+    },
+    {
+      "epoch": 0.5236307633073443,
+      "grad_norm": 0.30117401480674744,
+      "learning_rate": 4.8659465096248345e-06,
+      "loss": 0.8173,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5274809895081336,
+      "grad_norm": 0.29359200596809387,
+      "learning_rate": 4.803623033907169e-06,
+      "loss": 0.8175,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5313312157089229,
+      "grad_norm": 0.24939952790737152,
+      "learning_rate": 4.741330102359441e-06,
+      "loss": 0.8219,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5351814419097122,
+      "grad_norm": 0.28408271074295044,
+      "learning_rate": 4.679077403927996e-06,
+      "loss": 0.8021,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5390316681105015,
+      "grad_norm": 0.2781761586666107,
+      "learning_rate": 4.616874621301373e-06,
+      "loss": 0.8009,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5428818943112907,
+      "grad_norm": 0.28376200795173645,
+      "learning_rate": 4.554731429404293e-06,
+      "loss": 0.8159,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5467321205120801,
+      "grad_norm": 0.2923646569252014,
+      "learning_rate": 4.492657493892822e-06,
+      "loss": 0.8022,
+      "step": 1420
+    },
+    {
+      "epoch": 0.5505823467128694,
+      "grad_norm": 0.30053791403770447,
+      "learning_rate": 4.430662469650999e-06,
+      "loss": 0.816,
+      "step": 1430
+    },
+    {
+      "epoch": 0.5544325729136587,
+      "grad_norm": 0.2876088321208954,
+      "learning_rate": 4.368755999289118e-06,
+      "loss": 0.8178,
+      "step": 1440
+    },
+    {
+      "epoch": 0.558282799114448,
+      "grad_norm": 0.30126968026161194,
+      "learning_rate": 4.306947711643949e-06,
+      "loss": 0.8049,
+      "step": 1450
+    },
+    {
+      "epoch": 0.5621330253152372,
+      "grad_norm": 0.2741060256958008,
+      "learning_rate": 4.245247220281068e-06,
+      "loss": 0.8063,
+      "step": 1460
+    },
+    {
+      "epoch": 0.5659832515160266,
+      "grad_norm": 0.28501754999160767,
+      "learning_rate": 4.183664121999587e-06,
+      "loss": 0.8053,
+      "step": 1470
+    },
+    {
+      "epoch": 0.5698334777168159,
+      "grad_norm": 0.29287654161453247,
+      "learning_rate": 4.122207995339474e-06,
+      "loss": 0.8001,
+      "step": 1480
+    },
+    {
+      "epoch": 0.5736837039176051,
+      "grad_norm": 0.28070566058158875,
+      "learning_rate": 4.060888399091734e-06,
+      "loss": 0.8064,
+      "step": 1490
+    },
+    {
+      "epoch": 0.5775339301183945,
+      "grad_norm": 0.26964443922042847,
+      "learning_rate": 3.999714870811639e-06,
+      "loss": 0.8132,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5813841563191837,
+      "grad_norm": 0.2642393708229065,
+      "learning_rate": 3.938696925335275e-06,
+      "loss": 0.8073,
+      "step": 1510
+    },
+    {
+      "epoch": 0.5852343825199731,
+      "grad_norm": 0.2634388208389282,
+      "learning_rate": 3.87784405329962e-06,
+      "loss": 0.8152,
+      "step": 1520
+    },
+    {
+      "epoch": 0.5890846087207623,
+      "grad_norm": 0.2749176025390625,
+      "learning_rate": 3.817165719666382e-06,
+      "loss": 0.8185,
+      "step": 1530
+    },
+    {
+      "epoch": 0.5929348349215516,
+      "grad_norm": 0.2638886868953705,
+      "learning_rate": 3.756671362249841e-06,
+      "loss": 0.8204,
+      "step": 1540
+    },
+    {
+      "epoch": 0.596785061122341,
+      "grad_norm": 0.2644767463207245,
+      "learning_rate": 3.6963703902488907e-06,
+      "loss": 0.8074,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6006352873231302,
+      "grad_norm": 0.26660779118537903,
+      "learning_rate": 3.6362721827835655e-06,
+      "loss": 0.8274,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6044855135239195,
+      "grad_norm": 0.276185542345047,
+      "learning_rate": 3.576386087436212e-06,
+      "loss": 0.8195,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6083357397247088,
+      "grad_norm": 0.27970919013023376,
+      "learning_rate": 3.5167214187975913e-06,
+      "loss": 0.8101,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6121859659254981,
+      "grad_norm": 0.2271667867898941,
+      "learning_rate": 3.457287457018086e-06,
+      "loss": 0.8122,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6160361921262875,
+      "grad_norm": 0.26994943618774414,
+      "learning_rate": 3.3980934463643003e-06,
+      "loss": 0.7992,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6198864183270767,
+      "grad_norm": 0.2469833791255951,
+      "learning_rate": 3.3391485937812053e-06,
+      "loss": 0.7823,
+      "step": 1610
+    },
+    {
+      "epoch": 0.623736644527866,
+      "grad_norm": 0.26032140851020813,
+      "learning_rate": 3.280462067460115e-06,
+      "loss": 0.7815,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6275868707286553,
+      "grad_norm": 0.2644536793231964,
+      "learning_rate": 3.222042995412669e-06,
+      "loss": 0.8117,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6314370969294446,
+      "grad_norm": 0.24432280659675598,
+      "learning_rate": 3.163900464051089e-06,
+      "loss": 0.802,
+      "step": 1640
+    },
+    {
+      "epoch": 0.635287323130234,
+      "grad_norm": 0.2752269506454468,
+      "learning_rate": 3.1060435167748826e-06,
+      "loss": 0.8102,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6391375493310232,
+      "grad_norm": 0.2819995880126953,
+      "learning_rate": 3.048481152564253e-06,
+      "loss": 0.7907,
+      "step": 1660
+    },
+    {
+      "epoch": 0.6429877755318125,
+      "grad_norm": 0.2852069139480591,
+      "learning_rate": 2.9912223245804017e-06,
+      "loss": 0.8124,
+      "step": 1670
+    },
+    {
+      "epoch": 0.6468380017326018,
+      "grad_norm": 0.2759275436401367,
+      "learning_rate": 2.934275938772981e-06,
+      "loss": 0.8113,
+      "step": 1680
+    },
+    {
+      "epoch": 0.6506882279333911,
+      "grad_norm": 0.2798722982406616,
+      "learning_rate": 2.8776508524948676e-06,
+      "loss": 0.7998,
+      "step": 1690
+    },
+    {
+      "epoch": 0.6545384541341804,
+      "grad_norm": 0.2374049574136734,
+      "learning_rate": 2.821355873124503e-06,
+      "loss": 0.799,
+      "step": 1700
+    },
+    {
+      "epoch": 0.6583886803349697,
+      "grad_norm": 0.2805563807487488,
+      "learning_rate": 2.7653997566960144e-06,
+      "loss": 0.8123,
+      "step": 1710
+    },
+    {
+      "epoch": 0.662238906535759,
+      "grad_norm": 0.2795714735984802,
+      "learning_rate": 2.709791206537309e-06,
+      "loss": 0.8021,
+      "step": 1720
+    },
+    {
+      "epoch": 0.6660891327365482,
+      "grad_norm": 0.2684919536113739,
+      "learning_rate": 2.6545388719163747e-06,
+      "loss": 0.8101,
+      "step": 1730
+    },
+    {
+      "epoch": 0.6699393589373376,
+      "grad_norm": 0.2645135819911957,
+      "learning_rate": 2.599651346695979e-06,
+      "loss": 0.8071,
+      "step": 1740
+    },
+    {
+      "epoch": 0.6737895851381268,
+      "grad_norm": 0.27560359239578247,
+      "learning_rate": 2.545137167997007e-06,
+      "loss": 0.8026,
+      "step": 1750
+    },
+    {
+      "epoch": 0.6776398113389162,
+      "grad_norm": 0.25950711965560913,
+      "learning_rate": 2.4910048148705972e-06,
+      "loss": 0.7908,
+      "step": 1760
+    },
+    {
+      "epoch": 0.6814900375397055,
+      "grad_norm": 0.2866698205471039,
+      "learning_rate": 2.4372627069793327e-06,
+      "loss": 0.8334,
+      "step": 1770
+    },
+    {
+      "epoch": 0.6853402637404947,
+      "grad_norm": 0.2598281800746918,
+      "learning_rate": 2.3839192032876486e-06,
+      "loss": 0.8203,
+      "step": 1780
+    },
+    {
+      "epoch": 0.6891904899412841,
+      "grad_norm": 0.26622211933135986,
+      "learning_rate": 2.3309826007617047e-06,
+      "loss": 0.8308,
+      "step": 1790
+    },
+    {
+      "epoch": 0.6930407161420733,
+      "grad_norm": 0.26827993988990784,
+      "learning_rate": 2.278461133078881e-06,
+      "loss": 0.795,
+      "step": 1800
+    },
+    {
+      "epoch": 0.6968909423428626,
+      "grad_norm": 0.26965147256851196,
+      "learning_rate": 2.2263629693471258e-06,
+      "loss": 0.7856,
+      "step": 1810
+    },
+    {
+      "epoch": 0.700741168543652,
+      "grad_norm": 0.2599840462207794,
+      "learning_rate": 2.1746962128343414e-06,
+      "loss": 0.8112,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7045913947444412,
+      "grad_norm": 0.261130690574646,
+      "learning_rate": 2.1234688997080183e-06,
+      "loss": 0.799,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7084416209452306,
+      "grad_norm": 0.28573477268218994,
+      "learning_rate": 2.0726889977853053e-06,
+      "loss": 0.8037,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7122918471460198,
+      "grad_norm": 0.29170966148376465,
+      "learning_rate": 2.022364405293703e-06,
+      "loss": 0.8159,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7161420733468091,
+      "grad_norm": 0.2663772702217102,
+      "learning_rate": 1.9725029496425814e-06,
+      "loss": 0.808,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7199922995475985,
+      "grad_norm": 0.2721380591392517,
+      "learning_rate": 1.923112386205727e-06,
+      "loss": 0.806,
+      "step": 1870
+    },
+    {
+      "epoch": 0.7238425257483877,
+      "grad_norm": 0.2550492286682129,
+      "learning_rate": 1.874200397115079e-06,
+      "loss": 0.812,
+      "step": 1880
+    },
+    {
+      "epoch": 0.727692751949177,
+      "grad_norm": 0.2764725685119629,
+      "learning_rate": 1.825774590065857e-06,
+      "loss": 0.7967,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7315429781499663,
+      "grad_norm": 0.2732618451118469,
+      "learning_rate": 1.7778424971332813e-06,
+      "loss": 0.803,
+      "step": 1900
+    },
+    {
+      "epoch": 0.7353932043507556,
+      "grad_norm": 0.265154629945755,
+      "learning_rate": 1.7304115736010407e-06,
+      "loss": 0.8029,
+      "step": 1910
+    },
+    {
+      "epoch": 0.739243430551545,
+      "grad_norm": 0.2691805362701416,
+      "learning_rate": 1.6834891968017108e-06,
+      "loss": 0.7942,
+      "step": 1920
+    },
+    {
+      "epoch": 0.7430936567523342,
+      "grad_norm": 0.2905904948711395,
+      "learning_rate": 1.6370826649692884e-06,
+      "loss": 0.816,
+      "step": 1930
+    },
+    {
+      "epoch": 0.7469438829531235,
+      "grad_norm": 0.281703382730484,
+      "learning_rate": 1.5911991961040463e-06,
+      "loss": 0.822,
+      "step": 1940
+    },
+    {
+      "epoch": 0.7507941091539128,
+      "grad_norm": 0.2760506570339203,
+      "learning_rate": 1.545845926849851e-06,
+      "loss": 0.8071,
+      "step": 1950
+    },
+    {
+      "epoch": 0.7546443353547021,
+      "grad_norm": 0.27864885330200195,
+      "learning_rate": 1.5010299113841397e-06,
+      "loss": 0.8263,
+      "step": 1960
+    },
+    {
+      "epoch": 0.7584945615554913,
+      "grad_norm": 0.2792971134185791,
+      "learning_rate": 1.4567581203207238e-06,
+      "loss": 0.8134,
+      "step": 1970
+    },
+    {
+      "epoch": 0.7623447877562807,
+      "grad_norm": 0.28140127658843994,
+      "learning_rate": 1.413037439625593e-06,
+      "loss": 0.8143,
+      "step": 1980
+    },
+    {
+      "epoch": 0.76619501395707,
+      "grad_norm": 0.2912014424800873,
+      "learning_rate": 1.3698746695458847e-06,
+      "loss": 0.8067,
+      "step": 1990
+    },
+    {
+      "epoch": 0.7700452401578592,
+      "grad_norm": 0.26314154267311096,
+      "learning_rate": 1.3272765235521745e-06,
+      "loss": 0.7983,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7738954663586486,
+      "grad_norm": 0.2631196677684784,
+      "learning_rate": 1.2852496272942877e-06,
+      "loss": 0.8055,
+      "step": 2010
+    },
+    {
+      "epoch": 0.7777456925594378,
+      "grad_norm": 0.250620573759079,
+      "learning_rate": 1.2438005175707396e-06,
+      "loss": 0.7943,
+      "step": 2020
+    },
+    {
+      "epoch": 0.7815959187602272,
+      "grad_norm": 0.2585567533969879,
+      "learning_rate": 1.2029356413120297e-06,
+      "loss": 0.7818,
+      "step": 2030
+    },
+    {
+      "epoch": 0.7854461449610165,
+      "grad_norm": 0.2663929760456085,
+      "learning_rate": 1.1626613545778786e-06,
+      "loss": 0.7979,
+      "step": 2040
+    },
+    {
+      "epoch": 0.7892963711618057,
+      "grad_norm": 0.25810953974723816,
+      "learning_rate": 1.1229839215686294e-06,
+      "loss": 0.8017,
+      "step": 2050
+    },
+    {
+      "epoch": 0.7931465973625951,
+      "grad_norm": 0.2393358051776886,
+      "learning_rate": 1.0839095136509188e-06,
+      "loss": 0.7892,
+      "step": 2060
+    },
+    {
+      "epoch": 0.7969968235633843,
+      "grad_norm": 0.2972220480442047,
+      "learning_rate": 1.045444208397791e-06,
+      "loss": 0.8113,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8008470497641736,
+      "grad_norm": 0.2628802955150604,
+      "learning_rate": 1.007593988643402e-06,
+      "loss": 0.7895,
+      "step": 2080
+    },
+    {
+      "epoch": 0.804697275964963,
+      "grad_norm": 0.26266899704933167,
+      "learning_rate": 9.703647415524613e-07,
+      "loss": 0.8115,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8085475021657522,
+      "grad_norm": 0.27808743715286255,
+      "learning_rate": 9.337622577045485e-07,
+      "loss": 0.8008,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8123977283665416,
+      "grad_norm": 0.26716500520706177,
+      "learning_rate": 8.977922301934588e-07,
+      "loss": 0.808,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8162479545673308,
+      "grad_norm": 0.2683115601539612,
+      "learning_rate": 8.624602537417009e-07,
+      "loss": 0.8102,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8200981807681201,
+      "grad_norm": 0.2916320264339447,
+      "learning_rate": 8.277718238303123e-07,
+      "loss": 0.7999,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8239484069689095,
+      "grad_norm": 0.2687273621559143,
+      "learning_rate": 7.937323358440935e-07,
+      "loss": 0.8023,
+      "step": 2140
+    },
+    {
+      "epoch": 0.8277986331696987,
+      "grad_norm": 0.2944130003452301,
+      "learning_rate": 7.603470842324223e-07,
+      "loss": 0.7959,
+      "step": 2150
+    },
+    {
+      "epoch": 0.831648859370488,
+      "grad_norm": 0.2593955397605896,
+      "learning_rate": 7.276212616857575e-07,
+      "loss": 0.811,
+      "step": 2160
+    },
+    {
+      "epoch": 0.8354990855712773,
+      "grad_norm": 0.24906399846076965,
+      "learning_rate": 6.955599583279837e-07,
+      "loss": 0.8131,
+      "step": 2170
+    },
+    {
+      "epoch": 0.8393493117720666,
+      "grad_norm": 0.2700301706790924,
+      "learning_rate": 6.641681609246981e-07,
+      "loss": 0.8023,
+      "step": 2180
+    },
+    {
+      "epoch": 0.8431995379728559,
+      "grad_norm": 0.2754879593849182,
+      "learning_rate": 6.334507521075711e-07,
+      "loss": 0.8126,
+      "step": 2190
+    },
+    {
+      "epoch": 0.8470497641736452,
+      "grad_norm": 0.2713867723941803,
+      "learning_rate": 6.034125096149179e-07,
+      "loss": 0.804,
+      "step": 2200
+    },
+    {
+      "epoch": 0.8508999903744345,
+      "grad_norm": 0.27584150433540344,
+      "learning_rate": 5.740581055485722e-07,
+      "loss": 0.7761,
+      "step": 2210
+    },
+    {
+      "epoch": 0.8547502165752238,
+      "grad_norm": 0.25458014011383057,
+      "learning_rate": 5.453921056471961e-07,
+      "loss": 0.811,
+      "step": 2220
+    },
+    {
+      "epoch": 0.8586004427760131,
+      "grad_norm": 0.27776286005973816,
+      "learning_rate": 5.174189685761288e-07,
+      "loss": 0.8028,
+      "step": 2230
+    },
+    {
+      "epoch": 0.8624506689768023,
+      "grad_norm": 0.2683608829975128,
+      "learning_rate": 4.901430452338979e-07,
+      "loss": 0.8145,
+      "step": 2240
+    },
+    {
+      "epoch": 0.8663008951775917,
+      "grad_norm": 0.2655920088291168,
+      "learning_rate": 4.6356857807548286e-07,
+      "loss": 0.7998,
+      "step": 2250
+    },
+    {
+      "epoch": 0.870151121378381,
+      "grad_norm": 0.2635996639728546,
+      "learning_rate": 4.376997004524541e-07,
+      "loss": 0.7887,
+      "step": 2260
+    },
+    {
+      "epoch": 0.8740013475791703,
+      "grad_norm": 0.2690141499042511,
+      "learning_rate": 4.12540435970073e-07,
+      "loss": 0.8006,
+      "step": 2270
+    },
+    {
+      "epoch": 0.8778515737799596,
+      "grad_norm": 0.2486380934715271,
+      "learning_rate": 3.880946978614703e-07,
+      "loss": 0.8069,
+      "step": 2280
+    },
+    {
+      "epoch": 0.8817017999807488,
+      "grad_norm": 0.26038703322410583,
+      "learning_rate": 3.643662883789878e-07,
+      "loss": 0.8145,
+      "step": 2290
+    },
+    {
+      "epoch": 0.8855520261815382,
+      "grad_norm": 0.251383900642395,
+      "learning_rate": 3.41358898202781e-07,
+      "loss": 0.7901,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8894022523823275,
+      "grad_norm": 0.25031036138534546,
+      "learning_rate": 3.190761058667741e-07,
+      "loss": 0.8014,
+      "step": 2310
+    },
+    {
+      "epoch": 0.8932524785831167,
+      "grad_norm": 0.25500062108039856,
+      "learning_rate": 2.9752137720206463e-07,
+      "loss": 0.8093,
+      "step": 2320
+    },
+    {
+      "epoch": 0.8971027047839061,
+      "grad_norm": 0.26776403188705444,
+      "learning_rate": 2.7669806479785067e-07,
+      "loss": 0.813,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9009529309846953,
+      "grad_norm": 0.2724411189556122,
+      "learning_rate": 2.566094074799763e-07,
+      "loss": 0.8087,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9048031571854847,
+      "grad_norm": 0.2659168243408203,
+      "learning_rate": 2.3725852980716557e-07,
+      "loss": 0.7871,
+      "step": 2350
+    },
+    {
+      "epoch": 0.908653383386274,
+      "grad_norm": 0.26471224427223206,
+      "learning_rate": 2.1864844158503983e-07,
+      "loss": 0.8022,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9125036095870632,
+      "grad_norm": 0.29799941182136536,
+      "learning_rate": 2.007820373979702e-07,
+      "loss": 0.7924,
+      "step": 2370
+    },
+    {
+      "epoch": 0.9163538357878526,
+      "grad_norm": 0.26369351148605347,
+      "learning_rate": 1.8366209615886532e-07,
+      "loss": 0.8015,
+      "step": 2380
+    },
+    {
+      "epoch": 0.9202040619886418,
+      "grad_norm": 0.2651843726634979,
+      "learning_rate": 1.6729128067693613e-07,
+      "loss": 0.8106,
+      "step": 2390
+    },
+    {
+      "epoch": 0.9240542881894311,
+      "grad_norm": 0.2775970697402954,
+      "learning_rate": 1.5167213724353426e-07,
+      "loss": 0.8126,
+      "step": 2400
+    },
+    {
+      "epoch": 0.9279045143902205,
+      "grad_norm": 0.2602279782295227,
+      "learning_rate": 1.3680709523610058e-07,
+      "loss": 0.7952,
+      "step": 2410
+    },
+    {
+      "epoch": 0.9317547405910097,
+      "grad_norm": 0.2603018283843994,
+      "learning_rate": 1.2269846674030418e-07,
+      "loss": 0.8183,
+      "step": 2420
+    },
+    {
+      "epoch": 0.935604966791799,
+      "grad_norm": 0.26794055104255676,
+      "learning_rate": 1.0934844619042595e-07,
+      "loss": 0.8142,
+      "step": 2430
+    },
+    {
+      "epoch": 0.9394551929925883,
+      "grad_norm": 0.28293582797050476,
+      "learning_rate": 9.675911002803961e-08,
+      "loss": 0.8037,
+      "step": 2440
+    },
+    {
+      "epoch": 0.9433054191933776,
+      "grad_norm": 0.27812570333480835,
+      "learning_rate": 8.493241637904381e-08,
+      "loss": 0.8018,
+      "step": 2450
+    },
+    {
+      "epoch": 0.9471556453941669,
+      "grad_norm": 0.2765512764453888,
+      "learning_rate": 7.387020474909689e-08,
+      "loss": 0.8204,
+      "step": 2460
+    },
+    {
+      "epoch": 0.9510058715949562,
+      "grad_norm": 0.27378353476524353,
+      "learning_rate": 6.357419573750856e-08,
+      "loss": 0.8162,
+      "step": 2470
+    },
+    {
+      "epoch": 0.9548560977957455,
+      "grad_norm": 0.2842361032962799,
+      "learning_rate": 5.4045990769613365e-08,
+      "loss": 0.7987,
+      "step": 2480
+    },
+    {
+      "epoch": 0.9587063239965348,
+      "grad_norm": 0.3113895356655121,
+      "learning_rate": 4.5287071847694434e-08,
+      "loss": 0.8042,
+      "step": 2490
+    },
+    {
+      "epoch": 0.9625565501973241,
+      "grad_norm": 0.2691528797149658,
+      "learning_rate": 3.7298801320467817e-08,
+      "loss": 0.7966,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9664067763981133,
+      "grad_norm": 0.2688901424407959,
+      "learning_rate": 3.0082421671192576e-08,
+      "loss": 0.7898,
+      "step": 2510
+    },
+    {
+      "epoch": 0.9702570025989027,
+      "grad_norm": 0.26523804664611816,
+      "learning_rate": 2.3639055324410355e-08,
+      "loss": 0.812,
+      "step": 2520
+    },
+    {
+      "epoch": 0.974107228799692,
+      "grad_norm": 0.26079612970352173,
+      "learning_rate": 1.7969704471369477e-08,
+      "loss": 0.7962,
+      "step": 2530
+    },
+    {
+      "epoch": 0.9779574550004813,
+      "grad_norm": 0.24929016828536987,
+      "learning_rate": 1.3075250914142413e-08,
+      "loss": 0.8,
+      "step": 2540
+    },
+    {
+      "epoch": 0.9818076812012706,
+      "grad_norm": 0.2572931945323944,
+      "learning_rate": 8.956455928476049e-09,
+      "loss": 0.8105,
+      "step": 2550
+    },
+    {
+      "epoch": 0.9856579074020598,
+      "grad_norm": 0.26398971676826477,
+      "learning_rate": 5.6139601453786275e-09,
+      "loss": 0.8023,
+      "step": 2560
+    },
+    {
+      "epoch": 0.9895081336028492,
+      "grad_norm": 0.2694685757160187,
+      "learning_rate": 3.0482834514805602e-09,
+      "loss": 0.7899,
+      "step": 2570
+    },
+    {
+      "epoch": 0.9933583598036385,
+      "grad_norm": 0.2551470398902893,
+      "learning_rate": 1.2598249081713366e-09,
+      "loss": 0.7861,
+      "step": 2580
+    },
+    {
+      "epoch": 0.9972085860044277,
+      "grad_norm": 0.27522730827331543,
+      "learning_rate": 2.488626895302826e-10,
+      "loss": 0.8142,
+      "step": 2590
+    },
+    {
+      "epoch": 0.9999037443449803,
+      "step": 2597,
+      "total_flos": 3.6088477382690734e+18,
+      "train_loss": 0.8450901004319747,
+      "train_runtime": 12949.5265,
+      "train_samples_per_second": 51.346,
+      "train_steps_per_second": 0.201
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2597,
+  "num_input_tokens_seen": 418855352,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.6088477382690734e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

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

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:cc40b9cc75e165e025780ac05de8c09b9cbe620a87f100f525c3811c51b4e4df
+size 6264

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

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

continuous/qwen3/eupe_vit_s/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": 384,
+  "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": "vtb_ssl:eupe",
+  "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": "/cache/ckpt/download/tokenizer/continuous/eupe/EUPE-ViT-S/EUPE-ViT-S.pt",
+  "vocab_size": 151936
+}

continuous/qwen3/eupe_vit_s/pretrain/mm_projector.bin ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:278308ad77e42d1e653c5e8049f19c356da22adfa048c92e9a7cfb95637ec42d
+size 19941432

continuous/qwen3/eupe_vit_s/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": 0.9480026960372925,
+      "learning_rate": 0.00032142857142857147,
+      "loss": 10.7738,
+      "step": 10
+    },
+    {
+      "epoch": 0.03669724770642202,
+      "grad_norm": 0.833734929561615,
+      "learning_rate": 0.0006785714285714287,
+      "loss": 7.1843,
+      "step": 20
+    },
+    {
+      "epoch": 0.05504587155963303,
+      "grad_norm": 0.31272485852241516,
+      "learning_rate": 0.0009999907688176173,
+      "loss": 5.5634,
+      "step": 30
+    },
+    {
+      "epoch": 0.07339449541284404,
+      "grad_norm": 0.14902356266975403,
+      "learning_rate": 0.0009988834393115766,
+      "loss": 5.0267,
+      "step": 40
+    },
+    {
+      "epoch": 0.09174311926605505,
+      "grad_norm": 0.08119824528694153,
+      "learning_rate": 0.0009959345572940772,
+      "loss": 4.8704,
+      "step": 50
+    },
+    {
+      "epoch": 0.11009174311926606,
+      "grad_norm": 0.052465666085481644,
+      "learning_rate": 0.0009911550081155983,
+      "loss": 4.8028,
+      "step": 60
+    },
+    {
+      "epoch": 0.12844036697247707,
+      "grad_norm": 0.06398725509643555,
+      "learning_rate": 0.0009845624347566433,
+      "loss": 4.7484,
+      "step": 70
+    },
+    {
+      "epoch": 0.14678899082568808,
+      "grad_norm": 0.062098175287246704,
+      "learning_rate": 0.0009761811727013547,
+      "loss": 4.7212,
+      "step": 80
+    },
+    {
+      "epoch": 0.1651376146788991,
+      "grad_norm": 0.0815352201461792,
+      "learning_rate": 0.0009660421601067666,
+      "loss": 4.6865,
+      "step": 90
+    },
+    {
+      "epoch": 0.1834862385321101,
+      "grad_norm": 0.11144111305475235,
+      "learning_rate": 0.0009541828235992926,
+      "loss": 4.6671,
+      "step": 100
+    },
+    {
+      "epoch": 0.2018348623853211,
+      "grad_norm": 0.07532529532909393,
+      "learning_rate": 0.0009406469401200151,
+      "loss": 4.6367,
+      "step": 110
+    },
+    {
+      "epoch": 0.22018348623853212,
+      "grad_norm": 0.07476724684238434,
+      "learning_rate": 0.0009254844753287492,
+      "loss": 4.6257,
+      "step": 120
+    },
+    {
+      "epoch": 0.23853211009174313,
+      "grad_norm": 0.08246374875307083,
+      "learning_rate": 0.0009087513991633924,
+      "loss": 4.5956,
+      "step": 130
+    },
+    {
+      "epoch": 0.25688073394495414,
+      "grad_norm": 0.08919081091880798,
+      "learning_rate": 0.0008905094792353917,
+      "loss": 4.5742,
+      "step": 140
+    },
+    {
+      "epoch": 0.27522935779816515,
+      "grad_norm": 0.0858825072646141,
+      "learning_rate": 0.0008708260528239789,
+      "loss": 4.5659,
+      "step": 150
+    },
+    {
+      "epoch": 0.29357798165137616,
+      "grad_norm": 0.12631291151046753,
+      "learning_rate": 0.0008497737783108237,
+      "loss": 4.5476,
+      "step": 160
+    },
+    {
+      "epoch": 0.3119266055045872,
+      "grad_norm": 0.11523227393627167,
+      "learning_rate": 0.0008274303669726426,
+      "loss": 4.526,
+      "step": 170
+    },
+    {
+      "epoch": 0.3302752293577982,
+      "grad_norm": 0.08967792242765427,
+      "learning_rate": 0.0008038782961218137,
+      "loss": 4.5109,
+      "step": 180
+    },
+    {
+      "epoch": 0.3486238532110092,
+      "grad_norm": 0.10110028088092804,
+      "learning_rate": 0.000779204504653894,
+      "loss": 4.5029,
+      "step": 190
+    },
+    {
+      "epoch": 0.3669724770642202,
+      "grad_norm": 0.09860724955797195,
+      "learning_rate": 0.0007535000721258791,
+      "loss": 4.4791,
+      "step": 200
+    },
+    {
+      "epoch": 0.3853211009174312,
+      "grad_norm": 0.12845861911773682,
+      "learning_rate": 0.00072685988254984,
+      "loss": 4.4746,
+      "step": 210
+    },
+    {
+      "epoch": 0.4036697247706422,
+      "grad_norm": 0.09165429323911667,
+      "learning_rate": 0.0006993822741429906,
+      "loss": 4.487,
+      "step": 220
+    },
+    {
+      "epoch": 0.42201834862385323,
+      "grad_norm": 0.10490678995847702,
+      "learning_rate": 0.0006711686763270818,
+      "loss": 4.4798,
+      "step": 230
+    },
+    {
+      "epoch": 0.44036697247706424,
+      "grad_norm": 0.08985967189073563,
+      "learning_rate": 0.0006423232353170831,
+      "loss": 4.4618,
+      "step": 240
+    },
+    {
+      "epoch": 0.45871559633027525,
+      "grad_norm": 0.07988454401493073,
+      "learning_rate": 0.0006129524296812335,
+      "loss": 4.4454,
+      "step": 250
+    },
+    {
+      "epoch": 0.47706422018348627,
+      "grad_norm": 0.08549457043409348,
+      "learning_rate": 0.0005831646772915651,
+      "loss": 4.4518,
+      "step": 260
+    },
+    {
+      "epoch": 0.4954128440366973,
+      "grad_norm": 0.12344008684158325,
+      "learning_rate": 0.0005530699351157809,
+      "loss": 4.4341,
+      "step": 270
+    },
+    {
+      "epoch": 0.5137614678899083,
+      "grad_norm": 0.07952326536178589,
+      "learning_rate": 0.0005227792933277943,
+      "loss": 4.4284,
+      "step": 280
+    },
+    {
+      "epoch": 0.5321100917431193,
+      "grad_norm": 0.09339901059865952,
+      "learning_rate": 0.0004924045652352092,
+      "loss": 4.4112,
+      "step": 290
+    },
+    {
+      "epoch": 0.5504587155963303,
+      "grad_norm": 0.07493714988231659,
+      "learning_rate": 0.0004620578745374633,
+      "loss": 4.4095,
+      "step": 300
+    },
+    {
+      "epoch": 0.5688073394495413,
+      "grad_norm": 0.08445863425731659,
+      "learning_rate": 0.0004318512414382059,
+      "loss": 4.4256,
+      "step": 310
+    },
+    {
+      "epoch": 0.5871559633027523,
+      "grad_norm": 0.07300131767988205,
+      "learning_rate": 0.0004018961691397148,
+      "loss": 4.4162,
+      "step": 320
+    },
+    {
+      "epoch": 0.6055045871559633,
+      "grad_norm": 0.08370614796876907,
+      "learning_rate": 0.0003723032322457458,
+      "loss": 4.395,
+      "step": 330
+    },
+    {
+      "epoch": 0.6238532110091743,
+      "grad_norm": 0.08879134058952332,
+      "learning_rate": 0.0003431816685921625,
+      "loss": 4.3937,
+      "step": 340
+    },
+    {
+      "epoch": 0.6422018348623854,
+      "grad_norm": 0.09948540478944778,
+      "learning_rate": 0.0003146389760120416,
+      "loss": 4.4015,
+      "step": 350
+    },
+    {
+      "epoch": 0.6605504587155964,
+      "grad_norm": 0.07465502619743347,
+      "learning_rate": 0.00028678051552373485,
+      "loss": 4.4091,
+      "step": 360
+    },
+    {
+      "epoch": 0.6788990825688074,
+      "grad_norm": 0.08125840127468109,
+      "learning_rate": 0.0002597091224066581,
+      "loss": 4.3883,
+      "step": 370
+    },
+    {
+      "epoch": 0.6972477064220184,
+      "grad_norm": 0.07049981504678726,
+      "learning_rate": 0.0002335247266004629,
+      "loss": 4.4057,
+      "step": 380
+    },
+    {
+      "epoch": 0.7155963302752294,
+      "grad_norm": 0.07990957796573639,
+      "learning_rate": 0.000208323983828828,
+      "loss": 4.3883,
+      "step": 390
+    },
+    {
+      "epoch": 0.7339449541284404,
+      "grad_norm": 0.07220075279474258,
+      "learning_rate": 0.0001841999188095224,
+      "loss": 4.3872,
+      "step": 400
+    },
+    {
+      "epoch": 0.7522935779816514,
+      "grad_norm": 0.07628626376390457,
+      "learning_rate": 0.00016124158186777676,
+      "loss": 4.3884,
+      "step": 410
+    },
+    {
+      "epoch": 0.7706422018348624,
+      "grad_norm": 0.07147698104381561,
+      "learning_rate": 0.0001395337202205211,
+      "loss": 4.3797,
+      "step": 420
+    },
+    {
+      "epoch": 0.7889908256880734,
+      "grad_norm": 0.07267605513334274,
+      "learning_rate": 0.00011915646514489292,
+      "loss": 4.3868,
+      "step": 430
+    },
+    {
+      "epoch": 0.8073394495412844,
+      "grad_norm": 0.07186093926429749,
+      "learning_rate": 0.00010018503618578818,
+      "loss": 4.3903,
+      "step": 440
+    },
+    {
+      "epoch": 0.8256880733944955,
+      "grad_norm": 0.08037380129098892,
+      "learning_rate": 8.268946349432583e-05,
+      "loss": 4.3827,
+      "step": 450
+    },
+    {
+      "epoch": 0.8440366972477065,
+      "grad_norm": 0.07553426176309586,
+      "learning_rate": 6.673432932217117e-05,
+      "loss": 4.3747,
+      "step": 460
+    },
+    {
+      "epoch": 0.8623853211009175,
+      "grad_norm": 0.075074702501297,
+      "learning_rate": 5.2378529625954684e-05,
+      "loss": 4.3637,
+      "step": 470
+    },
+    {
+      "epoch": 0.8807339449541285,
+      "grad_norm": 0.07699332386255264,
+      "learning_rate": 3.9675056661785556e-05,
+      "loss": 4.374,
+      "step": 480
+    },
+    {
+      "epoch": 0.8990825688073395,
+      "grad_norm": 0.06940322369337082,
+      "learning_rate": 2.8670803372379538e-05,
+      "loss": 4.3743,
+      "step": 490
+    },
+    {
+      "epoch": 0.9174311926605505,
+      "grad_norm": 0.06667378544807434,
+      "learning_rate": 1.9406390288876586e-05,
+      "loss": 4.3865,
+      "step": 500
+    },
+    {
+      "epoch": 0.9357798165137615,
+      "grad_norm": 0.0664869025349617,
+      "learning_rate": 1.191601558631461e-05,
+      "loss": 4.3769,
+      "step": 510
+    },
+    {
+      "epoch": 0.9541284403669725,
+      "grad_norm": 0.06935941427946091,
+      "learning_rate": 6.2273288462559305e-06,
+      "loss": 4.3675,
+      "step": 520
+    },
+    {
+      "epoch": 0.9724770642201835,
+      "grad_norm": 0.06312169879674911,
+      "learning_rate": 2.361328992552314e-06,
+      "loss": 4.3687,
+      "step": 530
+    },
+    {
+      "epoch": 0.9908256880733946,
+      "grad_norm": 0.06299514323472977,
+      "learning_rate": 3.322867770044202e-07,
+      "loss": 4.3733,
+      "step": 540
+    },
+    {
+      "epoch": 1.0,
+      "step": 545,
+      "total_flos": 2.748198601925591e+17,
+      "train_loss": 4.671283760420773,
+      "train_runtime": 5093.8053,
+      "train_samples_per_second": 109.57,
+      "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": 2.748198601925591e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen3/eupe_vit_t/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/eupe_vit_t/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/eupe_vit_t/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": 192,
+  "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": "vtb_ssl:eupe",
+  "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": "/cache/ckpt/download/tokenizer/continuous/eupe/EUPE-ViT-T/EUPE-ViT-T.pt",
+  "vocab_size": 151936
+}

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

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

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

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:ff1fe61ea37db90cca5081bc658a3c66b259621a2aed1fa0d558038ad5973f1b
+size 4092904648

continuous/qwen3/eupe_vit_t/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/eupe_vit_t/finetune/tokenizer.json ADDED Viewed

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

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

continuous/qwen3/eupe_vit_t/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/eupe_vit_t/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": 17.662353515625,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 2.3233,
+      "step": 10
+    },
+    {
+      "epoch": 0.007700452401578592,
+      "grad_norm": 7.653783321380615,
+      "learning_rate": 2.435897435897436e-06,
+      "loss": 1.8937,
+      "step": 20
+    },
+    {
+      "epoch": 0.01155067860236789,
+      "grad_norm": 4.399328231811523,
+      "learning_rate": 3.7179487179487184e-06,
+      "loss": 1.5023,
+      "step": 30
+    },
+    {
+      "epoch": 0.015400904803157185,
+      "grad_norm": 1.3332048654556274,
+      "learning_rate": 5e-06,
+      "loss": 1.1859,
+      "step": 40
+    },
+    {
+      "epoch": 0.019251131003946483,
+      "grad_norm": 0.8267809152603149,
+      "learning_rate": 6.282051282051282e-06,
+      "loss": 1.0773,
+      "step": 50
+    },
+    {
+      "epoch": 0.02310135720473578,
+      "grad_norm": 0.4115471839904785,
+      "learning_rate": 7.564102564102564e-06,
+      "loss": 0.9906,
+      "step": 60
+    },
+    {
+      "epoch": 0.026951583405525073,
+      "grad_norm": 0.35859403014183044,
+      "learning_rate": 8.846153846153847e-06,
+      "loss": 0.943,
+      "step": 70
+    },
+    {
+      "epoch": 0.03080180960631437,
+      "grad_norm": 0.3009278178215027,
+      "learning_rate": 9.999996111488723e-06,
+      "loss": 0.9404,
+      "step": 80
+    },
+    {
+      "epoch": 0.034652035807103666,
+      "grad_norm": 0.2955552041530609,
+      "learning_rate": 9.999529497453782e-06,
+      "loss": 0.9071,
+      "step": 90
+    },
+    {
+      "epoch": 0.038502262007892966,
+      "grad_norm": 0.2648596167564392,
+      "learning_rate": 9.998285264324193e-06,
+      "loss": 0.8891,
+      "step": 100
+    },
+    {
+      "epoch": 0.04235248820868226,
+      "grad_norm": 0.27995195984840393,
+      "learning_rate": 9.996263605626052e-06,
+      "loss": 0.8697,
+      "step": 110
+    },
+    {
+      "epoch": 0.04620271440947156,
+      "grad_norm": 0.26050353050231934,
+      "learning_rate": 9.99346483580503e-06,
+      "loss": 0.8801,
+      "step": 120
+    },
+    {
+      "epoch": 0.05005294061026085,
+      "grad_norm": 0.25198447704315186,
+      "learning_rate": 9.989889390177467e-06,
+      "loss": 0.8764,
+      "step": 130
+    },
+    {
+      "epoch": 0.053903166811050146,
+      "grad_norm": 0.28501108288764954,
+      "learning_rate": 9.985537824862647e-06,
+      "loss": 0.8694,
+      "step": 140
+    },
+    {
+      "epoch": 0.057753393011839446,
+      "grad_norm": 0.29433903098106384,
+      "learning_rate": 9.980410816696317e-06,
+      "loss": 0.8555,
+      "step": 150
+    },
+    {
+      "epoch": 0.06160361921262874,
+      "grad_norm": 0.2684520184993744,
+      "learning_rate": 9.97450916312541e-06,
+      "loss": 0.8613,
+      "step": 160
+    },
+    {
+      "epoch": 0.06545384541341803,
+      "grad_norm": 0.2229306995868683,
+      "learning_rate": 9.967833782084004e-06,
+      "loss": 0.8612,
+      "step": 170
+    },
+    {
+      "epoch": 0.06930407161420733,
+      "grad_norm": 0.25912144780158997,
+      "learning_rate": 9.960385711850554e-06,
+      "loss": 0.8891,
+      "step": 180
+    },
+    {
+      "epoch": 0.07315429781499663,
+      "grad_norm": 0.2562401592731476,
+      "learning_rate": 9.952166110886394e-06,
+      "loss": 0.878,
+      "step": 190
+    },
+    {
+      "epoch": 0.07700452401578593,
+      "grad_norm": 0.3034636378288269,
+      "learning_rate": 9.943176257655567e-06,
+      "loss": 0.8686,
+      "step": 200
+    },
+    {
+      "epoch": 0.08085475021657522,
+      "grad_norm": 0.26456692814826965,
+      "learning_rate": 9.933417550425952e-06,
+      "loss": 0.8617,
+      "step": 210
+    },
+    {
+      "epoch": 0.08470497641736452,
+      "grad_norm": 0.2598437964916229,
+      "learning_rate": 9.9228915070518e-06,
+      "loss": 0.8507,
+      "step": 220
+    },
+    {
+      "epoch": 0.08855520261815382,
+      "grad_norm": 0.25099658966064453,
+      "learning_rate": 9.911599764737632e-06,
+      "loss": 0.8513,
+      "step": 230
+    },
+    {
+      "epoch": 0.09240542881894312,
+      "grad_norm": 0.2504135072231293,
+      "learning_rate": 9.899544079783606e-06,
+      "loss": 0.8456,
+      "step": 240
+    },
+    {
+      "epoch": 0.0962556550197324,
+      "grad_norm": 0.2926587164402008,
+      "learning_rate": 9.886726327312338e-06,
+      "loss": 0.8621,
+      "step": 250
+    },
+    {
+      "epoch": 0.1001058812205217,
+      "grad_norm": 0.2466915249824524,
+      "learning_rate": 9.87314850097724e-06,
+      "loss": 0.8598,
+      "step": 260
+    },
+    {
+      "epoch": 0.103956107421311,
+      "grad_norm": 0.2557885944843292,
+      "learning_rate": 9.858812712652452e-06,
+      "loss": 0.8466,
+      "step": 270
+    },
+    {
+      "epoch": 0.10780633362210029,
+      "grad_norm": 0.23679330945014954,
+      "learning_rate": 9.843721192104341e-06,
+      "loss": 0.847,
+      "step": 280
+    },
+    {
+      "epoch": 0.11165655982288959,
+      "grad_norm": 0.2475350797176361,
+      "learning_rate": 9.827876286644698e-06,
+      "loss": 0.8636,
+      "step": 290
+    },
+    {
+      "epoch": 0.11550678602367889,
+      "grad_norm": 0.22834450006484985,
+      "learning_rate": 9.811280460765643e-06,
+      "loss": 0.8241,
+      "step": 300
+    },
+    {
+      "epoch": 0.11935701222446819,
+      "grad_norm": 0.24299068748950958,
+      "learning_rate": 9.793936295756292e-06,
+      "loss": 0.8384,
+      "step": 310
+    },
+    {
+      "epoch": 0.12320723842525748,
+      "grad_norm": 0.2397458702325821,
+      "learning_rate": 9.775846489301279e-06,
+      "loss": 0.8383,
+      "step": 320
+    },
+    {
+      "epoch": 0.12705746462604678,
+      "grad_norm": 0.25913846492767334,
+      "learning_rate": 9.757013855061157e-06,
+      "loss": 0.8379,
+      "step": 330
+    },
+    {
+      "epoch": 0.13090769082683607,
+      "grad_norm": 0.2448691725730896,
+      "learning_rate": 9.737441322234758e-06,
+      "loss": 0.8498,
+      "step": 340
+    },
+    {
+      "epoch": 0.13475791702762538,
+      "grad_norm": 0.25775498151779175,
+      "learning_rate": 9.717131935103605e-06,
+      "loss": 0.8488,
+      "step": 350
+    },
+    {
+      "epoch": 0.13860814322841467,
+      "grad_norm": 0.24115166068077087,
+      "learning_rate": 9.696088852558398e-06,
+      "loss": 0.8358,
+      "step": 360
+    },
+    {
+      "epoch": 0.14245836942920398,
+      "grad_norm": 0.256876677274704,
+      "learning_rate": 9.674315347607686e-06,
+      "loss": 0.8618,
+      "step": 370
+    },
+    {
+      "epoch": 0.14630859562999327,
+      "grad_norm": 0.2374137043952942,
+      "learning_rate": 9.651814806868796e-06,
+      "loss": 0.8223,
+      "step": 380
+    },
+    {
+      "epoch": 0.15015882183078255,
+      "grad_norm": 0.23744648694992065,
+      "learning_rate": 9.62859073004107e-06,
+      "loss": 0.8311,
+      "step": 390
+    },
+    {
+      "epoch": 0.15400904803157187,
+      "grad_norm": 0.23627208173274994,
+      "learning_rate": 9.604646729361547e-06,
+      "loss": 0.8329,
+      "step": 400
+    },
+    {
+      "epoch": 0.15785927423236115,
+      "grad_norm": 0.2403443306684494,
+      "learning_rate": 9.579986529043099e-06,
+      "loss": 0.8222,
+      "step": 410
+    },
+    {
+      "epoch": 0.16170950043315044,
+      "grad_norm": 0.24101820588111877,
+      "learning_rate": 9.554613964695189e-06,
+      "loss": 0.8384,
+      "step": 420
+    },
+    {
+      "epoch": 0.16555972663393975,
+      "grad_norm": 0.247356578707695,
+      "learning_rate": 9.528532982727281e-06,
+      "loss": 0.8276,
+      "step": 430
+    },
+    {
+      "epoch": 0.16940995283472904,
+      "grad_norm": 0.2468992918729782,
+      "learning_rate": 9.501747639735018e-06,
+      "loss": 0.8261,
+      "step": 440
+    },
+    {
+      "epoch": 0.17326017903551832,
+      "grad_norm": 0.24569033086299896,
+      "learning_rate": 9.474262101869272e-06,
+      "loss": 0.8152,
+      "step": 450
+    },
+    {
+      "epoch": 0.17711040523630764,
+      "grad_norm": 0.24328067898750305,
+      "learning_rate": 9.446080644188141e-06,
+      "loss": 0.8207,
+      "step": 460
+    },
+    {
+      "epoch": 0.18096063143709692,
+      "grad_norm": 0.24548572301864624,
+      "learning_rate": 9.417207649992017e-06,
+      "loss": 0.829,
+      "step": 470
+    },
+    {
+      "epoch": 0.18481085763788624,
+      "grad_norm": 0.23838219046592712,
+      "learning_rate": 9.387647610141816e-06,
+      "loss": 0.8277,
+      "step": 480
+    },
+    {
+      "epoch": 0.18866108383867553,
+      "grad_norm": 0.23057948052883148,
+      "learning_rate": 9.35740512236047e-06,
+      "loss": 0.8145,
+      "step": 490
+    },
+    {
+      "epoch": 0.1925113100394648,
+      "grad_norm": 0.22752974927425385,
+      "learning_rate": 9.326484890517809e-06,
+      "loss": 0.8181,
+      "step": 500
+    },
+    {
+      "epoch": 0.19636153624025413,
+      "grad_norm": 0.2306179255247116,
+      "learning_rate": 9.294891723898926e-06,
+      "loss": 0.815,
+      "step": 510
+    },
+    {
+      "epoch": 0.2002117624410434,
+      "grad_norm": 0.24033881723880768,
+      "learning_rate": 9.262630536456153e-06,
+      "loss": 0.8443,
+      "step": 520
+    },
+    {
+      "epoch": 0.2040619886418327,
+      "grad_norm": 0.2578759491443634,
+      "learning_rate": 9.229706346044749e-06,
+      "loss": 0.8162,
+      "step": 530
+    },
+    {
+      "epoch": 0.207912214842622,
+      "grad_norm": 0.2539111375808716,
+      "learning_rate": 9.196124273642422e-06,
+      "loss": 0.8297,
+      "step": 540
+    },
+    {
+      "epoch": 0.2117624410434113,
+      "grad_norm": 0.22644346952438354,
+      "learning_rate": 9.161889542552842e-06,
+      "loss": 0.8311,
+      "step": 550
+    },
+    {
+      "epoch": 0.21561266724420058,
+      "grad_norm": 0.21902616322040558,
+      "learning_rate": 9.127007477593193e-06,
+      "loss": 0.8363,
+      "step": 560
+    },
+    {
+      "epoch": 0.2194628934449899,
+      "grad_norm": 0.2602729797363281,
+      "learning_rate": 9.091483504265974e-06,
+      "loss": 0.8457,
+      "step": 570
+    },
+    {
+      "epoch": 0.22331311964577918,
+      "grad_norm": 0.2533755600452423,
+      "learning_rate": 9.055323147915116e-06,
+      "loss": 0.8247,
+      "step": 580
+    },
+    {
+      "epoch": 0.2271633458465685,
+      "grad_norm": 0.2536756098270416,
+      "learning_rate": 9.018532032866585e-06,
+      "loss": 0.8354,
+      "step": 590
+    },
+    {
+      "epoch": 0.23101357204735778,
+      "grad_norm": 0.2278566211462021,
+      "learning_rate": 8.981115881553579e-06,
+      "loss": 0.8257,
+      "step": 600
+    },
+    {
+      "epoch": 0.23486379824814707,
+      "grad_norm": 0.2596656084060669,
+      "learning_rate": 8.943080513626476e-06,
+      "loss": 0.8469,
+      "step": 610
+    },
+    {
+      "epoch": 0.23871402444893638,
+      "grad_norm": 0.2356439083814621,
+      "learning_rate": 8.904431845047641e-06,
+      "loss": 0.8302,
+      "step": 620
+    },
+    {
+      "epoch": 0.24256425064972567,
+      "grad_norm": 0.2529607117176056,
+      "learning_rate": 8.86517588717128e-06,
+      "loss": 0.847,
+      "step": 630
+    },
+    {
+      "epoch": 0.24641447685051496,
+      "grad_norm": 0.23651540279388428,
+      "learning_rate": 8.82531874580844e-06,
+      "loss": 0.8352,
+      "step": 640
+    },
+    {
+      "epoch": 0.25026470305130427,
+      "grad_norm": 0.223335400223732,
+      "learning_rate": 8.784866620277314e-06,
+      "loss": 0.7923,
+      "step": 650
+    },
+    {
+      "epoch": 0.25411492925209356,
+      "grad_norm": 0.27132871747016907,
+      "learning_rate": 8.743825802439018e-06,
+      "loss": 0.8306,
+      "step": 660
+    },
+    {
+      "epoch": 0.25796515545288284,
+      "grad_norm": 0.2569819688796997,
+      "learning_rate": 8.702202675718956e-06,
+      "loss": 0.835,
+      "step": 670
+    },
+    {
+      "epoch": 0.26181538165367213,
+      "grad_norm": 0.22892408072948456,
+      "learning_rate": 8.660003714113959e-06,
+      "loss": 0.8054,
+      "step": 680
+    },
+    {
+      "epoch": 0.26566560785446147,
+      "grad_norm": 0.22995132207870483,
+      "learning_rate": 8.617235481185323e-06,
+      "loss": 0.8304,
+      "step": 690
+    },
+    {
+      "epoch": 0.26951583405525076,
+      "grad_norm": 0.2433670312166214,
+      "learning_rate": 8.573904629037931e-06,
+      "loss": 0.8347,
+      "step": 700
+    },
+    {
+      "epoch": 0.27336606025604004,
+      "grad_norm": 0.23720984160900116,
+      "learning_rate": 8.530017897285585e-06,
+      "loss": 0.8144,
+      "step": 710
+    },
+    {
+      "epoch": 0.27721628645682933,
+      "grad_norm": 0.2173091173171997,
+      "learning_rate": 8.485582112002746e-06,
+      "loss": 0.8188,
+      "step": 720
+    },
+    {
+      "epoch": 0.2810665126576186,
+      "grad_norm": 0.23379366099834442,
+      "learning_rate": 8.440604184662811e-06,
+      "loss": 0.8255,
+      "step": 730
+    },
+    {
+      "epoch": 0.28491673885840796,
+      "grad_norm": 0.3439956605434418,
+      "learning_rate": 8.395091111063118e-06,
+      "loss": 0.8138,
+      "step": 740
+    },
+    {
+      "epoch": 0.28876696505919724,
+      "grad_norm": 0.2201453298330307,
+      "learning_rate": 8.349049970236822e-06,
+      "loss": 0.8125,
+      "step": 750
+    },
+    {
+      "epoch": 0.29261719125998653,
+      "grad_norm": 0.23767344653606415,
+      "learning_rate": 8.302487923351842e-06,
+      "loss": 0.8059,
+      "step": 760
+    },
+    {
+      "epoch": 0.2964674174607758,
+      "grad_norm": 0.23543451726436615,
+      "learning_rate": 8.255412212597021e-06,
+      "loss": 0.8234,
+      "step": 770
+    },
+    {
+      "epoch": 0.3003176436615651,
+      "grad_norm": 0.23025886714458466,
+      "learning_rate": 8.207830160055688e-06,
+      "loss": 0.8202,
+      "step": 780
+    },
+    {
+      "epoch": 0.3041678698623544,
+      "grad_norm": 0.24769090116024017,
+      "learning_rate": 8.159749166566786e-06,
+      "loss": 0.8053,
+      "step": 790
+    },
+    {
+      "epoch": 0.30801809606314373,
+      "grad_norm": 0.23131048679351807,
+      "learning_rate": 8.11117671057377e-06,
+      "loss": 0.8331,
+      "step": 800
+    },
+    {
+      "epoch": 0.311868322263933,
+      "grad_norm": 0.236127108335495,
+      "learning_rate": 8.062120346961404e-06,
+      "loss": 0.8252,
+      "step": 810
+    },
+    {
+      "epoch": 0.3157185484647223,
+      "grad_norm": 0.23204800486564636,
+      "learning_rate": 8.012587705880705e-06,
+      "loss": 0.817,
+      "step": 820
+    },
+    {
+      "epoch": 0.3195687746655116,
+      "grad_norm": 0.23862417042255402,
+      "learning_rate": 7.962586491562146e-06,
+      "loss": 0.8043,
+      "step": 830
+    },
+    {
+      "epoch": 0.3234190008663009,
+      "grad_norm": 0.25034165382385254,
+      "learning_rate": 7.912124481117357e-06,
+      "loss": 0.8102,
+      "step": 840
+    },
+    {
+      "epoch": 0.3272692270670902,
+      "grad_norm": 0.23150968551635742,
+      "learning_rate": 7.861209523329494e-06,
+      "loss": 0.82,
+      "step": 850
+    },
+    {
+      "epoch": 0.3311194532678795,
+      "grad_norm": 0.23157379031181335,
+      "learning_rate": 7.809849537432432e-06,
+      "loss": 0.8042,
+      "step": 860
+    },
+    {
+      "epoch": 0.3349696794686688,
+      "grad_norm": 0.23144736886024475,
+      "learning_rate": 7.758052511879035e-06,
+      "loss": 0.8145,
+      "step": 870
+    },
+    {
+      "epoch": 0.3388199056694581,
+      "grad_norm": 0.21671704947948456,
+      "learning_rate": 7.705826503098642e-06,
+      "loss": 0.8081,
+      "step": 880
+    },
+    {
+      "epoch": 0.34267013187024736,
+      "grad_norm": 0.227386474609375,
+      "learning_rate": 7.653179634243979e-06,
+      "loss": 0.8085,
+      "step": 890
+    },
+    {
+      "epoch": 0.34652035807103665,
+      "grad_norm": 0.22326436638832092,
+      "learning_rate": 7.600120093927696e-06,
+      "loss": 0.8151,
+      "step": 900
+    },
+    {
+      "epoch": 0.350370584271826,
+      "grad_norm": 0.240731343626976,
+      "learning_rate": 7.546656134948729e-06,
+      "loss": 0.8053,
+      "step": 910
+    },
+    {
+      "epoch": 0.3542208104726153,
+      "grad_norm": 0.2353595495223999,
+      "learning_rate": 7.492796073008661e-06,
+      "loss": 0.8023,
+      "step": 920
+    },
+    {
+      "epoch": 0.35807103667340456,
+      "grad_norm": 0.25311657786369324,
+      "learning_rate": 7.438548285418326e-06,
+      "loss": 0.7885,
+      "step": 930
+    },
+    {
+      "epoch": 0.36192126287419385,
+      "grad_norm": 0.2513812482357025,
+      "learning_rate": 7.3839212097948e-06,
+      "loss": 0.8229,
+      "step": 940
+    },
+    {
+      "epoch": 0.36577148907498314,
+      "grad_norm": 0.2339373677968979,
+      "learning_rate": 7.328923342749039e-06,
+      "loss": 0.8374,
+      "step": 950
+    },
+    {
+      "epoch": 0.3696217152757725,
+      "grad_norm": 0.251939058303833,
+      "learning_rate": 7.273563238564319e-06,
+      "loss": 0.8214,
+      "step": 960
+    },
+    {
+      "epoch": 0.37347194147656176,
+      "grad_norm": 0.23422987759113312,
+      "learning_rate": 7.217849507865724e-06,
+      "loss": 0.8033,
+      "step": 970
+    },
+    {
+      "epoch": 0.37732216767735105,
+      "grad_norm": 0.23278078436851501,
+      "learning_rate": 7.161790816280858e-06,
+      "loss": 0.8069,
+      "step": 980
+    },
+    {
+      "epoch": 0.38117239387814034,
+      "grad_norm": 0.23938512802124023,
+      "learning_rate": 7.1053958830920125e-06,
+      "loss": 0.7954,
+      "step": 990
+    },
+    {
+      "epoch": 0.3850226200789296,
+      "grad_norm": 0.23269489407539368,
+      "learning_rate": 7.0486734798799715e-06,
+      "loss": 0.8101,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3888728462797189,
+      "grad_norm": 0.2308688759803772,
+      "learning_rate": 6.991632429159708e-06,
+      "loss": 0.8334,
+      "step": 1010
+    },
+    {
+      "epoch": 0.39272307248050825,
+      "grad_norm": 0.25654757022857666,
+      "learning_rate": 6.934281603008132e-06,
+      "loss": 0.8253,
+      "step": 1020
+    },
+    {
+      "epoch": 0.39657329868129754,
+      "grad_norm": 0.23077958822250366,
+      "learning_rate": 6.876629921684143e-06,
+      "loss": 0.8231,
+      "step": 1030
+    },
+    {
+      "epoch": 0.4004235248820868,
+      "grad_norm": 0.24254585802555084,
+      "learning_rate": 6.818686352241199e-06,
+      "loss": 0.8132,
+      "step": 1040
+    },
+    {
+      "epoch": 0.4042737510828761,
+      "grad_norm": 0.21668875217437744,
+      "learning_rate": 6.760459907132573e-06,
+      "loss": 0.8097,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4081239772836654,
+      "grad_norm": 0.22675132751464844,
+      "learning_rate": 6.701959642809589e-06,
+      "loss": 0.8165,
+      "step": 1060
+    },
+    {
+      "epoch": 0.41197420348445474,
+      "grad_norm": 0.2363034188747406,
+      "learning_rate": 6.643194658312987e-06,
+      "loss": 0.8082,
+      "step": 1070
+    },
+    {
+      "epoch": 0.415824429685244,
+      "grad_norm": 0.2258535623550415,
+      "learning_rate": 6.584174093857676e-06,
+      "loss": 0.8327,
+      "step": 1080
+    },
+    {
+      "epoch": 0.4196746558860333,
+      "grad_norm": 0.25662073493003845,
+      "learning_rate": 6.5249071294110735e-06,
+      "loss": 0.8121,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4235248820868226,
+      "grad_norm": 0.25288233160972595,
+      "learning_rate": 6.465402983265274e-06,
+      "loss": 0.8266,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4273751082876119,
+      "grad_norm": 0.2390207052230835,
+      "learning_rate": 6.405670910603243e-06,
+      "loss": 0.8231,
+      "step": 1110
+    },
+    {
+      "epoch": 0.43122533448840117,
+      "grad_norm": 0.25337085127830505,
+      "learning_rate": 6.345720202059291e-06,
+      "loss": 0.819,
+      "step": 1120
+    },
+    {
+      "epoch": 0.4350755606891905,
+      "grad_norm": 0.24532322585582733,
+      "learning_rate": 6.285560182274008e-06,
+      "loss": 0.7981,
+      "step": 1130
+    },
+    {
+      "epoch": 0.4389257868899798,
+      "grad_norm": 0.23423783481121063,
+      "learning_rate": 6.225200208443936e-06,
+      "loss": 0.8196,
+      "step": 1140
+    },
+    {
+      "epoch": 0.4427760130907691,
+      "grad_norm": 0.21852706372737885,
+      "learning_rate": 6.16464966886616e-06,
+      "loss": 0.8102,
+      "step": 1150
+    },
+    {
+      "epoch": 0.44662623929155837,
+      "grad_norm": 0.2554682791233063,
+      "learning_rate": 6.103917981478065e-06,
+      "loss": 0.8327,
+      "step": 1160
+    },
+    {
+      "epoch": 0.45047646549234766,
+      "grad_norm": 0.238169863820076,
+      "learning_rate": 6.043014592392489e-06,
+      "loss": 0.8448,
+      "step": 1170
+    },
+    {
+      "epoch": 0.454326691693137,
+      "grad_norm": 0.2269848883152008,
+      "learning_rate": 5.981948974428485e-06,
+      "loss": 0.8185,
+      "step": 1180
+    },
+    {
+      "epoch": 0.4581769178939263,
+      "grad_norm": 0.23975035548210144,
+      "learning_rate": 5.920730625637934e-06,
+      "loss": 0.799,
+      "step": 1190
+    },
+    {
+      "epoch": 0.46202714409471557,
+      "grad_norm": 0.24146078526973724,
+      "learning_rate": 5.8593690678282355e-06,
+      "loss": 0.8081,
+      "step": 1200
+    },
+    {
+      "epoch": 0.46587737029550486,
+      "grad_norm": 0.2489355355501175,
+      "learning_rate": 5.797873845081301e-06,
+      "loss": 0.812,
+      "step": 1210
+    },
+    {
+      "epoch": 0.46972759649629414,
+      "grad_norm": 0.24009259045124054,
+      "learning_rate": 5.736254522269075e-06,
+      "loss": 0.8295,
+      "step": 1220
+    },
+    {
+      "epoch": 0.47357782269708343,
+      "grad_norm": 0.2381322830915451,
+      "learning_rate": 5.6745206835658386e-06,
+      "loss": 0.8301,
+      "step": 1230
+    },
+    {
+      "epoch": 0.47742804889787277,
+      "grad_norm": 0.22687029838562012,
+      "learning_rate": 5.612681930957495e-06,
+      "loss": 0.8016,
+      "step": 1240
+    },
+    {
+      "epoch": 0.48127827509866206,
+      "grad_norm": 0.22578254342079163,
+      "learning_rate": 5.550747882748095e-06,
+      "loss": 0.8005,
+      "step": 1250
+    },
+    {
+      "epoch": 0.48512850129945134,
+      "grad_norm": 0.25458380579948425,
+      "learning_rate": 5.4887281720638195e-06,
+      "loss": 0.8049,
+      "step": 1260
+    },
+    {
+      "epoch": 0.48897872750024063,
+      "grad_norm": 0.26724380254745483,
+      "learning_rate": 5.42663244535466e-06,
+      "loss": 0.7942,
+      "step": 1270
+    },
+    {
+      "epoch": 0.4928289537010299,
+      "grad_norm": 0.25580331683158875,
+      "learning_rate": 5.3644703608940205e-06,
+      "loss": 0.7998,
+      "step": 1280
+    },
+    {
+      "epoch": 0.49667917990181926,
+      "grad_norm": 0.24235638976097107,
+      "learning_rate": 5.302251587276491e-06,
+      "loss": 0.8116,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5005294061026085,
+      "grad_norm": 0.21909165382385254,
+      "learning_rate": 5.2399858019140005e-06,
+      "loss": 0.8116,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5043796323033978,
+      "grad_norm": 0.23198756575584412,
+      "learning_rate": 5.177682689530615e-06,
+      "loss": 0.8261,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5082298585041871,
+      "grad_norm": 0.24189314246177673,
+      "learning_rate": 5.11535194065619e-06,
+      "loss": 0.7939,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5120800847049765,
+      "grad_norm": 0.257007360458374,
+      "learning_rate": 5.053003250119113e-06,
+      "loss": 0.8055,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5159303109057657,
+      "grad_norm": 0.23789845407009125,
+      "learning_rate": 4.9906463155383925e-06,
+      "loss": 0.8116,
+      "step": 1340
+    },
+    {
+      "epoch": 0.519780537106555,
+      "grad_norm": 0.2336827665567398,
+      "learning_rate": 4.9282908358153046e-06,
+      "loss": 0.8018,
+      "step": 1350
+    },
+    {
+      "epoch": 0.5236307633073443,
+      "grad_norm": 0.25947898626327515,
+      "learning_rate": 4.8659465096248345e-06,
+      "loss": 0.8217,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5274809895081336,
+      "grad_norm": 0.2617848515510559,
+      "learning_rate": 4.803623033907169e-06,
+      "loss": 0.8219,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5313312157089229,
+      "grad_norm": 0.23110990226268768,
+      "learning_rate": 4.741330102359441e-06,
+      "loss": 0.8253,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5351814419097122,
+      "grad_norm": 0.2514432370662689,
+      "learning_rate": 4.679077403927996e-06,
+      "loss": 0.8061,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5390316681105015,
+      "grad_norm": 0.23255065083503723,
+      "learning_rate": 4.616874621301373e-06,
+      "loss": 0.8041,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5428818943112907,
+      "grad_norm": 0.2419893592596054,
+      "learning_rate": 4.554731429404293e-06,
+      "loss": 0.8204,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5467321205120801,
+      "grad_norm": 0.2643243670463562,
+      "learning_rate": 4.492657493892822e-06,
+      "loss": 0.8065,
+      "step": 1420
+    },
+    {
+      "epoch": 0.5505823467128694,
+      "grad_norm": 0.2917819321155548,
+      "learning_rate": 4.430662469650999e-06,
+      "loss": 0.8208,
+      "step": 1430
+    },
+    {
+      "epoch": 0.5544325729136587,
+      "grad_norm": 0.2594841420650482,
+      "learning_rate": 4.368755999289118e-06,
+      "loss": 0.8217,
+      "step": 1440
+    },
+    {
+      "epoch": 0.558282799114448,
+      "grad_norm": 0.2630656361579895,
+      "learning_rate": 4.306947711643949e-06,
+      "loss": 0.8092,
+      "step": 1450
+    },
+    {
+      "epoch": 0.5621330253152372,
+      "grad_norm": 0.24098335206508636,
+      "learning_rate": 4.245247220281068e-06,
+      "loss": 0.8103,
+      "step": 1460
+    },
+    {
+      "epoch": 0.5659832515160266,
+      "grad_norm": 0.24143166840076447,
+      "learning_rate": 4.183664121999587e-06,
+      "loss": 0.8096,
+      "step": 1470
+    },
+    {
+      "epoch": 0.5698334777168159,
+      "grad_norm": 0.250322550535202,
+      "learning_rate": 4.122207995339474e-06,
+      "loss": 0.8044,
+      "step": 1480
+    },
+    {
+      "epoch": 0.5736837039176051,
+      "grad_norm": 0.23447555303573608,
+      "learning_rate": 4.060888399091734e-06,
+      "loss": 0.8106,
+      "step": 1490
+    },
+    {
+      "epoch": 0.5775339301183945,
+      "grad_norm": 0.23810453712940216,
+      "learning_rate": 3.999714870811639e-06,
+      "loss": 0.8169,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5813841563191837,
+      "grad_norm": 0.22955146431922913,
+      "learning_rate": 3.938696925335275e-06,
+      "loss": 0.8116,
+      "step": 1510
+    },
+    {
+      "epoch": 0.5852343825199731,
+      "grad_norm": 0.23246856033802032,
+      "learning_rate": 3.87784405329962e-06,
+      "loss": 0.8196,
+      "step": 1520
+    },
+    {
+      "epoch": 0.5890846087207623,
+      "grad_norm": 0.23737019300460815,
+      "learning_rate": 3.817165719666382e-06,
+      "loss": 0.8228,
+      "step": 1530
+    },
+    {
+      "epoch": 0.5929348349215516,
+      "grad_norm": 0.22887438535690308,
+      "learning_rate": 3.756671362249841e-06,
+      "loss": 0.825,
+      "step": 1540
+    },
+    {
+      "epoch": 0.596785061122341,
+      "grad_norm": 0.23459482192993164,
+      "learning_rate": 3.6963703902488907e-06,
+      "loss": 0.812,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6006352873231302,
+      "grad_norm": 0.23749199509620667,
+      "learning_rate": 3.6362721827835655e-06,
+      "loss": 0.8318,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6044855135239195,
+      "grad_norm": 0.25375521183013916,
+      "learning_rate": 3.576386087436212e-06,
+      "loss": 0.8244,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6083357397247088,
+      "grad_norm": 0.24759694933891296,
+      "learning_rate": 3.5167214187975913e-06,
+      "loss": 0.8141,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6121859659254981,
+      "grad_norm": 0.20647309720516205,
+      "learning_rate": 3.457287457018086e-06,
+      "loss": 0.8163,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6160361921262875,
+      "grad_norm": 0.23355813324451447,
+      "learning_rate": 3.3980934463643003e-06,
+      "loss": 0.8036,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6198864183270767,
+      "grad_norm": 0.2177426964044571,
+      "learning_rate": 3.3391485937812053e-06,
+      "loss": 0.7869,
+      "step": 1610
+    },
+    {
+      "epoch": 0.623736644527866,
+      "grad_norm": 0.23785145580768585,
+      "learning_rate": 3.280462067460115e-06,
+      "loss": 0.786,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6275868707286553,
+      "grad_norm": 0.23873941600322723,
+      "learning_rate": 3.222042995412669e-06,
+      "loss": 0.8166,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6314370969294446,
+      "grad_norm": 0.21843862533569336,
+      "learning_rate": 3.163900464051089e-06,
+      "loss": 0.8057,
+      "step": 1640
+    },
+    {
+      "epoch": 0.635287323130234,
+      "grad_norm": 0.2298617959022522,
+      "learning_rate": 3.1060435167748826e-06,
+      "loss": 0.8144,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6391375493310232,
+      "grad_norm": 0.23204311728477478,
+      "learning_rate": 3.048481152564253e-06,
+      "loss": 0.7948,
+      "step": 1660
+    },
+    {
+      "epoch": 0.6429877755318125,
+      "grad_norm": 0.24425549805164337,
+      "learning_rate": 2.9912223245804017e-06,
+      "loss": 0.8162,
+      "step": 1670
+    },
+    {
+      "epoch": 0.6468380017326018,
+      "grad_norm": 0.23694977164268494,
+      "learning_rate": 2.934275938772981e-06,
+      "loss": 0.8157,
+      "step": 1680
+    },
+    {
+      "epoch": 0.6506882279333911,
+      "grad_norm": 0.2520178556442261,
+      "learning_rate": 2.8776508524948676e-06,
+      "loss": 0.8039,
+      "step": 1690
+    },
+    {
+      "epoch": 0.6545384541341804,
+      "grad_norm": 0.21580561995506287,
+      "learning_rate": 2.821355873124503e-06,
+      "loss": 0.8032,
+      "step": 1700
+    },
+    {
+      "epoch": 0.6583886803349697,
+      "grad_norm": 0.24744398891925812,
+      "learning_rate": 2.7653997566960144e-06,
+      "loss": 0.8167,
+      "step": 1710
+    },
+    {
+      "epoch": 0.662238906535759,
+      "grad_norm": 0.2517266571521759,
+      "learning_rate": 2.709791206537309e-06,
+      "loss": 0.8065,
+      "step": 1720
+    },
+    {
+      "epoch": 0.6660891327365482,
+      "grad_norm": 0.24214737117290497,
+      "learning_rate": 2.6545388719163747e-06,
+      "loss": 0.8147,
+      "step": 1730
+    },
+    {
+      "epoch": 0.6699393589373376,
+      "grad_norm": 0.24518561363220215,
+      "learning_rate": 2.599651346695979e-06,
+      "loss": 0.8114,
+      "step": 1740
+    },
+    {
+      "epoch": 0.6737895851381268,
+      "grad_norm": 0.24320606887340546,
+      "learning_rate": 2.545137167997007e-06,
+      "loss": 0.8069,
+      "step": 1750
+    },
+    {
+      "epoch": 0.6776398113389162,
+      "grad_norm": 0.2278076410293579,
+      "learning_rate": 2.4910048148705972e-06,
+      "loss": 0.7959,
+      "step": 1760
+    },
+    {
+      "epoch": 0.6814900375397055,
+      "grad_norm": 0.2507089376449585,
+      "learning_rate": 2.4372627069793327e-06,
+      "loss": 0.8378,
+      "step": 1770
+    },
+    {
+      "epoch": 0.6853402637404947,
+      "grad_norm": 0.2347089648246765,
+      "learning_rate": 2.3839192032876486e-06,
+      "loss": 0.8251,
+      "step": 1780
+    },
+    {
+      "epoch": 0.6891904899412841,
+      "grad_norm": 0.22757399082183838,
+      "learning_rate": 2.3309826007617047e-06,
+      "loss": 0.8353,
+      "step": 1790
+    },
+    {
+      "epoch": 0.6930407161420733,
+      "grad_norm": 0.24116858839988708,
+      "learning_rate": 2.278461133078881e-06,
+      "loss": 0.7994,
+      "step": 1800
+    },
+    {
+      "epoch": 0.6968909423428626,
+      "grad_norm": 0.2388163059949875,
+      "learning_rate": 2.2263629693471258e-06,
+      "loss": 0.7906,
+      "step": 1810
+    },
+    {
+      "epoch": 0.700741168543652,
+      "grad_norm": 0.22614409029483795,
+      "learning_rate": 2.1746962128343414e-06,
+      "loss": 0.8161,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7045913947444412,
+      "grad_norm": 0.2273930311203003,
+      "learning_rate": 2.1234688997080183e-06,
+      "loss": 0.8038,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7084416209452306,
+      "grad_norm": 0.24915441870689392,
+      "learning_rate": 2.0726889977853053e-06,
+      "loss": 0.8076,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7122918471460198,
+      "grad_norm": 0.25916165113449097,
+      "learning_rate": 2.022364405293703e-06,
+      "loss": 0.8202,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7161420733468091,
+      "grad_norm": 0.2289314717054367,
+      "learning_rate": 1.9725029496425814e-06,
+      "loss": 0.8128,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7199922995475985,
+      "grad_norm": 0.24255646765232086,
+      "learning_rate": 1.923112386205727e-06,
+      "loss": 0.8103,
+      "step": 1870
+    },
+    {
+      "epoch": 0.7238425257483877,
+      "grad_norm": 0.22471974790096283,
+      "learning_rate": 1.874200397115079e-06,
+      "loss": 0.8164,
+      "step": 1880
+    },
+    {
+      "epoch": 0.727692751949177,
+      "grad_norm": 0.2440529763698578,
+      "learning_rate": 1.825774590065857e-06,
+      "loss": 0.8016,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7315429781499663,
+      "grad_norm": 0.23856337368488312,
+      "learning_rate": 1.7778424971332813e-06,
+      "loss": 0.8081,
+      "step": 1900
+    },
+    {
+      "epoch": 0.7353932043507556,
+      "grad_norm": 0.23438166081905365,
+      "learning_rate": 1.7304115736010407e-06,
+      "loss": 0.8072,
+      "step": 1910
+    },
+    {
+      "epoch": 0.739243430551545,
+      "grad_norm": 0.23538047075271606,
+      "learning_rate": 1.6834891968017108e-06,
+      "loss": 0.7991,
+      "step": 1920
+    },
+    {
+      "epoch": 0.7430936567523342,
+      "grad_norm": 0.24501019716262817,
+      "learning_rate": 1.6370826649692884e-06,
+      "loss": 0.8208,
+      "step": 1930
+    },
+    {
+      "epoch": 0.7469438829531235,
+      "grad_norm": 0.23601073026657104,
+      "learning_rate": 1.5911991961040463e-06,
+      "loss": 0.8264,
+      "step": 1940
+    },
+    {
+      "epoch": 0.7507941091539128,
+      "grad_norm": 0.24460498988628387,
+      "learning_rate": 1.545845926849851e-06,
+      "loss": 0.8119,
+      "step": 1950
+    },
+    {
+      "epoch": 0.7546443353547021,
+      "grad_norm": 0.2547614872455597,
+      "learning_rate": 1.5010299113841397e-06,
+      "loss": 0.8312,
+      "step": 1960
+    },
+    {
+      "epoch": 0.7584945615554913,
+      "grad_norm": 0.23811982572078705,
+      "learning_rate": 1.4567581203207238e-06,
+      "loss": 0.8185,
+      "step": 1970
+    },
+    {
+      "epoch": 0.7623447877562807,
+      "grad_norm": 0.2491694539785385,
+      "learning_rate": 1.413037439625593e-06,
+      "loss": 0.8189,
+      "step": 1980
+    },
+    {
+      "epoch": 0.76619501395707,
+      "grad_norm": 0.26107364892959595,
+      "learning_rate": 1.3698746695458847e-06,
+      "loss": 0.8112,
+      "step": 1990
+    },
+    {
+      "epoch": 0.7700452401578592,
+      "grad_norm": 0.23213285207748413,
+      "learning_rate": 1.3272765235521745e-06,
+      "loss": 0.8033,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7738954663586486,
+      "grad_norm": 0.22581815719604492,
+      "learning_rate": 1.2852496272942877e-06,
+      "loss": 0.8103,
+      "step": 2010
+    },
+    {
+      "epoch": 0.7777456925594378,
+      "grad_norm": 0.22374200820922852,
+      "learning_rate": 1.2438005175707396e-06,
+      "loss": 0.7996,
+      "step": 2020
+    },
+    {
+      "epoch": 0.7815959187602272,
+      "grad_norm": 0.2288970649242401,
+      "learning_rate": 1.2029356413120297e-06,
+      "loss": 0.786,
+      "step": 2030
+    },
+    {
+      "epoch": 0.7854461449610165,
+      "grad_norm": 0.23176926374435425,
+      "learning_rate": 1.1626613545778786e-06,
+      "loss": 0.8028,
+      "step": 2040
+    },
+    {
+      "epoch": 0.7892963711618057,
+      "grad_norm": 0.22512046992778778,
+      "learning_rate": 1.1229839215686294e-06,
+      "loss": 0.8069,
+      "step": 2050
+    },
+    {
+      "epoch": 0.7931465973625951,
+      "grad_norm": 0.2156192660331726,
+      "learning_rate": 1.0839095136509188e-06,
+      "loss": 0.7931,
+      "step": 2060
+    },
+    {
+      "epoch": 0.7969968235633843,
+      "grad_norm": 0.25544652342796326,
+      "learning_rate": 1.045444208397791e-06,
+      "loss": 0.8156,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8008470497641736,
+      "grad_norm": 0.22855247557163239,
+      "learning_rate": 1.007593988643402e-06,
+      "loss": 0.7939,
+      "step": 2080
+    },
+    {
+      "epoch": 0.804697275964963,
+      "grad_norm": 0.2302682250738144,
+      "learning_rate": 9.703647415524613e-07,
+      "loss": 0.8159,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8085475021657522,
+      "grad_norm": 0.2358298897743225,
+      "learning_rate": 9.337622577045485e-07,
+      "loss": 0.8058,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8123977283665416,
+      "grad_norm": 0.23154498636722565,
+      "learning_rate": 8.977922301934588e-07,
+      "loss": 0.8126,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8162479545673308,
+      "grad_norm": 0.2347831130027771,
+      "learning_rate": 8.624602537417009e-07,
+      "loss": 0.8148,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8200981807681201,
+      "grad_norm": 0.2523323595523834,
+      "learning_rate": 8.277718238303123e-07,
+      "loss": 0.8044,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8239484069689095,
+      "grad_norm": 0.23747844994068146,
+      "learning_rate": 7.937323358440935e-07,
+      "loss": 0.8071,
+      "step": 2140
+    },
+    {
+      "epoch": 0.8277986331696987,
+      "grad_norm": 0.23891286551952362,
+      "learning_rate": 7.603470842324223e-07,
+      "loss": 0.8002,
+      "step": 2150
+    },
+    {
+      "epoch": 0.831648859370488,
+      "grad_norm": 0.22864075005054474,
+      "learning_rate": 7.276212616857575e-07,
+      "loss": 0.816,
+      "step": 2160
+    },
+    {
+      "epoch": 0.8354990855712773,
+      "grad_norm": 0.22507299482822418,
+      "learning_rate": 6.955599583279837e-07,
+      "loss": 0.8185,
+      "step": 2170
+    },
+    {
+      "epoch": 0.8393493117720666,
+      "grad_norm": 0.23963169753551483,
+      "learning_rate": 6.641681609246981e-07,
+      "loss": 0.8071,
+      "step": 2180
+    },
+    {
+      "epoch": 0.8431995379728559,
+      "grad_norm": 0.22956469655036926,
+      "learning_rate": 6.334507521075711e-07,
+      "loss": 0.8181,
+      "step": 2190
+    },
+    {
+      "epoch": 0.8470497641736452,
+      "grad_norm": 0.25327354669570923,
+      "learning_rate": 6.034125096149179e-07,
+      "loss": 0.8082,
+      "step": 2200
+    },
+    {
+      "epoch": 0.8508999903744345,
+      "grad_norm": 0.23601125180721283,
+      "learning_rate": 5.740581055485722e-07,
+      "loss": 0.781,
+      "step": 2210
+    },
+    {
+      "epoch": 0.8547502165752238,
+      "grad_norm": 0.21925902366638184,
+      "learning_rate": 5.453921056471961e-07,
+      "loss": 0.8157,
+      "step": 2220
+    },
+    {
+      "epoch": 0.8586004427760131,
+      "grad_norm": 0.23972707986831665,
+      "learning_rate": 5.174189685761288e-07,
+      "loss": 0.8075,
+      "step": 2230
+    },
+    {
+      "epoch": 0.8624506689768023,
+      "grad_norm": 0.3530949652194977,
+      "learning_rate": 4.901430452338979e-07,
+      "loss": 0.8189,
+      "step": 2240
+    },
+    {
+      "epoch": 0.8663008951775917,
+      "grad_norm": 0.23236188292503357,
+      "learning_rate": 4.6356857807548286e-07,
+      "loss": 0.8041,
+      "step": 2250
+    },
+    {
+      "epoch": 0.870151121378381,
+      "grad_norm": 0.23859095573425293,
+      "learning_rate": 4.376997004524541e-07,
+      "loss": 0.7934,
+      "step": 2260
+    },
+    {
+      "epoch": 0.8740013475791703,
+      "grad_norm": 0.23183311522006989,
+      "learning_rate": 4.12540435970073e-07,
+      "loss": 0.8058,
+      "step": 2270
+    },
+    {
+      "epoch": 0.8778515737799596,
+      "grad_norm": 0.22914767265319824,
+      "learning_rate": 3.880946978614703e-07,
+      "loss": 0.8117,
+      "step": 2280
+    },
+    {
+      "epoch": 0.8817017999807488,
+      "grad_norm": 0.22884444892406464,
+      "learning_rate": 3.643662883789878e-07,
+      "loss": 0.8194,
+      "step": 2290
+    },
+    {
+      "epoch": 0.8855520261815382,
+      "grad_norm": 0.21766646206378937,
+      "learning_rate": 3.41358898202781e-07,
+      "loss": 0.795,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8894022523823275,
+      "grad_norm": 0.22319446504116058,
+      "learning_rate": 3.190761058667741e-07,
+      "loss": 0.8061,
+      "step": 2310
+    },
+    {
+      "epoch": 0.8932524785831167,
+      "grad_norm": 0.2301560342311859,
+      "learning_rate": 2.9752137720206463e-07,
+      "loss": 0.8135,
+      "step": 2320
+    },
+    {
+      "epoch": 0.8971027047839061,
+      "grad_norm": 0.23161546885967255,
+      "learning_rate": 2.7669806479785067e-07,
+      "loss": 0.8174,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9009529309846953,
+      "grad_norm": 0.2387039065361023,
+      "learning_rate": 2.566094074799763e-07,
+      "loss": 0.8131,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9048031571854847,
+      "grad_norm": 0.22804366052150726,
+      "learning_rate": 2.3725852980716557e-07,
+      "loss": 0.7922,
+      "step": 2350
+    },
+    {
+      "epoch": 0.908653383386274,
+      "grad_norm": 0.22775959968566895,
+      "learning_rate": 2.1864844158503983e-07,
+      "loss": 0.8063,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9125036095870632,
+      "grad_norm": 0.25649744272232056,
+      "learning_rate": 2.007820373979702e-07,
+      "loss": 0.798,
+      "step": 2370
+    },
+    {
+      "epoch": 0.9163538357878526,
+      "grad_norm": 0.23176084458827972,
+      "learning_rate": 1.8366209615886532e-07,
+      "loss": 0.8064,
+      "step": 2380
+    },
+    {
+      "epoch": 0.9202040619886418,
+      "grad_norm": 0.2286115139722824,
+      "learning_rate": 1.6729128067693613e-07,
+      "loss": 0.8151,
+      "step": 2390
+    },
+    {
+      "epoch": 0.9240542881894311,
+      "grad_norm": 0.2538682222366333,
+      "learning_rate": 1.5167213724353426e-07,
+      "loss": 0.817,
+      "step": 2400
+    },
+    {
+      "epoch": 0.9279045143902205,
+      "grad_norm": 0.2320171445608139,
+      "learning_rate": 1.3680709523610058e-07,
+      "loss": 0.8003,
+      "step": 2410
+    },
+    {
+      "epoch": 0.9317547405910097,
+      "grad_norm": 0.22797948122024536,
+      "learning_rate": 1.2269846674030418e-07,
+      "loss": 0.8226,
+      "step": 2420
+    },
+    {
+      "epoch": 0.935604966791799,
+      "grad_norm": 0.23333826661109924,
+      "learning_rate": 1.0934844619042595e-07,
+      "loss": 0.8181,
+      "step": 2430
+    },
+    {
+      "epoch": 0.9394551929925883,
+      "grad_norm": 0.2382403463125229,
+      "learning_rate": 9.675911002803961e-08,
+      "loss": 0.809,
+      "step": 2440
+    },
+    {
+      "epoch": 0.9433054191933776,
+      "grad_norm": 0.25066661834716797,
+      "learning_rate": 8.493241637904381e-08,
+      "loss": 0.8057,
+      "step": 2450
+    },
+    {
+      "epoch": 0.9471556453941669,
+      "grad_norm": 0.2384292185306549,
+      "learning_rate": 7.387020474909689e-08,
+      "loss": 0.8253,
+      "step": 2460
+    },
+    {
+      "epoch": 0.9510058715949562,
+      "grad_norm": 0.2511725127696991,
+      "learning_rate": 6.357419573750856e-08,
+      "loss": 0.8203,
+      "step": 2470
+    },
+    {
+      "epoch": 0.9548560977957455,
+      "grad_norm": 0.23605561256408691,
+      "learning_rate": 5.4045990769613365e-08,
+      "loss": 0.8039,
+      "step": 2480
+    },
+    {
+      "epoch": 0.9587063239965348,
+      "grad_norm": 0.2842068374156952,
+      "learning_rate": 4.5287071847694434e-08,
+      "loss": 0.8086,
+      "step": 2490
+    },
+    {
+      "epoch": 0.9625565501973241,
+      "grad_norm": 0.22939611971378326,
+      "learning_rate": 3.7298801320467817e-08,
+      "loss": 0.8002,
+      "step": 2500
+    },
+    {
+      "epoch": 0.9664067763981133,
+      "grad_norm": 0.23263263702392578,
+      "learning_rate": 3.0082421671192576e-08,
+      "loss": 0.7945,
+      "step": 2510
+    },
+    {
+      "epoch": 0.9702570025989027,
+      "grad_norm": 0.22831059992313385,
+      "learning_rate": 2.3639055324410355e-08,
+      "loss": 0.8166,
+      "step": 2520
+    },
+    {
+      "epoch": 0.974107228799692,
+      "grad_norm": 0.2354935109615326,
+      "learning_rate": 1.7969704471369477e-08,
+      "loss": 0.8012,
+      "step": 2530
+    },
+    {
+      "epoch": 0.9779574550004813,
+      "grad_norm": 0.2248469740152359,
+      "learning_rate": 1.3075250914142413e-08,
+      "loss": 0.8046,
+      "step": 2540
+    },
+    {
+      "epoch": 0.9818076812012706,
+      "grad_norm": 0.225304514169693,
+      "learning_rate": 8.956455928476049e-09,
+      "loss": 0.8154,
+      "step": 2550
+    },
+    {
+      "epoch": 0.9856579074020598,
+      "grad_norm": 0.2338920533657074,
+      "learning_rate": 5.6139601453786275e-09,
+      "loss": 0.8072,
+      "step": 2560
+    },
+    {
+      "epoch": 0.9895081336028492,
+      "grad_norm": 0.233718603849411,
+      "learning_rate": 3.0482834514805602e-09,
+      "loss": 0.7948,
+      "step": 2570
+    },
+    {
+      "epoch": 0.9933583598036385,
+      "grad_norm": 0.21753212809562683,
+      "learning_rate": 1.2598249081713366e-09,
+      "loss": 0.79,
+      "step": 2580
+    },
+    {
+      "epoch": 0.9972085860044277,
+      "grad_norm": 0.24563831090927124,
+      "learning_rate": 2.488626895302826e-10,
+      "loss": 0.8184,
+      "step": 2590
+    },
+    {
+      "epoch": 0.9999037443449803,
+      "step": 2597,
+      "total_flos": 3.56737551795277e+18,
+      "train_loss": 0.8354994926628903,
+      "train_runtime": 12954.6267,
+      "train_samples_per_second": 51.326,
+      "train_steps_per_second": 0.2
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2597,
+  "num_input_tokens_seen": 418855352,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.56737551795277e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

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

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:77363fe5117dd48e9e413c5c824620348a8c1190dadd98d111fa37e3b536aae9
+size 6264

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

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

continuous/qwen3/eupe_vit_t/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": 192,
+  "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": "vtb_ssl:eupe",
+  "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": "/cache/ckpt/download/tokenizer/continuous/eupe/EUPE-ViT-T/EUPE-ViT-T.pt",
+  "vocab_size": 151936
+}

continuous/qwen3/eupe_vit_t/pretrain/mm_projector.bin ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:f3911548f75f76223a2f3ab7af2e308b93d06c90dd4b7cc79263a6f59da7c3e1
+size 18368568

continuous/qwen3/eupe_vit_t/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": 4.631433486938477,
+      "learning_rate": 0.00032142857142857147,
+      "loss": 10.5516,
+      "step": 10
+    },
+    {
+      "epoch": 0.03669724770642202,
+      "grad_norm": 0.6552888751029968,
+      "learning_rate": 0.0006785714285714287,
+      "loss": 7.3171,
+      "step": 20
+    },
+    {
+      "epoch": 0.05504587155963303,
+      "grad_norm": 0.5749574899673462,
+      "learning_rate": 0.0009999907688176173,
+      "loss": 5.5234,
+      "step": 30
+    },
+    {
+      "epoch": 0.07339449541284404,
+      "grad_norm": 0.10169410705566406,
+      "learning_rate": 0.0009988834393115766,
+      "loss": 4.9525,
+      "step": 40
+    },
+    {
+      "epoch": 0.09174311926605505,
+      "grad_norm": 0.05236728489398956,
+      "learning_rate": 0.0009959345572940772,
+      "loss": 4.7766,
+      "step": 50
+    },
+    {
+      "epoch": 0.11009174311926606,
+      "grad_norm": 0.07881205528974533,
+      "learning_rate": 0.0009911550081155983,
+      "loss": 4.7014,
+      "step": 60
+    },
+    {
+      "epoch": 0.12844036697247707,
+      "grad_norm": 0.06380262225866318,
+      "learning_rate": 0.0009845624347566433,
+      "loss": 4.6436,
+      "step": 70
+    },
+    {
+      "epoch": 0.14678899082568808,
+      "grad_norm": 0.04670483246445656,
+      "learning_rate": 0.0009761811727013547,
+      "loss": 4.6082,
+      "step": 80
+    },
+    {
+      "epoch": 0.1651376146788991,
+      "grad_norm": 0.06421473622322083,
+      "learning_rate": 0.0009660421601067666,
+      "loss": 4.5707,
+      "step": 90
+    },
+    {
+      "epoch": 0.1834862385321101,
+      "grad_norm": 0.057049140334129333,
+      "learning_rate": 0.0009541828235992926,
+      "loss": 4.552,
+      "step": 100
+    },
+    {
+      "epoch": 0.2018348623853211,
+      "grad_norm": 0.07120932638645172,
+      "learning_rate": 0.0009406469401200151,
+      "loss": 4.5256,
+      "step": 110
+    },
+    {
+      "epoch": 0.22018348623853212,
+      "grad_norm": 0.05798851698637009,
+      "learning_rate": 0.0009254844753287492,
+      "loss": 4.5175,
+      "step": 120
+    },
+    {
+      "epoch": 0.23853211009174313,
+      "grad_norm": 0.04866760969161987,
+      "learning_rate": 0.0009087513991633924,
+      "loss": 4.4892,
+      "step": 130
+    },
+    {
+      "epoch": 0.25688073394495414,
+      "grad_norm": 0.0657285824418068,
+      "learning_rate": 0.0008905094792353917,
+      "loss": 4.4738,
+      "step": 140
+    },
+    {
+      "epoch": 0.27522935779816515,
+      "grad_norm": 0.06303583830595016,
+      "learning_rate": 0.0008708260528239789,
+      "loss": 4.473,
+      "step": 150
+    },
+    {
+      "epoch": 0.29357798165137616,
+      "grad_norm": 0.05447995290160179,
+      "learning_rate": 0.0008497737783108237,
+      "loss": 4.459,
+      "step": 160
+    },
+    {
+      "epoch": 0.3119266055045872,
+      "grad_norm": 0.05726242437958717,
+      "learning_rate": 0.0008274303669726426,
+      "loss": 4.4451,
+      "step": 170
+    },
+    {
+      "epoch": 0.3302752293577982,
+      "grad_norm": 0.05949408560991287,
+      "learning_rate": 0.0008038782961218137,
+      "loss": 4.4335,
+      "step": 180
+    },
+    {
+      "epoch": 0.3486238532110092,
+      "grad_norm": 0.07161182910203934,
+      "learning_rate": 0.000779204504653894,
+      "loss": 4.4302,
+      "step": 190
+    },
+    {
+      "epoch": 0.3669724770642202,
+      "grad_norm": 0.08935785293579102,
+      "learning_rate": 0.0007535000721258791,
+      "loss": 4.4096,
+      "step": 200
+    },
+    {
+      "epoch": 0.3853211009174312,
+      "grad_norm": 0.07697807252407074,
+      "learning_rate": 0.00072685988254984,
+      "loss": 4.4072,
+      "step": 210
+    },
+    {
+      "epoch": 0.4036697247706422,
+      "grad_norm": 0.06716114282608032,
+      "learning_rate": 0.0006993822741429906,
+      "loss": 4.4175,
+      "step": 220
+    },
+    {
+      "epoch": 0.42201834862385323,
+      "grad_norm": 0.05618627741932869,
+      "learning_rate": 0.0006711686763270818,
+      "loss": 4.4162,
+      "step": 230
+    },
+    {
+      "epoch": 0.44036697247706424,
+      "grad_norm": 0.05568604916334152,
+      "learning_rate": 0.0006423232353170831,
+      "loss": 4.3999,
+      "step": 240
+    },
+    {
+      "epoch": 0.45871559633027525,
+      "grad_norm": 0.05896687135100365,
+      "learning_rate": 0.0006129524296812335,
+      "loss": 4.3823,
+      "step": 250
+    },
+    {
+      "epoch": 0.47706422018348627,
+      "grad_norm": 0.052249837666749954,
+      "learning_rate": 0.0005831646772915651,
+      "loss": 4.396,
+      "step": 260
+    },
+    {
+      "epoch": 0.4954128440366973,
+      "grad_norm": 0.058744002133607864,
+      "learning_rate": 0.0005530699351157809,
+      "loss": 4.3792,
+      "step": 270
+    },
+    {
+      "epoch": 0.5137614678899083,
+      "grad_norm": 0.05226633697748184,
+      "learning_rate": 0.0005227792933277943,
+      "loss": 4.3769,
+      "step": 280
+    },
+    {
+      "epoch": 0.5321100917431193,
+      "grad_norm": 0.049597036093473434,
+      "learning_rate": 0.0004924045652352092,
+      "loss": 4.3606,
+      "step": 290
+    },
+    {
+      "epoch": 0.5504587155963303,
+      "grad_norm": 0.06306581944227219,
+      "learning_rate": 0.0004620578745374633,
+      "loss": 4.3612,
+      "step": 300
+    },
+    {
+      "epoch": 0.5688073394495413,
+      "grad_norm": 0.04717100411653519,
+      "learning_rate": 0.0004318512414382059,
+      "loss": 4.3756,
+      "step": 310
+    },
+    {
+      "epoch": 0.5871559633027523,
+      "grad_norm": 0.05522476136684418,
+      "learning_rate": 0.0004018961691397148,
+      "loss": 4.3682,
+      "step": 320
+    },
+    {
+      "epoch": 0.6055045871559633,
+      "grad_norm": 0.06528068333864212,
+      "learning_rate": 0.0003723032322457458,
+      "loss": 4.352,
+      "step": 330
+    },
+    {
+      "epoch": 0.6238532110091743,
+      "grad_norm": 0.062453512102365494,
+      "learning_rate": 0.0003431816685921625,
+      "loss": 4.3487,
+      "step": 340
+    },
+    {
+      "epoch": 0.6422018348623854,
+      "grad_norm": 0.06507401913404465,
+      "learning_rate": 0.0003146389760120416,
+      "loss": 4.3593,
+      "step": 350
+    },
+    {
+      "epoch": 0.6605504587155964,
+      "grad_norm": 0.05578480288386345,
+      "learning_rate": 0.00028678051552373485,
+      "loss": 4.367,
+      "step": 360
+    },
+    {
+      "epoch": 0.6788990825688074,
+      "grad_norm": 0.05135942995548248,
+      "learning_rate": 0.0002597091224066581,
+      "loss": 4.3487,
+      "step": 370
+    },
+    {
+      "epoch": 0.6972477064220184,
+      "grad_norm": 0.052246738225221634,
+      "learning_rate": 0.0002335247266004629,
+      "loss": 4.3652,
+      "step": 380
+    },
+    {
+      "epoch": 0.7155963302752294,
+      "grad_norm": 0.05735236778855324,
+      "learning_rate": 0.000208323983828828,
+      "loss": 4.3493,
+      "step": 390
+    },
+    {
+      "epoch": 0.7339449541284404,
+      "grad_norm": 0.05265427380800247,
+      "learning_rate": 0.0001841999188095224,
+      "loss": 4.3487,
+      "step": 400
+    },
+    {
+      "epoch": 0.7522935779816514,
+      "grad_norm": 0.054281312972307205,
+      "learning_rate": 0.00016124158186777676,
+      "loss": 4.349,
+      "step": 410
+    },
+    {
+      "epoch": 0.7706422018348624,
+      "grad_norm": 0.05127573758363724,
+      "learning_rate": 0.0001395337202205211,
+      "loss": 4.3409,
+      "step": 420
+    },
+    {
+      "epoch": 0.7889908256880734,
+      "grad_norm": 0.0551503524184227,
+      "learning_rate": 0.00011915646514489292,
+      "loss": 4.3475,
+      "step": 430
+    },
+    {
+      "epoch": 0.8073394495412844,
+      "grad_norm": 0.058349452912807465,
+      "learning_rate": 0.00010018503618578818,
+      "loss": 4.3536,
+      "step": 440
+    },
+    {
+      "epoch": 0.8256880733944955,
+      "grad_norm": 0.04724553972482681,
+      "learning_rate": 8.268946349432583e-05,
+      "loss": 4.3465,
+      "step": 450
+    },
+    {
+      "epoch": 0.8440366972477065,
+      "grad_norm": 0.04967445135116577,
+      "learning_rate": 6.673432932217117e-05,
+      "loss": 4.3369,
+      "step": 460
+    },
+    {
+      "epoch": 0.8623853211009175,
+      "grad_norm": 0.053452879190444946,
+      "learning_rate": 5.2378529625954684e-05,
+      "loss": 4.3276,
+      "step": 470
+    },
+    {
+      "epoch": 0.8807339449541285,
+      "grad_norm": 0.051092106848955154,
+      "learning_rate": 3.9675056661785556e-05,
+      "loss": 4.3388,
+      "step": 480
+    },
+    {
+      "epoch": 0.8990825688073395,
+      "grad_norm": 0.05085812881588936,
+      "learning_rate": 2.8670803372379538e-05,
+      "loss": 4.3402,
+      "step": 490
+    },
+    {
+      "epoch": 0.9174311926605505,
+      "grad_norm": 0.05015129595994949,
+      "learning_rate": 1.9406390288876586e-05,
+      "loss": 4.3515,
+      "step": 500
+    },
+    {
+      "epoch": 0.9357798165137615,
+      "grad_norm": 0.048234786838293076,
+      "learning_rate": 1.191601558631461e-05,
+      "loss": 4.3406,
+      "step": 510
+    },
+    {
+      "epoch": 0.9541284403669725,
+      "grad_norm": 0.0469999797642231,
+      "learning_rate": 6.2273288462559305e-06,
+      "loss": 4.3323,
+      "step": 520
+    },
+    {
+      "epoch": 0.9724770642201835,
+      "grad_norm": 0.04679225757718086,
+      "learning_rate": 2.361328992552314e-06,
+      "loss": 4.334,
+      "step": 530
+    },
+    {
+      "epoch": 0.9908256880733946,
+      "grad_norm": 0.05235099792480469,
+      "learning_rate": 3.322867770044202e-07,
+      "loss": 4.3397,
+      "step": 540
+    },
+    {
+      "epoch": 1.0,
+      "step": 545,
+      "total_flos": 2.716616829614162e+17,
+      "train_loss": 4.610839374787217,
+      "train_runtime": 5061.9346,
+      "train_samples_per_second": 110.26,
+      "train_steps_per_second": 0.108
+    }
+  ],
+  "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": 2.716616829614162e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

continuous/qwen3/ijepa_vith14/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/ijepa_vith14/finetune/config.json ADDED Viewed

	@@ -0,0 +1,59 @@

+{
+  "add_faster_video": false,
+  "add_time_instruction": false,
+  "architectures": [
+    "LlavaQwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "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,
+  "max_position_embeddings": 40960,
+  "max_window_layers": 28,
+  "mm_hidden_size": 1280,
+  "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": "vtb_ssl:ijepa",
+  "mm_vision_tower_lr": null,
+  "model_type": "qwen3",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 8,
+  "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",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "use_mm_proj": true,
+  "use_pos_skipping": false,
+  "use_sliding_window": false,
+  "vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/ijepa/ijepa_vith.pth",
+  "vocab_size": 151936
+}

continuous/qwen3/ijepa_vith14/finetune/generation_config.json ADDED Viewed

	@@ -0,0 +1,13 @@

+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "temperature": 0.6,
+  "top_k": 20,
+  "top_p": 0.95,
+  "transformers_version": "4.51.3"
+}

continuous/qwen3/ijepa_vith14/finetune/merges.txt ADDED Viewed

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