Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

added_tokens.json

@@ -0,0 +1,28 @@
+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

config.json

@@ -0,0 +1,38 @@
+{
+  "architectures": [
+    "Qwen3MoeForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "decoder_sparse_step": 1,
+  "eos_token_id": 151643,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 6144,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 48,
+  "mlp_only_layers": [],
+  "model_type": "qwen3_moe",
+  "moe_intermediate_size": 768,
+  "norm_topk_prob": true,
+  "num_attention_heads": 32,
+  "num_experts": 128,
+  "num_experts_per_tok": 8,
+  "num_hidden_layers": 48,
+  "num_key_value_heads": 4,
+  "output_router_logits": true,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "router_aux_loss_coef": 0.001,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "max_new_tokens": 2048,
+  "transformers_version": "4.51.3"
+}

model-00001-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc8bcc80d6c7938ecd0dccebf2d00f2cb306a1db71a878d469ffb76bb1d644e5
+size 4997184968

model-00002-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38dc4f8705db32ea56072528ab809f29a9fdf5ccb8aa28b5506cc3feb18731fa
+size 4997741608

model-00003-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6924c71133835571e2dad55220a1e9821d56dab460dd1b008ba5a68a6b891831
+size 4997742208

model-00004-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8829bce6e602b9466d5b660724156dfdf8a758ed7cedd774cf769621032c57d9
+size 4997743184

model-00005-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:460ff56a588e058bc5f798e8b43e566dba8d562a579d94e47b33ff1af5a6b934
+size 4997743184

model-00006-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c92fe24f23c70909b80757fdad3847365715533b3d57dba0fde20b2902edb068
+size 4997743184

model-00007-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f1926017288e00cf59bc1e65cce0f29d3eec764804559e5bf9e30e5b5640aeb
+size 4997743184

model-00008-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce6d3519c21ac629d398ea6c9a1aabd13051fe0012e5ffb0518ee8dc58cd6514
+size 4997743184

model-00009-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d4e9f336c46cb12170d19b93fbd43eda11b672d25c9ba3bc9737c929093b9cd
+size 4997743184

model-00010-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6ed80695c9646a4cfaa83e9c104266e6bf46454072b025ef530aeaef252f781
+size 4997743184

model-00011-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb7986bd248f37e978aee61a6b8250d07c14de73f1f258c095a1baef019c3169
+size 4997743184

model-00012-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9ec82ad53662effac24435e3ba06169f73087b0a48c04fc66e3a7d0f074772f
+size 4997743184

model-00013-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9b617ad84db7e8842f0b66ce15f196f95b5b048ba90be6f1b609555d9a26adb
+size 1094220288

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": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

tokenizer_config.json

@@ -0,0 +1,241 @@
+{
+  "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": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "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,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {{- messages[0].content + '\\n\\n' }}\n    {%- endif %}\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>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\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\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- if ns.multi_step_tool and message.role == \"user\" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set content = message.content %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in message.content %}\n                {%- set content = message.content.split('</think>')[-1].lstrip('\\n') %}\n                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n    {%- if enable_thinking is defined and enable_thinking is false %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_state.json

@@ -0,0 +1,2029 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 285,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.017543859649122806,
+      "grad_norm": 2.1743679382834613,
+      "learning_rate": 0.0,
+      "loss": 0.8567,
+      "step": 1
+    },
+    {
+      "epoch": 0.03508771929824561,
+      "grad_norm": 2.065936532100158,
+      "learning_rate": 6.666666666666667e-07,
+      "loss": 0.8846,
+      "step": 2
+    },
+    {
+      "epoch": 0.05263157894736842,
+      "grad_norm": 2.2592127802227417,
+      "learning_rate": 1.3333333333333334e-06,
+      "loss": 1.0293,
+      "step": 3
+    },
+    {
+      "epoch": 0.07017543859649122,
+      "grad_norm": 2.3463436185497395,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.0703,
+      "step": 4
+    },
+    {
+      "epoch": 0.08771929824561403,
+      "grad_norm": 2.1002183137442683,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 0.8644,
+      "step": 5
+    },
+    {
+      "epoch": 0.10526315789473684,
+      "grad_norm": 2.0847937124442115,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 1.0164,
+      "step": 6
+    },
+    {
+      "epoch": 0.12280701754385964,
+      "grad_norm": 1.7185938121052617,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.0055,
+      "step": 7
+    },
+    {
+      "epoch": 0.14035087719298245,
+      "grad_norm": 1.7374042391478925,
+      "learning_rate": 4.666666666666667e-06,
+      "loss": 0.9546,
+      "step": 8
+    },
+    {
+      "epoch": 0.15789473684210525,
+      "grad_norm": 1.5314517018400144,
+      "learning_rate": 5.333333333333334e-06,
+      "loss": 0.9692,
+      "step": 9
+    },
+    {
+      "epoch": 0.17543859649122806,
+      "grad_norm": 1.2378039020716545,
+      "learning_rate": 6e-06,
+      "loss": 0.9946,
+      "step": 10
+    },
+    {
+      "epoch": 0.19298245614035087,
+      "grad_norm": 1.29478501876491,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 1.0377,
+      "step": 11
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 1.4543552489625455,
+      "learning_rate": 7.333333333333333e-06,
+      "loss": 0.9065,
+      "step": 12
+    },
+    {
+      "epoch": 0.22807017543859648,
+      "grad_norm": 1.4859158978308853,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.9426,
+      "step": 13
+    },
+    {
+      "epoch": 0.24561403508771928,
+      "grad_norm": 1.437617755966179,
+      "learning_rate": 8.666666666666668e-06,
+      "loss": 0.9205,
+      "step": 14
+    },
+    {
+      "epoch": 0.2631578947368421,
+      "grad_norm": 1.280823019336999,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.9789,
+      "step": 15
+    },
+    {
+      "epoch": 0.2807017543859649,
+      "grad_norm": 1.3096294502362182,
+      "learning_rate": 1e-05,
+      "loss": 0.93,
+      "step": 16
+    },
+    {
+      "epoch": 0.2982456140350877,
+      "grad_norm": 1.3902944737940592,
+      "learning_rate": 9.999661540018812e-06,
+      "loss": 1.0273,
+      "step": 17
+    },
+    {
+      "epoch": 0.3157894736842105,
+      "grad_norm": 1.3866472808835544,
+      "learning_rate": 9.99864620589731e-06,
+      "loss": 0.9108,
+      "step": 18
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 1.0976269555809974,
+      "learning_rate": 9.99695413509548e-06,
+      "loss": 0.9311,
+      "step": 19
+    },
+    {
+      "epoch": 0.3508771929824561,
+      "grad_norm": 0.9581466026044247,
+      "learning_rate": 9.994585556692624e-06,
+      "loss": 0.9664,
+      "step": 20
+    },
+    {
+      "epoch": 0.3684210526315789,
+      "grad_norm": 0.8742061741880427,
+      "learning_rate": 9.991540791356342e-06,
+      "loss": 0.9042,
+      "step": 21
+    },
+    {
+      "epoch": 0.38596491228070173,
+      "grad_norm": 0.8127242286364288,
+      "learning_rate": 9.987820251299121e-06,
+      "loss": 0.7719,
+      "step": 22
+    },
+    {
+      "epoch": 0.40350877192982454,
+      "grad_norm": 0.9218648947648101,
+      "learning_rate": 9.98342444022253e-06,
+      "loss": 0.9057,
+      "step": 23
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.7985103108366662,
+      "learning_rate": 9.978353953249023e-06,
+      "loss": 0.8032,
+      "step": 24
+    },
+    {
+      "epoch": 0.43859649122807015,
+      "grad_norm": 0.8519004929440739,
+      "learning_rate": 9.972609476841368e-06,
+      "loss": 0.9328,
+      "step": 25
+    },
+    {
+      "epoch": 0.45614035087719296,
+      "grad_norm": 0.9321320761136158,
+      "learning_rate": 9.966191788709716e-06,
+      "loss": 0.897,
+      "step": 26
+    },
+    {
+      "epoch": 0.47368421052631576,
+      "grad_norm": 1.1881269834015444,
+      "learning_rate": 9.959101757706308e-06,
+      "loss": 0.8169,
+      "step": 27
+    },
+    {
+      "epoch": 0.49122807017543857,
+      "grad_norm": 0.7693003217933029,
+      "learning_rate": 9.951340343707852e-06,
+      "loss": 0.8854,
+      "step": 28
+    },
+    {
+      "epoch": 0.5087719298245614,
+      "grad_norm": 0.6825700058048908,
+      "learning_rate": 9.942908597485558e-06,
+      "loss": 0.9067,
+      "step": 29
+    },
+    {
+      "epoch": 0.5263157894736842,
+      "grad_norm": 0.5987346031153414,
+      "learning_rate": 9.933807660562898e-06,
+      "loss": 0.7924,
+      "step": 30
+    },
+    {
+      "epoch": 0.543859649122807,
+      "grad_norm": 0.5548910090458355,
+      "learning_rate": 9.924038765061042e-06,
+      "loss": 0.8482,
+      "step": 31
+    },
+    {
+      "epoch": 0.5614035087719298,
+      "grad_norm": 0.5383543839380202,
+      "learning_rate": 9.913603233532067e-06,
+      "loss": 0.7789,
+      "step": 32
+    },
+    {
+      "epoch": 0.5789473684210527,
+      "grad_norm": 0.7299192422724775,
+      "learning_rate": 9.902502478779897e-06,
+      "loss": 0.8773,
+      "step": 33
+    },
+    {
+      "epoch": 0.5964912280701754,
+      "grad_norm": 0.6922732156258199,
+      "learning_rate": 9.890738003669029e-06,
+      "loss": 0.9766,
+      "step": 34
+    },
+    {
+      "epoch": 0.6140350877192983,
+      "grad_norm": 0.6246419836484991,
+      "learning_rate": 9.878311400921072e-06,
+      "loss": 0.8212,
+      "step": 35
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 1.1220803760331017,
+      "learning_rate": 9.86522435289912e-06,
+      "loss": 0.8362,
+      "step": 36
+    },
+    {
+      "epoch": 0.6491228070175439,
+      "grad_norm": 0.5667320037510573,
+      "learning_rate": 9.851478631379982e-06,
+      "loss": 0.8713,
+      "step": 37
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.5410348279430676,
+      "learning_rate": 9.83707609731432e-06,
+      "loss": 0.8842,
+      "step": 38
+    },
+    {
+      "epoch": 0.6842105263157895,
+      "grad_norm": 0.641889612832866,
+      "learning_rate": 9.822018700574696e-06,
+      "loss": 0.9068,
+      "step": 39
+    },
+    {
+      "epoch": 0.7017543859649122,
+      "grad_norm": 0.4916947719279244,
+      "learning_rate": 9.806308479691595e-06,
+      "loss": 0.8728,
+      "step": 40
+    },
+    {
+      "epoch": 0.7192982456140351,
+      "grad_norm": 0.5400784001882519,
+      "learning_rate": 9.789947561577445e-06,
+      "loss": 0.8773,
+      "step": 41
+    },
+    {
+      "epoch": 0.7368421052631579,
+      "grad_norm": 0.5399508837893761,
+      "learning_rate": 9.77293816123866e-06,
+      "loss": 0.8126,
+      "step": 42
+    },
+    {
+      "epoch": 0.7543859649122807,
+      "grad_norm": 0.5248163734594365,
+      "learning_rate": 9.755282581475769e-06,
+      "loss": 0.8451,
+      "step": 43
+    },
+    {
+      "epoch": 0.7719298245614035,
+      "grad_norm": 0.5195803755313205,
+      "learning_rate": 9.736983212571646e-06,
+      "loss": 0.7615,
+      "step": 44
+    },
+    {
+      "epoch": 0.7894736842105263,
+      "grad_norm": 0.49652590728770457,
+      "learning_rate": 9.718042531967918e-06,
+      "loss": 0.8785,
+      "step": 45
+    },
+    {
+      "epoch": 0.8070175438596491,
+      "grad_norm": 0.5477983890922427,
+      "learning_rate": 9.698463103929542e-06,
+      "loss": 0.8549,
+      "step": 46
+    },
+    {
+      "epoch": 0.8245614035087719,
+      "grad_norm": 0.6631860895060182,
+      "learning_rate": 9.678247579197658e-06,
+      "loss": 0.8591,
+      "step": 47
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.4379914410432218,
+      "learning_rate": 9.657398694630713e-06,
+      "loss": 0.8438,
+      "step": 48
+    },
+    {
+      "epoch": 0.8596491228070176,
+      "grad_norm": 0.5118173798397813,
+      "learning_rate": 9.635919272833938e-06,
+      "loss": 0.9355,
+      "step": 49
+    },
+    {
+      "epoch": 0.8771929824561403,
+      "grad_norm": 0.4499149257988012,
+      "learning_rate": 9.613812221777212e-06,
+      "loss": 0.6753,
+      "step": 50
+    },
+    {
+      "epoch": 0.8947368421052632,
+      "grad_norm": 0.4918824458174192,
+      "learning_rate": 9.591080534401371e-06,
+      "loss": 0.9046,
+      "step": 51
+    },
+    {
+      "epoch": 0.9122807017543859,
+      "grad_norm": 0.48310664704359907,
+      "learning_rate": 9.567727288213005e-06,
+      "loss": 0.8728,
+      "step": 52
+    },
+    {
+      "epoch": 0.9298245614035088,
+      "grad_norm": 0.44537325137357436,
+      "learning_rate": 9.543755644867823e-06,
+      "loss": 0.8243,
+      "step": 53
+    },
+    {
+      "epoch": 0.9473684210526315,
+      "grad_norm": 0.49791673972055667,
+      "learning_rate": 9.519168849742603e-06,
+      "loss": 0.7796,
+      "step": 54
+    },
+    {
+      "epoch": 0.9649122807017544,
+      "grad_norm": 0.49148133350951356,
+      "learning_rate": 9.493970231495836e-06,
+      "loss": 0.8186,
+      "step": 55
+    },
+    {
+      "epoch": 0.9824561403508771,
+      "grad_norm": 0.4772135056244933,
+      "learning_rate": 9.468163201617063e-06,
+      "loss": 0.8303,
+      "step": 56
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5146092970588294,
+      "learning_rate": 9.441751253965022e-06,
+      "loss": 0.7413,
+      "step": 57
+    },
+    {
+      "epoch": 1.0175438596491229,
+      "grad_norm": 0.5707299611511895,
+      "learning_rate": 9.414737964294636e-06,
+      "loss": 0.8499,
+      "step": 58
+    },
+    {
+      "epoch": 1.0350877192982457,
+      "grad_norm": 0.5306849780410736,
+      "learning_rate": 9.38712698977291e-06,
+      "loss": 0.739,
+      "step": 59
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 0.5223538957045127,
+      "learning_rate": 9.358922068483813e-06,
+      "loss": 0.8245,
+      "step": 60
+    },
+    {
+      "epoch": 1.0701754385964912,
+      "grad_norm": 0.5042070796385456,
+      "learning_rate": 9.330127018922195e-06,
+      "loss": 0.7004,
+      "step": 61
+    },
+    {
+      "epoch": 1.087719298245614,
+      "grad_norm": 0.5009488991267903,
+      "learning_rate": 9.30074573947683e-06,
+      "loss": 0.7844,
+      "step": 62
+    },
+    {
+      "epoch": 1.1052631578947367,
+      "grad_norm": 0.6087253771381698,
+      "learning_rate": 9.27078220790263e-06,
+      "loss": 0.8109,
+      "step": 63
+    },
+    {
+      "epoch": 1.1228070175438596,
+      "grad_norm": 0.5456945142498564,
+      "learning_rate": 9.24024048078213e-06,
+      "loss": 0.8035,
+      "step": 64
+    },
+    {
+      "epoch": 1.1403508771929824,
+      "grad_norm": 0.44839717791010614,
+      "learning_rate": 9.209124692976287e-06,
+      "loss": 0.7257,
+      "step": 65
+    },
+    {
+      "epoch": 1.1578947368421053,
+      "grad_norm": 0.5097655229916522,
+      "learning_rate": 9.177439057064684e-06,
+      "loss": 0.7281,
+      "step": 66
+    },
+    {
+      "epoch": 1.1754385964912282,
+      "grad_norm": 0.517921546215267,
+      "learning_rate": 9.145187862775208e-06,
+      "loss": 0.7249,
+      "step": 67
+    },
+    {
+      "epoch": 1.1929824561403508,
+      "grad_norm": 0.47990375795295936,
+      "learning_rate": 9.112375476403313e-06,
+      "loss": 0.7714,
+      "step": 68
+    },
+    {
+      "epoch": 1.2105263157894737,
+      "grad_norm": 0.5045141372712146,
+      "learning_rate": 9.079006340220862e-06,
+      "loss": 0.8193,
+      "step": 69
+    },
+    {
+      "epoch": 1.2280701754385965,
+      "grad_norm": 0.5804566280703113,
+      "learning_rate": 9.045084971874738e-06,
+      "loss": 0.7695,
+      "step": 70
+    },
+    {
+      "epoch": 1.2456140350877192,
+      "grad_norm": 0.46340512291170205,
+      "learning_rate": 9.01061596377522e-06,
+      "loss": 0.8198,
+      "step": 71
+    },
+    {
+      "epoch": 1.263157894736842,
+      "grad_norm": 0.5509170504473769,
+      "learning_rate": 8.97560398247424e-06,
+      "loss": 0.7834,
+      "step": 72
+    },
+    {
+      "epoch": 1.280701754385965,
+      "grad_norm": 0.4925271038625461,
+      "learning_rate": 8.94005376803361e-06,
+      "loss": 0.7842,
+      "step": 73
+    },
+    {
+      "epoch": 1.2982456140350878,
+      "grad_norm": 0.4623728941867384,
+      "learning_rate": 8.903970133383297e-06,
+      "loss": 0.7045,
+      "step": 74
+    },
+    {
+      "epoch": 1.3157894736842106,
+      "grad_norm": 0.4622528793181321,
+      "learning_rate": 8.867357963669821e-06,
+      "loss": 0.8323,
+      "step": 75
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.44017925231276656,
+      "learning_rate": 8.83022221559489e-06,
+      "loss": 0.7274,
+      "step": 76
+    },
+    {
+      "epoch": 1.3508771929824561,
+      "grad_norm": 0.4855447339968991,
+      "learning_rate": 8.792567916744346e-06,
+      "loss": 0.6991,
+      "step": 77
+    },
+    {
+      "epoch": 1.368421052631579,
+      "grad_norm": 0.5862777150363095,
+      "learning_rate": 8.754400164907496e-06,
+      "loss": 0.8171,
+      "step": 78
+    },
+    {
+      "epoch": 1.3859649122807016,
+      "grad_norm": 0.5689053746090412,
+      "learning_rate": 8.715724127386971e-06,
+      "loss": 0.8002,
+      "step": 79
+    },
+    {
+      "epoch": 1.4035087719298245,
+      "grad_norm": 0.4743146314442524,
+      "learning_rate": 8.676545040299145e-06,
+      "loss": 0.7492,
+      "step": 80
+    },
+    {
+      "epoch": 1.4210526315789473,
+      "grad_norm": 0.4496553971365961,
+      "learning_rate": 8.636868207865244e-06,
+      "loss": 0.7745,
+      "step": 81
+    },
+    {
+      "epoch": 1.4385964912280702,
+      "grad_norm": 0.4841172726306871,
+      "learning_rate": 8.596699001693257e-06,
+      "loss": 0.759,
+      "step": 82
+    },
+    {
+      "epoch": 1.456140350877193,
+      "grad_norm": 0.5074636296001964,
+      "learning_rate": 8.556042860050686e-06,
+      "loss": 0.6932,
+      "step": 83
+    },
+    {
+      "epoch": 1.4736842105263157,
+      "grad_norm": 0.4644953408415979,
+      "learning_rate": 8.51490528712831e-06,
+      "loss": 0.6223,
+      "step": 84
+    },
+    {
+      "epoch": 1.4912280701754386,
+      "grad_norm": 0.4828551950683606,
+      "learning_rate": 8.473291852294986e-06,
+      "loss": 0.7761,
+      "step": 85
+    },
+    {
+      "epoch": 1.5087719298245614,
+      "grad_norm": 0.5143431748577654,
+      "learning_rate": 8.43120818934367e-06,
+      "loss": 0.7623,
+      "step": 86
+    },
+    {
+      "epoch": 1.526315789473684,
+      "grad_norm": 0.45414594274680353,
+      "learning_rate": 8.388659995728662e-06,
+      "loss": 0.8921,
+      "step": 87
+    },
+    {
+      "epoch": 1.543859649122807,
+      "grad_norm": 0.5858756826418567,
+      "learning_rate": 8.345653031794292e-06,
+      "loss": 0.778,
+      "step": 88
+    },
+    {
+      "epoch": 1.5614035087719298,
+      "grad_norm": 0.4357794588497438,
+      "learning_rate": 8.302193119995038e-06,
+      "loss": 0.8172,
+      "step": 89
+    },
+    {
+      "epoch": 1.5789473684210527,
+      "grad_norm": 0.4714293820655886,
+      "learning_rate": 8.258286144107277e-06,
+      "loss": 0.7186,
+      "step": 90
+    },
+    {
+      "epoch": 1.5964912280701755,
+      "grad_norm": 0.5293457619236406,
+      "learning_rate": 8.213938048432697e-06,
+      "loss": 0.754,
+      "step": 91
+    },
+    {
+      "epoch": 1.6140350877192984,
+      "grad_norm": 0.43517623229684854,
+      "learning_rate": 8.16915483699355e-06,
+      "loss": 0.7654,
+      "step": 92
+    },
+    {
+      "epoch": 1.631578947368421,
+      "grad_norm": 0.40391894529350186,
+      "learning_rate": 8.123942572719801e-06,
+      "loss": 0.7036,
+      "step": 93
+    },
+    {
+      "epoch": 1.6491228070175439,
+      "grad_norm": 0.4114610381039708,
+      "learning_rate": 8.078307376628292e-06,
+      "loss": 0.7112,
+      "step": 94
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.5229223132781197,
+      "learning_rate": 8.032255426994069e-06,
+      "loss": 0.8081,
+      "step": 95
+    },
+    {
+      "epoch": 1.6842105263157894,
+      "grad_norm": 0.4392295140626043,
+      "learning_rate": 7.985792958513932e-06,
+      "loss": 0.713,
+      "step": 96
+    },
+    {
+      "epoch": 1.7017543859649122,
+      "grad_norm": 0.4674308378876463,
+      "learning_rate": 7.938926261462366e-06,
+      "loss": 0.879,
+      "step": 97
+    },
+    {
+      "epoch": 1.719298245614035,
+      "grad_norm": 0.4354904050679196,
+      "learning_rate": 7.891661680839932e-06,
+      "loss": 0.758,
+      "step": 98
+    },
+    {
+      "epoch": 1.736842105263158,
+      "grad_norm": 0.4710923835605549,
+      "learning_rate": 7.84400561551426e-06,
+      "loss": 0.7942,
+      "step": 99
+    },
+    {
+      "epoch": 1.7543859649122808,
+      "grad_norm": 0.4541845202770017,
+      "learning_rate": 7.795964517353734e-06,
+      "loss": 0.7364,
+      "step": 100
+    },
+    {
+      "epoch": 1.7719298245614035,
+      "grad_norm": 0.5436400805712097,
+      "learning_rate": 7.747544890354031e-06,
+      "loss": 0.8059,
+      "step": 101
+    },
+    {
+      "epoch": 1.7894736842105263,
+      "grad_norm": 0.4519039016274186,
+      "learning_rate": 7.698753289757565e-06,
+      "loss": 0.8054,
+      "step": 102
+    },
+    {
+      "epoch": 1.807017543859649,
+      "grad_norm": 0.535411449854802,
+      "learning_rate": 7.649596321166024e-06,
+      "loss": 0.8224,
+      "step": 103
+    },
+    {
+      "epoch": 1.8245614035087718,
+      "grad_norm": 0.452937347654015,
+      "learning_rate": 7.600080639646077e-06,
+      "loss": 0.6789,
+      "step": 104
+    },
+    {
+      "epoch": 1.8421052631578947,
+      "grad_norm": 0.4632987744077176,
+      "learning_rate": 7.550212948828377e-06,
+      "loss": 0.6507,
+      "step": 105
+    },
+    {
+      "epoch": 1.8596491228070176,
+      "grad_norm": 0.5846132881467483,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 0.8404,
+      "step": 106
+    },
+    {
+      "epoch": 1.8771929824561404,
+      "grad_norm": 0.5221424183239722,
+      "learning_rate": 7.449448591190436e-06,
+      "loss": 0.7738,
+      "step": 107
+    },
+    {
+      "epoch": 1.8947368421052633,
+      "grad_norm": 0.4780212894923245,
+      "learning_rate": 7.398565566251232e-06,
+      "loss": 0.7942,
+      "step": 108
+    },
+    {
+      "epoch": 1.912280701754386,
+      "grad_norm": 0.4498797772942343,
+      "learning_rate": 7.347357813929455e-06,
+      "loss": 0.7079,
+      "step": 109
+    },
+    {
+      "epoch": 1.9298245614035088,
+      "grad_norm": 0.5134358225643797,
+      "learning_rate": 7.295832266935059e-06,
+      "loss": 0.8141,
+      "step": 110
+    },
+    {
+      "epoch": 1.9473684210526314,
+      "grad_norm": 0.4543724828212894,
+      "learning_rate": 7.243995901002312e-06,
+      "loss": 0.8115,
+      "step": 111
+    },
+    {
+      "epoch": 1.9649122807017543,
+      "grad_norm": 0.5432744513400934,
+      "learning_rate": 7.191855733945388e-06,
+      "loss": 0.7811,
+      "step": 112
+    },
+    {
+      "epoch": 1.9824561403508771,
+      "grad_norm": 0.4818873923535046,
+      "learning_rate": 7.1394188247082715e-06,
+      "loss": 0.8568,
+      "step": 113
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6085310263030772,
+      "learning_rate": 7.08669227240909e-06,
+      "loss": 0.7619,
+      "step": 114
+    },
+    {
+      "epoch": 2.017543859649123,
+      "grad_norm": 0.7002453670731734,
+      "learning_rate": 7.033683215379002e-06,
+      "loss": 0.638,
+      "step": 115
+    },
+    {
+      "epoch": 2.0350877192982457,
+      "grad_norm": 0.46139309364956993,
+      "learning_rate": 6.980398830195785e-06,
+      "loss": 0.6779,
+      "step": 116
+    },
+    {
+      "epoch": 2.0526315789473686,
+      "grad_norm": 0.6136343913874313,
+      "learning_rate": 6.9268463307122425e-06,
+      "loss": 0.8546,
+      "step": 117
+    },
+    {
+      "epoch": 2.0701754385964914,
+      "grad_norm": 1.114800056976964,
+      "learning_rate": 6.873032967079562e-06,
+      "loss": 0.719,
+      "step": 118
+    },
+    {
+      "epoch": 2.087719298245614,
+      "grad_norm": 0.6214013908852314,
+      "learning_rate": 6.818966024765758e-06,
+      "loss": 0.6997,
+      "step": 119
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 0.5200928133155619,
+      "learning_rate": 6.7646528235693445e-06,
+      "loss": 0.6798,
+      "step": 120
+    },
+    {
+      "epoch": 2.1228070175438596,
+      "grad_norm": 0.7091096851850384,
+      "learning_rate": 6.710100716628345e-06,
+      "loss": 0.673,
+      "step": 121
+    },
+    {
+      "epoch": 2.1403508771929824,
+      "grad_norm": 0.7985623618136684,
+      "learning_rate": 6.655317089424791e-06,
+      "loss": 0.6277,
+      "step": 122
+    },
+    {
+      "epoch": 2.1578947368421053,
+      "grad_norm": 0.5548347309189483,
+      "learning_rate": 6.600309358784858e-06,
+      "loss": 0.7752,
+      "step": 123
+    },
+    {
+      "epoch": 2.175438596491228,
+      "grad_norm": 0.5579527113721896,
+      "learning_rate": 6.545084971874738e-06,
+      "loss": 0.6762,
+      "step": 124
+    },
+    {
+      "epoch": 2.192982456140351,
+      "grad_norm": 0.9075303715532044,
+      "learning_rate": 6.48965140519241e-06,
+      "loss": 0.7744,
+      "step": 125
+    },
+    {
+      "epoch": 2.2105263157894735,
+      "grad_norm": 0.6426858187075343,
+      "learning_rate": 6.434016163555452e-06,
+      "loss": 0.6827,
+      "step": 126
+    },
+    {
+      "epoch": 2.2280701754385963,
+      "grad_norm": 0.5482869973492827,
+      "learning_rate": 6.378186779084996e-06,
+      "loss": 0.6293,
+      "step": 127
+    },
+    {
+      "epoch": 2.245614035087719,
+      "grad_norm": 0.5393148541564667,
+      "learning_rate": 6.322170810186013e-06,
+      "loss": 0.6549,
+      "step": 128
+    },
+    {
+      "epoch": 2.263157894736842,
+      "grad_norm": 0.5979082946371272,
+      "learning_rate": 6.26597584052401e-06,
+      "loss": 0.701,
+      "step": 129
+    },
+    {
+      "epoch": 2.280701754385965,
+      "grad_norm": 0.5769890620947821,
+      "learning_rate": 6.209609477998339e-06,
+      "loss": 0.7124,
+      "step": 130
+    },
+    {
+      "epoch": 2.2982456140350878,
+      "grad_norm": 0.5361207762048112,
+      "learning_rate": 6.153079353712201e-06,
+      "loss": 0.693,
+      "step": 131
+    },
+    {
+      "epoch": 2.3157894736842106,
+      "grad_norm": 0.4698632877266117,
+      "learning_rate": 6.0963931209395165e-06,
+      "loss": 0.6544,
+      "step": 132
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.7402806515712913,
+      "learning_rate": 6.039558454088796e-06,
+      "loss": 0.7508,
+      "step": 133
+    },
+    {
+      "epoch": 2.3508771929824563,
+      "grad_norm": 0.5887309002991683,
+      "learning_rate": 5.982583047664151e-06,
+      "loss": 0.6627,
+      "step": 134
+    },
+    {
+      "epoch": 2.3684210526315788,
+      "grad_norm": 0.5768193739422414,
+      "learning_rate": 5.925474615223573e-06,
+      "loss": 0.6794,
+      "step": 135
+    },
+    {
+      "epoch": 2.3859649122807016,
+      "grad_norm": 0.4900932586342072,
+      "learning_rate": 5.8682408883346535e-06,
+      "loss": 0.6742,
+      "step": 136
+    },
+    {
+      "epoch": 2.4035087719298245,
+      "grad_norm": 0.5909882725589083,
+      "learning_rate": 5.810889615527839e-06,
+      "loss": 0.7839,
+      "step": 137
+    },
+    {
+      "epoch": 2.4210526315789473,
+      "grad_norm": 0.5356191845844215,
+      "learning_rate": 5.753428561247416e-06,
+      "loss": 0.7685,
+      "step": 138
+    },
+    {
+      "epoch": 2.43859649122807,
+      "grad_norm": 0.5557393095861676,
+      "learning_rate": 5.695865504800328e-06,
+      "loss": 0.7255,
+      "step": 139
+    },
+    {
+      "epoch": 2.456140350877193,
+      "grad_norm": 0.4321783466948419,
+      "learning_rate": 5.638208239302975e-06,
+      "loss": 0.6827,
+      "step": 140
+    },
+    {
+      "epoch": 2.473684210526316,
+      "grad_norm": 0.5450014904476171,
+      "learning_rate": 5.5804645706261515e-06,
+      "loss": 0.7264,
+      "step": 141
+    },
+    {
+      "epoch": 2.4912280701754383,
+      "grad_norm": 0.49737092249787296,
+      "learning_rate": 5.522642316338268e-06,
+      "loss": 0.7297,
+      "step": 142
+    },
+    {
+      "epoch": 2.5087719298245617,
+      "grad_norm": 0.5031219221886872,
+      "learning_rate": 5.464749304646963e-06,
+      "loss": 0.578,
+      "step": 143
+    },
+    {
+      "epoch": 2.526315789473684,
+      "grad_norm": 0.4608703470897519,
+      "learning_rate": 5.406793373339292e-06,
+      "loss": 0.6669,
+      "step": 144
+    },
+    {
+      "epoch": 2.543859649122807,
+      "grad_norm": 0.4161725717756173,
+      "learning_rate": 5.348782368720627e-06,
+      "loss": 0.6811,
+      "step": 145
+    },
+    {
+      "epoch": 2.56140350877193,
+      "grad_norm": 0.6000163863512125,
+      "learning_rate": 5.290724144552379e-06,
+      "loss": 0.7432,
+      "step": 146
+    },
+    {
+      "epoch": 2.5789473684210527,
+      "grad_norm": 0.4603282356714494,
+      "learning_rate": 5.232626560988735e-06,
+      "loss": 0.6819,
+      "step": 147
+    },
+    {
+      "epoch": 2.5964912280701755,
+      "grad_norm": 0.4485140646183522,
+      "learning_rate": 5.174497483512506e-06,
+      "loss": 0.6995,
+      "step": 148
+    },
+    {
+      "epoch": 2.6140350877192984,
+      "grad_norm": 0.5294130333173804,
+      "learning_rate": 5.116344781870282e-06,
+      "loss": 0.6738,
+      "step": 149
+    },
+    {
+      "epoch": 2.6315789473684212,
+      "grad_norm": 1.674919136856814,
+      "learning_rate": 5.0581763290069865e-06,
+      "loss": 0.7646,
+      "step": 150
+    },
+    {
+      "epoch": 2.6491228070175437,
+      "grad_norm": 0.4555731387212146,
+      "learning_rate": 5e-06,
+      "loss": 0.7391,
+      "step": 151
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.5056597113054125,
+      "learning_rate": 4.941823670993016e-06,
+      "loss": 0.7389,
+      "step": 152
+    },
+    {
+      "epoch": 2.6842105263157894,
+      "grad_norm": 0.7379235969983658,
+      "learning_rate": 4.883655218129719e-06,
+      "loss": 0.7339,
+      "step": 153
+    },
+    {
+      "epoch": 2.7017543859649122,
+      "grad_norm": 0.5895150097062052,
+      "learning_rate": 4.825502516487497e-06,
+      "loss": 0.7294,
+      "step": 154
+    },
+    {
+      "epoch": 2.719298245614035,
+      "grad_norm": 0.4744494071190993,
+      "learning_rate": 4.767373439011267e-06,
+      "loss": 0.7421,
+      "step": 155
+    },
+    {
+      "epoch": 2.736842105263158,
+      "grad_norm": 0.6393532113710564,
+      "learning_rate": 4.7092758554476215e-06,
+      "loss": 0.6823,
+      "step": 156
+    },
+    {
+      "epoch": 2.754385964912281,
+      "grad_norm": 0.43542542519160277,
+      "learning_rate": 4.651217631279374e-06,
+      "loss": 0.6999,
+      "step": 157
+    },
+    {
+      "epoch": 2.7719298245614032,
+      "grad_norm": 0.4834621655771387,
+      "learning_rate": 4.59320662666071e-06,
+      "loss": 0.6113,
+      "step": 158
+    },
+    {
+      "epoch": 2.7894736842105265,
+      "grad_norm": 0.43827824482397315,
+      "learning_rate": 4.53525069535304e-06,
+      "loss": 0.6755,
+      "step": 159
+    },
+    {
+      "epoch": 2.807017543859649,
+      "grad_norm": 0.5114651303995619,
+      "learning_rate": 4.477357683661734e-06,
+      "loss": 0.5774,
+      "step": 160
+    },
+    {
+      "epoch": 2.824561403508772,
+      "grad_norm": 0.49630882513160696,
+      "learning_rate": 4.4195354293738484e-06,
+      "loss": 0.683,
+      "step": 161
+    },
+    {
+      "epoch": 2.8421052631578947,
+      "grad_norm": 0.5496029014377304,
+      "learning_rate": 4.361791760697027e-06,
+      "loss": 0.712,
+      "step": 162
+    },
+    {
+      "epoch": 2.8596491228070176,
+      "grad_norm": 0.4542060270968721,
+      "learning_rate": 4.304134495199675e-06,
+      "loss": 0.7218,
+      "step": 163
+    },
+    {
+      "epoch": 2.8771929824561404,
+      "grad_norm": 0.42080888911231823,
+      "learning_rate": 4.246571438752585e-06,
+      "loss": 0.6948,
+      "step": 164
+    },
+    {
+      "epoch": 2.8947368421052633,
+      "grad_norm": 0.4962570475835595,
+      "learning_rate": 4.189110384472164e-06,
+      "loss": 0.7592,
+      "step": 165
+    },
+    {
+      "epoch": 2.912280701754386,
+      "grad_norm": 0.431313111964283,
+      "learning_rate": 4.131759111665349e-06,
+      "loss": 0.6944,
+      "step": 166
+    },
+    {
+      "epoch": 2.9298245614035086,
+      "grad_norm": 0.4825307197434815,
+      "learning_rate": 4.074525384776428e-06,
+      "loss": 0.6351,
+      "step": 167
+    },
+    {
+      "epoch": 2.9473684210526314,
+      "grad_norm": 0.4499086142393376,
+      "learning_rate": 4.017416952335849e-06,
+      "loss": 0.6754,
+      "step": 168
+    },
+    {
+      "epoch": 2.9649122807017543,
+      "grad_norm": 0.43765899419426385,
+      "learning_rate": 3.960441545911205e-06,
+      "loss": 0.7014,
+      "step": 169
+    },
+    {
+      "epoch": 2.982456140350877,
+      "grad_norm": 0.46206138256747586,
+      "learning_rate": 3.903606879060483e-06,
+      "loss": 0.7157,
+      "step": 170
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.5575074870476582,
+      "learning_rate": 3.8469206462878e-06,
+      "loss": 0.5412,
+      "step": 171
+    },
+    {
+      "epoch": 3.017543859649123,
+      "grad_norm": 0.8960891403260335,
+      "learning_rate": 3.790390522001662e-06,
+      "loss": 0.5653,
+      "step": 172
+    },
+    {
+      "epoch": 3.0350877192982457,
+      "grad_norm": 0.429394565133977,
+      "learning_rate": 3.7340241594759917e-06,
+      "loss": 0.6847,
+      "step": 173
+    },
+    {
+      "epoch": 3.0526315789473686,
+      "grad_norm": 0.7458517128068287,
+      "learning_rate": 3.6778291898139907e-06,
+      "loss": 0.715,
+      "step": 174
+    },
+    {
+      "epoch": 3.0701754385964914,
+      "grad_norm": 0.7909906043291404,
+      "learning_rate": 3.6218132209150047e-06,
+      "loss": 0.7003,
+      "step": 175
+    },
+    {
+      "epoch": 3.087719298245614,
+      "grad_norm": 0.6830208055157037,
+      "learning_rate": 3.5659838364445505e-06,
+      "loss": 0.626,
+      "step": 176
+    },
+    {
+      "epoch": 3.1052631578947367,
+      "grad_norm": 0.5905786837952531,
+      "learning_rate": 3.51034859480759e-06,
+      "loss": 0.7365,
+      "step": 177
+    },
+    {
+      "epoch": 3.1228070175438596,
+      "grad_norm": 0.5503230235862585,
+      "learning_rate": 3.4549150281252635e-06,
+      "loss": 0.6986,
+      "step": 178
+    },
+    {
+      "epoch": 3.1403508771929824,
+      "grad_norm": 0.5423276784690628,
+      "learning_rate": 3.399690641215142e-06,
+      "loss": 0.6015,
+      "step": 179
+    },
+    {
+      "epoch": 3.1578947368421053,
+      "grad_norm": 0.7391674737539429,
+      "learning_rate": 3.3446829105752103e-06,
+      "loss": 0.6067,
+      "step": 180
+    },
+    {
+      "epoch": 3.175438596491228,
+      "grad_norm": 0.5210937495443324,
+      "learning_rate": 3.289899283371657e-06,
+      "loss": 0.6466,
+      "step": 181
+    },
+    {
+      "epoch": 3.192982456140351,
+      "grad_norm": 0.4941600655911879,
+      "learning_rate": 3.2353471764306567e-06,
+      "loss": 0.7822,
+      "step": 182
+    },
+    {
+      "epoch": 3.2105263157894735,
+      "grad_norm": 0.5391028040981926,
+      "learning_rate": 3.1810339752342446e-06,
+      "loss": 0.688,
+      "step": 183
+    },
+    {
+      "epoch": 3.2280701754385963,
+      "grad_norm": 0.6488593915381242,
+      "learning_rate": 3.12696703292044e-06,
+      "loss": 0.6278,
+      "step": 184
+    },
+    {
+      "epoch": 3.245614035087719,
+      "grad_norm": 0.42511310632264826,
+      "learning_rate": 3.0731536692877596e-06,
+      "loss": 0.6169,
+      "step": 185
+    },
+    {
+      "epoch": 3.263157894736842,
+      "grad_norm": 0.6834526113797037,
+      "learning_rate": 3.019601169804216e-06,
+      "loss": 0.5514,
+      "step": 186
+    },
+    {
+      "epoch": 3.280701754385965,
+      "grad_norm": 0.4857273254381776,
+      "learning_rate": 2.966316784621e-06,
+      "loss": 0.6412,
+      "step": 187
+    },
+    {
+      "epoch": 3.2982456140350878,
+      "grad_norm": 0.47134283803555077,
+      "learning_rate": 2.9133077275909112e-06,
+      "loss": 0.7009,
+      "step": 188
+    },
+    {
+      "epoch": 3.3157894736842106,
+      "grad_norm": 0.5879111417559143,
+      "learning_rate": 2.86058117529173e-06,
+      "loss": 0.6108,
+      "step": 189
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.5252081147359309,
+      "learning_rate": 2.8081442660546126e-06,
+      "loss": 0.6227,
+      "step": 190
+    },
+    {
+      "epoch": 3.3508771929824563,
+      "grad_norm": 0.5056820590798169,
+      "learning_rate": 2.7560040989976894e-06,
+      "loss": 0.654,
+      "step": 191
+    },
+    {
+      "epoch": 3.3684210526315788,
+      "grad_norm": 0.4958207647868106,
+      "learning_rate": 2.7041677330649408e-06,
+      "loss": 0.5542,
+      "step": 192
+    },
+    {
+      "epoch": 3.3859649122807016,
+      "grad_norm": 0.42921614594630214,
+      "learning_rate": 2.6526421860705474e-06,
+      "loss": 0.5723,
+      "step": 193
+    },
+    {
+      "epoch": 3.4035087719298245,
+      "grad_norm": 0.47547990937480933,
+      "learning_rate": 2.601434433748771e-06,
+      "loss": 0.6057,
+      "step": 194
+    },
+    {
+      "epoch": 3.4210526315789473,
+      "grad_norm": 0.9483349883795978,
+      "learning_rate": 2.550551408809566e-06,
+      "loss": 0.5332,
+      "step": 195
+    },
+    {
+      "epoch": 3.43859649122807,
+      "grad_norm": 0.530787300623137,
+      "learning_rate": 2.5000000000000015e-06,
+      "loss": 0.6533,
+      "step": 196
+    },
+    {
+      "epoch": 3.456140350877193,
+      "grad_norm": 0.6001850292167885,
+      "learning_rate": 2.4497870511716237e-06,
+      "loss": 0.4865,
+      "step": 197
+    },
+    {
+      "epoch": 3.473684210526316,
+      "grad_norm": 0.48448562586343535,
+      "learning_rate": 2.3999193603539234e-06,
+      "loss": 0.6026,
+      "step": 198
+    },
+    {
+      "epoch": 3.4912280701754383,
+      "grad_norm": 0.4753131480240871,
+      "learning_rate": 2.3504036788339763e-06,
+      "loss": 0.6397,
+      "step": 199
+    },
+    {
+      "epoch": 3.5087719298245617,
+      "grad_norm": 0.41785970596049077,
+      "learning_rate": 2.3012467102424373e-06,
+      "loss": 0.5908,
+      "step": 200
+    },
+    {
+      "epoch": 3.526315789473684,
+      "grad_norm": 0.5305842617772449,
+      "learning_rate": 2.2524551096459703e-06,
+      "loss": 0.6441,
+      "step": 201
+    },
+    {
+      "epoch": 3.543859649122807,
+      "grad_norm": 0.5712123142233522,
+      "learning_rate": 2.204035482646267e-06,
+      "loss": 0.5213,
+      "step": 202
+    },
+    {
+      "epoch": 3.56140350877193,
+      "grad_norm": 0.43115874181813724,
+      "learning_rate": 2.155994384485742e-06,
+      "loss": 0.7195,
+      "step": 203
+    },
+    {
+      "epoch": 3.5789473684210527,
+      "grad_norm": 0.5977556858646753,
+      "learning_rate": 2.1083383191600676e-06,
+      "loss": 0.5891,
+      "step": 204
+    },
+    {
+      "epoch": 3.5964912280701755,
+      "grad_norm": 0.5966648891745611,
+      "learning_rate": 2.061073738537635e-06,
+      "loss": 0.6022,
+      "step": 205
+    },
+    {
+      "epoch": 3.6140350877192984,
+      "grad_norm": 0.44035700208396406,
+      "learning_rate": 2.0142070414860704e-06,
+      "loss": 0.5315,
+      "step": 206
+    },
+    {
+      "epoch": 3.6315789473684212,
+      "grad_norm": 0.6020963303963573,
+      "learning_rate": 1.9677445730059348e-06,
+      "loss": 0.5479,
+      "step": 207
+    },
+    {
+      "epoch": 3.6491228070175437,
+      "grad_norm": 0.610172072779867,
+      "learning_rate": 1.9216926233717087e-06,
+      "loss": 0.5795,
+      "step": 208
+    },
+    {
+      "epoch": 3.6666666666666665,
+      "grad_norm": 0.5310472931422197,
+      "learning_rate": 1.8760574272802002e-06,
+      "loss": 0.6415,
+      "step": 209
+    },
+    {
+      "epoch": 3.6842105263157894,
+      "grad_norm": 0.5760708279783529,
+      "learning_rate": 1.8308451630064484e-06,
+      "loss": 0.571,
+      "step": 210
+    },
+    {
+      "epoch": 3.7017543859649122,
+      "grad_norm": 0.5554621456243964,
+      "learning_rate": 1.7860619515673034e-06,
+      "loss": 0.659,
+      "step": 211
+    },
+    {
+      "epoch": 3.719298245614035,
+      "grad_norm": 0.5057468988672211,
+      "learning_rate": 1.7417138558927244e-06,
+      "loss": 0.6274,
+      "step": 212
+    },
+    {
+      "epoch": 3.736842105263158,
+      "grad_norm": 0.4634242102552953,
+      "learning_rate": 1.6978068800049624e-06,
+      "loss": 0.686,
+      "step": 213
+    },
+    {
+      "epoch": 3.754385964912281,
+      "grad_norm": 0.4711471654335856,
+      "learning_rate": 1.6543469682057105e-06,
+      "loss": 0.6881,
+      "step": 214
+    },
+    {
+      "epoch": 3.7719298245614032,
+      "grad_norm": 0.44579346276413756,
+      "learning_rate": 1.611340004271339e-06,
+      "loss": 0.7255,
+      "step": 215
+    },
+    {
+      "epoch": 3.7894736842105265,
+      "grad_norm": 0.5506682698125349,
+      "learning_rate": 1.5687918106563326e-06,
+      "loss": 0.6243,
+      "step": 216
+    },
+    {
+      "epoch": 3.807017543859649,
+      "grad_norm": 0.4947665994574585,
+      "learning_rate": 1.5267081477050132e-06,
+      "loss": 0.6742,
+      "step": 217
+    },
+    {
+      "epoch": 3.824561403508772,
+      "grad_norm": 0.505923404111897,
+      "learning_rate": 1.4850947128716914e-06,
+      "loss": 0.6901,
+      "step": 218
+    },
+    {
+      "epoch": 3.8421052631578947,
+      "grad_norm": 0.5208839377384743,
+      "learning_rate": 1.4439571399493146e-06,
+      "loss": 0.6514,
+      "step": 219
+    },
+    {
+      "epoch": 3.8596491228070176,
+      "grad_norm": 0.4528805064363317,
+      "learning_rate": 1.4033009983067454e-06,
+      "loss": 0.5547,
+      "step": 220
+    },
+    {
+      "epoch": 3.8771929824561404,
+      "grad_norm": 0.45401366457289255,
+      "learning_rate": 1.3631317921347564e-06,
+      "loss": 0.6451,
+      "step": 221
+    },
+    {
+      "epoch": 3.8947368421052633,
+      "grad_norm": 0.5169718060442362,
+      "learning_rate": 1.3234549597008572e-06,
+      "loss": 0.636,
+      "step": 222
+    },
+    {
+      "epoch": 3.912280701754386,
+      "grad_norm": 0.4212973065101782,
+      "learning_rate": 1.2842758726130283e-06,
+      "loss": 0.5681,
+      "step": 223
+    },
+    {
+      "epoch": 3.9298245614035086,
+      "grad_norm": 0.5808214950684624,
+      "learning_rate": 1.2455998350925042e-06,
+      "loss": 0.5836,
+      "step": 224
+    },
+    {
+      "epoch": 3.9473684210526314,
+      "grad_norm": 0.8470975982923624,
+      "learning_rate": 1.2074320832556558e-06,
+      "loss": 0.6845,
+      "step": 225
+    },
+    {
+      "epoch": 3.9649122807017543,
+      "grad_norm": 0.651446990518644,
+      "learning_rate": 1.1697777844051105e-06,
+      "loss": 0.6585,
+      "step": 226
+    },
+    {
+      "epoch": 3.982456140350877,
+      "grad_norm": 0.5255363120908353,
+      "learning_rate": 1.132642036330181e-06,
+      "loss": 0.6802,
+      "step": 227
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.5068444505458819,
+      "learning_rate": 1.096029866616704e-06,
+      "loss": 0.5748,
+      "step": 228
+    },
+    {
+      "epoch": 4.017543859649122,
+      "grad_norm": 0.717072216096874,
+      "learning_rate": 1.0599462319663906e-06,
+      "loss": 0.6407,
+      "step": 229
+    },
+    {
+      "epoch": 4.035087719298246,
+      "grad_norm": 1.0385760459673252,
+      "learning_rate": 1.0243960175257605e-06,
+      "loss": 0.5482,
+      "step": 230
+    },
+    {
+      "epoch": 4.052631578947368,
+      "grad_norm": 0.7659544458424336,
+      "learning_rate": 9.893840362247809e-07,
+      "loss": 0.5831,
+      "step": 231
+    },
+    {
+      "epoch": 4.0701754385964914,
+      "grad_norm": 0.6460846484467759,
+      "learning_rate": 9.549150281252633e-07,
+      "loss": 0.5684,
+      "step": 232
+    },
+    {
+      "epoch": 4.087719298245614,
+      "grad_norm": 0.4387696389355417,
+      "learning_rate": 9.209936597791407e-07,
+      "loss": 0.6202,
+      "step": 233
+    },
+    {
+      "epoch": 4.105263157894737,
+      "grad_norm": 0.5104318901712891,
+      "learning_rate": 8.876245235966884e-07,
+      "loss": 0.6144,
+      "step": 234
+    },
+    {
+      "epoch": 4.12280701754386,
+      "grad_norm": 0.4263000982839917,
+      "learning_rate": 8.54812137224792e-07,
+      "loss": 0.5486,
+      "step": 235
+    },
+    {
+      "epoch": 4.140350877192983,
+      "grad_norm": 0.6032891578717126,
+      "learning_rate": 8.225609429353187e-07,
+      "loss": 0.6453,
+      "step": 236
+    },
+    {
+      "epoch": 4.157894736842105,
+      "grad_norm": 0.7811946598328325,
+      "learning_rate": 7.908753070237124e-07,
+      "loss": 0.507,
+      "step": 237
+    },
+    {
+      "epoch": 4.175438596491228,
+      "grad_norm": 0.7973182741337309,
+      "learning_rate": 7.597595192178702e-07,
+      "loss": 0.5967,
+      "step": 238
+    },
+    {
+      "epoch": 4.192982456140351,
+      "grad_norm": 0.6836397736571374,
+      "learning_rate": 7.292177920973726e-07,
+      "loss": 0.5821,
+      "step": 239
+    },
+    {
+      "epoch": 4.2105263157894735,
+      "grad_norm": 0.6951473287136067,
+      "learning_rate": 6.992542605231739e-07,
+      "loss": 0.5048,
+      "step": 240
+    },
+    {
+      "epoch": 4.228070175438597,
+      "grad_norm": 0.47509403704050995,
+      "learning_rate": 6.698729810778065e-07,
+      "loss": 0.578,
+      "step": 241
+    },
+    {
+      "epoch": 4.245614035087719,
+      "grad_norm": 0.5672633435740269,
+      "learning_rate": 6.410779315161885e-07,
+      "loss": 0.6184,
+      "step": 242
+    },
+    {
+      "epoch": 4.2631578947368425,
+      "grad_norm": 0.5254404527789579,
+      "learning_rate": 6.128730102270897e-07,
+      "loss": 0.5265,
+      "step": 243
+    },
+    {
+      "epoch": 4.280701754385965,
+      "grad_norm": 0.5743109910671067,
+      "learning_rate": 5.852620357053651e-07,
+      "loss": 0.6397,
+      "step": 244
+    },
+    {
+      "epoch": 4.298245614035087,
+      "grad_norm": 0.4557512708460931,
+      "learning_rate": 5.582487460349806e-07,
+      "loss": 0.6358,
+      "step": 245
+    },
+    {
+      "epoch": 4.315789473684211,
+      "grad_norm": 0.5806336006050219,
+      "learning_rate": 5.318367983829393e-07,
+      "loss": 0.6875,
+      "step": 246
+    },
+    {
+      "epoch": 4.333333333333333,
+      "grad_norm": 0.4937094896850357,
+      "learning_rate": 5.06029768504166e-07,
+      "loss": 0.637,
+      "step": 247
+    },
+    {
+      "epoch": 4.350877192982456,
+      "grad_norm": 0.4652629762222679,
+      "learning_rate": 4.808311502573976e-07,
+      "loss": 0.5903,
+      "step": 248
+    },
+    {
+      "epoch": 4.368421052631579,
+      "grad_norm": 0.46066078208125744,
+      "learning_rate": 4.562443551321788e-07,
+      "loss": 0.4891,
+      "step": 249
+    },
+    {
+      "epoch": 4.385964912280702,
+      "grad_norm": 0.46881139643834285,
+      "learning_rate": 4.322727117869951e-07,
+      "loss": 0.6732,
+      "step": 250
+    },
+    {
+      "epoch": 4.4035087719298245,
+      "grad_norm": 0.5127117943239673,
+      "learning_rate": 4.089194655986306e-07,
+      "loss": 0.6552,
+      "step": 251
+    },
+    {
+      "epoch": 4.421052631578947,
+      "grad_norm": 0.4162014214367883,
+      "learning_rate": 3.8618777822278854e-07,
+      "loss": 0.5416,
+      "step": 252
+    },
+    {
+      "epoch": 4.43859649122807,
+      "grad_norm": 0.48289206390401124,
+      "learning_rate": 3.6408072716606346e-07,
+      "loss": 0.5928,
+      "step": 253
+    },
+    {
+      "epoch": 4.456140350877193,
+      "grad_norm": 0.4850298646454562,
+      "learning_rate": 3.426013053692878e-07,
+      "loss": 0.6311,
+      "step": 254
+    },
+    {
+      "epoch": 4.473684210526316,
+      "grad_norm": 0.531210336532165,
+      "learning_rate": 3.2175242080234314e-07,
+      "loss": 0.7105,
+      "step": 255
+    },
+    {
+      "epoch": 4.491228070175438,
+      "grad_norm": 0.48333574225203335,
+      "learning_rate": 3.015368960704584e-07,
+      "loss": 0.6517,
+      "step": 256
+    },
+    {
+      "epoch": 4.508771929824562,
+      "grad_norm": 0.43103428929632426,
+      "learning_rate": 2.819574680320825e-07,
+      "loss": 0.6116,
+      "step": 257
+    },
+    {
+      "epoch": 4.526315789473684,
+      "grad_norm": 0.4778051001439487,
+      "learning_rate": 2.63016787428354e-07,
+      "loss": 0.5503,
+      "step": 258
+    },
+    {
+      "epoch": 4.543859649122807,
+      "grad_norm": 0.41974974922161384,
+      "learning_rate": 2.447174185242324e-07,
+      "loss": 0.6852,
+      "step": 259
+    },
+    {
+      "epoch": 4.56140350877193,
+      "grad_norm": 0.44246702992567427,
+      "learning_rate": 2.2706183876134047e-07,
+      "loss": 0.4739,
+      "step": 260
+    },
+    {
+      "epoch": 4.578947368421053,
+      "grad_norm": 0.4323307801121955,
+      "learning_rate": 2.1005243842255552e-07,
+      "loss": 0.538,
+      "step": 261
+    },
+    {
+      "epoch": 4.5964912280701755,
+      "grad_norm": 0.4347543756895765,
+      "learning_rate": 1.9369152030840553e-07,
+      "loss": 0.535,
+      "step": 262
+    },
+    {
+      "epoch": 4.614035087719298,
+      "grad_norm": 0.4420008972018591,
+      "learning_rate": 1.779812994253055e-07,
+      "loss": 0.4799,
+      "step": 263
+    },
+    {
+      "epoch": 4.631578947368421,
+      "grad_norm": 0.5179027214282402,
+      "learning_rate": 1.6292390268568103e-07,
+      "loss": 0.5518,
+      "step": 264
+    },
+    {
+      "epoch": 4.649122807017544,
+      "grad_norm": 0.44131211829977146,
+      "learning_rate": 1.4852136862001766e-07,
+      "loss": 0.6209,
+      "step": 265
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.4912194254345963,
+      "learning_rate": 1.3477564710088097e-07,
+      "loss": 0.5648,
+      "step": 266
+    },
+    {
+      "epoch": 4.684210526315789,
+      "grad_norm": 0.44383149142901357,
+      "learning_rate": 1.2168859907892904e-07,
+      "loss": 0.5855,
+      "step": 267
+    },
+    {
+      "epoch": 4.701754385964913,
+      "grad_norm": 0.5231452863668585,
+      "learning_rate": 1.0926199633097156e-07,
+      "loss": 0.4864,
+      "step": 268
+    },
+    {
+      "epoch": 4.719298245614035,
+      "grad_norm": 0.46974490276181063,
+      "learning_rate": 9.749752122010347e-08,
+      "loss": 0.6212,
+      "step": 269
+    },
+    {
+      "epoch": 4.7368421052631575,
+      "grad_norm": 0.44234244742840645,
+      "learning_rate": 8.639676646793382e-08,
+      "loss": 0.5871,
+      "step": 270
+    },
+    {
+      "epoch": 4.754385964912281,
+      "grad_norm": 0.429151879850756,
+      "learning_rate": 7.59612349389599e-08,
+      "loss": 0.6006,
+      "step": 271
+    },
+    {
+      "epoch": 4.771929824561403,
+      "grad_norm": 0.41027417971390234,
+      "learning_rate": 6.61923394371039e-08,
+      "loss": 0.7135,
+      "step": 272
+    },
+    {
+      "epoch": 4.7894736842105265,
+      "grad_norm": 0.44729992547643194,
+      "learning_rate": 5.709140251444201e-08,
+      "loss": 0.6072,
+      "step": 273
+    },
+    {
+      "epoch": 4.807017543859649,
+      "grad_norm": 0.5158805461859292,
+      "learning_rate": 4.865965629214819e-08,
+      "loss": 0.5775,
+      "step": 274
+    },
+    {
+      "epoch": 4.824561403508772,
+      "grad_norm": 0.4524237610322769,
+      "learning_rate": 4.0898242293691546e-08,
+      "loss": 0.5438,
+      "step": 275
+    },
+    {
+      "epoch": 4.842105263157895,
+      "grad_norm": 0.43714065174890604,
+      "learning_rate": 3.3808211290284886e-08,
+      "loss": 0.5923,
+      "step": 276
+    },
+    {
+      "epoch": 4.859649122807017,
+      "grad_norm": 0.542581817451324,
+      "learning_rate": 2.7390523158633552e-08,
+      "loss": 0.6038,
+      "step": 277
+    },
+    {
+      "epoch": 4.87719298245614,
+      "grad_norm": 0.3855811663651015,
+      "learning_rate": 2.1646046750978255e-08,
+      "loss": 0.6165,
+      "step": 278
+    },
+    {
+      "epoch": 4.894736842105263,
+      "grad_norm": 0.40204255230526825,
+      "learning_rate": 1.657555977746972e-08,
+      "loss": 0.5782,
+      "step": 279
+    },
+    {
+      "epoch": 4.912280701754386,
+      "grad_norm": 0.5070185485143726,
+      "learning_rate": 1.2179748700879013e-08,
+      "loss": 0.6097,
+      "step": 280
+    },
+    {
+      "epoch": 4.9298245614035086,
+      "grad_norm": 0.4546563631847268,
+      "learning_rate": 8.459208643659122e-09,
+      "loss": 0.6043,
+      "step": 281
+    },
+    {
+      "epoch": 4.947368421052632,
+      "grad_norm": 0.6307366966036172,
+      "learning_rate": 5.414443307377171e-09,
+      "loss": 0.7013,
+      "step": 282
+    },
+    {
+      "epoch": 4.964912280701754,
+      "grad_norm": 0.383640403567264,
+      "learning_rate": 3.0458649045211897e-09,
+      "loss": 0.5529,
+      "step": 283
+    },
+    {
+      "epoch": 4.982456140350877,
+      "grad_norm": 0.41920509047166143,
+      "learning_rate": 1.3537941026914302e-09,
+      "loss": 0.6274,
+      "step": 284
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.4160697744227897,
+      "learning_rate": 3.384599811889766e-10,
+      "loss": 0.6334,
+      "step": 285
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 285,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 65402569555968.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38a3e40f87d7c46db409da77aa1ba3c1c3d6ed6e16da374d1abef557aecc8171
+size 8081