add checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50

.gitattributes

@@ -78,3 +78,9 @@ checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/checkpoint-2000/tokenizer.json filt
 checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/checkpoint-2500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/wandb/offline-run-20260622_212221-ipdo7bip/run-ipdo7bip.wandb filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/wandb/offline-run-20260622_173630-kv8jgb2s/run-kv8jgb2s.wandb filter=lfs diff=lfs merge=lfs -text

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1000/added_tokens.json

@@ -0,0 +1,35 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|action_sep|>": 151670,
+  "<|arg_sep|>": 151671,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|call_sep|>": 151666,
+  "<|end_of_text|>": 151673,
+  "<|endoftext|>": 151643,
+  "<|exception_sep|>": 151669,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|frame_sep|>": 151672,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|latent_end|>": 151675,
+  "<|latent_start|>": 151674,
+  "<|line_sep|>": 151667,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|return_sep|>": 151668,
+  "<|trace_context_start|>": 151665,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1000/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1000/config.json

@@ -0,0 +1,66 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "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",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 36,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "pad_token_id": 151643,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "transformers_version": "4.57.6",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151676
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92b6ecef084d0d7945d6a753849efb3c291d5d22874041d6b8797e45bc816430
+size 6187858991

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1000/special_tokens_map.json

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

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1000/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64d06cfe78797eb8dbf0ff96165cf2ffad5090c082450a5aa013358d629c1bc3
+size 16788033

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1000/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83a790d654474f5dfe225f889afd0210313eb1083f942671f2c4b8e95a1c922b
+size 11424004

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1000/tokenizer_config.json

@@ -0,0 +1,295 @@
+{
+  "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": "<|trace_context_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151666": {
+      "content": "<|call_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151667": {
+      "content": "<|line_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151668": {
+      "content": "<|return_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151669": {
+      "content": "<|exception_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<|action_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<|arg_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151672": {
+      "content": "<|frame_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151673": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151674": {
+      "content": "<|latent_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151675": {
+      "content": "<|latent_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "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": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1000/trainer_state.json

@@ -0,0 +1,2234 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.28835063437139563,
+  "eval_steps": 500,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0014417531718569781,
+      "grad_norm": 64.5,
+      "kd_loss": 0.2863670885562897,
+      "learning_rate": 4.444444444444445e-07,
+      "loss": 0.5774,
+      "ss_p": 0.0013333333333333333,
+      "step": 5,
+      "student_loss": 0.19185857474803925,
+      "teacher_loss": 0.0020165962632745504
+    },
+    {
+      "epoch": 0.0028835063437139563,
+      "grad_norm": 51.0,
+      "kd_loss": 0.27077770233154297,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 0.5416,
+      "ss_p": 0.003,
+      "step": 10,
+      "student_loss": 0.22956214845180511,
+      "teacher_loss": 0.005406768061220646
+    },
+    {
+      "epoch": 0.004325259515570935,
+      "grad_norm": 52.5,
+      "kd_loss": 0.23106813430786133,
+      "learning_rate": 1.5555555555555558e-06,
+      "loss": 0.4933,
+      "ss_p": 0.004666666666666667,
+      "step": 15,
+      "student_loss": 0.08662421256303787,
+      "teacher_loss": 0.0011790532153099775
+    },
+    {
+      "epoch": 0.0057670126874279125,
+      "grad_norm": 41.5,
+      "kd_loss": 0.20822274684906006,
+      "learning_rate": 2.1111111111111114e-06,
+      "loss": 0.4413,
+      "ss_p": 0.006333333333333333,
+      "step": 20,
+      "student_loss": 0.0842326357960701,
+      "teacher_loss": 0.0005264537176117301
+    },
+    {
+      "epoch": 0.00720876585928489,
+      "grad_norm": 14.4375,
+      "kd_loss": 0.21752168238162994,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 0.3384,
+      "ss_p": 0.008,
+      "step": 25,
+      "student_loss": 0.02283354476094246,
+      "teacher_loss": 0.0008345277165062726
+    },
+    {
+      "epoch": 0.00865051903114187,
+      "grad_norm": 10.5,
+      "kd_loss": 0.17801165580749512,
+      "learning_rate": 3.2222222222222227e-06,
+      "loss": 0.2811,
+      "ss_p": 0.009666666666666667,
+      "step": 30,
+      "student_loss": 0.013038136065006256,
+      "teacher_loss": 0.00035301479510962963
+    },
+    {
+      "epoch": 0.010092272202998846,
+      "grad_norm": 5.46875,
+      "kd_loss": 0.14403901994228363,
+      "learning_rate": 3.777777777777778e-06,
+      "loss": 0.2465,
+      "ss_p": 0.011333333333333334,
+      "step": 35,
+      "student_loss": 0.0017115750815719366,
+      "teacher_loss": 0.00027838119422085583
+    },
+    {
+      "epoch": 0.011534025374855825,
+      "grad_norm": 3.890625,
+      "kd_loss": 0.11463554203510284,
+      "learning_rate": 4.333333333333334e-06,
+      "loss": 0.1987,
+      "ss_p": 0.013,
+      "step": 40,
+      "student_loss": 0.051267627626657486,
+      "teacher_loss": 0.043106820434331894
+    },
+    {
+      "epoch": 0.012975778546712802,
+      "grad_norm": 3.703125,
+      "kd_loss": 0.1978389024734497,
+      "learning_rate": 4.888888888888889e-06,
+      "loss": 0.1562,
+      "ss_p": 0.014666666666666666,
+      "step": 45,
+      "student_loss": 0.07474519312381744,
+      "teacher_loss": 0.002643088111653924
+    },
+    {
+      "epoch": 0.01441753171856978,
+      "grad_norm": 2.796875,
+      "kd_loss": 0.06925997883081436,
+      "learning_rate": 5.444444444444445e-06,
+      "loss": 0.1565,
+      "ss_p": 0.01633333333333333,
+      "step": 50,
+      "student_loss": 0.02035536803305149,
+      "teacher_loss": 0.007393870037049055
+    },
+    {
+      "epoch": 0.015859284890426758,
+      "grad_norm": 2.421875,
+      "kd_loss": 0.061659786850214005,
+      "learning_rate": 6e-06,
+      "loss": 0.133,
+      "ss_p": 0.018,
+      "step": 55,
+      "student_loss": 0.0011764049995690584,
+      "teacher_loss": 0.0005172735545784235
+    },
+    {
+      "epoch": 0.01730103806228374,
+      "grad_norm": 3.921875,
+      "kd_loss": 0.05249356850981712,
+      "learning_rate": 6.555555555555556e-06,
+      "loss": 0.1161,
+      "ss_p": 0.019666666666666666,
+      "step": 60,
+      "student_loss": 0.002592012519016862,
+      "teacher_loss": 0.0012550005922093987
+    },
+    {
+      "epoch": 0.018742791234140715,
+      "grad_norm": 5.65625,
+      "kd_loss": 0.05232969671487808,
+      "learning_rate": 7.111111111111112e-06,
+      "loss": 0.1122,
+      "ss_p": 0.021333333333333333,
+      "step": 65,
+      "student_loss": 0.026336967945098877,
+      "teacher_loss": 0.03840035945177078
+    },
+    {
+      "epoch": 0.020184544405997693,
+      "grad_norm": 1.71875,
+      "kd_loss": 0.09275423735380173,
+      "learning_rate": 7.666666666666667e-06,
+      "loss": 0.1067,
+      "ss_p": 0.023,
+      "step": 70,
+      "student_loss": 0.016458863392472267,
+      "teacher_loss": 0.0014907496515661478
+    },
+    {
+      "epoch": 0.02162629757785467,
+      "grad_norm": 2.546875,
+      "kd_loss": 0.03650372475385666,
+      "learning_rate": 8.222222222222222e-06,
+      "loss": 0.0971,
+      "ss_p": 0.024666666666666667,
+      "step": 75,
+      "student_loss": 0.0020554568618535995,
+      "teacher_loss": 0.00103466073051095
+    },
+    {
+      "epoch": 0.02306805074971165,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.03666790947318077,
+      "learning_rate": 8.777777777777778e-06,
+      "loss": 0.082,
+      "ss_p": 0.026333333333333334,
+      "step": 80,
+      "student_loss": 0.003596354741603136,
+      "teacher_loss": 0.0015790935140103102
+    },
+    {
+      "epoch": 0.024509803921568627,
+      "grad_norm": 22.875,
+      "kd_loss": 0.057585254311561584,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.1142,
+      "ss_p": 0.028,
+      "step": 85,
+      "student_loss": 0.0022186494898051023,
+      "teacher_loss": 0.0016249272739514709
+    },
+    {
+      "epoch": 0.025951557093425604,
+      "grad_norm": 2.9375,
+      "kd_loss": 0.08127611130475998,
+      "learning_rate": 9.88888888888889e-06,
+      "loss": 0.0849,
+      "ss_p": 0.029666666666666668,
+      "step": 90,
+      "student_loss": 0.023147715255618095,
+      "teacher_loss": 0.0030787428840994835
+    },
+    {
+      "epoch": 0.027393310265282585,
+      "grad_norm": 2.03125,
+      "kd_loss": 0.0529036745429039,
+      "learning_rate": 9.99995337991255e-06,
+      "loss": 0.0727,
+      "ss_p": 0.03133333333333333,
+      "step": 95,
+      "student_loss": 0.010709365829825401,
+      "teacher_loss": 0.007111336570233107
+    },
+    {
+      "epoch": 0.02883506343713956,
+      "grad_norm": 4.5625,
+      "kd_loss": 0.0342501662671566,
+      "learning_rate": 9.999763987297266e-06,
+      "loss": 0.0646,
+      "ss_p": 0.033,
+      "step": 100,
+      "student_loss": 0.0012194744776934385,
+      "teacher_loss": 0.0011596690164878964
+    },
+    {
+      "epoch": 0.03027681660899654,
+      "grad_norm": 4.25,
+      "kd_loss": 0.04341937229037285,
+      "learning_rate": 9.999428913912876e-06,
+      "loss": 0.0756,
+      "ss_p": 0.034666666666666665,
+      "step": 105,
+      "student_loss": 0.0035452761221677065,
+      "teacher_loss": 0.0006120483740232885
+    },
+    {
+      "epoch": 0.031718569780853516,
+      "grad_norm": 2.234375,
+      "kd_loss": 0.04445260763168335,
+      "learning_rate": 9.998948169522587e-06,
+      "loss": 0.0704,
+      "ss_p": 0.036333333333333336,
+      "step": 110,
+      "student_loss": 0.032347422093153,
+      "teacher_loss": 0.017396895214915276
+    },
+    {
+      "epoch": 0.03316032295271049,
+      "grad_norm": 1.2109375,
+      "kd_loss": 0.03784821927547455,
+      "learning_rate": 9.998321768134101e-06,
+      "loss": 0.0663,
+      "ss_p": 0.038,
+      "step": 115,
+      "student_loss": 0.0008873185142874718,
+      "teacher_loss": 0.0005071499035693705
+    },
+    {
+      "epoch": 0.03460207612456748,
+      "grad_norm": 1.734375,
+      "kd_loss": 0.02907339669764042,
+      "learning_rate": 9.997549727999212e-06,
+      "loss": 0.0649,
+      "ss_p": 0.03966666666666667,
+      "step": 120,
+      "student_loss": 0.0014366256073117256,
+      "teacher_loss": 0.0008944892324507236
+    },
+    {
+      "epoch": 0.036043829296424454,
+      "grad_norm": 1.4140625,
+      "kd_loss": 0.07474926859140396,
+      "learning_rate": 9.996632071613257e-06,
+      "loss": 0.0674,
+      "ss_p": 0.04133333333333333,
+      "step": 125,
+      "student_loss": 0.0798906609416008,
+      "teacher_loss": 0.00038791491533629596
+    },
+    {
+      "epoch": 0.03748558246828143,
+      "grad_norm": 1.109375,
+      "kd_loss": 0.02972075715661049,
+      "learning_rate": 9.995568825714479e-06,
+      "loss": 0.0615,
+      "ss_p": 0.043,
+      "step": 130,
+      "student_loss": 0.0008196862763725221,
+      "teacher_loss": 0.0007098562782630324
+    },
+    {
+      "epoch": 0.03892733564013841,
+      "grad_norm": 1.2421875,
+      "kd_loss": 0.04688045382499695,
+      "learning_rate": 9.994360021283235e-06,
+      "loss": 0.0688,
+      "ss_p": 0.04466666666666667,
+      "step": 135,
+      "student_loss": 0.0013504090020433068,
+      "teacher_loss": 0.0007031455752439797
+    },
+    {
+      "epoch": 0.040369088811995385,
+      "grad_norm": 1.3515625,
+      "kd_loss": 0.0252703670412302,
+      "learning_rate": 9.993005693541098e-06,
+      "loss": 0.0627,
+      "ss_p": 0.04633333333333333,
+      "step": 140,
+      "student_loss": 0.0008158929413184524,
+      "teacher_loss": 0.0006232777959667146
+    },
+    {
+      "epoch": 0.04181084198385236,
+      "grad_norm": 1.5,
+      "kd_loss": 0.0884011834859848,
+      "learning_rate": 9.991505881949837e-06,
+      "loss": 0.0571,
+      "ss_p": 0.048,
+      "step": 145,
+      "student_loss": 0.06209264695644379,
+      "teacher_loss": 0.006001627072691917
+    },
+    {
+      "epoch": 0.04325259515570934,
+      "grad_norm": 2.53125,
+      "kd_loss": 0.01753244921565056,
+      "learning_rate": 9.989860630210251e-06,
+      "loss": 0.0594,
+      "ss_p": 0.049666666666666665,
+      "step": 150,
+      "student_loss": 0.012924237176775932,
+      "teacher_loss": 0.010354846715927124
+    },
+    {
+      "epoch": 0.04469434832756632,
+      "grad_norm": 1.5,
+      "kd_loss": 0.022433122619986534,
+      "learning_rate": 9.98806998626091e-06,
+      "loss": 0.0643,
+      "ss_p": 0.051333333333333335,
+      "step": 155,
+      "student_loss": 0.0025498506147414446,
+      "teacher_loss": 0.0007692191284149885
+    },
+    {
+      "epoch": 0.0461361014994233,
+      "grad_norm": 3.515625,
+      "kd_loss": 0.03412366658449173,
+      "learning_rate": 9.98613400227676e-06,
+      "loss": 0.0647,
+      "ss_p": 0.053,
+      "step": 160,
+      "student_loss": 0.21487264335155487,
+      "teacher_loss": 0.04478331282734871
+    },
+    {
+      "epoch": 0.04757785467128028,
+      "grad_norm": 1.328125,
+      "kd_loss": 0.03570609912276268,
+      "learning_rate": 9.984052734667586e-06,
+      "loss": 0.0578,
+      "ss_p": 0.05466666666666667,
+      "step": 165,
+      "student_loss": 0.0005230513052083552,
+      "teacher_loss": 0.004346254747360945
+    },
+    {
+      "epoch": 0.049019607843137254,
+      "grad_norm": 2.84375,
+      "kd_loss": 0.02066916599869728,
+      "learning_rate": 9.981826244076387e-06,
+      "loss": 0.0679,
+      "ss_p": 0.05633333333333333,
+      "step": 170,
+      "student_loss": 0.0007387064397335052,
+      "teacher_loss": 0.0004792494000867009
+    },
+    {
+      "epoch": 0.05046136101499423,
+      "grad_norm": 1.375,
+      "kd_loss": 0.028709346428513527,
+      "learning_rate": 9.979454595377594e-06,
+      "loss": 0.0566,
+      "ss_p": 0.058,
+      "step": 175,
+      "student_loss": 0.0034709912724792957,
+      "teacher_loss": 0.00040501507464796305
+    },
+    {
+      "epoch": 0.05190311418685121,
+      "grad_norm": 1.4765625,
+      "kd_loss": 0.03572613000869751,
+      "learning_rate": 9.97693785767519e-06,
+      "loss": 0.057,
+      "ss_p": 0.059666666666666666,
+      "step": 180,
+      "student_loss": 0.0007740050787106156,
+      "teacher_loss": 0.0003084918425884098
+    },
+    {
+      "epoch": 0.05334486735870819,
+      "grad_norm": 1.75,
+      "kd_loss": 0.037760552018880844,
+      "learning_rate": 9.9742761043007e-06,
+      "loss": 0.06,
+      "ss_p": 0.06133333333333333,
+      "step": 185,
+      "student_loss": 0.03433312103152275,
+      "teacher_loss": 0.00026245793560519814
+    },
+    {
+      "epoch": 0.05478662053056517,
+      "grad_norm": 2.453125,
+      "kd_loss": 0.05671067163348198,
+      "learning_rate": 9.971469412811032e-06,
+      "loss": 0.0554,
+      "ss_p": 0.063,
+      "step": 190,
+      "student_loss": 0.018113840371370316,
+      "teacher_loss": 0.0010532024316489697
+    },
+    {
+      "epoch": 0.056228373702422146,
+      "grad_norm": 2.671875,
+      "kd_loss": 0.05576304718852043,
+      "learning_rate": 9.968517864986247e-06,
+      "loss": 0.0559,
+      "ss_p": 0.06466666666666666,
+      "step": 195,
+      "student_loss": 0.022675305604934692,
+      "teacher_loss": 0.0003246408887207508
+    },
+    {
+      "epoch": 0.05767012687427912,
+      "grad_norm": 1.609375,
+      "kd_loss": 0.03567410260438919,
+      "learning_rate": 9.965421546827153e-06,
+      "loss": 0.0666,
+      "ss_p": 0.06633333333333333,
+      "step": 200,
+      "student_loss": 0.000706703751347959,
+      "teacher_loss": 0.00023577056708745658
+    },
+    {
+      "epoch": 0.0591118800461361,
+      "grad_norm": 1.5625,
+      "kd_loss": 0.02744506113231182,
+      "learning_rate": 9.962180548552812e-06,
+      "loss": 0.0642,
+      "ss_p": 0.068,
+      "step": 205,
+      "student_loss": 0.03751567006111145,
+      "teacher_loss": 0.029027285054326057
+    },
+    {
+      "epoch": 0.06055363321799308,
+      "grad_norm": 1.8125,
+      "kd_loss": 0.03411497920751572,
+      "learning_rate": 9.958794964597906e-06,
+      "loss": 0.0561,
+      "ss_p": 0.06966666666666667,
+      "step": 210,
+      "student_loss": 0.0009551342809572816,
+      "teacher_loss": 0.0004929212154820561
+    },
+    {
+      "epoch": 0.061995386389850055,
+      "grad_norm": 1.328125,
+      "kd_loss": 0.037674449384212494,
+      "learning_rate": 9.955264893609986e-06,
+      "loss": 0.057,
+      "ss_p": 0.07133333333333333,
+      "step": 215,
+      "student_loss": 0.000515033199917525,
+      "teacher_loss": 0.00020458470680750906
+    },
+    {
+      "epoch": 0.06343713956170703,
+      "grad_norm": 1.75,
+      "kd_loss": 0.04471597447991371,
+      "learning_rate": 9.951590438446597e-06,
+      "loss": 0.0554,
+      "ss_p": 0.073,
+      "step": 220,
+      "student_loss": 0.033046673983335495,
+      "teacher_loss": 0.009949561208486557
+    },
+    {
+      "epoch": 0.06487889273356401,
+      "grad_norm": 1.2734375,
+      "kd_loss": 0.022534122690558434,
+      "learning_rate": 9.947771706172285e-06,
+      "loss": 0.0667,
+      "ss_p": 0.07466666666666667,
+      "step": 225,
+      "student_loss": 0.039918579161167145,
+      "teacher_loss": 0.0019707411993294954
+    },
+    {
+      "epoch": 0.06632064590542099,
+      "grad_norm": 4.25,
+      "kd_loss": 0.042446766048669815,
+      "learning_rate": 9.943808808055472e-06,
+      "loss": 0.0544,
+      "ss_p": 0.07633333333333334,
+      "step": 230,
+      "student_loss": 0.0027608482632786036,
+      "teacher_loss": 0.00353744113817811
+    },
+    {
+      "epoch": 0.06776239907727798,
+      "grad_norm": 1.859375,
+      "kd_loss": 0.08581197261810303,
+      "learning_rate": 9.93970185956522e-06,
+      "loss": 0.0592,
+      "ss_p": 0.078,
+      "step": 235,
+      "student_loss": 0.026898548007011414,
+      "teacher_loss": 0.0004704761959146708
+    },
+    {
+      "epoch": 0.06920415224913495,
+      "grad_norm": 1.328125,
+      "kd_loss": 0.036941006779670715,
+      "learning_rate": 9.93545098036786e-06,
+      "loss": 0.0608,
+      "ss_p": 0.07966666666666666,
+      "step": 240,
+      "student_loss": 0.004996009636670351,
+      "teacher_loss": 0.0005259682075120509
+    },
+    {
+      "epoch": 0.07064590542099193,
+      "grad_norm": 4.625,
+      "kd_loss": 0.020961780101060867,
+      "learning_rate": 9.931056294323507e-06,
+      "loss": 0.0577,
+      "ss_p": 0.08133333333333333,
+      "step": 245,
+      "student_loss": 0.01238278765231371,
+      "teacher_loss": 0.013973773457109928
+    },
+    {
+      "epoch": 0.07208765859284891,
+      "grad_norm": 1.6796875,
+      "kd_loss": 0.02463451586663723,
+      "learning_rate": 9.926517929482454e-06,
+      "loss": 0.0607,
+      "ss_p": 0.083,
+      "step": 250,
+      "student_loss": 0.0007480139029212296,
+      "teacher_loss": 0.000364149222150445
+    },
+    {
+      "epoch": 0.07352941176470588,
+      "grad_norm": 1.5078125,
+      "kd_loss": 0.019065026193857193,
+      "learning_rate": 9.921836018081438e-06,
+      "loss": 0.0564,
+      "ss_p": 0.08466666666666667,
+      "step": 255,
+      "student_loss": 0.00039553557871840894,
+      "teacher_loss": 0.000350873451679945
+    },
+    {
+      "epoch": 0.07497116493656286,
+      "grad_norm": 2.265625,
+      "kd_loss": 0.03508452698588371,
+      "learning_rate": 9.917010696539793e-06,
+      "loss": 0.0627,
+      "ss_p": 0.08633333333333333,
+      "step": 260,
+      "student_loss": 0.04029975086450577,
+      "teacher_loss": 0.03254583850502968
+    },
+    {
+      "epoch": 0.07641291810841984,
+      "grad_norm": 2.125,
+      "kd_loss": 0.02173917554318905,
+      "learning_rate": 9.912042105455462e-06,
+      "loss": 0.0661,
+      "ss_p": 0.088,
+      "step": 265,
+      "student_loss": 0.0005539674893952906,
+      "teacher_loss": 0.000338099169312045
+    },
+    {
+      "epoch": 0.07785467128027682,
+      "grad_norm": 2.59375,
+      "kd_loss": 0.022049423307180405,
+      "learning_rate": 9.906930389600917e-06,
+      "loss": 0.0652,
+      "ss_p": 0.08966666666666667,
+      "step": 270,
+      "student_loss": 0.0019419180462136865,
+      "teacher_loss": 0.0019364559557288885
+    },
+    {
+      "epoch": 0.07929642445213379,
+      "grad_norm": 2.8125,
+      "kd_loss": 0.0304579995572567,
+      "learning_rate": 9.901675697918925e-06,
+      "loss": 0.0561,
+      "ss_p": 0.09133333333333334,
+      "step": 275,
+      "student_loss": 0.04892721772193909,
+      "teacher_loss": 0.0351153239607811
+    },
+    {
+      "epoch": 0.08073817762399077,
+      "grad_norm": 3.1875,
+      "kd_loss": 0.038798850029706955,
+      "learning_rate": 9.896278183518216e-06,
+      "loss": 0.0584,
+      "ss_p": 0.093,
+      "step": 280,
+      "student_loss": 0.0007021792698651552,
+      "teacher_loss": 0.00022823568724561483
+    },
+    {
+      "epoch": 0.08217993079584775,
+      "grad_norm": 2.625,
+      "kd_loss": 0.024285076186060905,
+      "learning_rate": 9.890738003669029e-06,
+      "loss": 0.0655,
+      "ss_p": 0.09466666666666666,
+      "step": 285,
+      "student_loss": 0.001277840230613947,
+      "teacher_loss": 0.0009347093873657286
+    },
+    {
+      "epoch": 0.08362168396770472,
+      "grad_norm": 2.140625,
+      "kd_loss": 0.01619662716984749,
+      "learning_rate": 9.885055319798514e-06,
+      "loss": 0.0722,
+      "ss_p": 0.09633333333333334,
+      "step": 290,
+      "student_loss": 0.022745976224541664,
+      "teacher_loss": 0.018432462587952614
+    },
+    {
+      "epoch": 0.0850634371395617,
+      "grad_norm": 1.9609375,
+      "kd_loss": 0.020541999489068985,
+      "learning_rate": 9.879230297486034e-06,
+      "loss": 0.0651,
+      "ss_p": 0.098,
+      "step": 295,
+      "student_loss": 0.0011434606276452541,
+      "teacher_loss": 0.0008217736030928791
+    },
+    {
+      "epoch": 0.08650519031141868,
+      "grad_norm": 2.765625,
+      "kd_loss": 0.0563744381070137,
+      "learning_rate": 9.873263106458353e-06,
+      "loss": 0.0579,
+      "ss_p": 0.09966666666666667,
+      "step": 300,
+      "student_loss": 0.017663387581706047,
+      "teacher_loss": 0.001860561314970255
+    },
+    {
+      "epoch": 0.08794694348327567,
+      "grad_norm": 1.375,
+      "kd_loss": 0.027216989547014236,
+      "learning_rate": 9.86715392058467e-06,
+      "loss": 0.0594,
+      "ss_p": 0.10133333333333333,
+      "step": 305,
+      "student_loss": 0.012391547672450542,
+      "teacher_loss": 0.004124736879020929
+    },
+    {
+      "epoch": 0.08938869665513265,
+      "grad_norm": 1.7578125,
+      "kd_loss": 0.023861510679125786,
+      "learning_rate": 9.860902917871566e-06,
+      "loss": 0.067,
+      "ss_p": 0.103,
+      "step": 310,
+      "student_loss": 0.0006996596930548549,
+      "teacher_loss": 0.0005677860463038087
+    },
+    {
+      "epoch": 0.09083044982698962,
+      "grad_norm": 1.8828125,
+      "kd_loss": 0.04223452880978584,
+      "learning_rate": 9.854510280457819e-06,
+      "loss": 0.0582,
+      "ss_p": 0.10466666666666667,
+      "step": 315,
+      "student_loss": 0.006460036616772413,
+      "teacher_loss": 0.0005626050988212228
+    },
+    {
+      "epoch": 0.0922722029988466,
+      "grad_norm": 3.609375,
+      "kd_loss": 0.027444876730442047,
+      "learning_rate": 9.847976194609083e-06,
+      "loss": 0.0671,
+      "ss_p": 0.10633333333333334,
+      "step": 320,
+      "student_loss": 0.0007372965919785202,
+      "teacher_loss": 0.0003845398605335504
+    },
+    {
+      "epoch": 0.09371395617070358,
+      "grad_norm": 2.65625,
+      "kd_loss": 0.07274553924798965,
+      "learning_rate": 9.841300850712479e-06,
+      "loss": 0.0609,
+      "ss_p": 0.108,
+      "step": 325,
+      "student_loss": 0.0047082072123885155,
+      "teacher_loss": 0.01834835484623909
+    },
+    {
+      "epoch": 0.09515570934256055,
+      "grad_norm": 1.3125,
+      "kd_loss": 0.05378974229097366,
+      "learning_rate": 9.834484443271035e-06,
+      "loss": 0.0613,
+      "ss_p": 0.10966666666666666,
+      "step": 330,
+      "student_loss": 0.019354818388819695,
+      "teacher_loss": 0.003437924664467573
+    },
+    {
+      "epoch": 0.09659746251441753,
+      "grad_norm": 1.3125,
+      "kd_loss": 0.015957042574882507,
+      "learning_rate": 9.827527170898019e-06,
+      "loss": 0.0545,
+      "ss_p": 0.11133333333333334,
+      "step": 335,
+      "student_loss": 0.0012550931423902512,
+      "teacher_loss": 0.001135819940827787
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 1.4375,
+      "kd_loss": 0.031383149325847626,
+      "learning_rate": 9.820429236311158e-06,
+      "loss": 0.0637,
+      "ss_p": 0.113,
+      "step": 340,
+      "student_loss": 0.011534185148775578,
+      "teacher_loss": 0.007344625890254974
+    },
+    {
+      "epoch": 0.09948096885813149,
+      "grad_norm": 3.34375,
+      "kd_loss": 0.03594260290265083,
+      "learning_rate": 9.81319084632673e-06,
+      "loss": 0.063,
+      "ss_p": 0.11466666666666667,
+      "step": 345,
+      "student_loss": 0.010385124944150448,
+      "teacher_loss": 0.00046041407040320337
+    },
+    {
+      "epoch": 0.10092272202998846,
+      "grad_norm": 1.71875,
+      "kd_loss": 0.06371043622493744,
+      "learning_rate": 9.805812211853534e-06,
+      "loss": 0.0514,
+      "ss_p": 0.11633333333333333,
+      "step": 350,
+      "student_loss": 0.005106152966618538,
+      "teacher_loss": 0.0012367554008960724
+    },
+    {
+      "epoch": 0.10236447520184544,
+      "grad_norm": 2.015625,
+      "kd_loss": 0.017179591581225395,
+      "learning_rate": 9.798293547886748e-06,
+      "loss": 0.0683,
+      "ss_p": 0.118,
+      "step": 355,
+      "student_loss": 0.0007231732015497983,
+      "teacher_loss": 0.0005862703546881676
+    },
+    {
+      "epoch": 0.10380622837370242,
+      "grad_norm": 1.453125,
+      "kd_loss": 0.015755275264382362,
+      "learning_rate": 9.790635073501662e-06,
+      "loss": 0.058,
+      "ss_p": 0.11966666666666667,
+      "step": 360,
+      "student_loss": 0.0009628906846046448,
+      "teacher_loss": 0.0007298106211237609
+    },
+    {
+      "epoch": 0.1052479815455594,
+      "grad_norm": 3.46875,
+      "kd_loss": 0.0353640653192997,
+      "learning_rate": 9.782837011847294e-06,
+      "loss": 0.0613,
+      "ss_p": 0.12133333333333333,
+      "step": 365,
+      "student_loss": 0.050860460847616196,
+      "teacher_loss": 0.031099138781428337
+    },
+    {
+      "epoch": 0.10668973471741638,
+      "grad_norm": 3.296875,
+      "kd_loss": 0.0550801083445549,
+      "learning_rate": 9.774899590139897e-06,
+      "loss": 0.0697,
+      "ss_p": 0.123,
+      "step": 370,
+      "student_loss": 0.020883264020085335,
+      "teacher_loss": 0.0005534720839932561
+    },
+    {
+      "epoch": 0.10813148788927336,
+      "grad_norm": 2.296875,
+      "kd_loss": 0.033242158591747284,
+      "learning_rate": 9.766823039656328e-06,
+      "loss": 0.0579,
+      "ss_p": 0.12466666666666666,
+      "step": 375,
+      "student_loss": 0.033453598618507385,
+      "teacher_loss": 0.020838754251599312
+    },
+    {
+      "epoch": 0.10957324106113034,
+      "grad_norm": 3.46875,
+      "kd_loss": 0.019318392500281334,
+      "learning_rate": 9.758607595727309e-06,
+      "loss": 0.0585,
+      "ss_p": 0.12633333333333333,
+      "step": 380,
+      "student_loss": 0.0009146491065621376,
+      "teacher_loss": 0.00070954579859972
+    },
+    {
+      "epoch": 0.11101499423298732,
+      "grad_norm": 2.75,
+      "kd_loss": 0.029564542695879936,
+      "learning_rate": 9.75025349773058e-06,
+      "loss": 0.0522,
+      "ss_p": 0.128,
+      "step": 385,
+      "student_loss": 0.001163977780379355,
+      "teacher_loss": 0.0009751158067956567
+    },
+    {
+      "epoch": 0.11245674740484429,
+      "grad_norm": 1.40625,
+      "kd_loss": 0.018493372946977615,
+      "learning_rate": 9.741760989083914e-06,
+      "loss": 0.0517,
+      "ss_p": 0.12966666666666668,
+      "step": 390,
+      "student_loss": 0.00033564394107088447,
+      "teacher_loss": 0.00030597165459766984
+    },
+    {
+      "epoch": 0.11389850057670127,
+      "grad_norm": 1.3515625,
+      "kd_loss": 0.020225515589118004,
+      "learning_rate": 9.73313031723803e-06,
+      "loss": 0.0583,
+      "ss_p": 0.13133333333333333,
+      "step": 395,
+      "student_loss": 0.0009163676295429468,
+      "teacher_loss": 0.0006653889431618154
+    },
+    {
+      "epoch": 0.11534025374855825,
+      "grad_norm": 1.46875,
+      "kd_loss": 0.03176708519458771,
+      "learning_rate": 9.724361733669383e-06,
+      "loss": 0.0533,
+      "ss_p": 0.133,
+      "step": 400,
+      "student_loss": 0.0014075544895604253,
+      "teacher_loss": 0.00035167118767276406
+    },
+    {
+      "epoch": 0.11678200692041522,
+      "grad_norm": 1.984375,
+      "kd_loss": 0.030771590769290924,
+      "learning_rate": 9.715455493872827e-06,
+      "loss": 0.058,
+      "ss_p": 0.13466666666666666,
+      "step": 405,
+      "student_loss": 0.011284356005489826,
+      "teacher_loss": 0.0005015083006583154
+    },
+    {
+      "epoch": 0.1182237600922722,
+      "grad_norm": 1.3984375,
+      "kd_loss": 0.03774675726890564,
+      "learning_rate": 9.70641185735419e-06,
+      "loss": 0.0639,
+      "ss_p": 0.13633333333333333,
+      "step": 410,
+      "student_loss": 0.014033801853656769,
+      "teacher_loss": 0.007850556634366512
+    },
+    {
+      "epoch": 0.11966551326412918,
+      "grad_norm": 1.984375,
+      "kd_loss": 0.028400558978319168,
+      "learning_rate": 9.697231087622691e-06,
+      "loss": 0.0617,
+      "ss_p": 0.138,
+      "step": 415,
+      "student_loss": 0.006174962967634201,
+      "teacher_loss": 0.004486659541726112
+    },
+    {
+      "epoch": 0.12110726643598616,
+      "grad_norm": 2.21875,
+      "kd_loss": 0.0441397987306118,
+      "learning_rate": 9.687913452183277e-06,
+      "loss": 0.0684,
+      "ss_p": 0.13966666666666666,
+      "step": 420,
+      "student_loss": 0.07838834822177887,
+      "teacher_loss": 0.023644467815756798
+    },
+    {
+      "epoch": 0.12254901960784313,
+      "grad_norm": 3.859375,
+      "kd_loss": 0.013343766331672668,
+      "learning_rate": 9.678459222528825e-06,
+      "loss": 0.0533,
+      "ss_p": 0.14133333333333334,
+      "step": 425,
+      "student_loss": 0.0031237537041306496,
+      "teacher_loss": 0.0034240910317748785
+    },
+    {
+      "epoch": 0.12399077277970011,
+      "grad_norm": 4.25,
+      "kd_loss": 0.053346991539001465,
+      "learning_rate": 9.668868674132224e-06,
+      "loss": 0.0708,
+      "ss_p": 0.143,
+      "step": 430,
+      "student_loss": 0.15212653577327728,
+      "teacher_loss": 0.06338228285312653
+    },
+    {
+      "epoch": 0.1254325259515571,
+      "grad_norm": 4.46875,
+      "kd_loss": 0.017751501873135567,
+      "learning_rate": 9.65914208643836e-06,
+      "loss": 0.0564,
+      "ss_p": 0.14466666666666667,
+      "step": 435,
+      "student_loss": 0.0012140851467847824,
+      "teacher_loss": 0.0009782634442672133
+    },
+    {
+      "epoch": 0.12687427912341406,
+      "grad_norm": 5.03125,
+      "kd_loss": 0.04744955152273178,
+      "learning_rate": 9.649279742855964e-06,
+      "loss": 0.0617,
+      "ss_p": 0.14633333333333334,
+      "step": 440,
+      "student_loss": 0.012753048911690712,
+      "teacher_loss": 0.00758012430742383
+    },
+    {
+      "epoch": 0.12831603229527105,
+      "grad_norm": 6.125,
+      "kd_loss": 0.036048877984285355,
+      "learning_rate": 9.639281930749363e-06,
+      "loss": 0.0631,
+      "ss_p": 0.148,
+      "step": 445,
+      "student_loss": 0.031179705634713173,
+      "teacher_loss": 0.018292605876922607
+    },
+    {
+      "epoch": 0.12975778546712802,
+      "grad_norm": 3.375,
+      "kd_loss": 0.047337114810943604,
+      "learning_rate": 9.629148941430096e-06,
+      "loss": 0.0744,
+      "ss_p": 0.14966666666666667,
+      "step": 450,
+      "student_loss": 0.0051811751909554005,
+      "teacher_loss": 0.0033224786166101694
+    },
+    {
+      "epoch": 0.131199538638985,
+      "grad_norm": 1.2890625,
+      "kd_loss": 0.027121078222990036,
+      "learning_rate": 9.618881070148436e-06,
+      "loss": 0.0696,
+      "ss_p": 0.15133333333333332,
+      "step": 455,
+      "student_loss": 0.0013372941175475717,
+      "teacher_loss": 0.0007214359357021749
+    },
+    {
+      "epoch": 0.13264129181084197,
+      "grad_norm": 2.765625,
+      "kd_loss": 0.043845392763614655,
+      "learning_rate": 9.608478616084784e-06,
+      "loss": 0.0539,
+      "ss_p": 0.153,
+      "step": 460,
+      "student_loss": 0.015617738477885723,
+      "teacher_loss": 0.0007256992976181209
+    },
+    {
+      "epoch": 0.13408304498269896,
+      "grad_norm": 3.0,
+      "kd_loss": 0.034967243671417236,
+      "learning_rate": 9.597941882340947e-06,
+      "loss": 0.0678,
+      "ss_p": 0.15466666666666667,
+      "step": 465,
+      "student_loss": 0.0006019921856932342,
+      "teacher_loss": 0.0004377762961667031
+    },
+    {
+      "epoch": 0.13552479815455595,
+      "grad_norm": 2.109375,
+      "kd_loss": 0.02909771539270878,
+      "learning_rate": 9.587271175931313e-06,
+      "loss": 0.0631,
+      "ss_p": 0.15633333333333332,
+      "step": 470,
+      "student_loss": 0.0005127487238496542,
+      "teacher_loss": 0.0003308009763713926
+    },
+    {
+      "epoch": 0.13696655132641292,
+      "grad_norm": 1.84375,
+      "kd_loss": 0.039995044469833374,
+      "learning_rate": 9.5764668077739e-06,
+      "loss": 0.0547,
+      "ss_p": 0.158,
+      "step": 475,
+      "student_loss": 0.0009442830341868103,
+      "teacher_loss": 0.0007654255605302751
+    },
+    {
+      "epoch": 0.1384083044982699,
+      "grad_norm": 1.625,
+      "kd_loss": 0.035417329519987106,
+      "learning_rate": 9.5655290926813e-06,
+      "loss": 0.0793,
+      "ss_p": 0.15966666666666668,
+      "step": 480,
+      "student_loss": 0.03640369325876236,
+      "teacher_loss": 0.027462221682071686
+    },
+    {
+      "epoch": 0.13985005767012687,
+      "grad_norm": 2.34375,
+      "kd_loss": 0.014298087917268276,
+      "learning_rate": 9.554458349351507e-06,
+      "loss": 0.0634,
+      "ss_p": 0.16133333333333333,
+      "step": 485,
+      "student_loss": 0.0011409457074478269,
+      "teacher_loss": 0.0012796911178156734
+    },
+    {
+      "epoch": 0.14129181084198386,
+      "grad_norm": 2.5,
+      "kd_loss": 0.02429298497736454,
+      "learning_rate": 9.54325490035863e-06,
+      "loss": 0.0709,
+      "ss_p": 0.163,
+      "step": 490,
+      "student_loss": 0.0038263762835413218,
+      "teacher_loss": 0.0019276788225397468
+    },
+    {
+      "epoch": 0.14273356401384082,
+      "grad_norm": 2.828125,
+      "kd_loss": 0.040927041321992874,
+      "learning_rate": 9.531919072143488e-06,
+      "loss": 0.0777,
+      "ss_p": 0.16466666666666666,
+      "step": 495,
+      "student_loss": 0.006731952540576458,
+      "teacher_loss": 0.0031836475245654583
+    },
+    {
+      "epoch": 0.14417531718569782,
+      "grad_norm": 2.015625,
+      "kd_loss": 0.024556661024689674,
+      "learning_rate": 9.520451195004107e-06,
+      "loss": 0.0644,
+      "ss_p": 0.16633333333333333,
+      "step": 500,
+      "student_loss": 0.000614077493082732,
+      "teacher_loss": 0.0004563273978419602
+    },
+    {
+      "epoch": 0.14561707035755478,
+      "grad_norm": 3.390625,
+      "kd_loss": 0.024445993825793266,
+      "learning_rate": 9.508851603086094e-06,
+      "loss": 0.077,
+      "ss_p": 0.168,
+      "step": 505,
+      "student_loss": 0.000669357308652252,
+      "teacher_loss": 0.00045785080874338746
+    },
+    {
+      "epoch": 0.14705882352941177,
+      "grad_norm": 3.296875,
+      "kd_loss": 0.02244059555232525,
+      "learning_rate": 9.497120634372894e-06,
+      "loss": 0.063,
+      "ss_p": 0.16966666666666666,
+      "step": 510,
+      "student_loss": 0.000732982240151614,
+      "teacher_loss": 0.0004698011325672269
+    },
+    {
+      "epoch": 0.14850057670126873,
+      "grad_norm": 3.625,
+      "kd_loss": 0.02795323170721531,
+      "learning_rate": 9.48525863067595e-06,
+      "loss": 0.0557,
+      "ss_p": 0.17133333333333334,
+      "step": 515,
+      "student_loss": 0.017220236361026764,
+      "teacher_loss": 0.00040931650437414646
+    },
+    {
+      "epoch": 0.14994232987312572,
+      "grad_norm": 2.390625,
+      "kd_loss": 0.0944812074303627,
+      "learning_rate": 9.473265937624748e-06,
+      "loss": 0.0664,
+      "ss_p": 0.173,
+      "step": 520,
+      "student_loss": 0.042549774050712585,
+      "teacher_loss": 0.012764347717165947
+    },
+    {
+      "epoch": 0.1513840830449827,
+      "grad_norm": 1.8828125,
+      "kd_loss": 0.030559686943888664,
+      "learning_rate": 9.461142904656722e-06,
+      "loss": 0.0695,
+      "ss_p": 0.17466666666666666,
+      "step": 525,
+      "student_loss": 0.0007777044083923101,
+      "teacher_loss": 0.0003210337890777737
+    },
+    {
+      "epoch": 0.15282583621683968,
+      "grad_norm": 1.40625,
+      "kd_loss": 0.023660320788621902,
+      "learning_rate": 9.448889885007105e-06,
+      "loss": 0.0651,
+      "ss_p": 0.17633333333333334,
+      "step": 530,
+      "student_loss": 0.0008825063705444336,
+      "teacher_loss": 0.0006451026420108974
+    },
+    {
+      "epoch": 0.15426758938869667,
+      "grad_norm": 1.09375,
+      "kd_loss": 0.0332137756049633,
+      "learning_rate": 9.436507235698613e-06,
+      "loss": 0.0612,
+      "ss_p": 0.178,
+      "step": 535,
+      "student_loss": 0.0014239865122362971,
+      "teacher_loss": 0.0005905443686060607
+    },
+    {
+      "epoch": 0.15570934256055363,
+      "grad_norm": 2.109375,
+      "kd_loss": 0.054817572236061096,
+      "learning_rate": 9.423995317531049e-06,
+      "loss": 0.0605,
+      "ss_p": 0.17966666666666667,
+      "step": 540,
+      "student_loss": 0.017393875867128372,
+      "teacher_loss": 0.0006006712792441249
+    },
+    {
+      "epoch": 0.15715109573241062,
+      "grad_norm": 2.421875,
+      "kd_loss": 0.01688145473599434,
+      "learning_rate": 9.4113544950708e-06,
+      "loss": 0.0514,
+      "ss_p": 0.18133333333333335,
+      "step": 545,
+      "student_loss": 0.0035911172162741423,
+      "teacher_loss": 0.0023353511933237314
+    },
+    {
+      "epoch": 0.15859284890426759,
+      "grad_norm": 3.5,
+      "kd_loss": 0.05606849864125252,
+      "learning_rate": 9.398585136640195e-06,
+      "loss": 0.0535,
+      "ss_p": 0.183,
+      "step": 550,
+      "student_loss": 0.0003872204106301069,
+      "teacher_loss": 0.004548444412648678
+    },
+    {
+      "epoch": 0.16003460207612458,
+      "grad_norm": 3.015625,
+      "kd_loss": 0.07959499955177307,
+      "learning_rate": 9.38568761430679e-06,
+      "loss": 0.0598,
+      "ss_p": 0.18466666666666667,
+      "step": 555,
+      "student_loss": 0.032066237181425095,
+      "teacher_loss": 0.001948776887729764
+    },
+    {
+      "epoch": 0.16147635524798154,
+      "grad_norm": 2.25,
+      "kd_loss": 0.03467326983809471,
+      "learning_rate": 9.372662303872514e-06,
+      "loss": 0.0625,
+      "ss_p": 0.18633333333333332,
+      "step": 560,
+      "student_loss": 0.00195577391423285,
+      "teacher_loss": 0.0003902352473232895
+    },
+    {
+      "epoch": 0.16291810841983853,
+      "grad_norm": 2.75,
+      "kd_loss": 0.04329576715826988,
+      "learning_rate": 9.359509584862735e-06,
+      "loss": 0.0692,
+      "ss_p": 0.188,
+      "step": 565,
+      "student_loss": 0.0301277544349432,
+      "teacher_loss": 0.021337011829018593
+    },
+    {
+      "epoch": 0.1643598615916955,
+      "grad_norm": 3.9375,
+      "kd_loss": 0.03518754243850708,
+      "learning_rate": 9.346229840515186e-06,
+      "loss": 0.0664,
+      "ss_p": 0.18966666666666668,
+      "step": 570,
+      "student_loss": 0.011308778077363968,
+      "teacher_loss": 0.007574025541543961
+    },
+    {
+      "epoch": 0.16580161476355249,
+      "grad_norm": 1.1875,
+      "kd_loss": 0.025030676275491714,
+      "learning_rate": 9.332823457768805e-06,
+      "loss": 0.0588,
+      "ss_p": 0.19133333333333333,
+      "step": 575,
+      "student_loss": 0.0008461016695946455,
+      "teacher_loss": 0.00043505072244443
+    },
+    {
+      "epoch": 0.16724336793540945,
+      "grad_norm": 2.265625,
+      "kd_loss": 0.02528354711830616,
+      "learning_rate": 9.31929082725246e-06,
+      "loss": 0.0537,
+      "ss_p": 0.193,
+      "step": 580,
+      "student_loss": 0.0017597861588001251,
+      "teacher_loss": 0.002002239925786853
+    },
+    {
+      "epoch": 0.16868512110726644,
+      "grad_norm": 1.5859375,
+      "kd_loss": 0.029205303639173508,
+      "learning_rate": 9.305632343273567e-06,
+      "loss": 0.0539,
+      "ss_p": 0.19466666666666665,
+      "step": 585,
+      "student_loss": 0.0007650582119822502,
+      "teacher_loss": 0.0003985688090324402
+    },
+    {
+      "epoch": 0.1701268742791234,
+      "grad_norm": 1.2109375,
+      "kd_loss": 0.025012480095028877,
+      "learning_rate": 9.291848403806604e-06,
+      "loss": 0.0556,
+      "ss_p": 0.19633333333333333,
+      "step": 590,
+      "student_loss": 0.0013493653386831284,
+      "teacher_loss": 0.0016745957545936108
+    },
+    {
+      "epoch": 0.1715686274509804,
+      "grad_norm": 1.5625,
+      "kd_loss": 0.022375669330358505,
+      "learning_rate": 9.277939410481507e-06,
+      "loss": 0.0657,
+      "ss_p": 0.198,
+      "step": 595,
+      "student_loss": 0.0006021016160957515,
+      "teacher_loss": 0.0005392336170189083
+    },
+    {
+      "epoch": 0.17301038062283736,
+      "grad_norm": 2.1875,
+      "kd_loss": 0.043182745575904846,
+      "learning_rate": 9.263905768571977e-06,
+      "loss": 0.0663,
+      "ss_p": 0.19966666666666666,
+      "step": 600,
+      "student_loss": 0.04248778894543648,
+      "teacher_loss": 0.005614001303911209
+    },
+    {
+      "epoch": 0.17445213379469435,
+      "grad_norm": 1.9453125,
+      "kd_loss": 0.04084120690822601,
+      "learning_rate": 9.249747886983664e-06,
+      "loss": 0.0686,
+      "ss_p": 0.20133333333333334,
+      "step": 605,
+      "student_loss": 0.13167093694210052,
+      "teacher_loss": 0.034250061959028244
+    },
+    {
+      "epoch": 0.17589388696655134,
+      "grad_norm": 2.546875,
+      "kd_loss": 0.03859245032072067,
+      "learning_rate": 9.235466178242255e-06,
+      "loss": 0.0561,
+      "ss_p": 0.203,
+      "step": 610,
+      "student_loss": 0.012362505309283733,
+      "teacher_loss": 0.0005754792364314198
+    },
+    {
+      "epoch": 0.1773356401384083,
+      "grad_norm": 3.140625,
+      "kd_loss": 0.021341314539313316,
+      "learning_rate": 9.221061058481456e-06,
+      "loss": 0.0497,
+      "ss_p": 0.20466666666666666,
+      "step": 615,
+      "student_loss": 0.0007531295414082706,
+      "teacher_loss": 0.0006147727835923433
+    },
+    {
+      "epoch": 0.1787773933102653,
+      "grad_norm": 2.1875,
+      "kd_loss": 0.03471383824944496,
+      "learning_rate": 9.206532947430865e-06,
+      "loss": 0.0716,
+      "ss_p": 0.20633333333333334,
+      "step": 620,
+      "student_loss": 0.011007798835635185,
+      "teacher_loss": 0.0073872217908501625
+    },
+    {
+      "epoch": 0.18021914648212226,
+      "grad_norm": 3.125,
+      "kd_loss": 0.03351916745305061,
+      "learning_rate": 9.191882268403743e-06,
+      "loss": 0.0523,
+      "ss_p": 0.208,
+      "step": 625,
+      "student_loss": 0.03929333761334419,
+      "teacher_loss": 0.006530145648866892
+    },
+    {
+      "epoch": 0.18166089965397925,
+      "grad_norm": 6.09375,
+      "kd_loss": 0.04052945598959923,
+      "learning_rate": 9.177109448284673e-06,
+      "loss": 0.0551,
+      "ss_p": 0.20966666666666667,
+      "step": 630,
+      "student_loss": 0.0015670763095840812,
+      "teacher_loss": 0.0019278909312561154
+    },
+    {
+      "epoch": 0.1831026528258362,
+      "grad_norm": 1.390625,
+      "kd_loss": 0.03274727985262871,
+      "learning_rate": 9.162214917517139e-06,
+      "loss": 0.0835,
+      "ss_p": 0.21133333333333335,
+      "step": 635,
+      "student_loss": 0.0011002587852999568,
+      "teacher_loss": 0.00039536177064292133
+    },
+    {
+      "epoch": 0.1845444059976932,
+      "grad_norm": 2.359375,
+      "kd_loss": 0.037640687078237534,
+      "learning_rate": 9.14719911009096e-06,
+      "loss": 0.0669,
+      "ss_p": 0.213,
+      "step": 640,
+      "student_loss": 0.002206522738561034,
+      "teacher_loss": 0.0011659987503662705
+    },
+    {
+      "epoch": 0.18598615916955016,
+      "grad_norm": 2.359375,
+      "kd_loss": 0.04081936180591583,
+      "learning_rate": 9.132062463529666e-06,
+      "loss": 0.0649,
+      "ss_p": 0.21466666666666667,
+      "step": 645,
+      "student_loss": 0.0010437730234116316,
+      "teacher_loss": 0.0008008155273273587
+    },
+    {
+      "epoch": 0.18742791234140715,
+      "grad_norm": 3.15625,
+      "kd_loss": 0.01630258560180664,
+      "learning_rate": 9.11680541887774e-06,
+      "loss": 0.0663,
+      "ss_p": 0.21633333333333332,
+      "step": 650,
+      "student_loss": 0.0011455006897449493,
+      "teacher_loss": 0.0010825780918821692
+    },
+    {
+      "epoch": 0.18886966551326412,
+      "grad_norm": 1.5,
+      "kd_loss": 0.055530741810798645,
+      "learning_rate": 9.101428420687759e-06,
+      "loss": 0.0626,
+      "ss_p": 0.218,
+      "step": 655,
+      "student_loss": 0.0020486980210989714,
+      "teacher_loss": 0.0006226010737009346
+    },
+    {
+      "epoch": 0.1903114186851211,
+      "grad_norm": 2.28125,
+      "kd_loss": 0.024903684854507446,
+      "learning_rate": 9.085931917007464e-06,
+      "loss": 0.0746,
+      "ss_p": 0.21966666666666668,
+      "step": 660,
+      "student_loss": 0.0009118863381445408,
+      "teacher_loss": 0.0011200098088011146
+    },
+    {
+      "epoch": 0.19175317185697807,
+      "grad_norm": 2.421875,
+      "kd_loss": 0.09853949397802353,
+      "learning_rate": 9.070316359366681e-06,
+      "loss": 0.0734,
+      "ss_p": 0.22133333333333333,
+      "step": 665,
+      "student_loss": 0.007519611157476902,
+      "teacher_loss": 0.0017423606477677822
+    },
+    {
+      "epoch": 0.19319492502883506,
+      "grad_norm": 6.625,
+      "kd_loss": 0.039860885590314865,
+      "learning_rate": 9.054582202764175e-06,
+      "loss": 0.0665,
+      "ss_p": 0.223,
+      "step": 670,
+      "student_loss": 0.04489629715681076,
+      "teacher_loss": 0.011484071612358093
+    },
+    {
+      "epoch": 0.19463667820069205,
+      "grad_norm": 2.578125,
+      "kd_loss": 0.016214454546570778,
+      "learning_rate": 9.038729905654393e-06,
+      "loss": 0.071,
+      "ss_p": 0.22466666666666665,
+      "step": 675,
+      "student_loss": 0.0020811932627111673,
+      "teacher_loss": 0.0019474982982501388
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 5.0,
+      "kd_loss": 0.08082395792007446,
+      "learning_rate": 9.022759929934109e-06,
+      "loss": 0.0638,
+      "ss_p": 0.22633333333333333,
+      "step": 680,
+      "student_loss": 0.30400002002716064,
+      "teacher_loss": 0.04030829668045044
+    },
+    {
+      "epoch": 0.197520184544406,
+      "grad_norm": 1.859375,
+      "kd_loss": 0.025356216356158257,
+      "learning_rate": 9.006672740928952e-06,
+      "loss": 0.0672,
+      "ss_p": 0.228,
+      "step": 685,
+      "student_loss": 0.03995335474610329,
+      "teacher_loss": 0.03175293281674385
+    },
+    {
+      "epoch": 0.19896193771626297,
+      "grad_norm": 2.953125,
+      "kd_loss": 0.0686028003692627,
+      "learning_rate": 8.990468807379861e-06,
+      "loss": 0.0657,
+      "ss_p": 0.22966666666666666,
+      "step": 690,
+      "student_loss": 0.007285703904926777,
+      "teacher_loss": 0.011531146243214607
+    },
+    {
+      "epoch": 0.20040369088811996,
+      "grad_norm": 5.5,
+      "kd_loss": 0.049407538026571274,
+      "learning_rate": 8.974148601429426e-06,
+      "loss": 0.0789,
+      "ss_p": 0.23133333333333334,
+      "step": 695,
+      "student_loss": 0.003694782964885235,
+      "teacher_loss": 0.007218712009489536
+    },
+    {
+      "epoch": 0.20184544405997693,
+      "grad_norm": 4.625,
+      "kd_loss": 0.02625896967947483,
+      "learning_rate": 8.957712598608123e-06,
+      "loss": 0.0617,
+      "ss_p": 0.233,
+      "step": 700,
+      "student_loss": 0.001044552423991263,
+      "teacher_loss": 0.0004832793783862144
+    },
+    {
+      "epoch": 0.20328719723183392,
+      "grad_norm": 1.53125,
+      "kd_loss": 0.01772036775946617,
+      "learning_rate": 8.941161277820462e-06,
+      "loss": 0.0629,
+      "ss_p": 0.23466666666666666,
+      "step": 705,
+      "student_loss": 0.002373241586610675,
+      "teacher_loss": 0.0014788242988288403
+    },
+    {
+      "epoch": 0.20472895040369088,
+      "grad_norm": 4.625,
+      "kd_loss": 0.07706381380558014,
+      "learning_rate": 8.924495121331038e-06,
+      "loss": 0.0714,
+      "ss_p": 0.23633333333333334,
+      "step": 710,
+      "student_loss": 0.023498786613345146,
+      "teacher_loss": 0.004984163213521242
+    },
+    {
+      "epoch": 0.20617070357554787,
+      "grad_norm": 6.125,
+      "kd_loss": 0.028038812801241875,
+      "learning_rate": 8.907714614750473e-06,
+      "loss": 0.0707,
+      "ss_p": 0.238,
+      "step": 715,
+      "student_loss": 0.002961911726742983,
+      "teacher_loss": 0.006258171051740646
+    },
+    {
+      "epoch": 0.20761245674740483,
+      "grad_norm": 1.6875,
+      "kd_loss": 0.027723219245672226,
+      "learning_rate": 8.890820247021263e-06,
+      "loss": 0.0668,
+      "ss_p": 0.23966666666666667,
+      "step": 720,
+      "student_loss": 0.001304189208894968,
+      "teacher_loss": 0.0014796392060816288
+    },
+    {
+      "epoch": 0.20905420991926182,
+      "grad_norm": 3.375,
+      "kd_loss": 0.019339652732014656,
+      "learning_rate": 8.873812510403545e-06,
+      "loss": 0.0517,
+      "ss_p": 0.24133333333333334,
+      "step": 725,
+      "student_loss": 0.0005653988919220865,
+      "teacher_loss": 0.0005336694302968681
+    },
+    {
+      "epoch": 0.2104959630911188,
+      "grad_norm": 3.609375,
+      "kd_loss": 0.03018653392791748,
+      "learning_rate": 8.85669190046074e-06,
+      "loss": 0.0671,
+      "ss_p": 0.243,
+      "step": 730,
+      "student_loss": 0.0039001437835395336,
+      "teacher_loss": 0.0023783277720212936
+    },
+    {
+      "epoch": 0.21193771626297578,
+      "grad_norm": 2.671875,
+      "kd_loss": 0.03969630226492882,
+      "learning_rate": 8.839458916045117e-06,
+      "loss": 0.0734,
+      "ss_p": 0.24466666666666667,
+      "step": 735,
+      "student_loss": 0.0012796723749488592,
+      "teacher_loss": 0.0002167494676541537
+    },
+    {
+      "epoch": 0.21337946943483277,
+      "grad_norm": 3.109375,
+      "kd_loss": 0.04305332526564598,
+      "learning_rate": 8.822114059283267e-06,
+      "loss": 0.0607,
+      "ss_p": 0.24633333333333332,
+      "step": 740,
+      "student_loss": 0.0048164562322199345,
+      "teacher_loss": 0.0003611106367316097
+    },
+    {
+      "epoch": 0.21482122260668973,
+      "grad_norm": 3.421875,
+      "kd_loss": 0.03184329718351364,
+      "learning_rate": 8.804657835561456e-06,
+      "loss": 0.0608,
+      "ss_p": 0.248,
+      "step": 745,
+      "student_loss": 0.035390615463256836,
+      "teacher_loss": 0.019044777378439903
+    },
+    {
+      "epoch": 0.21626297577854672,
+      "grad_norm": 3.96875,
+      "kd_loss": 0.06939367204904556,
+      "learning_rate": 8.787090753510917e-06,
+      "loss": 0.0638,
+      "ss_p": 0.24966666666666668,
+      "step": 750,
+      "student_loss": 0.07072567939758301,
+      "teacher_loss": 0.007685464806854725
+    },
+    {
+      "epoch": 0.2177047289504037,
+      "grad_norm": 2.375,
+      "kd_loss": 0.030518613755702972,
+      "learning_rate": 8.76941332499301e-06,
+      "loss": 0.071,
+      "ss_p": 0.25133333333333335,
+      "step": 755,
+      "student_loss": 0.036837998777627945,
+      "teacher_loss": 0.008802128955721855
+    },
+    {
+      "epoch": 0.21914648212226068,
+      "grad_norm": 2.0625,
+      "kd_loss": 0.07421382516622543,
+      "learning_rate": 8.751626065084328e-06,
+      "loss": 0.0621,
+      "ss_p": 0.253,
+      "step": 760,
+      "student_loss": 0.049801260232925415,
+      "teacher_loss": 0.0030652873683720827
+    },
+    {
+      "epoch": 0.22058823529411764,
+      "grad_norm": 2.21875,
+      "kd_loss": 0.08238057792186737,
+      "learning_rate": 8.733729492061675e-06,
+      "loss": 0.0758,
+      "ss_p": 0.25466666666666665,
+      "step": 765,
+      "student_loss": 0.004180022981017828,
+      "teacher_loss": 0.0014872766332700849
+    },
+    {
+      "epoch": 0.22202998846597463,
+      "grad_norm": 5.3125,
+      "kd_loss": 0.05960839241743088,
+      "learning_rate": 8.715724127386971e-06,
+      "loss": 0.0618,
+      "ss_p": 0.25633333333333336,
+      "step": 770,
+      "student_loss": 0.0062656765803694725,
+      "teacher_loss": 0.003912253770977259
+    },
+    {
+      "epoch": 0.2234717416378316,
+      "grad_norm": 3.40625,
+      "kd_loss": 0.04264191910624504,
+      "learning_rate": 8.697610495692055e-06,
+      "loss": 0.0603,
+      "ss_p": 0.258,
+      "step": 775,
+      "student_loss": 0.0029328938107937574,
+      "teacher_loss": 0.002261607674881816
+    },
+    {
+      "epoch": 0.22491349480968859,
+      "grad_norm": 1.5234375,
+      "kd_loss": 0.036945097148418427,
+      "learning_rate": 8.679389124763396e-06,
+      "loss": 0.0721,
+      "ss_p": 0.25966666666666666,
+      "step": 780,
+      "student_loss": 0.0006587983807548881,
+      "teacher_loss": 0.0003180545463692397
+    },
+    {
+      "epoch": 0.22635524798154555,
+      "grad_norm": 6.875,
+      "kd_loss": 0.020621806383132935,
+      "learning_rate": 8.661060545526724e-06,
+      "loss": 0.063,
+      "ss_p": 0.2613333333333333,
+      "step": 785,
+      "student_loss": 0.0010969371069222689,
+      "teacher_loss": 0.0009253812022507191
+    },
+    {
+      "epoch": 0.22779700115340254,
+      "grad_norm": 2.1875,
+      "kd_loss": 0.015935583040118217,
+      "learning_rate": 8.64262529203155e-06,
+      "loss": 0.0574,
+      "ss_p": 0.263,
+      "step": 790,
+      "student_loss": 0.0019491399871185422,
+      "teacher_loss": 0.0011666042264550924
+    },
+    {
+      "epoch": 0.2292387543252595,
+      "grad_norm": 1.921875,
+      "kd_loss": 0.03370453044772148,
+      "learning_rate": 8.624083901435613e-06,
+      "loss": 0.0655,
+      "ss_p": 0.26466666666666666,
+      "step": 795,
+      "student_loss": 0.0008452001493424177,
+      "teacher_loss": 0.0005345051758922637
+    },
+    {
+      "epoch": 0.2306805074971165,
+      "grad_norm": 2.953125,
+      "kd_loss": 0.07227624952793121,
+      "learning_rate": 8.60543691398922e-06,
+      "loss": 0.0755,
+      "ss_p": 0.2663333333333333,
+      "step": 800,
+      "student_loss": 0.12735629081726074,
+      "teacher_loss": 0.035768210887908936
+    },
+    {
+      "epoch": 0.23212226066897348,
+      "grad_norm": 1.8515625,
+      "kd_loss": 0.023104339838027954,
+      "learning_rate": 8.586684873019513e-06,
+      "loss": 0.0702,
+      "ss_p": 0.268,
+      "step": 805,
+      "student_loss": 0.0012016247492283583,
+      "teacher_loss": 0.0010145728010684252
+    },
+    {
+      "epoch": 0.23356401384083045,
+      "grad_norm": 18.625,
+      "kd_loss": 0.04012278467416763,
+      "learning_rate": 8.567828324914633e-06,
+      "loss": 0.0665,
+      "ss_p": 0.26966666666666667,
+      "step": 810,
+      "student_loss": 0.0039580329321324825,
+      "teacher_loss": 0.001885153935290873
+    },
+    {
+      "epoch": 0.23500576701268744,
+      "grad_norm": 2.859375,
+      "kd_loss": 0.02250695787370205,
+      "learning_rate": 8.548867819107803e-06,
+      "loss": 0.0644,
+      "ss_p": 0.2713333333333333,
+      "step": 815,
+      "student_loss": 0.002193224150687456,
+      "teacher_loss": 0.012057735584676266
+    },
+    {
+      "epoch": 0.2364475201845444,
+      "grad_norm": 3.171875,
+      "kd_loss": 0.03591145575046539,
+      "learning_rate": 8.52980390806131e-06,
+      "loss": 0.0656,
+      "ss_p": 0.273,
+      "step": 820,
+      "student_loss": 0.04509356617927551,
+      "teacher_loss": 0.0014289942337200046
+    },
+    {
+      "epoch": 0.2378892733564014,
+      "grad_norm": 5.96875,
+      "kd_loss": 0.05026383697986603,
+      "learning_rate": 8.51063714725042e-06,
+      "loss": 0.0675,
+      "ss_p": 0.27466666666666667,
+      "step": 825,
+      "student_loss": 0.002324386965483427,
+      "teacher_loss": 0.017429033294320107
+    },
+    {
+      "epoch": 0.23933102652825836,
+      "grad_norm": 2.1875,
+      "kd_loss": 0.04187867417931557,
+      "learning_rate": 8.49136809514719e-06,
+      "loss": 0.0686,
+      "ss_p": 0.2763333333333333,
+      "step": 830,
+      "student_loss": 0.022419560700654984,
+      "teacher_loss": 0.013296069577336311
+    },
+    {
+      "epoch": 0.24077277970011535,
+      "grad_norm": 2.21875,
+      "kd_loss": 0.034400805830955505,
+      "learning_rate": 8.471997313204183e-06,
+      "loss": 0.0562,
+      "ss_p": 0.278,
+      "step": 835,
+      "student_loss": 0.014651641249656677,
+      "teacher_loss": 0.0035596999805420637
+    },
+    {
+      "epoch": 0.2422145328719723,
+      "grad_norm": 1.84375,
+      "kd_loss": 0.045325540006160736,
+      "learning_rate": 8.452525365838123e-06,
+      "loss": 0.0606,
+      "ss_p": 0.2796666666666667,
+      "step": 840,
+      "student_loss": 0.0005841721431352198,
+      "teacher_loss": 0.0002929263573605567
+    },
+    {
+      "epoch": 0.2436562860438293,
+      "grad_norm": 3.390625,
+      "kd_loss": 0.02255791425704956,
+      "learning_rate": 8.432952820413447e-06,
+      "loss": 0.0805,
+      "ss_p": 0.2813333333333333,
+      "step": 845,
+      "student_loss": 0.0009223757078871131,
+      "teacher_loss": 0.0006204976816661656
+    },
+    {
+      "epoch": 0.24509803921568626,
+      "grad_norm": 2.53125,
+      "kd_loss": 0.04509445279836655,
+      "learning_rate": 8.41328024722577e-06,
+      "loss": 0.0635,
+      "ss_p": 0.283,
+      "step": 850,
+      "student_loss": 0.009289605543017387,
+      "teacher_loss": 0.0005729056429117918
+    },
+    {
+      "epoch": 0.24653979238754326,
+      "grad_norm": 1.328125,
+      "kd_loss": 0.026419783011078835,
+      "learning_rate": 8.393508219485263e-06,
+      "loss": 0.062,
+      "ss_p": 0.2846666666666667,
+      "step": 855,
+      "student_loss": 0.00041390740079805255,
+      "teacher_loss": 0.00025891108089126647
+    },
+    {
+      "epoch": 0.24798154555940022,
+      "grad_norm": 4.0,
+      "kd_loss": 0.027005845680832863,
+      "learning_rate": 8.373637313299966e-06,
+      "loss": 0.074,
+      "ss_p": 0.28633333333333333,
+      "step": 860,
+      "student_loss": 0.0017632029484957457,
+      "teacher_loss": 0.0014858728973194957
+    },
+    {
+      "epoch": 0.2494232987312572,
+      "grad_norm": 3.5,
+      "kd_loss": 0.03546707332134247,
+      "learning_rate": 8.353668107658984e-06,
+      "loss": 0.0667,
+      "ss_p": 0.288,
+      "step": 865,
+      "student_loss": 0.000969983113463968,
+      "teacher_loss": 0.0006166775128804147
+    },
+    {
+      "epoch": 0.2508650519031142,
+      "grad_norm": 2.03125,
+      "kd_loss": 0.03606695681810379,
+      "learning_rate": 8.333601184415632e-06,
+      "loss": 0.0766,
+      "ss_p": 0.2896666666666667,
+      "step": 870,
+      "student_loss": 0.0029952137265354395,
+      "teacher_loss": 0.0022839908488094807
+    },
+    {
+      "epoch": 0.25230680507497116,
+      "grad_norm": 2.21875,
+      "kd_loss": 0.025676360353827477,
+      "learning_rate": 8.31343712827047e-06,
+      "loss": 0.0603,
+      "ss_p": 0.29133333333333333,
+      "step": 875,
+      "student_loss": 0.0007382369367405772,
+      "teacher_loss": 0.0004353248805273324
+    },
+    {
+      "epoch": 0.2537485582468281,
+      "grad_norm": 4.15625,
+      "kd_loss": 0.034421466290950775,
+      "learning_rate": 8.293176526754274e-06,
+      "loss": 0.0666,
+      "ss_p": 0.293,
+      "step": 880,
+      "student_loss": 0.004245162941515446,
+      "teacher_loss": 0.00153382308781147
+    },
+    {
+      "epoch": 0.25519031141868515,
+      "grad_norm": 2.8125,
+      "kd_loss": 0.07459811866283417,
+      "learning_rate": 8.27281997021091e-06,
+      "loss": 0.0555,
+      "ss_p": 0.2946666666666667,
+      "step": 885,
+      "student_loss": 0.04025157168507576,
+      "teacher_loss": 0.05041491240262985
+    },
+    {
+      "epoch": 0.2566320645905421,
+      "grad_norm": 2.40625,
+      "kd_loss": 0.020669063553214073,
+      "learning_rate": 8.25236805178014e-06,
+      "loss": 0.0643,
+      "ss_p": 0.29633333333333334,
+      "step": 890,
+      "student_loss": 0.0008314622682519257,
+      "teacher_loss": 0.0006236363551579416
+    },
+    {
+      "epoch": 0.25807381776239907,
+      "grad_norm": 4.125,
+      "kd_loss": 0.05192258208990097,
+      "learning_rate": 8.231821367380335e-06,
+      "loss": 0.0686,
+      "ss_p": 0.298,
+      "step": 895,
+      "student_loss": 0.07789942622184753,
+      "teacher_loss": 0.01009672973304987
+    },
+    {
+      "epoch": 0.25951557093425603,
+      "grad_norm": 3.03125,
+      "kd_loss": 0.03482178598642349,
+      "learning_rate": 8.211180515691108e-06,
+      "loss": 0.0723,
+      "ss_p": 0.2996666666666667,
+      "step": 900,
+      "student_loss": 0.06222911924123764,
+      "teacher_loss": 0.02422705851495266
+    },
+    {
+      "epoch": 0.26095732410611305,
+      "grad_norm": 1.8671875,
+      "kd_loss": 0.02488919161260128,
+      "learning_rate": 8.190446098135883e-06,
+      "loss": 0.0677,
+      "ss_p": 0.30133333333333334,
+      "step": 905,
+      "student_loss": 0.005900839809328318,
+      "teacher_loss": 0.004700648132711649
+    },
+    {
+      "epoch": 0.26239907727797,
+      "grad_norm": 3.109375,
+      "kd_loss": 0.018075549975037575,
+      "learning_rate": 8.16961871886435e-06,
+      "loss": 0.0633,
+      "ss_p": 0.303,
+      "step": 910,
+      "student_loss": 0.0010505489772185683,
+      "teacher_loss": 0.001017843489535153
+    },
+    {
+      "epoch": 0.263840830449827,
+      "grad_norm": 2.921875,
+      "kd_loss": 0.08231297880411148,
+      "learning_rate": 8.148698984734887e-06,
+      "loss": 0.0618,
+      "ss_p": 0.30466666666666664,
+      "step": 915,
+      "student_loss": 0.08048724383115768,
+      "teacher_loss": 0.0016023749485611916
+    },
+    {
+      "epoch": 0.26528258362168394,
+      "grad_norm": 2.3125,
+      "kd_loss": 0.024417152628302574,
+      "learning_rate": 8.127687505296852e-06,
+      "loss": 0.0676,
+      "ss_p": 0.30633333333333335,
+      "step": 920,
+      "student_loss": 0.018898461014032364,
+      "teacher_loss": 0.014232015237212181
+    },
+    {
+      "epoch": 0.26672433679354096,
+      "grad_norm": 5.3125,
+      "kd_loss": 0.019279776141047478,
+      "learning_rate": 8.106584892772844e-06,
+      "loss": 0.059,
+      "ss_p": 0.308,
+      "step": 925,
+      "student_loss": 0.00064336002105847,
+      "teacher_loss": 0.0005252147675491869
+    },
+    {
+      "epoch": 0.2681660899653979,
+      "grad_norm": 5.21875,
+      "kd_loss": 0.036257047206163406,
+      "learning_rate": 8.08539176204085e-06,
+      "loss": 0.0635,
+      "ss_p": 0.30966666666666665,
+      "step": 930,
+      "student_loss": 0.005282807629555464,
+      "teacher_loss": 0.001441403990611434
+    },
+    {
+      "epoch": 0.2696078431372549,
+      "grad_norm": 1.796875,
+      "kd_loss": 0.03358248993754387,
+      "learning_rate": 8.064108730616336e-06,
+      "loss": 0.0666,
+      "ss_p": 0.31133333333333335,
+      "step": 935,
+      "student_loss": 0.0055069089867174625,
+      "teacher_loss": 0.004551655147224665
+    },
+    {
+      "epoch": 0.2710495963091119,
+      "grad_norm": 2.4375,
+      "kd_loss": 0.055301859974861145,
+      "learning_rate": 8.042736418634252e-06,
+      "loss": 0.0684,
+      "ss_p": 0.313,
+      "step": 940,
+      "student_loss": 0.008183932863175869,
+      "teacher_loss": 0.0029486436396837234
+    },
+    {
+      "epoch": 0.27249134948096887,
+      "grad_norm": 1.8203125,
+      "kd_loss": 0.03048628568649292,
+      "learning_rate": 8.02127544883096e-06,
+      "loss": 0.0738,
+      "ss_p": 0.31466666666666665,
+      "step": 945,
+      "student_loss": 0.008925776928663254,
+      "teacher_loss": 0.0036037624813616276
+    },
+    {
+      "epoch": 0.27393310265282583,
+      "grad_norm": 2.921875,
+      "kd_loss": 0.03789523243904114,
+      "learning_rate": 7.999726446526095e-06,
+      "loss": 0.0668,
+      "ss_p": 0.31633333333333336,
+      "step": 950,
+      "student_loss": 0.0004505998804233968,
+      "teacher_loss": 0.0002629539812915027
+    },
+    {
+      "epoch": 0.2753748558246828,
+      "grad_norm": 2.203125,
+      "kd_loss": 0.033856287598609924,
+      "learning_rate": 7.978090039604342e-06,
+      "loss": 0.062,
+      "ss_p": 0.318,
+      "step": 955,
+      "student_loss": 0.0015073805116117,
+      "teacher_loss": 0.0004874717560596764
+    },
+    {
+      "epoch": 0.2768166089965398,
+      "grad_norm": 4.96875,
+      "kd_loss": 0.07367336750030518,
+      "learning_rate": 7.956366858497138e-06,
+      "loss": 0.0786,
+      "ss_p": 0.31966666666666665,
+      "step": 960,
+      "student_loss": 0.15931503474712372,
+      "teacher_loss": 0.002468140795826912
+    },
+    {
+      "epoch": 0.2782583621683968,
+      "grad_norm": 3.1875,
+      "kd_loss": 0.05005790293216705,
+      "learning_rate": 7.934557536164304e-06,
+      "loss": 0.0614,
+      "ss_p": 0.32133333333333336,
+      "step": 965,
+      "student_loss": 0.010769322514533997,
+      "teacher_loss": 0.006683016661554575
+    },
+    {
+      "epoch": 0.27970011534025374,
+      "grad_norm": 4.46875,
+      "kd_loss": 0.037505801767110825,
+      "learning_rate": 7.91266270807561e-06,
+      "loss": 0.0646,
+      "ss_p": 0.323,
+      "step": 970,
+      "student_loss": 0.00047873955918475986,
+      "teacher_loss": 0.00020865659462288022
+    },
+    {
+      "epoch": 0.2811418685121107,
+      "grad_norm": 2.578125,
+      "kd_loss": 0.026819568127393723,
+      "learning_rate": 7.890683012192249e-06,
+      "loss": 0.06,
+      "ss_p": 0.32466666666666666,
+      "step": 975,
+      "student_loss": 0.013693196699023247,
+      "teacher_loss": 0.0007407311932183802
+    },
+    {
+      "epoch": 0.2825836216839677,
+      "grad_norm": 3.984375,
+      "kd_loss": 0.040309686213731766,
+      "learning_rate": 7.868619088948246e-06,
+      "loss": 0.0616,
+      "ss_p": 0.3263333333333333,
+      "step": 980,
+      "student_loss": 0.0009199021733365953,
+      "teacher_loss": 0.00036276885657571256
+    },
+    {
+      "epoch": 0.2840253748558247,
+      "grad_norm": 3.171875,
+      "kd_loss": 0.033541709184646606,
+      "learning_rate": 7.846471581231814e-06,
+      "loss": 0.0701,
+      "ss_p": 0.328,
+      "step": 985,
+      "student_loss": 0.0004103538813069463,
+      "teacher_loss": 0.0003041589225176722
+    },
+    {
+      "epoch": 0.28546712802768165,
+      "grad_norm": 4.1875,
+      "kd_loss": 0.034742482006549835,
+      "learning_rate": 7.824241134366604e-06,
+      "loss": 0.0559,
+      "ss_p": 0.32966666666666666,
+      "step": 990,
+      "student_loss": 0.0004675070522353053,
+      "teacher_loss": 0.000328505237121135
+    },
+    {
+      "epoch": 0.2869088811995386,
+      "grad_norm": 2.75,
+      "kd_loss": 0.05086643248796463,
+      "learning_rate": 7.801928396092908e-06,
+      "loss": 0.0684,
+      "ss_p": 0.3313333333333333,
+      "step": 995,
+      "student_loss": 0.004713962785899639,
+      "teacher_loss": 0.003148426068946719
+    },
+    {
+      "epoch": 0.28835063437139563,
+      "grad_norm": 1.265625,
+      "kd_loss": 0.03605404496192932,
+      "learning_rate": 7.779534016548791e-06,
+      "loss": 0.0612,
+      "ss_p": 0.333,
+      "step": 1000,
+      "student_loss": 0.005091230850666761,
+      "teacher_loss": 0.00029147646273486316
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1500/added_tokens.json

@@ -0,0 +1,35 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|action_sep|>": 151670,
+  "<|arg_sep|>": 151671,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|call_sep|>": 151666,
+  "<|end_of_text|>": 151673,
+  "<|endoftext|>": 151643,
+  "<|exception_sep|>": 151669,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|frame_sep|>": 151672,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|latent_end|>": 151675,
+  "<|latent_start|>": 151674,
+  "<|line_sep|>": 151667,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|return_sep|>": 151668,
+  "<|trace_context_start|>": 151665,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1500/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1500/config.json

@@ -0,0 +1,66 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "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",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 36,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "pad_token_id": 151643,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "transformers_version": "4.57.6",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151676
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1500/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c9560c4722db8b2f116f989e3d21ba00b4cee51175b4cee41e920321af6556a
+size 6187858991

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1500/special_tokens_map.json

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

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1500/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f8f7b9a9e00da73442e32c377d551b32972d12377801224559282b59e3389ea
+size 16788033

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1500/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83a790d654474f5dfe225f889afd0210313eb1083f942671f2c4b8e95a1c922b
+size 11424004

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-1500/tokenizer_config.json

@@ -0,0 +1,295 @@
+{
+  "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": "<|trace_context_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151666": {
+      "content": "<|call_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151667": {
+      "content": "<|line_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151668": {
+      "content": "<|return_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151669": {
+      "content": "<|exception_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<|action_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<|arg_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151672": {
+      "content": "<|frame_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151673": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151674": {
+      "content": "<|latent_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151675": {
+      "content": "<|latent_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "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": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2000/added_tokens.json

@@ -0,0 +1,35 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|action_sep|>": 151670,
+  "<|arg_sep|>": 151671,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|call_sep|>": 151666,
+  "<|end_of_text|>": 151673,
+  "<|endoftext|>": 151643,
+  "<|exception_sep|>": 151669,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|frame_sep|>": 151672,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|latent_end|>": 151675,
+  "<|latent_start|>": 151674,
+  "<|line_sep|>": 151667,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|return_sep|>": 151668,
+  "<|trace_context_start|>": 151665,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2000/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2000/config.json

@@ -0,0 +1,66 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "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",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 36,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "pad_token_id": 151643,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "transformers_version": "4.57.6",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151676
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13753ff0f69fbcb580ced271c45ee4b0a261d0a0803a6585c9fe63f0bfccd692
+size 6187858991

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2000/special_tokens_map.json

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

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2000/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:066259fabbf158395b62a5587d6d742a745c46a9ad0029befba1e67654e8182d
+size 16788033

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2000/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83a790d654474f5dfe225f889afd0210313eb1083f942671f2c4b8e95a1c922b
+size 11424004

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2000/tokenizer_config.json

@@ -0,0 +1,295 @@
+{
+  "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": "<|trace_context_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151666": {
+      "content": "<|call_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151667": {
+      "content": "<|line_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151668": {
+      "content": "<|return_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151669": {
+      "content": "<|exception_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<|action_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<|arg_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151672": {
+      "content": "<|frame_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151673": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151674": {
+      "content": "<|latent_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151675": {
+      "content": "<|latent_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "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": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2500/added_tokens.json

@@ -0,0 +1,35 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|action_sep|>": 151670,
+  "<|arg_sep|>": 151671,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|call_sep|>": 151666,
+  "<|end_of_text|>": 151673,
+  "<|endoftext|>": 151643,
+  "<|exception_sep|>": 151669,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|frame_sep|>": 151672,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|latent_end|>": 151675,
+  "<|latent_start|>": 151674,
+  "<|line_sep|>": 151667,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|return_sep|>": 151668,
+  "<|trace_context_start|>": 151665,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2500/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2500/config.json

@@ -0,0 +1,66 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "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",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 36,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "pad_token_id": 151643,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "transformers_version": "4.57.6",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151676
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2500/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5342f6d5d1812683995e4ab098c2750f76a01fec131237b2eee0cc65cb33a30
+size 6187858991

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2500/special_tokens_map.json

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

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2500/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ad644efcf9cecaa52e23e4b7ba28eac2aa257c72911a4e1b09f25f6aa59d16f
+size 16788033

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2500/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83a790d654474f5dfe225f889afd0210313eb1083f942671f2c4b8e95a1c922b
+size 11424004

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-2500/tokenizer_config.json

@@ -0,0 +1,295 @@
+{
+  "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": "<|trace_context_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151666": {
+      "content": "<|call_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151667": {
+      "content": "<|line_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151668": {
+      "content": "<|return_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151669": {
+      "content": "<|exception_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<|action_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<|arg_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151672": {
+      "content": "<|frame_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151673": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151674": {
+      "content": "<|latent_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151675": {
+      "content": "<|latent_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "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": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-500/added_tokens.json

@@ -0,0 +1,35 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|action_sep|>": 151670,
+  "<|arg_sep|>": 151671,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|call_sep|>": 151666,
+  "<|end_of_text|>": 151673,
+  "<|endoftext|>": 151643,
+  "<|exception_sep|>": 151669,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|frame_sep|>": 151672,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|latent_end|>": 151675,
+  "<|latent_start|>": 151674,
+  "<|line_sep|>": 151667,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|return_sep|>": 151668,
+  "<|trace_context_start|>": 151665,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-500/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-500/config.json

@@ -0,0 +1,66 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "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",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 36,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "pad_token_id": 151643,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "transformers_version": "4.57.6",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151676
+}

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.50/checkpoint-500/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb456cf8e20f8a111e6fff29c59504015e22188782266c18142a111d5c5cc29e
+size 6187858991