Add codi_recon_delta_1.5b_ls2_rw0.03 checkpoints

.gitattributes

@@ -156,3 +156,9 @@ checkpoints/codi_recon_full_1.5b_ls2_rw0.1/checkpoint-300/tokenizer.json filter=
 checkpoints/codi_recon_full_1.5b_ls2_rw0.1/checkpoint-600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/codi_recon_full_1.5b_ls2_rw0.1/checkpoint-900/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/codi_recon_full_1.5b_ls2_rw0.1/wandb/offline-run-20260627_222021-7dbxalnj/run-7dbxalnj.wandb filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-900/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/wandb/offline-run-20260702_042011-gs3t81f2/run-gs3t81f2.wandb filter=lfs diff=lfs merge=lfs -text

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/added_tokens.json

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

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/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/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/config.json

@@ -0,0 +1,58 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "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,
+  "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": 151936
+}

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8a7de84d0e48b0644dfe0964440cf77eea2ea227e1e4f326328f31fb359f6ee
+size 4539424633

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eedb76566ba563e2b64ece55c65fa6d9d0dc3a88c5753cc2d35db2228467d2cb
+size 3097074235

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4aa03284661460f7588cf2ed51cae6f25c4941e23d7976dea3db0885dadb428
+size 15429

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97017e17d6838d8a09f84b48e28a5938a238a3e50e6a6671b6a2439f52a40fa6
+size 15429

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5a6d9e5e8614e5b7bcfddbab25b719e621f1ff67f400bc9a73dc5af42659928
+size 15429

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:841e9f5079dd6c6b4131e29e2e887ebf784990e17a6d8d8c7131ffb11723d069
+size 15429

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ca01a26e4215a06318d34f4837223ef5e3e71cfd78483b4d43840df556602f8
+size 1465

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/special_tokens_map.json

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

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13cd4f5618f4ea7df29116a4b2f266af065d08840299e335c9d3e1858533c190
+size 9445953

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/tokenizer.json

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

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/tokenizer_config.json

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

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/trainer_state.json

@@ -0,0 +1,2914 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.34594594594594597,
+  "eval_steps": 500,
+  "global_step": 1200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0014414414414414415,
+      "grad_norm": 144.0,
+      "kd_loss": 0.18668268620967865,
+      "learning_rate": 8.88888888888889e-07,
+      "loss": 5.0919,
+      "recon_loss": 11.27502727508545,
+      "recon_trunc": 0.0,
+      "step": 5,
+      "student_loss": 3.2272865772247314,
+      "teacher_loss": 1.4361733198165894
+    },
+    {
+      "epoch": 0.002882882882882883,
+      "grad_norm": 127.5,
+      "kd_loss": 0.20599040389060974,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 4.8166,
+      "recon_loss": 8.36417293548584,
+      "recon_trunc": 0.0,
+      "step": 10,
+      "student_loss": 2.726564407348633,
+      "teacher_loss": 1.1014676094055176
+    },
+    {
+      "epoch": 0.004324324324324324,
+      "grad_norm": 6176.0,
+      "kd_loss": 0.16678810119628906,
+      "learning_rate": 3.1111111111111116e-06,
+      "loss": 4.7608,
+      "recon_loss": 6.041282653808594,
+      "recon_trunc": 0.0,
+      "step": 15,
+      "student_loss": 1.5410536527633667,
+      "teacher_loss": 1.1915134191513062
+    },
+    {
+      "epoch": 0.005765765765765766,
+      "grad_norm": 135.0,
+      "kd_loss": 0.20393367111682892,
+      "learning_rate": 4.222222222222223e-06,
+      "loss": 4.8779,
+      "recon_loss": 5.446779251098633,
+      "recon_trunc": 0.0,
+      "step": 20,
+      "student_loss": 2.177321672439575,
+      "teacher_loss": 0.6389463543891907
+    },
+    {
+      "epoch": 0.007207207207207207,
+      "grad_norm": 253952.0,
+      "kd_loss": 0.22543171048164368,
+      "learning_rate": 5.333333333333334e-06,
+      "loss": 4.5538,
+      "recon_loss": 9.855738639831543,
+      "recon_trunc": 0.0,
+      "step": 25,
+      "student_loss": 2.061642646789551,
+      "teacher_loss": 1.7715281248092651
+    },
+    {
+      "epoch": 0.008648648648648649,
+      "grad_norm": 768.0,
+      "kd_loss": 0.19796203076839447,
+      "learning_rate": 6.444444444444445e-06,
+      "loss": 4.4295,
+      "recon_loss": 5.747690200805664,
+      "recon_trunc": 0.0,
+      "step": 30,
+      "student_loss": 2.255260944366455,
+      "teacher_loss": 1.9419420957565308
+    },
+    {
+      "epoch": 0.01009009009009009,
+      "grad_norm": 7808.0,
+      "kd_loss": 0.17444051802158356,
+      "learning_rate": 7.555555555555556e-06,
+      "loss": 4.3142,
+      "recon_loss": 9.594672203063965,
+      "recon_trunc": 0.0,
+      "step": 35,
+      "student_loss": 2.9477691650390625,
+      "teacher_loss": 2.030805826187134
+    },
+    {
+      "epoch": 0.011531531531531532,
+      "grad_norm": 84.0,
+      "kd_loss": 0.19227176904678345,
+      "learning_rate": 8.666666666666668e-06,
+      "loss": 4.4129,
+      "recon_loss": 6.2831501960754395,
+      "recon_trunc": 0.0,
+      "step": 40,
+      "student_loss": 2.40488338470459,
+      "teacher_loss": 0.9565104246139526
+    },
+    {
+      "epoch": 0.012972972972972972,
+      "grad_norm": 53.5,
+      "kd_loss": 0.16631732881069183,
+      "learning_rate": 9.777777777777779e-06,
+      "loss": 4.2628,
+      "recon_loss": 6.8525390625,
+      "recon_trunc": 0.0,
+      "step": 45,
+      "student_loss": 3.3176517486572266,
+      "teacher_loss": 2.6601927280426025
+    },
+    {
+      "epoch": 0.014414414414414415,
+      "grad_norm": 69.5,
+      "kd_loss": 0.150796577334404,
+      "learning_rate": 9.999813520519569e-06,
+      "loss": 3.4106,
+      "recon_loss": 3.702688455581665,
+      "recon_trunc": 0.0,
+      "step": 50,
+      "student_loss": 1.7257497310638428,
+      "teacher_loss": 1.1742305755615234
+    },
+    {
+      "epoch": 0.015855855855855857,
+      "grad_norm": 35.5,
+      "kd_loss": 0.13769835233688354,
+      "learning_rate": 9.999055971469864e-06,
+      "loss": 3.1423,
+      "recon_loss": 2.6614763736724854,
+      "recon_trunc": 0.0,
+      "step": 55,
+      "student_loss": 1.2244848012924194,
+      "teacher_loss": 0.9719807505607605
+    },
+    {
+      "epoch": 0.017297297297297298,
+      "grad_norm": 32.5,
+      "kd_loss": 0.11417189985513687,
+      "learning_rate": 9.997715786107228e-06,
+      "loss": 2.9039,
+      "recon_loss": 2.063100576400757,
+      "recon_trunc": 0.0,
+      "step": 60,
+      "student_loss": 1.9033191204071045,
+      "teacher_loss": 0.6805936694145203
+    },
+    {
+      "epoch": 0.018738738738738738,
+      "grad_norm": 34.75,
+      "kd_loss": 0.1221928671002388,
+      "learning_rate": 9.995793120629282e-06,
+      "loss": 2.7189,
+      "recon_loss": 2.155364513397217,
+      "recon_trunc": 0.0,
+      "step": 65,
+      "student_loss": 1.172133207321167,
+      "teacher_loss": 0.4049614667892456
+    },
+    {
+      "epoch": 0.02018018018018018,
+      "grad_norm": 36.5,
+      "kd_loss": 0.07467015832662582,
+      "learning_rate": 9.993288199121283e-06,
+      "loss": 2.4826,
+      "recon_loss": 2.2295444011688232,
+      "recon_trunc": 0.0,
+      "step": 70,
+      "student_loss": 1.2911094427108765,
+      "teacher_loss": 1.190304160118103
+    },
+    {
+      "epoch": 0.021621621621621623,
+      "grad_norm": 31.0,
+      "kd_loss": 0.09782560914754868,
+      "learning_rate": 9.990201313529995e-06,
+      "loss": 2.2692,
+      "recon_loss": 2.263787031173706,
+      "recon_trunc": 0.0,
+      "step": 75,
+      "student_loss": 0.8870642185211182,
+      "teacher_loss": 0.38332948088645935
+    },
+    {
+      "epoch": 0.023063063063063063,
+      "grad_norm": 31.25,
+      "kd_loss": 0.08544448763132095,
+      "learning_rate": 9.986532823629675e-06,
+      "loss": 1.9822,
+      "recon_loss": 2.109919548034668,
+      "recon_trunc": 0.0,
+      "step": 80,
+      "student_loss": 1.538913607597351,
+      "teacher_loss": 0.5950594544410706
+    },
+    {
+      "epoch": 0.024504504504504504,
+      "grad_norm": 30.5,
+      "kd_loss": 0.06714422255754471,
+      "learning_rate": 9.982283156980133e-06,
+      "loss": 1.7643,
+      "recon_loss": 3.1866095066070557,
+      "recon_trunc": 0.0,
+      "step": 85,
+      "student_loss": 1.037903904914856,
+      "teacher_loss": 0.8964719176292419
+    },
+    {
+      "epoch": 0.025945945945945945,
+      "grad_norm": 29.25,
+      "kd_loss": 0.06367570161819458,
+      "learning_rate": 9.977452808876907e-06,
+      "loss": 1.5629,
+      "recon_loss": 3.314178466796875,
+      "recon_trunc": 0.0,
+      "step": 90,
+      "student_loss": 0.7879223227500916,
+      "teacher_loss": 0.6495854258537292
+    },
+    {
+      "epoch": 0.02738738738738739,
+      "grad_norm": 25.5,
+      "kd_loss": 0.05303771421313286,
+      "learning_rate": 9.972042342293531e-06,
+      "loss": 1.405,
+      "recon_loss": 2.446079730987549,
+      "recon_trunc": 0.0,
+      "step": 95,
+      "student_loss": 0.6456317901611328,
+      "teacher_loss": 0.5170290470123291
+    },
+    {
+      "epoch": 0.02882882882882883,
+      "grad_norm": 23.875,
+      "kd_loss": 0.05943922698497772,
+      "learning_rate": 9.966052387815923e-06,
+      "loss": 1.2091,
+      "recon_loss": 1.8016548156738281,
+      "recon_trunc": 0.0,
+      "step": 100,
+      "student_loss": 0.6617249250411987,
+      "teacher_loss": 0.47058895230293274
+    },
+    {
+      "epoch": 0.03027027027027027,
+      "grad_norm": 20.375,
+      "kd_loss": 0.0777481347322464,
+      "learning_rate": 9.959483643568893e-06,
+      "loss": 1.0222,
+      "recon_loss": 1.344748616218567,
+      "recon_trunc": 0.0,
+      "step": 105,
+      "student_loss": 0.8499628901481628,
+      "teacher_loss": 0.5487883687019348
+    },
+    {
+      "epoch": 0.031711711711711714,
+      "grad_norm": 14.625,
+      "kd_loss": 0.0892779678106308,
+      "learning_rate": 9.95233687513477e-06,
+      "loss": 0.9259,
+      "recon_loss": 2.6645045280456543,
+      "recon_trunc": 0.0,
+      "step": 110,
+      "student_loss": 0.4784584939479828,
+      "teacher_loss": 0.21723365783691406
+    },
+    {
+      "epoch": 0.033153153153153155,
+      "grad_norm": 12.6875,
+      "kd_loss": 0.05595879256725311,
+      "learning_rate": 9.944612915464183e-06,
+      "loss": 0.8526,
+      "recon_loss": 1.4303338527679443,
+      "recon_trunc": 0.0,
+      "step": 115,
+      "student_loss": 0.42129674553871155,
+      "teacher_loss": 0.34839943051338196
+    },
+    {
+      "epoch": 0.034594594594594595,
+      "grad_norm": 9.75,
+      "kd_loss": 0.04541080817580223,
+      "learning_rate": 9.936312664778978e-06,
+      "loss": 0.8036,
+      "recon_loss": 1.4835879802703857,
+      "recon_trunc": 0.0,
+      "step": 120,
+      "student_loss": 0.2531515657901764,
+      "teacher_loss": 0.21365374326705933
+    },
+    {
+      "epoch": 0.036036036036036036,
+      "grad_norm": 6.78125,
+      "kd_loss": 0.04539714753627777,
+      "learning_rate": 9.92743709046729e-06,
+      "loss": 0.7758,
+      "recon_loss": 2.6002440452575684,
+      "recon_trunc": 0.0,
+      "step": 125,
+      "student_loss": 0.5731778740882874,
+      "teacher_loss": 0.4177762567996979
+    },
+    {
+      "epoch": 0.037477477477477476,
+      "grad_norm": 6.5,
+      "kd_loss": 0.0358523465692997,
+      "learning_rate": 9.917987226970811e-06,
+      "loss": 0.6983,
+      "recon_loss": 3.8052804470062256,
+      "recon_trunc": 0.0,
+      "step": 130,
+      "student_loss": 0.29267042875289917,
+      "teacher_loss": 0.2587020695209503
+    },
+    {
+      "epoch": 0.03891891891891892,
+      "grad_norm": 6.6875,
+      "kd_loss": 0.04369649663567543,
+      "learning_rate": 9.907964175664207e-06,
+      "loss": 0.6848,
+      "recon_loss": 1.7037179470062256,
+      "recon_trunc": 0.0,
+      "step": 135,
+      "student_loss": 0.31005415320396423,
+      "teacher_loss": 0.16338767111301422
+    },
+    {
+      "epoch": 0.04036036036036036,
+      "grad_norm": 8.8125,
+      "kd_loss": 0.030473196879029274,
+      "learning_rate": 9.897369104726772e-06,
+      "loss": 0.6438,
+      "recon_loss": 2.9579691886901855,
+      "recon_trunc": 0.0,
+      "step": 140,
+      "student_loss": 0.5350664258003235,
+      "teacher_loss": 0.45312559604644775
+    },
+    {
+      "epoch": 0.041801801801801805,
+      "grad_norm": 10.5,
+      "kd_loss": 0.04736681282520294,
+      "learning_rate": 9.886203249006265e-06,
+      "loss": 0.6417,
+      "recon_loss": 1.3402631282806396,
+      "recon_trunc": 0.0,
+      "step": 145,
+      "student_loss": 0.13968625664710999,
+      "teacher_loss": 0.13039764761924744
+    },
+    {
+      "epoch": 0.043243243243243246,
+      "grad_norm": 8.5625,
+      "kd_loss": 0.035465020686388016,
+      "learning_rate": 9.874467909874994e-06,
+      "loss": 0.5845,
+      "recon_loss": 1.7700623273849487,
+      "recon_trunc": 0.0,
+      "step": 150,
+      "student_loss": 0.346703439950943,
+      "teacher_loss": 0.16136543452739716
+    },
+    {
+      "epoch": 0.044684684684684686,
+      "grad_norm": 9.0625,
+      "kd_loss": 0.04292791336774826,
+      "learning_rate": 9.862164455078141e-06,
+      "loss": 0.5625,
+      "recon_loss": 1.3899339437484741,
+      "recon_trunc": 0.0,
+      "step": 155,
+      "student_loss": 0.25330430269241333,
+      "teacher_loss": 0.15239955484867096
+    },
+    {
+      "epoch": 0.04612612612612613,
+      "grad_norm": 8.375,
+      "kd_loss": 0.04669129103422165,
+      "learning_rate": 9.849294318574353e-06,
+      "loss": 0.5287,
+      "recon_loss": 2.322531223297119,
+      "recon_trunc": 0.0,
+      "step": 160,
+      "student_loss": 0.2261267602443695,
+      "teacher_loss": 0.08589927107095718
+    },
+    {
+      "epoch": 0.04756756756756757,
+      "grad_norm": 7.90625,
+      "kd_loss": 0.034078288823366165,
+      "learning_rate": 9.835859000368615e-06,
+      "loss": 0.4944,
+      "recon_loss": 2.0721518993377686,
+      "recon_trunc": 0.0,
+      "step": 165,
+      "student_loss": 0.2308652698993683,
+      "teacher_loss": 0.14940541982650757
+    },
+    {
+      "epoch": 0.04900900900900901,
+      "grad_norm": 9.1875,
+      "kd_loss": 0.042618874460458755,
+      "learning_rate": 9.821860066337427e-06,
+      "loss": 0.4743,
+      "recon_loss": 1.079853892326355,
+      "recon_trunc": 0.0,
+      "step": 170,
+      "student_loss": 0.13290715217590332,
+      "teacher_loss": 0.06264088302850723
+    },
+    {
+      "epoch": 0.05045045045045045,
+      "grad_norm": 6.71875,
+      "kd_loss": 0.038317374885082245,
+      "learning_rate": 9.807299148046301e-06,
+      "loss": 0.4088,
+      "recon_loss": 1.7100510597229004,
+      "recon_trunc": 0.1666666716337204,
+      "step": 175,
+      "student_loss": 0.2707020938396454,
+      "teacher_loss": 0.3391987979412079
+    },
+    {
+      "epoch": 0.05189189189189189,
+      "grad_norm": 7.71875,
+      "kd_loss": 0.03604669123888016,
+      "learning_rate": 9.792177942559598e-06,
+      "loss": 0.3935,
+      "recon_loss": 1.3424031734466553,
+      "recon_trunc": 0.0,
+      "step": 180,
+      "student_loss": 0.10768720507621765,
+      "teacher_loss": 0.05560263246297836
+    },
+    {
+      "epoch": 0.05333333333333334,
+      "grad_norm": 5.875,
+      "kd_loss": 0.046635184437036514,
+      "learning_rate": 9.776498212242748e-06,
+      "loss": 0.3469,
+      "recon_loss": 1.495619535446167,
+      "recon_trunc": 0.0,
+      "step": 185,
+      "student_loss": 0.18289713561534882,
+      "teacher_loss": 0.05616335570812225
+    },
+    {
+      "epoch": 0.05477477477477478,
+      "grad_norm": 5.78125,
+      "kd_loss": 0.04560407996177673,
+      "learning_rate": 9.76026178455684e-06,
+      "loss": 0.3421,
+      "recon_loss": 1.6703602075576782,
+      "recon_trunc": 0.0416666679084301,
+      "step": 190,
+      "student_loss": 0.10543839633464813,
+      "teacher_loss": 0.09451761096715927
+    },
+    {
+      "epoch": 0.05621621621621622,
+      "grad_norm": 4.46875,
+      "kd_loss": 0.023925229907035828,
+      "learning_rate": 9.743470551845625e-06,
+      "loss": 0.2909,
+      "recon_loss": 1.3105652332305908,
+      "recon_trunc": 0.0,
+      "step": 195,
+      "student_loss": 0.09683974087238312,
+      "teacher_loss": 0.060805100947618484
+    },
+    {
+      "epoch": 0.05765765765765766,
+      "grad_norm": 4.625,
+      "kd_loss": 0.020097926259040833,
+      "learning_rate": 9.726126471114985e-06,
+      "loss": 0.273,
+      "recon_loss": 1.6309553384780884,
+      "recon_trunc": 0.0,
+      "step": 200,
+      "student_loss": 0.1421128213405609,
+      "teacher_loss": 0.1302688717842102
+    },
+    {
+      "epoch": 0.0590990990990991,
+      "grad_norm": 6.53125,
+      "kd_loss": 0.03434700518846512,
+      "learning_rate": 9.708231563804828e-06,
+      "loss": 0.2626,
+      "recon_loss": 1.0981206893920898,
+      "recon_trunc": 0.0,
+      "step": 205,
+      "student_loss": 0.03550760820508003,
+      "teacher_loss": 0.03432264178991318
+    },
+    {
+      "epoch": 0.06054054054054054,
+      "grad_norm": 4.03125,
+      "kd_loss": 0.0490451417863369,
+      "learning_rate": 9.689787915553497e-06,
+      "loss": 0.2458,
+      "recon_loss": 1.5271836519241333,
+      "recon_trunc": 0.021276595070958138,
+      "step": 210,
+      "student_loss": 0.10633832961320877,
+      "teacher_loss": 0.04522276669740677
+    },
+    {
+      "epoch": 0.06198198198198198,
+      "grad_norm": 3.5625,
+      "kd_loss": 0.030574269592761993,
+      "learning_rate": 9.670797675954693e-06,
+      "loss": 0.2322,
+      "recon_loss": 1.0532352924346924,
+      "recon_trunc": 0.0,
+      "step": 215,
+      "student_loss": 0.03432159870862961,
+      "teacher_loss": 0.029644442722201347
+    },
+    {
+      "epoch": 0.06342342342342343,
+      "grad_norm": 3.578125,
+      "kd_loss": 0.02381826564669609,
+      "learning_rate": 9.651263058306932e-06,
+      "loss": 0.2022,
+      "recon_loss": 1.1155163049697876,
+      "recon_trunc": 0.0,
+      "step": 220,
+      "student_loss": 0.03244610130786896,
+      "teacher_loss": 0.03664662688970566
+    },
+    {
+      "epoch": 0.06486486486486487,
+      "grad_norm": 4.0625,
+      "kd_loss": 0.04193637892603874,
+      "learning_rate": 9.631186339355599e-06,
+      "loss": 0.1995,
+      "recon_loss": 1.2504764795303345,
+      "recon_trunc": 0.0,
+      "step": 225,
+      "student_loss": 0.052963703870773315,
+      "teacher_loss": 0.042675312608480453
+    },
+    {
+      "epoch": 0.06630630630630631,
+      "grad_norm": 3.71875,
+      "kd_loss": 0.022358015179634094,
+      "learning_rate": 9.610569859027585e-06,
+      "loss": 0.2074,
+      "recon_loss": 1.3408961296081543,
+      "recon_trunc": 0.0,
+      "step": 230,
+      "student_loss": 0.06527604162693024,
+      "teacher_loss": 0.06690996885299683
+    },
+    {
+      "epoch": 0.06774774774774775,
+      "grad_norm": 3.953125,
+      "kd_loss": 0.04554065316915512,
+      "learning_rate": 9.589416020158577e-06,
+      "loss": 0.2046,
+      "recon_loss": 1.0968328714370728,
+      "recon_trunc": 0.0,
+      "step": 235,
+      "student_loss": 0.037662506103515625,
+      "teacher_loss": 0.01968599110841751
+    },
+    {
+      "epoch": 0.06918918918918919,
+      "grad_norm": 3.609375,
+      "kd_loss": 0.03338759392499924,
+      "learning_rate": 9.567727288213005e-06,
+      "loss": 0.1812,
+      "recon_loss": 1.7976175546646118,
+      "recon_trunc": 0.0,
+      "step": 240,
+      "student_loss": 0.06893198937177658,
+      "teacher_loss": 0.03559521958231926
+    },
+    {
+      "epoch": 0.07063063063063063,
+      "grad_norm": 3.171875,
+      "kd_loss": 0.01759061962366104,
+      "learning_rate": 9.545506190996695e-06,
+      "loss": 0.169,
+      "recon_loss": 1.6089872121810913,
+      "recon_trunc": 0.0,
+      "step": 245,
+      "student_loss": 0.041334331035614014,
+      "teacher_loss": 0.030137302353978157
+    },
+    {
+      "epoch": 0.07207207207207207,
+      "grad_norm": 3.609375,
+      "kd_loss": 0.03067457303404808,
+      "learning_rate": 9.52275531836226e-06,
+      "loss": 0.1498,
+      "recon_loss": 1.2096911668777466,
+      "recon_trunc": 0.0,
+      "step": 250,
+      "student_loss": 0.07505633682012558,
+      "teacher_loss": 0.07239583134651184
+    },
+    {
+      "epoch": 0.07351351351351351,
+      "grad_norm": 3.15625,
+      "kd_loss": 0.02453661896288395,
+      "learning_rate": 9.499477321907243e-06,
+      "loss": 0.1445,
+      "recon_loss": 1.6908211708068848,
+      "recon_trunc": 0.0,
+      "step": 255,
+      "student_loss": 0.016758263111114502,
+      "teacher_loss": 0.013908207416534424
+    },
+    {
+      "epoch": 0.07495495495495495,
+      "grad_norm": 3.171875,
+      "kd_loss": 0.02927604876458645,
+      "learning_rate": 9.475674914665084e-06,
+      "loss": 0.1595,
+      "recon_loss": 1.4746726751327515,
+      "recon_trunc": 0.0,
+      "step": 260,
+      "student_loss": 0.05420182645320892,
+      "teacher_loss": 0.02283495105803013
+    },
+    {
+      "epoch": 0.0763963963963964,
+      "grad_norm": 3.125,
+      "kd_loss": 0.030070308595895767,
+      "learning_rate": 9.451350870788922e-06,
+      "loss": 0.1621,
+      "recon_loss": 1.2860867977142334,
+      "recon_trunc": 0.0,
+      "step": 265,
+      "student_loss": 0.041329968720674515,
+      "teacher_loss": 0.007841577753424644
+    },
+    {
+      "epoch": 0.07783783783783783,
+      "grad_norm": 3.234375,
+      "kd_loss": 0.01138228364288807,
+      "learning_rate": 9.426508025228255e-06,
+      "loss": 0.156,
+      "recon_loss": 1.46453857421875,
+      "recon_trunc": 0.0,
+      "step": 270,
+      "student_loss": 0.025395482778549194,
+      "teacher_loss": 0.024079594761133194
+    },
+    {
+      "epoch": 0.07927927927927927,
+      "grad_norm": 1.953125,
+      "kd_loss": 0.012232596054673195,
+      "learning_rate": 9.401149273398547e-06,
+      "loss": 0.1578,
+      "recon_loss": 1.163053035736084,
+      "recon_trunc": 0.0,
+      "step": 275,
+      "student_loss": 0.010497995652258396,
+      "teacher_loss": 0.009640091098845005
+    },
+    {
+      "epoch": 0.08072072072072072,
+      "grad_norm": 2.375,
+      "kd_loss": 0.04322246462106705,
+      "learning_rate": 9.37527757084375e-06,
+      "loss": 0.1418,
+      "recon_loss": 2.4887444972991943,
+      "recon_trunc": 0.013513513840734959,
+      "step": 280,
+      "student_loss": 0.041192568838596344,
+      "teacher_loss": 0.020611625164747238
+    },
+    {
+      "epoch": 0.08216216216216216,
+      "grad_norm": 4.0,
+      "kd_loss": 0.0165244247764349,
+      "learning_rate": 9.348895932891857e-06,
+      "loss": 0.1369,
+      "recon_loss": 1.9490307569503784,
+      "recon_trunc": 0.0,
+      "step": 285,
+      "student_loss": 0.06232248619198799,
+      "teacher_loss": 0.08218289911746979
+    },
+    {
+      "epoch": 0.08360360360360361,
+      "grad_norm": 3.53125,
+      "kd_loss": 0.01325551699846983,
+      "learning_rate": 9.322007434303449e-06,
+      "loss": 0.1411,
+      "recon_loss": 3.0142760276794434,
+      "recon_trunc": 0.0,
+      "step": 290,
+      "student_loss": 0.014739612117409706,
+      "teacher_loss": 0.014967009425163269
+    },
+    {
+      "epoch": 0.08504504504504505,
+      "grad_norm": 3.359375,
+      "kd_loss": 0.03525789454579353,
+      "learning_rate": 9.29461520891335e-06,
+      "loss": 0.1508,
+      "recon_loss": 1.8346909284591675,
+      "recon_trunc": 0.0,
+      "step": 295,
+      "student_loss": 0.05333559215068817,
+      "teacher_loss": 0.01834472082555294
+    },
+    {
+      "epoch": 0.08648648648648649,
+      "grad_norm": 2.78125,
+      "kd_loss": 0.020346438512206078,
+      "learning_rate": 9.266722449265371e-06,
+      "loss": 0.1323,
+      "recon_loss": 1.0011640787124634,
+      "recon_trunc": 0.0,
+      "step": 300,
+      "student_loss": 0.11178486794233322,
+      "teacher_loss": 0.10985656827688217
+    },
+    {
+      "epoch": 0.08792792792792793,
+      "grad_norm": 2.3125,
+      "kd_loss": 0.023063305765390396,
+      "learning_rate": 9.238332406240224e-06,
+      "loss": 0.1389,
+      "recon_loss": 1.3658217191696167,
+      "recon_trunc": 0.0,
+      "step": 305,
+      "student_loss": 0.09694303572177887,
+      "teacher_loss": 0.07988587766885757
+    },
+    {
+      "epoch": 0.08936936936936937,
+      "grad_norm": 2.0625,
+      "kd_loss": 0.014214077033102512,
+      "learning_rate": 9.209448388676636e-06,
+      "loss": 0.1389,
+      "recon_loss": 1.2214659452438354,
+      "recon_trunc": 0.0,
+      "step": 310,
+      "student_loss": 0.006524513941258192,
+      "teacher_loss": 0.02097470313310623
+    },
+    {
+      "epoch": 0.09081081081081081,
+      "grad_norm": 1.9140625,
+      "kd_loss": 0.01485999021679163,
+      "learning_rate": 9.180073762985701e-06,
+      "loss": 0.1395,
+      "recon_loss": 4.162618160247803,
+      "recon_trunc": 0.0,
+      "step": 315,
+      "student_loss": 0.04494684562087059,
+      "teacher_loss": 0.031888674944639206
+    },
+    {
+      "epoch": 0.09225225225225225,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.022917235270142555,
+      "learning_rate": 9.15021195275854e-06,
+      "loss": 0.1285,
+      "recon_loss": 1.1853079795837402,
+      "recon_trunc": 0.0,
+      "step": 320,
+      "student_loss": 0.021672768518328667,
+      "teacher_loss": 0.014209315180778503
+    },
+    {
+      "epoch": 0.0936936936936937,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.022615715861320496,
+      "learning_rate": 9.119866438367263e-06,
+      "loss": 0.1394,
+      "recon_loss": 1.818947196006775,
+      "recon_trunc": 0.0,
+      "step": 325,
+      "student_loss": 0.028340045362710953,
+      "teacher_loss": 0.013725091703236103
+    },
+    {
+      "epoch": 0.09513513513513513,
+      "grad_norm": 2.296875,
+      "kd_loss": 0.011119124479591846,
+      "learning_rate": 9.089040756559352e-06,
+      "loss": 0.1317,
+      "recon_loss": 2.282118558883667,
+      "recon_trunc": 0.0,
+      "step": 330,
+      "student_loss": 0.03408842533826828,
+      "teacher_loss": 0.04678163677453995
+    },
+    {
+      "epoch": 0.09657657657657658,
+      "grad_norm": 2.265625,
+      "kd_loss": 0.01260451227426529,
+      "learning_rate": 9.057738500045453e-06,
+      "loss": 0.1314,
+      "recon_loss": 0.6638595461845398,
+      "recon_trunc": 0.0,
+      "step": 335,
+      "student_loss": 0.019066525623202324,
+      "teacher_loss": 0.014849068596959114
+    },
+    {
+      "epoch": 0.09801801801801802,
+      "grad_norm": 2.5625,
+      "kd_loss": 0.02786502055823803,
+      "learning_rate": 9.025963317080641e-06,
+      "loss": 0.1279,
+      "recon_loss": 1.0848702192306519,
+      "recon_trunc": 0.0,
+      "step": 340,
+      "student_loss": 0.008073833771049976,
+      "teacher_loss": 0.005774135701358318
+    },
+    {
+      "epoch": 0.09945945945945946,
+      "grad_norm": 2.828125,
+      "kd_loss": 0.022951554507017136,
+      "learning_rate": 8.993718911039221e-06,
+      "loss": 0.1183,
+      "recon_loss": 1.0540446043014526,
+      "recon_trunc": 0.0,
+      "step": 345,
+      "student_loss": 0.0063321711495518684,
+      "teacher_loss": 0.006896166130900383
+    },
+    {
+      "epoch": 0.1009009009009009,
+      "grad_norm": 2.203125,
+      "kd_loss": 0.03470675274729729,
+      "learning_rate": 8.96100903998311e-06,
+      "loss": 0.1377,
+      "recon_loss": 1.5478863716125488,
+      "recon_trunc": 0.0,
+      "step": 350,
+      "student_loss": 0.04314170405268669,
+      "teacher_loss": 0.013373729772865772
+    },
+    {
+      "epoch": 0.10234234234234234,
+      "grad_norm": 3.296875,
+      "kd_loss": 0.0213958527892828,
+      "learning_rate": 8.927837516223824e-06,
+      "loss": 0.145,
+      "recon_loss": 0.774444580078125,
+      "recon_trunc": 0.0,
+      "step": 355,
+      "student_loss": 0.02257266826927662,
+      "teacher_loss": 0.016409652307629585
+    },
+    {
+      "epoch": 0.10378378378378378,
+      "grad_norm": 2.5625,
+      "kd_loss": 0.024546559900045395,
+      "learning_rate": 8.89420820587817e-06,
+      "loss": 0.1346,
+      "recon_loss": 1.2082504034042358,
+      "recon_trunc": 0.0,
+      "step": 360,
+      "student_loss": 0.04384709522128105,
+      "teacher_loss": 0.04615149646997452
+    },
+    {
+      "epoch": 0.10522522522522522,
+      "grad_norm": 1.859375,
+      "kd_loss": 0.03939265385270119,
+      "learning_rate": 8.86012502841764e-06,
+      "loss": 0.1226,
+      "recon_loss": 1.1720443964004517,
+      "recon_trunc": 0.01785714365541935,
+      "step": 365,
+      "student_loss": 0.05701206624507904,
+      "teacher_loss": 0.02054421976208687
+    },
+    {
+      "epoch": 0.10666666666666667,
+      "grad_norm": 1.6171875,
+      "kd_loss": 0.010782319121062756,
+      "learning_rate": 8.825591956211614e-06,
+      "loss": 0.114,
+      "recon_loss": 1.5270192623138428,
+      "recon_trunc": 0.0,
+      "step": 370,
+      "student_loss": 0.009973138570785522,
+      "teacher_loss": 0.010867908596992493
+    },
+    {
+      "epoch": 0.10810810810810811,
+      "grad_norm": 2.609375,
+      "kd_loss": 0.023110082373023033,
+      "learning_rate": 8.790613014064371e-06,
+      "loss": 0.128,
+      "recon_loss": 1.2032126188278198,
+      "recon_trunc": 0.0,
+      "step": 375,
+      "student_loss": 0.02264533005654812,
+      "teacher_loss": 0.0033990642987191677
+    },
+    {
+      "epoch": 0.10954954954954955,
+      "grad_norm": 2.015625,
+      "kd_loss": 0.02052084170281887,
+      "learning_rate": 8.755192278746e-06,
+      "loss": 0.1281,
+      "recon_loss": 1.5262211561203003,
+      "recon_trunc": 0.0,
+      "step": 380,
+      "student_loss": 0.1342981606721878,
+      "teacher_loss": 0.11457706987857819
+    },
+    {
+      "epoch": 0.110990990990991,
+      "grad_norm": 2.1875,
+      "kd_loss": 0.01637507975101471,
+      "learning_rate": 8.719333878517274e-06,
+      "loss": 0.1278,
+      "recon_loss": 1.517993688583374,
+      "recon_trunc": 0.0,
+      "step": 385,
+      "student_loss": 0.003994242288172245,
+      "teacher_loss": 0.003780895145609975
+    },
+    {
+      "epoch": 0.11243243243243244,
+      "grad_norm": 3.125,
+      "kd_loss": 0.01132834143936634,
+      "learning_rate": 8.683041992648474e-06,
+      "loss": 0.1264,
+      "recon_loss": 0.751715898513794,
+      "recon_trunc": 0.0,
+      "step": 390,
+      "student_loss": 0.00670157466083765,
+      "teacher_loss": 0.005965149030089378
+    },
+    {
+      "epoch": 0.11387387387387388,
+      "grad_norm": 2.125,
+      "kd_loss": 0.028760867193341255,
+      "learning_rate": 8.646320850932326e-06,
+      "loss": 0.1065,
+      "recon_loss": 0.6950639486312866,
+      "recon_trunc": 0.0,
+      "step": 395,
+      "student_loss": 0.02178291417658329,
+      "teacher_loss": 0.003591229673475027
+    },
+    {
+      "epoch": 0.11531531531531532,
+      "grad_norm": 1.734375,
+      "kd_loss": 0.022613229230046272,
+      "learning_rate": 8.609174733191012e-06,
+      "loss": 0.1199,
+      "recon_loss": 2.425037145614624,
+      "recon_trunc": 0.0,
+      "step": 400,
+      "student_loss": 0.03343844413757324,
+      "teacher_loss": 0.03374646604061127
+    },
+    {
+      "epoch": 0.11675675675675676,
+      "grad_norm": 2.46875,
+      "kd_loss": 0.008865836076438427,
+      "learning_rate": 8.571607968777349e-06,
+      "loss": 0.1292,
+      "recon_loss": 1.5487998723983765,
+      "recon_trunc": 0.0,
+      "step": 405,
+      "student_loss": 0.08472836017608643,
+      "teacher_loss": 0.08174186944961548
+    },
+    {
+      "epoch": 0.1181981981981982,
+      "grad_norm": 2.3125,
+      "kd_loss": 0.02681361325085163,
+      "learning_rate": 8.533624936070225e-06,
+      "loss": 0.1081,
+      "recon_loss": 1.6724849939346313,
+      "recon_trunc": 0.0,
+      "step": 410,
+      "student_loss": 0.05861333757638931,
+      "teacher_loss": 0.028042757883667946
+    },
+    {
+      "epoch": 0.11963963963963964,
+      "grad_norm": 1.8203125,
+      "kd_loss": 0.011062159202992916,
+      "learning_rate": 8.495230061964289e-06,
+      "loss": 0.1142,
+      "recon_loss": 0.9828051328659058,
+      "recon_trunc": 0.0,
+      "step": 415,
+      "student_loss": 0.021241258829832077,
+      "teacher_loss": 0.01005750522017479
+    },
+    {
+      "epoch": 0.12108108108108108,
+      "grad_norm": 2.265625,
+      "kd_loss": 0.016577551141381264,
+      "learning_rate": 8.456427821354001e-06,
+      "loss": 0.1154,
+      "recon_loss": 0.7684392929077148,
+      "recon_trunc": 0.0,
+      "step": 420,
+      "student_loss": 0.008126895874738693,
+      "teacher_loss": 0.0070959595032036304
+    },
+    {
+      "epoch": 0.12252252252252252,
+      "grad_norm": 2.171875,
+      "kd_loss": 0.01684112846851349,
+      "learning_rate": 8.417222736612092e-06,
+      "loss": 0.1155,
+      "recon_loss": 2.551865339279175,
+      "recon_trunc": 0.0,
+      "step": 425,
+      "student_loss": 0.029610300436615944,
+      "teacher_loss": 0.029854409396648407
+    },
+    {
+      "epoch": 0.12396396396396396,
+      "grad_norm": 2.5,
+      "kd_loss": 0.021863555535674095,
+      "learning_rate": 8.377619377062483e-06,
+      "loss": 0.1125,
+      "recon_loss": 0.8800704479217529,
+      "recon_trunc": 0.0,
+      "step": 430,
+      "student_loss": 0.006858013104647398,
+      "teacher_loss": 0.0050558242946863174
+    },
+    {
+      "epoch": 0.1254054054054054,
+      "grad_norm": 2.109375,
+      "kd_loss": 0.025641009211540222,
+      "learning_rate": 8.337622358447728e-06,
+      "loss": 0.11,
+      "recon_loss": 1.5023645162582397,
+      "recon_trunc": 0.0,
+      "step": 435,
+      "student_loss": 0.05605771392583847,
+      "teacher_loss": 0.05006556212902069
+    },
+    {
+      "epoch": 0.12684684684684686,
+      "grad_norm": 2.4375,
+      "kd_loss": 0.016024881973862648,
+      "learning_rate": 8.29723634239106e-06,
+      "loss": 0.1123,
+      "recon_loss": 1.9361077547073364,
+      "recon_trunc": 0.0,
+      "step": 440,
+      "student_loss": 0.005403150338679552,
+      "teacher_loss": 0.003332361578941345
+    },
+    {
+      "epoch": 0.12828828828828828,
+      "grad_norm": 3.125,
+      "kd_loss": 0.023422684520483017,
+      "learning_rate": 8.256466035853077e-06,
+      "loss": 0.1125,
+      "recon_loss": 1.2101949453353882,
+      "recon_trunc": 0.0,
+      "step": 445,
+      "student_loss": 0.009501632302999496,
+      "teacher_loss": 0.004478491377085447
+    },
+    {
+      "epoch": 0.12972972972972974,
+      "grad_norm": 3.078125,
+      "kd_loss": 0.02115996554493904,
+      "learning_rate": 8.215316190583153e-06,
+      "loss": 0.1223,
+      "recon_loss": 0.6691838502883911,
+      "recon_trunc": 0.0,
+      "step": 450,
+      "student_loss": 0.018543338403105736,
+      "teacher_loss": 0.018817907199263573
+    },
+    {
+      "epoch": 0.13117117117117116,
+      "grad_norm": 1.7421875,
+      "kd_loss": 0.011057443916797638,
+      "learning_rate": 8.173791602565622e-06,
+      "loss": 0.1278,
+      "recon_loss": 2.2723820209503174,
+      "recon_trunc": 0.0,
+      "step": 455,
+      "student_loss": 0.009742753580212593,
+      "teacher_loss": 0.01030263677239418
+    },
+    {
+      "epoch": 0.13261261261261262,
+      "grad_norm": 1.734375,
+      "kd_loss": 0.009716923348605633,
+      "learning_rate": 8.13189711146081e-06,
+      "loss": 0.1068,
+      "recon_loss": 1.779807448387146,
+      "recon_trunc": 0.0,
+      "step": 460,
+      "student_loss": 0.005108564160764217,
+      "teacher_loss": 0.00407342379912734
+    },
+    {
+      "epoch": 0.13405405405405404,
+      "grad_norm": 2.5625,
+      "kd_loss": 0.03107694536447525,
+      "learning_rate": 8.08963760004098e-06,
+      "loss": 0.1273,
+      "recon_loss": 2.5726194381713867,
+      "recon_trunc": 0.0,
+      "step": 465,
+      "student_loss": 0.056230369955301285,
+      "teacher_loss": 0.04770596697926521
+    },
+    {
+      "epoch": 0.1354954954954955,
+      "grad_norm": 2.421875,
+      "kd_loss": 0.01239880919456482,
+      "learning_rate": 8.047017993621242e-06,
+      "loss": 0.1193,
+      "recon_loss": 1.7926909923553467,
+      "recon_trunc": 0.0,
+      "step": 470,
+      "student_loss": 0.0654020830988884,
+      "teacher_loss": 0.06957650929689407
+    },
+    {
+      "epoch": 0.13693693693693693,
+      "grad_norm": 2.453125,
+      "kd_loss": 0.015770934522151947,
+      "learning_rate": 8.004043259485519e-06,
+      "loss": 0.1142,
+      "recon_loss": 1.5135294198989868,
+      "recon_trunc": 0.0,
+      "step": 475,
+      "student_loss": 0.017355499789118767,
+      "teacher_loss": 0.014955965802073479
+    },
+    {
+      "epoch": 0.13837837837837838,
+      "grad_norm": 2.875,
+      "kd_loss": 0.03178933635354042,
+      "learning_rate": 7.960718406307602e-06,
+      "loss": 0.1184,
+      "recon_loss": 1.991554856300354,
+      "recon_trunc": 0.1666666716337204,
+      "step": 480,
+      "student_loss": 0.033552687615156174,
+      "teacher_loss": 0.06758459657430649
+    },
+    {
+      "epoch": 0.1398198198198198,
+      "grad_norm": 3.4375,
+      "kd_loss": 0.016763880848884583,
+      "learning_rate": 7.917048483567408e-06,
+      "loss": 0.1147,
+      "recon_loss": 3.2356066703796387,
+      "recon_trunc": 0.0,
+      "step": 485,
+      "student_loss": 0.02738153003156185,
+      "teacher_loss": 0.008060683496296406
+    },
+    {
+      "epoch": 0.14126126126126126,
+      "grad_norm": 2.140625,
+      "kd_loss": 0.016219064593315125,
+      "learning_rate": 7.873038580962453e-06,
+      "loss": 0.1159,
+      "recon_loss": 1.9894565343856812,
+      "recon_trunc": 0.0,
+      "step": 490,
+      "student_loss": 0.02026814967393875,
+      "teacher_loss": 0.022603487595915794
+    },
+    {
+      "epoch": 0.14270270270270272,
+      "grad_norm": 1.9765625,
+      "kd_loss": 0.02329091913998127,
+      "learning_rate": 7.828693827814662e-06,
+      "loss": 0.1082,
+      "recon_loss": 1.8104088306427002,
+      "recon_trunc": 0.0,
+      "step": 495,
+      "student_loss": 0.041161973029375076,
+      "teacher_loss": 0.010950487107038498
+    },
+    {
+      "epoch": 0.14414414414414414,
+      "grad_norm": 1.7578125,
+      "kd_loss": 0.012948920950293541,
+      "learning_rate": 7.784019392472539e-06,
+      "loss": 0.1177,
+      "recon_loss": 1.9585274457931519,
+      "recon_trunc": 0.0,
+      "step": 500,
+      "student_loss": 0.005330843850970268,
+      "teacher_loss": 0.004721441306173801
+    },
+    {
+      "epoch": 0.1455855855855856,
+      "grad_norm": 1.7890625,
+      "kd_loss": 0.01527533121407032,
+      "learning_rate": 7.739020481708816e-06,
+      "loss": 0.1096,
+      "recon_loss": 1.361791729927063,
+      "recon_trunc": 0.0,
+      "step": 505,
+      "student_loss": 0.030247874557971954,
+      "teacher_loss": 0.0125796003267169
+    },
+    {
+      "epoch": 0.14702702702702702,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.018237337470054626,
+      "learning_rate": 7.693702340113587e-06,
+      "loss": 0.099,
+      "recon_loss": 2.694063425064087,
+      "recon_trunc": 0.0,
+      "step": 510,
+      "student_loss": 0.004448129795491695,
+      "teacher_loss": 0.005566176027059555
+    },
+    {
+      "epoch": 0.14846846846846848,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.01716817170381546,
+      "learning_rate": 7.648070249483074e-06,
+      "loss": 0.1057,
+      "recon_loss": 2.4092276096343994,
+      "recon_trunc": 0.0,
+      "step": 515,
+      "student_loss": 0.01954336278140545,
+      "teacher_loss": 0.01161402091383934
+    },
+    {
+      "epoch": 0.1499099099099099,
+      "grad_norm": 1.859375,
+      "kd_loss": 0.013520020060241222,
+      "learning_rate": 7.602129528204023e-06,
+      "loss": 0.1002,
+      "recon_loss": 2.7109413146972656,
+      "recon_trunc": 0.0,
+      "step": 520,
+      "student_loss": 0.007106412667781115,
+      "teacher_loss": 0.0036226226948201656
+    },
+    {
+      "epoch": 0.15135135135135136,
+      "grad_norm": 2.125,
+      "kd_loss": 0.0283492524176836,
+      "learning_rate": 7.555885530633864e-06,
+      "loss": 0.1057,
+      "recon_loss": 1.030029296875,
+      "recon_trunc": 0.0,
+      "step": 525,
+      "student_loss": 0.014120209962129593,
+      "teacher_loss": 0.007666388060897589
+    },
+    {
+      "epoch": 0.1527927927927928,
+      "grad_norm": 3.21875,
+      "kd_loss": 0.012168077751994133,
+      "learning_rate": 7.509343646476647e-06,
+      "loss": 0.1097,
+      "recon_loss": 2.056184768676758,
+      "recon_trunc": 0.0,
+      "step": 530,
+      "student_loss": 0.10587294399738312,
+      "teacher_loss": 0.06957944482564926
+    },
+    {
+      "epoch": 0.15423423423423424,
+      "grad_norm": 2.109375,
+      "kd_loss": 0.017548369243741035,
+      "learning_rate": 7.462509300154892e-06,
+      "loss": 0.1103,
+      "recon_loss": 1.127314805984497,
+      "recon_trunc": 0.0,
+      "step": 535,
+      "student_loss": 0.01272011548280716,
+      "teacher_loss": 0.030406273901462555
+    },
+    {
+      "epoch": 0.15567567567567567,
+      "grad_norm": 2.3125,
+      "kd_loss": 0.010825314559042454,
+      "learning_rate": 7.415387950177369e-06,
+      "loss": 0.0912,
+      "recon_loss": 0.7348822355270386,
+      "recon_trunc": 0.0,
+      "step": 540,
+      "student_loss": 0.005687657278031111,
+      "teacher_loss": 0.004439381416887045
+    },
+    {
+      "epoch": 0.15711711711711712,
+      "grad_norm": 1.5546875,
+      "kd_loss": 0.028425579890608788,
+      "learning_rate": 7.367985088502914e-06,
+      "loss": 0.1013,
+      "recon_loss": 1.6967211961746216,
+      "recon_trunc": 0.0,
+      "step": 545,
+      "student_loss": 0.032522182911634445,
+      "teacher_loss": 0.03722462058067322
+    },
+    {
+      "epoch": 0.15855855855855855,
+      "grad_norm": 1.8046875,
+      "kd_loss": 0.02775968424975872,
+      "learning_rate": 7.320306239900343e-06,
+      "loss": 0.1108,
+      "recon_loss": 1.2605125904083252,
+      "recon_trunc": 0.0,
+      "step": 550,
+      "student_loss": 0.029979035258293152,
+      "teacher_loss": 0.020589159801602364
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.012850400060415268,
+      "learning_rate": 7.27235696130455e-06,
+      "loss": 0.0939,
+      "recon_loss": 0.8966094255447388,
+      "recon_trunc": 0.0,
+      "step": 555,
+      "student_loss": 0.0039372630417346954,
+      "teacher_loss": 0.00253364653326571
+    },
+    {
+      "epoch": 0.16144144144144143,
+      "grad_norm": 3.984375,
+      "kd_loss": 0.01984316296875477,
+      "learning_rate": 7.2241428411688415e-06,
+      "loss": 0.1201,
+      "recon_loss": 0.6383526921272278,
+      "recon_trunc": 0.0,
+      "step": 560,
+      "student_loss": 0.013905559666454792,
+      "teacher_loss": 0.011228952556848526
+    },
+    {
+      "epoch": 0.16288288288288288,
+      "grad_norm": 2.03125,
+      "kd_loss": 0.014146905392408371,
+      "learning_rate": 7.1756694988136165e-06,
+      "loss": 0.1169,
+      "recon_loss": 0.9294888377189636,
+      "recon_trunc": 0.0,
+      "step": 565,
+      "student_loss": 0.004254485014826059,
+      "teacher_loss": 0.007349506486207247
+    },
+    {
+      "epoch": 0.1643243243243243,
+      "grad_norm": 1.984375,
+      "kd_loss": 0.01966790296137333,
+      "learning_rate": 7.126942583771433e-06,
+      "loss": 0.1011,
+      "recon_loss": 1.405023455619812,
+      "recon_trunc": 0.0,
+      "step": 570,
+      "student_loss": 0.032598767429590225,
+      "teacher_loss": 0.017219817265868187
+    },
+    {
+      "epoch": 0.16576576576576577,
+      "grad_norm": 2.6875,
+      "kd_loss": 0.009513165801763535,
+      "learning_rate": 7.077967775128558e-06,
+      "loss": 0.1139,
+      "recon_loss": 1.7493226528167725,
+      "recon_trunc": 0.0,
+      "step": 575,
+      "student_loss": 0.0040468815714120865,
+      "teacher_loss": 0.004172624088823795
+    },
+    {
+      "epoch": 0.16720720720720722,
+      "grad_norm": 3.75,
+      "kd_loss": 0.020738841965794563,
+      "learning_rate": 7.028750780863078e-06,
+      "loss": 0.1032,
+      "recon_loss": 0.6202591061592102,
+      "recon_trunc": 0.0,
+      "step": 580,
+      "student_loss": 0.011248578317463398,
+      "teacher_loss": 0.010976684279739857
+    },
+    {
+      "epoch": 0.16864864864864865,
+      "grad_norm": 2.3125,
+      "kd_loss": 0.025385674089193344,
+      "learning_rate": 6.979297337179632e-06,
+      "loss": 0.1025,
+      "recon_loss": 1.0138957500457764,
+      "recon_trunc": 0.0,
+      "step": 585,
+      "student_loss": 0.028503064066171646,
+      "teacher_loss": 0.010264373384416103
+    },
+    {
+      "epoch": 0.1700900900900901,
+      "grad_norm": 1.8671875,
+      "kd_loss": 0.0301853995770216,
+      "learning_rate": 6.929613207840875e-06,
+      "loss": 0.1018,
+      "recon_loss": 0.6474505066871643,
+      "recon_trunc": 0.13636364042758942,
+      "step": 590,
+      "student_loss": 0.008808792568743229,
+      "teacher_loss": 0.004761975258588791
+    },
+    {
+      "epoch": 0.17153153153153153,
+      "grad_norm": 2.78125,
+      "kd_loss": 0.01562667265534401,
+      "learning_rate": 6.8797041834956955e-06,
+      "loss": 0.1024,
+      "recon_loss": 0.8219525814056396,
+      "recon_trunc": 0.0,
+      "step": 595,
+      "student_loss": 0.00366596900857985,
+      "teacher_loss": 0.002013420220464468
+    },
+    {
+      "epoch": 0.17297297297297298,
+      "grad_norm": 2.359375,
+      "kd_loss": 0.019215423613786697,
+      "learning_rate": 6.829576081004339e-06,
+      "loss": 0.0961,
+      "recon_loss": 0.6283009648323059,
+      "recon_trunc": 0.0,
+      "step": 600,
+      "student_loss": 0.02321065217256546,
+      "teacher_loss": 0.022313382476568222
+    },
+    {
+      "epoch": 0.1744144144144144,
+      "grad_norm": 2.078125,
+      "kd_loss": 0.014276467263698578,
+      "learning_rate": 6.779234742760441e-06,
+      "loss": 0.0951,
+      "recon_loss": 0.869193434715271,
+      "recon_trunc": 0.0,
+      "step": 605,
+      "student_loss": 0.004768861923366785,
+      "teacher_loss": 0.0027686513494700193
+    },
+    {
+      "epoch": 0.17585585585585586,
+      "grad_norm": 6.5,
+      "kd_loss": 0.011466795578598976,
+      "learning_rate": 6.728686036010115e-06,
+      "loss": 0.0968,
+      "recon_loss": 1.282074213027954,
+      "recon_trunc": 0.0,
+      "step": 610,
+      "student_loss": 0.003399502718821168,
+      "teacher_loss": 0.002284495159983635
+    },
+    {
+      "epoch": 0.1772972972972973,
+      "grad_norm": 2.46875,
+      "kd_loss": 0.019078882411122322,
+      "learning_rate": 6.677935852168116e-06,
+      "loss": 0.0894,
+      "recon_loss": 0.6419486403465271,
+      "recon_trunc": 0.0,
+      "step": 615,
+      "student_loss": 0.012190327048301697,
+      "teacher_loss": 0.0023125908337533474
+    },
+    {
+      "epoch": 0.17873873873873874,
+      "grad_norm": 1.8515625,
+      "kd_loss": 0.027036365121603012,
+      "learning_rate": 6.62699010613121e-06,
+      "loss": 0.0996,
+      "recon_loss": 0.874606192111969,
+      "recon_trunc": 0.0,
+      "step": 620,
+      "student_loss": 0.0224213395267725,
+      "teacher_loss": 0.013729802332818508
+    },
+    {
+      "epoch": 0.18018018018018017,
+      "grad_norm": 3.328125,
+      "kd_loss": 0.007798908744007349,
+      "learning_rate": 6.5758547355887944e-06,
+      "loss": 0.0907,
+      "recon_loss": 2.5433475971221924,
+      "recon_trunc": 0.0,
+      "step": 625,
+      "student_loss": 0.011145239695906639,
+      "teacher_loss": 0.014219601638615131
+    },
+    {
+      "epoch": 0.18162162162162163,
+      "grad_norm": 2.375,
+      "kd_loss": 0.016567589715123177,
+      "learning_rate": 6.524535700330858e-06,
+      "loss": 0.1001,
+      "recon_loss": 1.1229078769683838,
+      "recon_trunc": 0.0,
+      "step": 630,
+      "student_loss": 0.06187977269291878,
+      "teacher_loss": 0.05378362536430359
+    },
+    {
+      "epoch": 0.18306306306306305,
+      "grad_norm": 1.6875,
+      "kd_loss": 0.02085557021200657,
+      "learning_rate": 6.4730389815533865e-06,
+      "loss": 0.0899,
+      "recon_loss": 1.187085509300232,
+      "recon_trunc": 0.0,
+      "step": 635,
+      "student_loss": 0.00996415875852108,
+      "teacher_loss": 0.006535361520946026
+    },
+    {
+      "epoch": 0.1845045045045045,
+      "grad_norm": 2.46875,
+      "kd_loss": 0.014909463934600353,
+      "learning_rate": 6.421370581161244e-06,
+      "loss": 0.0929,
+      "recon_loss": 0.9069868326187134,
+      "recon_trunc": 0.0,
+      "step": 640,
+      "student_loss": 0.019701724871993065,
+      "teacher_loss": 0.012945645488798618
+    },
+    {
+      "epoch": 0.18594594594594593,
+      "grad_norm": 2.375,
+      "kd_loss": 0.01683386042714119,
+      "learning_rate": 6.36953652106866e-06,
+      "loss": 0.1082,
+      "recon_loss": 0.494709312915802,
+      "recon_trunc": 0.0,
+      "step": 645,
+      "student_loss": 0.014453653246164322,
+      "teacher_loss": 0.010388009250164032
+    },
+    {
+      "epoch": 0.1873873873873874,
+      "grad_norm": 1.9453125,
+      "kd_loss": 0.0196087509393692,
+      "learning_rate": 6.317542842497384e-06,
+      "loss": 0.0994,
+      "recon_loss": 0.6906006336212158,
+      "recon_trunc": 0.0,
+      "step": 650,
+      "student_loss": 0.010469207540154457,
+      "teacher_loss": 0.015019979327917099
+    },
+    {
+      "epoch": 0.18882882882882882,
+      "grad_norm": 1.8046875,
+      "kd_loss": 0.022767694666981697,
+      "learning_rate": 6.265395605272581e-06,
+      "loss": 0.093,
+      "recon_loss": 0.925295352935791,
+      "recon_trunc": 0.0,
+      "step": 655,
+      "student_loss": 0.016291920095682144,
+      "teacher_loss": 0.012215238064527512
+    },
+    {
+      "epoch": 0.19027027027027027,
+      "grad_norm": 1.84375,
+      "kd_loss": 0.014528779312968254,
+      "learning_rate": 6.213100887116568e-06,
+      "loss": 0.1007,
+      "recon_loss": 0.7675014734268188,
+      "recon_trunc": 0.0,
+      "step": 660,
+      "student_loss": 0.028876151889562607,
+      "teacher_loss": 0.020355725660920143
+    },
+    {
+      "epoch": 0.19171171171171172,
+      "grad_norm": 2.3125,
+      "kd_loss": 0.013471566140651703,
+      "learning_rate": 6.1606647829404485e-06,
+      "loss": 0.0975,
+      "recon_loss": 0.9562106728553772,
+      "recon_trunc": 0.0,
+      "step": 665,
+      "student_loss": 0.0036644593346863985,
+      "teacher_loss": 0.0021427099127322435
+    },
+    {
+      "epoch": 0.19315315315315315,
+      "grad_norm": 1.9765625,
+      "kd_loss": 0.0173123087733984,
+      "learning_rate": 6.108093404133772e-06,
+      "loss": 0.1091,
+      "recon_loss": 1.563036322593689,
+      "recon_trunc": 0.0,
+      "step": 670,
+      "student_loss": 0.0046034688130021095,
+      "teacher_loss": 0.010668099857866764
+    },
+    {
+      "epoch": 0.1945945945945946,
+      "grad_norm": 1.921875,
+      "kd_loss": 0.020700708031654358,
+      "learning_rate": 6.0553928778522405e-06,
+      "loss": 0.1068,
+      "recon_loss": 1.0824533700942993,
+      "recon_trunc": 0.0,
+      "step": 675,
+      "student_loss": 0.020477328449487686,
+      "teacher_loss": 0.01928558386862278
+    },
+    {
+      "epoch": 0.19603603603603603,
+      "grad_norm": 3.140625,
+      "kd_loss": 0.011443321593105793,
+      "learning_rate": 6.002569346303605e-06,
+      "loss": 0.119,
+      "recon_loss": 0.7971557378768921,
+      "recon_trunc": 0.0,
+      "step": 680,
+      "student_loss": 0.0038715738337486982,
+      "teacher_loss": 0.009752260521054268
+    },
+    {
+      "epoch": 0.1974774774774775,
+      "grad_norm": 2.1875,
+      "kd_loss": 0.01789681799709797,
+      "learning_rate": 5.949628966031785e-06,
+      "loss": 0.0982,
+      "recon_loss": 1.0170581340789795,
+      "recon_trunc": 0.0,
+      "step": 685,
+      "student_loss": 0.011738591827452183,
+      "teacher_loss": 0.003596890950575471
+    },
+    {
+      "epoch": 0.1989189189189189,
+      "grad_norm": 1.515625,
+      "kd_loss": 0.010234168730676174,
+      "learning_rate": 5.8965779071993365e-06,
+      "loss": 0.0989,
+      "recon_loss": 1.3678561449050903,
+      "recon_trunc": 0.0,
+      "step": 690,
+      "student_loss": 0.04748813807964325,
+      "teacher_loss": 0.0339285247027874
+    },
+    {
+      "epoch": 0.20036036036036037,
+      "grad_norm": 1.28125,
+      "kd_loss": 0.008595343679189682,
+      "learning_rate": 5.843422352868321e-06,
+      "loss": 0.0935,
+      "recon_loss": 2.8541831970214844,
+      "recon_trunc": 0.0,
+      "step": 695,
+      "student_loss": 0.0029857722111046314,
+      "teacher_loss": 0.0022341967560350895
+    },
+    {
+      "epoch": 0.2018018018018018,
+      "grad_norm": 2.203125,
+      "kd_loss": 0.010823438875377178,
+      "learning_rate": 5.7901684982796716e-06,
+      "loss": 0.0968,
+      "recon_loss": 1.351580262184143,
+      "recon_trunc": 0.0,
+      "step": 700,
+      "student_loss": 0.07310483604669571,
+      "teacher_loss": 0.050734903663396835
+    },
+    {
+      "epoch": 0.20324324324324325,
+      "grad_norm": 1.9375,
+      "kd_loss": 0.013946866616606712,
+      "learning_rate": 5.736822550131147e-06,
+      "loss": 0.0949,
+      "recon_loss": 1.0093135833740234,
+      "recon_trunc": 0.0,
+      "step": 705,
+      "student_loss": 0.027368871495127678,
+      "teacher_loss": 0.04656485840678215
+    },
+    {
+      "epoch": 0.20468468468468468,
+      "grad_norm": 2.0,
+      "kd_loss": 0.017323924228549004,
+      "learning_rate": 5.68339072585394e-06,
+      "loss": 0.0892,
+      "recon_loss": 0.7244986295700073,
+      "recon_trunc": 0.0,
+      "step": 710,
+      "student_loss": 0.04973391443490982,
+      "teacher_loss": 0.04366318881511688
+    },
+    {
+      "epoch": 0.20612612612612613,
+      "grad_norm": 1.890625,
+      "kd_loss": 0.007980936206877232,
+      "learning_rate": 5.629879252888046e-06,
+      "loss": 0.0892,
+      "recon_loss": 0.8430204391479492,
+      "recon_trunc": 0.0,
+      "step": 715,
+      "student_loss": 0.003929371479898691,
+      "teacher_loss": 0.0031223194673657417
+    },
+    {
+      "epoch": 0.20756756756756756,
+      "grad_norm": 1.8671875,
+      "kd_loss": 0.009112298488616943,
+      "learning_rate": 5.576294367956455e-06,
+      "loss": 0.0955,
+      "recon_loss": 0.8602954149246216,
+      "recon_trunc": 0.0,
+      "step": 720,
+      "student_loss": 0.002911132061854005,
+      "teacher_loss": 0.002563533838838339
+    },
+    {
+      "epoch": 0.209009009009009,
+      "grad_norm": 2.609375,
+      "kd_loss": 0.0065227169543504715,
+      "learning_rate": 5.522642316338268e-06,
+      "loss": 0.1006,
+      "recon_loss": 0.4496604800224304,
+      "recon_trunc": 0.0,
+      "step": 725,
+      "student_loss": 0.004659323953092098,
+      "teacher_loss": 0.004054300021380186
+    },
+    {
+      "epoch": 0.21045045045045044,
+      "grad_norm": 2.765625,
+      "kd_loss": 0.013251368887722492,
+      "learning_rate": 5.4689293511408155e-06,
+      "loss": 0.1024,
+      "recon_loss": 1.2096127271652222,
+      "recon_trunc": 0.0,
+      "step": 730,
+      "student_loss": 0.006948236376047134,
+      "teacher_loss": 0.007843713276088238
+    },
+    {
+      "epoch": 0.2118918918918919,
+      "grad_norm": 2.015625,
+      "kd_loss": 0.013089313171803951,
+      "learning_rate": 5.4151617325708595e-06,
+      "loss": 0.098,
+      "recon_loss": 1.2019041776657104,
+      "recon_trunc": 0.0,
+      "step": 735,
+      "student_loss": 0.040870193392038345,
+      "teacher_loss": 0.027125149965286255
+    },
+    {
+      "epoch": 0.21333333333333335,
+      "grad_norm": 1.71875,
+      "kd_loss": 0.02157880924642086,
+      "learning_rate": 5.361345727204972e-06,
+      "loss": 0.1027,
+      "recon_loss": 1.120439052581787,
+      "recon_trunc": 0.02857142873108387,
+      "step": 740,
+      "student_loss": 0.06317839026451111,
+      "teacher_loss": 0.03955307975411415
+    },
+    {
+      "epoch": 0.21477477477477477,
+      "grad_norm": 2.5,
+      "kd_loss": 0.009251553565263748,
+      "learning_rate": 5.307487607259175e-06,
+      "loss": 0.101,
+      "recon_loss": 1.1851692199707031,
+      "recon_trunc": 0.0,
+      "step": 745,
+      "student_loss": 0.10666896402835846,
+      "teacher_loss": 0.06141383945941925
+    },
+    {
+      "epoch": 0.21621621621621623,
+      "grad_norm": 1.625,
+      "kd_loss": 0.012688029557466507,
+      "learning_rate": 5.253593649857907e-06,
+      "loss": 0.0965,
+      "recon_loss": 1.3027397394180298,
+      "recon_trunc": 0.0,
+      "step": 750,
+      "student_loss": 0.0028391056694090366,
+      "teacher_loss": 0.0029107951559126377
+    },
+    {
+      "epoch": 0.21765765765765765,
+      "grad_norm": 1.515625,
+      "kd_loss": 0.028317168354988098,
+      "learning_rate": 5.199670136302443e-06,
+      "loss": 0.0993,
+      "recon_loss": 1.194815754890442,
+      "recon_trunc": 0.0,
+      "step": 755,
+      "student_loss": 0.031675707548856735,
+      "teacher_loss": 0.021445225924253464
+    },
+    {
+      "epoch": 0.2190990990990991,
+      "grad_norm": 2.296875,
+      "kd_loss": 0.024305731058120728,
+      "learning_rate": 5.145723351338799e-06,
+      "loss": 0.0947,
+      "recon_loss": 0.7198876142501831,
+      "recon_trunc": 0.0,
+      "step": 760,
+      "student_loss": 0.004496672656387091,
+      "teacher_loss": 0.002695088740438223
+    },
+    {
+      "epoch": 0.22054054054054054,
+      "grad_norm": 2.28125,
+      "kd_loss": 0.009439276531338692,
+      "learning_rate": 5.091759582425263e-06,
+      "loss": 0.1063,
+      "recon_loss": 1.6393507719039917,
+      "recon_trunc": 0.0,
+      "step": 765,
+      "student_loss": 0.003358996007591486,
+      "teacher_loss": 0.002300855703651905
+    },
+    {
+      "epoch": 0.221981981981982,
+      "grad_norm": 1.9140625,
+      "kd_loss": 0.01450285129249096,
+      "learning_rate": 5.037785118999585e-06,
+      "loss": 0.1033,
+      "recon_loss": 2.684377431869507,
+      "recon_trunc": 0.0,
+      "step": 770,
+      "student_loss": 0.02557431533932686,
+      "teacher_loss": 0.017635051161050797
+    },
+    {
+      "epoch": 0.22342342342342342,
+      "grad_norm": 2.03125,
+      "kd_loss": 0.009869835339486599,
+      "learning_rate": 4.983806251745958e-06,
+      "loss": 0.104,
+      "recon_loss": 2.2494969367980957,
+      "recon_trunc": 0.0,
+      "step": 775,
+      "student_loss": 0.004350853618234396,
+      "teacher_loss": 0.011981328018009663
+    },
+    {
+      "epoch": 0.22486486486486487,
+      "grad_norm": 2.140625,
+      "kd_loss": 0.011742651462554932,
+      "learning_rate": 4.9298292718618295e-06,
+      "loss": 0.0987,
+      "recon_loss": 1.0973132848739624,
+      "recon_trunc": 0.0,
+      "step": 780,
+      "student_loss": 0.010750161483883858,
+      "teacher_loss": 0.0023801035713404417
+    },
+    {
+      "epoch": 0.2263063063063063,
+      "grad_norm": 1.9375,
+      "kd_loss": 0.014649628661572933,
+      "learning_rate": 4.875860470324685e-06,
+      "loss": 0.1004,
+      "recon_loss": 0.7299957275390625,
+      "recon_trunc": 0.0,
+      "step": 785,
+      "student_loss": 0.0033197549637407064,
+      "teacher_loss": 0.0042490423657000065
+    },
+    {
+      "epoch": 0.22774774774774775,
+      "grad_norm": 5.5,
+      "kd_loss": 0.014108363538980484,
+      "learning_rate": 4.821906137158822e-06,
+      "loss": 0.0943,
+      "recon_loss": 1.2130095958709717,
+      "recon_trunc": 0.0,
+      "step": 790,
+      "student_loss": 0.011785932816565037,
+      "teacher_loss": 0.0068002487532794476
+    },
+    {
+      "epoch": 0.22918918918918918,
+      "grad_norm": 3.5,
+      "kd_loss": 0.01567964069545269,
+      "learning_rate": 4.767972560702266e-06,
+      "loss": 0.1001,
+      "recon_loss": 2.4877049922943115,
+      "recon_trunc": 0.0,
+      "step": 795,
+      "student_loss": 0.19766555726528168,
+      "teacher_loss": 0.15711107850074768
+    },
+    {
+      "epoch": 0.23063063063063063,
+      "grad_norm": 1.859375,
+      "kd_loss": 0.018815049901604652,
+      "learning_rate": 4.7140660268738565e-06,
+      "loss": 0.0934,
+      "recon_loss": 1.1460384130477905,
+      "recon_trunc": 0.0,
+      "step": 800,
+      "student_loss": 0.0026388883125036955,
+      "teacher_loss": 0.021389150992035866
+    },
+    {
+      "epoch": 0.23207207207207206,
+      "grad_norm": 1.828125,
+      "kd_loss": 0.011156775057315826,
+      "learning_rate": 4.660192818440642e-06,
+      "loss": 0.0927,
+      "recon_loss": 1.1910074949264526,
+      "recon_trunc": 0.0,
+      "step": 805,
+      "student_loss": 0.003458657767623663,
+      "teacher_loss": 0.0030844355933368206
+    },
+    {
+      "epoch": 0.23351351351351352,
+      "grad_norm": 1.6640625,
+      "kd_loss": 0.017878390848636627,
+      "learning_rate": 4.606359214285615e-06,
+      "loss": 0.0953,
+      "recon_loss": 1.1707850694656372,
+      "recon_trunc": 0.0,
+      "step": 810,
+      "student_loss": 0.002790343714877963,
+      "teacher_loss": 0.001495691598393023
+    },
+    {
+      "epoch": 0.23495495495495494,
+      "grad_norm": 2.09375,
+      "kd_loss": 0.011960725300014019,
+      "learning_rate": 4.5525714886759214e-06,
+      "loss": 0.091,
+      "recon_loss": 3.316906213760376,
+      "recon_trunc": 0.0,
+      "step": 815,
+      "student_loss": 0.004394159186631441,
+      "teacher_loss": 0.0032252457458525896
+    },
+    {
+      "epoch": 0.2363963963963964,
+      "grad_norm": 1.890625,
+      "kd_loss": 0.007228494621813297,
+      "learning_rate": 4.498835910531595e-06,
+      "loss": 0.1048,
+      "recon_loss": 2.9495725631713867,
+      "recon_trunc": 0.0,
+      "step": 820,
+      "student_loss": 0.004381747916340828,
+      "teacher_loss": 0.00408752029761672
+    },
+    {
+      "epoch": 0.23783783783783785,
+      "grad_norm": 1.6328125,
+      "kd_loss": 0.010650701820850372,
+      "learning_rate": 4.445158742694919e-06,
+      "loss": 0.0962,
+      "recon_loss": 0.4567302465438843,
+      "recon_trunc": 0.0,
+      "step": 825,
+      "student_loss": 0.00786358118057251,
+      "teacher_loss": 0.008311954326927662
+    },
+    {
+      "epoch": 0.23927927927927928,
+      "grad_norm": 1.4609375,
+      "kd_loss": 0.02082512527704239,
+      "learning_rate": 4.391546241200503e-06,
+      "loss": 0.0767,
+      "recon_loss": 0.9175398349761963,
+      "recon_trunc": 0.0,
+      "step": 830,
+      "student_loss": 0.012589729391038418,
+      "teacher_loss": 0.00804829504340887
+    },
+    {
+      "epoch": 0.24072072072072073,
+      "grad_norm": 1.515625,
+      "kd_loss": 0.020304908975958824,
+      "learning_rate": 4.338004654546136e-06,
+      "loss": 0.0933,
+      "recon_loss": 1.0316662788391113,
+      "recon_trunc": 0.095238097012043,
+      "step": 835,
+      "student_loss": 0.07545731961727142,
+      "teacher_loss": 0.06839510798454285
+    },
+    {
+      "epoch": 0.24216216216216216,
+      "grad_norm": 1.9375,
+      "kd_loss": 0.013990873470902443,
+      "learning_rate": 4.2845402229645366e-06,
+      "loss": 0.0841,
+      "recon_loss": 1.8841662406921387,
+      "recon_trunc": 0.0,
+      "step": 840,
+      "student_loss": 0.005074299871921539,
+      "teacher_loss": 0.03514579311013222
+    },
+    {
+      "epoch": 0.2436036036036036,
+      "grad_norm": 1.5859375,
+      "kd_loss": 0.017677653580904007,
+      "learning_rate": 4.231159177696057e-06,
+      "loss": 0.0957,
+      "recon_loss": 1.6847047805786133,
+      "recon_trunc": 0.0,
+      "step": 845,
+      "student_loss": 0.056813523173332214,
+      "teacher_loss": 0.036687787622213364
+    },
+    {
+      "epoch": 0.24504504504504504,
+      "grad_norm": 1.8828125,
+      "kd_loss": 0.007710277568548918,
+      "learning_rate": 4.177867740262437e-06,
+      "loss": 0.0947,
+      "recon_loss": 0.8180376887321472,
+      "recon_trunc": 0.0,
+      "step": 850,
+      "student_loss": 0.07068512588739395,
+      "teacher_loss": 0.035924047231674194
+    },
+    {
+      "epoch": 0.2464864864864865,
+      "grad_norm": 1.5390625,
+      "kd_loss": 0.009996934793889523,
+      "learning_rate": 4.1246721217416846e-06,
+      "loss": 0.0956,
+      "recon_loss": 1.8882759809494019,
+      "recon_trunc": 0.0,
+      "step": 855,
+      "student_loss": 0.03555234149098396,
+      "teacher_loss": 0.015627330169081688
+    },
+    {
+      "epoch": 0.24792792792792792,
+      "grad_norm": 2.09375,
+      "kd_loss": 0.01746698096394539,
+      "learning_rate": 4.071578522044188e-06,
+      "loss": 0.1005,
+      "recon_loss": 0.6234949827194214,
+      "recon_trunc": 0.0,
+      "step": 860,
+      "student_loss": 0.01093185506761074,
+      "teacher_loss": 0.01649426482617855
+    },
+    {
+      "epoch": 0.24936936936936938,
+      "grad_norm": 2.171875,
+      "kd_loss": 0.015009298920631409,
+      "learning_rate": 4.018593129190113e-06,
+      "loss": 0.1077,
+      "recon_loss": 0.8162397742271423,
+      "recon_trunc": 0.0,
+      "step": 865,
+      "student_loss": 0.009281515143811703,
+      "teacher_loss": 0.0039774044416844845
+    },
+    {
+      "epoch": 0.2508108108108108,
+      "grad_norm": 1.3046875,
+      "kd_loss": 0.0070146722719073296,
+      "learning_rate": 3.965722118588199e-06,
+      "loss": 0.0894,
+      "recon_loss": 0.20341670513153076,
+      "recon_trunc": 0.0,
+      "step": 870,
+      "student_loss": 0.0033650461118668318,
+      "teacher_loss": 0.003060839604586363
+    },
+    {
+      "epoch": 0.25225225225225223,
+      "grad_norm": 1.5234375,
+      "kd_loss": 0.007941058836877346,
+      "learning_rate": 3.912971652316017e-06,
+      "loss": 0.0982,
+      "recon_loss": 1.1236732006072998,
+      "recon_trunc": 0.0,
+      "step": 875,
+      "student_loss": 0.0031773594673722982,
+      "teacher_loss": 0.002277960767969489
+    },
+    {
+      "epoch": 0.2536936936936937,
+      "grad_norm": 1.7265625,
+      "kd_loss": 0.03348002955317497,
+      "learning_rate": 3.8603478784017845e-06,
+      "loss": 0.0964,
+      "recon_loss": 1.726459264755249,
+      "recon_trunc": 0.0,
+      "step": 880,
+      "student_loss": 0.048676300793886185,
+      "teacher_loss": 0.024330031126737595
+    },
+    {
+      "epoch": 0.25513513513513514,
+      "grad_norm": 1.390625,
+      "kd_loss": 0.006867540534585714,
+      "learning_rate": 3.8078569301078165e-06,
+      "loss": 0.095,
+      "recon_loss": 1.1007535457611084,
+      "recon_trunc": 0.0,
+      "step": 885,
+      "student_loss": 0.05072127282619476,
+      "teacher_loss": 0.044561658054590225
+    },
+    {
+      "epoch": 0.25657657657657656,
+      "grad_norm": 1.7734375,
+      "kd_loss": 0.008706974796950817,
+      "learning_rate": 3.7555049252157006e-06,
+      "loss": 0.0983,
+      "recon_loss": 1.0271594524383545,
+      "recon_trunc": 0.0,
+      "step": 890,
+      "student_loss": 0.0025675473734736443,
+      "teacher_loss": 0.00493641709908843
+    },
+    {
+      "epoch": 0.258018018018018,
+      "grad_norm": 1.515625,
+      "kd_loss": 0.01960054226219654,
+      "learning_rate": 3.703297965313275e-06,
+      "loss": 0.0919,
+      "recon_loss": 1.547040343284607,
+      "recon_trunc": 0.0,
+      "step": 895,
+      "student_loss": 0.05576763674616814,
+      "teacher_loss": 0.0349409393966198
+    },
+    {
+      "epoch": 0.2594594594594595,
+      "grad_norm": 1.421875,
+      "kd_loss": 0.015275723300874233,
+      "learning_rate": 3.651242135083485e-06,
+      "loss": 0.0853,
+      "recon_loss": 1.4707047939300537,
+      "recon_trunc": 0.0,
+      "step": 900,
+      "student_loss": 0.01047065295279026,
+      "teacher_loss": 0.019112160429358482
+    },
+    {
+      "epoch": 0.2609009009009009,
+      "grad_norm": 1.6875,
+      "kd_loss": 0.01020866446197033,
+      "learning_rate": 3.599343501595229e-06,
+      "loss": 0.0974,
+      "recon_loss": 1.6204107999801636,
+      "recon_trunc": 0.0,
+      "step": 905,
+      "student_loss": 0.007498105987906456,
+      "teacher_loss": 0.005625428166240454
+    },
+    {
+      "epoch": 0.2623423423423423,
+      "grad_norm": 1.9296875,
+      "kd_loss": 0.022968482226133347,
+      "learning_rate": 3.5476081135962335e-06,
+      "loss": 0.1013,
+      "recon_loss": 1.4789589643478394,
+      "recon_trunc": 0.0,
+      "step": 910,
+      "student_loss": 0.17107854783535004,
+      "teacher_loss": 0.12146960198879242
+    },
+    {
+      "epoch": 0.2637837837837838,
+      "grad_norm": 2.015625,
+      "kd_loss": 0.008676648139953613,
+      "learning_rate": 3.496042000808091e-06,
+      "loss": 0.0954,
+      "recon_loss": 1.9014108180999756,
+      "recon_trunc": 0.0,
+      "step": 915,
+      "student_loss": 0.0032600320409983397,
+      "teacher_loss": 0.005251611117273569
+    },
+    {
+      "epoch": 0.26522522522522524,
+      "grad_norm": 2.0625,
+      "kd_loss": 0.006392610725015402,
+      "learning_rate": 3.4446511732234846e-06,
+      "loss": 0.0965,
+      "recon_loss": 3.175675392150879,
+      "recon_trunc": 0.0,
+      "step": 920,
+      "student_loss": 0.0030960082076489925,
+      "teacher_loss": 0.002726975828409195
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 2.625,
+      "kd_loss": 0.015356424264609814,
+      "learning_rate": 3.3934416204057396e-06,
+      "loss": 0.0957,
+      "recon_loss": 1.1342909336090088,
+      "recon_trunc": 0.0,
+      "step": 925,
+      "student_loss": 0.03828441724181175,
+      "teacher_loss": 0.02549426443874836
+    },
+    {
+      "epoch": 0.2681081081081081,
+      "grad_norm": 1.59375,
+      "kd_loss": 0.008220403455197811,
+      "learning_rate": 3.3424193107907386e-06,
+      "loss": 0.0932,
+      "recon_loss": 0.886868417263031,
+      "recon_trunc": 0.0,
+      "step": 930,
+      "student_loss": 0.0020137070678174496,
+      "teacher_loss": 0.0015112607507035136
+    },
+    {
+      "epoch": 0.26954954954954957,
+      "grad_norm": 2.4375,
+      "kd_loss": 0.015720056369900703,
+      "learning_rate": 3.291590190991306e-06,
+      "loss": 0.0939,
+      "recon_loss": 1.2992150783538818,
+      "recon_trunc": 0.0,
+      "step": 935,
+      "student_loss": 0.006933893077075481,
+      "teacher_loss": 0.0027362850960344076
+    },
+    {
+      "epoch": 0.270990990990991,
+      "grad_norm": 1.5234375,
+      "kd_loss": 0.012597126886248589,
+      "learning_rate": 3.240960185104137e-06,
+      "loss": 0.093,
+      "recon_loss": 1.9500657320022583,
+      "recon_trunc": 0.0,
+      "step": 940,
+      "student_loss": 0.014928960241377354,
+      "teacher_loss": 0.012803679332137108
+    },
+    {
+      "epoch": 0.2724324324324324,
+      "grad_norm": 1.8125,
+      "kd_loss": 0.008467881008982658,
+      "learning_rate": 3.1905351940193417e-06,
+      "loss": 0.094,
+      "recon_loss": 2.5965704917907715,
+      "recon_trunc": 0.0,
+      "step": 945,
+      "student_loss": 0.018712086603045464,
+      "teacher_loss": 0.011909306049346924
+    },
+    {
+      "epoch": 0.27387387387387385,
+      "grad_norm": 1.6484375,
+      "kd_loss": 0.007483259774744511,
+      "learning_rate": 3.14032109473271e-06,
+      "loss": 0.1115,
+      "recon_loss": 3.548143148422241,
+      "recon_trunc": 0.0,
+      "step": 950,
+      "student_loss": 0.007603426463901997,
+      "teacher_loss": 0.006821023300290108
+    },
+    {
+      "epoch": 0.27531531531531533,
+      "grad_norm": 1.8046875,
+      "kd_loss": 0.010757386684417725,
+      "learning_rate": 3.090323739660742e-06,
+      "loss": 0.0805,
+      "recon_loss": 0.7502933144569397,
+      "recon_trunc": 0.0,
+      "step": 955,
+      "student_loss": 0.00226012896746397,
+      "teacher_loss": 0.0012833380606025457
+    },
+    {
+      "epoch": 0.27675675675675676,
+      "grad_norm": 1.703125,
+      "kd_loss": 0.010780137963593006,
+      "learning_rate": 3.040548955958563e-06,
+      "loss": 0.092,
+      "recon_loss": 0.8820634484291077,
+      "recon_trunc": 0.0,
+      "step": 960,
+      "student_loss": 0.04735017940402031,
+      "teacher_loss": 0.01725047081708908
+    },
+    {
+      "epoch": 0.2781981981981982,
+      "grad_norm": 2.75,
+      "kd_loss": 0.006949195172637701,
+      "learning_rate": 2.991002544840763e-06,
+      "loss": 0.092,
+      "recon_loss": 3.699920177459717,
+      "recon_trunc": 0.0,
+      "step": 965,
+      "student_loss": 0.006522569339722395,
+      "teacher_loss": 0.005998547188937664
+    },
+    {
+      "epoch": 0.2796396396396396,
+      "grad_norm": 1.5546875,
+      "kd_loss": 0.011674799956381321,
+      "learning_rate": 2.9416902809052817e-06,
+      "loss": 0.0914,
+      "recon_loss": 1.6148827075958252,
+      "recon_trunc": 0.0,
+      "step": 970,
+      "student_loss": 0.002716856775805354,
+      "teacher_loss": 0.0016918969340622425
+    },
+    {
+      "epoch": 0.2810810810810811,
+      "grad_norm": 2.90625,
+      "kd_loss": 0.0183776393532753,
+      "learning_rate": 2.8926179114603704e-06,
+      "loss": 0.0961,
+      "recon_loss": 0.5997033715248108,
+      "recon_trunc": 0.0,
+      "step": 975,
+      "student_loss": 0.005472161341458559,
+      "teacher_loss": 0.001077870954759419
+    },
+    {
+      "epoch": 0.2825225225225225,
+      "grad_norm": 1.5859375,
+      "kd_loss": 0.017831645905971527,
+      "learning_rate": 2.8437911558547515e-06,
+      "loss": 0.0858,
+      "recon_loss": 1.7456401586532593,
+      "recon_trunc": 0.0,
+      "step": 980,
+      "student_loss": 0.029375635087490082,
+      "teacher_loss": 0.024158205837011337
+    },
+    {
+      "epoch": 0.28396396396396395,
+      "grad_norm": 2.03125,
+      "kd_loss": 0.009199182502925396,
+      "learning_rate": 2.7952157048110406e-06,
+      "loss": 0.0908,
+      "recon_loss": 2.176257610321045,
+      "recon_trunc": 0.0,
+      "step": 985,
+      "student_loss": 0.002366869244724512,
+      "teacher_loss": 0.002399787772446871
+    },
+    {
+      "epoch": 0.28540540540540543,
+      "grad_norm": 1.40625,
+      "kd_loss": 0.014838067814707756,
+      "learning_rate": 2.74689721976249e-06,
+      "loss": 0.09,
+      "recon_loss": 1.4003294706344604,
+      "recon_trunc": 0.0,
+      "step": 990,
+      "student_loss": 0.011432410217821598,
+      "teacher_loss": 0.01332931499928236
+    },
+    {
+      "epoch": 0.28684684684684686,
+      "grad_norm": 1.75,
+      "kd_loss": 0.018339650705456734,
+      "learning_rate": 2.698841332193148e-06,
+      "loss": 0.0876,
+      "recon_loss": 0.7649494409561157,
+      "recon_trunc": 0.0,
+      "step": 995,
+      "student_loss": 0.054880835115909576,
+      "teacher_loss": 0.022861065343022346
+    },
+    {
+      "epoch": 0.2882882882882883,
+      "grad_norm": 1.171875,
+      "kd_loss": 0.01832008734345436,
+      "learning_rate": 2.6510536429815224e-06,
+      "loss": 0.0849,
+      "recon_loss": 0.9259073138237,
+      "recon_trunc": 0.0,
+      "step": 1000,
+      "student_loss": 0.013930868357419968,
+      "teacher_loss": 0.010818339884281158
+    },
+    {
+      "epoch": 0.2897297297297297,
+      "grad_norm": 2.078125,
+      "kd_loss": 0.011398128233850002,
+      "learning_rate": 2.6035397217477977e-06,
+      "loss": 0.0922,
+      "recon_loss": 1.2647162675857544,
+      "recon_trunc": 0.0,
+      "step": 1005,
+      "student_loss": 0.03702947124838829,
+      "teacher_loss": 0.014146753586828709
+    },
+    {
+      "epoch": 0.2911711711711712,
+      "grad_norm": 1.7578125,
+      "kd_loss": 0.008698206394910812,
+      "learning_rate": 2.5563051062047034e-06,
+      "loss": 0.0931,
+      "recon_loss": 3.5339043140411377,
+      "recon_trunc": 0.0,
+      "step": 1010,
+      "student_loss": 0.002814521547406912,
+      "teacher_loss": 0.002307652495801449
+    },
+    {
+      "epoch": 0.2926126126126126,
+      "grad_norm": 2.09375,
+      "kd_loss": 0.008382459171116352,
+      "learning_rate": 2.5093553015120937e-06,
+      "loss": 0.0908,
+      "recon_loss": 0.8478963375091553,
+      "recon_trunc": 0.0,
+      "step": 1015,
+      "student_loss": 0.003397122025489807,
+      "teacher_loss": 0.002558506326749921
+    },
+    {
+      "epoch": 0.29405405405405405,
+      "grad_norm": 1.6875,
+      "kd_loss": 0.01787000708281994,
+      "learning_rate": 2.4626957796353257e-06,
+      "loss": 0.0946,
+      "recon_loss": 1.4376500844955444,
+      "recon_trunc": 0.0,
+      "step": 1020,
+      "student_loss": 0.03289255499839783,
+      "teacher_loss": 0.05749712511897087
+    },
+    {
+      "epoch": 0.2954954954954955,
+      "grad_norm": 2.0625,
+      "kd_loss": 0.008949090726673603,
+      "learning_rate": 2.416331978707506e-06,
+      "loss": 0.0915,
+      "recon_loss": 2.0869102478027344,
+      "recon_trunc": 0.0,
+      "step": 1025,
+      "student_loss": 0.045973047614097595,
+      "teacher_loss": 0.032188646495342255
+    },
+    {
+      "epoch": 0.29693693693693696,
+      "grad_norm": 1.328125,
+      "kd_loss": 0.023239919915795326,
+      "learning_rate": 2.3702693023956853e-06,
+      "loss": 0.0903,
+      "recon_loss": 1.1783219575881958,
+      "recon_trunc": 0.0,
+      "step": 1030,
+      "student_loss": 0.023714011535048485,
+      "teacher_loss": 0.004810755141079426
+    },
+    {
+      "epoch": 0.2983783783783784,
+      "grad_norm": 1.890625,
+      "kd_loss": 0.012070492841303349,
+      "learning_rate": 2.32451311927106e-06,
+      "loss": 0.0858,
+      "recon_loss": 1.1948258876800537,
+      "recon_trunc": 0.0,
+      "step": 1035,
+      "student_loss": 0.05023680254817009,
+      "teacher_loss": 0.04626806825399399
+    },
+    {
+      "epoch": 0.2998198198198198,
+      "grad_norm": 1.6796875,
+      "kd_loss": 0.01622183620929718,
+      "learning_rate": 2.279068762183267e-06,
+      "loss": 0.0936,
+      "recon_loss": 0.8670558333396912,
+      "recon_trunc": 0.0,
+      "step": 1040,
+      "student_loss": 0.028481584042310715,
+      "teacher_loss": 0.02142520621418953
+    },
+    {
+      "epoch": 0.30126126126126124,
+      "grad_norm": 2.46875,
+      "kd_loss": 0.007846716791391373,
+      "learning_rate": 2.233941527638848e-06,
+      "loss": 0.0918,
+      "recon_loss": 1.5890685319900513,
+      "recon_trunc": 0.0,
+      "step": 1045,
+      "student_loss": 0.0036064554005861282,
+      "teacher_loss": 0.002732039662078023
+    },
+    {
+      "epoch": 0.3027027027027027,
+      "grad_norm": 2.53125,
+      "kd_loss": 0.02530010975897312,
+      "learning_rate": 2.1891366751839443e-06,
+      "loss": 0.0895,
+      "recon_loss": 1.2964962720870972,
+      "recon_trunc": 0.0,
+      "step": 1050,
+      "student_loss": 0.01320843305438757,
+      "teacher_loss": 0.02027864195406437
+    },
+    {
+      "epoch": 0.30414414414414415,
+      "grad_norm": 2.125,
+      "kd_loss": 0.016451945528388023,
+      "learning_rate": 2.1446594267913003e-06,
+      "loss": 0.0823,
+      "recon_loss": 0.9878601431846619,
+      "recon_trunc": 0.0,
+      "step": 1055,
+      "student_loss": 0.00892066303640604,
+      "teacher_loss": 0.005814709234982729
+    },
+    {
+      "epoch": 0.3055855855855856,
+      "grad_norm": 2.015625,
+      "kd_loss": 0.014636109583079815,
+      "learning_rate": 2.1005149662516517e-06,
+      "loss": 0.0884,
+      "recon_loss": 0.5212307572364807,
+      "recon_trunc": 0.0,
+      "step": 1060,
+      "student_loss": 0.005706994794309139,
+      "teacher_loss": 0.004586569964885712
+    },
+    {
+      "epoch": 0.307027027027027,
+      "grad_norm": 1.34375,
+      "kd_loss": 0.018635490909218788,
+      "learning_rate": 2.0567084385695475e-06,
+      "loss": 0.093,
+      "recon_loss": 0.8483979105949402,
+      "recon_trunc": 0.0,
+      "step": 1065,
+      "student_loss": 0.014275050722062588,
+      "teacher_loss": 0.0028141001239418983
+    },
+    {
+      "epoch": 0.3084684684684685,
+      "grad_norm": 1.828125,
+      "kd_loss": 0.02133617177605629,
+      "learning_rate": 2.0132449493637095e-06,
+      "loss": 0.0863,
+      "recon_loss": 1.1203880310058594,
+      "recon_trunc": 0.029411764815449715,
+      "step": 1070,
+      "student_loss": 0.02277274988591671,
+      "teacher_loss": 0.017514826729893684
+    },
+    {
+      "epoch": 0.3099099099099099,
+      "grad_norm": 2.8125,
+      "kd_loss": 0.025110790506005287,
+      "learning_rate": 1.9701295642719836e-06,
+      "loss": 0.0908,
+      "recon_loss": 1.4889389276504517,
+      "recon_trunc": 0.0,
+      "step": 1075,
+      "student_loss": 0.02432236634194851,
+      "teacher_loss": 0.019872069358825684
+    },
+    {
+      "epoch": 0.31135135135135134,
+      "grad_norm": 2.484375,
+      "kd_loss": 0.01807084120810032,
+      "learning_rate": 1.927367308360926e-06,
+      "loss": 0.0989,
+      "recon_loss": 1.4147769212722778,
+      "recon_trunc": 0.0,
+      "step": 1080,
+      "student_loss": 0.002939870348200202,
+      "teacher_loss": 0.0010408149100840092
+    },
+    {
+      "epoch": 0.3127927927927928,
+      "grad_norm": 2.25,
+      "kd_loss": 0.011869323439896107,
+      "learning_rate": 1.8849631655401557e-06,
+      "loss": 0.0907,
+      "recon_loss": 1.709843397140503,
+      "recon_trunc": 0.0,
+      "step": 1085,
+      "student_loss": 0.0282153207808733,
+      "teacher_loss": 0.03757878765463829
+    },
+    {
+      "epoch": 0.31423423423423424,
+      "grad_norm": 1.5546875,
+      "kd_loss": 0.016320809721946716,
+      "learning_rate": 1.8429220779814654e-06,
+      "loss": 0.0899,
+      "recon_loss": 0.6446734070777893,
+      "recon_trunc": 0.0,
+      "step": 1090,
+      "student_loss": 0.05461004003882408,
+      "teacher_loss": 0.0350608229637146
+    },
+    {
+      "epoch": 0.31567567567567567,
+      "grad_norm": 2.28125,
+      "kd_loss": 0.020763752982020378,
+      "learning_rate": 1.8012489455428261e-06,
+      "loss": 0.0973,
+      "recon_loss": 0.9621430039405823,
+      "recon_trunc": 0.0,
+      "step": 1095,
+      "student_loss": 0.0130164111033082,
+      "teacher_loss": 0.0014654602855443954
+    },
+    {
+      "epoch": 0.3171171171171171,
+      "grad_norm": 1.6640625,
+      "kd_loss": 0.029585907235741615,
+      "learning_rate": 1.7599486251973057e-06,
+      "loss": 0.0917,
+      "recon_loss": 1.8840738534927368,
+      "recon_trunc": 0.125,
+      "step": 1100,
+      "student_loss": 0.03502124175429344,
+      "teacher_loss": 0.023600025102496147
+    },
+    {
+      "epoch": 0.3185585585585586,
+      "grad_norm": 2.078125,
+      "kd_loss": 0.01753348670899868,
+      "learning_rate": 1.7190259304670038e-06,
+      "loss": 0.0784,
+      "recon_loss": 0.582436740398407,
+      "recon_trunc": 0.0,
+      "step": 1105,
+      "student_loss": 0.020779673010110855,
+      "teacher_loss": 0.02027658000588417
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 2.609375,
+      "kd_loss": 0.020432377234101295,
+      "learning_rate": 1.6784856308620217e-06,
+      "loss": 0.0924,
+      "recon_loss": 0.4507274925708771,
+      "recon_trunc": 0.0,
+      "step": 1110,
+      "student_loss": 0.010371611453592777,
+      "teacher_loss": 0.0023979730904102325
+    },
+    {
+      "epoch": 0.32144144144144143,
+      "grad_norm": 1.390625,
+      "kd_loss": 0.008641150780022144,
+      "learning_rate": 1.638332451324593e-06,
+      "loss": 0.0912,
+      "recon_loss": 1.6399307250976562,
+      "recon_trunc": 0.0,
+      "step": 1115,
+      "student_loss": 0.008218732662498951,
+      "teacher_loss": 0.0056208232417702675
+    },
+    {
+      "epoch": 0.32288288288288286,
+      "grad_norm": 1.421875,
+      "kd_loss": 0.0075892792083323,
+      "learning_rate": 1.5985710716783936e-06,
+      "loss": 0.0937,
+      "recon_loss": 1.7890102863311768,
+      "recon_trunc": 0.0,
+      "step": 1120,
+      "student_loss": 0.002585415495559573,
+      "teacher_loss": 0.0025254059582948685
+    },
+    {
+      "epoch": 0.32432432432432434,
+      "grad_norm": 1.859375,
+      "kd_loss": 0.007906612008810043,
+      "learning_rate": 1.5592061260831026e-06,
+      "loss": 0.1223,
+      "recon_loss": 4.803988456726074,
+      "recon_trunc": 0.0,
+      "step": 1125,
+      "student_loss": 0.04124952480196953,
+      "teacher_loss": 0.03742600232362747
+    },
+    {
+      "epoch": 0.32576576576576577,
+      "grad_norm": 1.6484375,
+      "kd_loss": 0.007388543803244829,
+      "learning_rate": 1.5202422024943069e-06,
+      "loss": 0.0829,
+      "recon_loss": 0.67726731300354,
+      "recon_trunc": 0.0,
+      "step": 1130,
+      "student_loss": 0.002848874544724822,
+      "teacher_loss": 0.0026724361814558506
+    },
+    {
+      "epoch": 0.3272072072072072,
+      "grad_norm": 1.9375,
+      "kd_loss": 0.022917937487363815,
+      "learning_rate": 1.4816838421287693e-06,
+      "loss": 0.1084,
+      "recon_loss": 1.062599539756775,
+      "recon_trunc": 0.0,
+      "step": 1135,
+      "student_loss": 0.05018072575330734,
+      "teacher_loss": 0.023166941478848457
+    },
+    {
+      "epoch": 0.3286486486486486,
+      "grad_norm": 3.3125,
+      "kd_loss": 0.013231215067207813,
+      "learning_rate": 1.4435355389351553e-06,
+      "loss": 0.0935,
+      "recon_loss": 0.82977694272995,
+      "recon_trunc": 0.0,
+      "step": 1140,
+      "student_loss": 0.0035748393274843693,
+      "teacher_loss": 0.0023794621229171753
+    },
+    {
+      "epoch": 0.3300900900900901,
+      "grad_norm": 1.6875,
+      "kd_loss": 0.01594485342502594,
+      "learning_rate": 1.4058017390702671e-06,
+      "loss": 0.1028,
+      "recon_loss": 3.129906177520752,
+      "recon_trunc": 0.0,
+      "step": 1145,
+      "student_loss": 0.02241712436079979,
+      "teacher_loss": 0.004998465068638325
+    },
+    {
+      "epoch": 0.33153153153153153,
+      "grad_norm": 1.421875,
+      "kd_loss": 0.012003305368125439,
+      "learning_rate": 1.368486840380851e-06,
+      "loss": 0.0935,
+      "recon_loss": 0.7430909872055054,
+      "recon_trunc": 0.0,
+      "step": 1150,
+      "student_loss": 0.020740047097206116,
+      "teacher_loss": 0.008964942768216133
+    },
+    {
+      "epoch": 0.33297297297297296,
+      "grad_norm": 1.6640625,
+      "kd_loss": 0.011111510917544365,
+      "learning_rate": 1.3315951918910241e-06,
+      "loss": 0.091,
+      "recon_loss": 2.4856646060943604,
+      "recon_trunc": 0.0,
+      "step": 1155,
+      "student_loss": 0.0018154662102460861,
+      "teacher_loss": 0.001369763514958322
+    },
+    {
+      "epoch": 0.33441441441441444,
+      "grad_norm": 2.703125,
+      "kd_loss": 0.031182873994112015,
+      "learning_rate": 1.2951310932954024e-06,
+      "loss": 0.0878,
+      "recon_loss": 0.9922821521759033,
+      "recon_trunc": 0.0,
+      "step": 1160,
+      "student_loss": 0.030812330543994904,
+      "teacher_loss": 0.006668732967227697
+    },
+    {
+      "epoch": 0.33585585585585587,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.0226407703012228,
+      "learning_rate": 1.2590987944579808e-06,
+      "loss": 0.1046,
+      "recon_loss": 0.7255238890647888,
+      "recon_trunc": 0.0,
+      "step": 1165,
+      "student_loss": 0.027604632079601288,
+      "teacher_loss": 0.010866493918001652
+    },
+    {
+      "epoch": 0.3372972972972973,
+      "grad_norm": 2.0625,
+      "kd_loss": 0.01790717802941799,
+      "learning_rate": 1.2235024949167994e-06,
+      "loss": 0.0937,
+      "recon_loss": 1.1223076581954956,
+      "recon_trunc": 0.0,
+      "step": 1170,
+      "student_loss": 0.025606432929635048,
+      "teacher_loss": 0.009214171208441257
+    },
+    {
+      "epoch": 0.3387387387387387,
+      "grad_norm": 1.3359375,
+      "kd_loss": 0.02829998917877674,
+      "learning_rate": 1.1883463433945092e-06,
+      "loss": 0.102,
+      "recon_loss": 0.5901147723197937,
+      "recon_trunc": 0.0,
+      "step": 1175,
+      "student_loss": 0.028822554275393486,
+      "teacher_loss": 0.012383908964693546
+    },
+    {
+      "epoch": 0.3401801801801802,
+      "grad_norm": 1.6640625,
+      "kd_loss": 0.02160187065601349,
+      "learning_rate": 1.1536344373148245e-06,
+      "loss": 0.1012,
+      "recon_loss": 1.1110165119171143,
+      "recon_trunc": 0.0,
+      "step": 1180,
+      "student_loss": 0.050414130091667175,
+      "teacher_loss": 0.010841689072549343
+    },
+    {
+      "epoch": 0.34162162162162163,
+      "grad_norm": 1.8515625,
+      "kd_loss": 0.0186303723603487,
+      "learning_rate": 1.119370822324981e-06,
+      "loss": 0.0898,
+      "recon_loss": 0.6764245629310608,
+      "recon_trunc": 0.0,
+      "step": 1185,
+      "student_loss": 0.014354808256030083,
+      "teacher_loss": 0.011137118563055992
+    },
+    {
+      "epoch": 0.34306306306306306,
+      "grad_norm": 1.71875,
+      "kd_loss": 0.014475632458925247,
+      "learning_rate": 1.0855594918242157e-06,
+      "loss": 0.0911,
+      "recon_loss": 0.9524720907211304,
+      "recon_trunc": 0.0,
+      "step": 1190,
+      "student_loss": 0.03521260991692543,
+      "teacher_loss": 0.002256652805954218
+    },
+    {
+      "epoch": 0.3445045045045045,
+      "grad_norm": 2.15625,
+      "kd_loss": 0.01269450131803751,
+      "learning_rate": 1.0522043864983428e-06,
+      "loss": 0.0891,
+      "recon_loss": 0.35486042499542236,
+      "recon_trunc": 0.0,
+      "step": 1195,
+      "student_loss": 0.006026371382176876,
+      "teacher_loss": 0.0035609882324934006
+    },
+    {
+      "epoch": 0.34594594594594597,
+      "grad_norm": 1.5859375,
+      "kd_loss": 0.011818514205515385,
+      "learning_rate": 1.0193093938604627e-06,
+      "loss": 0.0969,
+      "recon_loss": 2.8682312965393066,
+      "recon_trunc": 0.0,
+      "step": 1200,
+      "student_loss": 0.005163633730262518,
+      "teacher_loss": 0.002091110683977604
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1500,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 300,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1200/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:536c3d53a07695cc7265e3fd9454ddd0c9804e9f9e367393c4c1e4c6530f3697
+size 5905

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/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/codi_recon_delta_1.5b_ls2_rw0.03/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/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1500/config.json

@@ -0,0 +1,58 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "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,
+  "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": 151936
+}

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:388fde385c7a9caab4e874ceebadb43d1d4d0775ab0fa5aa945056a5756ef197
+size 4539424633

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1500/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9bbf69501d45d07ca050732b88839cd16d99cb785e04d1dc756b4ccdb092b597
+size 3097074235

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1500/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4aa03284661460f7588cf2ed51cae6f25c4941e23d7976dea3db0885dadb428
+size 15429

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1500/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97017e17d6838d8a09f84b48e28a5938a238a3e50e6a6671b6a2439f52a40fa6
+size 15429

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1500/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5a6d9e5e8614e5b7bcfddbab25b719e621f1ff67f400bc9a73dc5af42659928
+size 15429

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1500/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:841e9f5079dd6c6b4131e29e2e887ebf784990e17a6d8d8c7131ffb11723d069
+size 15429

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:475539e52c6e83f2f5a22f5cabb5de882acee72c74676095cdf00d332b3d8c01
+size 1465

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/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/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1500/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74339125d1c951eff924a8924c572451fe9670e834978941b1b757921bc2b2a0
+size 9445953

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1500/tokenizer.json

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

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/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/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-1500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:536c3d53a07695cc7265e3fd9454ddd0c9804e9f9e367393c4c1e4c6530f3697
+size 5905

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/added_tokens.json

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

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/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/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/config.json

@@ -0,0 +1,58 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "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,
+  "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": 151936
+}

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:187f212a560244bf7c2f09792e78a9e50a0bf1fbaeec0a7ed53992f22790eb71
+size 4539424633

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bef820d756dca8e875e9d402f95e61cad73c4f1cfd8b0d4390488529f3f20f23
+size 3097074235

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4aa03284661460f7588cf2ed51cae6f25c4941e23d7976dea3db0885dadb428
+size 15429

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97017e17d6838d8a09f84b48e28a5938a238a3e50e6a6671b6a2439f52a40fa6
+size 15429

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5a6d9e5e8614e5b7bcfddbab25b719e621f1ff67f400bc9a73dc5af42659928
+size 15429

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:841e9f5079dd6c6b4131e29e2e887ebf784990e17a6d8d8c7131ffb11723d069
+size 15429

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a1cf7dc915fd8683200c4cab003d16bce9119c6dac8f6d21b4ba406ae5f3bfc
+size 1465

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/special_tokens_map.json

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

checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-300/thought_projector.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3188ed27f10bc655cc1acbc15d1a558b446ecbc20194a8ab932d062d05d357c7
+size 9445953