add checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25

.gitattributes

@@ -72,3 +72,9 @@ checkpoints/codi3b_a1.0_b1.0_g1.0_ls1/checkpoint-1000/tokenizer.json filter=lfs
 checkpoints/codi3b_a1.0_b1.0_g1.0_ls1/checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/codi3b_a1.0_b1.0_g1.0_ls1/checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/codi3b_a1.0_b1.0_g1.0_ls1/wandb/offline-run-20260619_215302-txanrr5l/run-txanrr5l.wandb filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+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_ss0.25/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_ss0.25/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_ss0.25/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_ss0.25/checkpoint-1000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e129a8c98cff3d2f9ce47fbdd6b88c4e526b320167337f749b624ec782f9cd10
+size 6187858991

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/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_ss0.25/checkpoint-1000/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e771727a967d8ace9a7f4aee86389d2fa33e0772024f27c1f0e3e3b5f13b809
+size 16788033

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/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_ss0.25/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_ss0.25/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": 51.75,
+      "kd_loss": 0.2911630868911743,
+      "learning_rate": 4.444444444444445e-07,
+      "loss": 0.5076,
+      "ss_p": 0.0006666666666666666,
+      "step": 5,
+      "student_loss": 0.1302940994501114,
+      "teacher_loss": 0.0019674289505928755
+    },
+    {
+      "epoch": 0.0028835063437139563,
+      "grad_norm": 38.0,
+      "kd_loss": 0.270911306142807,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 0.4808,
+      "ss_p": 0.0015,
+      "step": 10,
+      "student_loss": 0.18397922813892365,
+      "teacher_loss": 0.005486903712153435
+    },
+    {
+      "epoch": 0.004325259515570935,
+      "grad_norm": 35.75,
+      "kd_loss": 0.2324424684047699,
+      "learning_rate": 1.5555555555555558e-06,
+      "loss": 0.4406,
+      "ss_p": 0.0023333333333333335,
+      "step": 15,
+      "student_loss": 0.053055763244628906,
+      "teacher_loss": 0.001329739228822291
+    },
+    {
+      "epoch": 0.0057670126874279125,
+      "grad_norm": 27.875,
+      "kd_loss": 0.2010677456855774,
+      "learning_rate": 2.1111111111111114e-06,
+      "loss": 0.3963,
+      "ss_p": 0.0031666666666666666,
+      "step": 20,
+      "student_loss": 0.07125679403543472,
+      "teacher_loss": 0.0005395798943936825
+    },
+    {
+      "epoch": 0.00720876585928489,
+      "grad_norm": 13.125,
+      "kd_loss": 0.20564113557338715,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 0.319,
+      "ss_p": 0.004,
+      "step": 25,
+      "student_loss": 0.022935954853892326,
+      "teacher_loss": 0.0008126357570290565
+    },
+    {
+      "epoch": 0.00865051903114187,
+      "grad_norm": 8.5,
+      "kd_loss": 0.17410063743591309,
+      "learning_rate": 3.2222222222222227e-06,
+      "loss": 0.273,
+      "ss_p": 0.004833333333333334,
+      "step": 30,
+      "student_loss": 0.023023078218102455,
+      "teacher_loss": 0.0003572013520170003
+    },
+    {
+      "epoch": 0.010092272202998846,
+      "grad_norm": 5.3125,
+      "kd_loss": 0.14901787042617798,
+      "learning_rate": 3.777777777777778e-06,
+      "loss": 0.2393,
+      "ss_p": 0.005666666666666667,
+      "step": 35,
+      "student_loss": 0.003114640014246106,
+      "teacher_loss": 0.0002823838731274009
+    },
+    {
+      "epoch": 0.011534025374855825,
+      "grad_norm": 3.515625,
+      "kd_loss": 0.11224491149187088,
+      "learning_rate": 4.333333333333334e-06,
+      "loss": 0.1954,
+      "ss_p": 0.0065,
+      "step": 40,
+      "student_loss": 0.047119662165641785,
+      "teacher_loss": 0.04035693407058716
+    },
+    {
+      "epoch": 0.012975778546712802,
+      "grad_norm": 3.234375,
+      "kd_loss": 0.20430822670459747,
+      "learning_rate": 4.888888888888889e-06,
+      "loss": 0.1575,
+      "ss_p": 0.007333333333333333,
+      "step": 45,
+      "student_loss": 0.09196822345256805,
+      "teacher_loss": 0.002917634788900614
+    },
+    {
+      "epoch": 0.01441753171856978,
+      "grad_norm": 2.421875,
+      "kd_loss": 0.0694398581981659,
+      "learning_rate": 5.444444444444445e-06,
+      "loss": 0.1548,
+      "ss_p": 0.008166666666666666,
+      "step": 50,
+      "student_loss": 0.01751815713942051,
+      "teacher_loss": 0.007385170552879572
+    },
+    {
+      "epoch": 0.015859284890426758,
+      "grad_norm": 2.453125,
+      "kd_loss": 0.06267786026000977,
+      "learning_rate": 6e-06,
+      "loss": 0.1343,
+      "ss_p": 0.009,
+      "step": 55,
+      "student_loss": 0.001116996631026268,
+      "teacher_loss": 0.0005169984651729465
+    },
+    {
+      "epoch": 0.01730103806228374,
+      "grad_norm": 4.1875,
+      "kd_loss": 0.05475896596908569,
+      "learning_rate": 6.555555555555556e-06,
+      "loss": 0.1184,
+      "ss_p": 0.009833333333333333,
+      "step": 60,
+      "student_loss": 0.0025303340516984463,
+      "teacher_loss": 0.0013067334657534957
+    },
+    {
+      "epoch": 0.018742791234140715,
+      "grad_norm": 3.390625,
+      "kd_loss": 0.05040474236011505,
+      "learning_rate": 7.111111111111112e-06,
+      "loss": 0.1116,
+      "ss_p": 0.010666666666666666,
+      "step": 65,
+      "student_loss": 0.026373304426670074,
+      "teacher_loss": 0.03748839348554611
+    },
+    {
+      "epoch": 0.020184544405997693,
+      "grad_norm": 1.7109375,
+      "kd_loss": 0.09240958839654922,
+      "learning_rate": 7.666666666666667e-06,
+      "loss": 0.1025,
+      "ss_p": 0.0115,
+      "step": 70,
+      "student_loss": 0.01682574301958084,
+      "teacher_loss": 0.0015897674020379782
+    },
+    {
+      "epoch": 0.02162629757785467,
+      "grad_norm": 1.9296875,
+      "kd_loss": 0.0344117097556591,
+      "learning_rate": 8.222222222222222e-06,
+      "loss": 0.0931,
+      "ss_p": 0.012333333333333333,
+      "step": 75,
+      "student_loss": 0.0018068131757900119,
+      "teacher_loss": 0.0010662946151569486
+    },
+    {
+      "epoch": 0.02306805074971165,
+      "grad_norm": 2.125,
+      "kd_loss": 0.03444109857082367,
+      "learning_rate": 8.777777777777778e-06,
+      "loss": 0.0823,
+      "ss_p": 0.013166666666666667,
+      "step": 80,
+      "student_loss": 0.0033705250825732946,
+      "teacher_loss": 0.0015141806798055768
+    },
+    {
+      "epoch": 0.024509803921568627,
+      "grad_norm": 1.609375,
+      "kd_loss": 0.030533254146575928,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.082,
+      "ss_p": 0.014,
+      "step": 85,
+      "student_loss": 0.0010508184786885977,
+      "teacher_loss": 0.0009611450368538499
+    },
+    {
+      "epoch": 0.025951557093425604,
+      "grad_norm": 1.1796875,
+      "kd_loss": 0.08606406301259995,
+      "learning_rate": 9.88888888888889e-06,
+      "loss": 0.0772,
+      "ss_p": 0.014833333333333334,
+      "step": 90,
+      "student_loss": 0.027494383975863457,
+      "teacher_loss": 0.003187393769621849
+    },
+    {
+      "epoch": 0.027393310265282585,
+      "grad_norm": 1.8203125,
+      "kd_loss": 0.04909161105751991,
+      "learning_rate": 9.99995337991255e-06,
+      "loss": 0.0691,
+      "ss_p": 0.015666666666666666,
+      "step": 95,
+      "student_loss": 0.008900301530957222,
+      "teacher_loss": 0.005448868032544851
+    },
+    {
+      "epoch": 0.02883506343713956,
+      "grad_norm": 1.3046875,
+      "kd_loss": 0.023117493838071823,
+      "learning_rate": 9.999763987297266e-06,
+      "loss": 0.0613,
+      "ss_p": 0.0165,
+      "step": 100,
+      "student_loss": 0.0008615201222710311,
+      "teacher_loss": 0.000606034358497709
+    },
+    {
+      "epoch": 0.03027681660899654,
+      "grad_norm": 1.6953125,
+      "kd_loss": 0.03902491554617882,
+      "learning_rate": 9.999428913912876e-06,
+      "loss": 0.0638,
+      "ss_p": 0.017333333333333333,
+      "step": 105,
+      "student_loss": 0.003911722917109728,
+      "teacher_loss": 0.0003194632881786674
+    },
+    {
+      "epoch": 0.031718569780853516,
+      "grad_norm": 1.0,
+      "kd_loss": 0.04485023394227028,
+      "learning_rate": 9.998948169522587e-06,
+      "loss": 0.064,
+      "ss_p": 0.018166666666666668,
+      "step": 110,
+      "student_loss": 0.03466537222266197,
+      "teacher_loss": 0.01840282790362835
+    },
+    {
+      "epoch": 0.03316032295271049,
+      "grad_norm": 1.1015625,
+      "kd_loss": 0.04261879250407219,
+      "learning_rate": 9.998321768134101e-06,
+      "loss": 0.0618,
+      "ss_p": 0.019,
+      "step": 115,
+      "student_loss": 0.0023938817903399467,
+      "teacher_loss": 0.00034946223604492843
+    },
+    {
+      "epoch": 0.03460207612456748,
+      "grad_norm": 1.140625,
+      "kd_loss": 0.027965571731328964,
+      "learning_rate": 9.997549727999212e-06,
+      "loss": 0.0578,
+      "ss_p": 0.019833333333333335,
+      "step": 120,
+      "student_loss": 0.0012657151091843843,
+      "teacher_loss": 0.0008060845430009067
+    },
+    {
+      "epoch": 0.036043829296424454,
+      "grad_norm": 2.53125,
+      "kd_loss": 0.063721664249897,
+      "learning_rate": 9.996632071613257e-06,
+      "loss": 0.0634,
+      "ss_p": 0.020666666666666667,
+      "step": 125,
+      "student_loss": 0.03624747321009636,
+      "teacher_loss": 0.0003536214353516698
+    },
+    {
+      "epoch": 0.03748558246828143,
+      "grad_norm": 1.1875,
+      "kd_loss": 0.027761410921812057,
+      "learning_rate": 9.995568825714479e-06,
+      "loss": 0.059,
+      "ss_p": 0.0215,
+      "step": 130,
+      "student_loss": 0.0006298073567450047,
+      "teacher_loss": 0.000541882065590471
+    },
+    {
+      "epoch": 0.03892733564013841,
+      "grad_norm": 1.3125,
+      "kd_loss": 0.044490400701761246,
+      "learning_rate": 9.994360021283235e-06,
+      "loss": 0.0639,
+      "ss_p": 0.022333333333333334,
+      "step": 135,
+      "student_loss": 0.002881347667425871,
+      "teacher_loss": 0.0007115196785889566
+    },
+    {
+      "epoch": 0.040369088811995385,
+      "grad_norm": 1.78125,
+      "kd_loss": 0.023633815348148346,
+      "learning_rate": 9.993005693541098e-06,
+      "loss": 0.0604,
+      "ss_p": 0.023166666666666665,
+      "step": 140,
+      "student_loss": 0.0006920672603882849,
+      "teacher_loss": 0.0006185604142956436
+    },
+    {
+      "epoch": 0.04181084198385236,
+      "grad_norm": 0.87890625,
+      "kd_loss": 0.08394590765237808,
+      "learning_rate": 9.991505881949837e-06,
+      "loss": 0.0528,
+      "ss_p": 0.024,
+      "step": 145,
+      "student_loss": 0.06726142764091492,
+      "teacher_loss": 0.006664975546300411
+    },
+    {
+      "epoch": 0.04325259515570934,
+      "grad_norm": 0.94140625,
+      "kd_loss": 0.015645405277609825,
+      "learning_rate": 9.989860630210251e-06,
+      "loss": 0.0519,
+      "ss_p": 0.024833333333333332,
+      "step": 150,
+      "student_loss": 0.00325228413566947,
+      "teacher_loss": 0.001607410958968103
+    },
+    {
+      "epoch": 0.04469434832756632,
+      "grad_norm": 1.6171875,
+      "kd_loss": 0.02038741670548916,
+      "learning_rate": 9.98806998626091e-06,
+      "loss": 0.0605,
+      "ss_p": 0.025666666666666667,
+      "step": 155,
+      "student_loss": 0.0034302759449929,
+      "teacher_loss": 0.0009208041010424495
+    },
+    {
+      "epoch": 0.0461361014994233,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.03190923109650612,
+      "learning_rate": 9.98613400227676e-06,
+      "loss": 0.062,
+      "ss_p": 0.0265,
+      "step": 160,
+      "student_loss": 0.08172117918729782,
+      "teacher_loss": 0.045917265117168427
+    },
+    {
+      "epoch": 0.04757785467128028,
+      "grad_norm": 1.0,
+      "kd_loss": 0.030859047546982765,
+      "learning_rate": 9.984052734667586e-06,
+      "loss": 0.0532,
+      "ss_p": 0.027333333333333334,
+      "step": 165,
+      "student_loss": 0.00046156172174960375,
+      "teacher_loss": 0.004423642531037331
+    },
+    {
+      "epoch": 0.049019607843137254,
+      "grad_norm": 2.625,
+      "kd_loss": 0.01684083230793476,
+      "learning_rate": 9.981826244076387e-06,
+      "loss": 0.0582,
+      "ss_p": 0.028166666666666666,
+      "step": 170,
+      "student_loss": 0.0006269018049351871,
+      "teacher_loss": 0.000417911825934425
+    },
+    {
+      "epoch": 0.05046136101499423,
+      "grad_norm": 2.125,
+      "kd_loss": 0.02710077352821827,
+      "learning_rate": 9.979454595377594e-06,
+      "loss": 0.0541,
+      "ss_p": 0.029,
+      "step": 175,
+      "student_loss": 0.0026831587310880423,
+      "teacher_loss": 0.0005185891059227288
+    },
+    {
+      "epoch": 0.05190311418685121,
+      "grad_norm": 1.625,
+      "kd_loss": 0.03340298682451248,
+      "learning_rate": 9.97693785767519e-06,
+      "loss": 0.052,
+      "ss_p": 0.029833333333333333,
+      "step": 180,
+      "student_loss": 0.0029260488227009773,
+      "teacher_loss": 0.0007292699883691967
+    },
+    {
+      "epoch": 0.05334486735870819,
+      "grad_norm": 1.2734375,
+      "kd_loss": 0.03773834928870201,
+      "learning_rate": 9.9742761043007e-06,
+      "loss": 0.0572,
+      "ss_p": 0.030666666666666665,
+      "step": 185,
+      "student_loss": 0.03489382192492485,
+      "teacher_loss": 0.0003711420577019453
+    },
+    {
+      "epoch": 0.05478662053056517,
+      "grad_norm": 1.5859375,
+      "kd_loss": 0.05151769891381264,
+      "learning_rate": 9.971469412811032e-06,
+      "loss": 0.0514,
+      "ss_p": 0.0315,
+      "step": 190,
+      "student_loss": 0.004275215789675713,
+      "teacher_loss": 0.0015893314266577363
+    },
+    {
+      "epoch": 0.056228373702422146,
+      "grad_norm": 2.046875,
+      "kd_loss": 0.04925917834043503,
+      "learning_rate": 9.968517864986247e-06,
+      "loss": 0.0516,
+      "ss_p": 0.03233333333333333,
+      "step": 195,
+      "student_loss": 0.012135050259530544,
+      "teacher_loss": 0.00039562693564221263
+    },
+    {
+      "epoch": 0.05767012687427912,
+      "grad_norm": 2.078125,
+      "kd_loss": 0.03382336348295212,
+      "learning_rate": 9.965421546827153e-06,
+      "loss": 0.0598,
+      "ss_p": 0.033166666666666664,
+      "step": 200,
+      "student_loss": 0.0005725009832531214,
+      "teacher_loss": 0.0002131070359610021
+    },
+    {
+      "epoch": 0.0591118800461361,
+      "grad_norm": 1.4375,
+      "kd_loss": 0.0258716382086277,
+      "learning_rate": 9.962180548552812e-06,
+      "loss": 0.0578,
+      "ss_p": 0.034,
+      "step": 205,
+      "student_loss": 0.03533850237727165,
+      "teacher_loss": 0.027285216376185417
+    },
+    {
+      "epoch": 0.06055363321799308,
+      "grad_norm": 2.234375,
+      "kd_loss": 0.0328000970184803,
+      "learning_rate": 9.958794964597906e-06,
+      "loss": 0.0509,
+      "ss_p": 0.034833333333333334,
+      "step": 210,
+      "student_loss": 0.0008364069508388638,
+      "teacher_loss": 0.00042736949399113655
+    },
+    {
+      "epoch": 0.061995386389850055,
+      "grad_norm": 2.59375,
+      "kd_loss": 0.03772606700658798,
+      "learning_rate": 9.955264893609986e-06,
+      "loss": 0.0526,
+      "ss_p": 0.035666666666666666,
+      "step": 215,
+      "student_loss": 0.0005249156965874135,
+      "teacher_loss": 0.00020741825574077666
+    },
+    {
+      "epoch": 0.06343713956170703,
+      "grad_norm": 1.9609375,
+      "kd_loss": 0.04148948937654495,
+      "learning_rate": 9.951590438446597e-06,
+      "loss": 0.0497,
+      "ss_p": 0.0365,
+      "step": 220,
+      "student_loss": 0.03238594904541969,
+      "teacher_loss": 0.00991231668740511
+    },
+    {
+      "epoch": 0.06487889273356401,
+      "grad_norm": 2.484375,
+      "kd_loss": 0.02161458693444729,
+      "learning_rate": 9.947771706172285e-06,
+      "loss": 0.0608,
+      "ss_p": 0.037333333333333336,
+      "step": 225,
+      "student_loss": 0.04591212794184685,
+      "teacher_loss": 0.0021177390590310097
+    },
+    {
+      "epoch": 0.06632064590542099,
+      "grad_norm": 2.34375,
+      "kd_loss": 0.039198894053697586,
+      "learning_rate": 9.943808808055472e-06,
+      "loss": 0.0512,
+      "ss_p": 0.03816666666666667,
+      "step": 230,
+      "student_loss": 0.006647516507655382,
+      "teacher_loss": 0.003550174878910184
+    },
+    {
+      "epoch": 0.06776239907727798,
+      "grad_norm": 1.859375,
+      "kd_loss": 0.07571706175804138,
+      "learning_rate": 9.93970185956522e-06,
+      "loss": 0.0541,
+      "ss_p": 0.039,
+      "step": 235,
+      "student_loss": 0.02396474778652191,
+      "teacher_loss": 0.0004969856818206608
+    },
+    {
+      "epoch": 0.06920415224913495,
+      "grad_norm": 1.40625,
+      "kd_loss": 0.02857440896332264,
+      "learning_rate": 9.93545098036786e-06,
+      "loss": 0.054,
+      "ss_p": 0.03983333333333333,
+      "step": 240,
+      "student_loss": 0.0027293721213936806,
+      "teacher_loss": 0.00035359253524802625
+    },
+    {
+      "epoch": 0.07064590542099193,
+      "grad_norm": 1.6640625,
+      "kd_loss": 0.01553173828870058,
+      "learning_rate": 9.931056294323507e-06,
+      "loss": 0.0515,
+      "ss_p": 0.04066666666666666,
+      "step": 245,
+      "student_loss": 0.0008876210777089,
+      "teacher_loss": 0.0013351637171581388
+    },
+    {
+      "epoch": 0.07208765859284891,
+      "grad_norm": 1.7734375,
+      "kd_loss": 0.019031953066587448,
+      "learning_rate": 9.926517929482454e-06,
+      "loss": 0.0494,
+      "ss_p": 0.0415,
+      "step": 250,
+      "student_loss": 0.0006975256255827844,
+      "teacher_loss": 0.0003186521353200078
+    },
+    {
+      "epoch": 0.07352941176470588,
+      "grad_norm": 1.9921875,
+      "kd_loss": 0.015427670441567898,
+      "learning_rate": 9.921836018081438e-06,
+      "loss": 0.0497,
+      "ss_p": 0.042333333333333334,
+      "step": 255,
+      "student_loss": 0.00035640658461488783,
+      "teacher_loss": 0.0003770627954509109
+    },
+    {
+      "epoch": 0.07497116493656286,
+      "grad_norm": 1.796875,
+      "kd_loss": 0.027072442695498466,
+      "learning_rate": 9.917010696539793e-06,
+      "loss": 0.0553,
+      "ss_p": 0.043166666666666666,
+      "step": 260,
+      "student_loss": 0.03946039080619812,
+      "teacher_loss": 0.03156239539384842
+    },
+    {
+      "epoch": 0.07641291810841984,
+      "grad_norm": 1.890625,
+      "kd_loss": 0.01610456593334675,
+      "learning_rate": 9.912042105455462e-06,
+      "loss": 0.0564,
+      "ss_p": 0.044,
+      "step": 265,
+      "student_loss": 0.0003752708435058594,
+      "teacher_loss": 0.00027317472267895937
+    },
+    {
+      "epoch": 0.07785467128027682,
+      "grad_norm": 2.453125,
+      "kd_loss": 0.016607360914349556,
+      "learning_rate": 9.906930389600917e-06,
+      "loss": 0.0568,
+      "ss_p": 0.044833333333333336,
+      "step": 270,
+      "student_loss": 0.0018301332602277398,
+      "teacher_loss": 0.0015998801682144403
+    },
+    {
+      "epoch": 0.07929642445213379,
+      "grad_norm": 2.5,
+      "kd_loss": 0.024908045306801796,
+      "learning_rate": 9.901675697918925e-06,
+      "loss": 0.0515,
+      "ss_p": 0.04566666666666667,
+      "step": 275,
+      "student_loss": 0.04913569986820221,
+      "teacher_loss": 0.035234954208135605
+    },
+    {
+      "epoch": 0.08073817762399077,
+      "grad_norm": 3.28125,
+      "kd_loss": 0.034790776669979095,
+      "learning_rate": 9.896278183518216e-06,
+      "loss": 0.0507,
+      "ss_p": 0.0465,
+      "step": 280,
+      "student_loss": 0.0006408383487723768,
+      "teacher_loss": 0.00018051423830911517
+    },
+    {
+      "epoch": 0.08217993079584775,
+      "grad_norm": 1.9921875,
+      "kd_loss": 0.021553825587034225,
+      "learning_rate": 9.890738003669029e-06,
+      "loss": 0.0597,
+      "ss_p": 0.04733333333333333,
+      "step": 285,
+      "student_loss": 0.0006896372651681304,
+      "teacher_loss": 0.0004881985078100115
+    },
+    {
+      "epoch": 0.08362168396770472,
+      "grad_norm": 1.4453125,
+      "kd_loss": 0.017040127888321877,
+      "learning_rate": 9.885055319798514e-06,
+      "loss": 0.0663,
+      "ss_p": 0.04816666666666667,
+      "step": 290,
+      "student_loss": 0.015433781780302525,
+      "teacher_loss": 0.011667763814330101
+    },
+    {
+      "epoch": 0.0850634371395617,
+      "grad_norm": 1.8203125,
+      "kd_loss": 0.02004189044237137,
+      "learning_rate": 9.879230297486034e-06,
+      "loss": 0.0589,
+      "ss_p": 0.049,
+      "step": 295,
+      "student_loss": 0.0008422060054726899,
+      "teacher_loss": 0.0006135343573987484
+    },
+    {
+      "epoch": 0.08650519031141868,
+      "grad_norm": 1.8125,
+      "kd_loss": 0.05030286684632301,
+      "learning_rate": 9.873263106458353e-06,
+      "loss": 0.0549,
+      "ss_p": 0.049833333333333334,
+      "step": 300,
+      "student_loss": 0.017038553953170776,
+      "teacher_loss": 0.0017766954842954874
+    },
+    {
+      "epoch": 0.08794694348327567,
+      "grad_norm": 2.5625,
+      "kd_loss": 0.02125021256506443,
+      "learning_rate": 9.86715392058467e-06,
+      "loss": 0.05,
+      "ss_p": 0.050666666666666665,
+      "step": 305,
+      "student_loss": 0.011004108935594559,
+      "teacher_loss": 0.003803006373345852
+    },
+    {
+      "epoch": 0.08938869665513265,
+      "grad_norm": 3.53125,
+      "kd_loss": 0.018712202087044716,
+      "learning_rate": 9.860902917871566e-06,
+      "loss": 0.0526,
+      "ss_p": 0.0515,
+      "step": 310,
+      "student_loss": 0.000736552057787776,
+      "teacher_loss": 0.0005247860099188983
+    },
+    {
+      "epoch": 0.09083044982698962,
+      "grad_norm": 3.421875,
+      "kd_loss": 0.04219638183712959,
+      "learning_rate": 9.854510280457819e-06,
+      "loss": 0.0545,
+      "ss_p": 0.052333333333333336,
+      "step": 315,
+      "student_loss": 0.010509521700441837,
+      "teacher_loss": 0.0003759938117582351
+    },
+    {
+      "epoch": 0.0922722029988466,
+      "grad_norm": 2.859375,
+      "kd_loss": 0.02967497706413269,
+      "learning_rate": 9.847976194609083e-06,
+      "loss": 0.0631,
+      "ss_p": 0.05316666666666667,
+      "step": 320,
+      "student_loss": 0.0005802118103019893,
+      "teacher_loss": 0.000306545669445768
+    },
+    {
+      "epoch": 0.09371395617070358,
+      "grad_norm": 3.796875,
+      "kd_loss": 0.07338711619377136,
+      "learning_rate": 9.841300850712479e-06,
+      "loss": 0.0571,
+      "ss_p": 0.054,
+      "step": 325,
+      "student_loss": 0.01456360425800085,
+      "teacher_loss": 0.021586118265986443
+    },
+    {
+      "epoch": 0.09515570934256055,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.052099090069532394,
+      "learning_rate": 9.834484443271035e-06,
+      "loss": 0.057,
+      "ss_p": 0.05483333333333333,
+      "step": 330,
+      "student_loss": 0.0252448208630085,
+      "teacher_loss": 0.005709249060600996
+    },
+    {
+      "epoch": 0.09659746251441753,
+      "grad_norm": 2.265625,
+      "kd_loss": 0.01595858670771122,
+      "learning_rate": 9.827527170898019e-06,
+      "loss": 0.0498,
+      "ss_p": 0.05566666666666667,
+      "step": 335,
+      "student_loss": 0.0009765715803951025,
+      "teacher_loss": 0.0011430901940912008
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 2.21875,
+      "kd_loss": 0.02486472763121128,
+      "learning_rate": 9.820429236311158e-06,
+      "loss": 0.0569,
+      "ss_p": 0.0565,
+      "step": 340,
+      "student_loss": 0.01074217353016138,
+      "teacher_loss": 0.006791106890887022
+    },
+    {
+      "epoch": 0.09948096885813149,
+      "grad_norm": 2.359375,
+      "kd_loss": 0.03829123079776764,
+      "learning_rate": 9.81319084632673e-06,
+      "loss": 0.0592,
+      "ss_p": 0.05733333333333333,
+      "step": 345,
+      "student_loss": 0.014017965644598007,
+      "teacher_loss": 0.0004889628035016358
+    },
+    {
+      "epoch": 0.10092272202998846,
+      "grad_norm": 2.546875,
+      "kd_loss": 0.05929192528128624,
+      "learning_rate": 9.805812211853534e-06,
+      "loss": 0.0487,
+      "ss_p": 0.058166666666666665,
+      "step": 350,
+      "student_loss": 0.0034538835752755404,
+      "teacher_loss": 0.0020060648676007986
+    },
+    {
+      "epoch": 0.10236447520184544,
+      "grad_norm": 1.9609375,
+      "kd_loss": 0.01714949496090412,
+      "learning_rate": 9.798293547886748e-06,
+      "loss": 0.0633,
+      "ss_p": 0.059,
+      "step": 355,
+      "student_loss": 0.0006743644480593503,
+      "teacher_loss": 0.0005169557989574969
+    },
+    {
+      "epoch": 0.10380622837370242,
+      "grad_norm": 1.65625,
+      "kd_loss": 0.017548151314258575,
+      "learning_rate": 9.790635073501662e-06,
+      "loss": 0.0542,
+      "ss_p": 0.059833333333333336,
+      "step": 360,
+      "student_loss": 0.0008041374385356903,
+      "teacher_loss": 0.0005842456594109535
+    },
+    {
+      "epoch": 0.1052479815455594,
+      "grad_norm": 1.6171875,
+      "kd_loss": 0.029959363862872124,
+      "learning_rate": 9.782837011847294e-06,
+      "loss": 0.0575,
+      "ss_p": 0.06066666666666667,
+      "step": 365,
+      "student_loss": 0.04379178583621979,
+      "teacher_loss": 0.028138814494013786
+    },
+    {
+      "epoch": 0.10668973471741638,
+      "grad_norm": 3.328125,
+      "kd_loss": 0.05084262043237686,
+      "learning_rate": 9.774899590139897e-06,
+      "loss": 0.0644,
+      "ss_p": 0.0615,
+      "step": 370,
+      "student_loss": 0.024329576641321182,
+      "teacher_loss": 0.0006152085261419415
+    },
+    {
+      "epoch": 0.10813148788927336,
+      "grad_norm": 3.375,
+      "kd_loss": 0.03078492544591427,
+      "learning_rate": 9.766823039656328e-06,
+      "loss": 0.0549,
+      "ss_p": 0.06233333333333333,
+      "step": 375,
+      "student_loss": 0.0323534831404686,
+      "teacher_loss": 0.020907964557409286
+    },
+    {
+      "epoch": 0.10957324106113034,
+      "grad_norm": 2.515625,
+      "kd_loss": 0.019612271338701248,
+      "learning_rate": 9.758607595727309e-06,
+      "loss": 0.0555,
+      "ss_p": 0.06316666666666666,
+      "step": 380,
+      "student_loss": 0.0007233266951516271,
+      "teacher_loss": 0.0004633954958990216
+    },
+    {
+      "epoch": 0.11101499423298732,
+      "grad_norm": 2.28125,
+      "kd_loss": 0.03037400171160698,
+      "learning_rate": 9.75025349773058e-06,
+      "loss": 0.05,
+      "ss_p": 0.064,
+      "step": 385,
+      "student_loss": 0.0013653574278578162,
+      "teacher_loss": 0.0010349624790251255
+    },
+    {
+      "epoch": 0.11245674740484429,
+      "grad_norm": 1.4609375,
+      "kd_loss": 0.017816657200455666,
+      "learning_rate": 9.741760989083914e-06,
+      "loss": 0.0467,
+      "ss_p": 0.06483333333333334,
+      "step": 390,
+      "student_loss": 0.0003158455074299127,
+      "teacher_loss": 0.00028364264289848506
+    },
+    {
+      "epoch": 0.11389850057670127,
+      "grad_norm": 1.1953125,
+      "kd_loss": 0.018483463674783707,
+      "learning_rate": 9.73313031723803e-06,
+      "loss": 0.0528,
+      "ss_p": 0.06566666666666666,
+      "step": 395,
+      "student_loss": 0.000659606303088367,
+      "teacher_loss": 0.0004989048466086388
+    },
+    {
+      "epoch": 0.11534025374855825,
+      "grad_norm": 1.4609375,
+      "kd_loss": 0.027655545622110367,
+      "learning_rate": 9.724361733669383e-06,
+      "loss": 0.0457,
+      "ss_p": 0.0665,
+      "step": 400,
+      "student_loss": 0.0011931798653677106,
+      "teacher_loss": 0.00031232176115736365
+    },
+    {
+      "epoch": 0.11678200692041522,
+      "grad_norm": 2.1875,
+      "kd_loss": 0.026839647442102432,
+      "learning_rate": 9.715455493872827e-06,
+      "loss": 0.0546,
+      "ss_p": 0.06733333333333333,
+      "step": 405,
+      "student_loss": 0.011741007678210735,
+      "teacher_loss": 0.00042350406874902546
+    },
+    {
+      "epoch": 0.1182237600922722,
+      "grad_norm": 1.234375,
+      "kd_loss": 0.03331433981657028,
+      "learning_rate": 9.70641185735419e-06,
+      "loss": 0.0581,
+      "ss_p": 0.06816666666666667,
+      "step": 410,
+      "student_loss": 0.014031303115189075,
+      "teacher_loss": 0.008188456296920776
+    },
+    {
+      "epoch": 0.11966551326412918,
+      "grad_norm": 1.4453125,
+      "kd_loss": 0.024215172976255417,
+      "learning_rate": 9.697231087622691e-06,
+      "loss": 0.0571,
+      "ss_p": 0.069,
+      "step": 415,
+      "student_loss": 0.005363308358937502,
+      "teacher_loss": 0.0043652174063026905
+    },
+    {
+      "epoch": 0.12110726643598616,
+      "grad_norm": 1.9375,
+      "kd_loss": 0.039297956973314285,
+      "learning_rate": 9.687913452183277e-06,
+      "loss": 0.0538,
+      "ss_p": 0.06983333333333333,
+      "step": 420,
+      "student_loss": 0.04513926804065704,
+      "teacher_loss": 0.02523217350244522
+    },
+    {
+      "epoch": 0.12254901960784313,
+      "grad_norm": 1.6171875,
+      "kd_loss": 0.014175890013575554,
+      "learning_rate": 9.678459222528825e-06,
+      "loss": 0.048,
+      "ss_p": 0.07066666666666667,
+      "step": 425,
+      "student_loss": 0.0021354646887630224,
+      "teacher_loss": 0.002133190631866455
+    },
+    {
+      "epoch": 0.12399077277970011,
+      "grad_norm": 3.21875,
+      "kd_loss": 0.050583723932504654,
+      "learning_rate": 9.668868674132224e-06,
+      "loss": 0.0602,
+      "ss_p": 0.0715,
+      "step": 430,
+      "student_loss": 0.04505128040909767,
+      "teacher_loss": 0.059542153030633926
+    },
+    {
+      "epoch": 0.1254325259515571,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.01723223365843296,
+      "learning_rate": 9.65914208643836e-06,
+      "loss": 0.0501,
+      "ss_p": 0.07233333333333333,
+      "step": 435,
+      "student_loss": 0.0008684992790222168,
+      "teacher_loss": 0.0006412562215700746
+    },
+    {
+      "epoch": 0.12687427912341406,
+      "grad_norm": 1.3671875,
+      "kd_loss": 0.03685420751571655,
+      "learning_rate": 9.649279742855964e-06,
+      "loss": 0.0534,
+      "ss_p": 0.07316666666666667,
+      "step": 440,
+      "student_loss": 0.013619957491755486,
+      "teacher_loss": 0.007216887082904577
+    },
+    {
+      "epoch": 0.12831603229527105,
+      "grad_norm": 1.34375,
+      "kd_loss": 0.02925006113946438,
+      "learning_rate": 9.639281930749363e-06,
+      "loss": 0.0519,
+      "ss_p": 0.074,
+      "step": 445,
+      "student_loss": 0.0380571112036705,
+      "teacher_loss": 0.01941097341477871
+    },
+    {
+      "epoch": 0.12975778546712802,
+      "grad_norm": 2.296875,
+      "kd_loss": 0.03925434127449989,
+      "learning_rate": 9.629148941430096e-06,
+      "loss": 0.0645,
+      "ss_p": 0.07483333333333334,
+      "step": 450,
+      "student_loss": 0.0027915090322494507,
+      "teacher_loss": 0.003204664448276162
+    },
+    {
+      "epoch": 0.131199538638985,
+      "grad_norm": 1.265625,
+      "kd_loss": 0.02644190937280655,
+      "learning_rate": 9.618881070148436e-06,
+      "loss": 0.0593,
+      "ss_p": 0.07566666666666666,
+      "step": 455,
+      "student_loss": 0.001614358858205378,
+      "teacher_loss": 0.0007554752519354224
+    },
+    {
+      "epoch": 0.13264129181084197,
+      "grad_norm": 1.8828125,
+      "kd_loss": 0.03925043344497681,
+      "learning_rate": 9.608478616084784e-06,
+      "loss": 0.0493,
+      "ss_p": 0.0765,
+      "step": 460,
+      "student_loss": 0.008012035861611366,
+      "teacher_loss": 0.0013571040472015738
+    },
+    {
+      "epoch": 0.13408304498269896,
+      "grad_norm": 1.8046875,
+      "kd_loss": 0.03139708191156387,
+      "learning_rate": 9.597941882340947e-06,
+      "loss": 0.056,
+      "ss_p": 0.07733333333333334,
+      "step": 465,
+      "student_loss": 0.0007810476236045361,
+      "teacher_loss": 0.000705737853422761
+    },
+    {
+      "epoch": 0.13552479815455595,
+      "grad_norm": 10.5625,
+      "kd_loss": 0.02820807509124279,
+      "learning_rate": 9.587271175931313e-06,
+      "loss": 0.0548,
+      "ss_p": 0.07816666666666666,
+      "step": 470,
+      "student_loss": 0.0006233889725990593,
+      "teacher_loss": 0.009306474588811398
+    },
+    {
+      "epoch": 0.13696655132641292,
+      "grad_norm": 5.4375,
+      "kd_loss": 0.03380393236875534,
+      "learning_rate": 9.5764668077739e-06,
+      "loss": 0.0547,
+      "ss_p": 0.079,
+      "step": 475,
+      "student_loss": 0.0008260513423010707,
+      "teacher_loss": 0.0004428833199199289
+    },
+    {
+      "epoch": 0.1384083044982699,
+      "grad_norm": 1.9140625,
+      "kd_loss": 0.03291467949748039,
+      "learning_rate": 9.5655290926813e-06,
+      "loss": 0.0731,
+      "ss_p": 0.07983333333333334,
+      "step": 480,
+      "student_loss": 0.027618875727057457,
+      "teacher_loss": 0.02329031378030777
+    },
+    {
+      "epoch": 0.13985005767012687,
+      "grad_norm": 2.34375,
+      "kd_loss": 0.01908169314265251,
+      "learning_rate": 9.554458349351507e-06,
+      "loss": 0.0565,
+      "ss_p": 0.08066666666666666,
+      "step": 485,
+      "student_loss": 0.0010344330221414566,
+      "teacher_loss": 0.0014974202495068312
+    },
+    {
+      "epoch": 0.14129181084198386,
+      "grad_norm": 3.21875,
+      "kd_loss": 0.02358153834939003,
+      "learning_rate": 9.54325490035863e-06,
+      "loss": 0.0605,
+      "ss_p": 0.0815,
+      "step": 490,
+      "student_loss": 0.0019428008235991001,
+      "teacher_loss": 0.0019422578625380993
+    },
+    {
+      "epoch": 0.14273356401384082,
+      "grad_norm": 1.9375,
+      "kd_loss": 0.03615574538707733,
+      "learning_rate": 9.531919072143488e-06,
+      "loss": 0.0678,
+      "ss_p": 0.08233333333333333,
+      "step": 495,
+      "student_loss": 0.004113656468689442,
+      "teacher_loss": 0.0020528335589915514
+    },
+    {
+      "epoch": 0.14417531718569782,
+      "grad_norm": 1.859375,
+      "kd_loss": 0.02917010523378849,
+      "learning_rate": 9.520451195004107e-06,
+      "loss": 0.0549,
+      "ss_p": 0.08316666666666667,
+      "step": 500,
+      "student_loss": 0.0012471633963286877,
+      "teacher_loss": 0.0007156505598686635
+    },
+    {
+      "epoch": 0.14561707035755478,
+      "grad_norm": 3.796875,
+      "kd_loss": 0.028955919668078423,
+      "learning_rate": 9.508851603086094e-06,
+      "loss": 0.0756,
+      "ss_p": 0.084,
+      "step": 505,
+      "student_loss": 0.00103476969525218,
+      "teacher_loss": 0.0007179073872976005
+    },
+    {
+      "epoch": 0.14705882352941177,
+      "grad_norm": 4.3125,
+      "kd_loss": 0.03181787207722664,
+      "learning_rate": 9.497120634372894e-06,
+      "loss": 0.0652,
+      "ss_p": 0.08483333333333333,
+      "step": 510,
+      "student_loss": 0.0008358138729818165,
+      "teacher_loss": 0.0009078999282792211
+    },
+    {
+      "epoch": 0.14850057670126873,
+      "grad_norm": 2.75,
+      "kd_loss": 0.030279649421572685,
+      "learning_rate": 9.48525863067595e-06,
+      "loss": 0.0561,
+      "ss_p": 0.08566666666666667,
+      "step": 515,
+      "student_loss": 0.01584705524146557,
+      "teacher_loss": 0.0004912562435492873
+    },
+    {
+      "epoch": 0.14994232987312572,
+      "grad_norm": 1.8125,
+      "kd_loss": 0.09092690050601959,
+      "learning_rate": 9.473265937624748e-06,
+      "loss": 0.0652,
+      "ss_p": 0.0865,
+      "step": 520,
+      "student_loss": 0.04334317892789841,
+      "teacher_loss": 0.01329742930829525
+    },
+    {
+      "epoch": 0.1513840830449827,
+      "grad_norm": 1.9609375,
+      "kd_loss": 0.035351403057575226,
+      "learning_rate": 9.461142904656722e-06,
+      "loss": 0.0676,
+      "ss_p": 0.08733333333333333,
+      "step": 525,
+      "student_loss": 0.0008091186173260212,
+      "teacher_loss": 0.0003106557414866984
+    },
+    {
+      "epoch": 0.15282583621683968,
+      "grad_norm": 3.265625,
+      "kd_loss": 0.03477087989449501,
+      "learning_rate": 9.448889885007105e-06,
+      "loss": 0.0664,
+      "ss_p": 0.08816666666666667,
+      "step": 530,
+      "student_loss": 0.000741212978027761,
+      "teacher_loss": 0.0006058380822651088
+    },
+    {
+      "epoch": 0.15426758938869667,
+      "grad_norm": 1.9453125,
+      "kd_loss": 0.04042341187596321,
+      "learning_rate": 9.436507235698613e-06,
+      "loss": 0.067,
+      "ss_p": 0.089,
+      "step": 535,
+      "student_loss": 0.001875737914815545,
+      "teacher_loss": 0.0009398771217092872
+    },
+    {
+      "epoch": 0.15570934256055363,
+      "grad_norm": 2.375,
+      "kd_loss": 0.052647072821855545,
+      "learning_rate": 9.423995317531049e-06,
+      "loss": 0.0685,
+      "ss_p": 0.08983333333333333,
+      "step": 540,
+      "student_loss": 0.018440421670675278,
+      "teacher_loss": 0.0006669810973107815
+    },
+    {
+      "epoch": 0.15715109573241062,
+      "grad_norm": 1.8203125,
+      "kd_loss": 0.02273302711546421,
+      "learning_rate": 9.4113544950708e-06,
+      "loss": 0.0592,
+      "ss_p": 0.09066666666666667,
+      "step": 545,
+      "student_loss": 0.004481257405132055,
+      "teacher_loss": 0.00299642700701952
+    },
+    {
+      "epoch": 0.15859284890426759,
+      "grad_norm": 2.21875,
+      "kd_loss": 0.06994060426950455,
+      "learning_rate": 9.398585136640195e-06,
+      "loss": 0.06,
+      "ss_p": 0.0915,
+      "step": 550,
+      "student_loss": 0.0002918424434028566,
+      "teacher_loss": 0.0043050614185631275
+    },
+    {
+      "epoch": 0.16003460207612458,
+      "grad_norm": 2.609375,
+      "kd_loss": 0.09039658308029175,
+      "learning_rate": 9.38568761430679e-06,
+      "loss": 0.0645,
+      "ss_p": 0.09233333333333334,
+      "step": 555,
+      "student_loss": 0.03315426781773567,
+      "teacher_loss": 0.0027825473807752132
+    },
+    {
+      "epoch": 0.16147635524798154,
+      "grad_norm": 2.4375,
+      "kd_loss": 0.04773836210370064,
+      "learning_rate": 9.372662303872514e-06,
+      "loss": 0.0699,
+      "ss_p": 0.09316666666666666,
+      "step": 560,
+      "student_loss": 0.0019605644047260284,
+      "teacher_loss": 0.00033071250072680414
+    },
+    {
+      "epoch": 0.16291810841983853,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.055092114955186844,
+      "learning_rate": 9.359509584862735e-06,
+      "loss": 0.0722,
+      "ss_p": 0.094,
+      "step": 565,
+      "student_loss": 0.06662841886281967,
+      "teacher_loss": 0.021180005744099617
+    },
+    {
+      "epoch": 0.1643598615916955,
+      "grad_norm": 2.25,
+      "kd_loss": 0.03327203169465065,
+      "learning_rate": 9.346229840515186e-06,
+      "loss": 0.0715,
+      "ss_p": 0.09483333333333334,
+      "step": 570,
+      "student_loss": 0.011727867648005486,
+      "teacher_loss": 0.007846700958907604
+    },
+    {
+      "epoch": 0.16580161476355249,
+      "grad_norm": 1.640625,
+      "kd_loss": 0.03131795674562454,
+      "learning_rate": 9.332823457768805e-06,
+      "loss": 0.0632,
+      "ss_p": 0.09566666666666666,
+      "step": 575,
+      "student_loss": 0.0012770607136189938,
+      "teacher_loss": 0.0006532660336233675
+    },
+    {
+      "epoch": 0.16724336793540945,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.02394150011241436,
+      "learning_rate": 9.31929082725246e-06,
+      "loss": 0.0559,
+      "ss_p": 0.0965,
+      "step": 580,
+      "student_loss": 0.0018843002617359161,
+      "teacher_loss": 0.002092719078063965
+    },
+    {
+      "epoch": 0.16868512110726644,
+      "grad_norm": 1.375,
+      "kd_loss": 0.030440978705883026,
+      "learning_rate": 9.305632343273567e-06,
+      "loss": 0.0504,
+      "ss_p": 0.09733333333333333,
+      "step": 585,
+      "student_loss": 0.0007624205318279564,
+      "teacher_loss": 0.00039506866596639156
+    },
+    {
+      "epoch": 0.1701268742791234,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.024257397279143333,
+      "learning_rate": 9.291848403806604e-06,
+      "loss": 0.0564,
+      "ss_p": 0.09816666666666667,
+      "step": 590,
+      "student_loss": 0.0005771798314526677,
+      "teacher_loss": 0.0008993460214696825
+    },
+    {
+      "epoch": 0.1715686274509804,
+      "grad_norm": 4.125,
+      "kd_loss": 0.020785460248589516,
+      "learning_rate": 9.277939410481507e-06,
+      "loss": 0.0662,
+      "ss_p": 0.099,
+      "step": 595,
+      "student_loss": 0.0006834843079559505,
+      "teacher_loss": 0.0006025595939718187
+    },
+    {
+      "epoch": 0.17301038062283736,
+      "grad_norm": 4.3125,
+      "kd_loss": 0.04758751019835472,
+      "learning_rate": 9.263905768571977e-06,
+      "loss": 0.0681,
+      "ss_p": 0.09983333333333333,
+      "step": 600,
+      "student_loss": 0.04168753698468208,
+      "teacher_loss": 0.004258541855961084
+    },
+    {
+      "epoch": 0.17445213379469435,
+      "grad_norm": 3.578125,
+      "kd_loss": 0.05026453733444214,
+      "learning_rate": 9.249747886983664e-06,
+      "loss": 0.0608,
+      "ss_p": 0.10066666666666667,
+      "step": 605,
+      "student_loss": 0.12914426624774933,
+      "teacher_loss": 0.0332721583545208
+    },
+    {
+      "epoch": 0.17589388696655134,
+      "grad_norm": 1.5625,
+      "kd_loss": 0.04372970014810562,
+      "learning_rate": 9.235466178242255e-06,
+      "loss": 0.0553,
+      "ss_p": 0.1015,
+      "step": 610,
+      "student_loss": 0.015209294855594635,
+      "teacher_loss": 0.0004905562964268029
+    },
+    {
+      "epoch": 0.1773356401384083,
+      "grad_norm": 1.5703125,
+      "kd_loss": 0.022922828793525696,
+      "learning_rate": 9.221061058481456e-06,
+      "loss": 0.0521,
+      "ss_p": 0.10233333333333333,
+      "step": 615,
+      "student_loss": 0.0005627846694551408,
+      "teacher_loss": 0.0006025371258147061
+    },
+    {
+      "epoch": 0.1787773933102653,
+      "grad_norm": 3.09375,
+      "kd_loss": 0.035977646708488464,
+      "learning_rate": 9.206532947430865e-06,
+      "loss": 0.0733,
+      "ss_p": 0.10316666666666667,
+      "step": 620,
+      "student_loss": 0.010981826111674309,
+      "teacher_loss": 0.0073308078572154045
+    },
+    {
+      "epoch": 0.18021914648212226,
+      "grad_norm": 4.28125,
+      "kd_loss": 0.027282772585749626,
+      "learning_rate": 9.191882268403743e-06,
+      "loss": 0.0504,
+      "ss_p": 0.104,
+      "step": 625,
+      "student_loss": 0.013904711231589317,
+      "teacher_loss": 0.004958369769155979
+    },
+    {
+      "epoch": 0.18166089965397925,
+      "grad_norm": 2.421875,
+      "kd_loss": 0.0420038215816021,
+      "learning_rate": 9.177109448284673e-06,
+      "loss": 0.0551,
+      "ss_p": 0.10483333333333333,
+      "step": 630,
+      "student_loss": 0.0015013068914413452,
+      "teacher_loss": 0.000932888186071068
+    },
+    {
+      "epoch": 0.1831026528258362,
+      "grad_norm": 1.5703125,
+      "kd_loss": 0.031230047345161438,
+      "learning_rate": 9.162214917517139e-06,
+      "loss": 0.0816,
+      "ss_p": 0.10566666666666667,
+      "step": 635,
+      "student_loss": 0.0012656699400395155,
+      "teacher_loss": 0.0003248720895498991
+    },
+    {
+      "epoch": 0.1845444059976932,
+      "grad_norm": 1.8359375,
+      "kd_loss": 0.036213044077157974,
+      "learning_rate": 9.14719911009096e-06,
+      "loss": 0.059,
+      "ss_p": 0.1065,
+      "step": 640,
+      "student_loss": 0.0015696294140070677,
+      "teacher_loss": 0.0015752669423818588
+    },
+    {
+      "epoch": 0.18598615916955016,
+      "grad_norm": 2.40625,
+      "kd_loss": 0.04032283276319504,
+      "learning_rate": 9.132062463529666e-06,
+      "loss": 0.0596,
+      "ss_p": 0.10733333333333334,
+      "step": 645,
+      "student_loss": 0.00094438815722242,
+      "teacher_loss": 0.0006934748380444944
+    },
+    {
+      "epoch": 0.18742791234140715,
+      "grad_norm": 3.875,
+      "kd_loss": 0.021463755518198013,
+      "learning_rate": 9.11680541887774e-06,
+      "loss": 0.0562,
+      "ss_p": 0.10816666666666666,
+      "step": 650,
+      "student_loss": 0.01918548159301281,
+      "teacher_loss": 0.015398788265883923
+    },
+    {
+      "epoch": 0.18886966551326412,
+      "grad_norm": 2.78125,
+      "kd_loss": 0.04955601319670677,
+      "learning_rate": 9.101428420687759e-06,
+      "loss": 0.0589,
+      "ss_p": 0.109,
+      "step": 655,
+      "student_loss": 0.0006307393778115511,
+      "teacher_loss": 0.00035819882759824395
+    },
+    {
+      "epoch": 0.1903114186851211,
+      "grad_norm": 2.078125,
+      "kd_loss": 0.021997841075062752,
+      "learning_rate": 9.085931917007464e-06,
+      "loss": 0.0617,
+      "ss_p": 0.10983333333333334,
+      "step": 660,
+      "student_loss": 0.0008303869399242103,
+      "teacher_loss": 0.0007437756285071373
+    },
+    {
+      "epoch": 0.19175317185697807,
+      "grad_norm": 1.8828125,
+      "kd_loss": 0.0701814517378807,
+      "learning_rate": 9.070316359366681e-06,
+      "loss": 0.0623,
+      "ss_p": 0.11066666666666666,
+      "step": 665,
+      "student_loss": 0.00640520267188549,
+      "teacher_loss": 0.0016227789456024766
+    },
+    {
+      "epoch": 0.19319492502883506,
+      "grad_norm": 2.0625,
+      "kd_loss": 0.0384909063577652,
+      "learning_rate": 9.054582202764175e-06,
+      "loss": 0.0591,
+      "ss_p": 0.1115,
+      "step": 670,
+      "student_loss": 0.024686485528945923,
+      "teacher_loss": 0.010944215580821037
+    },
+    {
+      "epoch": 0.19463667820069205,
+      "grad_norm": 2.53125,
+      "kd_loss": 0.019254416227340698,
+      "learning_rate": 9.038729905654393e-06,
+      "loss": 0.0635,
+      "ss_p": 0.11233333333333333,
+      "step": 675,
+      "student_loss": 0.0009956193389371037,
+      "teacher_loss": 0.0009586027590557933
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 3.171875,
+      "kd_loss": 0.06988031417131424,
+      "learning_rate": 9.022759929934109e-06,
+      "loss": 0.0576,
+      "ss_p": 0.11316666666666667,
+      "step": 680,
+      "student_loss": 0.10042662918567657,
+      "teacher_loss": 0.037429697811603546
+    },
+    {
+      "epoch": 0.197520184544406,
+      "grad_norm": 2.875,
+      "kd_loss": 0.023996418341994286,
+      "learning_rate": 9.006672740928952e-06,
+      "loss": 0.0567,
+      "ss_p": 0.114,
+      "step": 685,
+      "student_loss": 0.03923623636364937,
+      "teacher_loss": 0.030981319025158882
+    },
+    {
+      "epoch": 0.19896193771626297,
+      "grad_norm": 1.5078125,
+      "kd_loss": 0.05567062273621559,
+      "learning_rate": 8.990468807379861e-06,
+      "loss": 0.0566,
+      "ss_p": 0.11483333333333333,
+      "step": 690,
+      "student_loss": 0.0040661352686584,
+      "teacher_loss": 0.011099128983914852
+    },
+    {
+      "epoch": 0.20040369088811996,
+      "grad_norm": 3.859375,
+      "kd_loss": 0.04607200622558594,
+      "learning_rate": 8.974148601429426e-06,
+      "loss": 0.0649,
+      "ss_p": 0.11566666666666667,
+      "step": 695,
+      "student_loss": 0.0025057813618332148,
+      "teacher_loss": 0.0075830984860658646
+    },
+    {
+      "epoch": 0.20184544405997693,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.024678010493516922,
+      "learning_rate": 8.957712598608123e-06,
+      "loss": 0.0556,
+      "ss_p": 0.1165,
+      "step": 700,
+      "student_loss": 0.0005524636944755912,
+      "teacher_loss": 0.0003281493845861405
+    },
+    {
+      "epoch": 0.20328719723183392,
+      "grad_norm": 2.75,
+      "kd_loss": 0.01949162222445011,
+      "learning_rate": 8.941161277820462e-06,
+      "loss": 0.0587,
+      "ss_p": 0.11733333333333333,
+      "step": 705,
+      "student_loss": 0.002285210881382227,
+      "teacher_loss": 0.0009470953955315053
+    },
+    {
+      "epoch": 0.20472895040369088,
+      "grad_norm": 1.4375,
+      "kd_loss": 0.049595337361097336,
+      "learning_rate": 8.924495121331038e-06,
+      "loss": 0.0597,
+      "ss_p": 0.11816666666666667,
+      "step": 710,
+      "student_loss": 0.004098474979400635,
+      "teacher_loss": 0.0049895416013896465
+    },
+    {
+      "epoch": 0.20617070357554787,
+      "grad_norm": 2.09375,
+      "kd_loss": 0.027050882577896118,
+      "learning_rate": 8.907714614750473e-06,
+      "loss": 0.0619,
+      "ss_p": 0.119,
+      "step": 715,
+      "student_loss": 0.002861611545085907,
+      "teacher_loss": 0.0046501075848937035
+    },
+    {
+      "epoch": 0.20761245674740483,
+      "grad_norm": 2.96875,
+      "kd_loss": 0.025670425966382027,
+      "learning_rate": 8.890820247021263e-06,
+      "loss": 0.0632,
+      "ss_p": 0.11983333333333333,
+      "step": 720,
+      "student_loss": 0.001188535476103425,
+      "teacher_loss": 0.0028166016563773155
+    },
+    {
+      "epoch": 0.20905420991926182,
+      "grad_norm": 1.640625,
+      "kd_loss": 0.018999872729182243,
+      "learning_rate": 8.873812510403545e-06,
+      "loss": 0.0475,
+      "ss_p": 0.12066666666666667,
+      "step": 725,
+      "student_loss": 0.0006456644041463733,
+      "teacher_loss": 0.0005741088534705341
+    },
+    {
+      "epoch": 0.2104959630911188,
+      "grad_norm": 3.421875,
+      "kd_loss": 0.04065852612257004,
+      "learning_rate": 8.85669190046074e-06,
+      "loss": 0.0611,
+      "ss_p": 0.1215,
+      "step": 730,
+      "student_loss": 0.004662148654460907,
+      "teacher_loss": 0.0029930805321782827
+    },
+    {
+      "epoch": 0.21193771626297578,
+      "grad_norm": 3.046875,
+      "kd_loss": 0.031192727386951447,
+      "learning_rate": 8.839458916045117e-06,
+      "loss": 0.0646,
+      "ss_p": 0.12233333333333334,
+      "step": 735,
+      "student_loss": 0.000892693642526865,
+      "teacher_loss": 0.00016416113066952676
+    },
+    {
+      "epoch": 0.21337946943483277,
+      "grad_norm": 1.890625,
+      "kd_loss": 0.036736395210027695,
+      "learning_rate": 8.822114059283267e-06,
+      "loss": 0.0521,
+      "ss_p": 0.12316666666666666,
+      "step": 740,
+      "student_loss": 0.0019187863217666745,
+      "teacher_loss": 0.00022182970133144408
+    },
+    {
+      "epoch": 0.21482122260668973,
+      "grad_norm": 1.6015625,
+      "kd_loss": 0.026628436520695686,
+      "learning_rate": 8.804657835561456e-06,
+      "loss": 0.0551,
+      "ss_p": 0.124,
+      "step": 745,
+      "student_loss": 0.04293574392795563,
+      "teacher_loss": 0.01928584836423397
+    },
+    {
+      "epoch": 0.21626297577854672,
+      "grad_norm": 2.765625,
+      "kd_loss": 0.06700080633163452,
+      "learning_rate": 8.787090753510917e-06,
+      "loss": 0.0528,
+      "ss_p": 0.12483333333333334,
+      "step": 750,
+      "student_loss": 0.03903910145163536,
+      "teacher_loss": 0.005492281634360552
+    },
+    {
+      "epoch": 0.2177047289504037,
+      "grad_norm": 2.8125,
+      "kd_loss": 0.026075949892401695,
+      "learning_rate": 8.76941332499301e-06,
+      "loss": 0.0665,
+      "ss_p": 0.12566666666666668,
+      "step": 755,
+      "student_loss": 0.03722135350108147,
+      "teacher_loss": 0.00845416635274887
+    },
+    {
+      "epoch": 0.21914648212226068,
+      "grad_norm": 1.9921875,
+      "kd_loss": 0.0730191171169281,
+      "learning_rate": 8.751626065084328e-06,
+      "loss": 0.0513,
+      "ss_p": 0.1265,
+      "step": 760,
+      "student_loss": 0.03827729821205139,
+      "teacher_loss": 0.001533320639282465
+    },
+    {
+      "epoch": 0.22058823529411764,
+      "grad_norm": 1.9453125,
+      "kd_loss": 0.07695702463388443,
+      "learning_rate": 8.733729492061675e-06,
+      "loss": 0.0678,
+      "ss_p": 0.12733333333333333,
+      "step": 765,
+      "student_loss": 0.0033891198690980673,
+      "teacher_loss": 0.0018518184078857303
+    },
+    {
+      "epoch": 0.22202998846597463,
+      "grad_norm": 2.296875,
+      "kd_loss": 0.0629807561635971,
+      "learning_rate": 8.715724127386971e-06,
+      "loss": 0.0562,
+      "ss_p": 0.12816666666666668,
+      "step": 770,
+      "student_loss": 0.009104887954890728,
+      "teacher_loss": 0.003206773428246379
+    },
+    {
+      "epoch": 0.2234717416378316,
+      "grad_norm": 1.8671875,
+      "kd_loss": 0.04077409580349922,
+      "learning_rate": 8.697610495692055e-06,
+      "loss": 0.0569,
+      "ss_p": 0.129,
+      "step": 775,
+      "student_loss": 0.0035035060718655586,
+      "teacher_loss": 0.002162383636459708
+    },
+    {
+      "epoch": 0.22491349480968859,
+      "grad_norm": 2.671875,
+      "kd_loss": 0.04463198035955429,
+      "learning_rate": 8.679389124763396e-06,
+      "loss": 0.0692,
+      "ss_p": 0.12983333333333333,
+      "step": 780,
+      "student_loss": 0.000573027296923101,
+      "teacher_loss": 0.0003768337774090469
+    },
+    {
+      "epoch": 0.22635524798154555,
+      "grad_norm": 2.171875,
+      "kd_loss": 0.02057073451578617,
+      "learning_rate": 8.661060545526724e-06,
+      "loss": 0.0579,
+      "ss_p": 0.13066666666666665,
+      "step": 785,
+      "student_loss": 0.00046177307376638055,
+      "teacher_loss": 0.0009386134333908558
+    },
+    {
+      "epoch": 0.22779700115340254,
+      "grad_norm": 1.7578125,
+      "kd_loss": 0.0165251474827528,
+      "learning_rate": 8.64262529203155e-06,
+      "loss": 0.0561,
+      "ss_p": 0.1315,
+      "step": 790,
+      "student_loss": 0.00909686740487814,
+      "teacher_loss": 0.007585103157907724
+    },
+    {
+      "epoch": 0.2292387543252595,
+      "grad_norm": 2.59375,
+      "kd_loss": 0.036473266780376434,
+      "learning_rate": 8.624083901435613e-06,
+      "loss": 0.0665,
+      "ss_p": 0.13233333333333333,
+      "step": 795,
+      "student_loss": 0.001641886425204575,
+      "teacher_loss": 0.0007439031032845378
+    },
+    {
+      "epoch": 0.2306805074971165,
+      "grad_norm": 2.234375,
+      "kd_loss": 0.07317814230918884,
+      "learning_rate": 8.60543691398922e-06,
+      "loss": 0.07,
+      "ss_p": 0.13316666666666666,
+      "step": 800,
+      "student_loss": 0.14681710302829742,
+      "teacher_loss": 0.0409475602209568
+    },
+    {
+      "epoch": 0.23212226066897348,
+      "grad_norm": 1.6640625,
+      "kd_loss": 0.021199576556682587,
+      "learning_rate": 8.586684873019513e-06,
+      "loss": 0.065,
+      "ss_p": 0.134,
+      "step": 805,
+      "student_loss": 0.0005692061386071146,
+      "teacher_loss": 0.0005178118008188903
+    },
+    {
+      "epoch": 0.23356401384083045,
+      "grad_norm": 37.75,
+      "kd_loss": 0.04357064887881279,
+      "learning_rate": 8.567828324914633e-06,
+      "loss": 0.0747,
+      "ss_p": 0.13483333333333333,
+      "step": 810,
+      "student_loss": 0.0022272157948464155,
+      "teacher_loss": 0.0013909294502809644
+    },
+    {
+      "epoch": 0.23500576701268744,
+      "grad_norm": 3.125,
+      "kd_loss": 0.02284885011613369,
+      "learning_rate": 8.548867819107803e-06,
+      "loss": 0.062,
+      "ss_p": 0.13566666666666666,
+      "step": 815,
+      "student_loss": 0.001657223328948021,
+      "teacher_loss": 0.0122824190184474
+    },
+    {
+      "epoch": 0.2364475201845444,
+      "grad_norm": 2.875,
+      "kd_loss": 0.051225535571575165,
+      "learning_rate": 8.52980390806131e-06,
+      "loss": 0.0624,
+      "ss_p": 0.1365,
+      "step": 820,
+      "student_loss": 0.005345536395907402,
+      "teacher_loss": 0.0022601906675845385
+    },
+    {
+      "epoch": 0.2378892733564014,
+      "grad_norm": 2.28125,
+      "kd_loss": 0.0469605028629303,
+      "learning_rate": 8.51063714725042e-06,
+      "loss": 0.0613,
+      "ss_p": 0.13733333333333334,
+      "step": 825,
+      "student_loss": 0.0024987931828945875,
+      "teacher_loss": 0.017313949763774872
+    },
+    {
+      "epoch": 0.23933102652825836,
+      "grad_norm": 3.0625,
+      "kd_loss": 0.04128275439143181,
+      "learning_rate": 8.49136809514719e-06,
+      "loss": 0.0602,
+      "ss_p": 0.13816666666666666,
+      "step": 830,
+      "student_loss": 0.021648574620485306,
+      "teacher_loss": 0.008686312474310398
+    },
+    {
+      "epoch": 0.24077277970011535,
+      "grad_norm": 2.90625,
+      "kd_loss": 0.0319555401802063,
+      "learning_rate": 8.471997313204183e-06,
+      "loss": 0.0531,
+      "ss_p": 0.139,
+      "step": 835,
+      "student_loss": 0.013425268232822418,
+      "teacher_loss": 0.005469009280204773
+    },
+    {
+      "epoch": 0.2422145328719723,
+      "grad_norm": 3.515625,
+      "kd_loss": 0.045133091509342194,
+      "learning_rate": 8.452525365838123e-06,
+      "loss": 0.0587,
+      "ss_p": 0.13983333333333334,
+      "step": 840,
+      "student_loss": 0.0007193881901912391,
+      "teacher_loss": 0.00027729570865631104
+    },
+    {
+      "epoch": 0.2436562860438293,
+      "grad_norm": 4.0,
+      "kd_loss": 0.02394530363380909,
+      "learning_rate": 8.432952820413447e-06,
+      "loss": 0.0792,
+      "ss_p": 0.14066666666666666,
+      "step": 845,
+      "student_loss": 0.0009191186982207,
+      "teacher_loss": 0.0006061987951397896
+    },
+    {
+      "epoch": 0.24509803921568626,
+      "grad_norm": 2.546875,
+      "kd_loss": 0.04272513836622238,
+      "learning_rate": 8.41328024722577e-06,
+      "loss": 0.0642,
+      "ss_p": 0.1415,
+      "step": 850,
+      "student_loss": 0.010309615172445774,
+      "teacher_loss": 0.0007157528307288885
+    },
+    {
+      "epoch": 0.24653979238754326,
+      "grad_norm": 2.3125,
+      "kd_loss": 0.03682299330830574,
+      "learning_rate": 8.393508219485263e-06,
+      "loss": 0.0614,
+      "ss_p": 0.14233333333333334,
+      "step": 855,
+      "student_loss": 0.0003967257507611066,
+      "teacher_loss": 0.00033078939304687083
+    },
+    {
+      "epoch": 0.24798154555940022,
+      "grad_norm": 4.75,
+      "kd_loss": 0.034355249255895615,
+      "learning_rate": 8.373637313299966e-06,
+      "loss": 0.0762,
+      "ss_p": 0.14316666666666666,
+      "step": 860,
+      "student_loss": 0.0028368483763188124,
+      "teacher_loss": 0.0024731862358748913
+    },
+    {
+      "epoch": 0.2494232987312572,
+      "grad_norm": 3.0,
+      "kd_loss": 0.040522415190935135,
+      "learning_rate": 8.353668107658984e-06,
+      "loss": 0.0652,
+      "ss_p": 0.144,
+      "step": 865,
+      "student_loss": 0.0005781393847428262,
+      "teacher_loss": 0.0006893461686559021
+    },
+    {
+      "epoch": 0.2508650519031142,
+      "grad_norm": 1.84375,
+      "kd_loss": 0.0422118715941906,
+      "learning_rate": 8.333601184415632e-06,
+      "loss": 0.0663,
+      "ss_p": 0.14483333333333334,
+      "step": 870,
+      "student_loss": 0.006354041863232851,
+      "teacher_loss": 0.003884020959958434
+    },
+    {
+      "epoch": 0.25230680507497116,
+      "grad_norm": 2.375,
+      "kd_loss": 0.02563890814781189,
+      "learning_rate": 8.31343712827047e-06,
+      "loss": 0.0605,
+      "ss_p": 0.14566666666666667,
+      "step": 875,
+      "student_loss": 0.000642551458440721,
+      "teacher_loss": 0.0004050025308970362
+    },
+    {
+      "epoch": 0.2537485582468281,
+      "grad_norm": 3.390625,
+      "kd_loss": 0.03512357175350189,
+      "learning_rate": 8.293176526754274e-06,
+      "loss": 0.066,
+      "ss_p": 0.1465,
+      "step": 880,
+      "student_loss": 0.0049323407001793385,
+      "teacher_loss": 0.0015710614388808608
+    },
+    {
+      "epoch": 0.25519031141868515,
+      "grad_norm": 1.875,
+      "kd_loss": 0.07576354593038559,
+      "learning_rate": 8.27281997021091e-06,
+      "loss": 0.0517,
+      "ss_p": 0.14733333333333334,
+      "step": 885,
+      "student_loss": 0.05895135924220085,
+      "teacher_loss": 0.047839391976594925
+    },
+    {
+      "epoch": 0.2566320645905421,
+      "grad_norm": 2.453125,
+      "kd_loss": 0.01908315159380436,
+      "learning_rate": 8.25236805178014e-06,
+      "loss": 0.0638,
+      "ss_p": 0.14816666666666667,
+      "step": 890,
+      "student_loss": 0.0004396491567604244,
+      "teacher_loss": 0.0004117630305700004
+    },
+    {
+      "epoch": 0.25807381776239907,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.04943661764264107,
+      "learning_rate": 8.231821367380335e-06,
+      "loss": 0.0603,
+      "ss_p": 0.149,
+      "step": 895,
+      "student_loss": 0.0752253606915474,
+      "teacher_loss": 0.009776663035154343
+    },
+    {
+      "epoch": 0.25951557093425603,
+      "grad_norm": 2.28125,
+      "kd_loss": 0.035139672458171844,
+      "learning_rate": 8.211180515691108e-06,
+      "loss": 0.0669,
+      "ss_p": 0.14983333333333335,
+      "step": 900,
+      "student_loss": 0.031900305300951004,
+      "teacher_loss": 0.02295292727649212
+    },
+    {
+      "epoch": 0.26095732410611305,
+      "grad_norm": 2.390625,
+      "kd_loss": 0.024371637031435966,
+      "learning_rate": 8.190446098135883e-06,
+      "loss": 0.0559,
+      "ss_p": 0.15066666666666667,
+      "step": 905,
+      "student_loss": 0.002926761982962489,
+      "teacher_loss": 0.0024585167411714792
+    },
+    {
+      "epoch": 0.26239907727797,
+      "grad_norm": 2.03125,
+      "kd_loss": 0.015997495502233505,
+      "learning_rate": 8.16961871886435e-06,
+      "loss": 0.059,
+      "ss_p": 0.1515,
+      "step": 910,
+      "student_loss": 0.0006850226200185716,
+      "teacher_loss": 0.0009627231047488749
+    },
+    {
+      "epoch": 0.263840830449827,
+      "grad_norm": 3.5,
+      "kd_loss": 0.06692399829626083,
+      "learning_rate": 8.148698984734887e-06,
+      "loss": 0.0565,
+      "ss_p": 0.15233333333333332,
+      "step": 915,
+      "student_loss": 0.02510847896337509,
+      "teacher_loss": 0.0017064232379198074
+    },
+    {
+      "epoch": 0.26528258362168394,
+      "grad_norm": 2.015625,
+      "kd_loss": 0.023355849087238312,
+      "learning_rate": 8.127687505296852e-06,
+      "loss": 0.0584,
+      "ss_p": 0.15316666666666667,
+      "step": 920,
+      "student_loss": 0.014558948576450348,
+      "teacher_loss": 0.012419975362718105
+    },
+    {
+      "epoch": 0.26672433679354096,
+      "grad_norm": 2.28125,
+      "kd_loss": 0.019888903945684433,
+      "learning_rate": 8.106584892772844e-06,
+      "loss": 0.0562,
+      "ss_p": 0.154,
+      "step": 925,
+      "student_loss": 0.00041061241063289344,
+      "teacher_loss": 0.00035547054721973836
+    },
+    {
+      "epoch": 0.2681660899653979,
+      "grad_norm": 2.0625,
+      "kd_loss": 0.031213538721203804,
+      "learning_rate": 8.08539176204085e-06,
+      "loss": 0.0595,
+      "ss_p": 0.15483333333333332,
+      "step": 930,
+      "student_loss": 0.002920044120401144,
+      "teacher_loss": 0.0016776882112026215
+    },
+    {
+      "epoch": 0.2696078431372549,
+      "grad_norm": 1.84375,
+      "kd_loss": 0.027394674718379974,
+      "learning_rate": 8.064108730616336e-06,
+      "loss": 0.0563,
+      "ss_p": 0.15566666666666668,
+      "step": 935,
+      "student_loss": 0.008324102498590946,
+      "teacher_loss": 0.009070378728210926
+    },
+    {
+      "epoch": 0.2710495963091119,
+      "grad_norm": 2.09375,
+      "kd_loss": 0.05151185765862465,
+      "learning_rate": 8.042736418634252e-06,
+      "loss": 0.061,
+      "ss_p": 0.1565,
+      "step": 940,
+      "student_loss": 0.01357505563646555,
+      "teacher_loss": 0.002036648103967309
+    },
+    {
+      "epoch": 0.27249134948096887,
+      "grad_norm": 1.3984375,
+      "kd_loss": 0.02468743361532688,
+      "learning_rate": 8.02127544883096e-06,
+      "loss": 0.0621,
+      "ss_p": 0.15733333333333333,
+      "step": 945,
+      "student_loss": 0.007080715615302324,
+      "teacher_loss": 0.0022822311148047447
+    },
+    {
+      "epoch": 0.27393310265282583,
+      "grad_norm": 1.4375,
+      "kd_loss": 0.03817794844508171,
+      "learning_rate": 7.999726446526095e-06,
+      "loss": 0.0545,
+      "ss_p": 0.15816666666666668,
+      "step": 950,
+      "student_loss": 0.00028919827309437096,
+      "teacher_loss": 0.0002177249116357416
+    },
+    {
+      "epoch": 0.2753748558246828,
+      "grad_norm": 2.390625,
+      "kd_loss": 0.03385411575436592,
+      "learning_rate": 7.978090039604342e-06,
+      "loss": 0.0578,
+      "ss_p": 0.159,
+      "step": 955,
+      "student_loss": 0.001508592627942562,
+      "teacher_loss": 0.00042277705506421626
+    },
+    {
+      "epoch": 0.2768166089965398,
+      "grad_norm": 3.15625,
+      "kd_loss": 0.04712347686290741,
+      "learning_rate": 7.956366858497138e-06,
+      "loss": 0.0619,
+      "ss_p": 0.15983333333333333,
+      "step": 960,
+      "student_loss": 0.058671582490205765,
+      "teacher_loss": 0.019312357529997826
+    },
+    {
+      "epoch": 0.2782583621683968,
+      "grad_norm": 6.09375,
+      "kd_loss": 0.043683819472789764,
+      "learning_rate": 7.934557536164304e-06,
+      "loss": 0.0567,
+      "ss_p": 0.16066666666666668,
+      "step": 965,
+      "student_loss": 0.010448629036545753,
+      "teacher_loss": 0.006408644840121269
+    },
+    {
+      "epoch": 0.27970011534025374,
+      "grad_norm": 3.953125,
+      "kd_loss": 0.03605993092060089,
+      "learning_rate": 7.91266270807561e-06,
+      "loss": 0.0572,
+      "ss_p": 0.1615,
+      "step": 970,
+      "student_loss": 0.00036959629505872726,
+      "teacher_loss": 0.00015978794544935226
+    },
+    {
+      "epoch": 0.2811418685121107,
+      "grad_norm": 6.71875,
+      "kd_loss": 0.028087493032217026,
+      "learning_rate": 7.890683012192249e-06,
+      "loss": 0.0564,
+      "ss_p": 0.16233333333333333,
+      "step": 975,
+      "student_loss": 0.011479988694190979,
+      "teacher_loss": 0.0005568081978708506
+    },
+    {
+      "epoch": 0.2825836216839677,
+      "grad_norm": 3.90625,
+      "kd_loss": 0.03656884282827377,
+      "learning_rate": 7.868619088948246e-06,
+      "loss": 0.057,
+      "ss_p": 0.16316666666666665,
+      "step": 980,
+      "student_loss": 0.001271704793907702,
+      "teacher_loss": 0.0011149536585435271
+    },
+    {
+      "epoch": 0.2840253748558247,
+      "grad_norm": 4.28125,
+      "kd_loss": 0.030445413663983345,
+      "learning_rate": 7.846471581231814e-06,
+      "loss": 0.0597,
+      "ss_p": 0.164,
+      "step": 985,
+      "student_loss": 0.00041054701432585716,
+      "teacher_loss": 0.00036504954914562404
+    },
+    {
+      "epoch": 0.28546712802768165,
+      "grad_norm": 2.734375,
+      "kd_loss": 0.02900032326579094,
+      "learning_rate": 7.824241134366604e-06,
+      "loss": 0.0506,
+      "ss_p": 0.16483333333333333,
+      "step": 990,
+      "student_loss": 0.0003092332335654646,
+      "teacher_loss": 0.00026412648730911314
+    },
+    {
+      "epoch": 0.2869088811995386,
+      "grad_norm": 2.125,
+      "kd_loss": 0.043829672038555145,
+      "learning_rate": 7.801928396092908e-06,
+      "loss": 0.059,
+      "ss_p": 0.16566666666666666,
+      "step": 995,
+      "student_loss": 0.0030654766596853733,
+      "teacher_loss": 0.0021067524794489145
+    },
+    {
+      "epoch": 0.28835063437139563,
+      "grad_norm": 1.4296875,
+      "kd_loss": 0.03440792113542557,
+      "learning_rate": 7.779534016548791e-06,
+      "loss": 0.0591,
+      "ss_p": 0.1665,
+      "step": 1000,
+      "student_loss": 0.006035493221133947,
+      "teacher_loss": 0.0004984359838999808
+    }
+  ],
+  "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_ss0.25/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_ss0.25/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_ss0.25/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_ss0.25/checkpoint-1500/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd88d0fb76d78f5a88d7b3c8124c573da78c733564c66af90fb855a95ac87bdf
+size 6187858991

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/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_ss0.25/checkpoint-1500/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8932ecc74ffe708ae69123f43a19804f2734013531d5ba6a59576384b99432d6
+size 16788033

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/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_ss0.25/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_ss0.25/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_ss0.25/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_ss0.25/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_ss0.25/checkpoint-2000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8da30a6cad0fa0927fe53c259e48cb542488b48aebb594b0285949534627fa0
+size 6187858991

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/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_ss0.25/checkpoint-2000/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7532801cfb3e55589e923c2fdf4dffce08182db495ca316de83b6459cee0ab4d
+size 16788033

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/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_ss0.25/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_ss0.25/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_ss0.25/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_ss0.25/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_ss0.25/checkpoint-2500/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40492009633d4b33991cc1cea710361173df3183e5852c117bcb659c04a2b6b2
+size 6187858991

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/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_ss0.25/checkpoint-2500/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:023cad5b4884cf2e551db16c1648a171c77250766dcd4f0c1c3343ffb5e896c6
+size 16788033

checkpoints/codi3b_a1.0_b1.0_g1.0_ls1_ss0.25/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_ss0.25/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_ss0.25/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_ss0.25/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_ss0.25/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_ss0.25/checkpoint-500/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75c01dd32b9f36be2d5653159ba0afcac581a8a3c35b46029215425ea9413b11
+size 6187858991