add checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1

.gitattributes

@@ -44,3 +44,8 @@ checkpoints/codi1.5b_a0.5_b1.0_g0.5_ls2/checkpoint-200/tokenizer.json filter=lfs
 checkpoints/codi1.5b_a0.5_b1.0_g0.5_ls2/checkpoint-400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/codi1.5b_a0.5_b1.0_g0.5_ls2/checkpoint-600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/codi1.5b_a0.5_b1.0_g0.5_ls2/checkpoint-800/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/wandb/offline-run-20260619_215302-q1wgok6t/run-q1wgok6t.wandb filter=lfs diff=lfs merge=lfs -text

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/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/codi1.5b_a1.0_b1.0_g1.0_ls1/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 Qwen, created by Alibaba Cloud. 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 Qwen, created by Alibaba Cloud. You 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/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1000/config.json

@@ -0,0 +1,58 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "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/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c5b0ad2330893bc3c119aa9e549651b6caa6bb02928f5c144887381d77c39ac
+size 3096275515

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/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/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1000/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eadb45b07a8938d958ae27c0033af0c6d73b81872cc6d44c8e202878d4422c6
+size 9445953

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1000/tokenizer.json

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

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/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/codi1.5b_a1.0_b1.0_g1.0_ls1/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": 34.25,
+      "kd_loss": 0.2726191282272339,
+      "learning_rate": 4.444444444444445e-07,
+      "loss": 0.3835,
+      "step": 5,
+      "student_loss": 0.053846802562475204,
+      "teacher_loss": 0.00042048963950946927
+    },
+    {
+      "epoch": 0.0028835063437139563,
+      "grad_norm": 48.0,
+      "kd_loss": 0.24393108487129211,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 0.3789,
+      "step": 10,
+      "student_loss": 0.09970586746931076,
+      "teacher_loss": 0.0007505021058022976
+    },
+    {
+      "epoch": 0.004325259515570935,
+      "grad_norm": 21.75,
+      "kd_loss": 0.21377132833003998,
+      "learning_rate": 1.5555555555555558e-06,
+      "loss": 0.3585,
+      "step": 15,
+      "student_loss": 0.029451103881001472,
+      "teacher_loss": 0.0006585777155123651
+    },
+    {
+      "epoch": 0.0057670126874279125,
+      "grad_norm": 27.5,
+      "kd_loss": 0.2140207439661026,
+      "learning_rate": 2.1111111111111114e-06,
+      "loss": 0.3372,
+      "step": 20,
+      "student_loss": 0.1321348249912262,
+      "teacher_loss": 0.000296285143122077
+    },
+    {
+      "epoch": 0.00720876585928489,
+      "grad_norm": 14.5625,
+      "kd_loss": 0.18640007078647614,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 0.2845,
+      "step": 25,
+      "student_loss": 0.018293429166078568,
+      "teacher_loss": 0.0010102478554472327
+    },
+    {
+      "epoch": 0.00865051903114187,
+      "grad_norm": 7.1875,
+      "kd_loss": 0.14892512559890747,
+      "learning_rate": 3.2222222222222227e-06,
+      "loss": 0.2387,
+      "step": 30,
+      "student_loss": 0.005030089057981968,
+      "teacher_loss": 0.0003383651783224195
+    },
+    {
+      "epoch": 0.010092272202998846,
+      "grad_norm": 5.5,
+      "kd_loss": 0.1269177347421646,
+      "learning_rate": 3.777777777777778e-06,
+      "loss": 0.2117,
+      "step": 35,
+      "student_loss": 0.0014062788104638457,
+      "teacher_loss": 0.00025501433992758393
+    },
+    {
+      "epoch": 0.011534025374855825,
+      "grad_norm": 4.65625,
+      "kd_loss": 0.12731066346168518,
+      "learning_rate": 4.333333333333334e-06,
+      "loss": 0.1863,
+      "step": 40,
+      "student_loss": 0.049713075160980225,
+      "teacher_loss": 0.040033385157585144
+    },
+    {
+      "epoch": 0.012975778546712802,
+      "grad_norm": 7.375,
+      "kd_loss": 0.1775631606578827,
+      "learning_rate": 4.888888888888889e-06,
+      "loss": 0.1456,
+      "step": 45,
+      "student_loss": 0.06933413445949554,
+      "teacher_loss": 0.0009156111045740545
+    },
+    {
+      "epoch": 0.01441753171856978,
+      "grad_norm": 4.15625,
+      "kd_loss": 0.06588269770145416,
+      "learning_rate": 5.444444444444445e-06,
+      "loss": 0.1417,
+      "step": 50,
+      "student_loss": 0.008036560378968716,
+      "teacher_loss": 0.004471915308386087
+    },
+    {
+      "epoch": 0.015859284890426758,
+      "grad_norm": 8.375,
+      "kd_loss": 0.05523190647363663,
+      "learning_rate": 6e-06,
+      "loss": 0.1225,
+      "step": 55,
+      "student_loss": 0.0012776668881997466,
+      "teacher_loss": 0.00044586719013750553
+    },
+    {
+      "epoch": 0.01730103806228374,
+      "grad_norm": 5.125,
+      "kd_loss": 0.049259770661592484,
+      "learning_rate": 6.555555555555556e-06,
+      "loss": 0.1047,
+      "step": 60,
+      "student_loss": 0.01390351727604866,
+      "teacher_loss": 0.002199392532929778
+    },
+    {
+      "epoch": 0.018742791234140715,
+      "grad_norm": 7.375,
+      "kd_loss": 0.047981150448322296,
+      "learning_rate": 7.111111111111112e-06,
+      "loss": 0.1006,
+      "step": 65,
+      "student_loss": 0.04912174493074417,
+      "teacher_loss": 0.05355874449014664
+    },
+    {
+      "epoch": 0.020184544405997693,
+      "grad_norm": 5.21875,
+      "kd_loss": 0.07241208106279373,
+      "learning_rate": 7.666666666666667e-06,
+      "loss": 0.0935,
+      "step": 70,
+      "student_loss": 0.03459472954273224,
+      "teacher_loss": 0.0029977799858897924
+    },
+    {
+      "epoch": 0.02162629757785467,
+      "grad_norm": 5.125,
+      "kd_loss": 0.0307459756731987,
+      "learning_rate": 8.222222222222222e-06,
+      "loss": 0.0907,
+      "step": 75,
+      "student_loss": 0.0016565826954320073,
+      "teacher_loss": 0.0008985538734123111
+    },
+    {
+      "epoch": 0.02306805074971165,
+      "grad_norm": 1.9453125,
+      "kd_loss": 0.026722930371761322,
+      "learning_rate": 8.777777777777778e-06,
+      "loss": 0.0788,
+      "step": 80,
+      "student_loss": 0.0021714135073125362,
+      "teacher_loss": 0.0012323743430897593
+    },
+    {
+      "epoch": 0.024509803921568627,
+      "grad_norm": 3.71875,
+      "kd_loss": 0.023355644196271896,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.0733,
+      "step": 85,
+      "student_loss": 0.0012368966126814485,
+      "teacher_loss": 0.0009521314641460776
+    },
+    {
+      "epoch": 0.025951557093425604,
+      "grad_norm": 1.75,
+      "kd_loss": 0.0727810338139534,
+      "learning_rate": 9.88888888888889e-06,
+      "loss": 0.0676,
+      "step": 90,
+      "student_loss": 0.028416095301508904,
+      "teacher_loss": 0.005437986925244331
+    },
+    {
+      "epoch": 0.027393310265282585,
+      "grad_norm": 2.75,
+      "kd_loss": 0.04388673976063728,
+      "learning_rate": 9.99995337991255e-06,
+      "loss": 0.0638,
+      "step": 95,
+      "student_loss": 0.009086629375815392,
+      "teacher_loss": 0.005068701226264238
+    },
+    {
+      "epoch": 0.02883506343713956,
+      "grad_norm": 3.328125,
+      "kd_loss": 0.018161524087190628,
+      "learning_rate": 9.999763987297266e-06,
+      "loss": 0.0537,
+      "step": 100,
+      "student_loss": 0.0007679096888750792,
+      "teacher_loss": 0.0005449313321150839
+    },
+    {
+      "epoch": 0.03027681660899654,
+      "grad_norm": 2.0,
+      "kd_loss": 0.041785843670368195,
+      "learning_rate": 9.999428913912876e-06,
+      "loss": 0.0561,
+      "step": 105,
+      "student_loss": 0.005221863742917776,
+      "teacher_loss": 0.0004642533021979034
+    },
+    {
+      "epoch": 0.031718569780853516,
+      "grad_norm": 1.90625,
+      "kd_loss": 0.03158900886774063,
+      "learning_rate": 9.998948169522587e-06,
+      "loss": 0.0634,
+      "step": 110,
+      "student_loss": 0.02624746784567833,
+      "teacher_loss": 0.019401099532842636
+    },
+    {
+      "epoch": 0.03316032295271049,
+      "grad_norm": 1.84375,
+      "kd_loss": 0.025746585801243782,
+      "learning_rate": 9.998321768134101e-06,
+      "loss": 0.058,
+      "step": 115,
+      "student_loss": 0.0007584464619867504,
+      "teacher_loss": 0.00038955273339524865
+    },
+    {
+      "epoch": 0.03460207612456748,
+      "grad_norm": 2.5,
+      "kd_loss": 0.017939826473593712,
+      "learning_rate": 9.997549727999212e-06,
+      "loss": 0.0505,
+      "step": 120,
+      "student_loss": 0.0029337499290704727,
+      "teacher_loss": 0.0009877075208351016
+    },
+    {
+      "epoch": 0.036043829296424454,
+      "grad_norm": 2.046875,
+      "kd_loss": 0.0580839179456234,
+      "learning_rate": 9.996632071613257e-06,
+      "loss": 0.0538,
+      "step": 125,
+      "student_loss": 0.042769238352775574,
+      "teacher_loss": 0.0003408969205338508
+    },
+    {
+      "epoch": 0.03748558246828143,
+      "grad_norm": 11.75,
+      "kd_loss": 0.017885997891426086,
+      "learning_rate": 9.995568825714479e-06,
+      "loss": 0.0556,
+      "step": 130,
+      "student_loss": 0.0008774802554398775,
+      "teacher_loss": 0.000655525189358741
+    },
+    {
+      "epoch": 0.03892733564013841,
+      "grad_norm": 1.1875,
+      "kd_loss": 0.032637353986501694,
+      "learning_rate": 9.994360021283235e-06,
+      "loss": 0.054,
+      "step": 135,
+      "student_loss": 0.009718491695821285,
+      "teacher_loss": 0.0008957762038335204
+    },
+    {
+      "epoch": 0.040369088811995385,
+      "grad_norm": 1.015625,
+      "kd_loss": 0.01658937893807888,
+      "learning_rate": 9.993005693541098e-06,
+      "loss": 0.0484,
+      "step": 140,
+      "student_loss": 0.0006278170621953905,
+      "teacher_loss": 0.00041464308742433786
+    },
+    {
+      "epoch": 0.04181084198385236,
+      "grad_norm": 3.09375,
+      "kd_loss": 0.06331603974103928,
+      "learning_rate": 9.991505881949837e-06,
+      "loss": 0.0455,
+      "step": 145,
+      "student_loss": 0.04779814928770065,
+      "teacher_loss": 0.008776809088885784
+    },
+    {
+      "epoch": 0.04325259515570934,
+      "grad_norm": 1.1875,
+      "kd_loss": 0.012085826136171818,
+      "learning_rate": 9.989860630210251e-06,
+      "loss": 0.0447,
+      "step": 150,
+      "student_loss": 0.0007606842555105686,
+      "teacher_loss": 0.0006636636680923402
+    },
+    {
+      "epoch": 0.04469434832756632,
+      "grad_norm": 1.8515625,
+      "kd_loss": 0.015217100270092487,
+      "learning_rate": 9.98806998626091e-06,
+      "loss": 0.0574,
+      "step": 155,
+      "student_loss": 0.0006297255749814212,
+      "teacher_loss": 0.0004516748304013163
+    },
+    {
+      "epoch": 0.0461361014994233,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.021681347861886024,
+      "learning_rate": 9.98613400227676e-06,
+      "loss": 0.0499,
+      "step": 160,
+      "student_loss": 0.08383286744356155,
+      "teacher_loss": 0.0425245501101017
+    },
+    {
+      "epoch": 0.04757785467128028,
+      "grad_norm": 1.0625,
+      "kd_loss": 0.02377716824412346,
+      "learning_rate": 9.984052734667586e-06,
+      "loss": 0.0438,
+      "step": 165,
+      "student_loss": 0.000517472974024713,
+      "teacher_loss": 0.0054916865192353725
+    },
+    {
+      "epoch": 0.049019607843137254,
+      "grad_norm": 1.296875,
+      "kd_loss": 0.012254269793629646,
+      "learning_rate": 9.981826244076387e-06,
+      "loss": 0.045,
+      "step": 170,
+      "student_loss": 0.0007025015656836331,
+      "teacher_loss": 0.0005940330447629094
+    },
+    {
+      "epoch": 0.05046136101499423,
+      "grad_norm": 1.2265625,
+      "kd_loss": 0.022828614339232445,
+      "learning_rate": 9.979454595377594e-06,
+      "loss": 0.0449,
+      "step": 175,
+      "student_loss": 0.002857188694179058,
+      "teacher_loss": 0.00044675302342511714
+    },
+    {
+      "epoch": 0.05190311418685121,
+      "grad_norm": 1.15625,
+      "kd_loss": 0.02279340662062168,
+      "learning_rate": 9.97693785767519e-06,
+      "loss": 0.0416,
+      "step": 180,
+      "student_loss": 0.0009870725916698575,
+      "teacher_loss": 0.0002556022081989795
+    },
+    {
+      "epoch": 0.05334486735870819,
+      "grad_norm": 1.484375,
+      "kd_loss": 0.02873876690864563,
+      "learning_rate": 9.9742761043007e-06,
+      "loss": 0.0472,
+      "step": 185,
+      "student_loss": 0.04467737674713135,
+      "teacher_loss": 0.0002736605820246041
+    },
+    {
+      "epoch": 0.05478662053056517,
+      "grad_norm": 1.5078125,
+      "kd_loss": 0.03800494968891144,
+      "learning_rate": 9.971469412811032e-06,
+      "loss": 0.0432,
+      "step": 190,
+      "student_loss": 0.006019711960107088,
+      "teacher_loss": 0.005745577625930309
+    },
+    {
+      "epoch": 0.056228373702422146,
+      "grad_norm": 2.453125,
+      "kd_loss": 0.03193611651659012,
+      "learning_rate": 9.968517864986247e-06,
+      "loss": 0.044,
+      "step": 195,
+      "student_loss": 0.010869728401303291,
+      "teacher_loss": 0.0007882199133746326
+    },
+    {
+      "epoch": 0.05767012687427912,
+      "grad_norm": 0.703125,
+      "kd_loss": 0.025101784616708755,
+      "learning_rate": 9.965421546827153e-06,
+      "loss": 0.0501,
+      "step": 200,
+      "student_loss": 0.0018625810043886304,
+      "teacher_loss": 0.00027628077077679336
+    },
+    {
+      "epoch": 0.0591118800461361,
+      "grad_norm": 1.359375,
+      "kd_loss": 0.02438431791961193,
+      "learning_rate": 9.962180548552812e-06,
+      "loss": 0.0458,
+      "step": 205,
+      "student_loss": 0.03717027232050896,
+      "teacher_loss": 0.02886415459215641
+    },
+    {
+      "epoch": 0.06055363321799308,
+      "grad_norm": 1.84375,
+      "kd_loss": 0.02243652194738388,
+      "learning_rate": 9.958794964597906e-06,
+      "loss": 0.043,
+      "step": 210,
+      "student_loss": 0.0013577835634350777,
+      "teacher_loss": 0.000932522292714566
+    },
+    {
+      "epoch": 0.061995386389850055,
+      "grad_norm": 1.078125,
+      "kd_loss": 0.028962837532162666,
+      "learning_rate": 9.955264893609986e-06,
+      "loss": 0.0453,
+      "step": 215,
+      "student_loss": 0.001151631586253643,
+      "teacher_loss": 0.0003005610778927803
+    },
+    {
+      "epoch": 0.06343713956170703,
+      "grad_norm": 1.9375,
+      "kd_loss": 0.03575044497847557,
+      "learning_rate": 9.951590438446597e-06,
+      "loss": 0.0453,
+      "step": 220,
+      "student_loss": 0.048845142126083374,
+      "teacher_loss": 0.009562678635120392
+    },
+    {
+      "epoch": 0.06487889273356401,
+      "grad_norm": 1.375,
+      "kd_loss": 0.015835396945476532,
+      "learning_rate": 9.947771706172285e-06,
+      "loss": 0.0506,
+      "step": 225,
+      "student_loss": 0.0546843521296978,
+      "teacher_loss": 0.0008852278115227818
+    },
+    {
+      "epoch": 0.06632064590542099,
+      "grad_norm": 2.125,
+      "kd_loss": 0.02423272468149662,
+      "learning_rate": 9.943808808055472e-06,
+      "loss": 0.0398,
+      "step": 230,
+      "student_loss": 0.0009826251771301031,
+      "teacher_loss": 0.002756665926426649
+    },
+    {
+      "epoch": 0.06776239907727798,
+      "grad_norm": 0.93359375,
+      "kd_loss": 0.04541483893990517,
+      "learning_rate": 9.93970185956522e-06,
+      "loss": 0.0415,
+      "step": 235,
+      "student_loss": 0.014311765320599079,
+      "teacher_loss": 0.0015107382787391543
+    },
+    {
+      "epoch": 0.06920415224913495,
+      "grad_norm": 1.296875,
+      "kd_loss": 0.018931882455945015,
+      "learning_rate": 9.93545098036786e-06,
+      "loss": 0.0432,
+      "step": 240,
+      "student_loss": 0.0005313835572451353,
+      "teacher_loss": 0.00029519537929445505
+    },
+    {
+      "epoch": 0.07064590542099193,
+      "grad_norm": 1.2890625,
+      "kd_loss": 0.012201114557683468,
+      "learning_rate": 9.931056294323507e-06,
+      "loss": 0.0392,
+      "step": 245,
+      "student_loss": 0.0008615658152848482,
+      "teacher_loss": 0.0008678268059156835
+    },
+    {
+      "epoch": 0.07208765859284891,
+      "grad_norm": 1.9296875,
+      "kd_loss": 0.013639300130307674,
+      "learning_rate": 9.926517929482454e-06,
+      "loss": 0.0416,
+      "step": 250,
+      "student_loss": 0.000850351236294955,
+      "teacher_loss": 0.0004967341665178537
+    },
+    {
+      "epoch": 0.07352941176470588,
+      "grad_norm": 0.82421875,
+      "kd_loss": 0.010670178569853306,
+      "learning_rate": 9.921836018081438e-06,
+      "loss": 0.0421,
+      "step": 255,
+      "student_loss": 0.0005453266203403473,
+      "teacher_loss": 0.0004909925046376884
+    },
+    {
+      "epoch": 0.07497116493656286,
+      "grad_norm": 1.375,
+      "kd_loss": 0.019146086648106575,
+      "learning_rate": 9.917010696539793e-06,
+      "loss": 0.0448,
+      "step": 260,
+      "student_loss": 0.040390945971012115,
+      "teacher_loss": 0.033758774399757385
+    },
+    {
+      "epoch": 0.07641291810841984,
+      "grad_norm": 1.78125,
+      "kd_loss": 0.011872848495841026,
+      "learning_rate": 9.912042105455462e-06,
+      "loss": 0.0422,
+      "step": 265,
+      "student_loss": 0.0005196996498852968,
+      "teacher_loss": 0.00042158958967775106
+    },
+    {
+      "epoch": 0.07785467128027682,
+      "grad_norm": 1.28125,
+      "kd_loss": 0.012035544961690903,
+      "learning_rate": 9.906930389600917e-06,
+      "loss": 0.0458,
+      "step": 270,
+      "student_loss": 0.0009468707139603794,
+      "teacher_loss": 0.000693841720931232
+    },
+    {
+      "epoch": 0.07929642445213379,
+      "grad_norm": 1.7890625,
+      "kd_loss": 0.01897946372628212,
+      "learning_rate": 9.901675697918925e-06,
+      "loss": 0.0432,
+      "step": 275,
+      "student_loss": 0.047525323927402496,
+      "teacher_loss": 0.03367537260055542
+    },
+    {
+      "epoch": 0.08073817762399077,
+      "grad_norm": 1.5859375,
+      "kd_loss": 0.021627889946103096,
+      "learning_rate": 9.896278183518216e-06,
+      "loss": 0.041,
+      "step": 280,
+      "student_loss": 0.0006324543501250446,
+      "teacher_loss": 0.0002107525651808828
+    },
+    {
+      "epoch": 0.08217993079584775,
+      "grad_norm": 1.46875,
+      "kd_loss": 0.013249333947896957,
+      "learning_rate": 9.890738003669029e-06,
+      "loss": 0.0444,
+      "step": 285,
+      "student_loss": 0.0005354061140678823,
+      "teacher_loss": 0.0004488772538024932
+    },
+    {
+      "epoch": 0.08362168396770472,
+      "grad_norm": 1.3359375,
+      "kd_loss": 0.011662454344332218,
+      "learning_rate": 9.885055319798514e-06,
+      "loss": 0.0476,
+      "step": 290,
+      "student_loss": 0.0025434105191379786,
+      "teacher_loss": 0.0015740139642730355
+    },
+    {
+      "epoch": 0.0850634371395617,
+      "grad_norm": 1.828125,
+      "kd_loss": 0.01183430477976799,
+      "learning_rate": 9.879230297486034e-06,
+      "loss": 0.0425,
+      "step": 295,
+      "student_loss": 0.0004690758651122451,
+      "teacher_loss": 0.00038755135028623044
+    },
+    {
+      "epoch": 0.08650519031141868,
+      "grad_norm": 1.03125,
+      "kd_loss": 0.03465069457888603,
+      "learning_rate": 9.873263106458353e-06,
+      "loss": 0.0394,
+      "step": 300,
+      "student_loss": 0.018424537032842636,
+      "teacher_loss": 0.002826885087415576
+    },
+    {
+      "epoch": 0.08794694348327567,
+      "grad_norm": 1.125,
+      "kd_loss": 0.013110927306115627,
+      "learning_rate": 9.86715392058467e-06,
+      "loss": 0.0357,
+      "step": 305,
+      "student_loss": 0.01486227661371231,
+      "teacher_loss": 0.0012473311508074403
+    },
+    {
+      "epoch": 0.08938869665513265,
+      "grad_norm": 1.4765625,
+      "kd_loss": 0.010091043077409267,
+      "learning_rate": 9.860902917871566e-06,
+      "loss": 0.039,
+      "step": 310,
+      "student_loss": 0.00043867999920621514,
+      "teacher_loss": 0.00034055308788083494
+    },
+    {
+      "epoch": 0.09083044982698962,
+      "grad_norm": 1.0625,
+      "kd_loss": 0.02157684415578842,
+      "learning_rate": 9.854510280457819e-06,
+      "loss": 0.0391,
+      "step": 315,
+      "student_loss": 0.014208661392331123,
+      "teacher_loss": 0.001196117140352726
+    },
+    {
+      "epoch": 0.0922722029988466,
+      "grad_norm": 1.1953125,
+      "kd_loss": 0.0127296457067132,
+      "learning_rate": 9.847976194609083e-06,
+      "loss": 0.0432,
+      "step": 320,
+      "student_loss": 0.0004476152826100588,
+      "teacher_loss": 0.000240055174799636
+    },
+    {
+      "epoch": 0.09371395617070358,
+      "grad_norm": 1.5546875,
+      "kd_loss": 0.032895129173994064,
+      "learning_rate": 9.841300850712479e-06,
+      "loss": 0.0377,
+      "step": 325,
+      "student_loss": 0.008438433520495892,
+      "teacher_loss": 0.015148998238146305
+    },
+    {
+      "epoch": 0.09515570934256055,
+      "grad_norm": 1.3046875,
+      "kd_loss": 0.04177394509315491,
+      "learning_rate": 9.834484443271035e-06,
+      "loss": 0.0389,
+      "step": 330,
+      "student_loss": 0.01764899119734764,
+      "teacher_loss": 0.001141766319051385
+    },
+    {
+      "epoch": 0.09659746251441753,
+      "grad_norm": 0.9296875,
+      "kd_loss": 0.008085020817816257,
+      "learning_rate": 9.827527170898019e-06,
+      "loss": 0.0328,
+      "step": 335,
+      "student_loss": 0.0030857992824167013,
+      "teacher_loss": 0.0029651003424078226
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 1.5625,
+      "kd_loss": 0.014680181629955769,
+      "learning_rate": 9.820429236311158e-06,
+      "loss": 0.0397,
+      "step": 340,
+      "student_loss": 0.010333855636417866,
+      "teacher_loss": 0.004541066940873861
+    },
+    {
+      "epoch": 0.09948096885813149,
+      "grad_norm": 0.9921875,
+      "kd_loss": 0.01903965324163437,
+      "learning_rate": 9.81319084632673e-06,
+      "loss": 0.043,
+      "step": 345,
+      "student_loss": 0.012247292324900627,
+      "teacher_loss": 0.00043584281229414046
+    },
+    {
+      "epoch": 0.10092272202998846,
+      "grad_norm": 1.1171875,
+      "kd_loss": 0.037925247102975845,
+      "learning_rate": 9.805812211853534e-06,
+      "loss": 0.034,
+      "step": 350,
+      "student_loss": 0.0080111650750041,
+      "teacher_loss": 0.0011970563791692257
+    },
+    {
+      "epoch": 0.10236447520184544,
+      "grad_norm": 1.4296875,
+      "kd_loss": 0.00829800870269537,
+      "learning_rate": 9.798293547886748e-06,
+      "loss": 0.0396,
+      "step": 355,
+      "student_loss": 0.0005787648260593414,
+      "teacher_loss": 0.0005460750544443727
+    },
+    {
+      "epoch": 0.10380622837370242,
+      "grad_norm": 1.03125,
+      "kd_loss": 0.008213132619857788,
+      "learning_rate": 9.790635073501662e-06,
+      "loss": 0.0393,
+      "step": 360,
+      "student_loss": 0.00047661116695962846,
+      "teacher_loss": 0.00041378207970410585
+    },
+    {
+      "epoch": 0.1052479815455594,
+      "grad_norm": 1.53125,
+      "kd_loss": 0.014927592128515244,
+      "learning_rate": 9.782837011847294e-06,
+      "loss": 0.0436,
+      "step": 365,
+      "student_loss": 0.043972667306661606,
+      "teacher_loss": 0.02826174721121788
+    },
+    {
+      "epoch": 0.10668973471741638,
+      "grad_norm": 1.9375,
+      "kd_loss": 0.03636546432971954,
+      "learning_rate": 9.774899590139897e-06,
+      "loss": 0.0423,
+      "step": 370,
+      "student_loss": 0.021089306101202965,
+      "teacher_loss": 0.0004860333865508437
+    },
+    {
+      "epoch": 0.10813148788927336,
+      "grad_norm": 1.140625,
+      "kd_loss": 0.015318617224693298,
+      "learning_rate": 9.766823039656328e-06,
+      "loss": 0.0357,
+      "step": 375,
+      "student_loss": 0.03581877425312996,
+      "teacher_loss": 0.02248113416135311
+    },
+    {
+      "epoch": 0.10957324106113034,
+      "grad_norm": 0.74609375,
+      "kd_loss": 0.009901116602122784,
+      "learning_rate": 9.758607595727309e-06,
+      "loss": 0.0372,
+      "step": 380,
+      "student_loss": 0.00044091136078350246,
+      "teacher_loss": 0.0003208389098290354
+    },
+    {
+      "epoch": 0.11101499423298732,
+      "grad_norm": 1.5390625,
+      "kd_loss": 0.015398882329463959,
+      "learning_rate": 9.75025349773058e-06,
+      "loss": 0.0348,
+      "step": 385,
+      "student_loss": 0.0006440416909754276,
+      "teacher_loss": 0.0005293422727845609
+    },
+    {
+      "epoch": 0.11245674740484429,
+      "grad_norm": 1.1171875,
+      "kd_loss": 0.009845156222581863,
+      "learning_rate": 9.741760989083914e-06,
+      "loss": 0.0326,
+      "step": 390,
+      "student_loss": 0.00028396621928550303,
+      "teacher_loss": 0.00030337049975059927
+    },
+    {
+      "epoch": 0.11389850057670127,
+      "grad_norm": 1.28125,
+      "kd_loss": 0.011232681572437286,
+      "learning_rate": 9.73313031723803e-06,
+      "loss": 0.038,
+      "step": 395,
+      "student_loss": 0.003544658189639449,
+      "teacher_loss": 0.0032308201771229506
+    },
+    {
+      "epoch": 0.11534025374855825,
+      "grad_norm": 1.609375,
+      "kd_loss": 0.01615743711590767,
+      "learning_rate": 9.724361733669383e-06,
+      "loss": 0.0357,
+      "step": 400,
+      "student_loss": 0.0006839327397756279,
+      "teacher_loss": 0.000942843733355403
+    },
+    {
+      "epoch": 0.11678200692041522,
+      "grad_norm": 1.078125,
+      "kd_loss": 0.013564227148890495,
+      "learning_rate": 9.715455493872827e-06,
+      "loss": 0.0415,
+      "step": 405,
+      "student_loss": 0.005195911042392254,
+      "teacher_loss": 0.00030810022144578397
+    },
+    {
+      "epoch": 0.1182237600922722,
+      "grad_norm": 1.5703125,
+      "kd_loss": 0.025074705481529236,
+      "learning_rate": 9.70641185735419e-06,
+      "loss": 0.0411,
+      "step": 410,
+      "student_loss": 0.01386551558971405,
+      "teacher_loss": 0.00967031717300415
+    },
+    {
+      "epoch": 0.11966551326412918,
+      "grad_norm": 1.1875,
+      "kd_loss": 0.015891144052147865,
+      "learning_rate": 9.697231087622691e-06,
+      "loss": 0.0411,
+      "step": 415,
+      "student_loss": 0.0021395094227045774,
+      "teacher_loss": 0.0021108894143253565
+    },
+    {
+      "epoch": 0.12110726643598616,
+      "grad_norm": 1.40625,
+      "kd_loss": 0.021059690043330193,
+      "learning_rate": 9.687913452183277e-06,
+      "loss": 0.0379,
+      "step": 420,
+      "student_loss": 0.04692540317773819,
+      "teacher_loss": 0.02177572436630726
+    },
+    {
+      "epoch": 0.12254901960784313,
+      "grad_norm": 1.3125,
+      "kd_loss": 0.007998880930244923,
+      "learning_rate": 9.678459222528825e-06,
+      "loss": 0.0366,
+      "step": 425,
+      "student_loss": 0.0025826317723840475,
+      "teacher_loss": 0.002152319997549057
+    },
+    {
+      "epoch": 0.12399077277970011,
+      "grad_norm": 1.5390625,
+      "kd_loss": 0.026353655382990837,
+      "learning_rate": 9.668868674132224e-06,
+      "loss": 0.0448,
+      "step": 430,
+      "student_loss": 0.04476851597428322,
+      "teacher_loss": 0.06174921244382858
+    },
+    {
+      "epoch": 0.1254325259515571,
+      "grad_norm": 1.3359375,
+      "kd_loss": 0.008578015491366386,
+      "learning_rate": 9.65914208643836e-06,
+      "loss": 0.0355,
+      "step": 435,
+      "student_loss": 0.0006695028278045356,
+      "teacher_loss": 0.0005997278494760394
+    },
+    {
+      "epoch": 0.12687427912341406,
+      "grad_norm": 1.1640625,
+      "kd_loss": 0.025591254234313965,
+      "learning_rate": 9.649279742855964e-06,
+      "loss": 0.039,
+      "step": 440,
+      "student_loss": 0.020308705046772957,
+      "teacher_loss": 0.005524828098714352
+    },
+    {
+      "epoch": 0.12831603229527105,
+      "grad_norm": 1.875,
+      "kd_loss": 0.019290607422590256,
+      "learning_rate": 9.639281930749363e-06,
+      "loss": 0.0384,
+      "step": 445,
+      "student_loss": 0.0254460871219635,
+      "teacher_loss": 0.01746205799281597
+    },
+    {
+      "epoch": 0.12975778546712802,
+      "grad_norm": 1.3046875,
+      "kd_loss": 0.01817379891872406,
+      "learning_rate": 9.629148941430096e-06,
+      "loss": 0.0398,
+      "step": 450,
+      "student_loss": 0.0006312711630016565,
+      "teacher_loss": 0.0004661971179302782
+    },
+    {
+      "epoch": 0.131199538638985,
+      "grad_norm": 1.390625,
+      "kd_loss": 0.01092479843646288,
+      "learning_rate": 9.618881070148436e-06,
+      "loss": 0.039,
+      "step": 455,
+      "student_loss": 0.005430892109870911,
+      "teacher_loss": 0.0015254098689183593
+    },
+    {
+      "epoch": 0.13264129181084197,
+      "grad_norm": 1.6484375,
+      "kd_loss": 0.014401745982468128,
+      "learning_rate": 9.608478616084784e-06,
+      "loss": 0.0329,
+      "step": 460,
+      "student_loss": 0.002872939221560955,
+      "teacher_loss": 0.0004281107976567
+    },
+    {
+      "epoch": 0.13408304498269896,
+      "grad_norm": 1.25,
+      "kd_loss": 0.02247895859181881,
+      "learning_rate": 9.597941882340947e-06,
+      "loss": 0.0409,
+      "step": 465,
+      "student_loss": 0.00037059103487990797,
+      "teacher_loss": 0.0019852325785905123
+    },
+    {
+      "epoch": 0.13552479815455595,
+      "grad_norm": 1.2890625,
+      "kd_loss": 0.011446633376181126,
+      "learning_rate": 9.587271175931313e-06,
+      "loss": 0.0372,
+      "step": 470,
+      "student_loss": 0.00029464755789376795,
+      "teacher_loss": 0.00020733379642479122
+    },
+    {
+      "epoch": 0.13696655132641292,
+      "grad_norm": 2.21875,
+      "kd_loss": 0.016915325075387955,
+      "learning_rate": 9.5764668077739e-06,
+      "loss": 0.0312,
+      "step": 475,
+      "student_loss": 0.0005512129282578826,
+      "teacher_loss": 0.00031623090035282075
+    },
+    {
+      "epoch": 0.1384083044982699,
+      "grad_norm": 1.2890625,
+      "kd_loss": 0.016255535185337067,
+      "learning_rate": 9.5655290926813e-06,
+      "loss": 0.0457,
+      "step": 480,
+      "student_loss": 0.015682687982916832,
+      "teacher_loss": 0.013726728036999702
+    },
+    {
+      "epoch": 0.13985005767012687,
+      "grad_norm": 1.109375,
+      "kd_loss": 0.007256851997226477,
+      "learning_rate": 9.554458349351507e-06,
+      "loss": 0.0341,
+      "step": 485,
+      "student_loss": 0.0005507866735570133,
+      "teacher_loss": 0.0005500169354490936
+    },
+    {
+      "epoch": 0.14129181084198386,
+      "grad_norm": 1.9765625,
+      "kd_loss": 0.018985357135534286,
+      "learning_rate": 9.54325490035863e-06,
+      "loss": 0.0445,
+      "step": 490,
+      "student_loss": 0.015820659697055817,
+      "teacher_loss": 0.00336285843513906
+    },
+    {
+      "epoch": 0.14273356401384082,
+      "grad_norm": 1.875,
+      "kd_loss": 0.025550341233611107,
+      "learning_rate": 9.531919072143488e-06,
+      "loss": 0.0483,
+      "step": 495,
+      "student_loss": 0.026045799255371094,
+      "teacher_loss": 0.004566465504467487
+    },
+    {
+      "epoch": 0.14417531718569782,
+      "grad_norm": 1.640625,
+      "kd_loss": 0.017328884452581406,
+      "learning_rate": 9.520451195004107e-06,
+      "loss": 0.0369,
+      "step": 500,
+      "student_loss": 0.0017473739571869373,
+      "teacher_loss": 0.000323916639899835
+    },
+    {
+      "epoch": 0.14561707035755478,
+      "grad_norm": 1.046875,
+      "kd_loss": 0.011184470728039742,
+      "learning_rate": 9.508851603086094e-06,
+      "loss": 0.0436,
+      "step": 505,
+      "student_loss": 0.0003846609324682504,
+      "teacher_loss": 0.0002943307626992464
+    },
+    {
+      "epoch": 0.14705882352941177,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.011295544914901257,
+      "learning_rate": 9.497120634372894e-06,
+      "loss": 0.0376,
+      "step": 510,
+      "student_loss": 0.00031946305534802377,
+      "teacher_loss": 0.0002625001943670213
+    },
+    {
+      "epoch": 0.14850057670126873,
+      "grad_norm": 1.8984375,
+      "kd_loss": 0.01340380497276783,
+      "learning_rate": 9.48525863067595e-06,
+      "loss": 0.0331,
+      "step": 515,
+      "student_loss": 0.007424864452332258,
+      "teacher_loss": 0.00023525822325609624
+    },
+    {
+      "epoch": 0.14994232987312572,
+      "grad_norm": 2.875,
+      "kd_loss": 0.032959483563899994,
+      "learning_rate": 9.473265937624748e-06,
+      "loss": 0.0408,
+      "step": 520,
+      "student_loss": 0.04309098795056343,
+      "teacher_loss": 0.011560679413378239
+    },
+    {
+      "epoch": 0.1513840830449827,
+      "grad_norm": 0.8828125,
+      "kd_loss": 0.01784936524927616,
+      "learning_rate": 9.461142904656722e-06,
+      "loss": 0.0398,
+      "step": 525,
+      "student_loss": 0.0003528800734784454,
+      "teacher_loss": 0.0001730480435071513
+    },
+    {
+      "epoch": 0.15282583621683968,
+      "grad_norm": 0.8515625,
+      "kd_loss": 0.01350821927189827,
+      "learning_rate": 9.448889885007105e-06,
+      "loss": 0.0353,
+      "step": 530,
+      "student_loss": 0.00033670777338556945,
+      "teacher_loss": 0.0002543032169342041
+    },
+    {
+      "epoch": 0.15426758938869667,
+      "grad_norm": 0.890625,
+      "kd_loss": 0.027389032766222954,
+      "learning_rate": 9.436507235698613e-06,
+      "loss": 0.0344,
+      "step": 535,
+      "student_loss": 0.0007846449734643102,
+      "teacher_loss": 0.0004379330202937126
+    },
+    {
+      "epoch": 0.15570934256055363,
+      "grad_norm": 1.734375,
+      "kd_loss": 0.014288514852523804,
+      "learning_rate": 9.423995317531049e-06,
+      "loss": 0.0341,
+      "step": 540,
+      "student_loss": 0.0010072099976241589,
+      "teacher_loss": 0.0003058096335735172
+    },
+    {
+      "epoch": 0.15715109573241062,
+      "grad_norm": 1.796875,
+      "kd_loss": 0.008625246584415436,
+      "learning_rate": 9.4113544950708e-06,
+      "loss": 0.0308,
+      "step": 545,
+      "student_loss": 0.010159535333514214,
+      "teacher_loss": 0.007676934357732534
+    },
+    {
+      "epoch": 0.15859284890426759,
+      "grad_norm": 1.9296875,
+      "kd_loss": 0.02502039261162281,
+      "learning_rate": 9.398585136640195e-06,
+      "loss": 0.0307,
+      "step": 550,
+      "student_loss": 0.00014985659800004214,
+      "teacher_loss": 0.014202730730175972
+    },
+    {
+      "epoch": 0.16003460207612458,
+      "grad_norm": 1.09375,
+      "kd_loss": 0.042002685368061066,
+      "learning_rate": 9.38568761430679e-06,
+      "loss": 0.0389,
+      "step": 555,
+      "student_loss": 0.02874353528022766,
+      "teacher_loss": 0.003934269305318594
+    },
+    {
+      "epoch": 0.16147635524798154,
+      "grad_norm": 1.6796875,
+      "kd_loss": 0.03426447883248329,
+      "learning_rate": 9.372662303872514e-06,
+      "loss": 0.0405,
+      "step": 560,
+      "student_loss": 0.03105512261390686,
+      "teacher_loss": 0.0003118969907518476
+    },
+    {
+      "epoch": 0.16291810841983853,
+      "grad_norm": 1.2109375,
+      "kd_loss": 0.01886710338294506,
+      "learning_rate": 9.359509584862735e-06,
+      "loss": 0.0345,
+      "step": 565,
+      "student_loss": 0.03195136412978172,
+      "teacher_loss": 0.021795904263854027
+    },
+    {
+      "epoch": 0.1643598615916955,
+      "grad_norm": 1.484375,
+      "kd_loss": 0.021855715662240982,
+      "learning_rate": 9.346229840515186e-06,
+      "loss": 0.0418,
+      "step": 570,
+      "student_loss": 0.011120815761387348,
+      "teacher_loss": 0.007288818247616291
+    },
+    {
+      "epoch": 0.16580161476355249,
+      "grad_norm": 1.734375,
+      "kd_loss": 0.013769465498626232,
+      "learning_rate": 9.332823457768805e-06,
+      "loss": 0.0386,
+      "step": 575,
+      "student_loss": 0.0005883977864868939,
+      "teacher_loss": 0.0003054222615901381
+    },
+    {
+      "epoch": 0.16724336793540945,
+      "grad_norm": 1.0703125,
+      "kd_loss": 0.012690847739577293,
+      "learning_rate": 9.31929082725246e-06,
+      "loss": 0.0337,
+      "step": 580,
+      "student_loss": 0.0022747719194740057,
+      "teacher_loss": 0.00218041124753654
+    },
+    {
+      "epoch": 0.16868512110726644,
+      "grad_norm": 0.85546875,
+      "kd_loss": 0.017418578267097473,
+      "learning_rate": 9.305632343273567e-06,
+      "loss": 0.0316,
+      "step": 585,
+      "student_loss": 0.0005172074306756258,
+      "teacher_loss": 0.0002657689619809389
+    },
+    {
+      "epoch": 0.1701268742791234,
+      "grad_norm": 1.1640625,
+      "kd_loss": 0.013570615090429783,
+      "learning_rate": 9.291848403806604e-06,
+      "loss": 0.037,
+      "step": 590,
+      "student_loss": 0.0005453322664834559,
+      "teacher_loss": 0.0003946890647057444
+    },
+    {
+      "epoch": 0.1715686274509804,
+      "grad_norm": 1.265625,
+      "kd_loss": 0.010840360075235367,
+      "learning_rate": 9.277939410481507e-06,
+      "loss": 0.0353,
+      "step": 595,
+      "student_loss": 0.00048185884952545166,
+      "teacher_loss": 0.00042216762085445225
+    },
+    {
+      "epoch": 0.17301038062283736,
+      "grad_norm": 1.390625,
+      "kd_loss": 0.02166227251291275,
+      "learning_rate": 9.263905768571977e-06,
+      "loss": 0.0417,
+      "step": 600,
+      "student_loss": 0.05149306356906891,
+      "teacher_loss": 0.009222068823873997
+    },
+    {
+      "epoch": 0.17445213379469435,
+      "grad_norm": 1.109375,
+      "kd_loss": 0.021928608417510986,
+      "learning_rate": 9.249747886983664e-06,
+      "loss": 0.0387,
+      "step": 605,
+      "student_loss": 0.06224099546670914,
+      "teacher_loss": 0.03538208454847336
+    },
+    {
+      "epoch": 0.17589388696655134,
+      "grad_norm": 1.359375,
+      "kd_loss": 0.0258643738925457,
+      "learning_rate": 9.235466178242255e-06,
+      "loss": 0.0353,
+      "step": 610,
+      "student_loss": 0.002640328835695982,
+      "teacher_loss": 0.0018086276249960065
+    },
+    {
+      "epoch": 0.1773356401384083,
+      "grad_norm": 0.9453125,
+      "kd_loss": 0.013348888605833054,
+      "learning_rate": 9.221061058481456e-06,
+      "loss": 0.0322,
+      "step": 615,
+      "student_loss": 0.000250897224759683,
+      "teacher_loss": 0.00027990140370093286
+    },
+    {
+      "epoch": 0.1787773933102653,
+      "grad_norm": 0.6875,
+      "kd_loss": 0.019072921946644783,
+      "learning_rate": 9.206532947430865e-06,
+      "loss": 0.0416,
+      "step": 620,
+      "student_loss": 0.010811248794198036,
+      "teacher_loss": 0.007016193587332964
+    },
+    {
+      "epoch": 0.18021914648212226,
+      "grad_norm": 0.98046875,
+      "kd_loss": 0.014861532486975193,
+      "learning_rate": 9.191882268403743e-06,
+      "loss": 0.0335,
+      "step": 625,
+      "student_loss": 0.002400943310931325,
+      "teacher_loss": 0.0028365827165544033
+    },
+    {
+      "epoch": 0.18166089965397925,
+      "grad_norm": 2.359375,
+      "kd_loss": 0.02572929672896862,
+      "learning_rate": 9.177109448284673e-06,
+      "loss": 0.0401,
+      "step": 630,
+      "student_loss": 0.028089625760912895,
+      "teacher_loss": 0.012286370620131493
+    },
+    {
+      "epoch": 0.1831026528258362,
+      "grad_norm": 1.203125,
+      "kd_loss": 0.01528047863394022,
+      "learning_rate": 9.162214917517139e-06,
+      "loss": 0.0528,
+      "step": 635,
+      "student_loss": 0.002457874594256282,
+      "teacher_loss": 0.0002723166544456035
+    },
+    {
+      "epoch": 0.1845444059976932,
+      "grad_norm": 1.6640625,
+      "kd_loss": 0.016974100843071938,
+      "learning_rate": 9.14719911009096e-06,
+      "loss": 0.0327,
+      "step": 640,
+      "student_loss": 0.0006965341162867844,
+      "teacher_loss": 0.0008102107094600797
+    },
+    {
+      "epoch": 0.18598615916955016,
+      "grad_norm": 1.9609375,
+      "kd_loss": 0.01875746250152588,
+      "learning_rate": 9.132062463529666e-06,
+      "loss": 0.033,
+      "step": 645,
+      "student_loss": 0.0016925447853282094,
+      "teacher_loss": 0.005389066878706217
+    },
+    {
+      "epoch": 0.18742791234140715,
+      "grad_norm": 1.578125,
+      "kd_loss": 0.009372042492032051,
+      "learning_rate": 9.11680541887774e-06,
+      "loss": 0.0324,
+      "step": 650,
+      "student_loss": 0.0010497953044250607,
+      "teacher_loss": 0.0008989371126517653
+    },
+    {
+      "epoch": 0.18886966551326412,
+      "grad_norm": 3.9375,
+      "kd_loss": 0.019143380224704742,
+      "learning_rate": 9.101428420687759e-06,
+      "loss": 0.0368,
+      "step": 655,
+      "student_loss": 0.0008799026254564524,
+      "teacher_loss": 0.00043919464224018157
+    },
+    {
+      "epoch": 0.1903114186851211,
+      "grad_norm": 1.921875,
+      "kd_loss": 0.013875705190002918,
+      "learning_rate": 9.085931917007464e-06,
+      "loss": 0.0405,
+      "step": 660,
+      "student_loss": 0.000673408736474812,
+      "teacher_loss": 0.0005781051004305482
+    },
+    {
+      "epoch": 0.19175317185697807,
+      "grad_norm": 1.6015625,
+      "kd_loss": 0.043592095375061035,
+      "learning_rate": 9.070316359366681e-06,
+      "loss": 0.0413,
+      "step": 665,
+      "student_loss": 0.0031369002535939217,
+      "teacher_loss": 0.001360239228233695
+    },
+    {
+      "epoch": 0.19319492502883506,
+      "grad_norm": 8.5,
+      "kd_loss": 0.01952010951936245,
+      "learning_rate": 9.054582202764175e-06,
+      "loss": 0.042,
+      "step": 670,
+      "student_loss": 0.012506276369094849,
+      "teacher_loss": 0.0028849972877651453
+    },
+    {
+      "epoch": 0.19463667820069205,
+      "grad_norm": 0.9609375,
+      "kd_loss": 0.008069864474236965,
+      "learning_rate": 9.038729905654393e-06,
+      "loss": 0.0365,
+      "step": 675,
+      "student_loss": 0.0004877475730609149,
+      "teacher_loss": 0.0004426161467563361
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 1.59375,
+      "kd_loss": 0.021699784323573112,
+      "learning_rate": 9.022759929934109e-06,
+      "loss": 0.0317,
+      "step": 680,
+      "student_loss": 0.07352294772863388,
+      "teacher_loss": 0.033334095031023026
+    },
+    {
+      "epoch": 0.197520184544406,
+      "grad_norm": 1.125,
+      "kd_loss": 0.014409462921321392,
+      "learning_rate": 9.006672740928952e-06,
+      "loss": 0.0353,
+      "step": 685,
+      "student_loss": 0.03678691014647484,
+      "teacher_loss": 0.02936466597020626
+    },
+    {
+      "epoch": 0.19896193771626297,
+      "grad_norm": 2.140625,
+      "kd_loss": 0.034733690321445465,
+      "learning_rate": 8.990468807379861e-06,
+      "loss": 0.0368,
+      "step": 690,
+      "student_loss": 0.0369374081492424,
+      "teacher_loss": 0.008476306684315205
+    },
+    {
+      "epoch": 0.20040369088811996,
+      "grad_norm": 1.453125,
+      "kd_loss": 0.016646580770611763,
+      "learning_rate": 8.974148601429426e-06,
+      "loss": 0.0392,
+      "step": 695,
+      "student_loss": 0.007193500176072121,
+      "teacher_loss": 0.005883329547941685
+    },
+    {
+      "epoch": 0.20184544405997693,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.01161164604127407,
+      "learning_rate": 8.957712598608123e-06,
+      "loss": 0.0341,
+      "step": 700,
+      "student_loss": 0.0006814897060394287,
+      "teacher_loss": 0.00038474920438602567
+    },
+    {
+      "epoch": 0.20328719723183392,
+      "grad_norm": 1.1015625,
+      "kd_loss": 0.008321454748511314,
+      "learning_rate": 8.941161277820462e-06,
+      "loss": 0.0348,
+      "step": 705,
+      "student_loss": 0.0013638287782669067,
+      "teacher_loss": 0.0012245745165273547
+    },
+    {
+      "epoch": 0.20472895040369088,
+      "grad_norm": 1.28125,
+      "kd_loss": 0.036699049174785614,
+      "learning_rate": 8.924495121331038e-06,
+      "loss": 0.0377,
+      "step": 710,
+      "student_loss": 0.011128552258014679,
+      "teacher_loss": 0.0038417454343289137
+    },
+    {
+      "epoch": 0.20617070357554787,
+      "grad_norm": 1.984375,
+      "kd_loss": 0.01310927327722311,
+      "learning_rate": 8.907714614750473e-06,
+      "loss": 0.0459,
+      "step": 715,
+      "student_loss": 0.003388170152902603,
+      "teacher_loss": 0.00937573704868555
+    },
+    {
+      "epoch": 0.20761245674740483,
+      "grad_norm": 1.640625,
+      "kd_loss": 0.016016153618693352,
+      "learning_rate": 8.890820247021263e-06,
+      "loss": 0.0374,
+      "step": 720,
+      "student_loss": 0.00045992882223799825,
+      "teacher_loss": 0.0004658737452700734
+    },
+    {
+      "epoch": 0.20905420991926182,
+      "grad_norm": 1.1640625,
+      "kd_loss": 0.009306555613875389,
+      "learning_rate": 8.873812510403545e-06,
+      "loss": 0.029,
+      "step": 725,
+      "student_loss": 0.0005832040915265679,
+      "teacher_loss": 0.0004866627568844706
+    },
+    {
+      "epoch": 0.2104959630911188,
+      "grad_norm": 0.953125,
+      "kd_loss": 0.019967591390013695,
+      "learning_rate": 8.85669190046074e-06,
+      "loss": 0.034,
+      "step": 730,
+      "student_loss": 0.007923947647213936,
+      "teacher_loss": 0.005159601103514433
+    },
+    {
+      "epoch": 0.21193771626297578,
+      "grad_norm": 1.453125,
+      "kd_loss": 0.014439092949032784,
+      "learning_rate": 8.839458916045117e-06,
+      "loss": 0.041,
+      "step": 735,
+      "student_loss": 0.00028902897611260414,
+      "teacher_loss": 0.00017782236682251096
+    },
+    {
+      "epoch": 0.21337946943483277,
+      "grad_norm": 1.7890625,
+      "kd_loss": 0.02189582586288452,
+      "learning_rate": 8.822114059283267e-06,
+      "loss": 0.0323,
+      "step": 740,
+      "student_loss": 0.0005453451885841787,
+      "teacher_loss": 0.00017786624084692448
+    },
+    {
+      "epoch": 0.21482122260668973,
+      "grad_norm": 1.53125,
+      "kd_loss": 0.013907963410019875,
+      "learning_rate": 8.804657835561456e-06,
+      "loss": 0.0349,
+      "step": 745,
+      "student_loss": 0.038910508155822754,
+      "teacher_loss": 0.020556708797812462
+    },
+    {
+      "epoch": 0.21626297577854672,
+      "grad_norm": 1.3046875,
+      "kd_loss": 0.03312985599040985,
+      "learning_rate": 8.787090753510917e-06,
+      "loss": 0.0353,
+      "step": 750,
+      "student_loss": 0.01961570419371128,
+      "teacher_loss": 0.005801274906843901
+    },
+    {
+      "epoch": 0.2177047289504037,
+      "grad_norm": 2.5,
+      "kd_loss": 0.01383906975388527,
+      "learning_rate": 8.76941332499301e-06,
+      "loss": 0.0397,
+      "step": 755,
+      "student_loss": 0.027118289843201637,
+      "teacher_loss": 0.001979502383619547
+    },
+    {
+      "epoch": 0.21914648212226068,
+      "grad_norm": 1.1484375,
+      "kd_loss": 0.026841050013899803,
+      "learning_rate": 8.751626065084328e-06,
+      "loss": 0.0304,
+      "step": 760,
+      "student_loss": 0.006133708171546459,
+      "teacher_loss": 0.0004118960350751877
+    },
+    {
+      "epoch": 0.22058823529411764,
+      "grad_norm": 0.984375,
+      "kd_loss": 0.028871504589915276,
+      "learning_rate": 8.733729492061675e-06,
+      "loss": 0.0377,
+      "step": 765,
+      "student_loss": 0.005369402468204498,
+      "teacher_loss": 0.0012096863938495517
+    },
+    {
+      "epoch": 0.22202998846597463,
+      "grad_norm": 0.921875,
+      "kd_loss": 0.028200991451740265,
+      "learning_rate": 8.715724127386971e-06,
+      "loss": 0.0337,
+      "step": 770,
+      "student_loss": 0.007010896224528551,
+      "teacher_loss": 0.001293784473091364
+    },
+    {
+      "epoch": 0.2234717416378316,
+      "grad_norm": 1.09375,
+      "kd_loss": 0.015337961725890636,
+      "learning_rate": 8.697610495692055e-06,
+      "loss": 0.0314,
+      "step": 775,
+      "student_loss": 0.007272190414369106,
+      "teacher_loss": 0.002003727713599801
+    },
+    {
+      "epoch": 0.22491349480968859,
+      "grad_norm": 1.140625,
+      "kd_loss": 0.014271454885601997,
+      "learning_rate": 8.679389124763396e-06,
+      "loss": 0.0357,
+      "step": 780,
+      "student_loss": 0.0013534394092857838,
+      "teacher_loss": 0.0002126321487594396
+    },
+    {
+      "epoch": 0.22635524798154555,
+      "grad_norm": 2.328125,
+      "kd_loss": 0.007646733429282904,
+      "learning_rate": 8.661060545526724e-06,
+      "loss": 0.0312,
+      "step": 785,
+      "student_loss": 0.000303611159324646,
+      "teacher_loss": 0.0003056841087527573
+    },
+    {
+      "epoch": 0.22779700115340254,
+      "grad_norm": 1.7890625,
+      "kd_loss": 0.009472047910094261,
+      "learning_rate": 8.64262529203155e-06,
+      "loss": 0.03,
+      "step": 790,
+      "student_loss": 0.00036868476308882236,
+      "teacher_loss": 0.0003439165884628892
+    },
+    {
+      "epoch": 0.2292387543252595,
+      "grad_norm": 1.1015625,
+      "kd_loss": 0.013042761944234371,
+      "learning_rate": 8.624083901435613e-06,
+      "loss": 0.0345,
+      "step": 795,
+      "student_loss": 0.0004417412565089762,
+      "teacher_loss": 0.0002294621808687225
+    },
+    {
+      "epoch": 0.2306805074971165,
+      "grad_norm": 1.7578125,
+      "kd_loss": 0.02265016920864582,
+      "learning_rate": 8.60543691398922e-06,
+      "loss": 0.0413,
+      "step": 800,
+      "student_loss": 0.08563552051782608,
+      "teacher_loss": 0.039897482842206955
+    },
+    {
+      "epoch": 0.23212226066897348,
+      "grad_norm": 1.3125,
+      "kd_loss": 0.009194138459861279,
+      "learning_rate": 8.586684873019513e-06,
+      "loss": 0.0381,
+      "step": 805,
+      "student_loss": 0.00031980325002223253,
+      "teacher_loss": 0.0002621679159346968
+    },
+    {
+      "epoch": 0.23356401384083045,
+      "grad_norm": 15.8125,
+      "kd_loss": 0.015975888818502426,
+      "learning_rate": 8.567828324914633e-06,
+      "loss": 0.038,
+      "step": 810,
+      "student_loss": 0.0005471970071084797,
+      "teacher_loss": 0.00027335959021002054
+    },
+    {
+      "epoch": 0.23500576701268744,
+      "grad_norm": 2.25,
+      "kd_loss": 0.009164939634501934,
+      "learning_rate": 8.548867819107803e-06,
+      "loss": 0.0324,
+      "step": 815,
+      "student_loss": 0.0015802276320755482,
+      "teacher_loss": 0.008377115242183208
+    },
+    {
+      "epoch": 0.2364475201845444,
+      "grad_norm": 1.296875,
+      "kd_loss": 0.018547918647527695,
+      "learning_rate": 8.52980390806131e-06,
+      "loss": 0.0343,
+      "step": 820,
+      "student_loss": 0.008644208312034607,
+      "teacher_loss": 0.0001676520478213206
+    },
+    {
+      "epoch": 0.2378892733564014,
+      "grad_norm": 1.390625,
+      "kd_loss": 0.019806109368801117,
+      "learning_rate": 8.51063714725042e-06,
+      "loss": 0.032,
+      "step": 825,
+      "student_loss": 0.0006446156185120344,
+      "teacher_loss": 0.016100004315376282
+    },
+    {
+      "epoch": 0.23933102652825836,
+      "grad_norm": 1.296875,
+      "kd_loss": 0.01830875873565674,
+      "learning_rate": 8.49136809514719e-06,
+      "loss": 0.0352,
+      "step": 830,
+      "student_loss": 0.00992797315120697,
+      "teacher_loss": 0.004923146218061447
+    },
+    {
+      "epoch": 0.24077277970011535,
+      "grad_norm": 1.265625,
+      "kd_loss": 0.015559202991425991,
+      "learning_rate": 8.471997313204183e-06,
+      "loss": 0.0308,
+      "step": 835,
+      "student_loss": 0.01591889001429081,
+      "teacher_loss": 0.0025253912899643183
+    },
+    {
+      "epoch": 0.2422145328719723,
+      "grad_norm": 1.46875,
+      "kd_loss": 0.01866515353322029,
+      "learning_rate": 8.452525365838123e-06,
+      "loss": 0.0346,
+      "step": 840,
+      "student_loss": 0.001301256357692182,
+      "teacher_loss": 0.00011679113958962262
+    },
+    {
+      "epoch": 0.2436562860438293,
+      "grad_norm": 0.859375,
+      "kd_loss": 0.010034815408289433,
+      "learning_rate": 8.432952820413447e-06,
+      "loss": 0.0498,
+      "step": 845,
+      "student_loss": 0.0005439711967483163,
+      "teacher_loss": 0.0004172325134277344
+    },
+    {
+      "epoch": 0.24509803921568626,
+      "grad_norm": 1.78125,
+      "kd_loss": 0.023002484813332558,
+      "learning_rate": 8.41328024722577e-06,
+      "loss": 0.0335,
+      "step": 850,
+      "student_loss": 0.00615985132753849,
+      "teacher_loss": 0.00031692718039266765
+    },
+    {
+      "epoch": 0.24653979238754326,
+      "grad_norm": 1.140625,
+      "kd_loss": 0.011961855925619602,
+      "learning_rate": 8.393508219485263e-06,
+      "loss": 0.034,
+      "step": 855,
+      "student_loss": 0.0002558475243858993,
+      "teacher_loss": 0.000246651004999876
+    },
+    {
+      "epoch": 0.24798154555940022,
+      "grad_norm": 0.80859375,
+      "kd_loss": 0.012495153583586216,
+      "learning_rate": 8.373637313299966e-06,
+      "loss": 0.0399,
+      "step": 860,
+      "student_loss": 0.0005405155243352056,
+      "teacher_loss": 0.0004368047812022269
+    },
+    {
+      "epoch": 0.2494232987312572,
+      "grad_norm": 1.0078125,
+      "kd_loss": 0.014805592596530914,
+      "learning_rate": 8.353668107658984e-06,
+      "loss": 0.0337,
+      "step": 865,
+      "student_loss": 0.0005246512009762228,
+      "teacher_loss": 0.00047315956908278167
+    },
+    {
+      "epoch": 0.2508650519031142,
+      "grad_norm": 1.1171875,
+      "kd_loss": 0.015027549117803574,
+      "learning_rate": 8.333601184415632e-06,
+      "loss": 0.0341,
+      "step": 870,
+      "student_loss": 0.0015372103080153465,
+      "teacher_loss": 0.0004421547637321055
+    },
+    {
+      "epoch": 0.25230680507497116,
+      "grad_norm": 1.1171875,
+      "kd_loss": 0.010914859361946583,
+      "learning_rate": 8.31343712827047e-06,
+      "loss": 0.0345,
+      "step": 875,
+      "student_loss": 0.0001896051544463262,
+      "teacher_loss": 0.00020151768694631755
+    },
+    {
+      "epoch": 0.2537485582468281,
+      "grad_norm": 2.015625,
+      "kd_loss": 0.01555558666586876,
+      "learning_rate": 8.293176526754274e-06,
+      "loss": 0.0384,
+      "step": 880,
+      "student_loss": 0.002947386121377349,
+      "teacher_loss": 0.0017182360170409083
+    },
+    {
+      "epoch": 0.25519031141868515,
+      "grad_norm": 1.2265625,
+      "kd_loss": 0.04539031907916069,
+      "learning_rate": 8.27281997021091e-06,
+      "loss": 0.0301,
+      "step": 885,
+      "student_loss": 0.014408277347683907,
+      "teacher_loss": 0.045094698667526245
+    },
+    {
+      "epoch": 0.2566320645905421,
+      "grad_norm": 1.046875,
+      "kd_loss": 0.007991768419742584,
+      "learning_rate": 8.25236805178014e-06,
+      "loss": 0.0355,
+      "step": 890,
+      "student_loss": 0.0003164609370287508,
+      "teacher_loss": 0.0002852486213669181
+    },
+    {
+      "epoch": 0.25807381776239907,
+      "grad_norm": 1.1953125,
+      "kd_loss": 0.02082483097910881,
+      "learning_rate": 8.231821367380335e-06,
+      "loss": 0.0331,
+      "step": 895,
+      "student_loss": 0.011601348407566547,
+      "teacher_loss": 0.004591627512127161
+    },
+    {
+      "epoch": 0.25951557093425603,
+      "grad_norm": 0.9609375,
+      "kd_loss": 0.014409971423447132,
+      "learning_rate": 8.211180515691108e-06,
+      "loss": 0.0354,
+      "step": 900,
+      "student_loss": 0.03723900392651558,
+      "teacher_loss": 0.027190692722797394
+    },
+    {
+      "epoch": 0.26095732410611305,
+      "grad_norm": 1.4375,
+      "kd_loss": 0.014778131619095802,
+      "learning_rate": 8.190446098135883e-06,
+      "loss": 0.0322,
+      "step": 905,
+      "student_loss": 0.018417518585920334,
+      "teacher_loss": 0.015774844214320183
+    },
+    {
+      "epoch": 0.26239907727797,
+      "grad_norm": 1.015625,
+      "kd_loss": 0.007474376354366541,
+      "learning_rate": 8.16961871886435e-06,
+      "loss": 0.0291,
+      "step": 910,
+      "student_loss": 0.00027324067195877433,
+      "teacher_loss": 0.000292893877485767
+    },
+    {
+      "epoch": 0.263840830449827,
+      "grad_norm": 2.796875,
+      "kd_loss": 0.05165571719408035,
+      "learning_rate": 8.148698984734887e-06,
+      "loss": 0.0351,
+      "step": 915,
+      "student_loss": 0.05543224513530731,
+      "teacher_loss": 0.0008424259722232819
+    },
+    {
+      "epoch": 0.26528258362168394,
+      "grad_norm": 1.109375,
+      "kd_loss": 0.013850719667971134,
+      "learning_rate": 8.127687505296852e-06,
+      "loss": 0.0326,
+      "step": 920,
+      "student_loss": 0.013253331184387207,
+      "teacher_loss": 0.011715217493474483
+    },
+    {
+      "epoch": 0.26672433679354096,
+      "grad_norm": 3.28125,
+      "kd_loss": 0.00828032847493887,
+      "learning_rate": 8.106584892772844e-06,
+      "loss": 0.0316,
+      "step": 925,
+      "student_loss": 0.00028670401661656797,
+      "teacher_loss": 0.0004561161040328443
+    },
+    {
+      "epoch": 0.2681660899653979,
+      "grad_norm": 2.203125,
+      "kd_loss": 0.013576381839811802,
+      "learning_rate": 8.08539176204085e-06,
+      "loss": 0.0326,
+      "step": 930,
+      "student_loss": 0.002268102252855897,
+      "teacher_loss": 0.001438588136807084
+    },
+    {
+      "epoch": 0.2696078431372549,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.009203169494867325,
+      "learning_rate": 8.064108730616336e-06,
+      "loss": 0.0329,
+      "step": 935,
+      "student_loss": 0.054588161408901215,
+      "teacher_loss": 0.04182029515504837
+    },
+    {
+      "epoch": 0.2710495963091119,
+      "grad_norm": 0.96484375,
+      "kd_loss": 0.022441105917096138,
+      "learning_rate": 8.042736418634252e-06,
+      "loss": 0.0339,
+      "step": 940,
+      "student_loss": 0.003998172469437122,
+      "teacher_loss": 0.0004027044924441725
+    },
+    {
+      "epoch": 0.27249134948096887,
+      "grad_norm": 2.078125,
+      "kd_loss": 0.009604145772755146,
+      "learning_rate": 8.02127544883096e-06,
+      "loss": 0.0358,
+      "step": 945,
+      "student_loss": 0.0019527329131960869,
+      "teacher_loss": 0.001390090910717845
+    },
+    {
+      "epoch": 0.27393310265282583,
+      "grad_norm": 0.8828125,
+      "kd_loss": 0.01388311106711626,
+      "learning_rate": 7.999726446526095e-06,
+      "loss": 0.0306,
+      "step": 950,
+      "student_loss": 0.00017252952966373414,
+      "teacher_loss": 0.00010741472942754626
+    },
+    {
+      "epoch": 0.2753748558246828,
+      "grad_norm": 1.7109375,
+      "kd_loss": 0.01476337481290102,
+      "learning_rate": 7.978090039604342e-06,
+      "loss": 0.0348,
+      "step": 955,
+      "student_loss": 0.0003931497922167182,
+      "teacher_loss": 0.00022118802007753402
+    },
+    {
+      "epoch": 0.2768166089965398,
+      "grad_norm": 2.046875,
+      "kd_loss": 0.017180806025862694,
+      "learning_rate": 7.956366858497138e-06,
+      "loss": 0.0379,
+      "step": 960,
+      "student_loss": 0.008577854372560978,
+      "teacher_loss": 0.0037893427070230246
+    },
+    {
+      "epoch": 0.2782583621683968,
+      "grad_norm": 1.828125,
+      "kd_loss": 0.024987826123833656,
+      "learning_rate": 7.934557536164304e-06,
+      "loss": 0.0349,
+      "step": 965,
+      "student_loss": 0.010798306204378605,
+      "teacher_loss": 0.006717974785715342
+    },
+    {
+      "epoch": 0.27970011534025374,
+      "grad_norm": 1.390625,
+      "kd_loss": 0.01618223264813423,
+      "learning_rate": 7.91266270807561e-06,
+      "loss": 0.0336,
+      "step": 970,
+      "student_loss": 0.00025495997397229075,
+      "teacher_loss": 0.00014070281758904457
+    },
+    {
+      "epoch": 0.2811418685121107,
+      "grad_norm": 1.3828125,
+      "kd_loss": 0.011907395906746387,
+      "learning_rate": 7.890683012192249e-06,
+      "loss": 0.0329,
+      "step": 975,
+      "student_loss": 0.016343005001544952,
+      "teacher_loss": 0.0006543407216668129
+    },
+    {
+      "epoch": 0.2825836216839677,
+      "grad_norm": 1.125,
+      "kd_loss": 0.019409775733947754,
+      "learning_rate": 7.868619088948246e-06,
+      "loss": 0.0339,
+      "step": 980,
+      "student_loss": 0.00046765804290771484,
+      "teacher_loss": 0.00030171783873811364
+    },
+    {
+      "epoch": 0.2840253748558247,
+      "grad_norm": 2.328125,
+      "kd_loss": 0.014747396111488342,
+      "learning_rate": 7.846471581231814e-06,
+      "loss": 0.0359,
+      "step": 985,
+      "student_loss": 0.00035184339503757656,
+      "teacher_loss": 0.0015139622846618295
+    },
+    {
+      "epoch": 0.28546712802768165,
+      "grad_norm": 0.91796875,
+      "kd_loss": 0.011445100419223309,
+      "learning_rate": 7.824241134366604e-06,
+      "loss": 0.029,
+      "step": 990,
+      "student_loss": 0.00024842782295309007,
+      "teacher_loss": 0.00020773368305526674
+    },
+    {
+      "epoch": 0.2869088811995386,
+      "grad_norm": 1.0703125,
+      "kd_loss": 0.025197861716151237,
+      "learning_rate": 7.801928396092908e-06,
+      "loss": 0.0406,
+      "step": 995,
+      "student_loss": 0.00697642145678401,
+      "teacher_loss": 0.0003467589558567852
+    },
+    {
+      "epoch": 0.28835063437139563,
+      "grad_norm": 1.1953125,
+      "kd_loss": 0.02068255841732025,
+      "learning_rate": 7.779534016548791e-06,
+      "loss": 0.0341,
+      "step": 1000,
+      "student_loss": 0.006020409055054188,
+      "teacher_loss": 0.0005438941298052669
+    }
+  ],
+  "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/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1500/added_tokens.json

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

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1500/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. 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 Qwen, created by Alibaba Cloud. You 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/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1500/config.json

@@ -0,0 +1,58 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "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/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1500/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bcf268c20104db5a1773f1ddc6a03880435dbc67ce8c30567c8e6934c714a16d
+size 3096275515

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1500/special_tokens_map.json

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

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1500/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ce224d584d9566df132439041a2b3d6d164dda6f1bc52652d85582be32506f1
+size 9445953

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1500/tokenizer.json

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

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1500/tokenizer_config.json

@@ -0,0 +1,295 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<|trace_context_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151666": {
+      "content": "<|call_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151667": {
+      "content": "<|line_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151668": {
+      "content": "<|return_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151669": {
+      "content": "<|exception_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<|action_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<|arg_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151672": {
+      "content": "<|frame_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151673": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151674": {
+      "content": "<|latent_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151675": {
+      "content": "<|latent_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-1500/trainer_state.json

@@ -0,0 +1,3034 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.43252595155709345,
+  "eval_steps": 500,
+  "global_step": 1500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0014417531718569781,
+      "grad_norm": 34.25,
+      "kd_loss": 0.2726191282272339,
+      "learning_rate": 4.444444444444445e-07,
+      "loss": 0.3835,
+      "step": 5,
+      "student_loss": 0.053846802562475204,
+      "teacher_loss": 0.00042048963950946927
+    },
+    {
+      "epoch": 0.0028835063437139563,
+      "grad_norm": 48.0,
+      "kd_loss": 0.24393108487129211,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 0.3789,
+      "step": 10,
+      "student_loss": 0.09970586746931076,
+      "teacher_loss": 0.0007505021058022976
+    },
+    {
+      "epoch": 0.004325259515570935,
+      "grad_norm": 21.75,
+      "kd_loss": 0.21377132833003998,
+      "learning_rate": 1.5555555555555558e-06,
+      "loss": 0.3585,
+      "step": 15,
+      "student_loss": 0.029451103881001472,
+      "teacher_loss": 0.0006585777155123651
+    },
+    {
+      "epoch": 0.0057670126874279125,
+      "grad_norm": 27.5,
+      "kd_loss": 0.2140207439661026,
+      "learning_rate": 2.1111111111111114e-06,
+      "loss": 0.3372,
+      "step": 20,
+      "student_loss": 0.1321348249912262,
+      "teacher_loss": 0.000296285143122077
+    },
+    {
+      "epoch": 0.00720876585928489,
+      "grad_norm": 14.5625,
+      "kd_loss": 0.18640007078647614,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 0.2845,
+      "step": 25,
+      "student_loss": 0.018293429166078568,
+      "teacher_loss": 0.0010102478554472327
+    },
+    {
+      "epoch": 0.00865051903114187,
+      "grad_norm": 7.1875,
+      "kd_loss": 0.14892512559890747,
+      "learning_rate": 3.2222222222222227e-06,
+      "loss": 0.2387,
+      "step": 30,
+      "student_loss": 0.005030089057981968,
+      "teacher_loss": 0.0003383651783224195
+    },
+    {
+      "epoch": 0.010092272202998846,
+      "grad_norm": 5.5,
+      "kd_loss": 0.1269177347421646,
+      "learning_rate": 3.777777777777778e-06,
+      "loss": 0.2117,
+      "step": 35,
+      "student_loss": 0.0014062788104638457,
+      "teacher_loss": 0.00025501433992758393
+    },
+    {
+      "epoch": 0.011534025374855825,
+      "grad_norm": 4.65625,
+      "kd_loss": 0.12731066346168518,
+      "learning_rate": 4.333333333333334e-06,
+      "loss": 0.1863,
+      "step": 40,
+      "student_loss": 0.049713075160980225,
+      "teacher_loss": 0.040033385157585144
+    },
+    {
+      "epoch": 0.012975778546712802,
+      "grad_norm": 7.375,
+      "kd_loss": 0.1775631606578827,
+      "learning_rate": 4.888888888888889e-06,
+      "loss": 0.1456,
+      "step": 45,
+      "student_loss": 0.06933413445949554,
+      "teacher_loss": 0.0009156111045740545
+    },
+    {
+      "epoch": 0.01441753171856978,
+      "grad_norm": 4.15625,
+      "kd_loss": 0.06588269770145416,
+      "learning_rate": 5.444444444444445e-06,
+      "loss": 0.1417,
+      "step": 50,
+      "student_loss": 0.008036560378968716,
+      "teacher_loss": 0.004471915308386087
+    },
+    {
+      "epoch": 0.015859284890426758,
+      "grad_norm": 8.375,
+      "kd_loss": 0.05523190647363663,
+      "learning_rate": 6e-06,
+      "loss": 0.1225,
+      "step": 55,
+      "student_loss": 0.0012776668881997466,
+      "teacher_loss": 0.00044586719013750553
+    },
+    {
+      "epoch": 0.01730103806228374,
+      "grad_norm": 5.125,
+      "kd_loss": 0.049259770661592484,
+      "learning_rate": 6.555555555555556e-06,
+      "loss": 0.1047,
+      "step": 60,
+      "student_loss": 0.01390351727604866,
+      "teacher_loss": 0.002199392532929778
+    },
+    {
+      "epoch": 0.018742791234140715,
+      "grad_norm": 7.375,
+      "kd_loss": 0.047981150448322296,
+      "learning_rate": 7.111111111111112e-06,
+      "loss": 0.1006,
+      "step": 65,
+      "student_loss": 0.04912174493074417,
+      "teacher_loss": 0.05355874449014664
+    },
+    {
+      "epoch": 0.020184544405997693,
+      "grad_norm": 5.21875,
+      "kd_loss": 0.07241208106279373,
+      "learning_rate": 7.666666666666667e-06,
+      "loss": 0.0935,
+      "step": 70,
+      "student_loss": 0.03459472954273224,
+      "teacher_loss": 0.0029977799858897924
+    },
+    {
+      "epoch": 0.02162629757785467,
+      "grad_norm": 5.125,
+      "kd_loss": 0.0307459756731987,
+      "learning_rate": 8.222222222222222e-06,
+      "loss": 0.0907,
+      "step": 75,
+      "student_loss": 0.0016565826954320073,
+      "teacher_loss": 0.0008985538734123111
+    },
+    {
+      "epoch": 0.02306805074971165,
+      "grad_norm": 1.9453125,
+      "kd_loss": 0.026722930371761322,
+      "learning_rate": 8.777777777777778e-06,
+      "loss": 0.0788,
+      "step": 80,
+      "student_loss": 0.0021714135073125362,
+      "teacher_loss": 0.0012323743430897593
+    },
+    {
+      "epoch": 0.024509803921568627,
+      "grad_norm": 3.71875,
+      "kd_loss": 0.023355644196271896,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.0733,
+      "step": 85,
+      "student_loss": 0.0012368966126814485,
+      "teacher_loss": 0.0009521314641460776
+    },
+    {
+      "epoch": 0.025951557093425604,
+      "grad_norm": 1.75,
+      "kd_loss": 0.0727810338139534,
+      "learning_rate": 9.88888888888889e-06,
+      "loss": 0.0676,
+      "step": 90,
+      "student_loss": 0.028416095301508904,
+      "teacher_loss": 0.005437986925244331
+    },
+    {
+      "epoch": 0.027393310265282585,
+      "grad_norm": 2.75,
+      "kd_loss": 0.04388673976063728,
+      "learning_rate": 9.99995337991255e-06,
+      "loss": 0.0638,
+      "step": 95,
+      "student_loss": 0.009086629375815392,
+      "teacher_loss": 0.005068701226264238
+    },
+    {
+      "epoch": 0.02883506343713956,
+      "grad_norm": 3.328125,
+      "kd_loss": 0.018161524087190628,
+      "learning_rate": 9.999763987297266e-06,
+      "loss": 0.0537,
+      "step": 100,
+      "student_loss": 0.0007679096888750792,
+      "teacher_loss": 0.0005449313321150839
+    },
+    {
+      "epoch": 0.03027681660899654,
+      "grad_norm": 2.0,
+      "kd_loss": 0.041785843670368195,
+      "learning_rate": 9.999428913912876e-06,
+      "loss": 0.0561,
+      "step": 105,
+      "student_loss": 0.005221863742917776,
+      "teacher_loss": 0.0004642533021979034
+    },
+    {
+      "epoch": 0.031718569780853516,
+      "grad_norm": 1.90625,
+      "kd_loss": 0.03158900886774063,
+      "learning_rate": 9.998948169522587e-06,
+      "loss": 0.0634,
+      "step": 110,
+      "student_loss": 0.02624746784567833,
+      "teacher_loss": 0.019401099532842636
+    },
+    {
+      "epoch": 0.03316032295271049,
+      "grad_norm": 1.84375,
+      "kd_loss": 0.025746585801243782,
+      "learning_rate": 9.998321768134101e-06,
+      "loss": 0.058,
+      "step": 115,
+      "student_loss": 0.0007584464619867504,
+      "teacher_loss": 0.00038955273339524865
+    },
+    {
+      "epoch": 0.03460207612456748,
+      "grad_norm": 2.5,
+      "kd_loss": 0.017939826473593712,
+      "learning_rate": 9.997549727999212e-06,
+      "loss": 0.0505,
+      "step": 120,
+      "student_loss": 0.0029337499290704727,
+      "teacher_loss": 0.0009877075208351016
+    },
+    {
+      "epoch": 0.036043829296424454,
+      "grad_norm": 2.046875,
+      "kd_loss": 0.0580839179456234,
+      "learning_rate": 9.996632071613257e-06,
+      "loss": 0.0538,
+      "step": 125,
+      "student_loss": 0.042769238352775574,
+      "teacher_loss": 0.0003408969205338508
+    },
+    {
+      "epoch": 0.03748558246828143,
+      "grad_norm": 11.75,
+      "kd_loss": 0.017885997891426086,
+      "learning_rate": 9.995568825714479e-06,
+      "loss": 0.0556,
+      "step": 130,
+      "student_loss": 0.0008774802554398775,
+      "teacher_loss": 0.000655525189358741
+    },
+    {
+      "epoch": 0.03892733564013841,
+      "grad_norm": 1.1875,
+      "kd_loss": 0.032637353986501694,
+      "learning_rate": 9.994360021283235e-06,
+      "loss": 0.054,
+      "step": 135,
+      "student_loss": 0.009718491695821285,
+      "teacher_loss": 0.0008957762038335204
+    },
+    {
+      "epoch": 0.040369088811995385,
+      "grad_norm": 1.015625,
+      "kd_loss": 0.01658937893807888,
+      "learning_rate": 9.993005693541098e-06,
+      "loss": 0.0484,
+      "step": 140,
+      "student_loss": 0.0006278170621953905,
+      "teacher_loss": 0.00041464308742433786
+    },
+    {
+      "epoch": 0.04181084198385236,
+      "grad_norm": 3.09375,
+      "kd_loss": 0.06331603974103928,
+      "learning_rate": 9.991505881949837e-06,
+      "loss": 0.0455,
+      "step": 145,
+      "student_loss": 0.04779814928770065,
+      "teacher_loss": 0.008776809088885784
+    },
+    {
+      "epoch": 0.04325259515570934,
+      "grad_norm": 1.1875,
+      "kd_loss": 0.012085826136171818,
+      "learning_rate": 9.989860630210251e-06,
+      "loss": 0.0447,
+      "step": 150,
+      "student_loss": 0.0007606842555105686,
+      "teacher_loss": 0.0006636636680923402
+    },
+    {
+      "epoch": 0.04469434832756632,
+      "grad_norm": 1.8515625,
+      "kd_loss": 0.015217100270092487,
+      "learning_rate": 9.98806998626091e-06,
+      "loss": 0.0574,
+      "step": 155,
+      "student_loss": 0.0006297255749814212,
+      "teacher_loss": 0.0004516748304013163
+    },
+    {
+      "epoch": 0.0461361014994233,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.021681347861886024,
+      "learning_rate": 9.98613400227676e-06,
+      "loss": 0.0499,
+      "step": 160,
+      "student_loss": 0.08383286744356155,
+      "teacher_loss": 0.0425245501101017
+    },
+    {
+      "epoch": 0.04757785467128028,
+      "grad_norm": 1.0625,
+      "kd_loss": 0.02377716824412346,
+      "learning_rate": 9.984052734667586e-06,
+      "loss": 0.0438,
+      "step": 165,
+      "student_loss": 0.000517472974024713,
+      "teacher_loss": 0.0054916865192353725
+    },
+    {
+      "epoch": 0.049019607843137254,
+      "grad_norm": 1.296875,
+      "kd_loss": 0.012254269793629646,
+      "learning_rate": 9.981826244076387e-06,
+      "loss": 0.045,
+      "step": 170,
+      "student_loss": 0.0007025015656836331,
+      "teacher_loss": 0.0005940330447629094
+    },
+    {
+      "epoch": 0.05046136101499423,
+      "grad_norm": 1.2265625,
+      "kd_loss": 0.022828614339232445,
+      "learning_rate": 9.979454595377594e-06,
+      "loss": 0.0449,
+      "step": 175,
+      "student_loss": 0.002857188694179058,
+      "teacher_loss": 0.00044675302342511714
+    },
+    {
+      "epoch": 0.05190311418685121,
+      "grad_norm": 1.15625,
+      "kd_loss": 0.02279340662062168,
+      "learning_rate": 9.97693785767519e-06,
+      "loss": 0.0416,
+      "step": 180,
+      "student_loss": 0.0009870725916698575,
+      "teacher_loss": 0.0002556022081989795
+    },
+    {
+      "epoch": 0.05334486735870819,
+      "grad_norm": 1.484375,
+      "kd_loss": 0.02873876690864563,
+      "learning_rate": 9.9742761043007e-06,
+      "loss": 0.0472,
+      "step": 185,
+      "student_loss": 0.04467737674713135,
+      "teacher_loss": 0.0002736605820246041
+    },
+    {
+      "epoch": 0.05478662053056517,
+      "grad_norm": 1.5078125,
+      "kd_loss": 0.03800494968891144,
+      "learning_rate": 9.971469412811032e-06,
+      "loss": 0.0432,
+      "step": 190,
+      "student_loss": 0.006019711960107088,
+      "teacher_loss": 0.005745577625930309
+    },
+    {
+      "epoch": 0.056228373702422146,
+      "grad_norm": 2.453125,
+      "kd_loss": 0.03193611651659012,
+      "learning_rate": 9.968517864986247e-06,
+      "loss": 0.044,
+      "step": 195,
+      "student_loss": 0.010869728401303291,
+      "teacher_loss": 0.0007882199133746326
+    },
+    {
+      "epoch": 0.05767012687427912,
+      "grad_norm": 0.703125,
+      "kd_loss": 0.025101784616708755,
+      "learning_rate": 9.965421546827153e-06,
+      "loss": 0.0501,
+      "step": 200,
+      "student_loss": 0.0018625810043886304,
+      "teacher_loss": 0.00027628077077679336
+    },
+    {
+      "epoch": 0.0591118800461361,
+      "grad_norm": 1.359375,
+      "kd_loss": 0.02438431791961193,
+      "learning_rate": 9.962180548552812e-06,
+      "loss": 0.0458,
+      "step": 205,
+      "student_loss": 0.03717027232050896,
+      "teacher_loss": 0.02886415459215641
+    },
+    {
+      "epoch": 0.06055363321799308,
+      "grad_norm": 1.84375,
+      "kd_loss": 0.02243652194738388,
+      "learning_rate": 9.958794964597906e-06,
+      "loss": 0.043,
+      "step": 210,
+      "student_loss": 0.0013577835634350777,
+      "teacher_loss": 0.000932522292714566
+    },
+    {
+      "epoch": 0.061995386389850055,
+      "grad_norm": 1.078125,
+      "kd_loss": 0.028962837532162666,
+      "learning_rate": 9.955264893609986e-06,
+      "loss": 0.0453,
+      "step": 215,
+      "student_loss": 0.001151631586253643,
+      "teacher_loss": 0.0003005610778927803
+    },
+    {
+      "epoch": 0.06343713956170703,
+      "grad_norm": 1.9375,
+      "kd_loss": 0.03575044497847557,
+      "learning_rate": 9.951590438446597e-06,
+      "loss": 0.0453,
+      "step": 220,
+      "student_loss": 0.048845142126083374,
+      "teacher_loss": 0.009562678635120392
+    },
+    {
+      "epoch": 0.06487889273356401,
+      "grad_norm": 1.375,
+      "kd_loss": 0.015835396945476532,
+      "learning_rate": 9.947771706172285e-06,
+      "loss": 0.0506,
+      "step": 225,
+      "student_loss": 0.0546843521296978,
+      "teacher_loss": 0.0008852278115227818
+    },
+    {
+      "epoch": 0.06632064590542099,
+      "grad_norm": 2.125,
+      "kd_loss": 0.02423272468149662,
+      "learning_rate": 9.943808808055472e-06,
+      "loss": 0.0398,
+      "step": 230,
+      "student_loss": 0.0009826251771301031,
+      "teacher_loss": 0.002756665926426649
+    },
+    {
+      "epoch": 0.06776239907727798,
+      "grad_norm": 0.93359375,
+      "kd_loss": 0.04541483893990517,
+      "learning_rate": 9.93970185956522e-06,
+      "loss": 0.0415,
+      "step": 235,
+      "student_loss": 0.014311765320599079,
+      "teacher_loss": 0.0015107382787391543
+    },
+    {
+      "epoch": 0.06920415224913495,
+      "grad_norm": 1.296875,
+      "kd_loss": 0.018931882455945015,
+      "learning_rate": 9.93545098036786e-06,
+      "loss": 0.0432,
+      "step": 240,
+      "student_loss": 0.0005313835572451353,
+      "teacher_loss": 0.00029519537929445505
+    },
+    {
+      "epoch": 0.07064590542099193,
+      "grad_norm": 1.2890625,
+      "kd_loss": 0.012201114557683468,
+      "learning_rate": 9.931056294323507e-06,
+      "loss": 0.0392,
+      "step": 245,
+      "student_loss": 0.0008615658152848482,
+      "teacher_loss": 0.0008678268059156835
+    },
+    {
+      "epoch": 0.07208765859284891,
+      "grad_norm": 1.9296875,
+      "kd_loss": 0.013639300130307674,
+      "learning_rate": 9.926517929482454e-06,
+      "loss": 0.0416,
+      "step": 250,
+      "student_loss": 0.000850351236294955,
+      "teacher_loss": 0.0004967341665178537
+    },
+    {
+      "epoch": 0.07352941176470588,
+      "grad_norm": 0.82421875,
+      "kd_loss": 0.010670178569853306,
+      "learning_rate": 9.921836018081438e-06,
+      "loss": 0.0421,
+      "step": 255,
+      "student_loss": 0.0005453266203403473,
+      "teacher_loss": 0.0004909925046376884
+    },
+    {
+      "epoch": 0.07497116493656286,
+      "grad_norm": 1.375,
+      "kd_loss": 0.019146086648106575,
+      "learning_rate": 9.917010696539793e-06,
+      "loss": 0.0448,
+      "step": 260,
+      "student_loss": 0.040390945971012115,
+      "teacher_loss": 0.033758774399757385
+    },
+    {
+      "epoch": 0.07641291810841984,
+      "grad_norm": 1.78125,
+      "kd_loss": 0.011872848495841026,
+      "learning_rate": 9.912042105455462e-06,
+      "loss": 0.0422,
+      "step": 265,
+      "student_loss": 0.0005196996498852968,
+      "teacher_loss": 0.00042158958967775106
+    },
+    {
+      "epoch": 0.07785467128027682,
+      "grad_norm": 1.28125,
+      "kd_loss": 0.012035544961690903,
+      "learning_rate": 9.906930389600917e-06,
+      "loss": 0.0458,
+      "step": 270,
+      "student_loss": 0.0009468707139603794,
+      "teacher_loss": 0.000693841720931232
+    },
+    {
+      "epoch": 0.07929642445213379,
+      "grad_norm": 1.7890625,
+      "kd_loss": 0.01897946372628212,
+      "learning_rate": 9.901675697918925e-06,
+      "loss": 0.0432,
+      "step": 275,
+      "student_loss": 0.047525323927402496,
+      "teacher_loss": 0.03367537260055542
+    },
+    {
+      "epoch": 0.08073817762399077,
+      "grad_norm": 1.5859375,
+      "kd_loss": 0.021627889946103096,
+      "learning_rate": 9.896278183518216e-06,
+      "loss": 0.041,
+      "step": 280,
+      "student_loss": 0.0006324543501250446,
+      "teacher_loss": 0.0002107525651808828
+    },
+    {
+      "epoch": 0.08217993079584775,
+      "grad_norm": 1.46875,
+      "kd_loss": 0.013249333947896957,
+      "learning_rate": 9.890738003669029e-06,
+      "loss": 0.0444,
+      "step": 285,
+      "student_loss": 0.0005354061140678823,
+      "teacher_loss": 0.0004488772538024932
+    },
+    {
+      "epoch": 0.08362168396770472,
+      "grad_norm": 1.3359375,
+      "kd_loss": 0.011662454344332218,
+      "learning_rate": 9.885055319798514e-06,
+      "loss": 0.0476,
+      "step": 290,
+      "student_loss": 0.0025434105191379786,
+      "teacher_loss": 0.0015740139642730355
+    },
+    {
+      "epoch": 0.0850634371395617,
+      "grad_norm": 1.828125,
+      "kd_loss": 0.01183430477976799,
+      "learning_rate": 9.879230297486034e-06,
+      "loss": 0.0425,
+      "step": 295,
+      "student_loss": 0.0004690758651122451,
+      "teacher_loss": 0.00038755135028623044
+    },
+    {
+      "epoch": 0.08650519031141868,
+      "grad_norm": 1.03125,
+      "kd_loss": 0.03465069457888603,
+      "learning_rate": 9.873263106458353e-06,
+      "loss": 0.0394,
+      "step": 300,
+      "student_loss": 0.018424537032842636,
+      "teacher_loss": 0.002826885087415576
+    },
+    {
+      "epoch": 0.08794694348327567,
+      "grad_norm": 1.125,
+      "kd_loss": 0.013110927306115627,
+      "learning_rate": 9.86715392058467e-06,
+      "loss": 0.0357,
+      "step": 305,
+      "student_loss": 0.01486227661371231,
+      "teacher_loss": 0.0012473311508074403
+    },
+    {
+      "epoch": 0.08938869665513265,
+      "grad_norm": 1.4765625,
+      "kd_loss": 0.010091043077409267,
+      "learning_rate": 9.860902917871566e-06,
+      "loss": 0.039,
+      "step": 310,
+      "student_loss": 0.00043867999920621514,
+      "teacher_loss": 0.00034055308788083494
+    },
+    {
+      "epoch": 0.09083044982698962,
+      "grad_norm": 1.0625,
+      "kd_loss": 0.02157684415578842,
+      "learning_rate": 9.854510280457819e-06,
+      "loss": 0.0391,
+      "step": 315,
+      "student_loss": 0.014208661392331123,
+      "teacher_loss": 0.001196117140352726
+    },
+    {
+      "epoch": 0.0922722029988466,
+      "grad_norm": 1.1953125,
+      "kd_loss": 0.0127296457067132,
+      "learning_rate": 9.847976194609083e-06,
+      "loss": 0.0432,
+      "step": 320,
+      "student_loss": 0.0004476152826100588,
+      "teacher_loss": 0.000240055174799636
+    },
+    {
+      "epoch": 0.09371395617070358,
+      "grad_norm": 1.5546875,
+      "kd_loss": 0.032895129173994064,
+      "learning_rate": 9.841300850712479e-06,
+      "loss": 0.0377,
+      "step": 325,
+      "student_loss": 0.008438433520495892,
+      "teacher_loss": 0.015148998238146305
+    },
+    {
+      "epoch": 0.09515570934256055,
+      "grad_norm": 1.3046875,
+      "kd_loss": 0.04177394509315491,
+      "learning_rate": 9.834484443271035e-06,
+      "loss": 0.0389,
+      "step": 330,
+      "student_loss": 0.01764899119734764,
+      "teacher_loss": 0.001141766319051385
+    },
+    {
+      "epoch": 0.09659746251441753,
+      "grad_norm": 0.9296875,
+      "kd_loss": 0.008085020817816257,
+      "learning_rate": 9.827527170898019e-06,
+      "loss": 0.0328,
+      "step": 335,
+      "student_loss": 0.0030857992824167013,
+      "teacher_loss": 0.0029651003424078226
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 1.5625,
+      "kd_loss": 0.014680181629955769,
+      "learning_rate": 9.820429236311158e-06,
+      "loss": 0.0397,
+      "step": 340,
+      "student_loss": 0.010333855636417866,
+      "teacher_loss": 0.004541066940873861
+    },
+    {
+      "epoch": 0.09948096885813149,
+      "grad_norm": 0.9921875,
+      "kd_loss": 0.01903965324163437,
+      "learning_rate": 9.81319084632673e-06,
+      "loss": 0.043,
+      "step": 345,
+      "student_loss": 0.012247292324900627,
+      "teacher_loss": 0.00043584281229414046
+    },
+    {
+      "epoch": 0.10092272202998846,
+      "grad_norm": 1.1171875,
+      "kd_loss": 0.037925247102975845,
+      "learning_rate": 9.805812211853534e-06,
+      "loss": 0.034,
+      "step": 350,
+      "student_loss": 0.0080111650750041,
+      "teacher_loss": 0.0011970563791692257
+    },
+    {
+      "epoch": 0.10236447520184544,
+      "grad_norm": 1.4296875,
+      "kd_loss": 0.00829800870269537,
+      "learning_rate": 9.798293547886748e-06,
+      "loss": 0.0396,
+      "step": 355,
+      "student_loss": 0.0005787648260593414,
+      "teacher_loss": 0.0005460750544443727
+    },
+    {
+      "epoch": 0.10380622837370242,
+      "grad_norm": 1.03125,
+      "kd_loss": 0.008213132619857788,
+      "learning_rate": 9.790635073501662e-06,
+      "loss": 0.0393,
+      "step": 360,
+      "student_loss": 0.00047661116695962846,
+      "teacher_loss": 0.00041378207970410585
+    },
+    {
+      "epoch": 0.1052479815455594,
+      "grad_norm": 1.53125,
+      "kd_loss": 0.014927592128515244,
+      "learning_rate": 9.782837011847294e-06,
+      "loss": 0.0436,
+      "step": 365,
+      "student_loss": 0.043972667306661606,
+      "teacher_loss": 0.02826174721121788
+    },
+    {
+      "epoch": 0.10668973471741638,
+      "grad_norm": 1.9375,
+      "kd_loss": 0.03636546432971954,
+      "learning_rate": 9.774899590139897e-06,
+      "loss": 0.0423,
+      "step": 370,
+      "student_loss": 0.021089306101202965,
+      "teacher_loss": 0.0004860333865508437
+    },
+    {
+      "epoch": 0.10813148788927336,
+      "grad_norm": 1.140625,
+      "kd_loss": 0.015318617224693298,
+      "learning_rate": 9.766823039656328e-06,
+      "loss": 0.0357,
+      "step": 375,
+      "student_loss": 0.03581877425312996,
+      "teacher_loss": 0.02248113416135311
+    },
+    {
+      "epoch": 0.10957324106113034,
+      "grad_norm": 0.74609375,
+      "kd_loss": 0.009901116602122784,
+      "learning_rate": 9.758607595727309e-06,
+      "loss": 0.0372,
+      "step": 380,
+      "student_loss": 0.00044091136078350246,
+      "teacher_loss": 0.0003208389098290354
+    },
+    {
+      "epoch": 0.11101499423298732,
+      "grad_norm": 1.5390625,
+      "kd_loss": 0.015398882329463959,
+      "learning_rate": 9.75025349773058e-06,
+      "loss": 0.0348,
+      "step": 385,
+      "student_loss": 0.0006440416909754276,
+      "teacher_loss": 0.0005293422727845609
+    },
+    {
+      "epoch": 0.11245674740484429,
+      "grad_norm": 1.1171875,
+      "kd_loss": 0.009845156222581863,
+      "learning_rate": 9.741760989083914e-06,
+      "loss": 0.0326,
+      "step": 390,
+      "student_loss": 0.00028396621928550303,
+      "teacher_loss": 0.00030337049975059927
+    },
+    {
+      "epoch": 0.11389850057670127,
+      "grad_norm": 1.28125,
+      "kd_loss": 0.011232681572437286,
+      "learning_rate": 9.73313031723803e-06,
+      "loss": 0.038,
+      "step": 395,
+      "student_loss": 0.003544658189639449,
+      "teacher_loss": 0.0032308201771229506
+    },
+    {
+      "epoch": 0.11534025374855825,
+      "grad_norm": 1.609375,
+      "kd_loss": 0.01615743711590767,
+      "learning_rate": 9.724361733669383e-06,
+      "loss": 0.0357,
+      "step": 400,
+      "student_loss": 0.0006839327397756279,
+      "teacher_loss": 0.000942843733355403
+    },
+    {
+      "epoch": 0.11678200692041522,
+      "grad_norm": 1.078125,
+      "kd_loss": 0.013564227148890495,
+      "learning_rate": 9.715455493872827e-06,
+      "loss": 0.0415,
+      "step": 405,
+      "student_loss": 0.005195911042392254,
+      "teacher_loss": 0.00030810022144578397
+    },
+    {
+      "epoch": 0.1182237600922722,
+      "grad_norm": 1.5703125,
+      "kd_loss": 0.025074705481529236,
+      "learning_rate": 9.70641185735419e-06,
+      "loss": 0.0411,
+      "step": 410,
+      "student_loss": 0.01386551558971405,
+      "teacher_loss": 0.00967031717300415
+    },
+    {
+      "epoch": 0.11966551326412918,
+      "grad_norm": 1.1875,
+      "kd_loss": 0.015891144052147865,
+      "learning_rate": 9.697231087622691e-06,
+      "loss": 0.0411,
+      "step": 415,
+      "student_loss": 0.0021395094227045774,
+      "teacher_loss": 0.0021108894143253565
+    },
+    {
+      "epoch": 0.12110726643598616,
+      "grad_norm": 1.40625,
+      "kd_loss": 0.021059690043330193,
+      "learning_rate": 9.687913452183277e-06,
+      "loss": 0.0379,
+      "step": 420,
+      "student_loss": 0.04692540317773819,
+      "teacher_loss": 0.02177572436630726
+    },
+    {
+      "epoch": 0.12254901960784313,
+      "grad_norm": 1.3125,
+      "kd_loss": 0.007998880930244923,
+      "learning_rate": 9.678459222528825e-06,
+      "loss": 0.0366,
+      "step": 425,
+      "student_loss": 0.0025826317723840475,
+      "teacher_loss": 0.002152319997549057
+    },
+    {
+      "epoch": 0.12399077277970011,
+      "grad_norm": 1.5390625,
+      "kd_loss": 0.026353655382990837,
+      "learning_rate": 9.668868674132224e-06,
+      "loss": 0.0448,
+      "step": 430,
+      "student_loss": 0.04476851597428322,
+      "teacher_loss": 0.06174921244382858
+    },
+    {
+      "epoch": 0.1254325259515571,
+      "grad_norm": 1.3359375,
+      "kd_loss": 0.008578015491366386,
+      "learning_rate": 9.65914208643836e-06,
+      "loss": 0.0355,
+      "step": 435,
+      "student_loss": 0.0006695028278045356,
+      "teacher_loss": 0.0005997278494760394
+    },
+    {
+      "epoch": 0.12687427912341406,
+      "grad_norm": 1.1640625,
+      "kd_loss": 0.025591254234313965,
+      "learning_rate": 9.649279742855964e-06,
+      "loss": 0.039,
+      "step": 440,
+      "student_loss": 0.020308705046772957,
+      "teacher_loss": 0.005524828098714352
+    },
+    {
+      "epoch": 0.12831603229527105,
+      "grad_norm": 1.875,
+      "kd_loss": 0.019290607422590256,
+      "learning_rate": 9.639281930749363e-06,
+      "loss": 0.0384,
+      "step": 445,
+      "student_loss": 0.0254460871219635,
+      "teacher_loss": 0.01746205799281597
+    },
+    {
+      "epoch": 0.12975778546712802,
+      "grad_norm": 1.3046875,
+      "kd_loss": 0.01817379891872406,
+      "learning_rate": 9.629148941430096e-06,
+      "loss": 0.0398,
+      "step": 450,
+      "student_loss": 0.0006312711630016565,
+      "teacher_loss": 0.0004661971179302782
+    },
+    {
+      "epoch": 0.131199538638985,
+      "grad_norm": 1.390625,
+      "kd_loss": 0.01092479843646288,
+      "learning_rate": 9.618881070148436e-06,
+      "loss": 0.039,
+      "step": 455,
+      "student_loss": 0.005430892109870911,
+      "teacher_loss": 0.0015254098689183593
+    },
+    {
+      "epoch": 0.13264129181084197,
+      "grad_norm": 1.6484375,
+      "kd_loss": 0.014401745982468128,
+      "learning_rate": 9.608478616084784e-06,
+      "loss": 0.0329,
+      "step": 460,
+      "student_loss": 0.002872939221560955,
+      "teacher_loss": 0.0004281107976567
+    },
+    {
+      "epoch": 0.13408304498269896,
+      "grad_norm": 1.25,
+      "kd_loss": 0.02247895859181881,
+      "learning_rate": 9.597941882340947e-06,
+      "loss": 0.0409,
+      "step": 465,
+      "student_loss": 0.00037059103487990797,
+      "teacher_loss": 0.0019852325785905123
+    },
+    {
+      "epoch": 0.13552479815455595,
+      "grad_norm": 1.2890625,
+      "kd_loss": 0.011446633376181126,
+      "learning_rate": 9.587271175931313e-06,
+      "loss": 0.0372,
+      "step": 470,
+      "student_loss": 0.00029464755789376795,
+      "teacher_loss": 0.00020733379642479122
+    },
+    {
+      "epoch": 0.13696655132641292,
+      "grad_norm": 2.21875,
+      "kd_loss": 0.016915325075387955,
+      "learning_rate": 9.5764668077739e-06,
+      "loss": 0.0312,
+      "step": 475,
+      "student_loss": 0.0005512129282578826,
+      "teacher_loss": 0.00031623090035282075
+    },
+    {
+      "epoch": 0.1384083044982699,
+      "grad_norm": 1.2890625,
+      "kd_loss": 0.016255535185337067,
+      "learning_rate": 9.5655290926813e-06,
+      "loss": 0.0457,
+      "step": 480,
+      "student_loss": 0.015682687982916832,
+      "teacher_loss": 0.013726728036999702
+    },
+    {
+      "epoch": 0.13985005767012687,
+      "grad_norm": 1.109375,
+      "kd_loss": 0.007256851997226477,
+      "learning_rate": 9.554458349351507e-06,
+      "loss": 0.0341,
+      "step": 485,
+      "student_loss": 0.0005507866735570133,
+      "teacher_loss": 0.0005500169354490936
+    },
+    {
+      "epoch": 0.14129181084198386,
+      "grad_norm": 1.9765625,
+      "kd_loss": 0.018985357135534286,
+      "learning_rate": 9.54325490035863e-06,
+      "loss": 0.0445,
+      "step": 490,
+      "student_loss": 0.015820659697055817,
+      "teacher_loss": 0.00336285843513906
+    },
+    {
+      "epoch": 0.14273356401384082,
+      "grad_norm": 1.875,
+      "kd_loss": 0.025550341233611107,
+      "learning_rate": 9.531919072143488e-06,
+      "loss": 0.0483,
+      "step": 495,
+      "student_loss": 0.026045799255371094,
+      "teacher_loss": 0.004566465504467487
+    },
+    {
+      "epoch": 0.14417531718569782,
+      "grad_norm": 1.640625,
+      "kd_loss": 0.017328884452581406,
+      "learning_rate": 9.520451195004107e-06,
+      "loss": 0.0369,
+      "step": 500,
+      "student_loss": 0.0017473739571869373,
+      "teacher_loss": 0.000323916639899835
+    },
+    {
+      "epoch": 0.14561707035755478,
+      "grad_norm": 1.046875,
+      "kd_loss": 0.011184470728039742,
+      "learning_rate": 9.508851603086094e-06,
+      "loss": 0.0436,
+      "step": 505,
+      "student_loss": 0.0003846609324682504,
+      "teacher_loss": 0.0002943307626992464
+    },
+    {
+      "epoch": 0.14705882352941177,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.011295544914901257,
+      "learning_rate": 9.497120634372894e-06,
+      "loss": 0.0376,
+      "step": 510,
+      "student_loss": 0.00031946305534802377,
+      "teacher_loss": 0.0002625001943670213
+    },
+    {
+      "epoch": 0.14850057670126873,
+      "grad_norm": 1.8984375,
+      "kd_loss": 0.01340380497276783,
+      "learning_rate": 9.48525863067595e-06,
+      "loss": 0.0331,
+      "step": 515,
+      "student_loss": 0.007424864452332258,
+      "teacher_loss": 0.00023525822325609624
+    },
+    {
+      "epoch": 0.14994232987312572,
+      "grad_norm": 2.875,
+      "kd_loss": 0.032959483563899994,
+      "learning_rate": 9.473265937624748e-06,
+      "loss": 0.0408,
+      "step": 520,
+      "student_loss": 0.04309098795056343,
+      "teacher_loss": 0.011560679413378239
+    },
+    {
+      "epoch": 0.1513840830449827,
+      "grad_norm": 0.8828125,
+      "kd_loss": 0.01784936524927616,
+      "learning_rate": 9.461142904656722e-06,
+      "loss": 0.0398,
+      "step": 525,
+      "student_loss": 0.0003528800734784454,
+      "teacher_loss": 0.0001730480435071513
+    },
+    {
+      "epoch": 0.15282583621683968,
+      "grad_norm": 0.8515625,
+      "kd_loss": 0.01350821927189827,
+      "learning_rate": 9.448889885007105e-06,
+      "loss": 0.0353,
+      "step": 530,
+      "student_loss": 0.00033670777338556945,
+      "teacher_loss": 0.0002543032169342041
+    },
+    {
+      "epoch": 0.15426758938869667,
+      "grad_norm": 0.890625,
+      "kd_loss": 0.027389032766222954,
+      "learning_rate": 9.436507235698613e-06,
+      "loss": 0.0344,
+      "step": 535,
+      "student_loss": 0.0007846449734643102,
+      "teacher_loss": 0.0004379330202937126
+    },
+    {
+      "epoch": 0.15570934256055363,
+      "grad_norm": 1.734375,
+      "kd_loss": 0.014288514852523804,
+      "learning_rate": 9.423995317531049e-06,
+      "loss": 0.0341,
+      "step": 540,
+      "student_loss": 0.0010072099976241589,
+      "teacher_loss": 0.0003058096335735172
+    },
+    {
+      "epoch": 0.15715109573241062,
+      "grad_norm": 1.796875,
+      "kd_loss": 0.008625246584415436,
+      "learning_rate": 9.4113544950708e-06,
+      "loss": 0.0308,
+      "step": 545,
+      "student_loss": 0.010159535333514214,
+      "teacher_loss": 0.007676934357732534
+    },
+    {
+      "epoch": 0.15859284890426759,
+      "grad_norm": 1.9296875,
+      "kd_loss": 0.02502039261162281,
+      "learning_rate": 9.398585136640195e-06,
+      "loss": 0.0307,
+      "step": 550,
+      "student_loss": 0.00014985659800004214,
+      "teacher_loss": 0.014202730730175972
+    },
+    {
+      "epoch": 0.16003460207612458,
+      "grad_norm": 1.09375,
+      "kd_loss": 0.042002685368061066,
+      "learning_rate": 9.38568761430679e-06,
+      "loss": 0.0389,
+      "step": 555,
+      "student_loss": 0.02874353528022766,
+      "teacher_loss": 0.003934269305318594
+    },
+    {
+      "epoch": 0.16147635524798154,
+      "grad_norm": 1.6796875,
+      "kd_loss": 0.03426447883248329,
+      "learning_rate": 9.372662303872514e-06,
+      "loss": 0.0405,
+      "step": 560,
+      "student_loss": 0.03105512261390686,
+      "teacher_loss": 0.0003118969907518476
+    },
+    {
+      "epoch": 0.16291810841983853,
+      "grad_norm": 1.2109375,
+      "kd_loss": 0.01886710338294506,
+      "learning_rate": 9.359509584862735e-06,
+      "loss": 0.0345,
+      "step": 565,
+      "student_loss": 0.03195136412978172,
+      "teacher_loss": 0.021795904263854027
+    },
+    {
+      "epoch": 0.1643598615916955,
+      "grad_norm": 1.484375,
+      "kd_loss": 0.021855715662240982,
+      "learning_rate": 9.346229840515186e-06,
+      "loss": 0.0418,
+      "step": 570,
+      "student_loss": 0.011120815761387348,
+      "teacher_loss": 0.007288818247616291
+    },
+    {
+      "epoch": 0.16580161476355249,
+      "grad_norm": 1.734375,
+      "kd_loss": 0.013769465498626232,
+      "learning_rate": 9.332823457768805e-06,
+      "loss": 0.0386,
+      "step": 575,
+      "student_loss": 0.0005883977864868939,
+      "teacher_loss": 0.0003054222615901381
+    },
+    {
+      "epoch": 0.16724336793540945,
+      "grad_norm": 1.0703125,
+      "kd_loss": 0.012690847739577293,
+      "learning_rate": 9.31929082725246e-06,
+      "loss": 0.0337,
+      "step": 580,
+      "student_loss": 0.0022747719194740057,
+      "teacher_loss": 0.00218041124753654
+    },
+    {
+      "epoch": 0.16868512110726644,
+      "grad_norm": 0.85546875,
+      "kd_loss": 0.017418578267097473,
+      "learning_rate": 9.305632343273567e-06,
+      "loss": 0.0316,
+      "step": 585,
+      "student_loss": 0.0005172074306756258,
+      "teacher_loss": 0.0002657689619809389
+    },
+    {
+      "epoch": 0.1701268742791234,
+      "grad_norm": 1.1640625,
+      "kd_loss": 0.013570615090429783,
+      "learning_rate": 9.291848403806604e-06,
+      "loss": 0.037,
+      "step": 590,
+      "student_loss": 0.0005453322664834559,
+      "teacher_loss": 0.0003946890647057444
+    },
+    {
+      "epoch": 0.1715686274509804,
+      "grad_norm": 1.265625,
+      "kd_loss": 0.010840360075235367,
+      "learning_rate": 9.277939410481507e-06,
+      "loss": 0.0353,
+      "step": 595,
+      "student_loss": 0.00048185884952545166,
+      "teacher_loss": 0.00042216762085445225
+    },
+    {
+      "epoch": 0.17301038062283736,
+      "grad_norm": 1.390625,
+      "kd_loss": 0.02166227251291275,
+      "learning_rate": 9.263905768571977e-06,
+      "loss": 0.0417,
+      "step": 600,
+      "student_loss": 0.05149306356906891,
+      "teacher_loss": 0.009222068823873997
+    },
+    {
+      "epoch": 0.17445213379469435,
+      "grad_norm": 1.109375,
+      "kd_loss": 0.021928608417510986,
+      "learning_rate": 9.249747886983664e-06,
+      "loss": 0.0387,
+      "step": 605,
+      "student_loss": 0.06224099546670914,
+      "teacher_loss": 0.03538208454847336
+    },
+    {
+      "epoch": 0.17589388696655134,
+      "grad_norm": 1.359375,
+      "kd_loss": 0.0258643738925457,
+      "learning_rate": 9.235466178242255e-06,
+      "loss": 0.0353,
+      "step": 610,
+      "student_loss": 0.002640328835695982,
+      "teacher_loss": 0.0018086276249960065
+    },
+    {
+      "epoch": 0.1773356401384083,
+      "grad_norm": 0.9453125,
+      "kd_loss": 0.013348888605833054,
+      "learning_rate": 9.221061058481456e-06,
+      "loss": 0.0322,
+      "step": 615,
+      "student_loss": 0.000250897224759683,
+      "teacher_loss": 0.00027990140370093286
+    },
+    {
+      "epoch": 0.1787773933102653,
+      "grad_norm": 0.6875,
+      "kd_loss": 0.019072921946644783,
+      "learning_rate": 9.206532947430865e-06,
+      "loss": 0.0416,
+      "step": 620,
+      "student_loss": 0.010811248794198036,
+      "teacher_loss": 0.007016193587332964
+    },
+    {
+      "epoch": 0.18021914648212226,
+      "grad_norm": 0.98046875,
+      "kd_loss": 0.014861532486975193,
+      "learning_rate": 9.191882268403743e-06,
+      "loss": 0.0335,
+      "step": 625,
+      "student_loss": 0.002400943310931325,
+      "teacher_loss": 0.0028365827165544033
+    },
+    {
+      "epoch": 0.18166089965397925,
+      "grad_norm": 2.359375,
+      "kd_loss": 0.02572929672896862,
+      "learning_rate": 9.177109448284673e-06,
+      "loss": 0.0401,
+      "step": 630,
+      "student_loss": 0.028089625760912895,
+      "teacher_loss": 0.012286370620131493
+    },
+    {
+      "epoch": 0.1831026528258362,
+      "grad_norm": 1.203125,
+      "kd_loss": 0.01528047863394022,
+      "learning_rate": 9.162214917517139e-06,
+      "loss": 0.0528,
+      "step": 635,
+      "student_loss": 0.002457874594256282,
+      "teacher_loss": 0.0002723166544456035
+    },
+    {
+      "epoch": 0.1845444059976932,
+      "grad_norm": 1.6640625,
+      "kd_loss": 0.016974100843071938,
+      "learning_rate": 9.14719911009096e-06,
+      "loss": 0.0327,
+      "step": 640,
+      "student_loss": 0.0006965341162867844,
+      "teacher_loss": 0.0008102107094600797
+    },
+    {
+      "epoch": 0.18598615916955016,
+      "grad_norm": 1.9609375,
+      "kd_loss": 0.01875746250152588,
+      "learning_rate": 9.132062463529666e-06,
+      "loss": 0.033,
+      "step": 645,
+      "student_loss": 0.0016925447853282094,
+      "teacher_loss": 0.005389066878706217
+    },
+    {
+      "epoch": 0.18742791234140715,
+      "grad_norm": 1.578125,
+      "kd_loss": 0.009372042492032051,
+      "learning_rate": 9.11680541887774e-06,
+      "loss": 0.0324,
+      "step": 650,
+      "student_loss": 0.0010497953044250607,
+      "teacher_loss": 0.0008989371126517653
+    },
+    {
+      "epoch": 0.18886966551326412,
+      "grad_norm": 3.9375,
+      "kd_loss": 0.019143380224704742,
+      "learning_rate": 9.101428420687759e-06,
+      "loss": 0.0368,
+      "step": 655,
+      "student_loss": 0.0008799026254564524,
+      "teacher_loss": 0.00043919464224018157
+    },
+    {
+      "epoch": 0.1903114186851211,
+      "grad_norm": 1.921875,
+      "kd_loss": 0.013875705190002918,
+      "learning_rate": 9.085931917007464e-06,
+      "loss": 0.0405,
+      "step": 660,
+      "student_loss": 0.000673408736474812,
+      "teacher_loss": 0.0005781051004305482
+    },
+    {
+      "epoch": 0.19175317185697807,
+      "grad_norm": 1.6015625,
+      "kd_loss": 0.043592095375061035,
+      "learning_rate": 9.070316359366681e-06,
+      "loss": 0.0413,
+      "step": 665,
+      "student_loss": 0.0031369002535939217,
+      "teacher_loss": 0.001360239228233695
+    },
+    {
+      "epoch": 0.19319492502883506,
+      "grad_norm": 8.5,
+      "kd_loss": 0.01952010951936245,
+      "learning_rate": 9.054582202764175e-06,
+      "loss": 0.042,
+      "step": 670,
+      "student_loss": 0.012506276369094849,
+      "teacher_loss": 0.0028849972877651453
+    },
+    {
+      "epoch": 0.19463667820069205,
+      "grad_norm": 0.9609375,
+      "kd_loss": 0.008069864474236965,
+      "learning_rate": 9.038729905654393e-06,
+      "loss": 0.0365,
+      "step": 675,
+      "student_loss": 0.0004877475730609149,
+      "teacher_loss": 0.0004426161467563361
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 1.59375,
+      "kd_loss": 0.021699784323573112,
+      "learning_rate": 9.022759929934109e-06,
+      "loss": 0.0317,
+      "step": 680,
+      "student_loss": 0.07352294772863388,
+      "teacher_loss": 0.033334095031023026
+    },
+    {
+      "epoch": 0.197520184544406,
+      "grad_norm": 1.125,
+      "kd_loss": 0.014409462921321392,
+      "learning_rate": 9.006672740928952e-06,
+      "loss": 0.0353,
+      "step": 685,
+      "student_loss": 0.03678691014647484,
+      "teacher_loss": 0.02936466597020626
+    },
+    {
+      "epoch": 0.19896193771626297,
+      "grad_norm": 2.140625,
+      "kd_loss": 0.034733690321445465,
+      "learning_rate": 8.990468807379861e-06,
+      "loss": 0.0368,
+      "step": 690,
+      "student_loss": 0.0369374081492424,
+      "teacher_loss": 0.008476306684315205
+    },
+    {
+      "epoch": 0.20040369088811996,
+      "grad_norm": 1.453125,
+      "kd_loss": 0.016646580770611763,
+      "learning_rate": 8.974148601429426e-06,
+      "loss": 0.0392,
+      "step": 695,
+      "student_loss": 0.007193500176072121,
+      "teacher_loss": 0.005883329547941685
+    },
+    {
+      "epoch": 0.20184544405997693,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.01161164604127407,
+      "learning_rate": 8.957712598608123e-06,
+      "loss": 0.0341,
+      "step": 700,
+      "student_loss": 0.0006814897060394287,
+      "teacher_loss": 0.00038474920438602567
+    },
+    {
+      "epoch": 0.20328719723183392,
+      "grad_norm": 1.1015625,
+      "kd_loss": 0.008321454748511314,
+      "learning_rate": 8.941161277820462e-06,
+      "loss": 0.0348,
+      "step": 705,
+      "student_loss": 0.0013638287782669067,
+      "teacher_loss": 0.0012245745165273547
+    },
+    {
+      "epoch": 0.20472895040369088,
+      "grad_norm": 1.28125,
+      "kd_loss": 0.036699049174785614,
+      "learning_rate": 8.924495121331038e-06,
+      "loss": 0.0377,
+      "step": 710,
+      "student_loss": 0.011128552258014679,
+      "teacher_loss": 0.0038417454343289137
+    },
+    {
+      "epoch": 0.20617070357554787,
+      "grad_norm": 1.984375,
+      "kd_loss": 0.01310927327722311,
+      "learning_rate": 8.907714614750473e-06,
+      "loss": 0.0459,
+      "step": 715,
+      "student_loss": 0.003388170152902603,
+      "teacher_loss": 0.00937573704868555
+    },
+    {
+      "epoch": 0.20761245674740483,
+      "grad_norm": 1.640625,
+      "kd_loss": 0.016016153618693352,
+      "learning_rate": 8.890820247021263e-06,
+      "loss": 0.0374,
+      "step": 720,
+      "student_loss": 0.00045992882223799825,
+      "teacher_loss": 0.0004658737452700734
+    },
+    {
+      "epoch": 0.20905420991926182,
+      "grad_norm": 1.1640625,
+      "kd_loss": 0.009306555613875389,
+      "learning_rate": 8.873812510403545e-06,
+      "loss": 0.029,
+      "step": 725,
+      "student_loss": 0.0005832040915265679,
+      "teacher_loss": 0.0004866627568844706
+    },
+    {
+      "epoch": 0.2104959630911188,
+      "grad_norm": 0.953125,
+      "kd_loss": 0.019967591390013695,
+      "learning_rate": 8.85669190046074e-06,
+      "loss": 0.034,
+      "step": 730,
+      "student_loss": 0.007923947647213936,
+      "teacher_loss": 0.005159601103514433
+    },
+    {
+      "epoch": 0.21193771626297578,
+      "grad_norm": 1.453125,
+      "kd_loss": 0.014439092949032784,
+      "learning_rate": 8.839458916045117e-06,
+      "loss": 0.041,
+      "step": 735,
+      "student_loss": 0.00028902897611260414,
+      "teacher_loss": 0.00017782236682251096
+    },
+    {
+      "epoch": 0.21337946943483277,
+      "grad_norm": 1.7890625,
+      "kd_loss": 0.02189582586288452,
+      "learning_rate": 8.822114059283267e-06,
+      "loss": 0.0323,
+      "step": 740,
+      "student_loss": 0.0005453451885841787,
+      "teacher_loss": 0.00017786624084692448
+    },
+    {
+      "epoch": 0.21482122260668973,
+      "grad_norm": 1.53125,
+      "kd_loss": 0.013907963410019875,
+      "learning_rate": 8.804657835561456e-06,
+      "loss": 0.0349,
+      "step": 745,
+      "student_loss": 0.038910508155822754,
+      "teacher_loss": 0.020556708797812462
+    },
+    {
+      "epoch": 0.21626297577854672,
+      "grad_norm": 1.3046875,
+      "kd_loss": 0.03312985599040985,
+      "learning_rate": 8.787090753510917e-06,
+      "loss": 0.0353,
+      "step": 750,
+      "student_loss": 0.01961570419371128,
+      "teacher_loss": 0.005801274906843901
+    },
+    {
+      "epoch": 0.2177047289504037,
+      "grad_norm": 2.5,
+      "kd_loss": 0.01383906975388527,
+      "learning_rate": 8.76941332499301e-06,
+      "loss": 0.0397,
+      "step": 755,
+      "student_loss": 0.027118289843201637,
+      "teacher_loss": 0.001979502383619547
+    },
+    {
+      "epoch": 0.21914648212226068,
+      "grad_norm": 1.1484375,
+      "kd_loss": 0.026841050013899803,
+      "learning_rate": 8.751626065084328e-06,
+      "loss": 0.0304,
+      "step": 760,
+      "student_loss": 0.006133708171546459,
+      "teacher_loss": 0.0004118960350751877
+    },
+    {
+      "epoch": 0.22058823529411764,
+      "grad_norm": 0.984375,
+      "kd_loss": 0.028871504589915276,
+      "learning_rate": 8.733729492061675e-06,
+      "loss": 0.0377,
+      "step": 765,
+      "student_loss": 0.005369402468204498,
+      "teacher_loss": 0.0012096863938495517
+    },
+    {
+      "epoch": 0.22202998846597463,
+      "grad_norm": 0.921875,
+      "kd_loss": 0.028200991451740265,
+      "learning_rate": 8.715724127386971e-06,
+      "loss": 0.0337,
+      "step": 770,
+      "student_loss": 0.007010896224528551,
+      "teacher_loss": 0.001293784473091364
+    },
+    {
+      "epoch": 0.2234717416378316,
+      "grad_norm": 1.09375,
+      "kd_loss": 0.015337961725890636,
+      "learning_rate": 8.697610495692055e-06,
+      "loss": 0.0314,
+      "step": 775,
+      "student_loss": 0.007272190414369106,
+      "teacher_loss": 0.002003727713599801
+    },
+    {
+      "epoch": 0.22491349480968859,
+      "grad_norm": 1.140625,
+      "kd_loss": 0.014271454885601997,
+      "learning_rate": 8.679389124763396e-06,
+      "loss": 0.0357,
+      "step": 780,
+      "student_loss": 0.0013534394092857838,
+      "teacher_loss": 0.0002126321487594396
+    },
+    {
+      "epoch": 0.22635524798154555,
+      "grad_norm": 2.328125,
+      "kd_loss": 0.007646733429282904,
+      "learning_rate": 8.661060545526724e-06,
+      "loss": 0.0312,
+      "step": 785,
+      "student_loss": 0.000303611159324646,
+      "teacher_loss": 0.0003056841087527573
+    },
+    {
+      "epoch": 0.22779700115340254,
+      "grad_norm": 1.7890625,
+      "kd_loss": 0.009472047910094261,
+      "learning_rate": 8.64262529203155e-06,
+      "loss": 0.03,
+      "step": 790,
+      "student_loss": 0.00036868476308882236,
+      "teacher_loss": 0.0003439165884628892
+    },
+    {
+      "epoch": 0.2292387543252595,
+      "grad_norm": 1.1015625,
+      "kd_loss": 0.013042761944234371,
+      "learning_rate": 8.624083901435613e-06,
+      "loss": 0.0345,
+      "step": 795,
+      "student_loss": 0.0004417412565089762,
+      "teacher_loss": 0.0002294621808687225
+    },
+    {
+      "epoch": 0.2306805074971165,
+      "grad_norm": 1.7578125,
+      "kd_loss": 0.02265016920864582,
+      "learning_rate": 8.60543691398922e-06,
+      "loss": 0.0413,
+      "step": 800,
+      "student_loss": 0.08563552051782608,
+      "teacher_loss": 0.039897482842206955
+    },
+    {
+      "epoch": 0.23212226066897348,
+      "grad_norm": 1.3125,
+      "kd_loss": 0.009194138459861279,
+      "learning_rate": 8.586684873019513e-06,
+      "loss": 0.0381,
+      "step": 805,
+      "student_loss": 0.00031980325002223253,
+      "teacher_loss": 0.0002621679159346968
+    },
+    {
+      "epoch": 0.23356401384083045,
+      "grad_norm": 15.8125,
+      "kd_loss": 0.015975888818502426,
+      "learning_rate": 8.567828324914633e-06,
+      "loss": 0.038,
+      "step": 810,
+      "student_loss": 0.0005471970071084797,
+      "teacher_loss": 0.00027335959021002054
+    },
+    {
+      "epoch": 0.23500576701268744,
+      "grad_norm": 2.25,
+      "kd_loss": 0.009164939634501934,
+      "learning_rate": 8.548867819107803e-06,
+      "loss": 0.0324,
+      "step": 815,
+      "student_loss": 0.0015802276320755482,
+      "teacher_loss": 0.008377115242183208
+    },
+    {
+      "epoch": 0.2364475201845444,
+      "grad_norm": 1.296875,
+      "kd_loss": 0.018547918647527695,
+      "learning_rate": 8.52980390806131e-06,
+      "loss": 0.0343,
+      "step": 820,
+      "student_loss": 0.008644208312034607,
+      "teacher_loss": 0.0001676520478213206
+    },
+    {
+      "epoch": 0.2378892733564014,
+      "grad_norm": 1.390625,
+      "kd_loss": 0.019806109368801117,
+      "learning_rate": 8.51063714725042e-06,
+      "loss": 0.032,
+      "step": 825,
+      "student_loss": 0.0006446156185120344,
+      "teacher_loss": 0.016100004315376282
+    },
+    {
+      "epoch": 0.23933102652825836,
+      "grad_norm": 1.296875,
+      "kd_loss": 0.01830875873565674,
+      "learning_rate": 8.49136809514719e-06,
+      "loss": 0.0352,
+      "step": 830,
+      "student_loss": 0.00992797315120697,
+      "teacher_loss": 0.004923146218061447
+    },
+    {
+      "epoch": 0.24077277970011535,
+      "grad_norm": 1.265625,
+      "kd_loss": 0.015559202991425991,
+      "learning_rate": 8.471997313204183e-06,
+      "loss": 0.0308,
+      "step": 835,
+      "student_loss": 0.01591889001429081,
+      "teacher_loss": 0.0025253912899643183
+    },
+    {
+      "epoch": 0.2422145328719723,
+      "grad_norm": 1.46875,
+      "kd_loss": 0.01866515353322029,
+      "learning_rate": 8.452525365838123e-06,
+      "loss": 0.0346,
+      "step": 840,
+      "student_loss": 0.001301256357692182,
+      "teacher_loss": 0.00011679113958962262
+    },
+    {
+      "epoch": 0.2436562860438293,
+      "grad_norm": 0.859375,
+      "kd_loss": 0.010034815408289433,
+      "learning_rate": 8.432952820413447e-06,
+      "loss": 0.0498,
+      "step": 845,
+      "student_loss": 0.0005439711967483163,
+      "teacher_loss": 0.0004172325134277344
+    },
+    {
+      "epoch": 0.24509803921568626,
+      "grad_norm": 1.78125,
+      "kd_loss": 0.023002484813332558,
+      "learning_rate": 8.41328024722577e-06,
+      "loss": 0.0335,
+      "step": 850,
+      "student_loss": 0.00615985132753849,
+      "teacher_loss": 0.00031692718039266765
+    },
+    {
+      "epoch": 0.24653979238754326,
+      "grad_norm": 1.140625,
+      "kd_loss": 0.011961855925619602,
+      "learning_rate": 8.393508219485263e-06,
+      "loss": 0.034,
+      "step": 855,
+      "student_loss": 0.0002558475243858993,
+      "teacher_loss": 0.000246651004999876
+    },
+    {
+      "epoch": 0.24798154555940022,
+      "grad_norm": 0.80859375,
+      "kd_loss": 0.012495153583586216,
+      "learning_rate": 8.373637313299966e-06,
+      "loss": 0.0399,
+      "step": 860,
+      "student_loss": 0.0005405155243352056,
+      "teacher_loss": 0.0004368047812022269
+    },
+    {
+      "epoch": 0.2494232987312572,
+      "grad_norm": 1.0078125,
+      "kd_loss": 0.014805592596530914,
+      "learning_rate": 8.353668107658984e-06,
+      "loss": 0.0337,
+      "step": 865,
+      "student_loss": 0.0005246512009762228,
+      "teacher_loss": 0.00047315956908278167
+    },
+    {
+      "epoch": 0.2508650519031142,
+      "grad_norm": 1.1171875,
+      "kd_loss": 0.015027549117803574,
+      "learning_rate": 8.333601184415632e-06,
+      "loss": 0.0341,
+      "step": 870,
+      "student_loss": 0.0015372103080153465,
+      "teacher_loss": 0.0004421547637321055
+    },
+    {
+      "epoch": 0.25230680507497116,
+      "grad_norm": 1.1171875,
+      "kd_loss": 0.010914859361946583,
+      "learning_rate": 8.31343712827047e-06,
+      "loss": 0.0345,
+      "step": 875,
+      "student_loss": 0.0001896051544463262,
+      "teacher_loss": 0.00020151768694631755
+    },
+    {
+      "epoch": 0.2537485582468281,
+      "grad_norm": 2.015625,
+      "kd_loss": 0.01555558666586876,
+      "learning_rate": 8.293176526754274e-06,
+      "loss": 0.0384,
+      "step": 880,
+      "student_loss": 0.002947386121377349,
+      "teacher_loss": 0.0017182360170409083
+    },
+    {
+      "epoch": 0.25519031141868515,
+      "grad_norm": 1.2265625,
+      "kd_loss": 0.04539031907916069,
+      "learning_rate": 8.27281997021091e-06,
+      "loss": 0.0301,
+      "step": 885,
+      "student_loss": 0.014408277347683907,
+      "teacher_loss": 0.045094698667526245
+    },
+    {
+      "epoch": 0.2566320645905421,
+      "grad_norm": 1.046875,
+      "kd_loss": 0.007991768419742584,
+      "learning_rate": 8.25236805178014e-06,
+      "loss": 0.0355,
+      "step": 890,
+      "student_loss": 0.0003164609370287508,
+      "teacher_loss": 0.0002852486213669181
+    },
+    {
+      "epoch": 0.25807381776239907,
+      "grad_norm": 1.1953125,
+      "kd_loss": 0.02082483097910881,
+      "learning_rate": 8.231821367380335e-06,
+      "loss": 0.0331,
+      "step": 895,
+      "student_loss": 0.011601348407566547,
+      "teacher_loss": 0.004591627512127161
+    },
+    {
+      "epoch": 0.25951557093425603,
+      "grad_norm": 0.9609375,
+      "kd_loss": 0.014409971423447132,
+      "learning_rate": 8.211180515691108e-06,
+      "loss": 0.0354,
+      "step": 900,
+      "student_loss": 0.03723900392651558,
+      "teacher_loss": 0.027190692722797394
+    },
+    {
+      "epoch": 0.26095732410611305,
+      "grad_norm": 1.4375,
+      "kd_loss": 0.014778131619095802,
+      "learning_rate": 8.190446098135883e-06,
+      "loss": 0.0322,
+      "step": 905,
+      "student_loss": 0.018417518585920334,
+      "teacher_loss": 0.015774844214320183
+    },
+    {
+      "epoch": 0.26239907727797,
+      "grad_norm": 1.015625,
+      "kd_loss": 0.007474376354366541,
+      "learning_rate": 8.16961871886435e-06,
+      "loss": 0.0291,
+      "step": 910,
+      "student_loss": 0.00027324067195877433,
+      "teacher_loss": 0.000292893877485767
+    },
+    {
+      "epoch": 0.263840830449827,
+      "grad_norm": 2.796875,
+      "kd_loss": 0.05165571719408035,
+      "learning_rate": 8.148698984734887e-06,
+      "loss": 0.0351,
+      "step": 915,
+      "student_loss": 0.05543224513530731,
+      "teacher_loss": 0.0008424259722232819
+    },
+    {
+      "epoch": 0.26528258362168394,
+      "grad_norm": 1.109375,
+      "kd_loss": 0.013850719667971134,
+      "learning_rate": 8.127687505296852e-06,
+      "loss": 0.0326,
+      "step": 920,
+      "student_loss": 0.013253331184387207,
+      "teacher_loss": 0.011715217493474483
+    },
+    {
+      "epoch": 0.26672433679354096,
+      "grad_norm": 3.28125,
+      "kd_loss": 0.00828032847493887,
+      "learning_rate": 8.106584892772844e-06,
+      "loss": 0.0316,
+      "step": 925,
+      "student_loss": 0.00028670401661656797,
+      "teacher_loss": 0.0004561161040328443
+    },
+    {
+      "epoch": 0.2681660899653979,
+      "grad_norm": 2.203125,
+      "kd_loss": 0.013576381839811802,
+      "learning_rate": 8.08539176204085e-06,
+      "loss": 0.0326,
+      "step": 930,
+      "student_loss": 0.002268102252855897,
+      "teacher_loss": 0.001438588136807084
+    },
+    {
+      "epoch": 0.2696078431372549,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.009203169494867325,
+      "learning_rate": 8.064108730616336e-06,
+      "loss": 0.0329,
+      "step": 935,
+      "student_loss": 0.054588161408901215,
+      "teacher_loss": 0.04182029515504837
+    },
+    {
+      "epoch": 0.2710495963091119,
+      "grad_norm": 0.96484375,
+      "kd_loss": 0.022441105917096138,
+      "learning_rate": 8.042736418634252e-06,
+      "loss": 0.0339,
+      "step": 940,
+      "student_loss": 0.003998172469437122,
+      "teacher_loss": 0.0004027044924441725
+    },
+    {
+      "epoch": 0.27249134948096887,
+      "grad_norm": 2.078125,
+      "kd_loss": 0.009604145772755146,
+      "learning_rate": 8.02127544883096e-06,
+      "loss": 0.0358,
+      "step": 945,
+      "student_loss": 0.0019527329131960869,
+      "teacher_loss": 0.001390090910717845
+    },
+    {
+      "epoch": 0.27393310265282583,
+      "grad_norm": 0.8828125,
+      "kd_loss": 0.01388311106711626,
+      "learning_rate": 7.999726446526095e-06,
+      "loss": 0.0306,
+      "step": 950,
+      "student_loss": 0.00017252952966373414,
+      "teacher_loss": 0.00010741472942754626
+    },
+    {
+      "epoch": 0.2753748558246828,
+      "grad_norm": 1.7109375,
+      "kd_loss": 0.01476337481290102,
+      "learning_rate": 7.978090039604342e-06,
+      "loss": 0.0348,
+      "step": 955,
+      "student_loss": 0.0003931497922167182,
+      "teacher_loss": 0.00022118802007753402
+    },
+    {
+      "epoch": 0.2768166089965398,
+      "grad_norm": 2.046875,
+      "kd_loss": 0.017180806025862694,
+      "learning_rate": 7.956366858497138e-06,
+      "loss": 0.0379,
+      "step": 960,
+      "student_loss": 0.008577854372560978,
+      "teacher_loss": 0.0037893427070230246
+    },
+    {
+      "epoch": 0.2782583621683968,
+      "grad_norm": 1.828125,
+      "kd_loss": 0.024987826123833656,
+      "learning_rate": 7.934557536164304e-06,
+      "loss": 0.0349,
+      "step": 965,
+      "student_loss": 0.010798306204378605,
+      "teacher_loss": 0.006717974785715342
+    },
+    {
+      "epoch": 0.27970011534025374,
+      "grad_norm": 1.390625,
+      "kd_loss": 0.01618223264813423,
+      "learning_rate": 7.91266270807561e-06,
+      "loss": 0.0336,
+      "step": 970,
+      "student_loss": 0.00025495997397229075,
+      "teacher_loss": 0.00014070281758904457
+    },
+    {
+      "epoch": 0.2811418685121107,
+      "grad_norm": 1.3828125,
+      "kd_loss": 0.011907395906746387,
+      "learning_rate": 7.890683012192249e-06,
+      "loss": 0.0329,
+      "step": 975,
+      "student_loss": 0.016343005001544952,
+      "teacher_loss": 0.0006543407216668129
+    },
+    {
+      "epoch": 0.2825836216839677,
+      "grad_norm": 1.125,
+      "kd_loss": 0.019409775733947754,
+      "learning_rate": 7.868619088948246e-06,
+      "loss": 0.0339,
+      "step": 980,
+      "student_loss": 0.00046765804290771484,
+      "teacher_loss": 0.00030171783873811364
+    },
+    {
+      "epoch": 0.2840253748558247,
+      "grad_norm": 2.328125,
+      "kd_loss": 0.014747396111488342,
+      "learning_rate": 7.846471581231814e-06,
+      "loss": 0.0359,
+      "step": 985,
+      "student_loss": 0.00035184339503757656,
+      "teacher_loss": 0.0015139622846618295
+    },
+    {
+      "epoch": 0.28546712802768165,
+      "grad_norm": 0.91796875,
+      "kd_loss": 0.011445100419223309,
+      "learning_rate": 7.824241134366604e-06,
+      "loss": 0.029,
+      "step": 990,
+      "student_loss": 0.00024842782295309007,
+      "teacher_loss": 0.00020773368305526674
+    },
+    {
+      "epoch": 0.2869088811995386,
+      "grad_norm": 1.0703125,
+      "kd_loss": 0.025197861716151237,
+      "learning_rate": 7.801928396092908e-06,
+      "loss": 0.0406,
+      "step": 995,
+      "student_loss": 0.00697642145678401,
+      "teacher_loss": 0.0003467589558567852
+    },
+    {
+      "epoch": 0.28835063437139563,
+      "grad_norm": 1.1953125,
+      "kd_loss": 0.02068255841732025,
+      "learning_rate": 7.779534016548791e-06,
+      "loss": 0.0341,
+      "step": 1000,
+      "student_loss": 0.006020409055054188,
+      "teacher_loss": 0.0005438941298052669
+    },
+    {
+      "epoch": 0.2897923875432526,
+      "grad_norm": 2.359375,
+      "kd_loss": 0.04980943724513054,
+      "learning_rate": 7.75705864825114e-06,
+      "loss": 0.0396,
+      "step": 1005,
+      "student_loss": 0.029118359088897705,
+      "teacher_loss": 0.06065075471997261
+    },
+    {
+      "epoch": 0.29123414071510956,
+      "grad_norm": 0.97265625,
+      "kd_loss": 0.008475052192807198,
+      "learning_rate": 7.734502946076656e-06,
+      "loss": 0.0343,
+      "step": 1010,
+      "student_loss": 0.00034027942456305027,
+      "teacher_loss": 0.0003257795178797096
+    },
+    {
+      "epoch": 0.2926758938869666,
+      "grad_norm": 1.4296875,
+      "kd_loss": 0.03711966052651405,
+      "learning_rate": 7.711867567242769e-06,
+      "loss": 0.0371,
+      "step": 1015,
+      "student_loss": 0.015291261486709118,
+      "teacher_loss": 0.008538170717656612
+    },
+    {
+      "epoch": 0.29411764705882354,
+      "grad_norm": 1.2890625,
+      "kd_loss": 0.017934992909431458,
+      "learning_rate": 7.689153171288487e-06,
+      "loss": 0.0307,
+      "step": 1020,
+      "student_loss": 0.04458937421441078,
+      "teacher_loss": 0.02570411190390587
+    },
+    {
+      "epoch": 0.2955594002306805,
+      "grad_norm": 0.9921875,
+      "kd_loss": 0.013871895149350166,
+      "learning_rate": 7.666360420055188e-06,
+      "loss": 0.0311,
+      "step": 1025,
+      "student_loss": 0.00020963574934285134,
+      "teacher_loss": 0.0001333217223873362
+    },
+    {
+      "epoch": 0.29700115340253747,
+      "grad_norm": 1.2265625,
+      "kd_loss": 0.026775527745485306,
+      "learning_rate": 7.643489977667327e-06,
+      "loss": 0.0368,
+      "step": 1030,
+      "student_loss": 0.007289777044206858,
+      "teacher_loss": 0.001790010486729443
+    },
+    {
+      "epoch": 0.2984429065743945,
+      "grad_norm": 1.09375,
+      "kd_loss": 0.015691982582211494,
+      "learning_rate": 7.6205425105130855e-06,
+      "loss": 0.0329,
+      "step": 1035,
+      "student_loss": 0.002382444217801094,
+      "teacher_loss": 0.003481772495433688
+    },
+    {
+      "epoch": 0.29988465974625145,
+      "grad_norm": 1.25,
+      "kd_loss": 0.008980272337794304,
+      "learning_rate": 7.597518687224959e-06,
+      "loss": 0.0351,
+      "step": 1040,
+      "student_loss": 0.004677724093198776,
+      "teacher_loss": 0.00020806775137316436
+    },
+    {
+      "epoch": 0.3013264129181084,
+      "grad_norm": 1.3125,
+      "kd_loss": 0.009131739847362041,
+      "learning_rate": 7.574419178660269e-06,
+      "loss": 0.0355,
+      "step": 1045,
+      "student_loss": 0.0002944975276477635,
+      "teacher_loss": 0.0002657297009136528
+    },
+    {
+      "epoch": 0.3027681660899654,
+      "grad_norm": 1.4296875,
+      "kd_loss": 0.012904422357678413,
+      "learning_rate": 7.551244657881618e-06,
+      "loss": 0.0361,
+      "step": 1050,
+      "student_loss": 0.042283061891794205,
+      "teacher_loss": 0.00028089716215617955
+    },
+    {
+      "epoch": 0.3042099192618224,
+      "grad_norm": 0.84375,
+      "kd_loss": 0.014016334898769855,
+      "learning_rate": 7.527995800137287e-06,
+      "loss": 0.035,
+      "step": 1055,
+      "student_loss": 0.00039159393054433167,
+      "teacher_loss": 0.003277128329500556
+    },
+    {
+      "epoch": 0.30565167243367936,
+      "grad_norm": 0.83984375,
+      "kd_loss": 0.01896512135863304,
+      "learning_rate": 7.504673282841544e-06,
+      "loss": 0.0336,
+      "step": 1060,
+      "student_loss": 0.006642750930041075,
+      "teacher_loss": 0.00026191832148469985
+    },
+    {
+      "epoch": 0.3070934256055363,
+      "grad_norm": 1.5234375,
+      "kd_loss": 0.04735545441508293,
+      "learning_rate": 7.481277785554918e-06,
+      "loss": 0.0347,
+      "step": 1065,
+      "student_loss": 0.029522676020860672,
+      "teacher_loss": 0.0019539925269782543
+    },
+    {
+      "epoch": 0.30853517877739334,
+      "grad_norm": 1.7421875,
+      "kd_loss": 0.021157141774892807,
+      "learning_rate": 7.457809989964393e-06,
+      "loss": 0.0364,
+      "step": 1070,
+      "student_loss": 0.007406193763017654,
+      "teacher_loss": 0.0001315005065407604
+    },
+    {
+      "epoch": 0.3099769319492503,
+      "grad_norm": 1.8515625,
+      "kd_loss": 0.009599569253623486,
+      "learning_rate": 7.434270579863549e-06,
+      "loss": 0.0375,
+      "step": 1075,
+      "student_loss": 0.004805834032595158,
+      "teacher_loss": 0.007656469475477934
+    },
+    {
+      "epoch": 0.31141868512110726,
+      "grad_norm": 1.2421875,
+      "kd_loss": 0.009782467968761921,
+      "learning_rate": 7.4106602411326345e-06,
+      "loss": 0.0353,
+      "step": 1080,
+      "student_loss": 0.0002771034778561443,
+      "teacher_loss": 0.008469615131616592
+    },
+    {
+      "epoch": 0.3128604382929642,
+      "grad_norm": 1.0859375,
+      "kd_loss": 0.009013949893414974,
+      "learning_rate": 7.386979661718585e-06,
+      "loss": 0.0379,
+      "step": 1085,
+      "student_loss": 0.00043321328121237457,
+      "teacher_loss": 0.00037873408291488886
+    },
+    {
+      "epoch": 0.31430219146482125,
+      "grad_norm": 1.34375,
+      "kd_loss": 0.01487668789923191,
+      "learning_rate": 7.363229531614973e-06,
+      "loss": 0.037,
+      "step": 1090,
+      "student_loss": 0.002832320285961032,
+      "teacher_loss": 0.0009056431008502841
+    },
+    {
+      "epoch": 0.3157439446366782,
+      "grad_norm": 1.078125,
+      "kd_loss": 0.013107807375490665,
+      "learning_rate": 7.339410542841906e-06,
+      "loss": 0.0392,
+      "step": 1095,
+      "student_loss": 0.048266466706991196,
+      "teacher_loss": 0.007708544377237558
+    },
+    {
+      "epoch": 0.31718569780853517,
+      "grad_norm": 2.5,
+      "kd_loss": 0.015548629686236382,
+      "learning_rate": 7.315523389425867e-06,
+      "loss": 0.038,
+      "step": 1100,
+      "student_loss": 0.0002608558861538768,
+      "teacher_loss": 0.000203106232220307
+    },
+    {
+      "epoch": 0.31862745098039214,
+      "grad_norm": 2.96875,
+      "kd_loss": 0.020934898406267166,
+      "learning_rate": 7.291568767379484e-06,
+      "loss": 0.0321,
+      "step": 1105,
+      "student_loss": 0.00109296350274235,
+      "teacher_loss": 0.00021800097601953894
+    },
+    {
+      "epoch": 0.32006920415224915,
+      "grad_norm": 1.140625,
+      "kd_loss": 0.03160605579614639,
+      "learning_rate": 7.267547374681259e-06,
+      "loss": 0.0303,
+      "step": 1110,
+      "student_loss": 0.000475409091450274,
+      "teacher_loss": 0.0006200398784130812
+    },
+    {
+      "epoch": 0.3215109573241061,
+      "grad_norm": 0.86328125,
+      "kd_loss": 0.017369039356708527,
+      "learning_rate": 7.24345991125522e-06,
+      "loss": 0.0365,
+      "step": 1115,
+      "student_loss": 0.0005862113903276622,
+      "teacher_loss": 0.000243371760006994
+    },
+    {
+      "epoch": 0.3229527104959631,
+      "grad_norm": 1.4453125,
+      "kd_loss": 0.016101138666272163,
+      "learning_rate": 7.219307078950536e-06,
+      "loss": 0.0338,
+      "step": 1120,
+      "student_loss": 0.004134031943976879,
+      "teacher_loss": 0.0011220399755984545
+    },
+    {
+      "epoch": 0.32439446366782004,
+      "grad_norm": 1.015625,
+      "kd_loss": 0.020290130749344826,
+      "learning_rate": 7.195089581521064e-06,
+      "loss": 0.037,
+      "step": 1125,
+      "student_loss": 0.03510711342096329,
+      "teacher_loss": 0.019309232011437416
+    },
+    {
+      "epoch": 0.32583621683967706,
+      "grad_norm": 1.2734375,
+      "kd_loss": 0.034219104796648026,
+      "learning_rate": 7.170808124604842e-06,
+      "loss": 0.0337,
+      "step": 1130,
+      "student_loss": 0.011434311978518963,
+      "teacher_loss": 0.0005800156504847109
+    },
+    {
+      "epoch": 0.327277970011534,
+      "grad_norm": 1.9296875,
+      "kd_loss": 0.02713492140173912,
+      "learning_rate": 7.14646341570353e-06,
+      "loss": 0.0363,
+      "step": 1135,
+      "student_loss": 0.031025355681777,
+      "teacher_loss": 0.015047783963382244
+    },
+    {
+      "epoch": 0.328719723183391,
+      "grad_norm": 1.234375,
+      "kd_loss": 0.007938395254313946,
+      "learning_rate": 7.122056164161795e-06,
+      "loss": 0.0323,
+      "step": 1140,
+      "student_loss": 0.0022050540428608656,
+      "teacher_loss": 0.00227121589705348
+    },
+    {
+      "epoch": 0.330161476355248,
+      "grad_norm": 1.2265625,
+      "kd_loss": 0.007772063836455345,
+      "learning_rate": 7.097587081146636e-06,
+      "loss": 0.0282,
+      "step": 1145,
+      "student_loss": 0.0010676963720470667,
+      "teacher_loss": 0.00042732557631097734
+    },
+    {
+      "epoch": 0.33160322952710497,
+      "grad_norm": 1.421875,
+      "kd_loss": 0.01249232329428196,
+      "learning_rate": 7.073056879626681e-06,
+      "loss": 0.0396,
+      "step": 1150,
+      "student_loss": 0.004681647755205631,
+      "teacher_loss": 0.0003934359410777688
+    },
+    {
+      "epoch": 0.33304498269896193,
+      "grad_norm": 0.98828125,
+      "kd_loss": 0.014167334884405136,
+      "learning_rate": 7.048466274351389e-06,
+      "loss": 0.0306,
+      "step": 1155,
+      "student_loss": 0.0022710193879902363,
+      "teacher_loss": 0.00011216364509891719
+    },
+    {
+      "epoch": 0.3344867358708189,
+      "grad_norm": 1.3984375,
+      "kd_loss": 0.013238955289125443,
+      "learning_rate": 7.023815981830236e-06,
+      "loss": 0.0422,
+      "step": 1160,
+      "student_loss": 0.03383457288146019,
+      "teacher_loss": 0.02717623859643936
+    },
+    {
+      "epoch": 0.3359284890426759,
+      "grad_norm": 0.8125,
+      "kd_loss": 0.013019035570323467,
+      "learning_rate": 6.999106720311846e-06,
+      "loss": 0.029,
+      "step": 1165,
+      "student_loss": 0.01575121283531189,
+      "teacher_loss": 0.0001449387928005308
+    },
+    {
+      "epoch": 0.3373702422145329,
+      "grad_norm": 1.015625,
+      "kd_loss": 0.012287464924156666,
+      "learning_rate": 6.974339209763043e-06,
+      "loss": 0.0314,
+      "step": 1170,
+      "student_loss": 0.01835404522716999,
+      "teacher_loss": 0.016026774421334267
+    },
+    {
+      "epoch": 0.33881199538638984,
+      "grad_norm": 4.28125,
+      "kd_loss": 0.019668571650981903,
+      "learning_rate": 6.949514171847891e-06,
+      "loss": 0.0322,
+      "step": 1175,
+      "student_loss": 0.0011080754920840263,
+      "teacher_loss": 0.0030183272901922464
+    },
+    {
+      "epoch": 0.3402537485582468,
+      "grad_norm": 1.5859375,
+      "kd_loss": 0.01167343370616436,
+      "learning_rate": 6.924632329906657e-06,
+      "loss": 0.0366,
+      "step": 1180,
+      "student_loss": 0.007453799247741699,
+      "teacher_loss": 0.004300081171095371
+    },
+    {
+      "epoch": 0.3416955017301038,
+      "grad_norm": 0.984375,
+      "kd_loss": 0.013878893107175827,
+      "learning_rate": 6.899694408934734e-06,
+      "loss": 0.0308,
+      "step": 1185,
+      "student_loss": 0.0019671425689011812,
+      "teacher_loss": 0.0009088900405913591
+    },
+    {
+      "epoch": 0.3431372549019608,
+      "grad_norm": 1.0703125,
+      "kd_loss": 0.01681910641491413,
+      "learning_rate": 6.874701135561524e-06,
+      "loss": 0.0324,
+      "step": 1190,
+      "student_loss": 0.0003535433206707239,
+      "teacher_loss": 0.0001539337681606412
+    },
+    {
+      "epoch": 0.34457900807381775,
+      "grad_norm": 1.5234375,
+      "kd_loss": 0.013083124533295631,
+      "learning_rate": 6.849653238029261e-06,
+      "loss": 0.0301,
+      "step": 1195,
+      "student_loss": 0.0008335649035871029,
+      "teacher_loss": 0.00023550030891783535
+    },
+    {
+      "epoch": 0.3460207612456747,
+      "grad_norm": 1.3515625,
+      "kd_loss": 0.007454675156623125,
+      "learning_rate": 6.824551446171788e-06,
+      "loss": 0.0362,
+      "step": 1200,
+      "student_loss": 0.0001837015151977539,
+      "teacher_loss": 0.00018060124421026558
+    },
+    {
+      "epoch": 0.34746251441753173,
+      "grad_norm": 0.88671875,
+      "kd_loss": 0.01776004582643509,
+      "learning_rate": 6.7993964913932975e-06,
+      "loss": 0.0315,
+      "step": 1205,
+      "student_loss": 0.0069564338773489,
+      "teacher_loss": 0.0001847606763476506
+    },
+    {
+      "epoch": 0.3489042675893887,
+      "grad_norm": 1.4140625,
+      "kd_loss": 0.013348796404898167,
+      "learning_rate": 6.774189106647021e-06,
+      "loss": 0.033,
+      "step": 1210,
+      "student_loss": 0.0007556963828392327,
+      "teacher_loss": 0.0002864951966330409
+    },
+    {
+      "epoch": 0.35034602076124566,
+      "grad_norm": 0.83984375,
+      "kd_loss": 0.020416274666786194,
+      "learning_rate": 6.748930026413865e-06,
+      "loss": 0.0336,
+      "step": 1215,
+      "student_loss": 0.009767523035407066,
+      "teacher_loss": 0.000396502495277673
+    },
+    {
+      "epoch": 0.3517877739331027,
+      "grad_norm": 1.5390625,
+      "kd_loss": 0.021770689636468887,
+      "learning_rate": 6.7236199866810185e-06,
+      "loss": 0.0335,
+      "step": 1220,
+      "student_loss": 0.026510940864682198,
+      "teacher_loss": 0.009501306340098381
+    },
+    {
+      "epoch": 0.35322952710495964,
+      "grad_norm": 1.453125,
+      "kd_loss": 0.008824316784739494,
+      "learning_rate": 6.698259724920503e-06,
+      "loss": 0.0364,
+      "step": 1225,
+      "student_loss": 0.0010392538970336318,
+      "teacher_loss": 0.005005348473787308
+    },
+    {
+      "epoch": 0.3546712802768166,
+      "grad_norm": 1.984375,
+      "kd_loss": 0.00727895786985755,
+      "learning_rate": 6.672849980067685e-06,
+      "loss": 0.0361,
+      "step": 1230,
+      "student_loss": 0.00048274945584125817,
+      "teacher_loss": 0.0004853134450968355
+    },
+    {
+      "epoch": 0.35611303344867357,
+      "grad_norm": 1.109375,
+      "kd_loss": 0.007448536343872547,
+      "learning_rate": 6.647391492499746e-06,
+      "loss": 0.0339,
+      "step": 1235,
+      "student_loss": 0.000482535979244858,
+      "teacher_loss": 0.0004550137382466346
+    },
+    {
+      "epoch": 0.3575547866205306,
+      "grad_norm": 1.828125,
+      "kd_loss": 0.009491190314292908,
+      "learning_rate": 6.621885004014113e-06,
+      "loss": 0.0416,
+      "step": 1240,
+      "student_loss": 0.0003805011510848999,
+      "teacher_loss": 0.0002974580565933138
+    },
+    {
+      "epoch": 0.35899653979238755,
+      "grad_norm": 1.5625,
+      "kd_loss": 0.012377781793475151,
+      "learning_rate": 6.596331257806837e-06,
+      "loss": 0.0387,
+      "step": 1245,
+      "student_loss": 0.00041664595482870936,
+      "teacher_loss": 0.0004409518151078373
+    },
+    {
+      "epoch": 0.3604382929642445,
+      "grad_norm": 1.1953125,
+      "kd_loss": 0.018127569928765297,
+      "learning_rate": 6.570730998450945e-06,
+      "loss": 0.0359,
+      "step": 1250,
+      "student_loss": 0.026891598477959633,
+      "teacher_loss": 0.003611157648265362
+    },
+    {
+      "epoch": 0.3618800461361015,
+      "grad_norm": 1.5625,
+      "kd_loss": 0.011977880261838436,
+      "learning_rate": 6.545084971874738e-06,
+      "loss": 0.0364,
+      "step": 1255,
+      "student_loss": 0.013014866970479488,
+      "teacher_loss": 0.00670029828324914
+    },
+    {
+      "epoch": 0.3633217993079585,
+      "grad_norm": 0.95703125,
+      "kd_loss": 0.00777229480445385,
+      "learning_rate": 6.519393925340067e-06,
+      "loss": 0.0378,
+      "step": 1260,
+      "student_loss": 0.00020224940089974552,
+      "teacher_loss": 0.0001866645470727235
+    },
+    {
+      "epoch": 0.36476355247981546,
+      "grad_norm": 2.5,
+      "kd_loss": 0.01976393535733223,
+      "learning_rate": 6.49365860742055e-06,
+      "loss": 0.0339,
+      "step": 1265,
+      "student_loss": 0.019368577748537064,
+      "teacher_loss": 0.019037866964936256
+    },
+    {
+      "epoch": 0.3662053056516724,
+      "grad_norm": 1.140625,
+      "kd_loss": 0.008854568935930729,
+      "learning_rate": 6.467879767979764e-06,
+      "loss": 0.0318,
+      "step": 1270,
+      "student_loss": 0.0003118886088486761,
+      "teacher_loss": 0.00024117485736496747
+    },
+    {
+      "epoch": 0.36764705882352944,
+      "grad_norm": 0.76953125,
+      "kd_loss": 0.014369281940162182,
+      "learning_rate": 6.442058158149396e-06,
+      "loss": 0.0263,
+      "step": 1275,
+      "student_loss": 0.005814644042402506,
+      "teacher_loss": 0.0010704882442951202
+    },
+    {
+      "epoch": 0.3690888119953864,
+      "grad_norm": 0.9453125,
+      "kd_loss": 0.014556691981852055,
+      "learning_rate": 6.4161945303073535e-06,
+      "loss": 0.0302,
+      "step": 1280,
+      "student_loss": 0.0008491647895425558,
+      "teacher_loss": 0.0003583270590752363
+    },
+    {
+      "epoch": 0.37053056516724336,
+      "grad_norm": 0.9453125,
+      "kd_loss": 0.009234513156116009,
+      "learning_rate": 6.390289638055851e-06,
+      "loss": 0.0314,
+      "step": 1285,
+      "student_loss": 0.0003543966740835458,
+      "teacher_loss": 0.0002642351610120386
+    },
+    {
+      "epoch": 0.3719723183391003,
+      "grad_norm": 1.09375,
+      "kd_loss": 0.013533290475606918,
+      "learning_rate": 6.364344236199441e-06,
+      "loss": 0.0349,
+      "step": 1290,
+      "student_loss": 0.04509873315691948,
+      "teacher_loss": 0.011920562013983727
+    },
+    {
+      "epoch": 0.37341407151095735,
+      "grad_norm": 1.71875,
+      "kd_loss": 0.010070240125060081,
+      "learning_rate": 6.3383590807230264e-06,
+      "loss": 0.0401,
+      "step": 1295,
+      "student_loss": 0.00038660253630951047,
+      "teacher_loss": 0.012159357778728008
+    },
+    {
+      "epoch": 0.3748558246828143,
+      "grad_norm": 0.765625,
+      "kd_loss": 0.01310427114367485,
+      "learning_rate": 6.3123349287698345e-06,
+      "loss": 0.0294,
+      "step": 1300,
+      "student_loss": 0.0010391304967924953,
+      "teacher_loss": 0.000431177846621722
+    },
+    {
+      "epoch": 0.3762975778546713,
+      "grad_norm": 1.8125,
+      "kd_loss": 0.03833131864666939,
+      "learning_rate": 6.286272538619351e-06,
+      "loss": 0.0384,
+      "step": 1305,
+      "student_loss": 0.0008277529268525541,
+      "teacher_loss": 0.3033102750778198
+    },
+    {
+      "epoch": 0.37773933102652824,
+      "grad_norm": 0.9921875,
+      "kd_loss": 0.008595611900091171,
+      "learning_rate": 6.260172669665233e-06,
+      "loss": 0.0275,
+      "step": 1310,
+      "student_loss": 0.0005548794870264828,
+      "teacher_loss": 0.00045184255577623844
+    },
+    {
+      "epoch": 0.37918108419838525,
+      "grad_norm": 1.0390625,
+      "kd_loss": 0.011356363072991371,
+      "learning_rate": 6.234036082393171e-06,
+      "loss": 0.0347,
+      "step": 1315,
+      "student_loss": 0.0020568380132317543,
+      "teacher_loss": 0.0015367335872724652
+    },
+    {
+      "epoch": 0.3806228373702422,
+      "grad_norm": 2.6875,
+      "kd_loss": 0.014263247139751911,
+      "learning_rate": 6.207863538358741e-06,
+      "loss": 0.029,
+      "step": 1320,
+      "student_loss": 0.0005206407513469458,
+      "teacher_loss": 0.00019755015091504902
+    },
+    {
+      "epoch": 0.3820645905420992,
+      "grad_norm": 1.0078125,
+      "kd_loss": 0.01969161257147789,
+      "learning_rate": 6.181655800165207e-06,
+      "loss": 0.0312,
+      "step": 1325,
+      "student_loss": 0.00045133603271096945,
+      "teacher_loss": 0.0016385306371375918
+    },
+    {
+      "epoch": 0.38350634371395614,
+      "grad_norm": 0.90234375,
+      "kd_loss": 0.0192131157964468,
+      "learning_rate": 6.155413631441307e-06,
+      "loss": 0.0329,
+      "step": 1330,
+      "student_loss": 0.005141856614500284,
+      "teacher_loss": 0.00033870694460347295
+    },
+    {
+      "epoch": 0.38494809688581316,
+      "grad_norm": 1.234375,
+      "kd_loss": 0.012846595607697964,
+      "learning_rate": 6.129137796818997e-06,
+      "loss": 0.0334,
+      "step": 1335,
+      "student_loss": 0.0005814098403789103,
+      "teacher_loss": 0.00034370762296020985
+    },
+    {
+      "epoch": 0.3863898500576701,
+      "grad_norm": 1.2265625,
+      "kd_loss": 0.025651609525084496,
+      "learning_rate": 6.102829061911176e-06,
+      "loss": 0.0374,
+      "step": 1340,
+      "student_loss": 0.016245299950242043,
+      "teacher_loss": 0.005765223875641823
+    },
+    {
+      "epoch": 0.3878316032295271,
+      "grad_norm": 1.34375,
+      "kd_loss": 0.007320974953472614,
+      "learning_rate": 6.076488193289375e-06,
+      "loss": 0.0328,
+      "step": 1345,
+      "student_loss": 0.00045855625648982823,
+      "teacher_loss": 0.00039081749855540693
+    },
+    {
+      "epoch": 0.3892733564013841,
+      "grad_norm": 1.2265625,
+      "kd_loss": 0.013206405565142632,
+      "learning_rate": 6.050115958461423e-06,
+      "loss": 0.0305,
+      "step": 1350,
+      "student_loss": 0.007964587770402431,
+      "teacher_loss": 0.00033924292074516416
+    },
+    {
+      "epoch": 0.39071510957324107,
+      "grad_norm": 1.625,
+      "kd_loss": 0.03144938498735428,
+      "learning_rate": 6.02371312584908e-06,
+      "loss": 0.0304,
+      "step": 1355,
+      "student_loss": 0.010693615302443504,
+      "teacher_loss": 0.0015053875977173448
+    },
+    {
+      "epoch": 0.39215686274509803,
+      "grad_norm": 1.1015625,
+      "kd_loss": 0.018121734261512756,
+      "learning_rate": 5.997280464765655e-06,
+      "loss": 0.0333,
+      "step": 1360,
+      "student_loss": 0.0034864251501858234,
+      "teacher_loss": 0.0005948848556727171
+    },
+    {
+      "epoch": 0.393598615916955,
+      "grad_norm": 1.65625,
+      "kd_loss": 0.017842859029769897,
+      "learning_rate": 5.970818745393579e-06,
+      "loss": 0.0317,
+      "step": 1365,
+      "student_loss": 0.033628303557634354,
+      "teacher_loss": 0.018931448459625244
+    },
+    {
+      "epoch": 0.395040369088812,
+      "grad_norm": 1.2109375,
+      "kd_loss": 0.011731956154108047,
+      "learning_rate": 5.9443287387619754e-06,
+      "loss": 0.0361,
+      "step": 1370,
+      "student_loss": 0.03157558664679527,
+      "teacher_loss": 0.015083356760442257
+    },
+    {
+      "epoch": 0.396482122260669,
+      "grad_norm": 0.984375,
+      "kd_loss": 0.029046373441815376,
+      "learning_rate": 5.9178112167241805e-06,
+      "loss": 0.0301,
+      "step": 1375,
+      "student_loss": 0.011733638122677803,
+      "teacher_loss": 0.0005304464721120894
+    },
+    {
+      "epoch": 0.39792387543252594,
+      "grad_norm": 1.046875,
+      "kd_loss": 0.007890443317592144,
+      "learning_rate": 5.8912669519352725e-06,
+      "loss": 0.0292,
+      "step": 1380,
+      "student_loss": 0.003315612906590104,
+      "teacher_loss": 0.0023998727556318045
+    },
+    {
+      "epoch": 0.3993656286043829,
+      "grad_norm": 1.21875,
+      "kd_loss": 0.01195371150970459,
+      "learning_rate": 5.864696717829539e-06,
+      "loss": 0.0311,
+      "step": 1385,
+      "student_loss": 0.025962568819522858,
+      "teacher_loss": 0.016366755589842796
+    },
+    {
+      "epoch": 0.4008073817762399,
+      "grad_norm": 1.53125,
+      "kd_loss": 0.014304417185485363,
+      "learning_rate": 5.838101288597951e-06,
+      "loss": 0.0327,
+      "step": 1390,
+      "student_loss": 0.000303708016872406,
+      "teacher_loss": 0.0004957292112521827
+    },
+    {
+      "epoch": 0.4022491349480969,
+      "grad_norm": 1.25,
+      "kd_loss": 0.021597102284431458,
+      "learning_rate": 5.8114814391656046e-06,
+      "loss": 0.0363,
+      "step": 1395,
+      "student_loss": 0.0030263301450759172,
+      "teacher_loss": 0.0013618755619972944
+    },
+    {
+      "epoch": 0.40369088811995385,
+      "grad_norm": 1.5390625,
+      "kd_loss": 0.014713764190673828,
+      "learning_rate": 5.78483794516914e-06,
+      "loss": 0.0313,
+      "step": 1400,
+      "student_loss": 0.001098829903639853,
+      "teacher_loss": 0.000684740487486124
+    },
+    {
+      "epoch": 0.40513264129181087,
+      "grad_norm": 1.6640625,
+      "kd_loss": 0.009366289712488651,
+      "learning_rate": 5.75817158293414e-06,
+      "loss": 0.0371,
+      "step": 1405,
+      "student_loss": 0.0001972005848074332,
+      "teacher_loss": 0.0001850561093306169
+    },
+    {
+      "epoch": 0.40657439446366783,
+      "grad_norm": 0.859375,
+      "kd_loss": 0.016328996047377586,
+      "learning_rate": 5.731483129452514e-06,
+      "loss": 0.0295,
+      "step": 1410,
+      "student_loss": 0.0008536606328561902,
+      "teacher_loss": 0.0004398300079628825
+    },
+    {
+      "epoch": 0.4080161476355248,
+      "grad_norm": 1.1328125,
+      "kd_loss": 0.010848389007151127,
+      "learning_rate": 5.704773362359854e-06,
+      "loss": 0.0293,
+      "step": 1415,
+      "student_loss": 0.002838744781911373,
+      "teacher_loss": 0.014628366567194462
+    },
+    {
+      "epoch": 0.40945790080738176,
+      "grad_norm": 2.453125,
+      "kd_loss": 0.019908664748072624,
+      "learning_rate": 5.678043059912776e-06,
+      "loss": 0.0359,
+      "step": 1420,
+      "student_loss": 0.0002800961665343493,
+      "teacher_loss": 0.0010707759065553546
+    },
+    {
+      "epoch": 0.4108996539792388,
+      "grad_norm": 1.3671875,
+      "kd_loss": 0.018275344744324684,
+      "learning_rate": 5.6512930009662524e-06,
+      "loss": 0.0328,
+      "step": 1425,
+      "student_loss": 0.02479003742337227,
+      "teacher_loss": 0.0003009576175827533
+    },
+    {
+      "epoch": 0.41234140715109574,
+      "grad_norm": 1.1328125,
+      "kd_loss": 0.011150456964969635,
+      "learning_rate": 5.624523964950903e-06,
+      "loss": 0.0273,
+      "step": 1430,
+      "student_loss": 0.0004835776053369045,
+      "teacher_loss": 0.00030788168078288436
+    },
+    {
+      "epoch": 0.4137831603229527,
+      "grad_norm": 1.21875,
+      "kd_loss": 0.008235611952841282,
+      "learning_rate": 5.597736731850295e-06,
+      "loss": 0.0301,
+      "step": 1435,
+      "student_loss": 0.001612766063772142,
+      "teacher_loss": 0.0010425216751173139
+    },
+    {
+      "epoch": 0.41522491349480967,
+      "grad_norm": 2.0,
+      "kd_loss": 0.013700158335268497,
+      "learning_rate": 5.570932082178219e-06,
+      "loss": 0.0373,
+      "step": 1440,
+      "student_loss": 0.00047144806012511253,
+      "teacher_loss": 0.0001182597188744694
+    },
+    {
+      "epoch": 0.4166666666666667,
+      "grad_norm": 1.0078125,
+      "kd_loss": 0.03291871026158333,
+      "learning_rate": 5.5441107969559315e-06,
+      "loss": 0.0314,
+      "step": 1445,
+      "student_loss": 0.02213055081665516,
+      "teacher_loss": 0.008454972878098488
+    },
+    {
+      "epoch": 0.41810841983852365,
+      "grad_norm": 1.2890625,
+      "kd_loss": 0.01569937914609909,
+      "learning_rate": 5.517273657689419e-06,
+      "loss": 0.0304,
+      "step": 1450,
+      "student_loss": 0.0019605839625000954,
+      "teacher_loss": 0.0004140445380471647
+    },
+    {
+      "epoch": 0.4195501730103806,
+      "grad_norm": 1.546875,
+      "kd_loss": 0.01067360769957304,
+      "learning_rate": 5.490421446346608e-06,
+      "loss": 0.032,
+      "step": 1455,
+      "student_loss": 0.00032557090162299573,
+      "teacher_loss": 0.000694004469551146
+    },
+    {
+      "epoch": 0.4209919261822376,
+      "grad_norm": 1.15625,
+      "kd_loss": 0.016091808676719666,
+      "learning_rate": 5.463554945334589e-06,
+      "loss": 0.0365,
+      "step": 1460,
+      "student_loss": 0.0004189903847873211,
+      "teacher_loss": 0.00016767717897891998
+    },
+    {
+      "epoch": 0.4224336793540946,
+      "grad_norm": 1.7890625,
+      "kd_loss": 0.010027039796113968,
+      "learning_rate": 5.43667493747682e-06,
+      "loss": 0.0421,
+      "step": 1465,
+      "student_loss": 0.013300910592079163,
+      "teacher_loss": 0.01153824757784605
+    },
+    {
+      "epoch": 0.42387543252595156,
+      "grad_norm": 1.9921875,
+      "kd_loss": 0.015040472149848938,
+      "learning_rate": 5.409782205990317e-06,
+      "loss": 0.0425,
+      "step": 1470,
+      "student_loss": 0.0007107509300112724,
+      "teacher_loss": 0.002018202794715762
+    },
+    {
+      "epoch": 0.4253171856978085,
+      "grad_norm": 1.21875,
+      "kd_loss": 0.006902145221829414,
+      "learning_rate": 5.3828775344628245e-06,
+      "loss": 0.0274,
+      "step": 1475,
+      "student_loss": 0.0003844336897600442,
+      "teacher_loss": 0.00035014282912015915
+    },
+    {
+      "epoch": 0.42675893886966554,
+      "grad_norm": 1.7265625,
+      "kd_loss": 0.006593468599021435,
+      "learning_rate": 5.355961706829997e-06,
+      "loss": 0.0431,
+      "step": 1480,
+      "student_loss": 0.00037691558827646077,
+      "teacher_loss": 0.0003957096778322011
+    },
+    {
+      "epoch": 0.4282006920415225,
+      "grad_norm": 1.96875,
+      "kd_loss": 0.0073732552118599415,
+      "learning_rate": 5.329035507352548e-06,
+      "loss": 0.0324,
+      "step": 1485,
+      "student_loss": 0.00028754115919582546,
+      "teacher_loss": 0.00032248100615106523
+    },
+    {
+      "epoch": 0.42964244521337946,
+      "grad_norm": 1.859375,
+      "kd_loss": 0.006218737456947565,
+      "learning_rate": 5.3020997205933985e-06,
+      "loss": 0.0314,
+      "step": 1490,
+      "student_loss": 0.0031746244058012962,
+      "teacher_loss": 0.002896807389333844
+    },
+    {
+      "epoch": 0.43108419838523643,
+      "grad_norm": 2.265625,
+      "kd_loss": 0.009453610517084599,
+      "learning_rate": 5.275155131394825e-06,
+      "loss": 0.0345,
+      "step": 1495,
+      "student_loss": 0.0003277304058428854,
+      "teacher_loss": 0.00025752728106454015
+    },
+    {
+      "epoch": 0.43252595155709345,
+      "grad_norm": 0.96875,
+      "kd_loss": 0.0312759093940258,
+      "learning_rate": 5.248202524855578e-06,
+      "loss": 0.0285,
+      "step": 1500,
+      "student_loss": 0.016945546492934227,
+      "teacher_loss": 0.006624266505241394
+    }
+  ],
+  "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/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-2000/added_tokens.json

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

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-2000/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. 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 Qwen, created by Alibaba Cloud. You 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/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-2000/config.json

@@ -0,0 +1,58 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "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/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-2000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9686f3a02f1a7fdd0c7c3136b2b3725533f059964bc42be07eaf0b00b9404041
+size 3096275515

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-2000/special_tokens_map.json

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

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-2000/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2bf3b0c709ed754c4ba18e84982d73bd1ec7455d713bfc48eb1672cf886a1b53
+size 9445953

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-2000/tokenizer.json

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

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-2000/tokenizer_config.json

@@ -0,0 +1,295 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<|trace_context_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151666": {
+      "content": "<|call_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151667": {
+      "content": "<|line_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151668": {
+      "content": "<|return_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151669": {
+      "content": "<|exception_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<|action_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<|arg_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151672": {
+      "content": "<|frame_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151673": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151674": {
+      "content": "<|latent_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151675": {
+      "content": "<|latent_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/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/codi1.5b_a1.0_b1.0_g1.0_ls1/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 Qwen, created by Alibaba Cloud. 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 Qwen, created by Alibaba Cloud. You 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/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-500/config.json

@@ -0,0 +1,58 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "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/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-500/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc3f71d977530ec1327bfb71e3217e9a65cf478fa98bff3f35ae724d3f4b503f
+size 3096275515

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/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/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-500/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a0aaf90e4d6c1ca000d4dc4464b6bb575f5d8a5b6539885bd0e7a8a6fdc1045
+size 9445953

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/checkpoint-500/tokenizer.json

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

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/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/codi1.5b_a1.0_b1.0_g1.0_ls1/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": 34.25,
+      "kd_loss": 0.2726191282272339,
+      "learning_rate": 4.444444444444445e-07,
+      "loss": 0.3835,
+      "step": 5,
+      "student_loss": 0.053846802562475204,
+      "teacher_loss": 0.00042048963950946927
+    },
+    {
+      "epoch": 0.0028835063437139563,
+      "grad_norm": 48.0,
+      "kd_loss": 0.24393108487129211,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 0.3789,
+      "step": 10,
+      "student_loss": 0.09970586746931076,
+      "teacher_loss": 0.0007505021058022976
+    },
+    {
+      "epoch": 0.004325259515570935,
+      "grad_norm": 21.75,
+      "kd_loss": 0.21377132833003998,
+      "learning_rate": 1.5555555555555558e-06,
+      "loss": 0.3585,
+      "step": 15,
+      "student_loss": 0.029451103881001472,
+      "teacher_loss": 0.0006585777155123651
+    },
+    {
+      "epoch": 0.0057670126874279125,
+      "grad_norm": 27.5,
+      "kd_loss": 0.2140207439661026,
+      "learning_rate": 2.1111111111111114e-06,
+      "loss": 0.3372,
+      "step": 20,
+      "student_loss": 0.1321348249912262,
+      "teacher_loss": 0.000296285143122077
+    },
+    {
+      "epoch": 0.00720876585928489,
+      "grad_norm": 14.5625,
+      "kd_loss": 0.18640007078647614,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 0.2845,
+      "step": 25,
+      "student_loss": 0.018293429166078568,
+      "teacher_loss": 0.0010102478554472327
+    },
+    {
+      "epoch": 0.00865051903114187,
+      "grad_norm": 7.1875,
+      "kd_loss": 0.14892512559890747,
+      "learning_rate": 3.2222222222222227e-06,
+      "loss": 0.2387,
+      "step": 30,
+      "student_loss": 0.005030089057981968,
+      "teacher_loss": 0.0003383651783224195
+    },
+    {
+      "epoch": 0.010092272202998846,
+      "grad_norm": 5.5,
+      "kd_loss": 0.1269177347421646,
+      "learning_rate": 3.777777777777778e-06,
+      "loss": 0.2117,
+      "step": 35,
+      "student_loss": 0.0014062788104638457,
+      "teacher_loss": 0.00025501433992758393
+    },
+    {
+      "epoch": 0.011534025374855825,
+      "grad_norm": 4.65625,
+      "kd_loss": 0.12731066346168518,
+      "learning_rate": 4.333333333333334e-06,
+      "loss": 0.1863,
+      "step": 40,
+      "student_loss": 0.049713075160980225,
+      "teacher_loss": 0.040033385157585144
+    },
+    {
+      "epoch": 0.012975778546712802,
+      "grad_norm": 7.375,
+      "kd_loss": 0.1775631606578827,
+      "learning_rate": 4.888888888888889e-06,
+      "loss": 0.1456,
+      "step": 45,
+      "student_loss": 0.06933413445949554,
+      "teacher_loss": 0.0009156111045740545
+    },
+    {
+      "epoch": 0.01441753171856978,
+      "grad_norm": 4.15625,
+      "kd_loss": 0.06588269770145416,
+      "learning_rate": 5.444444444444445e-06,
+      "loss": 0.1417,
+      "step": 50,
+      "student_loss": 0.008036560378968716,
+      "teacher_loss": 0.004471915308386087
+    },
+    {
+      "epoch": 0.015859284890426758,
+      "grad_norm": 8.375,
+      "kd_loss": 0.05523190647363663,
+      "learning_rate": 6e-06,
+      "loss": 0.1225,
+      "step": 55,
+      "student_loss": 0.0012776668881997466,
+      "teacher_loss": 0.00044586719013750553
+    },
+    {
+      "epoch": 0.01730103806228374,
+      "grad_norm": 5.125,
+      "kd_loss": 0.049259770661592484,
+      "learning_rate": 6.555555555555556e-06,
+      "loss": 0.1047,
+      "step": 60,
+      "student_loss": 0.01390351727604866,
+      "teacher_loss": 0.002199392532929778
+    },
+    {
+      "epoch": 0.018742791234140715,
+      "grad_norm": 7.375,
+      "kd_loss": 0.047981150448322296,
+      "learning_rate": 7.111111111111112e-06,
+      "loss": 0.1006,
+      "step": 65,
+      "student_loss": 0.04912174493074417,
+      "teacher_loss": 0.05355874449014664
+    },
+    {
+      "epoch": 0.020184544405997693,
+      "grad_norm": 5.21875,
+      "kd_loss": 0.07241208106279373,
+      "learning_rate": 7.666666666666667e-06,
+      "loss": 0.0935,
+      "step": 70,
+      "student_loss": 0.03459472954273224,
+      "teacher_loss": 0.0029977799858897924
+    },
+    {
+      "epoch": 0.02162629757785467,
+      "grad_norm": 5.125,
+      "kd_loss": 0.0307459756731987,
+      "learning_rate": 8.222222222222222e-06,
+      "loss": 0.0907,
+      "step": 75,
+      "student_loss": 0.0016565826954320073,
+      "teacher_loss": 0.0008985538734123111
+    },
+    {
+      "epoch": 0.02306805074971165,
+      "grad_norm": 1.9453125,
+      "kd_loss": 0.026722930371761322,
+      "learning_rate": 8.777777777777778e-06,
+      "loss": 0.0788,
+      "step": 80,
+      "student_loss": 0.0021714135073125362,
+      "teacher_loss": 0.0012323743430897593
+    },
+    {
+      "epoch": 0.024509803921568627,
+      "grad_norm": 3.71875,
+      "kd_loss": 0.023355644196271896,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.0733,
+      "step": 85,
+      "student_loss": 0.0012368966126814485,
+      "teacher_loss": 0.0009521314641460776
+    },
+    {
+      "epoch": 0.025951557093425604,
+      "grad_norm": 1.75,
+      "kd_loss": 0.0727810338139534,
+      "learning_rate": 9.88888888888889e-06,
+      "loss": 0.0676,
+      "step": 90,
+      "student_loss": 0.028416095301508904,
+      "teacher_loss": 0.005437986925244331
+    },
+    {
+      "epoch": 0.027393310265282585,
+      "grad_norm": 2.75,
+      "kd_loss": 0.04388673976063728,
+      "learning_rate": 9.99995337991255e-06,
+      "loss": 0.0638,
+      "step": 95,
+      "student_loss": 0.009086629375815392,
+      "teacher_loss": 0.005068701226264238
+    },
+    {
+      "epoch": 0.02883506343713956,
+      "grad_norm": 3.328125,
+      "kd_loss": 0.018161524087190628,
+      "learning_rate": 9.999763987297266e-06,
+      "loss": 0.0537,
+      "step": 100,
+      "student_loss": 0.0007679096888750792,
+      "teacher_loss": 0.0005449313321150839
+    },
+    {
+      "epoch": 0.03027681660899654,
+      "grad_norm": 2.0,
+      "kd_loss": 0.041785843670368195,
+      "learning_rate": 9.999428913912876e-06,
+      "loss": 0.0561,
+      "step": 105,
+      "student_loss": 0.005221863742917776,
+      "teacher_loss": 0.0004642533021979034
+    },
+    {
+      "epoch": 0.031718569780853516,
+      "grad_norm": 1.90625,
+      "kd_loss": 0.03158900886774063,
+      "learning_rate": 9.998948169522587e-06,
+      "loss": 0.0634,
+      "step": 110,
+      "student_loss": 0.02624746784567833,
+      "teacher_loss": 0.019401099532842636
+    },
+    {
+      "epoch": 0.03316032295271049,
+      "grad_norm": 1.84375,
+      "kd_loss": 0.025746585801243782,
+      "learning_rate": 9.998321768134101e-06,
+      "loss": 0.058,
+      "step": 115,
+      "student_loss": 0.0007584464619867504,
+      "teacher_loss": 0.00038955273339524865
+    },
+    {
+      "epoch": 0.03460207612456748,
+      "grad_norm": 2.5,
+      "kd_loss": 0.017939826473593712,
+      "learning_rate": 9.997549727999212e-06,
+      "loss": 0.0505,
+      "step": 120,
+      "student_loss": 0.0029337499290704727,
+      "teacher_loss": 0.0009877075208351016
+    },
+    {
+      "epoch": 0.036043829296424454,
+      "grad_norm": 2.046875,
+      "kd_loss": 0.0580839179456234,
+      "learning_rate": 9.996632071613257e-06,
+      "loss": 0.0538,
+      "step": 125,
+      "student_loss": 0.042769238352775574,
+      "teacher_loss": 0.0003408969205338508
+    },
+    {
+      "epoch": 0.03748558246828143,
+      "grad_norm": 11.75,
+      "kd_loss": 0.017885997891426086,
+      "learning_rate": 9.995568825714479e-06,
+      "loss": 0.0556,
+      "step": 130,
+      "student_loss": 0.0008774802554398775,
+      "teacher_loss": 0.000655525189358741
+    },
+    {
+      "epoch": 0.03892733564013841,
+      "grad_norm": 1.1875,
+      "kd_loss": 0.032637353986501694,
+      "learning_rate": 9.994360021283235e-06,
+      "loss": 0.054,
+      "step": 135,
+      "student_loss": 0.009718491695821285,
+      "teacher_loss": 0.0008957762038335204
+    },
+    {
+      "epoch": 0.040369088811995385,
+      "grad_norm": 1.015625,
+      "kd_loss": 0.01658937893807888,
+      "learning_rate": 9.993005693541098e-06,
+      "loss": 0.0484,
+      "step": 140,
+      "student_loss": 0.0006278170621953905,
+      "teacher_loss": 0.00041464308742433786
+    },
+    {
+      "epoch": 0.04181084198385236,
+      "grad_norm": 3.09375,
+      "kd_loss": 0.06331603974103928,
+      "learning_rate": 9.991505881949837e-06,
+      "loss": 0.0455,
+      "step": 145,
+      "student_loss": 0.04779814928770065,
+      "teacher_loss": 0.008776809088885784
+    },
+    {
+      "epoch": 0.04325259515570934,
+      "grad_norm": 1.1875,
+      "kd_loss": 0.012085826136171818,
+      "learning_rate": 9.989860630210251e-06,
+      "loss": 0.0447,
+      "step": 150,
+      "student_loss": 0.0007606842555105686,
+      "teacher_loss": 0.0006636636680923402
+    },
+    {
+      "epoch": 0.04469434832756632,
+      "grad_norm": 1.8515625,
+      "kd_loss": 0.015217100270092487,
+      "learning_rate": 9.98806998626091e-06,
+      "loss": 0.0574,
+      "step": 155,
+      "student_loss": 0.0006297255749814212,
+      "teacher_loss": 0.0004516748304013163
+    },
+    {
+      "epoch": 0.0461361014994233,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.021681347861886024,
+      "learning_rate": 9.98613400227676e-06,
+      "loss": 0.0499,
+      "step": 160,
+      "student_loss": 0.08383286744356155,
+      "teacher_loss": 0.0425245501101017
+    },
+    {
+      "epoch": 0.04757785467128028,
+      "grad_norm": 1.0625,
+      "kd_loss": 0.02377716824412346,
+      "learning_rate": 9.984052734667586e-06,
+      "loss": 0.0438,
+      "step": 165,
+      "student_loss": 0.000517472974024713,
+      "teacher_loss": 0.0054916865192353725
+    },
+    {
+      "epoch": 0.049019607843137254,
+      "grad_norm": 1.296875,
+      "kd_loss": 0.012254269793629646,
+      "learning_rate": 9.981826244076387e-06,
+      "loss": 0.045,
+      "step": 170,
+      "student_loss": 0.0007025015656836331,
+      "teacher_loss": 0.0005940330447629094
+    },
+    {
+      "epoch": 0.05046136101499423,
+      "grad_norm": 1.2265625,
+      "kd_loss": 0.022828614339232445,
+      "learning_rate": 9.979454595377594e-06,
+      "loss": 0.0449,
+      "step": 175,
+      "student_loss": 0.002857188694179058,
+      "teacher_loss": 0.00044675302342511714
+    },
+    {
+      "epoch": 0.05190311418685121,
+      "grad_norm": 1.15625,
+      "kd_loss": 0.02279340662062168,
+      "learning_rate": 9.97693785767519e-06,
+      "loss": 0.0416,
+      "step": 180,
+      "student_loss": 0.0009870725916698575,
+      "teacher_loss": 0.0002556022081989795
+    },
+    {
+      "epoch": 0.05334486735870819,
+      "grad_norm": 1.484375,
+      "kd_loss": 0.02873876690864563,
+      "learning_rate": 9.9742761043007e-06,
+      "loss": 0.0472,
+      "step": 185,
+      "student_loss": 0.04467737674713135,
+      "teacher_loss": 0.0002736605820246041
+    },
+    {
+      "epoch": 0.05478662053056517,
+      "grad_norm": 1.5078125,
+      "kd_loss": 0.03800494968891144,
+      "learning_rate": 9.971469412811032e-06,
+      "loss": 0.0432,
+      "step": 190,
+      "student_loss": 0.006019711960107088,
+      "teacher_loss": 0.005745577625930309
+    },
+    {
+      "epoch": 0.056228373702422146,
+      "grad_norm": 2.453125,
+      "kd_loss": 0.03193611651659012,
+      "learning_rate": 9.968517864986247e-06,
+      "loss": 0.044,
+      "step": 195,
+      "student_loss": 0.010869728401303291,
+      "teacher_loss": 0.0007882199133746326
+    },
+    {
+      "epoch": 0.05767012687427912,
+      "grad_norm": 0.703125,
+      "kd_loss": 0.025101784616708755,
+      "learning_rate": 9.965421546827153e-06,
+      "loss": 0.0501,
+      "step": 200,
+      "student_loss": 0.0018625810043886304,
+      "teacher_loss": 0.00027628077077679336
+    },
+    {
+      "epoch": 0.0591118800461361,
+      "grad_norm": 1.359375,
+      "kd_loss": 0.02438431791961193,
+      "learning_rate": 9.962180548552812e-06,
+      "loss": 0.0458,
+      "step": 205,
+      "student_loss": 0.03717027232050896,
+      "teacher_loss": 0.02886415459215641
+    },
+    {
+      "epoch": 0.06055363321799308,
+      "grad_norm": 1.84375,
+      "kd_loss": 0.02243652194738388,
+      "learning_rate": 9.958794964597906e-06,
+      "loss": 0.043,
+      "step": 210,
+      "student_loss": 0.0013577835634350777,
+      "teacher_loss": 0.000932522292714566
+    },
+    {
+      "epoch": 0.061995386389850055,
+      "grad_norm": 1.078125,
+      "kd_loss": 0.028962837532162666,
+      "learning_rate": 9.955264893609986e-06,
+      "loss": 0.0453,
+      "step": 215,
+      "student_loss": 0.001151631586253643,
+      "teacher_loss": 0.0003005610778927803
+    },
+    {
+      "epoch": 0.06343713956170703,
+      "grad_norm": 1.9375,
+      "kd_loss": 0.03575044497847557,
+      "learning_rate": 9.951590438446597e-06,
+      "loss": 0.0453,
+      "step": 220,
+      "student_loss": 0.048845142126083374,
+      "teacher_loss": 0.009562678635120392
+    },
+    {
+      "epoch": 0.06487889273356401,
+      "grad_norm": 1.375,
+      "kd_loss": 0.015835396945476532,
+      "learning_rate": 9.947771706172285e-06,
+      "loss": 0.0506,
+      "step": 225,
+      "student_loss": 0.0546843521296978,
+      "teacher_loss": 0.0008852278115227818
+    },
+    {
+      "epoch": 0.06632064590542099,
+      "grad_norm": 2.125,
+      "kd_loss": 0.02423272468149662,
+      "learning_rate": 9.943808808055472e-06,
+      "loss": 0.0398,
+      "step": 230,
+      "student_loss": 0.0009826251771301031,
+      "teacher_loss": 0.002756665926426649
+    },
+    {
+      "epoch": 0.06776239907727798,
+      "grad_norm": 0.93359375,
+      "kd_loss": 0.04541483893990517,
+      "learning_rate": 9.93970185956522e-06,
+      "loss": 0.0415,
+      "step": 235,
+      "student_loss": 0.014311765320599079,
+      "teacher_loss": 0.0015107382787391543
+    },
+    {
+      "epoch": 0.06920415224913495,
+      "grad_norm": 1.296875,
+      "kd_loss": 0.018931882455945015,
+      "learning_rate": 9.93545098036786e-06,
+      "loss": 0.0432,
+      "step": 240,
+      "student_loss": 0.0005313835572451353,
+      "teacher_loss": 0.00029519537929445505
+    },
+    {
+      "epoch": 0.07064590542099193,
+      "grad_norm": 1.2890625,
+      "kd_loss": 0.012201114557683468,
+      "learning_rate": 9.931056294323507e-06,
+      "loss": 0.0392,
+      "step": 245,
+      "student_loss": 0.0008615658152848482,
+      "teacher_loss": 0.0008678268059156835
+    },
+    {
+      "epoch": 0.07208765859284891,
+      "grad_norm": 1.9296875,
+      "kd_loss": 0.013639300130307674,
+      "learning_rate": 9.926517929482454e-06,
+      "loss": 0.0416,
+      "step": 250,
+      "student_loss": 0.000850351236294955,
+      "teacher_loss": 0.0004967341665178537
+    },
+    {
+      "epoch": 0.07352941176470588,
+      "grad_norm": 0.82421875,
+      "kd_loss": 0.010670178569853306,
+      "learning_rate": 9.921836018081438e-06,
+      "loss": 0.0421,
+      "step": 255,
+      "student_loss": 0.0005453266203403473,
+      "teacher_loss": 0.0004909925046376884
+    },
+    {
+      "epoch": 0.07497116493656286,
+      "grad_norm": 1.375,
+      "kd_loss": 0.019146086648106575,
+      "learning_rate": 9.917010696539793e-06,
+      "loss": 0.0448,
+      "step": 260,
+      "student_loss": 0.040390945971012115,
+      "teacher_loss": 0.033758774399757385
+    },
+    {
+      "epoch": 0.07641291810841984,
+      "grad_norm": 1.78125,
+      "kd_loss": 0.011872848495841026,
+      "learning_rate": 9.912042105455462e-06,
+      "loss": 0.0422,
+      "step": 265,
+      "student_loss": 0.0005196996498852968,
+      "teacher_loss": 0.00042158958967775106
+    },
+    {
+      "epoch": 0.07785467128027682,
+      "grad_norm": 1.28125,
+      "kd_loss": 0.012035544961690903,
+      "learning_rate": 9.906930389600917e-06,
+      "loss": 0.0458,
+      "step": 270,
+      "student_loss": 0.0009468707139603794,
+      "teacher_loss": 0.000693841720931232
+    },
+    {
+      "epoch": 0.07929642445213379,
+      "grad_norm": 1.7890625,
+      "kd_loss": 0.01897946372628212,
+      "learning_rate": 9.901675697918925e-06,
+      "loss": 0.0432,
+      "step": 275,
+      "student_loss": 0.047525323927402496,
+      "teacher_loss": 0.03367537260055542
+    },
+    {
+      "epoch": 0.08073817762399077,
+      "grad_norm": 1.5859375,
+      "kd_loss": 0.021627889946103096,
+      "learning_rate": 9.896278183518216e-06,
+      "loss": 0.041,
+      "step": 280,
+      "student_loss": 0.0006324543501250446,
+      "teacher_loss": 0.0002107525651808828
+    },
+    {
+      "epoch": 0.08217993079584775,
+      "grad_norm": 1.46875,
+      "kd_loss": 0.013249333947896957,
+      "learning_rate": 9.890738003669029e-06,
+      "loss": 0.0444,
+      "step": 285,
+      "student_loss": 0.0005354061140678823,
+      "teacher_loss": 0.0004488772538024932
+    },
+    {
+      "epoch": 0.08362168396770472,
+      "grad_norm": 1.3359375,
+      "kd_loss": 0.011662454344332218,
+      "learning_rate": 9.885055319798514e-06,
+      "loss": 0.0476,
+      "step": 290,
+      "student_loss": 0.0025434105191379786,
+      "teacher_loss": 0.0015740139642730355
+    },
+    {
+      "epoch": 0.0850634371395617,
+      "grad_norm": 1.828125,
+      "kd_loss": 0.01183430477976799,
+      "learning_rate": 9.879230297486034e-06,
+      "loss": 0.0425,
+      "step": 295,
+      "student_loss": 0.0004690758651122451,
+      "teacher_loss": 0.00038755135028623044
+    },
+    {
+      "epoch": 0.08650519031141868,
+      "grad_norm": 1.03125,
+      "kd_loss": 0.03465069457888603,
+      "learning_rate": 9.873263106458353e-06,
+      "loss": 0.0394,
+      "step": 300,
+      "student_loss": 0.018424537032842636,
+      "teacher_loss": 0.002826885087415576
+    },
+    {
+      "epoch": 0.08794694348327567,
+      "grad_norm": 1.125,
+      "kd_loss": 0.013110927306115627,
+      "learning_rate": 9.86715392058467e-06,
+      "loss": 0.0357,
+      "step": 305,
+      "student_loss": 0.01486227661371231,
+      "teacher_loss": 0.0012473311508074403
+    },
+    {
+      "epoch": 0.08938869665513265,
+      "grad_norm": 1.4765625,
+      "kd_loss": 0.010091043077409267,
+      "learning_rate": 9.860902917871566e-06,
+      "loss": 0.039,
+      "step": 310,
+      "student_loss": 0.00043867999920621514,
+      "teacher_loss": 0.00034055308788083494
+    },
+    {
+      "epoch": 0.09083044982698962,
+      "grad_norm": 1.0625,
+      "kd_loss": 0.02157684415578842,
+      "learning_rate": 9.854510280457819e-06,
+      "loss": 0.0391,
+      "step": 315,
+      "student_loss": 0.014208661392331123,
+      "teacher_loss": 0.001196117140352726
+    },
+    {
+      "epoch": 0.0922722029988466,
+      "grad_norm": 1.1953125,
+      "kd_loss": 0.0127296457067132,
+      "learning_rate": 9.847976194609083e-06,
+      "loss": 0.0432,
+      "step": 320,
+      "student_loss": 0.0004476152826100588,
+      "teacher_loss": 0.000240055174799636
+    },
+    {
+      "epoch": 0.09371395617070358,
+      "grad_norm": 1.5546875,
+      "kd_loss": 0.032895129173994064,
+      "learning_rate": 9.841300850712479e-06,
+      "loss": 0.0377,
+      "step": 325,
+      "student_loss": 0.008438433520495892,
+      "teacher_loss": 0.015148998238146305
+    },
+    {
+      "epoch": 0.09515570934256055,
+      "grad_norm": 1.3046875,
+      "kd_loss": 0.04177394509315491,
+      "learning_rate": 9.834484443271035e-06,
+      "loss": 0.0389,
+      "step": 330,
+      "student_loss": 0.01764899119734764,
+      "teacher_loss": 0.001141766319051385
+    },
+    {
+      "epoch": 0.09659746251441753,
+      "grad_norm": 0.9296875,
+      "kd_loss": 0.008085020817816257,
+      "learning_rate": 9.827527170898019e-06,
+      "loss": 0.0328,
+      "step": 335,
+      "student_loss": 0.0030857992824167013,
+      "teacher_loss": 0.0029651003424078226
+    },
+    {
+      "epoch": 0.09803921568627451,
+      "grad_norm": 1.5625,
+      "kd_loss": 0.014680181629955769,
+      "learning_rate": 9.820429236311158e-06,
+      "loss": 0.0397,
+      "step": 340,
+      "student_loss": 0.010333855636417866,
+      "teacher_loss": 0.004541066940873861
+    },
+    {
+      "epoch": 0.09948096885813149,
+      "grad_norm": 0.9921875,
+      "kd_loss": 0.01903965324163437,
+      "learning_rate": 9.81319084632673e-06,
+      "loss": 0.043,
+      "step": 345,
+      "student_loss": 0.012247292324900627,
+      "teacher_loss": 0.00043584281229414046
+    },
+    {
+      "epoch": 0.10092272202998846,
+      "grad_norm": 1.1171875,
+      "kd_loss": 0.037925247102975845,
+      "learning_rate": 9.805812211853534e-06,
+      "loss": 0.034,
+      "step": 350,
+      "student_loss": 0.0080111650750041,
+      "teacher_loss": 0.0011970563791692257
+    },
+    {
+      "epoch": 0.10236447520184544,
+      "grad_norm": 1.4296875,
+      "kd_loss": 0.00829800870269537,
+      "learning_rate": 9.798293547886748e-06,
+      "loss": 0.0396,
+      "step": 355,
+      "student_loss": 0.0005787648260593414,
+      "teacher_loss": 0.0005460750544443727
+    },
+    {
+      "epoch": 0.10380622837370242,
+      "grad_norm": 1.03125,
+      "kd_loss": 0.008213132619857788,
+      "learning_rate": 9.790635073501662e-06,
+      "loss": 0.0393,
+      "step": 360,
+      "student_loss": 0.00047661116695962846,
+      "teacher_loss": 0.00041378207970410585
+    },
+    {
+      "epoch": 0.1052479815455594,
+      "grad_norm": 1.53125,
+      "kd_loss": 0.014927592128515244,
+      "learning_rate": 9.782837011847294e-06,
+      "loss": 0.0436,
+      "step": 365,
+      "student_loss": 0.043972667306661606,
+      "teacher_loss": 0.02826174721121788
+    },
+    {
+      "epoch": 0.10668973471741638,
+      "grad_norm": 1.9375,
+      "kd_loss": 0.03636546432971954,
+      "learning_rate": 9.774899590139897e-06,
+      "loss": 0.0423,
+      "step": 370,
+      "student_loss": 0.021089306101202965,
+      "teacher_loss": 0.0004860333865508437
+    },
+    {
+      "epoch": 0.10813148788927336,
+      "grad_norm": 1.140625,
+      "kd_loss": 0.015318617224693298,
+      "learning_rate": 9.766823039656328e-06,
+      "loss": 0.0357,
+      "step": 375,
+      "student_loss": 0.03581877425312996,
+      "teacher_loss": 0.02248113416135311
+    },
+    {
+      "epoch": 0.10957324106113034,
+      "grad_norm": 0.74609375,
+      "kd_loss": 0.009901116602122784,
+      "learning_rate": 9.758607595727309e-06,
+      "loss": 0.0372,
+      "step": 380,
+      "student_loss": 0.00044091136078350246,
+      "teacher_loss": 0.0003208389098290354
+    },
+    {
+      "epoch": 0.11101499423298732,
+      "grad_norm": 1.5390625,
+      "kd_loss": 0.015398882329463959,
+      "learning_rate": 9.75025349773058e-06,
+      "loss": 0.0348,
+      "step": 385,
+      "student_loss": 0.0006440416909754276,
+      "teacher_loss": 0.0005293422727845609
+    },
+    {
+      "epoch": 0.11245674740484429,
+      "grad_norm": 1.1171875,
+      "kd_loss": 0.009845156222581863,
+      "learning_rate": 9.741760989083914e-06,
+      "loss": 0.0326,
+      "step": 390,
+      "student_loss": 0.00028396621928550303,
+      "teacher_loss": 0.00030337049975059927
+    },
+    {
+      "epoch": 0.11389850057670127,
+      "grad_norm": 1.28125,
+      "kd_loss": 0.011232681572437286,
+      "learning_rate": 9.73313031723803e-06,
+      "loss": 0.038,
+      "step": 395,
+      "student_loss": 0.003544658189639449,
+      "teacher_loss": 0.0032308201771229506
+    },
+    {
+      "epoch": 0.11534025374855825,
+      "grad_norm": 1.609375,
+      "kd_loss": 0.01615743711590767,
+      "learning_rate": 9.724361733669383e-06,
+      "loss": 0.0357,
+      "step": 400,
+      "student_loss": 0.0006839327397756279,
+      "teacher_loss": 0.000942843733355403
+    },
+    {
+      "epoch": 0.11678200692041522,
+      "grad_norm": 1.078125,
+      "kd_loss": 0.013564227148890495,
+      "learning_rate": 9.715455493872827e-06,
+      "loss": 0.0415,
+      "step": 405,
+      "student_loss": 0.005195911042392254,
+      "teacher_loss": 0.00030810022144578397
+    },
+    {
+      "epoch": 0.1182237600922722,
+      "grad_norm": 1.5703125,
+      "kd_loss": 0.025074705481529236,
+      "learning_rate": 9.70641185735419e-06,
+      "loss": 0.0411,
+      "step": 410,
+      "student_loss": 0.01386551558971405,
+      "teacher_loss": 0.00967031717300415
+    },
+    {
+      "epoch": 0.11966551326412918,
+      "grad_norm": 1.1875,
+      "kd_loss": 0.015891144052147865,
+      "learning_rate": 9.697231087622691e-06,
+      "loss": 0.0411,
+      "step": 415,
+      "student_loss": 0.0021395094227045774,
+      "teacher_loss": 0.0021108894143253565
+    },
+    {
+      "epoch": 0.12110726643598616,
+      "grad_norm": 1.40625,
+      "kd_loss": 0.021059690043330193,
+      "learning_rate": 9.687913452183277e-06,
+      "loss": 0.0379,
+      "step": 420,
+      "student_loss": 0.04692540317773819,
+      "teacher_loss": 0.02177572436630726
+    },
+    {
+      "epoch": 0.12254901960784313,
+      "grad_norm": 1.3125,
+      "kd_loss": 0.007998880930244923,
+      "learning_rate": 9.678459222528825e-06,
+      "loss": 0.0366,
+      "step": 425,
+      "student_loss": 0.0025826317723840475,
+      "teacher_loss": 0.002152319997549057
+    },
+    {
+      "epoch": 0.12399077277970011,
+      "grad_norm": 1.5390625,
+      "kd_loss": 0.026353655382990837,
+      "learning_rate": 9.668868674132224e-06,
+      "loss": 0.0448,
+      "step": 430,
+      "student_loss": 0.04476851597428322,
+      "teacher_loss": 0.06174921244382858
+    },
+    {
+      "epoch": 0.1254325259515571,
+      "grad_norm": 1.3359375,
+      "kd_loss": 0.008578015491366386,
+      "learning_rate": 9.65914208643836e-06,
+      "loss": 0.0355,
+      "step": 435,
+      "student_loss": 0.0006695028278045356,
+      "teacher_loss": 0.0005997278494760394
+    },
+    {
+      "epoch": 0.12687427912341406,
+      "grad_norm": 1.1640625,
+      "kd_loss": 0.025591254234313965,
+      "learning_rate": 9.649279742855964e-06,
+      "loss": 0.039,
+      "step": 440,
+      "student_loss": 0.020308705046772957,
+      "teacher_loss": 0.005524828098714352
+    },
+    {
+      "epoch": 0.12831603229527105,
+      "grad_norm": 1.875,
+      "kd_loss": 0.019290607422590256,
+      "learning_rate": 9.639281930749363e-06,
+      "loss": 0.0384,
+      "step": 445,
+      "student_loss": 0.0254460871219635,
+      "teacher_loss": 0.01746205799281597
+    },
+    {
+      "epoch": 0.12975778546712802,
+      "grad_norm": 1.3046875,
+      "kd_loss": 0.01817379891872406,
+      "learning_rate": 9.629148941430096e-06,
+      "loss": 0.0398,
+      "step": 450,
+      "student_loss": 0.0006312711630016565,
+      "teacher_loss": 0.0004661971179302782
+    },
+    {
+      "epoch": 0.131199538638985,
+      "grad_norm": 1.390625,
+      "kd_loss": 0.01092479843646288,
+      "learning_rate": 9.618881070148436e-06,
+      "loss": 0.039,
+      "step": 455,
+      "student_loss": 0.005430892109870911,
+      "teacher_loss": 0.0015254098689183593
+    },
+    {
+      "epoch": 0.13264129181084197,
+      "grad_norm": 1.6484375,
+      "kd_loss": 0.014401745982468128,
+      "learning_rate": 9.608478616084784e-06,
+      "loss": 0.0329,
+      "step": 460,
+      "student_loss": 0.002872939221560955,
+      "teacher_loss": 0.0004281107976567
+    },
+    {
+      "epoch": 0.13408304498269896,
+      "grad_norm": 1.25,
+      "kd_loss": 0.02247895859181881,
+      "learning_rate": 9.597941882340947e-06,
+      "loss": 0.0409,
+      "step": 465,
+      "student_loss": 0.00037059103487990797,
+      "teacher_loss": 0.0019852325785905123
+    },
+    {
+      "epoch": 0.13552479815455595,
+      "grad_norm": 1.2890625,
+      "kd_loss": 0.011446633376181126,
+      "learning_rate": 9.587271175931313e-06,
+      "loss": 0.0372,
+      "step": 470,
+      "student_loss": 0.00029464755789376795,
+      "teacher_loss": 0.00020733379642479122
+    },
+    {
+      "epoch": 0.13696655132641292,
+      "grad_norm": 2.21875,
+      "kd_loss": 0.016915325075387955,
+      "learning_rate": 9.5764668077739e-06,
+      "loss": 0.0312,
+      "step": 475,
+      "student_loss": 0.0005512129282578826,
+      "teacher_loss": 0.00031623090035282075
+    },
+    {
+      "epoch": 0.1384083044982699,
+      "grad_norm": 1.2890625,
+      "kd_loss": 0.016255535185337067,
+      "learning_rate": 9.5655290926813e-06,
+      "loss": 0.0457,
+      "step": 480,
+      "student_loss": 0.015682687982916832,
+      "teacher_loss": 0.013726728036999702
+    },
+    {
+      "epoch": 0.13985005767012687,
+      "grad_norm": 1.109375,
+      "kd_loss": 0.007256851997226477,
+      "learning_rate": 9.554458349351507e-06,
+      "loss": 0.0341,
+      "step": 485,
+      "student_loss": 0.0005507866735570133,
+      "teacher_loss": 0.0005500169354490936
+    },
+    {
+      "epoch": 0.14129181084198386,
+      "grad_norm": 1.9765625,
+      "kd_loss": 0.018985357135534286,
+      "learning_rate": 9.54325490035863e-06,
+      "loss": 0.0445,
+      "step": 490,
+      "student_loss": 0.015820659697055817,
+      "teacher_loss": 0.00336285843513906
+    },
+    {
+      "epoch": 0.14273356401384082,
+      "grad_norm": 1.875,
+      "kd_loss": 0.025550341233611107,
+      "learning_rate": 9.531919072143488e-06,
+      "loss": 0.0483,
+      "step": 495,
+      "student_loss": 0.026045799255371094,
+      "teacher_loss": 0.004566465504467487
+    },
+    {
+      "epoch": 0.14417531718569782,
+      "grad_norm": 1.640625,
+      "kd_loss": 0.017328884452581406,
+      "learning_rate": 9.520451195004107e-06,
+      "loss": 0.0369,
+      "step": 500,
+      "student_loss": 0.0017473739571869373,
+      "teacher_loss": 0.000323916639899835
+    }
+  ],
+  "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/codi1.5b_a1.0_b1.0_g1.0_ls1/wandb/debug-internal.log

@@ -0,0 +1,11 @@
+{"time":"2026-06-19T21:53:03.29682652+02:00","level":"INFO","msg":"wandb-core"}
+{"time":"2026-06-19T21:53:03.296839186+02:00","level":"INFO","msg":"stream: starting","core version":"0.27.2"}
+{"time":"2026-06-19T21:53:03.42032708+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-19T21:53:03.420348544+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-19T21:53:03.420370789+02:00","level":"INFO","msg":"stream: created new stream","id":"q1wgok6t"}
+{"time":"2026-06-19T21:53:03.420523753+02:00","level":"INFO","msg":"handler: started"}
+{"time":"2026-06-19T21:53:03.421778058+02:00","level":"INFO","msg":"stream: started"}
+{"time":"2026-06-19T21:53:03.421903179+02:00","level":"INFO","msg":"writer: started","stream_id":"q1wgok6t"}
+{"time":"2026-06-19T21:53:03.42193772+02:00","level":"INFO","msg":"sender: started"}
+{"time":"2026-06-19T21:53:03.43943152+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-19T21:53:03.439444571+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/codi1.5b_a1.0_b1.0_g1.0_ls1/wandb/debug.log

@@ -0,0 +1,20 @@
+2026-06-19 21:53:02,866 INFO    MainThread:906312 [wandb_setup.py:_flush():81] Current SDK version is 0.27.2
+2026-06-19 21:53:02,866 INFO    MainThread:906312 [wandb_setup.py:_flush():81] Configure stats pid to 906312
+2026-06-19 21:53:02,866 INFO    MainThread:906312 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-06-19 21:53:02,866 INFO    MainThread:906312 [wandb_init.py:setup_run_log_directory():723] Logging user logs to model_weights/codi-multi-1.5b/wandb/offline-run-20260619_215302-q1wgok6t/logs/debug.log
+2026-06-19 21:53:02,867 INFO    MainThread:906312 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to model_weights/codi-multi-1.5b/wandb/offline-run-20260619_215302-q1wgok6t/logs/debug-internal.log
+2026-06-19 21:53:02,867 INFO    MainThread:906312 [wandb_init.py:init():766] calling init triggers
+2026-06-19 21:53:02,867 INFO    MainThread:906312 [wandb_init.py:init():771] wandb.init called with sweep_config: {}
+config: {'model': 'model_weights/sft-coder-1.5b/checkpoint-6936', 'output_dir': 'model_weights/codi-multi-1.5b', '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': 1.0, 'beta': 1.0, 'gamma': 1.0, 'kd_layers': None, 'single_anchor': False, '_wandb': {}}
+2026-06-19 21:53:02,867 INFO    MainThread:906312 [wandb_init.py:init():814] starting backend
+2026-06-19 21:53:03,283 INFO    MainThread:906312 [wandb_init.py:init():829] sending inform_init request
+2026-06-19 21:53:03,426 INFO    MainThread:906312 [wandb_init.py:init():834] backend started and connected
+2026-06-19 21:53:03,428 INFO    MainThread:906312 [wandb_init.py:init():904] updated telemetry
+2026-06-19 21:53:03,437 INFO    MainThread:906312 [wandb_init.py:init():927] communicating run to backend with 90.0 second timeout
+2026-06-19 21:53:03,440 INFO    MainThread:906312 [wandb_init.py:init():972] starting run threads in backend
+2026-06-19 21:53:03,863 INFO    MainThread:906312 [wandb_run.py:_console_start():2523] atexit reg
+2026-06-19 21:53:03,863 INFO    MainThread:906312 [wandb_run.py:_redirect():2373] redirect: wrap_raw
+2026-06-19 21:53:03,863 INFO    MainThread:906312 [wandb_run.py:_redirect():2442] Wrapping output streams.
+2026-06-19 21:53:03,863 INFO    MainThread:906312 [wandb_run.py:_redirect():2465] Redirects installed.
+2026-06-19 21:53:03,869 INFO    MainThread:906312 [wandb_init.py:init():1010] run started, returning control to user process
+2026-06-19 21:53:08,544 INFO    MainThread:906312 [wandb_run.py:_config_callback():1346] config_cb None None {'output_dir': 'model_weights/codi-multi-1.5b', '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-multi-1.5b/runs/Jun19_21-53-03_alvis4-39', '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/codi1.5b_a1.0_b1.0_g1.0_ls1/wandb/offline-run-20260619_215302-q1wgok6t/files/requirements.txt

@@ -0,0 +1,222 @@
+wheel==0.47.0
+triton==3.3.1
+ptyprocess==0.7.0
+pyzmq==27.1.0
+nvidia-cublas-cu12==12.6.4.1
+python-dateutil==2.9.0.post0
+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
+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
+executing==2.2.1
+gitdb==4.0.12
+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
+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
+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/codi1.5b_a1.0_b1.0_g1.0_ls1/wandb/offline-run-20260619_215302-q1wgok6t/logs/debug-core.log

@@ -0,0 +1,12 @@
+{"time":"2026-06-19T21:53:03.094929463+02:00","level":"INFO","msg":"main: starting server","port-filename":"/local/tmp.6767841/tmptc5jh63i/port-2910938.txt","pid":2910938,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-06-19T21:53:03.095783268+02:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":2910938}
+{"time":"2026-06-19T21:53:03.095779775+02:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/local/tmp.6767841/wandb-2910938-2911302-2012556736/socket","Net":"unix"}}
+{"time":"2026-06-19T21:53:03.10475933+02:00","level":"INFO","msg":"main: starting server","port-filename":"/local/tmp.6767840/tmpzkap0g5c/port-906312.txt","pid":906312,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-06-19T21:53:03.108209915+02:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/local/tmp.6767840/wandb-906312-906672-88537129/socket","Net":"unix"}}
+{"time":"2026-06-19T21:53:03.108404089+02:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":906312}
+{"time":"2026-06-19T21:53:03.276797796+02:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-06-19T21:53:03.2833645+02:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-06-19T21:53:03.287130346+02:00","level":"INFO","msg":"handleInformInit: received","streamId":"txanrr5l","id":"1(@)"}
+{"time":"2026-06-19T21:53:03.293822712+02:00","level":"INFO","msg":"handleInformInit: received","streamId":"q1wgok6t","id":"1(@)"}
+{"time":"2026-06-19T21:53:03.418697302+02:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"txanrr5l","id":"1(@)"}
+{"time":"2026-06-19T21:53:03.421786591+02:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"q1wgok6t","id":"1(@)"}

checkpoints/codi1.5b_a1.0_b1.0_g1.0_ls1/wandb/offline-run-20260619_215302-q1wgok6t/logs/debug-internal.log

@@ -0,0 +1,11 @@
+{"time":"2026-06-19T21:53:03.29682652+02:00","level":"INFO","msg":"wandb-core"}
+{"time":"2026-06-19T21:53:03.296839186+02:00","level":"INFO","msg":"stream: starting","core version":"0.27.2"}
+{"time":"2026-06-19T21:53:03.42032708+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-19T21:53:03.420348544+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-19T21:53:03.420370789+02:00","level":"INFO","msg":"stream: created new stream","id":"q1wgok6t"}
+{"time":"2026-06-19T21:53:03.420523753+02:00","level":"INFO","msg":"handler: started"}
+{"time":"2026-06-19T21:53:03.421778058+02:00","level":"INFO","msg":"stream: started"}
+{"time":"2026-06-19T21:53:03.421903179+02:00","level":"INFO","msg":"writer: started","stream_id":"q1wgok6t"}
+{"time":"2026-06-19T21:53:03.42193772+02:00","level":"INFO","msg":"sender: started"}
+{"time":"2026-06-19T21:53:03.43943152+02:00","level":"WARN","msg":"featurechecker: GraphQL client is nil, skipping feature loading"}
+{"time":"2026-06-19T21:53:03.439444571+02:00","level":"WARN","msg":"runupserter: server does not expand metric globs but the x_server_side_expand_glob_metrics setting is set; ignoring"}