add checkpoints/codi_frozen_3b_logit

.gitattributes

@@ -100,3 +100,6 @@ checkpoints/codi_frozen_1.5b_logit/wandb/offline-run-20260624_215746-xw4jltek/ru
 checkpoints/codi_frozen_3b_hidden/checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/codi_frozen_3b_hidden/checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/codi_frozen_3b_hidden/wandb/offline-run-20260624_182251-t1mc3kud/run-t1mc3kud.wandb filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi_frozen_3b_logit/checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi_frozen_3b_logit/checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi_frozen_3b_logit/wandb/offline-run-20260624_183654-h7tylfot/run-h7tylfot.wandb filter=lfs diff=lfs merge=lfs -text

checkpoints/codi_frozen_3b_logit/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/codi_frozen_3b_logit/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/codi_frozen_3b_logit/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/codi_frozen_3b_logit/checkpoint-1000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5079b9c7c7dbdeaf2ac0281a52f84d7b13abc8101d96f85db35b3766ebb4ba54
+size 6187858991

checkpoints/codi_frozen_3b_logit/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/codi_frozen_3b_logit/checkpoint-1000/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf230cc47076e296896a872f7002ad21b69b838dbc9963295ddae9dfcf60e5b1
+size 16788033

checkpoints/codi_frozen_3b_logit/checkpoint-1000/tokenizer.json

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

checkpoints/codi_frozen_3b_logit/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/codi_frozen_3b_logit/checkpoint-1000/trainer_state.json

@@ -0,0 +1,2034 @@
+{
+  "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": 1584.0,
+      "kd_loss": 1.4788804054260254,
+      "learning_rate": 4.444444444444445e-07,
+      "loss": 2.2647,
+      "step": 5,
+      "student_loss": 0.12291660904884338,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.0028835063437139563,
+      "grad_norm": 796.0,
+      "kd_loss": 1.487825870513916,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 1.9392,
+      "step": 10,
+      "student_loss": 0.198484867811203,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.004325259515570935,
+      "grad_norm": 260.0,
+      "kd_loss": 0.07314793765544891,
+      "learning_rate": 1.5555555555555558e-06,
+      "loss": 1.2376,
+      "step": 15,
+      "student_loss": 0.08613421022891998,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.0057670126874279125,
+      "grad_norm": 85.0,
+      "kd_loss": 0.023559529334306717,
+      "learning_rate": 2.1111111111111114e-06,
+      "loss": 0.6343,
+      "step": 20,
+      "student_loss": 0.23392575979232788,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.00720876585928489,
+      "grad_norm": 33.75,
+      "kd_loss": 0.0026166969910264015,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 0.6107,
+      "step": 25,
+      "student_loss": 0.2554866075515747,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.00865051903114187,
+      "grad_norm": 24.125,
+      "kd_loss": 0.022094376385211945,
+      "learning_rate": 3.2222222222222227e-06,
+      "loss": 0.498,
+      "step": 30,
+      "student_loss": 0.2493494749069214,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.010092272202998846,
+      "grad_norm": 39.0,
+      "kd_loss": 0.04670819640159607,
+      "learning_rate": 3.777777777777778e-06,
+      "loss": 0.3781,
+      "step": 35,
+      "student_loss": 0.13782167434692383,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.011534025374855825,
+      "grad_norm": 18.125,
+      "kd_loss": 0.24180860817432404,
+      "learning_rate": 4.333333333333334e-06,
+      "loss": 0.3341,
+      "step": 40,
+      "student_loss": 0.09410303086042404,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.012975778546712802,
+      "grad_norm": 20.25,
+      "kd_loss": 0.4184195101261139,
+      "learning_rate": 4.888888888888889e-06,
+      "loss": 0.2388,
+      "step": 45,
+      "student_loss": 0.09438568353652954,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.01441753171856978,
+      "grad_norm": 32.75,
+      "kd_loss": 0.8934061527252197,
+      "learning_rate": 5.444444444444445e-06,
+      "loss": 0.3727,
+      "step": 50,
+      "student_loss": 0.020176472142338753,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.015859284890426758,
+      "grad_norm": 28.5,
+      "kd_loss": 0.018707511946558952,
+      "learning_rate": 6e-06,
+      "loss": 0.2167,
+      "step": 55,
+      "student_loss": 0.0011619763681665063,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.01730103806228374,
+      "grad_norm": 27.25,
+      "kd_loss": 0.007137665059417486,
+      "learning_rate": 6.555555555555556e-06,
+      "loss": 0.1682,
+      "step": 60,
+      "student_loss": 0.001694640377536416,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.018742791234140715,
+      "grad_norm": 15.75,
+      "kd_loss": 0.0016208974411711097,
+      "learning_rate": 7.111111111111112e-06,
+      "loss": 0.1815,
+      "step": 65,
+      "student_loss": 0.029718520119786263,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.020184544405997693,
+      "grad_norm": 12.1875,
+      "kd_loss": 0.2541334331035614,
+      "learning_rate": 7.666666666666667e-06,
+      "loss": 0.1369,
+      "step": 70,
+      "student_loss": 0.010810595005750656,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.02162629757785467,
+      "grad_norm": 18.375,
+      "kd_loss": 0.007482057902961969,
+      "learning_rate": 8.222222222222222e-06,
+      "loss": 0.2322,
+      "step": 75,
+      "student_loss": 0.00097510009072721,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.02306805074971165,
+      "grad_norm": 10.3125,
+      "kd_loss": 0.006171244662255049,
+      "learning_rate": 8.777777777777778e-06,
+      "loss": 0.1828,
+      "step": 80,
+      "student_loss": 0.000972377136349678,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.024509803921568627,
+      "grad_norm": 18.125,
+      "kd_loss": 0.005768964998424053,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.1856,
+      "step": 85,
+      "student_loss": 0.0005292362766340375,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.025951557093425604,
+      "grad_norm": 5.65625,
+      "kd_loss": 0.060487739741802216,
+      "learning_rate": 9.88888888888889e-06,
+      "loss": 0.1726,
+      "step": 90,
+      "student_loss": 0.007604286074638367,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.027393310265282585,
+      "grad_norm": 13.5625,
+      "kd_loss": 0.01542923878878355,
+      "learning_rate": 9.99995337991255e-06,
+      "loss": 0.1586,
+      "step": 95,
+      "student_loss": 0.015073101036250591,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.02883506343713956,
+      "grad_norm": 15.8125,
+      "kd_loss": 0.0029005378019064665,
+      "learning_rate": 9.999763987297266e-06,
+      "loss": 0.1247,
+      "step": 100,
+      "student_loss": 0.0006029795622453094,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.03027681660899654,
+      "grad_norm": 15.375,
+      "kd_loss": 0.007598078344017267,
+      "learning_rate": 9.999428913912876e-06,
+      "loss": 0.1266,
+      "step": 105,
+      "student_loss": 0.003934374079108238,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.031718569780853516,
+      "grad_norm": 16.0,
+      "kd_loss": 0.00252303434535861,
+      "learning_rate": 9.998948169522587e-06,
+      "loss": 0.1767,
+      "step": 110,
+      "student_loss": 0.04128251224756241,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.03316032295271049,
+      "grad_norm": 13.3125,
+      "kd_loss": 0.19726258516311646,
+      "learning_rate": 9.998321768134101e-06,
+      "loss": 0.1655,
+      "step": 115,
+      "student_loss": 0.001500162878073752,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.03460207612456748,
+      "grad_norm": 5.875,
+      "kd_loss": 0.008542308583855629,
+      "learning_rate": 9.997549727999212e-06,
+      "loss": 0.0858,
+      "step": 120,
+      "student_loss": 0.0007994729094207287,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.036043829296424454,
+      "grad_norm": 14.875,
+      "kd_loss": 0.7844982743263245,
+      "learning_rate": 9.996632071613257e-06,
+      "loss": 0.1328,
+      "step": 125,
+      "student_loss": 0.029885753989219666,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.03748558246828143,
+      "grad_norm": 14.0625,
+      "kd_loss": 0.011637737974524498,
+      "learning_rate": 9.995568825714479e-06,
+      "loss": 0.1454,
+      "step": 130,
+      "student_loss": 0.0005009593442082405,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.03892733564013841,
+      "grad_norm": 19.5,
+      "kd_loss": 0.12578554451465607,
+      "learning_rate": 9.994360021283235e-06,
+      "loss": 0.1434,
+      "step": 135,
+      "student_loss": 0.0007728396449238062,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.040369088811995385,
+      "grad_norm": 4.0625,
+      "kd_loss": 0.0028691363986581564,
+      "learning_rate": 9.993005693541098e-06,
+      "loss": 0.1367,
+      "step": 140,
+      "student_loss": 0.0006042163586243987,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.04181084198385236,
+      "grad_norm": 11.0,
+      "kd_loss": 1.0329420566558838,
+      "learning_rate": 9.991505881949837e-06,
+      "loss": 0.1066,
+      "step": 145,
+      "student_loss": 0.041140101850032806,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.04325259515570934,
+      "grad_norm": 7.71875,
+      "kd_loss": 0.013778705149888992,
+      "learning_rate": 9.989860630210251e-06,
+      "loss": 0.1259,
+      "step": 150,
+      "student_loss": 0.0009030756773427129,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.04469434832756632,
+      "grad_norm": 4.15625,
+      "kd_loss": 0.028440266847610474,
+      "learning_rate": 9.98806998626091e-06,
+      "loss": 0.1208,
+      "step": 155,
+      "student_loss": 0.0017314451979473233,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.0461361014994233,
+      "grad_norm": 20.75,
+      "kd_loss": 0.005404896102845669,
+      "learning_rate": 9.98613400227676e-06,
+      "loss": 0.1872,
+      "step": 160,
+      "student_loss": 0.10157372802495956,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.04757785467128028,
+      "grad_norm": 26.25,
+      "kd_loss": 0.008982345461845398,
+      "learning_rate": 9.984052734667586e-06,
+      "loss": 0.1018,
+      "step": 165,
+      "student_loss": 0.00033849035389721394,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.049019607843137254,
+      "grad_norm": 18.125,
+      "kd_loss": 0.005132909398525953,
+      "learning_rate": 9.981826244076387e-06,
+      "loss": 0.1339,
+      "step": 170,
+      "student_loss": 0.0005440761451609433,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.05046136101499423,
+      "grad_norm": 18.0,
+      "kd_loss": 0.021587209776043892,
+      "learning_rate": 9.979454595377594e-06,
+      "loss": 0.1247,
+      "step": 175,
+      "student_loss": 0.014186372049152851,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.05190311418685121,
+      "grad_norm": 8.5,
+      "kd_loss": 0.011625917628407478,
+      "learning_rate": 9.97693785767519e-06,
+      "loss": 0.0741,
+      "step": 180,
+      "student_loss": 0.0005693296552635729,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.05334486735870819,
+      "grad_norm": 7.21875,
+      "kd_loss": 0.0012370528420433402,
+      "learning_rate": 9.9742761043007e-06,
+      "loss": 0.1795,
+      "step": 185,
+      "student_loss": 0.043101612478494644,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.05478662053056517,
+      "grad_norm": 7.34375,
+      "kd_loss": 0.1722288429737091,
+      "learning_rate": 9.971469412811032e-06,
+      "loss": 0.093,
+      "step": 190,
+      "student_loss": 0.006316277664154768,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.056228373702422146,
+      "grad_norm": 44.25,
+      "kd_loss": 0.6569585800170898,
+      "learning_rate": 9.968517864986247e-06,
+      "loss": 0.1118,
+      "step": 195,
+      "student_loss": 0.023583440110087395,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.05767012687427912,
+      "grad_norm": 4.9375,
+      "kd_loss": 0.0010104529792442918,
+      "learning_rate": 9.965421546827153e-06,
+      "loss": 0.1094,
+      "step": 200,
+      "student_loss": 0.0004585790156852454,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.0591118800461361,
+      "grad_norm": 4.59375,
+      "kd_loss": 0.0006818274850957096,
+      "learning_rate": 9.962180548552812e-06,
+      "loss": 0.0885,
+      "step": 205,
+      "student_loss": 0.046175092458724976,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.06055363321799308,
+      "grad_norm": 58.0,
+      "kd_loss": 0.4983484745025635,
+      "learning_rate": 9.958794964597906e-06,
+      "loss": 0.1317,
+      "step": 210,
+      "student_loss": 0.007553795352578163,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.061995386389850055,
+      "grad_norm": 6.46875,
+      "kd_loss": 0.0009208342526108027,
+      "learning_rate": 9.955264893609986e-06,
+      "loss": 0.1522,
+      "step": 215,
+      "student_loss": 0.00047464462113566697,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.06343713956170703,
+      "grad_norm": 18.125,
+      "kd_loss": 0.0944160670042038,
+      "learning_rate": 9.951590438446597e-06,
+      "loss": 0.0774,
+      "step": 220,
+      "student_loss": 0.03491729497909546,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.06487889273356401,
+      "grad_norm": 10.25,
+      "kd_loss": 0.001832562847994268,
+      "learning_rate": 9.947771706172285e-06,
+      "loss": 0.0976,
+      "step": 225,
+      "student_loss": 0.05962345376610756,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.06632064590542099,
+      "grad_norm": 21.75,
+      "kd_loss": 0.10277731716632843,
+      "learning_rate": 9.943808808055472e-06,
+      "loss": 0.0911,
+      "step": 230,
+      "student_loss": 0.0012677249033004045,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.06776239907727798,
+      "grad_norm": 17.25,
+      "kd_loss": 0.4876050353050232,
+      "learning_rate": 9.93970185956522e-06,
+      "loss": 0.1077,
+      "step": 235,
+      "student_loss": 0.006362705025821924,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.06920415224913495,
+      "grad_norm": 1.46875,
+      "kd_loss": 0.010914070531725883,
+      "learning_rate": 9.93545098036786e-06,
+      "loss": 0.1107,
+      "step": 240,
+      "student_loss": 0.0032385734375566244,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07064590542099193,
+      "grad_norm": 10.3125,
+      "kd_loss": 0.02172888070344925,
+      "learning_rate": 9.931056294323507e-06,
+      "loss": 0.081,
+      "step": 245,
+      "student_loss": 0.0005711687845177948,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07208765859284891,
+      "grad_norm": 5.125,
+      "kd_loss": 0.005143388640135527,
+      "learning_rate": 9.926517929482454e-06,
+      "loss": 0.1052,
+      "step": 250,
+      "student_loss": 0.0005271504633128643,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07352941176470588,
+      "grad_norm": 6.5,
+      "kd_loss": 0.0008525780867785215,
+      "learning_rate": 9.921836018081438e-06,
+      "loss": 0.0589,
+      "step": 255,
+      "student_loss": 0.0003723589761648327,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07497116493656286,
+      "grad_norm": 17.0,
+      "kd_loss": 0.0010641744593158364,
+      "learning_rate": 9.917010696539793e-06,
+      "loss": 0.1453,
+      "step": 260,
+      "student_loss": 0.0432959608733654,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07641291810841984,
+      "grad_norm": 16.0,
+      "kd_loss": 0.001827583066187799,
+      "learning_rate": 9.912042105455462e-06,
+      "loss": 0.0908,
+      "step": 265,
+      "student_loss": 0.00034234311897307634,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07785467128027682,
+      "grad_norm": 14.0,
+      "kd_loss": 0.000661063939332962,
+      "learning_rate": 9.906930389600917e-06,
+      "loss": 0.0894,
+      "step": 270,
+      "student_loss": 0.0010115442564710975,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07929642445213379,
+      "grad_norm": 13.0,
+      "kd_loss": 0.0016069641569629312,
+      "learning_rate": 9.901675697918925e-06,
+      "loss": 0.1173,
+      "step": 275,
+      "student_loss": 0.05398955196142197,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.08073817762399077,
+      "grad_norm": 29.0,
+      "kd_loss": 0.0010736712720245123,
+      "learning_rate": 9.896278183518216e-06,
+      "loss": 0.1427,
+      "step": 280,
+      "student_loss": 0.0005599395371973515,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.08217993079584775,
+      "grad_norm": 17.5,
+      "kd_loss": 0.017228417098522186,
+      "learning_rate": 9.890738003669029e-06,
+      "loss": 0.1333,
+      "step": 285,
+      "student_loss": 0.0005250528338365257,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.08362168396770472,
+      "grad_norm": 35.5,
+      "kd_loss": 0.0026075318455696106,
+      "learning_rate": 9.885055319798514e-06,
+      "loss": 0.1752,
+      "step": 290,
+      "student_loss": 0.0017553431680426002,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.0850634371395617,
+      "grad_norm": 26.25,
+      "kd_loss": 0.004095411393791437,
+      "learning_rate": 9.879230297486034e-06,
+      "loss": 0.158,
+      "step": 295,
+      "student_loss": 0.0005752203287556767,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.08650519031141868,
+      "grad_norm": 4.96875,
+      "kd_loss": 0.2717151343822479,
+      "learning_rate": 9.873263106458353e-06,
+      "loss": 0.0941,
+      "step": 300,
+      "student_loss": 0.023448897525668144,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.08794694348327567,
+      "grad_norm": 6.65625,
+      "kd_loss": 0.019775602966547012,
+      "learning_rate": 9.86715392058467e-06,
+      "loss": 0.1302,
+      "step": 305,
+      "student_loss": 0.0025525283999741077,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.08938869665513265,
+      "grad_norm": 7.0,
+      "kd_loss": 0.0005424432456493378,
+      "learning_rate": 9.860902917871566e-06,
+      "loss": 0.1096,
+      "step": 310,
+      "student_loss": 0.000471778039354831,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.09083044982698962,
+      "grad_norm": 10.25,
+      "kd_loss": 0.021720539778470993,
+      "learning_rate": 9.854510280457819e-06,
+      "loss": 0.0897,
+      "step": 315,
+      "student_loss": 0.015475576743483543,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.0922722029988466,
+      "grad_norm": 6.40625,
+      "kd_loss": 0.0012843874283134937,
+      "learning_rate": 9.847976194609083e-06,
+      "loss": 0.091,
+      "step": 320,
+      "student_loss": 0.0004686483880504966,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.09371395617070358,
+      "grad_norm": 26.375,
+      "kd_loss": 0.6982781887054443,
+      "learning_rate": 9.841300850712479e-06,
+      "loss": 0.103,
+      "step": 325,
+      "student_loss": 0.017696071416139603,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.09515570934256055,
+      "grad_norm": 8.6875,
+      "kd_loss": 0.15215124189853668,
+      "learning_rate": 9.834484443271035e-06,
+      "loss": 0.0715,
+      "step": 330,
+      "student_loss": 0.005886889062821865,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.09659746251441753,
+      "grad_norm": 2.875,
+      "kd_loss": 0.0015391090419143438,
+      "learning_rate": 9.827527170898019e-06,
+      "loss": 0.0852,
+      "step": 335,
+      "student_loss": 0.0005516717210412025,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 15.0,
+      "kd_loss": 0.07792548835277557,
+      "learning_rate": 9.820429236311158e-06,
+      "loss": 0.1289,
+      "step": 340,
+      "student_loss": 0.010181530378758907,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.09948096885813149,
+      "grad_norm": 6.0625,
+      "kd_loss": 0.013684757985174656,
+      "learning_rate": 9.81319084632673e-06,
+      "loss": 0.093,
+      "step": 345,
+      "student_loss": 0.01164299063384533,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.10092272202998846,
+      "grad_norm": 5.3125,
+      "kd_loss": 0.0955321341753006,
+      "learning_rate": 9.805812211853534e-06,
+      "loss": 0.0668,
+      "step": 350,
+      "student_loss": 0.0006315007922239602,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.10236447520184544,
+      "grad_norm": 15.75,
+      "kd_loss": 0.000853118603117764,
+      "learning_rate": 9.798293547886748e-06,
+      "loss": 0.1354,
+      "step": 355,
+      "student_loss": 0.00041306763887405396,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.10380622837370242,
+      "grad_norm": 4.125,
+      "kd_loss": 0.0023229331709444523,
+      "learning_rate": 9.790635073501662e-06,
+      "loss": 0.0929,
+      "step": 360,
+      "student_loss": 0.0004519273934420198,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1052479815455594,
+      "grad_norm": 12.875,
+      "kd_loss": 0.004534870386123657,
+      "learning_rate": 9.782837011847294e-06,
+      "loss": 0.177,
+      "step": 365,
+      "student_loss": 0.048408832401037216,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.10668973471741638,
+      "grad_norm": 5.75,
+      "kd_loss": 0.5334176421165466,
+      "learning_rate": 9.774899590139897e-06,
+      "loss": 0.0711,
+      "step": 370,
+      "student_loss": 0.0273645780980587,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.10813148788927336,
+      "grad_norm": 12.25,
+      "kd_loss": 0.006150795612484217,
+      "learning_rate": 9.766823039656328e-06,
+      "loss": 0.1286,
+      "step": 375,
+      "student_loss": 0.03481242433190346,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.10957324106113034,
+      "grad_norm": 0.69140625,
+      "kd_loss": 0.0020576389506459236,
+      "learning_rate": 9.758607595727309e-06,
+      "loss": 0.0763,
+      "step": 380,
+      "student_loss": 0.0004342672764323652,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.11101499423298732,
+      "grad_norm": 3.375,
+      "kd_loss": 0.01198313757777214,
+      "learning_rate": 9.75025349773058e-06,
+      "loss": 0.0684,
+      "step": 385,
+      "student_loss": 0.00040114219882525504,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.11245674740484429,
+      "grad_norm": 42.25,
+      "kd_loss": 0.0029099821113049984,
+      "learning_rate": 9.741760989083914e-06,
+      "loss": 0.0966,
+      "step": 390,
+      "student_loss": 0.00024992373073473573,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.11389850057670127,
+      "grad_norm": 10.4375,
+      "kd_loss": 0.0029170780908316374,
+      "learning_rate": 9.73313031723803e-06,
+      "loss": 0.0846,
+      "step": 395,
+      "student_loss": 0.0004683209699578583,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.11534025374855825,
+      "grad_norm": 2.390625,
+      "kd_loss": 0.002679860917851329,
+      "learning_rate": 9.724361733669383e-06,
+      "loss": 0.0425,
+      "step": 400,
+      "student_loss": 0.002333736512809992,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.11678200692041522,
+      "grad_norm": 11.6875,
+      "kd_loss": 0.0017421097727492452,
+      "learning_rate": 9.715455493872827e-06,
+      "loss": 0.1317,
+      "step": 405,
+      "student_loss": 0.011753321625292301,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1182237600922722,
+      "grad_norm": 5.96875,
+      "kd_loss": 0.0038202686700969934,
+      "learning_rate": 9.70641185735419e-06,
+      "loss": 0.0947,
+      "step": 410,
+      "student_loss": 0.015384526923298836,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.11966551326412918,
+      "grad_norm": 15.625,
+      "kd_loss": 0.007967401295900345,
+      "learning_rate": 9.697231087622691e-06,
+      "loss": 0.1609,
+      "step": 415,
+      "student_loss": 0.003767178161069751,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.12110726643598616,
+      "grad_norm": 7.78125,
+      "kd_loss": 0.3626428246498108,
+      "learning_rate": 9.687913452183277e-06,
+      "loss": 0.0883,
+      "step": 420,
+      "student_loss": 0.06143607571721077,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.12254901960784313,
+      "grad_norm": 9.1875,
+      "kd_loss": 0.0016940389759838581,
+      "learning_rate": 9.678459222528825e-06,
+      "loss": 0.0493,
+      "step": 425,
+      "student_loss": 0.002240637084469199,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.12399077277970011,
+      "grad_norm": 19.75,
+      "kd_loss": 0.9260356426239014,
+      "learning_rate": 9.668868674132224e-06,
+      "loss": 0.1153,
+      "step": 430,
+      "student_loss": 0.0827086940407753,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1254325259515571,
+      "grad_norm": 12.8125,
+      "kd_loss": 0.0019437845330685377,
+      "learning_rate": 9.65914208643836e-06,
+      "loss": 0.0634,
+      "step": 435,
+      "student_loss": 0.0006262208335101604,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.12687427912341406,
+      "grad_norm": 4.25,
+      "kd_loss": 0.00489834975451231,
+      "learning_rate": 9.649279742855964e-06,
+      "loss": 0.103,
+      "step": 440,
+      "student_loss": 0.013814084231853485,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.12831603229527105,
+      "grad_norm": 22.375,
+      "kd_loss": 0.0016376337734982371,
+      "learning_rate": 9.639281930749363e-06,
+      "loss": 0.079,
+      "step": 445,
+      "student_loss": 0.03539160266518593,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.12975778546712802,
+      "grad_norm": 7.375,
+      "kd_loss": 0.0012233811430633068,
+      "learning_rate": 9.629148941430096e-06,
+      "loss": 0.1054,
+      "step": 450,
+      "student_loss": 0.0006881144363433123,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.131199538638985,
+      "grad_norm": 6.0,
+      "kd_loss": 0.0003836078103631735,
+      "learning_rate": 9.618881070148436e-06,
+      "loss": 0.1035,
+      "step": 455,
+      "student_loss": 0.001039310242049396,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.13264129181084197,
+      "grad_norm": 7.15625,
+      "kd_loss": 0.039876699447631836,
+      "learning_rate": 9.608478616084784e-06,
+      "loss": 0.0547,
+      "step": 460,
+      "student_loss": 0.008442185819149017,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.13408304498269896,
+      "grad_norm": 15.4375,
+      "kd_loss": 0.08534516394138336,
+      "learning_rate": 9.597941882340947e-06,
+      "loss": 0.1407,
+      "step": 465,
+      "student_loss": 0.00036318166530691087,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.13552479815455595,
+      "grad_norm": 9.125,
+      "kd_loss": 0.000860957894474268,
+      "learning_rate": 9.587271175931313e-06,
+      "loss": 0.0872,
+      "step": 470,
+      "student_loss": 0.00031289560138247907,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.13696655132641292,
+      "grad_norm": 3.21875,
+      "kd_loss": 0.021515244618058205,
+      "learning_rate": 9.5764668077739e-06,
+      "loss": 0.0627,
+      "step": 475,
+      "student_loss": 0.0004178633389528841,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1384083044982699,
+      "grad_norm": 11.5625,
+      "kd_loss": 0.01755513623356819,
+      "learning_rate": 9.5655290926813e-06,
+      "loss": 0.0742,
+      "step": 480,
+      "student_loss": 0.041500866413116455,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.13985005767012687,
+      "grad_norm": 18.875,
+      "kd_loss": 0.0021918262355029583,
+      "learning_rate": 9.554458349351507e-06,
+      "loss": 0.0853,
+      "step": 485,
+      "student_loss": 0.0005753549048677087,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.14129181084198386,
+      "grad_norm": 16.0,
+      "kd_loss": 0.005263358820229769,
+      "learning_rate": 9.54325490035863e-06,
+      "loss": 0.1613,
+      "step": 490,
+      "student_loss": 0.0003423037414904684,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.14273356401384082,
+      "grad_norm": 6.75,
+      "kd_loss": 0.0663486197590828,
+      "learning_rate": 9.531919072143488e-06,
+      "loss": 0.0925,
+      "step": 495,
+      "student_loss": 0.0005914232460781932,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.14417531718569782,
+      "grad_norm": 7.40625,
+      "kd_loss": 0.032540202140808105,
+      "learning_rate": 9.520451195004107e-06,
+      "loss": 0.0773,
+      "step": 500,
+      "student_loss": 0.00025544269010424614,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.14561707035755478,
+      "grad_norm": 2.671875,
+      "kd_loss": 0.15188878774642944,
+      "learning_rate": 9.508851603086094e-06,
+      "loss": 0.098,
+      "step": 505,
+      "student_loss": 0.0003635012835729867,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.14705882352941177,
+      "grad_norm": 9.3125,
+      "kd_loss": 0.0018157297745347023,
+      "learning_rate": 9.497120634372894e-06,
+      "loss": 0.1185,
+      "step": 510,
+      "student_loss": 0.00028236082289367914,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.14850057670126873,
+      "grad_norm": 9.6875,
+      "kd_loss": 0.015397800132632256,
+      "learning_rate": 9.48525863067595e-06,
+      "loss": 0.0661,
+      "step": 515,
+      "student_loss": 0.013613639399409294,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.14994232987312572,
+      "grad_norm": 9.0,
+      "kd_loss": 0.012119506485760212,
+      "learning_rate": 9.473265937624748e-06,
+      "loss": 0.0543,
+      "step": 520,
+      "student_loss": 0.041264429688453674,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1513840830449827,
+      "grad_norm": 1.078125,
+      "kd_loss": 0.0020418986678123474,
+      "learning_rate": 9.461142904656722e-06,
+      "loss": 0.0978,
+      "step": 525,
+      "student_loss": 0.0004116448399145156,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.15282583621683968,
+      "grad_norm": 7.5,
+      "kd_loss": 0.013642538338899612,
+      "learning_rate": 9.448889885007105e-06,
+      "loss": 0.0735,
+      "step": 530,
+      "student_loss": 0.00035193507210351527,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.15426758938869667,
+      "grad_norm": 2.953125,
+      "kd_loss": 0.01120305061340332,
+      "learning_rate": 9.436507235698613e-06,
+      "loss": 0.0934,
+      "step": 535,
+      "student_loss": 0.0004887144896201789,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.15570934256055363,
+      "grad_norm": 5.15625,
+      "kd_loss": 0.004150867462158203,
+      "learning_rate": 9.423995317531049e-06,
+      "loss": 0.091,
+      "step": 540,
+      "student_loss": 0.025533407926559448,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.15715109573241062,
+      "grad_norm": 4.8125,
+      "kd_loss": 0.018951416015625,
+      "learning_rate": 9.4113544950708e-06,
+      "loss": 0.0753,
+      "step": 545,
+      "student_loss": 0.018380453810095787,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.15859284890426759,
+      "grad_norm": 7.1875,
+      "kd_loss": 0.007083653472363949,
+      "learning_rate": 9.398585136640195e-06,
+      "loss": 0.0734,
+      "step": 550,
+      "student_loss": 0.00018351977632846683,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.16003460207612458,
+      "grad_norm": 13.375,
+      "kd_loss": 1.1156541109085083,
+      "learning_rate": 9.38568761430679e-06,
+      "loss": 0.1165,
+      "step": 555,
+      "student_loss": 0.03308179974555969,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.16147635524798154,
+      "grad_norm": 3.390625,
+      "kd_loss": 0.023138834163546562,
+      "learning_rate": 9.372662303872514e-06,
+      "loss": 0.1369,
+      "step": 560,
+      "student_loss": 0.0008326194947585464,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.16291810841983853,
+      "grad_norm": 9.125,
+      "kd_loss": 0.006360532250255346,
+      "learning_rate": 9.359509584862735e-06,
+      "loss": 0.0822,
+      "step": 565,
+      "student_loss": 0.030178692191839218,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1643598615916955,
+      "grad_norm": 23.625,
+      "kd_loss": 0.0031671514734625816,
+      "learning_rate": 9.346229840515186e-06,
+      "loss": 0.1555,
+      "step": 570,
+      "student_loss": 0.01120324432849884,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.16580161476355249,
+      "grad_norm": 4.1875,
+      "kd_loss": 0.0011146231554448605,
+      "learning_rate": 9.332823457768805e-06,
+      "loss": 0.0603,
+      "step": 575,
+      "student_loss": 0.00039536861004307866,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.16724336793540945,
+      "grad_norm": 4.46875,
+      "kd_loss": 0.002190124476328492,
+      "learning_rate": 9.31929082725246e-06,
+      "loss": 0.09,
+      "step": 580,
+      "student_loss": 0.0019336750265210867,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.16868512110726644,
+      "grad_norm": 4.75,
+      "kd_loss": 0.11950447410345078,
+      "learning_rate": 9.305632343273567e-06,
+      "loss": 0.0721,
+      "step": 585,
+      "student_loss": 0.0013314392417669296,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1701268742791234,
+      "grad_norm": 3.265625,
+      "kd_loss": 0.025959620252251625,
+      "learning_rate": 9.291848403806604e-06,
+      "loss": 0.0403,
+      "step": 590,
+      "student_loss": 0.0005966908065602183,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1715686274509804,
+      "grad_norm": 9.5625,
+      "kd_loss": 0.025158416479825974,
+      "learning_rate": 9.277939410481507e-06,
+      "loss": 0.1159,
+      "step": 595,
+      "student_loss": 0.00030659351614303887,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.17301038062283736,
+      "grad_norm": 26.125,
+      "kd_loss": 0.011669105850160122,
+      "learning_rate": 9.263905768571977e-06,
+      "loss": 0.1474,
+      "step": 600,
+      "student_loss": 0.04554896056652069,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.17445213379469435,
+      "grad_norm": 10.375,
+      "kd_loss": 0.004631172399967909,
+      "learning_rate": 9.249747886983664e-06,
+      "loss": 0.0784,
+      "step": 605,
+      "student_loss": 0.11249730736017227,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.17589388696655134,
+      "grad_norm": 6.78125,
+      "kd_loss": 0.16099677979946136,
+      "learning_rate": 9.235466178242255e-06,
+      "loss": 0.0878,
+      "step": 610,
+      "student_loss": 0.007015160284936428,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1773356401384083,
+      "grad_norm": 4.28125,
+      "kd_loss": 0.0017117339884862304,
+      "learning_rate": 9.221061058481456e-06,
+      "loss": 0.0695,
+      "step": 615,
+      "student_loss": 0.00019477600289974362,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1787773933102653,
+      "grad_norm": 8.6875,
+      "kd_loss": 0.0050481888465583324,
+      "learning_rate": 9.206532947430865e-06,
+      "loss": 0.0756,
+      "step": 620,
+      "student_loss": 0.01180206798017025,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.18021914648212226,
+      "grad_norm": 4.15625,
+      "kd_loss": 0.0011246080975979567,
+      "learning_rate": 9.191882268403743e-06,
+      "loss": 0.0468,
+      "step": 625,
+      "student_loss": 0.013564247637987137,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.18166089965397925,
+      "grad_norm": 25.75,
+      "kd_loss": 0.6801473498344421,
+      "learning_rate": 9.177109448284673e-06,
+      "loss": 0.1347,
+      "step": 630,
+      "student_loss": 0.013638689182698727,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1831026528258362,
+      "grad_norm": 7.96875,
+      "kd_loss": 0.0010556690394878387,
+      "learning_rate": 9.162214917517139e-06,
+      "loss": 0.101,
+      "step": 635,
+      "student_loss": 0.0004484328965190798,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1845444059976932,
+      "grad_norm": 3.96875,
+      "kd_loss": 0.008011292666196823,
+      "learning_rate": 9.14719911009096e-06,
+      "loss": 0.0764,
+      "step": 640,
+      "student_loss": 0.0003311619511805475,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.18598615916955016,
+      "grad_norm": 7.21875,
+      "kd_loss": 0.030139261856675148,
+      "learning_rate": 9.132062463529666e-06,
+      "loss": 0.0689,
+      "step": 645,
+      "student_loss": 0.0003902459575328976,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.18742791234140715,
+      "grad_norm": 5.59375,
+      "kd_loss": 0.0013374610571190715,
+      "learning_rate": 9.11680541887774e-06,
+      "loss": 0.0663,
+      "step": 650,
+      "student_loss": 0.0005999109125696123,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.18886966551326412,
+      "grad_norm": 36.25,
+      "kd_loss": 0.007892305962741375,
+      "learning_rate": 9.101428420687759e-06,
+      "loss": 0.0766,
+      "step": 655,
+      "student_loss": 0.00040867438656277955,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1903114186851211,
+      "grad_norm": 12.8125,
+      "kd_loss": 0.004360968712717295,
+      "learning_rate": 9.085931917007464e-06,
+      "loss": 0.0645,
+      "step": 660,
+      "student_loss": 0.00044171695481054485,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.19175317185697807,
+      "grad_norm": 5.90625,
+      "kd_loss": 0.023713773116469383,
+      "learning_rate": 9.070316359366681e-06,
+      "loss": 0.0818,
+      "step": 665,
+      "student_loss": 0.00041215657256543636,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.19319492502883506,
+      "grad_norm": 9.3125,
+      "kd_loss": 0.14988374710083008,
+      "learning_rate": 9.054582202764175e-06,
+      "loss": 0.0605,
+      "step": 670,
+      "student_loss": 0.002806498669087887,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.19463667820069205,
+      "grad_norm": 3.984375,
+      "kd_loss": 0.006893543992191553,
+      "learning_rate": 9.038729905654393e-06,
+      "loss": 0.0936,
+      "step": 675,
+      "student_loss": 0.0019069729605689645,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 17.5,
+      "kd_loss": 0.6314095258712769,
+      "learning_rate": 9.022759929934109e-06,
+      "loss": 0.0876,
+      "step": 680,
+      "student_loss": 0.10397554188966751,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.197520184544406,
+      "grad_norm": 10.75,
+      "kd_loss": 0.005205235444009304,
+      "learning_rate": 9.006672740928952e-06,
+      "loss": 0.0805,
+      "step": 685,
+      "student_loss": 0.04243871197104454,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.19896193771626297,
+      "grad_norm": 20.875,
+      "kd_loss": 0.9131407141685486,
+      "learning_rate": 8.990468807379861e-06,
+      "loss": 0.1088,
+      "step": 690,
+      "student_loss": 0.015043575316667557,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.20040369088811996,
+      "grad_norm": 17.625,
+      "kd_loss": 0.0014001968083903193,
+      "learning_rate": 8.974148601429426e-06,
+      "loss": 0.1066,
+      "step": 695,
+      "student_loss": 0.004736350383609533,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.20184544405997693,
+      "grad_norm": 13.125,
+      "kd_loss": 0.009080437943339348,
+      "learning_rate": 8.957712598608123e-06,
+      "loss": 0.122,
+      "step": 700,
+      "student_loss": 0.0004191877378616482,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.20328719723183392,
+      "grad_norm": 7.59375,
+      "kd_loss": 0.008329123258590698,
+      "learning_rate": 8.941161277820462e-06,
+      "loss": 0.0587,
+      "step": 705,
+      "student_loss": 0.00034082072670571506,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.20472895040369088,
+      "grad_norm": 5.0,
+      "kd_loss": 0.3375152051448822,
+      "learning_rate": 8.924495121331038e-06,
+      "loss": 0.0808,
+      "step": 710,
+      "student_loss": 0.0070212045684456825,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.20617070357554787,
+      "grad_norm": 17.125,
+      "kd_loss": 0.0006333683268167078,
+      "learning_rate": 8.907714614750473e-06,
+      "loss": 0.1079,
+      "step": 715,
+      "student_loss": 0.0029529642779380083,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.20761245674740483,
+      "grad_norm": 3.109375,
+      "kd_loss": 0.008093835785984993,
+      "learning_rate": 8.890820247021263e-06,
+      "loss": 0.1075,
+      "step": 720,
+      "student_loss": 0.0004524969554040581,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.20905420991926182,
+      "grad_norm": 3.4375,
+      "kd_loss": 0.02861846052110195,
+      "learning_rate": 8.873812510403545e-06,
+      "loss": 0.0413,
+      "step": 725,
+      "student_loss": 0.00030948614585213363,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2104959630911188,
+      "grad_norm": 18.375,
+      "kd_loss": 0.013474239967763424,
+      "learning_rate": 8.85669190046074e-06,
+      "loss": 0.09,
+      "step": 730,
+      "student_loss": 0.007885397411882877,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.21193771626297578,
+      "grad_norm": 11.25,
+      "kd_loss": 0.0050799660384655,
+      "learning_rate": 8.839458916045117e-06,
+      "loss": 0.0757,
+      "step": 735,
+      "student_loss": 0.0007924353703856468,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.21337946943483277,
+      "grad_norm": 19.375,
+      "kd_loss": 0.03126922622323036,
+      "learning_rate": 8.822114059283267e-06,
+      "loss": 0.0763,
+      "step": 740,
+      "student_loss": 0.0034000398591160774,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.21482122260668973,
+      "grad_norm": 15.8125,
+      "kd_loss": 0.001199642545543611,
+      "learning_rate": 8.804657835561456e-06,
+      "loss": 0.0755,
+      "step": 745,
+      "student_loss": 0.04256457835435867,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.21626297577854672,
+      "grad_norm": 10.4375,
+      "kd_loss": 0.5467672944068909,
+      "learning_rate": 8.787090753510917e-06,
+      "loss": 0.0635,
+      "step": 750,
+      "student_loss": 0.011441598646342754,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2177047289504037,
+      "grad_norm": 16.5,
+      "kd_loss": 0.0009971132967621088,
+      "learning_rate": 8.76941332499301e-06,
+      "loss": 0.1129,
+      "step": 755,
+      "student_loss": 0.03640365228056908,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.21914648212226068,
+      "grad_norm": 5.6875,
+      "kd_loss": 0.003233318915590644,
+      "learning_rate": 8.751626065084328e-06,
+      "loss": 0.0701,
+      "step": 760,
+      "student_loss": 0.04764896258711815,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.22058823529411764,
+      "grad_norm": 10.6875,
+      "kd_loss": 0.011508336290717125,
+      "learning_rate": 8.733729492061675e-06,
+      "loss": 0.0648,
+      "step": 765,
+      "student_loss": 0.0028325854800641537,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.22202998846597463,
+      "grad_norm": 4.0,
+      "kd_loss": 0.009445695206522942,
+      "learning_rate": 8.715724127386971e-06,
+      "loss": 0.1065,
+      "step": 770,
+      "student_loss": 0.0005283040809445083,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2234717416378316,
+      "grad_norm": 4.59375,
+      "kd_loss": 0.09654000401496887,
+      "learning_rate": 8.697610495692055e-06,
+      "loss": 0.0785,
+      "step": 775,
+      "student_loss": 0.002290144795551896,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.22491349480968859,
+      "grad_norm": 7.96875,
+      "kd_loss": 0.010726231150329113,
+      "learning_rate": 8.679389124763396e-06,
+      "loss": 0.1146,
+      "step": 780,
+      "student_loss": 0.0004144510894548148,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.22635524798154555,
+      "grad_norm": 9.4375,
+      "kd_loss": 0.0006008287309668958,
+      "learning_rate": 8.661060545526724e-06,
+      "loss": 0.0778,
+      "step": 785,
+      "student_loss": 0.0003037274000234902,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.22779700115340254,
+      "grad_norm": 8.625,
+      "kd_loss": 0.0009737013024277985,
+      "learning_rate": 8.64262529203155e-06,
+      "loss": 0.0607,
+      "step": 790,
+      "student_loss": 0.010159829631447792,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2292387543252595,
+      "grad_norm": 4.09375,
+      "kd_loss": 0.005302474834024906,
+      "learning_rate": 8.624083901435613e-06,
+      "loss": 0.08,
+      "step": 795,
+      "student_loss": 0.0003047943173442036,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2306805074971165,
+      "grad_norm": 17.125,
+      "kd_loss": 0.34581008553504944,
+      "learning_rate": 8.60543691398922e-06,
+      "loss": 0.0716,
+      "step": 800,
+      "student_loss": 0.058034349232912064,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.23212226066897348,
+      "grad_norm": 6.6875,
+      "kd_loss": 0.000975580420345068,
+      "learning_rate": 8.586684873019513e-06,
+      "loss": 0.0915,
+      "step": 805,
+      "student_loss": 0.0003182948858011514,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.23356401384083045,
+      "grad_norm": 7.8125,
+      "kd_loss": 0.0011780632194131613,
+      "learning_rate": 8.567828324914633e-06,
+      "loss": 0.0413,
+      "step": 810,
+      "student_loss": 0.0003564381622709334,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.23500576701268744,
+      "grad_norm": 4.6875,
+      "kd_loss": 0.0033866853918880224,
+      "learning_rate": 8.548867819107803e-06,
+      "loss": 0.0969,
+      "step": 815,
+      "student_loss": 0.0008914784411899745,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2364475201845444,
+      "grad_norm": 33.5,
+      "kd_loss": 0.0014539997791871428,
+      "learning_rate": 8.52980390806131e-06,
+      "loss": 0.0657,
+      "step": 820,
+      "student_loss": 0.005444650072604418,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2378892733564014,
+      "grad_norm": 7.71875,
+      "kd_loss": 0.0013552566524595022,
+      "learning_rate": 8.51063714725042e-06,
+      "loss": 0.0936,
+      "step": 825,
+      "student_loss": 0.012255060486495495,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.23933102652825836,
+      "grad_norm": 11.625,
+      "kd_loss": 0.16935427486896515,
+      "learning_rate": 8.49136809514719e-06,
+      "loss": 0.0793,
+      "step": 830,
+      "student_loss": 0.00629698159173131,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.24077277970011535,
+      "grad_norm": 5.25,
+      "kd_loss": 0.0005752237630076706,
+      "learning_rate": 8.471997313204183e-06,
+      "loss": 0.0423,
+      "step": 835,
+      "student_loss": 0.014124217443168163,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2422145328719723,
+      "grad_norm": 4.96875,
+      "kd_loss": 0.0006229743594303727,
+      "learning_rate": 8.452525365838123e-06,
+      "loss": 0.0537,
+      "step": 840,
+      "student_loss": 0.00042915187077596784,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2436562860438293,
+      "grad_norm": 10.9375,
+      "kd_loss": 0.012905994430184364,
+      "learning_rate": 8.432952820413447e-06,
+      "loss": 0.0811,
+      "step": 845,
+      "student_loss": 0.00039613110129721463,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.24509803921568626,
+      "grad_norm": 19.375,
+      "kd_loss": 0.18080615997314453,
+      "learning_rate": 8.41328024722577e-06,
+      "loss": 0.1201,
+      "step": 850,
+      "student_loss": 0.011291730217635632,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.24653979238754326,
+      "grad_norm": 16.75,
+      "kd_loss": 0.0011529283365234733,
+      "learning_rate": 8.393508219485263e-06,
+      "loss": 0.1001,
+      "step": 855,
+      "student_loss": 0.00018943188479170203,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.24798154555940022,
+      "grad_norm": 8.3125,
+      "kd_loss": 0.010198691859841347,
+      "learning_rate": 8.373637313299966e-06,
+      "loss": 0.1369,
+      "step": 860,
+      "student_loss": 0.0010680719278752804,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2494232987312572,
+      "grad_norm": 4.78125,
+      "kd_loss": 0.0014139733975753188,
+      "learning_rate": 8.353668107658984e-06,
+      "loss": 0.0713,
+      "step": 865,
+      "student_loss": 0.0002495216904208064,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2508650519031142,
+      "grad_norm": 6.53125,
+      "kd_loss": 0.01792648807168007,
+      "learning_rate": 8.333601184415632e-06,
+      "loss": 0.0735,
+      "step": 870,
+      "student_loss": 0.0004213955835439265,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.25230680507497116,
+      "grad_norm": 7.65625,
+      "kd_loss": 0.0010694258380681276,
+      "learning_rate": 8.31343712827047e-06,
+      "loss": 0.0712,
+      "step": 875,
+      "student_loss": 0.00016646686708554626,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2537485582468281,
+      "grad_norm": 10.3125,
+      "kd_loss": 0.006990634836256504,
+      "learning_rate": 8.293176526754274e-06,
+      "loss": 0.0956,
+      "step": 880,
+      "student_loss": 0.00040896443533711135,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.25519031141868515,
+      "grad_norm": 6.96875,
+      "kd_loss": 0.19721470773220062,
+      "learning_rate": 8.27281997021091e-06,
+      "loss": 0.0412,
+      "step": 885,
+      "student_loss": 0.0062887901440262794,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2566320645905421,
+      "grad_norm": 9.9375,
+      "kd_loss": 0.0008049776079133153,
+      "learning_rate": 8.25236805178014e-06,
+      "loss": 0.0867,
+      "step": 890,
+      "student_loss": 0.0003136158047709614,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.25807381776239907,
+      "grad_norm": 19.375,
+      "kd_loss": 0.12581539154052734,
+      "learning_rate": 8.231821367380335e-06,
+      "loss": 0.1217,
+      "step": 895,
+      "student_loss": 0.008851185441017151,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.25951557093425603,
+      "grad_norm": 6.8125,
+      "kd_loss": 0.0027310107834637165,
+      "learning_rate": 8.211180515691108e-06,
+      "loss": 0.0815,
+      "step": 900,
+      "student_loss": 0.03639969602227211,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.26095732410611305,
+      "grad_norm": 5.96875,
+      "kd_loss": 0.12574803829193115,
+      "learning_rate": 8.190446098135883e-06,
+      "loss": 0.0819,
+      "step": 905,
+      "student_loss": 0.004850549623370171,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.26239907727797,
+      "grad_norm": 8.9375,
+      "kd_loss": 0.0006859266432002187,
+      "learning_rate": 8.16961871886435e-06,
+      "loss": 0.0665,
+      "step": 910,
+      "student_loss": 0.00025130261201411486,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.263840830449827,
+      "grad_norm": 12.1875,
+      "kd_loss": 0.3031837046146393,
+      "learning_rate": 8.148698984734887e-06,
+      "loss": 0.0546,
+      "step": 915,
+      "student_loss": 0.002404158003628254,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.26528258362168394,
+      "grad_norm": 8.1875,
+      "kd_loss": 0.03748767077922821,
+      "learning_rate": 8.127687505296852e-06,
+      "loss": 0.0541,
+      "step": 920,
+      "student_loss": 0.014545704238116741,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.26672433679354096,
+      "grad_norm": 2.8125,
+      "kd_loss": 0.0011403196258470416,
+      "learning_rate": 8.106584892772844e-06,
+      "loss": 0.0912,
+      "step": 925,
+      "student_loss": 0.0002578723942860961,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2681660899653979,
+      "grad_norm": 10.0625,
+      "kd_loss": 0.0013566795969381928,
+      "learning_rate": 8.08539176204085e-06,
+      "loss": 0.0604,
+      "step": 930,
+      "student_loss": 0.007105275988578796,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2696078431372549,
+      "grad_norm": 3.84375,
+      "kd_loss": 0.03631063550710678,
+      "learning_rate": 8.064108730616336e-06,
+      "loss": 0.055,
+      "step": 935,
+      "student_loss": 0.0022572344169020653,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2710495963091119,
+      "grad_norm": 5.09375,
+      "kd_loss": 0.030622560530900955,
+      "learning_rate": 8.042736418634252e-06,
+      "loss": 0.081,
+      "step": 940,
+      "student_loss": 0.00955814030021429,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.27249134948096887,
+      "grad_norm": 3.96875,
+      "kd_loss": 0.004221701063215733,
+      "learning_rate": 8.02127544883096e-06,
+      "loss": 0.0567,
+      "step": 945,
+      "student_loss": 0.009679676033556461,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.27393310265282583,
+      "grad_norm": 2.078125,
+      "kd_loss": 0.004411423113197088,
+      "learning_rate": 7.999726446526095e-06,
+      "loss": 0.0472,
+      "step": 950,
+      "student_loss": 0.00019772372615989298,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2753748558246828,
+      "grad_norm": 4.71875,
+      "kd_loss": 0.003913923166692257,
+      "learning_rate": 7.978090039604342e-06,
+      "loss": 0.0528,
+      "step": 955,
+      "student_loss": 0.0007541932282038033,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2768166089965398,
+      "grad_norm": 2.875,
+      "kd_loss": 0.030727535486221313,
+      "learning_rate": 7.956366858497138e-06,
+      "loss": 0.0601,
+      "step": 960,
+      "student_loss": 0.00035272131208330393,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2782583621683968,
+      "grad_norm": 17.125,
+      "kd_loss": 0.00199729367159307,
+      "learning_rate": 7.934557536164304e-06,
+      "loss": 0.0965,
+      "step": 965,
+      "student_loss": 0.01171483937650919,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.27970011534025374,
+      "grad_norm": 10.875,
+      "kd_loss": 0.0011206807103008032,
+      "learning_rate": 7.91266270807561e-06,
+      "loss": 0.0549,
+      "step": 970,
+      "student_loss": 0.00026892279856838286,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2811418685121107,
+      "grad_norm": 9.4375,
+      "kd_loss": 0.004792070481926203,
+      "learning_rate": 7.890683012192249e-06,
+      "loss": 0.0831,
+      "step": 975,
+      "student_loss": 0.016710802912712097,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2825836216839677,
+      "grad_norm": 1.1171875,
+      "kd_loss": 0.0018690243596211076,
+      "learning_rate": 7.868619088948246e-06,
+      "loss": 0.0694,
+      "step": 980,
+      "student_loss": 0.00033925892785191536,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2840253748558247,
+      "grad_norm": 11.3125,
+      "kd_loss": 0.00043922787881456316,
+      "learning_rate": 7.846471581231814e-06,
+      "loss": 0.0819,
+      "step": 985,
+      "student_loss": 0.0002598507853690535,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.28546712802768165,
+      "grad_norm": 2.390625,
+      "kd_loss": 0.01619037427008152,
+      "learning_rate": 7.824241134366604e-06,
+      "loss": 0.0307,
+      "step": 990,
+      "student_loss": 0.0017275860300287604,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.2869088811995386,
+      "grad_norm": 10.25,
+      "kd_loss": 0.005464199930429459,
+      "learning_rate": 7.801928396092908e-06,
+      "loss": 0.0678,
+      "step": 995,
+      "student_loss": 0.001406969386152923,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.28835063437139563,
+      "grad_norm": 7.25,
+      "kd_loss": 0.013948436826467514,
+      "learning_rate": 7.779534016548791e-06,
+      "loss": 0.0719,
+      "step": 1000,
+      "student_loss": 0.0060744076035916805,
+      "teacher_loss": 0.0
+    }
+  ],
+  "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/codi_frozen_3b_logit/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/codi_frozen_3b_logit/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/codi_frozen_3b_logit/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/codi_frozen_3b_logit/checkpoint-500/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d474fc70ed16a4da714894b40d2c5120d98b72f84f047ce1fb21b151e5a0a94f
+size 6187858991

checkpoints/codi_frozen_3b_logit/checkpoint-500/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/codi_frozen_3b_logit/checkpoint-500/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0740be2f16e95f428d7d914c7179aa2f3693e150a0d40fa09f9855cabdbb668c
+size 16788033

checkpoints/codi_frozen_3b_logit/checkpoint-500/tokenizer.json

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

checkpoints/codi_frozen_3b_logit/checkpoint-500/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/codi_frozen_3b_logit/checkpoint-500/trainer_state.json

@@ -0,0 +1,1034 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.14417531718569782,
+  "eval_steps": 500,
+  "global_step": 500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0014417531718569781,
+      "grad_norm": 1584.0,
+      "kd_loss": 1.4788804054260254,
+      "learning_rate": 4.444444444444445e-07,
+      "loss": 2.2647,
+      "step": 5,
+      "student_loss": 0.12291660904884338,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.0028835063437139563,
+      "grad_norm": 796.0,
+      "kd_loss": 1.487825870513916,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 1.9392,
+      "step": 10,
+      "student_loss": 0.198484867811203,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.004325259515570935,
+      "grad_norm": 260.0,
+      "kd_loss": 0.07314793765544891,
+      "learning_rate": 1.5555555555555558e-06,
+      "loss": 1.2376,
+      "step": 15,
+      "student_loss": 0.08613421022891998,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.0057670126874279125,
+      "grad_norm": 85.0,
+      "kd_loss": 0.023559529334306717,
+      "learning_rate": 2.1111111111111114e-06,
+      "loss": 0.6343,
+      "step": 20,
+      "student_loss": 0.23392575979232788,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.00720876585928489,
+      "grad_norm": 33.75,
+      "kd_loss": 0.0026166969910264015,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 0.6107,
+      "step": 25,
+      "student_loss": 0.2554866075515747,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.00865051903114187,
+      "grad_norm": 24.125,
+      "kd_loss": 0.022094376385211945,
+      "learning_rate": 3.2222222222222227e-06,
+      "loss": 0.498,
+      "step": 30,
+      "student_loss": 0.2493494749069214,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.010092272202998846,
+      "grad_norm": 39.0,
+      "kd_loss": 0.04670819640159607,
+      "learning_rate": 3.777777777777778e-06,
+      "loss": 0.3781,
+      "step": 35,
+      "student_loss": 0.13782167434692383,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.011534025374855825,
+      "grad_norm": 18.125,
+      "kd_loss": 0.24180860817432404,
+      "learning_rate": 4.333333333333334e-06,
+      "loss": 0.3341,
+      "step": 40,
+      "student_loss": 0.09410303086042404,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.012975778546712802,
+      "grad_norm": 20.25,
+      "kd_loss": 0.4184195101261139,
+      "learning_rate": 4.888888888888889e-06,
+      "loss": 0.2388,
+      "step": 45,
+      "student_loss": 0.09438568353652954,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.01441753171856978,
+      "grad_norm": 32.75,
+      "kd_loss": 0.8934061527252197,
+      "learning_rate": 5.444444444444445e-06,
+      "loss": 0.3727,
+      "step": 50,
+      "student_loss": 0.020176472142338753,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.015859284890426758,
+      "grad_norm": 28.5,
+      "kd_loss": 0.018707511946558952,
+      "learning_rate": 6e-06,
+      "loss": 0.2167,
+      "step": 55,
+      "student_loss": 0.0011619763681665063,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.01730103806228374,
+      "grad_norm": 27.25,
+      "kd_loss": 0.007137665059417486,
+      "learning_rate": 6.555555555555556e-06,
+      "loss": 0.1682,
+      "step": 60,
+      "student_loss": 0.001694640377536416,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.018742791234140715,
+      "grad_norm": 15.75,
+      "kd_loss": 0.0016208974411711097,
+      "learning_rate": 7.111111111111112e-06,
+      "loss": 0.1815,
+      "step": 65,
+      "student_loss": 0.029718520119786263,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.020184544405997693,
+      "grad_norm": 12.1875,
+      "kd_loss": 0.2541334331035614,
+      "learning_rate": 7.666666666666667e-06,
+      "loss": 0.1369,
+      "step": 70,
+      "student_loss": 0.010810595005750656,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.02162629757785467,
+      "grad_norm": 18.375,
+      "kd_loss": 0.007482057902961969,
+      "learning_rate": 8.222222222222222e-06,
+      "loss": 0.2322,
+      "step": 75,
+      "student_loss": 0.00097510009072721,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.02306805074971165,
+      "grad_norm": 10.3125,
+      "kd_loss": 0.006171244662255049,
+      "learning_rate": 8.777777777777778e-06,
+      "loss": 0.1828,
+      "step": 80,
+      "student_loss": 0.000972377136349678,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.024509803921568627,
+      "grad_norm": 18.125,
+      "kd_loss": 0.005768964998424053,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.1856,
+      "step": 85,
+      "student_loss": 0.0005292362766340375,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.025951557093425604,
+      "grad_norm": 5.65625,
+      "kd_loss": 0.060487739741802216,
+      "learning_rate": 9.88888888888889e-06,
+      "loss": 0.1726,
+      "step": 90,
+      "student_loss": 0.007604286074638367,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.027393310265282585,
+      "grad_norm": 13.5625,
+      "kd_loss": 0.01542923878878355,
+      "learning_rate": 9.99995337991255e-06,
+      "loss": 0.1586,
+      "step": 95,
+      "student_loss": 0.015073101036250591,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.02883506343713956,
+      "grad_norm": 15.8125,
+      "kd_loss": 0.0029005378019064665,
+      "learning_rate": 9.999763987297266e-06,
+      "loss": 0.1247,
+      "step": 100,
+      "student_loss": 0.0006029795622453094,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.03027681660899654,
+      "grad_norm": 15.375,
+      "kd_loss": 0.007598078344017267,
+      "learning_rate": 9.999428913912876e-06,
+      "loss": 0.1266,
+      "step": 105,
+      "student_loss": 0.003934374079108238,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.031718569780853516,
+      "grad_norm": 16.0,
+      "kd_loss": 0.00252303434535861,
+      "learning_rate": 9.998948169522587e-06,
+      "loss": 0.1767,
+      "step": 110,
+      "student_loss": 0.04128251224756241,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.03316032295271049,
+      "grad_norm": 13.3125,
+      "kd_loss": 0.19726258516311646,
+      "learning_rate": 9.998321768134101e-06,
+      "loss": 0.1655,
+      "step": 115,
+      "student_loss": 0.001500162878073752,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.03460207612456748,
+      "grad_norm": 5.875,
+      "kd_loss": 0.008542308583855629,
+      "learning_rate": 9.997549727999212e-06,
+      "loss": 0.0858,
+      "step": 120,
+      "student_loss": 0.0007994729094207287,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.036043829296424454,
+      "grad_norm": 14.875,
+      "kd_loss": 0.7844982743263245,
+      "learning_rate": 9.996632071613257e-06,
+      "loss": 0.1328,
+      "step": 125,
+      "student_loss": 0.029885753989219666,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.03748558246828143,
+      "grad_norm": 14.0625,
+      "kd_loss": 0.011637737974524498,
+      "learning_rate": 9.995568825714479e-06,
+      "loss": 0.1454,
+      "step": 130,
+      "student_loss": 0.0005009593442082405,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.03892733564013841,
+      "grad_norm": 19.5,
+      "kd_loss": 0.12578554451465607,
+      "learning_rate": 9.994360021283235e-06,
+      "loss": 0.1434,
+      "step": 135,
+      "student_loss": 0.0007728396449238062,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.040369088811995385,
+      "grad_norm": 4.0625,
+      "kd_loss": 0.0028691363986581564,
+      "learning_rate": 9.993005693541098e-06,
+      "loss": 0.1367,
+      "step": 140,
+      "student_loss": 0.0006042163586243987,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.04181084198385236,
+      "grad_norm": 11.0,
+      "kd_loss": 1.0329420566558838,
+      "learning_rate": 9.991505881949837e-06,
+      "loss": 0.1066,
+      "step": 145,
+      "student_loss": 0.041140101850032806,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.04325259515570934,
+      "grad_norm": 7.71875,
+      "kd_loss": 0.013778705149888992,
+      "learning_rate": 9.989860630210251e-06,
+      "loss": 0.1259,
+      "step": 150,
+      "student_loss": 0.0009030756773427129,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.04469434832756632,
+      "grad_norm": 4.15625,
+      "kd_loss": 0.028440266847610474,
+      "learning_rate": 9.98806998626091e-06,
+      "loss": 0.1208,
+      "step": 155,
+      "student_loss": 0.0017314451979473233,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.0461361014994233,
+      "grad_norm": 20.75,
+      "kd_loss": 0.005404896102845669,
+      "learning_rate": 9.98613400227676e-06,
+      "loss": 0.1872,
+      "step": 160,
+      "student_loss": 0.10157372802495956,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.04757785467128028,
+      "grad_norm": 26.25,
+      "kd_loss": 0.008982345461845398,
+      "learning_rate": 9.984052734667586e-06,
+      "loss": 0.1018,
+      "step": 165,
+      "student_loss": 0.00033849035389721394,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.049019607843137254,
+      "grad_norm": 18.125,
+      "kd_loss": 0.005132909398525953,
+      "learning_rate": 9.981826244076387e-06,
+      "loss": 0.1339,
+      "step": 170,
+      "student_loss": 0.0005440761451609433,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.05046136101499423,
+      "grad_norm": 18.0,
+      "kd_loss": 0.021587209776043892,
+      "learning_rate": 9.979454595377594e-06,
+      "loss": 0.1247,
+      "step": 175,
+      "student_loss": 0.014186372049152851,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.05190311418685121,
+      "grad_norm": 8.5,
+      "kd_loss": 0.011625917628407478,
+      "learning_rate": 9.97693785767519e-06,
+      "loss": 0.0741,
+      "step": 180,
+      "student_loss": 0.0005693296552635729,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.05334486735870819,
+      "grad_norm": 7.21875,
+      "kd_loss": 0.0012370528420433402,
+      "learning_rate": 9.9742761043007e-06,
+      "loss": 0.1795,
+      "step": 185,
+      "student_loss": 0.043101612478494644,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.05478662053056517,
+      "grad_norm": 7.34375,
+      "kd_loss": 0.1722288429737091,
+      "learning_rate": 9.971469412811032e-06,
+      "loss": 0.093,
+      "step": 190,
+      "student_loss": 0.006316277664154768,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.056228373702422146,
+      "grad_norm": 44.25,
+      "kd_loss": 0.6569585800170898,
+      "learning_rate": 9.968517864986247e-06,
+      "loss": 0.1118,
+      "step": 195,
+      "student_loss": 0.023583440110087395,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.05767012687427912,
+      "grad_norm": 4.9375,
+      "kd_loss": 0.0010104529792442918,
+      "learning_rate": 9.965421546827153e-06,
+      "loss": 0.1094,
+      "step": 200,
+      "student_loss": 0.0004585790156852454,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.0591118800461361,
+      "grad_norm": 4.59375,
+      "kd_loss": 0.0006818274850957096,
+      "learning_rate": 9.962180548552812e-06,
+      "loss": 0.0885,
+      "step": 205,
+      "student_loss": 0.046175092458724976,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.06055363321799308,
+      "grad_norm": 58.0,
+      "kd_loss": 0.4983484745025635,
+      "learning_rate": 9.958794964597906e-06,
+      "loss": 0.1317,
+      "step": 210,
+      "student_loss": 0.007553795352578163,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.061995386389850055,
+      "grad_norm": 6.46875,
+      "kd_loss": 0.0009208342526108027,
+      "learning_rate": 9.955264893609986e-06,
+      "loss": 0.1522,
+      "step": 215,
+      "student_loss": 0.00047464462113566697,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.06343713956170703,
+      "grad_norm": 18.125,
+      "kd_loss": 0.0944160670042038,
+      "learning_rate": 9.951590438446597e-06,
+      "loss": 0.0774,
+      "step": 220,
+      "student_loss": 0.03491729497909546,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.06487889273356401,
+      "grad_norm": 10.25,
+      "kd_loss": 0.001832562847994268,
+      "learning_rate": 9.947771706172285e-06,
+      "loss": 0.0976,
+      "step": 225,
+      "student_loss": 0.05962345376610756,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.06632064590542099,
+      "grad_norm": 21.75,
+      "kd_loss": 0.10277731716632843,
+      "learning_rate": 9.943808808055472e-06,
+      "loss": 0.0911,
+      "step": 230,
+      "student_loss": 0.0012677249033004045,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.06776239907727798,
+      "grad_norm": 17.25,
+      "kd_loss": 0.4876050353050232,
+      "learning_rate": 9.93970185956522e-06,
+      "loss": 0.1077,
+      "step": 235,
+      "student_loss": 0.006362705025821924,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.06920415224913495,
+      "grad_norm": 1.46875,
+      "kd_loss": 0.010914070531725883,
+      "learning_rate": 9.93545098036786e-06,
+      "loss": 0.1107,
+      "step": 240,
+      "student_loss": 0.0032385734375566244,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07064590542099193,
+      "grad_norm": 10.3125,
+      "kd_loss": 0.02172888070344925,
+      "learning_rate": 9.931056294323507e-06,
+      "loss": 0.081,
+      "step": 245,
+      "student_loss": 0.0005711687845177948,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07208765859284891,
+      "grad_norm": 5.125,
+      "kd_loss": 0.005143388640135527,
+      "learning_rate": 9.926517929482454e-06,
+      "loss": 0.1052,
+      "step": 250,
+      "student_loss": 0.0005271504633128643,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07352941176470588,
+      "grad_norm": 6.5,
+      "kd_loss": 0.0008525780867785215,
+      "learning_rate": 9.921836018081438e-06,
+      "loss": 0.0589,
+      "step": 255,
+      "student_loss": 0.0003723589761648327,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07497116493656286,
+      "grad_norm": 17.0,
+      "kd_loss": 0.0010641744593158364,
+      "learning_rate": 9.917010696539793e-06,
+      "loss": 0.1453,
+      "step": 260,
+      "student_loss": 0.0432959608733654,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07641291810841984,
+      "grad_norm": 16.0,
+      "kd_loss": 0.001827583066187799,
+      "learning_rate": 9.912042105455462e-06,
+      "loss": 0.0908,
+      "step": 265,
+      "student_loss": 0.00034234311897307634,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07785467128027682,
+      "grad_norm": 14.0,
+      "kd_loss": 0.000661063939332962,
+      "learning_rate": 9.906930389600917e-06,
+      "loss": 0.0894,
+      "step": 270,
+      "student_loss": 0.0010115442564710975,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.07929642445213379,
+      "grad_norm": 13.0,
+      "kd_loss": 0.0016069641569629312,
+      "learning_rate": 9.901675697918925e-06,
+      "loss": 0.1173,
+      "step": 275,
+      "student_loss": 0.05398955196142197,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.08073817762399077,
+      "grad_norm": 29.0,
+      "kd_loss": 0.0010736712720245123,
+      "learning_rate": 9.896278183518216e-06,
+      "loss": 0.1427,
+      "step": 280,
+      "student_loss": 0.0005599395371973515,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.08217993079584775,
+      "grad_norm": 17.5,
+      "kd_loss": 0.017228417098522186,
+      "learning_rate": 9.890738003669029e-06,
+      "loss": 0.1333,
+      "step": 285,
+      "student_loss": 0.0005250528338365257,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.08362168396770472,
+      "grad_norm": 35.5,
+      "kd_loss": 0.0026075318455696106,
+      "learning_rate": 9.885055319798514e-06,
+      "loss": 0.1752,
+      "step": 290,
+      "student_loss": 0.0017553431680426002,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.0850634371395617,
+      "grad_norm": 26.25,
+      "kd_loss": 0.004095411393791437,
+      "learning_rate": 9.879230297486034e-06,
+      "loss": 0.158,
+      "step": 295,
+      "student_loss": 0.0005752203287556767,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.08650519031141868,
+      "grad_norm": 4.96875,
+      "kd_loss": 0.2717151343822479,
+      "learning_rate": 9.873263106458353e-06,
+      "loss": 0.0941,
+      "step": 300,
+      "student_loss": 0.023448897525668144,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.08794694348327567,
+      "grad_norm": 6.65625,
+      "kd_loss": 0.019775602966547012,
+      "learning_rate": 9.86715392058467e-06,
+      "loss": 0.1302,
+      "step": 305,
+      "student_loss": 0.0025525283999741077,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.08938869665513265,
+      "grad_norm": 7.0,
+      "kd_loss": 0.0005424432456493378,
+      "learning_rate": 9.860902917871566e-06,
+      "loss": 0.1096,
+      "step": 310,
+      "student_loss": 0.000471778039354831,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.09083044982698962,
+      "grad_norm": 10.25,
+      "kd_loss": 0.021720539778470993,
+      "learning_rate": 9.854510280457819e-06,
+      "loss": 0.0897,
+      "step": 315,
+      "student_loss": 0.015475576743483543,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.0922722029988466,
+      "grad_norm": 6.40625,
+      "kd_loss": 0.0012843874283134937,
+      "learning_rate": 9.847976194609083e-06,
+      "loss": 0.091,
+      "step": 320,
+      "student_loss": 0.0004686483880504966,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.09371395617070358,
+      "grad_norm": 26.375,
+      "kd_loss": 0.6982781887054443,
+      "learning_rate": 9.841300850712479e-06,
+      "loss": 0.103,
+      "step": 325,
+      "student_loss": 0.017696071416139603,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.09515570934256055,
+      "grad_norm": 8.6875,
+      "kd_loss": 0.15215124189853668,
+      "learning_rate": 9.834484443271035e-06,
+      "loss": 0.0715,
+      "step": 330,
+      "student_loss": 0.005886889062821865,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.09659746251441753,
+      "grad_norm": 2.875,
+      "kd_loss": 0.0015391090419143438,
+      "learning_rate": 9.827527170898019e-06,
+      "loss": 0.0852,
+      "step": 335,
+      "student_loss": 0.0005516717210412025,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 15.0,
+      "kd_loss": 0.07792548835277557,
+      "learning_rate": 9.820429236311158e-06,
+      "loss": 0.1289,
+      "step": 340,
+      "student_loss": 0.010181530378758907,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.09948096885813149,
+      "grad_norm": 6.0625,
+      "kd_loss": 0.013684757985174656,
+      "learning_rate": 9.81319084632673e-06,
+      "loss": 0.093,
+      "step": 345,
+      "student_loss": 0.01164299063384533,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.10092272202998846,
+      "grad_norm": 5.3125,
+      "kd_loss": 0.0955321341753006,
+      "learning_rate": 9.805812211853534e-06,
+      "loss": 0.0668,
+      "step": 350,
+      "student_loss": 0.0006315007922239602,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.10236447520184544,
+      "grad_norm": 15.75,
+      "kd_loss": 0.000853118603117764,
+      "learning_rate": 9.798293547886748e-06,
+      "loss": 0.1354,
+      "step": 355,
+      "student_loss": 0.00041306763887405396,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.10380622837370242,
+      "grad_norm": 4.125,
+      "kd_loss": 0.0023229331709444523,
+      "learning_rate": 9.790635073501662e-06,
+      "loss": 0.0929,
+      "step": 360,
+      "student_loss": 0.0004519273934420198,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1052479815455594,
+      "grad_norm": 12.875,
+      "kd_loss": 0.004534870386123657,
+      "learning_rate": 9.782837011847294e-06,
+      "loss": 0.177,
+      "step": 365,
+      "student_loss": 0.048408832401037216,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.10668973471741638,
+      "grad_norm": 5.75,
+      "kd_loss": 0.5334176421165466,
+      "learning_rate": 9.774899590139897e-06,
+      "loss": 0.0711,
+      "step": 370,
+      "student_loss": 0.0273645780980587,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.10813148788927336,
+      "grad_norm": 12.25,
+      "kd_loss": 0.006150795612484217,
+      "learning_rate": 9.766823039656328e-06,
+      "loss": 0.1286,
+      "step": 375,
+      "student_loss": 0.03481242433190346,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.10957324106113034,
+      "grad_norm": 0.69140625,
+      "kd_loss": 0.0020576389506459236,
+      "learning_rate": 9.758607595727309e-06,
+      "loss": 0.0763,
+      "step": 380,
+      "student_loss": 0.0004342672764323652,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.11101499423298732,
+      "grad_norm": 3.375,
+      "kd_loss": 0.01198313757777214,
+      "learning_rate": 9.75025349773058e-06,
+      "loss": 0.0684,
+      "step": 385,
+      "student_loss": 0.00040114219882525504,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.11245674740484429,
+      "grad_norm": 42.25,
+      "kd_loss": 0.0029099821113049984,
+      "learning_rate": 9.741760989083914e-06,
+      "loss": 0.0966,
+      "step": 390,
+      "student_loss": 0.00024992373073473573,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.11389850057670127,
+      "grad_norm": 10.4375,
+      "kd_loss": 0.0029170780908316374,
+      "learning_rate": 9.73313031723803e-06,
+      "loss": 0.0846,
+      "step": 395,
+      "student_loss": 0.0004683209699578583,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.11534025374855825,
+      "grad_norm": 2.390625,
+      "kd_loss": 0.002679860917851329,
+      "learning_rate": 9.724361733669383e-06,
+      "loss": 0.0425,
+      "step": 400,
+      "student_loss": 0.002333736512809992,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.11678200692041522,
+      "grad_norm": 11.6875,
+      "kd_loss": 0.0017421097727492452,
+      "learning_rate": 9.715455493872827e-06,
+      "loss": 0.1317,
+      "step": 405,
+      "student_loss": 0.011753321625292301,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1182237600922722,
+      "grad_norm": 5.96875,
+      "kd_loss": 0.0038202686700969934,
+      "learning_rate": 9.70641185735419e-06,
+      "loss": 0.0947,
+      "step": 410,
+      "student_loss": 0.015384526923298836,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.11966551326412918,
+      "grad_norm": 15.625,
+      "kd_loss": 0.007967401295900345,
+      "learning_rate": 9.697231087622691e-06,
+      "loss": 0.1609,
+      "step": 415,
+      "student_loss": 0.003767178161069751,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.12110726643598616,
+      "grad_norm": 7.78125,
+      "kd_loss": 0.3626428246498108,
+      "learning_rate": 9.687913452183277e-06,
+      "loss": 0.0883,
+      "step": 420,
+      "student_loss": 0.06143607571721077,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.12254901960784313,
+      "grad_norm": 9.1875,
+      "kd_loss": 0.0016940389759838581,
+      "learning_rate": 9.678459222528825e-06,
+      "loss": 0.0493,
+      "step": 425,
+      "student_loss": 0.002240637084469199,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.12399077277970011,
+      "grad_norm": 19.75,
+      "kd_loss": 0.9260356426239014,
+      "learning_rate": 9.668868674132224e-06,
+      "loss": 0.1153,
+      "step": 430,
+      "student_loss": 0.0827086940407753,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1254325259515571,
+      "grad_norm": 12.8125,
+      "kd_loss": 0.0019437845330685377,
+      "learning_rate": 9.65914208643836e-06,
+      "loss": 0.0634,
+      "step": 435,
+      "student_loss": 0.0006262208335101604,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.12687427912341406,
+      "grad_norm": 4.25,
+      "kd_loss": 0.00489834975451231,
+      "learning_rate": 9.649279742855964e-06,
+      "loss": 0.103,
+      "step": 440,
+      "student_loss": 0.013814084231853485,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.12831603229527105,
+      "grad_norm": 22.375,
+      "kd_loss": 0.0016376337734982371,
+      "learning_rate": 9.639281930749363e-06,
+      "loss": 0.079,
+      "step": 445,
+      "student_loss": 0.03539160266518593,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.12975778546712802,
+      "grad_norm": 7.375,
+      "kd_loss": 0.0012233811430633068,
+      "learning_rate": 9.629148941430096e-06,
+      "loss": 0.1054,
+      "step": 450,
+      "student_loss": 0.0006881144363433123,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.131199538638985,
+      "grad_norm": 6.0,
+      "kd_loss": 0.0003836078103631735,
+      "learning_rate": 9.618881070148436e-06,
+      "loss": 0.1035,
+      "step": 455,
+      "student_loss": 0.001039310242049396,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.13264129181084197,
+      "grad_norm": 7.15625,
+      "kd_loss": 0.039876699447631836,
+      "learning_rate": 9.608478616084784e-06,
+      "loss": 0.0547,
+      "step": 460,
+      "student_loss": 0.008442185819149017,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.13408304498269896,
+      "grad_norm": 15.4375,
+      "kd_loss": 0.08534516394138336,
+      "learning_rate": 9.597941882340947e-06,
+      "loss": 0.1407,
+      "step": 465,
+      "student_loss": 0.00036318166530691087,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.13552479815455595,
+      "grad_norm": 9.125,
+      "kd_loss": 0.000860957894474268,
+      "learning_rate": 9.587271175931313e-06,
+      "loss": 0.0872,
+      "step": 470,
+      "student_loss": 0.00031289560138247907,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.13696655132641292,
+      "grad_norm": 3.21875,
+      "kd_loss": 0.021515244618058205,
+      "learning_rate": 9.5764668077739e-06,
+      "loss": 0.0627,
+      "step": 475,
+      "student_loss": 0.0004178633389528841,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.1384083044982699,
+      "grad_norm": 11.5625,
+      "kd_loss": 0.01755513623356819,
+      "learning_rate": 9.5655290926813e-06,
+      "loss": 0.0742,
+      "step": 480,
+      "student_loss": 0.041500866413116455,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.13985005767012687,
+      "grad_norm": 18.875,
+      "kd_loss": 0.0021918262355029583,
+      "learning_rate": 9.554458349351507e-06,
+      "loss": 0.0853,
+      "step": 485,
+      "student_loss": 0.0005753549048677087,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.14129181084198386,
+      "grad_norm": 16.0,
+      "kd_loss": 0.005263358820229769,
+      "learning_rate": 9.54325490035863e-06,
+      "loss": 0.1613,
+      "step": 490,
+      "student_loss": 0.0003423037414904684,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.14273356401384082,
+      "grad_norm": 6.75,
+      "kd_loss": 0.0663486197590828,
+      "learning_rate": 9.531919072143488e-06,
+      "loss": 0.0925,
+      "step": 495,
+      "student_loss": 0.0005914232460781932,
+      "teacher_loss": 0.0
+    },
+    {
+      "epoch": 0.14417531718569782,
+      "grad_norm": 7.40625,
+      "kd_loss": 0.032540202140808105,
+      "learning_rate": 9.520451195004107e-06,
+      "loss": 0.0773,
+      "step": 500,
+      "student_loss": 0.00025544269010424614,
+      "teacher_loss": 0.0
+    }
+  ],
+  "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/codi_frozen_3b_logit/wandb/debug-internal.log

@@ -0,0 +1,11 @@
+{"time":"2026-06-25T13:55:14.865357126+02:00","level":"INFO","msg":"wandb-core"}
+{"time":"2026-06-25T13:55:14.865374279+02:00","level":"INFO","msg":"stream: starting","core version":"0.27.2"}
+{"time":"2026-06-25T13:55:14.988281216+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-25T13:55:14.988307148+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-25T13:55:14.988325352+02:00","level":"INFO","msg":"stream: created new stream","id":"i10qt7cx"}
+{"time":"2026-06-25T13:55:14.988514907+02:00","level":"INFO","msg":"handler: started"}
+{"time":"2026-06-25T13:55:14.989316587+02:00","level":"INFO","msg":"stream: started"}
+{"time":"2026-06-25T13:55:14.989459496+02:00","level":"INFO","msg":"writer: started","stream_id":"i10qt7cx"}
+{"time":"2026-06-25T13:55:14.989483136+02:00","level":"INFO","msg":"sender: started"}
+{"time":"2026-06-25T13:55:15.00239695+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-25T13:55:15.002412933+02:00","level":"WARN","msg":"runupserter: server does not expand metric globs but the x_server_side_expand_glob_metrics setting is set; ignoring"}

checkpoints/codi_frozen_3b_logit/wandb/debug.log

@@ -0,0 +1,20 @@
+2026-06-25 13:55:13,760 INFO    MainThread:1545486 [wandb_setup.py:_flush():81] Current SDK version is 0.27.2
+2026-06-25 13:55:13,760 INFO    MainThread:1545486 [wandb_setup.py:_flush():81] Configure stats pid to 1545486
+2026-06-25 13:55:13,760 INFO    MainThread:1545486 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-06-25 13:55:13,761 INFO    MainThread:1545486 [wandb_init.py:setup_run_log_directory():723] Logging user logs to model_weights/codi_frozen_3b_logit/wandb/offline-run-20260625_135513-i10qt7cx/logs/debug.log
+2026-06-25 13:55:13,761 INFO    MainThread:1545486 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to model_weights/codi_frozen_3b_logit/wandb/offline-run-20260625_135513-i10qt7cx/logs/debug-internal.log
+2026-06-25 13:55:13,761 INFO    MainThread:1545486 [wandb_init.py:init():766] calling init triggers
+2026-06-25 13:55:13,761 INFO    MainThread:1545486 [wandb_init.py:init():771] wandb.init called with sweep_config: {}
+config: {'model': 'model_weights/sft3b_lr1e5_bs32/checkpoint-3000', 'output_dir': 'model_weights/codi_frozen_3b_logit', 'sources': ['mbpp', 'humaneval', 'pyx'], 'cache_dir': 'data/cache/codi_train', 'n_samples': -1, 'max_seq_len': 3072, 'max_frames': -1, 'latent_steps': 1, 'epochs': 5.0, 'lr': 1e-05, 'batch_size': 1, 'grad_accum': 4, 'max_steps': 2500, 'save_steps': 500, 'alpha': 0.0, 'beta': 1.0, 'gamma': 1.0, 'kd_layers': None, 'frozen_teacher': 'model_weights/sft3b_lr1e5_bs32/checkpoint-3000', 'kd_target': 'logit', 'kd_temp': 2.0, 'single_anchor': False, 'ss_prob': 0.0, 'ss_ramp_frac': 0.5, '_wandb': {}}
+2026-06-25 13:55:13,761 INFO    MainThread:1545486 [wandb_init.py:init():814] starting backend
+2026-06-25 13:55:14,850 INFO    MainThread:1545486 [wandb_init.py:init():829] sending inform_init request
+2026-06-25 13:55:14,990 INFO    MainThread:1545486 [wandb_init.py:init():834] backend started and connected
+2026-06-25 13:55:14,991 INFO    MainThread:1545486 [wandb_init.py:init():904] updated telemetry
+2026-06-25 13:55:15,000 INFO    MainThread:1545486 [wandb_init.py:init():927] communicating run to backend with 90.0 second timeout
+2026-06-25 13:55:15,003 INFO    MainThread:1545486 [wandb_init.py:init():972] starting run threads in backend
+2026-06-25 13:55:15,387 INFO    MainThread:1545486 [wandb_run.py:_console_start():2523] atexit reg
+2026-06-25 13:55:15,387 INFO    MainThread:1545486 [wandb_run.py:_redirect():2373] redirect: wrap_raw
+2026-06-25 13:55:15,387 INFO    MainThread:1545486 [wandb_run.py:_redirect():2442] Wrapping output streams.
+2026-06-25 13:55:15,387 INFO    MainThread:1545486 [wandb_run.py:_redirect():2465] Redirects installed.
+2026-06-25 13:55:15,393 INFO    MainThread:1545486 [wandb_init.py:init():1010] run started, returning control to user process
+2026-06-25 13:55:25,930 INFO    MainThread:1545486 [wandb_run.py:_config_callback():1346] config_cb None None {'output_dir': 'model_weights/codi_frozen_3b_logit', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': 'no', 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 8, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 4, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.1, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 5.0, 'max_steps': 2500, 'lr_scheduler_type': 'cosine', 'lr_scheduler_kwargs': None, 'warmup_ratio': 0.03, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'model_weights/codi_frozen_3b_logit/runs/Jun25_13-55-15_alvis4-42', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 5, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': None, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': None, 'disable_tqdm': False, 'remove_unused_columns': False, 'label_names': [], 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'paged_adamw_8bit', 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'project': 'huggingface', 'trackio_space_id': 'trackio', 'ddp_find_unused_parameters': False, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'hub_revision': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': 'no', 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'liger_kernel_config': None, 'eval_use_gather_object': False, 'average_tokens_across_devices': True}

checkpoints/codi_frozen_3b_logit/wandb/offline-run-20260624_183654-h7tylfot/files/requirements.txt

@@ -0,0 +1,229 @@
+wheel==0.47.0
+triton==3.3.1
+ptyprocess==0.7.0
+kiwisolver==1.5.0
+pyzmq==27.1.0
+nvidia-cublas-cu12==12.6.4.1
+python-dateutil==2.9.0.post0
+cycler==0.12.1
+rfc3339-validator==0.1.4
+pluggy==1.6.0
+grpclib==0.4.9
+attrs==26.1.0
+urllib3==2.7.0
+cffi==2.0.0
+openai==1.86.0
+async-lru==2.3.0
+cfgv==3.5.0
+httpcore==1.0.9
+bitsandbytes==0.49.2
+iniconfig==2.3.0
+nvidia-cuda-runtime-cu12==12.6.77
+jupyterlab_server==2.28.0
+latex2sympy2_extended==1.11.0
+jupyter-events==0.12.1
+docker==7.1.0
+ipywidgets==8.1.8
+blinker==1.9.0
+cbor2==6.1.2
+nvidia-cufft-cu12==11.3.0.4
+nvidia-cusolver-cu12==11.7.1.2
+multiprocess==0.70.16
+anyio==4.14.0
+itsdangerous==2.2.0
+matplotlib==3.11.0
+widgetsnbextension==4.0.15
+watchfiles==1.2.0
+swebench==4.1.0
+blobfile==3.2.0
+jupyter==1.1.1
+certifi==2026.6.17
+argon2-cffi==25.1.0
+argon2-cffi-bindings==25.1.0
+overrides==7.7.0
+propcache==0.5.2
+ipython==8.33.0
+nbconvert==7.17.1
+rich==15.0.0
+asttokens==3.0.1
+tokenizers==0.22.2
+synchronicity==0.12.4
+tqdm==4.68.3
+nvidia-nvjitlink-cu12==12.6.85
+nvidia-nvtx-cu12==12.6.77
+jsonschema-specifications==2025.9.1
+jupyterlab==4.5.8
+pure_eval==0.2.3
+uri-template==1.3.0
+sentry-sdk==2.63.0
+pycryptodomex==3.23.0
+yarl==1.24.2
+h2==4.3.0
+h11==0.16.0
+requests==2.34.2
+Jinja2==3.1.6
+xformers==0.0.31.post1
+rfc3987-syntax==1.1.0
+antlr4-python3-runtime==4.11.0
+nvidia-cusparselt-cu12==0.6.3
+safetensors==0.8.0
+notebook_shim==0.2.4
+Pygments==2.20.0
+python-json-logger==4.1.0
+isoduration==20.11.0
+smmap==5.0.3
+click==8.3.3
+transformers==4.57.6
+aiohappyeyeballs==2.6.2
+regex==2026.5.9
+jsonpointer==3.1.1
+fqdn==1.5.1
+jupyter-lsp==2.3.1
+rfc3986-validator==0.1.1
+prometheus_client==0.25.0
+mistune==3.2.1
+sentencepiece==0.2.1
+types-toml==0.10.8.20260518
+wcwidth==0.8.1
+identify==2.6.19
+protobuf==6.33.6
+datasets==3.6.0
+accelerate==1.14.0
+networkx==3.6.1
+pydantic==2.13.4
+multidict==6.7.1
+chardet==7.4.3
+pytest==9.1.0
+babel==2.18.0
+pip==26.1.2
+dill==0.3.8
+referencing==0.37.0
+pre_commit==4.6.0
+numpy==2.4.6
+math-verify==0.9.0
+traitlets==5.15.1
+nvidia-curand-cu12==10.3.7.77
+websocket-client==1.9.0
+hf-xet==1.5.1
+unidiff==0.7.5
+nest-asyncio2==1.7.2
+typing_extensions==4.15.0
+platformdirs==4.10.0
+torch==2.7.1
+tblib==3.2.2
+sqlfluff==4.2.2
+modal==1.5.0
+ipykernel==7.3.0
+msgspec==0.21.1
+virtualenv==21.5.1
+json5==0.14.0
+pathlib_abc==0.5.2
+aiosignal==1.4.0
+moodist==0.5.1
+rpds-py==2026.5.1
+webencodings==0.5.1
+sniffio==1.3.1
+annotated-types==0.7.0
+python-discovery==1.4.2
+scipy==1.17.1
+GitPython==3.1.50
+defusedxml==0.7.1
+nvidia-cufile-cu12==1.11.1.6
+nvidia-nccl-cu12==2.26.2
+sympy==1.14.0
+tinycss2==1.5.1
+soupsieve==2.8.4
+ninja==1.13.0
+python-dotenv==1.2.2
+ghapi==1.0.13
+jupyter_server_terminals==0.5.4
+markdown-it-py==4.2.0
+nbclient==0.11.0
+mpmath==1.3.0
+frozenlist==1.8.0
+Werkzeug==3.1.8
+pillow==12.2.0
+executing==2.2.1
+gitdb==4.0.12
+pyparsing==3.3.2
+setuptools==82.0.1
+Flask==3.1.1
+MarkupSafe==3.0.3
+notebook==7.5.7
+jupyter_core==5.9.1
+jedi==0.20.0
+mdurl==0.1.2
+lark==1.3.1
+bleach==6.4.0
+hpack==4.1.0
+google-re2==1.1.20240702
+beautifulsoup4==4.15.0
+jsonschema==4.26.0
+psutil==7.2.2
+nvidia-cuda-nvrtc-cu12==12.6.77
+narwhals==2.22.1
+fonttools==4.63.0
+lxml==6.1.1
+fastjsonschema==2.21.2
+tornado==6.5.7
+nodeenv==1.10.0
+typing-inspection==0.4.2
+httpx==0.28.1
+distro==1.9.0
+colorama==0.4.6
+jupyter-console==6.6.3
+jupyterlab_widgets==3.0.16
+PyYAML==6.0.3
+fsspec==2025.3.0
+hyperframe==6.1.0
+nvidia-cuda-cupti-cu12==12.6.80
+charset-normalizer==3.4.7
+wandb==0.27.2
+nbformat==5.10.4
+nvidia-cudnn-cu12==9.5.1.17
+peft==0.19.1
+pathspec==1.1.1
+matplotlib-inline==0.2.2
+huggingface_hub==0.36.2
+pyarrow==24.0.0
+nvidia-cusparse-cu12==12.5.4.2
+cwm==0.0.1
+prompt_toolkit==3.0.52
+omegaconf==2.4.0.dev4
+tiktoken==0.13.0
+contourpy==1.3.3
+comm==0.2.3
+threadpoolctl==3.6.0
+pydantic_core==2.46.4
+tenacity==9.1.4
+parso==0.8.7
+pandas==3.0.3
+webcolors==25.10.0
+idna==3.18
+terminado==0.18.1
+pandocfilters==1.5.1
+aiohttp==3.14.1
+arrow==1.4.0
+packaging==26.2
+tzdata==2026.2
+distlib==0.4.3
+Send2Trash==2.1.0
+jupyterlab_pygments==0.3.0
+decorator==5.3.1
+jupyter_server==2.20.0
+scikit-learn==1.9.0
+six==1.17.0
+diff_cover==10.3.0
+pycparser==3.0
+types-certifi==2021.10.8.3
+toml==0.10.2
+stack-data==0.6.3
+fastcore==1.13.5
+filelock==3.29.4
+xxhash==3.7.0
+pexpect==4.9.0
+jupyter_client==8.9.1
+jiter==0.15.0
+debugpy==1.8.21
+universal_pathlib==0.3.10
+joblib==1.5.3

checkpoints/codi_frozen_3b_logit/wandb/offline-run-20260624_183654-h7tylfot/logs/debug-core.log

@@ -0,0 +1,15 @@
+{"time":"2026-06-24T18:36:55.287588537+02:00","level":"INFO","msg":"main: starting server","port-filename":"/local/tmp.6776480/tmp7v8iripp/port-3862836.txt","pid":3862836,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-06-24T18:36:55.288128529+02:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":3862836}
+{"time":"2026-06-24T18:36:55.288123286+02:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/local/tmp.6776480/wandb-3862836-3863309-2072640302/socket","Net":"unix"}}
+{"time":"2026-06-24T18:36:55.465759065+02:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-06-24T18:36:55.47743414+02:00","level":"INFO","msg":"handleInformInit: received","streamId":"h7tylfot","id":"1(@)"}
+{"time":"2026-06-24T18:36:55.601794194+02:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"h7tylfot","id":"1(@)"}
+{"time":"2026-06-25T08:22:53.59650799+02:00","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2026-06-25T08:22:53.59778409+02:00","level":"INFO","msg":"server is shutting down"}
+{"time":"2026-06-25T08:22:53.597894033+02:00","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-06-25T08:22:53.597926737+02:00","level":"INFO","msg":"server: listener closed","addr":{"Name":"/local/tmp.6776480/wandb-3862836-3863309-2072640302/socket","Net":"unix"}}
+{"time":"2026-06-25T08:22:53.597928241+02:00","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-06-25T08:22:53.59801602+02:00","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-06-25T08:22:53.610862413+02:00","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2026-06-25T08:22:53.610874022+02:00","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-06-25T08:22:53.610880152+02:00","level":"INFO","msg":"server is closed"}

checkpoints/codi_frozen_3b_logit/wandb/offline-run-20260624_183654-h7tylfot/logs/debug-internal.log

@@ -0,0 +1,11 @@
+{"time":"2026-06-24T18:36:55.480406596+02:00","level":"INFO","msg":"wandb-core"}
+{"time":"2026-06-24T18:36:55.480419535+02:00","level":"INFO","msg":"stream: starting","core version":"0.27.2"}
+{"time":"2026-06-24T18:36:55.600937635+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-24T18:36:55.600951703+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-24T18:36:55.600965787+02:00","level":"INFO","msg":"stream: created new stream","id":"h7tylfot"}
+{"time":"2026-06-24T18:36:55.601092756+02:00","level":"INFO","msg":"handler: started"}
+{"time":"2026-06-24T18:36:55.601789295+02:00","level":"INFO","msg":"stream: started"}
+{"time":"2026-06-24T18:36:55.601906907+02:00","level":"INFO","msg":"writer: started","stream_id":"h7tylfot"}
+{"time":"2026-06-24T18:36:55.601933505+02:00","level":"INFO","msg":"sender: started"}
+{"time":"2026-06-24T18:36:55.61937679+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-24T18:36:55.619392725+02:00","level":"WARN","msg":"runupserter: server does not expand metric globs but the x_server_side_expand_glob_metrics setting is set; ignoring"}

checkpoints/codi_frozen_3b_logit/wandb/offline-run-20260624_183654-h7tylfot/logs/debug.log

@@ -0,0 +1,20 @@
+2026-06-24 18:36:54,468 INFO    MainThread:3862836 [wandb_setup.py:_flush():81] Current SDK version is 0.27.2
+2026-06-24 18:36:54,469 INFO    MainThread:3862836 [wandb_setup.py:_flush():81] Configure stats pid to 3862836
+2026-06-24 18:36:54,469 INFO    MainThread:3862836 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-06-24 18:36:54,469 INFO    MainThread:3862836 [wandb_init.py:setup_run_log_directory():723] Logging user logs to model_weights/codi_frozen_3b_logit/wandb/offline-run-20260624_183654-h7tylfot/logs/debug.log
+2026-06-24 18:36:54,469 INFO    MainThread:3862836 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to model_weights/codi_frozen_3b_logit/wandb/offline-run-20260624_183654-h7tylfot/logs/debug-internal.log
+2026-06-24 18:36:54,469 INFO    MainThread:3862836 [wandb_init.py:init():766] calling init triggers
+2026-06-24 18:36:54,469 INFO    MainThread:3862836 [wandb_init.py:init():771] wandb.init called with sweep_config: {}
+config: {'model': 'model_weights/sft3b_lr1e5_bs32/checkpoint-3000', 'output_dir': 'model_weights/codi_frozen_3b_logit', 'sources': ['mbpp', 'humaneval', 'pyx'], 'cache_dir': 'data/cache/codi_train', 'n_samples': -1, 'max_seq_len': 3072, 'max_frames': -1, 'latent_steps': 1, 'epochs': 5.0, 'lr': 1e-05, 'batch_size': 1, 'grad_accum': 4, 'max_steps': 3000, 'save_steps': 500, 'alpha': 0.0, 'beta': 1.0, 'gamma': 1.0, 'kd_layers': [-1], 'frozen_teacher': 'model_weights/sft3b_lr1e5_bs32/checkpoint-3000', 'kd_target': 'logit', 'kd_temp': 2.0, 'single_anchor': False, 'ss_prob': 0.0, 'ss_ramp_frac': 0.5, '_wandb': {}}
+2026-06-24 18:36:54,469 INFO    MainThread:3862836 [wandb_init.py:init():814] starting backend
+2026-06-24 18:36:55,466 INFO    MainThread:3862836 [wandb_init.py:init():829] sending inform_init request
+2026-06-24 18:36:55,602 INFO    MainThread:3862836 [wandb_init.py:init():834] backend started and connected
+2026-06-24 18:36:55,604 INFO    MainThread:3862836 [wandb_init.py:init():904] updated telemetry
+2026-06-24 18:36:55,616 INFO    MainThread:3862836 [wandb_init.py:init():927] communicating run to backend with 90.0 second timeout
+2026-06-24 18:36:55,620 INFO    MainThread:3862836 [wandb_init.py:init():972] starting run threads in backend
+2026-06-24 18:36:55,889 INFO    MainThread:3862836 [wandb_run.py:_console_start():2523] atexit reg
+2026-06-24 18:36:55,889 INFO    MainThread:3862836 [wandb_run.py:_redirect():2373] redirect: wrap_raw
+2026-06-24 18:36:55,889 INFO    MainThread:3862836 [wandb_run.py:_redirect():2442] Wrapping output streams.
+2026-06-24 18:36:55,889 INFO    MainThread:3862836 [wandb_run.py:_redirect():2465] Redirects installed.
+2026-06-24 18:36:55,894 INFO    MainThread:3862836 [wandb_init.py:init():1010] run started, returning control to user process
+2026-06-24 18:37:01,262 INFO    MainThread:3862836 [wandb_run.py:_config_callback():1346] config_cb None None {'output_dir': 'model_weights/codi_frozen_3b_logit', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': 'no', 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 8, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 4, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.1, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 5.0, 'max_steps': 3000, 'lr_scheduler_type': 'cosine', 'lr_scheduler_kwargs': None, 'warmup_ratio': 0.03, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'model_weights/codi_frozen_3b_logit/runs/Jun24_18-36-55_alvis3-40', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 5, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': None, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': None, 'disable_tqdm': False, 'remove_unused_columns': False, 'label_names': [], 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'paged_adamw_8bit', 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'project': 'huggingface', 'trackio_space_id': 'trackio', 'ddp_find_unused_parameters': False, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'hub_revision': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': 'no', 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'liger_kernel_config': None, 'eval_use_gather_object': False, 'average_tokens_across_devices': True}

checkpoints/codi_frozen_3b_logit/wandb/offline-run-20260624_183654-h7tylfot/run-h7tylfot.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7408af03741efa23d8dc224dda9718c93c7d6eecc206c5e428860bb262fd20be
+size 7077888

checkpoints/codi_frozen_3b_logit/wandb/offline-run-20260625_135513-i10qt7cx/files/requirements.txt

@@ -0,0 +1,229 @@
+wheel==0.47.0
+triton==3.3.1
+ptyprocess==0.7.0
+kiwisolver==1.5.0
+pyzmq==27.1.0
+nvidia-cublas-cu12==12.6.4.1
+python-dateutil==2.9.0.post0
+cycler==0.12.1
+rfc3339-validator==0.1.4
+pluggy==1.6.0
+grpclib==0.4.9
+attrs==26.1.0
+urllib3==2.7.0
+cffi==2.0.0
+openai==1.86.0
+async-lru==2.3.0
+cfgv==3.5.0
+httpcore==1.0.9
+bitsandbytes==0.49.2
+iniconfig==2.3.0
+nvidia-cuda-runtime-cu12==12.6.77
+jupyterlab_server==2.28.0
+latex2sympy2_extended==1.11.0
+jupyter-events==0.12.1
+docker==7.1.0
+ipywidgets==8.1.8
+blinker==1.9.0
+cbor2==6.1.2
+nvidia-cufft-cu12==11.3.0.4
+nvidia-cusolver-cu12==11.7.1.2
+multiprocess==0.70.16
+anyio==4.14.0
+itsdangerous==2.2.0
+matplotlib==3.11.0
+widgetsnbextension==4.0.15
+watchfiles==1.2.0
+swebench==4.1.0
+blobfile==3.2.0
+jupyter==1.1.1
+certifi==2026.6.17
+argon2-cffi==25.1.0
+argon2-cffi-bindings==25.1.0
+overrides==7.7.0
+propcache==0.5.2
+ipython==8.33.0
+nbconvert==7.17.1
+rich==15.0.0
+asttokens==3.0.1
+tokenizers==0.22.2
+synchronicity==0.12.4
+tqdm==4.68.3
+nvidia-nvjitlink-cu12==12.6.85
+nvidia-nvtx-cu12==12.6.77
+jsonschema-specifications==2025.9.1
+jupyterlab==4.5.8
+pure_eval==0.2.3
+uri-template==1.3.0
+sentry-sdk==2.63.0
+pycryptodomex==3.23.0
+yarl==1.24.2
+h2==4.3.0
+h11==0.16.0
+requests==2.34.2
+Jinja2==3.1.6
+xformers==0.0.31.post1
+rfc3987-syntax==1.1.0
+antlr4-python3-runtime==4.11.0
+nvidia-cusparselt-cu12==0.6.3
+safetensors==0.8.0
+notebook_shim==0.2.4
+Pygments==2.20.0
+python-json-logger==4.1.0
+isoduration==20.11.0
+smmap==5.0.3
+click==8.3.3
+transformers==4.57.6
+aiohappyeyeballs==2.6.2
+regex==2026.5.9
+jsonpointer==3.1.1
+fqdn==1.5.1
+jupyter-lsp==2.3.1
+rfc3986-validator==0.1.1
+prometheus_client==0.25.0
+mistune==3.2.1
+sentencepiece==0.2.1
+types-toml==0.10.8.20260518
+wcwidth==0.8.1
+identify==2.6.19
+protobuf==6.33.6
+datasets==3.6.0
+accelerate==1.14.0
+networkx==3.6.1
+pydantic==2.13.4
+multidict==6.7.1
+chardet==7.4.3
+pytest==9.1.0
+babel==2.18.0
+pip==26.1.2
+dill==0.3.8
+referencing==0.37.0
+pre_commit==4.6.0
+numpy==2.4.6
+math-verify==0.9.0
+traitlets==5.15.1
+nvidia-curand-cu12==10.3.7.77
+websocket-client==1.9.0
+hf-xet==1.5.1
+unidiff==0.7.5
+nest-asyncio2==1.7.2
+typing_extensions==4.15.0
+platformdirs==4.10.0
+torch==2.7.1
+tblib==3.2.2
+sqlfluff==4.2.2
+modal==1.5.0
+ipykernel==7.3.0
+msgspec==0.21.1
+virtualenv==21.5.1
+json5==0.14.0
+pathlib_abc==0.5.2
+aiosignal==1.4.0
+moodist==0.5.1
+rpds-py==2026.5.1
+webencodings==0.5.1
+sniffio==1.3.1
+annotated-types==0.7.0
+python-discovery==1.4.2
+scipy==1.17.1
+GitPython==3.1.50
+defusedxml==0.7.1
+nvidia-cufile-cu12==1.11.1.6
+nvidia-nccl-cu12==2.26.2
+sympy==1.14.0
+tinycss2==1.5.1
+soupsieve==2.8.4
+ninja==1.13.0
+python-dotenv==1.2.2
+ghapi==1.0.13
+jupyter_server_terminals==0.5.4
+markdown-it-py==4.2.0
+nbclient==0.11.0
+mpmath==1.3.0
+frozenlist==1.8.0
+Werkzeug==3.1.8
+pillow==12.2.0
+executing==2.2.1
+gitdb==4.0.12
+pyparsing==3.3.2
+setuptools==82.0.1
+Flask==3.1.1
+MarkupSafe==3.0.3
+notebook==7.5.7
+jupyter_core==5.9.1
+jedi==0.20.0
+mdurl==0.1.2
+lark==1.3.1
+bleach==6.4.0
+hpack==4.1.0
+google-re2==1.1.20240702
+beautifulsoup4==4.15.0
+jsonschema==4.26.0
+psutil==7.2.2
+nvidia-cuda-nvrtc-cu12==12.6.77
+narwhals==2.22.1
+fonttools==4.63.0
+lxml==6.1.1
+fastjsonschema==2.21.2
+tornado==6.5.7
+nodeenv==1.10.0
+typing-inspection==0.4.2
+httpx==0.28.1
+distro==1.9.0
+colorama==0.4.6
+jupyter-console==6.6.3
+jupyterlab_widgets==3.0.16
+PyYAML==6.0.3
+fsspec==2025.3.0
+hyperframe==6.1.0
+nvidia-cuda-cupti-cu12==12.6.80
+charset-normalizer==3.4.7
+wandb==0.27.2
+nbformat==5.10.4
+nvidia-cudnn-cu12==9.5.1.17
+peft==0.19.1
+pathspec==1.1.1
+matplotlib-inline==0.2.2
+huggingface_hub==0.36.2
+pyarrow==24.0.0
+nvidia-cusparse-cu12==12.5.4.2
+cwm==0.0.1
+prompt_toolkit==3.0.52
+omegaconf==2.4.0.dev4
+tiktoken==0.13.0
+contourpy==1.3.3
+comm==0.2.3
+threadpoolctl==3.6.0
+pydantic_core==2.46.4
+tenacity==9.1.4
+parso==0.8.7
+pandas==3.0.3
+webcolors==25.10.0
+idna==3.18
+terminado==0.18.1
+pandocfilters==1.5.1
+aiohttp==3.14.1
+arrow==1.4.0
+packaging==26.2
+tzdata==2026.2
+distlib==0.4.3
+Send2Trash==2.1.0
+jupyterlab_pygments==0.3.0
+decorator==5.3.1
+jupyter_server==2.20.0
+scikit-learn==1.9.0
+six==1.17.0
+diff_cover==10.3.0
+pycparser==3.0
+types-certifi==2021.10.8.3
+toml==0.10.2
+stack-data==0.6.3
+fastcore==1.13.5
+filelock==3.29.4
+xxhash==3.7.0
+pexpect==4.9.0
+jupyter_client==8.9.1
+jiter==0.15.0
+debugpy==1.8.21
+universal_pathlib==0.3.10
+joblib==1.5.3

checkpoints/codi_frozen_3b_logit/wandb/offline-run-20260625_135513-i10qt7cx/logs/debug-core.log

@@ -0,0 +1,6 @@
+{"time":"2026-06-25T13:55:14.667979879+02:00","level":"INFO","msg":"main: starting server","port-filename":"/local/tmp.6777513/tmp8vosadi2/port-1545486.txt","pid":1545486,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-06-25T13:55:14.668540103+02:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/local/tmp.6777513/wandb-1545486-1545932-3192363471/socket","Net":"unix"}}
+{"time":"2026-06-25T13:55:14.668680652+02:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":1545486}
+{"time":"2026-06-25T13:55:14.850522925+02:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-06-25T13:55:14.863309797+02:00","level":"INFO","msg":"handleInformInit: received","streamId":"i10qt7cx","id":"1(@)"}
+{"time":"2026-06-25T13:55:14.989324567+02:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"i10qt7cx","id":"1(@)"}

checkpoints/codi_frozen_3b_logit/wandb/offline-run-20260625_135513-i10qt7cx/logs/debug-internal.log

@@ -0,0 +1,11 @@
+{"time":"2026-06-25T13:55:14.865357126+02:00","level":"INFO","msg":"wandb-core"}
+{"time":"2026-06-25T13:55:14.865374279+02:00","level":"INFO","msg":"stream: starting","core version":"0.27.2"}
+{"time":"2026-06-25T13:55:14.988281216+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-25T13:55:14.988307148+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-25T13:55:14.988325352+02:00","level":"INFO","msg":"stream: created new stream","id":"i10qt7cx"}
+{"time":"2026-06-25T13:55:14.988514907+02:00","level":"INFO","msg":"handler: started"}
+{"time":"2026-06-25T13:55:14.989316587+02:00","level":"INFO","msg":"stream: started"}
+{"time":"2026-06-25T13:55:14.989459496+02:00","level":"INFO","msg":"writer: started","stream_id":"i10qt7cx"}
+{"time":"2026-06-25T13:55:14.989483136+02:00","level":"INFO","msg":"sender: started"}
+{"time":"2026-06-25T13:55:15.00239695+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-25T13:55:15.002412933+02:00","level":"WARN","msg":"runupserter: server does not expand metric globs but the x_server_side_expand_glob_metrics setting is set; ignoring"}

checkpoints/codi_frozen_3b_logit/wandb/offline-run-20260625_135513-i10qt7cx/logs/debug.log

@@ -0,0 +1,20 @@
+2026-06-25 13:55:13,760 INFO    MainThread:1545486 [wandb_setup.py:_flush():81] Current SDK version is 0.27.2
+2026-06-25 13:55:13,760 INFO    MainThread:1545486 [wandb_setup.py:_flush():81] Configure stats pid to 1545486
+2026-06-25 13:55:13,760 INFO    MainThread:1545486 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-06-25 13:55:13,761 INFO    MainThread:1545486 [wandb_init.py:setup_run_log_directory():723] Logging user logs to model_weights/codi_frozen_3b_logit/wandb/offline-run-20260625_135513-i10qt7cx/logs/debug.log
+2026-06-25 13:55:13,761 INFO    MainThread:1545486 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to model_weights/codi_frozen_3b_logit/wandb/offline-run-20260625_135513-i10qt7cx/logs/debug-internal.log
+2026-06-25 13:55:13,761 INFO    MainThread:1545486 [wandb_init.py:init():766] calling init triggers
+2026-06-25 13:55:13,761 INFO    MainThread:1545486 [wandb_init.py:init():771] wandb.init called with sweep_config: {}
+config: {'model': 'model_weights/sft3b_lr1e5_bs32/checkpoint-3000', 'output_dir': 'model_weights/codi_frozen_3b_logit', 'sources': ['mbpp', 'humaneval', 'pyx'], 'cache_dir': 'data/cache/codi_train', 'n_samples': -1, 'max_seq_len': 3072, 'max_frames': -1, 'latent_steps': 1, 'epochs': 5.0, 'lr': 1e-05, 'batch_size': 1, 'grad_accum': 4, 'max_steps': 2500, 'save_steps': 500, 'alpha': 0.0, 'beta': 1.0, 'gamma': 1.0, 'kd_layers': None, 'frozen_teacher': 'model_weights/sft3b_lr1e5_bs32/checkpoint-3000', 'kd_target': 'logit', 'kd_temp': 2.0, 'single_anchor': False, 'ss_prob': 0.0, 'ss_ramp_frac': 0.5, '_wandb': {}}
+2026-06-25 13:55:13,761 INFO    MainThread:1545486 [wandb_init.py:init():814] starting backend
+2026-06-25 13:55:14,850 INFO    MainThread:1545486 [wandb_init.py:init():829] sending inform_init request
+2026-06-25 13:55:14,990 INFO    MainThread:1545486 [wandb_init.py:init():834] backend started and connected
+2026-06-25 13:55:14,991 INFO    MainThread:1545486 [wandb_init.py:init():904] updated telemetry
+2026-06-25 13:55:15,000 INFO    MainThread:1545486 [wandb_init.py:init():927] communicating run to backend with 90.0 second timeout
+2026-06-25 13:55:15,003 INFO    MainThread:1545486 [wandb_init.py:init():972] starting run threads in backend
+2026-06-25 13:55:15,387 INFO    MainThread:1545486 [wandb_run.py:_console_start():2523] atexit reg
+2026-06-25 13:55:15,387 INFO    MainThread:1545486 [wandb_run.py:_redirect():2373] redirect: wrap_raw
+2026-06-25 13:55:15,387 INFO    MainThread:1545486 [wandb_run.py:_redirect():2442] Wrapping output streams.
+2026-06-25 13:55:15,387 INFO    MainThread:1545486 [wandb_run.py:_redirect():2465] Redirects installed.
+2026-06-25 13:55:15,393 INFO    MainThread:1545486 [wandb_init.py:init():1010] run started, returning control to user process
+2026-06-25 13:55:25,930 INFO    MainThread:1545486 [wandb_run.py:_config_callback():1346] config_cb None None {'output_dir': 'model_weights/codi_frozen_3b_logit', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': False, 'do_predict': False, 'eval_strategy': 'no', 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 8, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 4, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.1, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 5.0, 'max_steps': 2500, 'lr_scheduler_type': 'cosine', 'lr_scheduler_kwargs': None, 'warmup_ratio': 0.03, 'warmup_steps': 0, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': 'model_weights/codi_frozen_3b_logit/runs/Jun25_13-55-15_alvis4-42', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 5, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': None, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': None, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': None, 'disable_tqdm': False, 'remove_unused_columns': False, 'label_names': [], 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'paged_adamw_8bit', 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'project': 'huggingface', 'trackio_space_id': 'trackio', 'ddp_find_unused_parameters': False, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'hub_revision': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': 'no', 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'liger_kernel_config': None, 'eval_use_gather_object': False, 'average_tokens_across_devices': True}