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": false,
+  "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:c4c662aaa6472d1e3b1f190f32f5cddf1cb32d7f5797b055bf9354cda499f7a9
+size 4997184968

model-00002-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fe03faf1d40744be9e92b98865995d525ec57a1fd931f45b9f96928cf533d7f
+size 4997741608

model-00003-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65efdb5949f62b2e2f9aa19efd6cfa47e29a69ed6e4a7c18ff899de4292cb482
+size 4997742208

model-00004-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b98cb14f94c8c68067f5ab6b9c0c28224d5b5f52273fbd98bc1cdc41c1359c7b
+size 4997743184

model-00005-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e8fb82697baa5ebd7395a954925e532a1beba0c89c0e477c3e0fdc8e2ec5b79
+size 4997743184

model-00006-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:060db6b3554a60639322a4a25ca06726823226125beac89488d51cacc89fb229
+size 4997743184

model-00007-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c0317e661ba972e7a352fe7e31033f0da7c92285e5989903465f591fa9e4686
+size 4997743184

model-00008-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d6e6475e961edb599e1025dd38a9436b2f284805180f4a7101ed821d0bcb550
+size 4997743184

model-00009-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed4999da4c81f1fd035669c9718f0a83fb790ca70f4414633314eef9901d3a9c
+size 4997743184

model-00010-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6381cbf68446b5482b8d26c7a2f010ad411506c88c101f144ff038733a3aa0bb
+size 4997743184

model-00011-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48d7abe39762a7baf4043cf179c32ebecd3be3a099596c66f519155311afefdc
+size 4997743184

model-00012-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b4a755e18e6deca2396e3c91b4f04fb95e5186e86b9a61dfe570a9bd5780d40
+size 4997743184

model-00013-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2af357c7d5203f1a563b06945d1325bcbd34d70f6a965c338824a41e53ffee9
+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": 2.1658732413844044,
+      "learning_rate": 0.0,
+      "loss": 0.8483,
+      "step": 1
+    },
+    {
+      "epoch": 0.03508771929824561,
+      "grad_norm": 2.092006940468524,
+      "learning_rate": 6.666666666666667e-07,
+      "loss": 0.8765,
+      "step": 2
+    },
+    {
+      "epoch": 0.05263157894736842,
+      "grad_norm": 2.275864451889245,
+      "learning_rate": 1.3333333333333334e-06,
+      "loss": 1.0208,
+      "step": 3
+    },
+    {
+      "epoch": 0.07017543859649122,
+      "grad_norm": 2.3835659363654167,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.062,
+      "step": 4
+    },
+    {
+      "epoch": 0.08771929824561403,
+      "grad_norm": 1.8668614513642336,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 0.8564,
+      "step": 5
+    },
+    {
+      "epoch": 0.10526315789473684,
+      "grad_norm": 2.042827888630328,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 1.0079,
+      "step": 6
+    },
+    {
+      "epoch": 0.12280701754385964,
+      "grad_norm": 1.7108395348483465,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.9971,
+      "step": 7
+    },
+    {
+      "epoch": 0.14035087719298245,
+      "grad_norm": 1.7285480717242914,
+      "learning_rate": 4.666666666666667e-06,
+      "loss": 0.9464,
+      "step": 8
+    },
+    {
+      "epoch": 0.15789473684210525,
+      "grad_norm": 1.4329487505630585,
+      "learning_rate": 5.333333333333334e-06,
+      "loss": 0.9607,
+      "step": 9
+    },
+    {
+      "epoch": 0.17543859649122806,
+      "grad_norm": 1.2795124499162036,
+      "learning_rate": 6e-06,
+      "loss": 0.9863,
+      "step": 10
+    },
+    {
+      "epoch": 0.19298245614035087,
+      "grad_norm": 1.2605076330876026,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 1.0293,
+      "step": 11
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 1.3904361544294626,
+      "learning_rate": 7.333333333333333e-06,
+      "loss": 0.8981,
+      "step": 12
+    },
+    {
+      "epoch": 0.22807017543859648,
+      "grad_norm": 1.4934089139203042,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.9339,
+      "step": 13
+    },
+    {
+      "epoch": 0.24561403508771928,
+      "grad_norm": 1.426034050887767,
+      "learning_rate": 8.666666666666668e-06,
+      "loss": 0.9114,
+      "step": 14
+    },
+    {
+      "epoch": 0.2631578947368421,
+      "grad_norm": 1.2195840821985733,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.9704,
+      "step": 15
+    },
+    {
+      "epoch": 0.2807017543859649,
+      "grad_norm": 1.3112838302645187,
+      "learning_rate": 1e-05,
+      "loss": 0.921,
+      "step": 16
+    },
+    {
+      "epoch": 0.2982456140350877,
+      "grad_norm": 1.4147363764085927,
+      "learning_rate": 9.999661540018812e-06,
+      "loss": 1.0178,
+      "step": 17
+    },
+    {
+      "epoch": 0.3157894736842105,
+      "grad_norm": 3.0364757241727425,
+      "learning_rate": 9.99864620589731e-06,
+      "loss": 0.9016,
+      "step": 18
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 1.1966597334373839,
+      "learning_rate": 9.99695413509548e-06,
+      "loss": 0.9241,
+      "step": 19
+    },
+    {
+      "epoch": 0.3508771929824561,
+      "grad_norm": 1.0673860693223811,
+      "learning_rate": 9.994585556692624e-06,
+      "loss": 0.9605,
+      "step": 20
+    },
+    {
+      "epoch": 0.3684210526315789,
+      "grad_norm": 0.9159275821464556,
+      "learning_rate": 9.991540791356342e-06,
+      "loss": 0.8978,
+      "step": 21
+    },
+    {
+      "epoch": 0.38596491228070173,
+      "grad_norm": 0.8662523471810823,
+      "learning_rate": 9.987820251299121e-06,
+      "loss": 0.7651,
+      "step": 22
+    },
+    {
+      "epoch": 0.40350877192982454,
+      "grad_norm": 0.9013821931274979,
+      "learning_rate": 9.98342444022253e-06,
+      "loss": 0.8986,
+      "step": 23
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.7877029407981232,
+      "learning_rate": 9.978353953249023e-06,
+      "loss": 0.7958,
+      "step": 24
+    },
+    {
+      "epoch": 0.43859649122807015,
+      "grad_norm": 0.8224354382712344,
+      "learning_rate": 9.972609476841368e-06,
+      "loss": 0.9247,
+      "step": 25
+    },
+    {
+      "epoch": 0.45614035087719296,
+      "grad_norm": 0.908198561067024,
+      "learning_rate": 9.966191788709716e-06,
+      "loss": 0.8886,
+      "step": 26
+    },
+    {
+      "epoch": 0.47368421052631576,
+      "grad_norm": 0.8471005617195208,
+      "learning_rate": 9.959101757706308e-06,
+      "loss": 0.8091,
+      "step": 27
+    },
+    {
+      "epoch": 0.49122807017543857,
+      "grad_norm": 0.7441058742609257,
+      "learning_rate": 9.951340343707852e-06,
+      "loss": 0.8774,
+      "step": 28
+    },
+    {
+      "epoch": 0.5087719298245614,
+      "grad_norm": 0.6415749968368839,
+      "learning_rate": 9.942908597485558e-06,
+      "loss": 0.8987,
+      "step": 29
+    },
+    {
+      "epoch": 0.5263157894736842,
+      "grad_norm": 0.5945595943682435,
+      "learning_rate": 9.933807660562898e-06,
+      "loss": 0.7845,
+      "step": 30
+    },
+    {
+      "epoch": 0.543859649122807,
+      "grad_norm": 0.5348299561393488,
+      "learning_rate": 9.924038765061042e-06,
+      "loss": 0.8401,
+      "step": 31
+    },
+    {
+      "epoch": 0.5614035087719298,
+      "grad_norm": 0.5449412249752669,
+      "learning_rate": 9.913603233532067e-06,
+      "loss": 0.7708,
+      "step": 32
+    },
+    {
+      "epoch": 0.5789473684210527,
+      "grad_norm": 0.7606361821473685,
+      "learning_rate": 9.902502478779897e-06,
+      "loss": 0.8696,
+      "step": 33
+    },
+    {
+      "epoch": 0.5964912280701754,
+      "grad_norm": 0.6987567739630257,
+      "learning_rate": 9.890738003669029e-06,
+      "loss": 0.9685,
+      "step": 34
+    },
+    {
+      "epoch": 0.6140350877192983,
+      "grad_norm": 0.6577860614813181,
+      "learning_rate": 9.878311400921072e-06,
+      "loss": 0.8129,
+      "step": 35
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.613409929779683,
+      "learning_rate": 9.86522435289912e-06,
+      "loss": 0.8279,
+      "step": 36
+    },
+    {
+      "epoch": 0.6491228070175439,
+      "grad_norm": 0.8310936385078747,
+      "learning_rate": 9.851478631379982e-06,
+      "loss": 0.8629,
+      "step": 37
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.5401034869147716,
+      "learning_rate": 9.83707609731432e-06,
+      "loss": 0.8763,
+      "step": 38
+    },
+    {
+      "epoch": 0.6842105263157895,
+      "grad_norm": 0.5645732604776249,
+      "learning_rate": 9.822018700574696e-06,
+      "loss": 0.8987,
+      "step": 39
+    },
+    {
+      "epoch": 0.7017543859649122,
+      "grad_norm": 0.48907017080577125,
+      "learning_rate": 9.806308479691595e-06,
+      "loss": 0.8645,
+      "step": 40
+    },
+    {
+      "epoch": 0.7192982456140351,
+      "grad_norm": 0.5336160010922303,
+      "learning_rate": 9.789947561577445e-06,
+      "loss": 0.8694,
+      "step": 41
+    },
+    {
+      "epoch": 0.7368421052631579,
+      "grad_norm": 0.5616846107481482,
+      "learning_rate": 9.77293816123866e-06,
+      "loss": 0.8044,
+      "step": 42
+    },
+    {
+      "epoch": 0.7543859649122807,
+      "grad_norm": 0.5285876380934174,
+      "learning_rate": 9.755282581475769e-06,
+      "loss": 0.8373,
+      "step": 43
+    },
+    {
+      "epoch": 0.7719298245614035,
+      "grad_norm": 0.5339805004772071,
+      "learning_rate": 9.736983212571646e-06,
+      "loss": 0.7534,
+      "step": 44
+    },
+    {
+      "epoch": 0.7894736842105263,
+      "grad_norm": 0.4965160387028941,
+      "learning_rate": 9.718042531967918e-06,
+      "loss": 0.87,
+      "step": 45
+    },
+    {
+      "epoch": 0.8070175438596491,
+      "grad_norm": 0.5512110498718402,
+      "learning_rate": 9.698463103929542e-06,
+      "loss": 0.8465,
+      "step": 46
+    },
+    {
+      "epoch": 0.8245614035087719,
+      "grad_norm": 0.5556768293250771,
+      "learning_rate": 9.678247579197658e-06,
+      "loss": 0.8509,
+      "step": 47
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.43424614496891145,
+      "learning_rate": 9.657398694630713e-06,
+      "loss": 0.8351,
+      "step": 48
+    },
+    {
+      "epoch": 0.8596491228070176,
+      "grad_norm": 0.517819637053559,
+      "learning_rate": 9.635919272833938e-06,
+      "loss": 0.9268,
+      "step": 49
+    },
+    {
+      "epoch": 0.8771929824561403,
+      "grad_norm": 0.4573125121358173,
+      "learning_rate": 9.613812221777212e-06,
+      "loss": 0.6669,
+      "step": 50
+    },
+    {
+      "epoch": 0.8947368421052632,
+      "grad_norm": 0.4918423165455613,
+      "learning_rate": 9.591080534401371e-06,
+      "loss": 0.896,
+      "step": 51
+    },
+    {
+      "epoch": 0.9122807017543859,
+      "grad_norm": 0.4873519914477964,
+      "learning_rate": 9.567727288213005e-06,
+      "loss": 0.8642,
+      "step": 52
+    },
+    {
+      "epoch": 0.9298245614035088,
+      "grad_norm": 0.4959554344589731,
+      "learning_rate": 9.543755644867823e-06,
+      "loss": 0.8161,
+      "step": 53
+    },
+    {
+      "epoch": 0.9473684210526315,
+      "grad_norm": 0.5053982708053872,
+      "learning_rate": 9.519168849742603e-06,
+      "loss": 0.7712,
+      "step": 54
+    },
+    {
+      "epoch": 0.9649122807017544,
+      "grad_norm": 0.4999876957766244,
+      "learning_rate": 9.493970231495836e-06,
+      "loss": 0.8104,
+      "step": 55
+    },
+    {
+      "epoch": 0.9824561403508771,
+      "grad_norm": 0.4802283149352191,
+      "learning_rate": 9.468163201617063e-06,
+      "loss": 0.8219,
+      "step": 56
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5730089270450355,
+      "learning_rate": 9.441751253965022e-06,
+      "loss": 0.733,
+      "step": 57
+    },
+    {
+      "epoch": 1.0175438596491229,
+      "grad_norm": 0.5933645174516611,
+      "learning_rate": 9.414737964294636e-06,
+      "loss": 0.8412,
+      "step": 58
+    },
+    {
+      "epoch": 1.0350877192982457,
+      "grad_norm": 0.5372883465461307,
+      "learning_rate": 9.38712698977291e-06,
+      "loss": 0.7292,
+      "step": 59
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 0.5446643188183766,
+      "learning_rate": 9.358922068483813e-06,
+      "loss": 0.8166,
+      "step": 60
+    },
+    {
+      "epoch": 1.0701754385964912,
+      "grad_norm": 0.5166103907170196,
+      "learning_rate": 9.330127018922195e-06,
+      "loss": 0.6909,
+      "step": 61
+    },
+    {
+      "epoch": 1.087719298245614,
+      "grad_norm": 0.5072015393223445,
+      "learning_rate": 9.30074573947683e-06,
+      "loss": 0.7737,
+      "step": 62
+    },
+    {
+      "epoch": 1.1052631578947367,
+      "grad_norm": 0.6007962503647915,
+      "learning_rate": 9.27078220790263e-06,
+      "loss": 0.8027,
+      "step": 63
+    },
+    {
+      "epoch": 1.1228070175438596,
+      "grad_norm": 0.6163080503533958,
+      "learning_rate": 9.24024048078213e-06,
+      "loss": 0.7946,
+      "step": 64
+    },
+    {
+      "epoch": 1.1403508771929824,
+      "grad_norm": 0.45627196367257494,
+      "learning_rate": 9.209124692976287e-06,
+      "loss": 0.7166,
+      "step": 65
+    },
+    {
+      "epoch": 1.1578947368421053,
+      "grad_norm": 0.5254958255330286,
+      "learning_rate": 9.177439057064684e-06,
+      "loss": 0.7189,
+      "step": 66
+    },
+    {
+      "epoch": 1.1754385964912282,
+      "grad_norm": 0.5327511599889234,
+      "learning_rate": 9.145187862775208e-06,
+      "loss": 0.716,
+      "step": 67
+    },
+    {
+      "epoch": 1.1929824561403508,
+      "grad_norm": 0.499495987047289,
+      "learning_rate": 9.112375476403313e-06,
+      "loss": 0.7628,
+      "step": 68
+    },
+    {
+      "epoch": 1.2105263157894737,
+      "grad_norm": 0.5345682406386708,
+      "learning_rate": 9.079006340220862e-06,
+      "loss": 0.8137,
+      "step": 69
+    },
+    {
+      "epoch": 1.2280701754385965,
+      "grad_norm": 0.5012428722506891,
+      "learning_rate": 9.045084971874738e-06,
+      "loss": 0.7596,
+      "step": 70
+    },
+    {
+      "epoch": 1.2456140350877192,
+      "grad_norm": 0.4718075088336184,
+      "learning_rate": 9.01061596377522e-06,
+      "loss": 0.8108,
+      "step": 71
+    },
+    {
+      "epoch": 1.263157894736842,
+      "grad_norm": 0.5691867690955263,
+      "learning_rate": 8.97560398247424e-06,
+      "loss": 0.7761,
+      "step": 72
+    },
+    {
+      "epoch": 1.280701754385965,
+      "grad_norm": 0.5056803795224231,
+      "learning_rate": 8.94005376803361e-06,
+      "loss": 0.7753,
+      "step": 73
+    },
+    {
+      "epoch": 1.2982456140350878,
+      "grad_norm": 0.4584931407709478,
+      "learning_rate": 8.903970133383297e-06,
+      "loss": 0.6956,
+      "step": 74
+    },
+    {
+      "epoch": 1.3157894736842106,
+      "grad_norm": 0.48200571497679806,
+      "learning_rate": 8.867357963669821e-06,
+      "loss": 0.8248,
+      "step": 75
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.44308375696335517,
+      "learning_rate": 8.83022221559489e-06,
+      "loss": 0.7184,
+      "step": 76
+    },
+    {
+      "epoch": 1.3508771929824561,
+      "grad_norm": 0.5092871749212193,
+      "learning_rate": 8.792567916744346e-06,
+      "loss": 0.6901,
+      "step": 77
+    },
+    {
+      "epoch": 1.368421052631579,
+      "grad_norm": 0.5922504470496844,
+      "learning_rate": 8.754400164907496e-06,
+      "loss": 0.8101,
+      "step": 78
+    },
+    {
+      "epoch": 1.3859649122807016,
+      "grad_norm": 0.5502021539392707,
+      "learning_rate": 8.715724127386971e-06,
+      "loss": 0.7933,
+      "step": 79
+    },
+    {
+      "epoch": 1.4035087719298245,
+      "grad_norm": 0.4739099022562294,
+      "learning_rate": 8.676545040299145e-06,
+      "loss": 0.7403,
+      "step": 80
+    },
+    {
+      "epoch": 1.4210526315789473,
+      "grad_norm": 0.5675822777531978,
+      "learning_rate": 8.636868207865244e-06,
+      "loss": 0.7654,
+      "step": 81
+    },
+    {
+      "epoch": 1.4385964912280702,
+      "grad_norm": 0.5343699447843498,
+      "learning_rate": 8.596699001693257e-06,
+      "loss": 0.7506,
+      "step": 82
+    },
+    {
+      "epoch": 1.456140350877193,
+      "grad_norm": 0.5135553837334624,
+      "learning_rate": 8.556042860050686e-06,
+      "loss": 0.6845,
+      "step": 83
+    },
+    {
+      "epoch": 1.4736842105263157,
+      "grad_norm": 0.5104930262488498,
+      "learning_rate": 8.51490528712831e-06,
+      "loss": 0.6129,
+      "step": 84
+    },
+    {
+      "epoch": 1.4912280701754386,
+      "grad_norm": 0.48637450571652296,
+      "learning_rate": 8.473291852294986e-06,
+      "loss": 0.7671,
+      "step": 85
+    },
+    {
+      "epoch": 1.5087719298245614,
+      "grad_norm": 0.51822441523707,
+      "learning_rate": 8.43120818934367e-06,
+      "loss": 0.7525,
+      "step": 86
+    },
+    {
+      "epoch": 1.526315789473684,
+      "grad_norm": 0.5245677712135876,
+      "learning_rate": 8.388659995728662e-06,
+      "loss": 0.8875,
+      "step": 87
+    },
+    {
+      "epoch": 1.543859649122807,
+      "grad_norm": 0.43219245811714907,
+      "learning_rate": 8.345653031794292e-06,
+      "loss": 0.7701,
+      "step": 88
+    },
+    {
+      "epoch": 1.5614035087719298,
+      "grad_norm": 0.43939935575665046,
+      "learning_rate": 8.302193119995038e-06,
+      "loss": 0.8084,
+      "step": 89
+    },
+    {
+      "epoch": 1.5789473684210527,
+      "grad_norm": 0.4804381141499697,
+      "learning_rate": 8.258286144107277e-06,
+      "loss": 0.7098,
+      "step": 90
+    },
+    {
+      "epoch": 1.5964912280701755,
+      "grad_norm": 0.520892331675854,
+      "learning_rate": 8.213938048432697e-06,
+      "loss": 0.747,
+      "step": 91
+    },
+    {
+      "epoch": 1.6140350877192984,
+      "grad_norm": 0.4430987555281133,
+      "learning_rate": 8.16915483699355e-06,
+      "loss": 0.758,
+      "step": 92
+    },
+    {
+      "epoch": 1.631578947368421,
+      "grad_norm": 0.3984274654952433,
+      "learning_rate": 8.123942572719801e-06,
+      "loss": 0.6948,
+      "step": 93
+    },
+    {
+      "epoch": 1.6491228070175439,
+      "grad_norm": 0.40432424672862693,
+      "learning_rate": 8.078307376628292e-06,
+      "loss": 0.7014,
+      "step": 94
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.5274290119353522,
+      "learning_rate": 8.032255426994069e-06,
+      "loss": 0.799,
+      "step": 95
+    },
+    {
+      "epoch": 1.6842105263157894,
+      "grad_norm": 0.4521630882545329,
+      "learning_rate": 7.985792958513932e-06,
+      "loss": 0.7079,
+      "step": 96
+    },
+    {
+      "epoch": 1.7017543859649122,
+      "grad_norm": 0.4660903384203027,
+      "learning_rate": 7.938926261462366e-06,
+      "loss": 0.8728,
+      "step": 97
+    },
+    {
+      "epoch": 1.719298245614035,
+      "grad_norm": 0.45027355953908643,
+      "learning_rate": 7.891661680839932e-06,
+      "loss": 0.7493,
+      "step": 98
+    },
+    {
+      "epoch": 1.736842105263158,
+      "grad_norm": 0.4755932897762567,
+      "learning_rate": 7.84400561551426e-06,
+      "loss": 0.7842,
+      "step": 99
+    },
+    {
+      "epoch": 1.7543859649122808,
+      "grad_norm": 0.4568461130515591,
+      "learning_rate": 7.795964517353734e-06,
+      "loss": 0.7278,
+      "step": 100
+    },
+    {
+      "epoch": 1.7719298245614035,
+      "grad_norm": 0.5334470147252376,
+      "learning_rate": 7.747544890354031e-06,
+      "loss": 0.7968,
+      "step": 101
+    },
+    {
+      "epoch": 1.7894736842105263,
+      "grad_norm": 0.4467766719927512,
+      "learning_rate": 7.698753289757565e-06,
+      "loss": 0.7962,
+      "step": 102
+    },
+    {
+      "epoch": 1.807017543859649,
+      "grad_norm": 0.5469179857216717,
+      "learning_rate": 7.649596321166024e-06,
+      "loss": 0.8132,
+      "step": 103
+    },
+    {
+      "epoch": 1.8245614035087718,
+      "grad_norm": 0.4569639347578776,
+      "learning_rate": 7.600080639646077e-06,
+      "loss": 0.6674,
+      "step": 104
+    },
+    {
+      "epoch": 1.8421052631578947,
+      "grad_norm": 0.46133635915522847,
+      "learning_rate": 7.550212948828377e-06,
+      "loss": 0.6431,
+      "step": 105
+    },
+    {
+      "epoch": 1.8596491228070176,
+      "grad_norm": 0.6123883071807754,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 0.8316,
+      "step": 106
+    },
+    {
+      "epoch": 1.8771929824561404,
+      "grad_norm": 0.5321187481349282,
+      "learning_rate": 7.449448591190436e-06,
+      "loss": 0.7628,
+      "step": 107
+    },
+    {
+      "epoch": 1.8947368421052633,
+      "grad_norm": 0.4735343474078112,
+      "learning_rate": 7.398565566251232e-06,
+      "loss": 0.784,
+      "step": 108
+    },
+    {
+      "epoch": 1.912280701754386,
+      "grad_norm": 0.458871553795796,
+      "learning_rate": 7.347357813929455e-06,
+      "loss": 0.6992,
+      "step": 109
+    },
+    {
+      "epoch": 1.9298245614035088,
+      "grad_norm": 0.5166436399930521,
+      "learning_rate": 7.295832266935059e-06,
+      "loss": 0.8054,
+      "step": 110
+    },
+    {
+      "epoch": 1.9473684210526314,
+      "grad_norm": 0.4516516353858229,
+      "learning_rate": 7.243995901002312e-06,
+      "loss": 0.8028,
+      "step": 111
+    },
+    {
+      "epoch": 1.9649122807017543,
+      "grad_norm": 0.52235491154167,
+      "learning_rate": 7.191855733945388e-06,
+      "loss": 0.7757,
+      "step": 112
+    },
+    {
+      "epoch": 1.9824561403508771,
+      "grad_norm": 0.4882398007983572,
+      "learning_rate": 7.1394188247082715e-06,
+      "loss": 0.8476,
+      "step": 113
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6282799373205465,
+      "learning_rate": 7.08669227240909e-06,
+      "loss": 0.7525,
+      "step": 114
+    },
+    {
+      "epoch": 2.017543859649123,
+      "grad_norm": 0.7063022001030749,
+      "learning_rate": 7.033683215379002e-06,
+      "loss": 0.6293,
+      "step": 115
+    },
+    {
+      "epoch": 2.0350877192982457,
+      "grad_norm": 0.468983383039873,
+      "learning_rate": 6.980398830195785e-06,
+      "loss": 0.6701,
+      "step": 116
+    },
+    {
+      "epoch": 2.0526315789473686,
+      "grad_norm": 0.6188215145922576,
+      "learning_rate": 6.9268463307122425e-06,
+      "loss": 0.8455,
+      "step": 117
+    },
+    {
+      "epoch": 2.0701754385964914,
+      "grad_norm": 1.1251110646181128,
+      "learning_rate": 6.873032967079562e-06,
+      "loss": 0.7125,
+      "step": 118
+    },
+    {
+      "epoch": 2.087719298245614,
+      "grad_norm": 0.8216966597385386,
+      "learning_rate": 6.818966024765758e-06,
+      "loss": 0.6899,
+      "step": 119
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 0.5247344441518691,
+      "learning_rate": 6.7646528235693445e-06,
+      "loss": 0.6706,
+      "step": 120
+    },
+    {
+      "epoch": 2.1228070175438596,
+      "grad_norm": 0.6891125392002603,
+      "learning_rate": 6.710100716628345e-06,
+      "loss": 0.6626,
+      "step": 121
+    },
+    {
+      "epoch": 2.1403508771929824,
+      "grad_norm": 0.8358975338472591,
+      "learning_rate": 6.655317089424791e-06,
+      "loss": 0.6183,
+      "step": 122
+    },
+    {
+      "epoch": 2.1578947368421053,
+      "grad_norm": 0.5405885869120298,
+      "learning_rate": 6.600309358784858e-06,
+      "loss": 0.7665,
+      "step": 123
+    },
+    {
+      "epoch": 2.175438596491228,
+      "grad_norm": 0.5806840526288823,
+      "learning_rate": 6.545084971874738e-06,
+      "loss": 0.6672,
+      "step": 124
+    },
+    {
+      "epoch": 2.192982456140351,
+      "grad_norm": 0.9120373679846502,
+      "learning_rate": 6.48965140519241e-06,
+      "loss": 0.7703,
+      "step": 125
+    },
+    {
+      "epoch": 2.2105263157894735,
+      "grad_norm": 0.6099997404267848,
+      "learning_rate": 6.434016163555452e-06,
+      "loss": 0.6729,
+      "step": 126
+    },
+    {
+      "epoch": 2.2280701754385963,
+      "grad_norm": 0.5476577394097183,
+      "learning_rate": 6.378186779084996e-06,
+      "loss": 0.6227,
+      "step": 127
+    },
+    {
+      "epoch": 2.245614035087719,
+      "grad_norm": 0.5138251066313695,
+      "learning_rate": 6.322170810186013e-06,
+      "loss": 0.6467,
+      "step": 128
+    },
+    {
+      "epoch": 2.263157894736842,
+      "grad_norm": 0.5916517716545399,
+      "learning_rate": 6.26597584052401e-06,
+      "loss": 0.6922,
+      "step": 129
+    },
+    {
+      "epoch": 2.280701754385965,
+      "grad_norm": 0.6311673974692158,
+      "learning_rate": 6.209609477998339e-06,
+      "loss": 0.7033,
+      "step": 130
+    },
+    {
+      "epoch": 2.2982456140350878,
+      "grad_norm": 0.5111012056097551,
+      "learning_rate": 6.153079353712201e-06,
+      "loss": 0.6841,
+      "step": 131
+    },
+    {
+      "epoch": 2.3157894736842106,
+      "grad_norm": 0.47416385243899933,
+      "learning_rate": 6.0963931209395165e-06,
+      "loss": 0.6454,
+      "step": 132
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.7154325052135133,
+      "learning_rate": 6.039558454088796e-06,
+      "loss": 0.7397,
+      "step": 133
+    },
+    {
+      "epoch": 2.3508771929824563,
+      "grad_norm": 0.6168844874657449,
+      "learning_rate": 5.982583047664151e-06,
+      "loss": 0.6534,
+      "step": 134
+    },
+    {
+      "epoch": 2.3684210526315788,
+      "grad_norm": 0.5752768710415526,
+      "learning_rate": 5.925474615223573e-06,
+      "loss": 0.6714,
+      "step": 135
+    },
+    {
+      "epoch": 2.3859649122807016,
+      "grad_norm": 0.500790932370114,
+      "learning_rate": 5.8682408883346535e-06,
+      "loss": 0.6658,
+      "step": 136
+    },
+    {
+      "epoch": 2.4035087719298245,
+      "grad_norm": 0.5691603857043621,
+      "learning_rate": 5.810889615527839e-06,
+      "loss": 0.7758,
+      "step": 137
+    },
+    {
+      "epoch": 2.4210526315789473,
+      "grad_norm": 0.5152378969652822,
+      "learning_rate": 5.753428561247416e-06,
+      "loss": 0.7618,
+      "step": 138
+    },
+    {
+      "epoch": 2.43859649122807,
+      "grad_norm": 0.5480249747263273,
+      "learning_rate": 5.695865504800328e-06,
+      "loss": 0.7163,
+      "step": 139
+    },
+    {
+      "epoch": 2.456140350877193,
+      "grad_norm": 0.46592300857001084,
+      "learning_rate": 5.638208239302975e-06,
+      "loss": 0.6735,
+      "step": 140
+    },
+    {
+      "epoch": 2.473684210526316,
+      "grad_norm": 0.5585608105245704,
+      "learning_rate": 5.5804645706261515e-06,
+      "loss": 0.7168,
+      "step": 141
+    },
+    {
+      "epoch": 2.4912280701754383,
+      "grad_norm": 0.478293606299736,
+      "learning_rate": 5.522642316338268e-06,
+      "loss": 0.7201,
+      "step": 142
+    },
+    {
+      "epoch": 2.5087719298245617,
+      "grad_norm": 0.5182777283508911,
+      "learning_rate": 5.464749304646963e-06,
+      "loss": 0.5704,
+      "step": 143
+    },
+    {
+      "epoch": 2.526315789473684,
+      "grad_norm": 0.46197548463372434,
+      "learning_rate": 5.406793373339292e-06,
+      "loss": 0.6573,
+      "step": 144
+    },
+    {
+      "epoch": 2.543859649122807,
+      "grad_norm": 0.43760805892411686,
+      "learning_rate": 5.348782368720627e-06,
+      "loss": 0.6725,
+      "step": 145
+    },
+    {
+      "epoch": 2.56140350877193,
+      "grad_norm": 0.6480646434757844,
+      "learning_rate": 5.290724144552379e-06,
+      "loss": 0.7335,
+      "step": 146
+    },
+    {
+      "epoch": 2.5789473684210527,
+      "grad_norm": 0.4774170630432071,
+      "learning_rate": 5.232626560988735e-06,
+      "loss": 0.6719,
+      "step": 147
+    },
+    {
+      "epoch": 2.5964912280701755,
+      "grad_norm": 0.47776520475395945,
+      "learning_rate": 5.174497483512506e-06,
+      "loss": 0.6944,
+      "step": 148
+    },
+    {
+      "epoch": 2.6140350877192984,
+      "grad_norm": 0.5445922223824958,
+      "learning_rate": 5.116344781870282e-06,
+      "loss": 0.6642,
+      "step": 149
+    },
+    {
+      "epoch": 2.6315789473684212,
+      "grad_norm": 1.5575866638410665,
+      "learning_rate": 5.0581763290069865e-06,
+      "loss": 0.7562,
+      "step": 150
+    },
+    {
+      "epoch": 2.6491228070175437,
+      "grad_norm": 0.44398282748234136,
+      "learning_rate": 5e-06,
+      "loss": 0.7303,
+      "step": 151
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.516466445832163,
+      "learning_rate": 4.941823670993016e-06,
+      "loss": 0.729,
+      "step": 152
+    },
+    {
+      "epoch": 2.6842105263157894,
+      "grad_norm": 0.6489061203378778,
+      "learning_rate": 4.883655218129719e-06,
+      "loss": 0.7245,
+      "step": 153
+    },
+    {
+      "epoch": 2.7017543859649122,
+      "grad_norm": 0.5455156755078856,
+      "learning_rate": 4.825502516487497e-06,
+      "loss": 0.7215,
+      "step": 154
+    },
+    {
+      "epoch": 2.719298245614035,
+      "grad_norm": 0.5113865044304958,
+      "learning_rate": 4.767373439011267e-06,
+      "loss": 0.7339,
+      "step": 155
+    },
+    {
+      "epoch": 2.736842105263158,
+      "grad_norm": 0.4222662620066933,
+      "learning_rate": 4.7092758554476215e-06,
+      "loss": 0.6739,
+      "step": 156
+    },
+    {
+      "epoch": 2.754385964912281,
+      "grad_norm": 0.43319020198089503,
+      "learning_rate": 4.651217631279374e-06,
+      "loss": 0.6892,
+      "step": 157
+    },
+    {
+      "epoch": 2.7719298245614032,
+      "grad_norm": 0.49241784511467307,
+      "learning_rate": 4.59320662666071e-06,
+      "loss": 0.6023,
+      "step": 158
+    },
+    {
+      "epoch": 2.7894736842105265,
+      "grad_norm": 0.45036125478517564,
+      "learning_rate": 4.53525069535304e-06,
+      "loss": 0.6663,
+      "step": 159
+    },
+    {
+      "epoch": 2.807017543859649,
+      "grad_norm": 0.5284115898387266,
+      "learning_rate": 4.477357683661734e-06,
+      "loss": 0.5687,
+      "step": 160
+    },
+    {
+      "epoch": 2.824561403508772,
+      "grad_norm": 0.5123465047513207,
+      "learning_rate": 4.4195354293738484e-06,
+      "loss": 0.6727,
+      "step": 161
+    },
+    {
+      "epoch": 2.8421052631578947,
+      "grad_norm": 0.5692049007560822,
+      "learning_rate": 4.361791760697027e-06,
+      "loss": 0.7027,
+      "step": 162
+    },
+    {
+      "epoch": 2.8596491228070176,
+      "grad_norm": 0.5267840248423996,
+      "learning_rate": 4.304134495199675e-06,
+      "loss": 0.7125,
+      "step": 163
+    },
+    {
+      "epoch": 2.8771929824561404,
+      "grad_norm": 0.4476798167406488,
+      "learning_rate": 4.246571438752585e-06,
+      "loss": 0.6858,
+      "step": 164
+    },
+    {
+      "epoch": 2.8947368421052633,
+      "grad_norm": 0.4877239792429488,
+      "learning_rate": 4.189110384472164e-06,
+      "loss": 0.7493,
+      "step": 165
+    },
+    {
+      "epoch": 2.912280701754386,
+      "grad_norm": 0.434099272030412,
+      "learning_rate": 4.131759111665349e-06,
+      "loss": 0.6845,
+      "step": 166
+    },
+    {
+      "epoch": 2.9298245614035086,
+      "grad_norm": 0.5804563462704657,
+      "learning_rate": 4.074525384776428e-06,
+      "loss": 0.6247,
+      "step": 167
+    },
+    {
+      "epoch": 2.9473684210526314,
+      "grad_norm": 0.4594980725557255,
+      "learning_rate": 4.017416952335849e-06,
+      "loss": 0.6679,
+      "step": 168
+    },
+    {
+      "epoch": 2.9649122807017543,
+      "grad_norm": 0.46871324608323867,
+      "learning_rate": 3.960441545911205e-06,
+      "loss": 0.6963,
+      "step": 169
+    },
+    {
+      "epoch": 2.982456140350877,
+      "grad_norm": 0.5908231692826603,
+      "learning_rate": 3.903606879060483e-06,
+      "loss": 0.7056,
+      "step": 170
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.5882241768547103,
+      "learning_rate": 3.8469206462878e-06,
+      "loss": 0.5314,
+      "step": 171
+    },
+    {
+      "epoch": 3.017543859649123,
+      "grad_norm": 0.9204123171996009,
+      "learning_rate": 3.790390522001662e-06,
+      "loss": 0.5576,
+      "step": 172
+    },
+    {
+      "epoch": 3.0350877192982457,
+      "grad_norm": 0.4343635198218746,
+      "learning_rate": 3.7340241594759917e-06,
+      "loss": 0.6756,
+      "step": 173
+    },
+    {
+      "epoch": 3.0526315789473686,
+      "grad_norm": 0.724604261743242,
+      "learning_rate": 3.6778291898139907e-06,
+      "loss": 0.7069,
+      "step": 174
+    },
+    {
+      "epoch": 3.0701754385964914,
+      "grad_norm": 0.8051279733139357,
+      "learning_rate": 3.6218132209150047e-06,
+      "loss": 0.6936,
+      "step": 175
+    },
+    {
+      "epoch": 3.087719298245614,
+      "grad_norm": 0.8593841140435677,
+      "learning_rate": 3.5659838364445505e-06,
+      "loss": 0.6195,
+      "step": 176
+    },
+    {
+      "epoch": 3.1052631578947367,
+      "grad_norm": 0.6082526627163497,
+      "learning_rate": 3.51034859480759e-06,
+      "loss": 0.7254,
+      "step": 177
+    },
+    {
+      "epoch": 3.1228070175438596,
+      "grad_norm": 0.553107734464842,
+      "learning_rate": 3.4549150281252635e-06,
+      "loss": 0.6935,
+      "step": 178
+    },
+    {
+      "epoch": 3.1403508771929824,
+      "grad_norm": 0.6259105806074109,
+      "learning_rate": 3.399690641215142e-06,
+      "loss": 0.5926,
+      "step": 179
+    },
+    {
+      "epoch": 3.1578947368421053,
+      "grad_norm": 0.8279774578774789,
+      "learning_rate": 3.3446829105752103e-06,
+      "loss": 0.597,
+      "step": 180
+    },
+    {
+      "epoch": 3.175438596491228,
+      "grad_norm": 0.5650831808352024,
+      "learning_rate": 3.289899283371657e-06,
+      "loss": 0.6382,
+      "step": 181
+    },
+    {
+      "epoch": 3.192982456140351,
+      "grad_norm": 0.4884733646140478,
+      "learning_rate": 3.2353471764306567e-06,
+      "loss": 0.7769,
+      "step": 182
+    },
+    {
+      "epoch": 3.2105263157894735,
+      "grad_norm": 0.5607761451916414,
+      "learning_rate": 3.1810339752342446e-06,
+      "loss": 0.68,
+      "step": 183
+    },
+    {
+      "epoch": 3.2280701754385963,
+      "grad_norm": 0.6418674845376859,
+      "learning_rate": 3.12696703292044e-06,
+      "loss": 0.6189,
+      "step": 184
+    },
+    {
+      "epoch": 3.245614035087719,
+      "grad_norm": 0.44836764225924963,
+      "learning_rate": 3.0731536692877596e-06,
+      "loss": 0.6083,
+      "step": 185
+    },
+    {
+      "epoch": 3.263157894736842,
+      "grad_norm": 0.4659021515168168,
+      "learning_rate": 3.019601169804216e-06,
+      "loss": 0.5413,
+      "step": 186
+    },
+    {
+      "epoch": 3.280701754385965,
+      "grad_norm": 0.46113074765941303,
+      "learning_rate": 2.966316784621e-06,
+      "loss": 0.6309,
+      "step": 187
+    },
+    {
+      "epoch": 3.2982456140350878,
+      "grad_norm": 0.4640366156738567,
+      "learning_rate": 2.9133077275909112e-06,
+      "loss": 0.6916,
+      "step": 188
+    },
+    {
+      "epoch": 3.3157894736842106,
+      "grad_norm": 0.5701252737387132,
+      "learning_rate": 2.86058117529173e-06,
+      "loss": 0.6016,
+      "step": 189
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.5159025710066555,
+      "learning_rate": 2.8081442660546126e-06,
+      "loss": 0.6142,
+      "step": 190
+    },
+    {
+      "epoch": 3.3508771929824563,
+      "grad_norm": 0.5141665185455173,
+      "learning_rate": 2.7560040989976894e-06,
+      "loss": 0.6463,
+      "step": 191
+    },
+    {
+      "epoch": 3.3684210526315788,
+      "grad_norm": 0.44717813123595995,
+      "learning_rate": 2.7041677330649408e-06,
+      "loss": 0.5449,
+      "step": 192
+    },
+    {
+      "epoch": 3.3859649122807016,
+      "grad_norm": 0.43385553399677956,
+      "learning_rate": 2.6526421860705474e-06,
+      "loss": 0.5635,
+      "step": 193
+    },
+    {
+      "epoch": 3.4035087719298245,
+      "grad_norm": 0.5031797225541397,
+      "learning_rate": 2.601434433748771e-06,
+      "loss": 0.5954,
+      "step": 194
+    },
+    {
+      "epoch": 3.4210526315789473,
+      "grad_norm": 0.9140713143718528,
+      "learning_rate": 2.550551408809566e-06,
+      "loss": 0.5228,
+      "step": 195
+    },
+    {
+      "epoch": 3.43859649122807,
+      "grad_norm": 0.4859704318245885,
+      "learning_rate": 2.5000000000000015e-06,
+      "loss": 0.6438,
+      "step": 196
+    },
+    {
+      "epoch": 3.456140350877193,
+      "grad_norm": 0.6542605053552696,
+      "learning_rate": 2.4497870511716237e-06,
+      "loss": 0.4806,
+      "step": 197
+    },
+    {
+      "epoch": 3.473684210526316,
+      "grad_norm": 0.49249625575639394,
+      "learning_rate": 2.3999193603539234e-06,
+      "loss": 0.5957,
+      "step": 198
+    },
+    {
+      "epoch": 3.4912280701754383,
+      "grad_norm": 0.5034055187641854,
+      "learning_rate": 2.3504036788339763e-06,
+      "loss": 0.6309,
+      "step": 199
+    },
+    {
+      "epoch": 3.5087719298245617,
+      "grad_norm": 0.42476198096216705,
+      "learning_rate": 2.3012467102424373e-06,
+      "loss": 0.5814,
+      "step": 200
+    },
+    {
+      "epoch": 3.526315789473684,
+      "grad_norm": 0.5214191130229293,
+      "learning_rate": 2.2524551096459703e-06,
+      "loss": 0.6347,
+      "step": 201
+    },
+    {
+      "epoch": 3.543859649122807,
+      "grad_norm": 0.6829927292880883,
+      "learning_rate": 2.204035482646267e-06,
+      "loss": 0.5123,
+      "step": 202
+    },
+    {
+      "epoch": 3.56140350877193,
+      "grad_norm": 0.4302781137797259,
+      "learning_rate": 2.155994384485742e-06,
+      "loss": 0.7108,
+      "step": 203
+    },
+    {
+      "epoch": 3.5789473684210527,
+      "grad_norm": 0.5763052347364854,
+      "learning_rate": 2.1083383191600676e-06,
+      "loss": 0.5797,
+      "step": 204
+    },
+    {
+      "epoch": 3.5964912280701755,
+      "grad_norm": 0.5659704372185881,
+      "learning_rate": 2.061073738537635e-06,
+      "loss": 0.5944,
+      "step": 205
+    },
+    {
+      "epoch": 3.6140350877192984,
+      "grad_norm": 0.41487096065603857,
+      "learning_rate": 2.0142070414860704e-06,
+      "loss": 0.5244,
+      "step": 206
+    },
+    {
+      "epoch": 3.6315789473684212,
+      "grad_norm": 0.5870004898342148,
+      "learning_rate": 1.9677445730059348e-06,
+      "loss": 0.538,
+      "step": 207
+    },
+    {
+      "epoch": 3.6491228070175437,
+      "grad_norm": 0.5329927474864604,
+      "learning_rate": 1.9216926233717087e-06,
+      "loss": 0.5704,
+      "step": 208
+    },
+    {
+      "epoch": 3.6666666666666665,
+      "grad_norm": 0.5309898738066137,
+      "learning_rate": 1.8760574272802002e-06,
+      "loss": 0.6323,
+      "step": 209
+    },
+    {
+      "epoch": 3.6842105263157894,
+      "grad_norm": 0.4723299355114008,
+      "learning_rate": 1.8308451630064484e-06,
+      "loss": 0.5625,
+      "step": 210
+    },
+    {
+      "epoch": 3.7017543859649122,
+      "grad_norm": 0.5647620157506328,
+      "learning_rate": 1.7860619515673034e-06,
+      "loss": 0.6485,
+      "step": 211
+    },
+    {
+      "epoch": 3.719298245614035,
+      "grad_norm": 0.5197824522938973,
+      "learning_rate": 1.7417138558927244e-06,
+      "loss": 0.617,
+      "step": 212
+    },
+    {
+      "epoch": 3.736842105263158,
+      "grad_norm": 0.46400200187546625,
+      "learning_rate": 1.6978068800049624e-06,
+      "loss": 0.6766,
+      "step": 213
+    },
+    {
+      "epoch": 3.754385964912281,
+      "grad_norm": 0.5140751610952029,
+      "learning_rate": 1.6543469682057105e-06,
+      "loss": 0.6786,
+      "step": 214
+    },
+    {
+      "epoch": 3.7719298245614032,
+      "grad_norm": 0.4501937787808013,
+      "learning_rate": 1.611340004271339e-06,
+      "loss": 0.7169,
+      "step": 215
+    },
+    {
+      "epoch": 3.7894736842105265,
+      "grad_norm": 0.49277258899480275,
+      "learning_rate": 1.5687918106563326e-06,
+      "loss": 0.6159,
+      "step": 216
+    },
+    {
+      "epoch": 3.807017543859649,
+      "grad_norm": 0.4835193138921562,
+      "learning_rate": 1.5267081477050132e-06,
+      "loss": 0.6639,
+      "step": 217
+    },
+    {
+      "epoch": 3.824561403508772,
+      "grad_norm": 0.4934861837773664,
+      "learning_rate": 1.4850947128716914e-06,
+      "loss": 0.6838,
+      "step": 218
+    },
+    {
+      "epoch": 3.8421052631578947,
+      "grad_norm": 0.5337251099094051,
+      "learning_rate": 1.4439571399493146e-06,
+      "loss": 0.6425,
+      "step": 219
+    },
+    {
+      "epoch": 3.8596491228070176,
+      "grad_norm": 0.4776899649583783,
+      "learning_rate": 1.4033009983067454e-06,
+      "loss": 0.5455,
+      "step": 220
+    },
+    {
+      "epoch": 3.8771929824561404,
+      "grad_norm": 0.47502311351020604,
+      "learning_rate": 1.3631317921347564e-06,
+      "loss": 0.6362,
+      "step": 221
+    },
+    {
+      "epoch": 3.8947368421052633,
+      "grad_norm": 0.5104894499026789,
+      "learning_rate": 1.3234549597008572e-06,
+      "loss": 0.6279,
+      "step": 222
+    },
+    {
+      "epoch": 3.912280701754386,
+      "grad_norm": 0.4082083559752492,
+      "learning_rate": 1.2842758726130283e-06,
+      "loss": 0.5586,
+      "step": 223
+    },
+    {
+      "epoch": 3.9298245614035086,
+      "grad_norm": 0.5509227614337333,
+      "learning_rate": 1.2455998350925042e-06,
+      "loss": 0.5738,
+      "step": 224
+    },
+    {
+      "epoch": 3.9473684210526314,
+      "grad_norm": 0.5752399702323144,
+      "learning_rate": 1.2074320832556558e-06,
+      "loss": 0.6763,
+      "step": 225
+    },
+    {
+      "epoch": 3.9649122807017543,
+      "grad_norm": 0.6677084420144763,
+      "learning_rate": 1.1697777844051105e-06,
+      "loss": 0.65,
+      "step": 226
+    },
+    {
+      "epoch": 3.982456140350877,
+      "grad_norm": 0.5697982145058713,
+      "learning_rate": 1.132642036330181e-06,
+      "loss": 0.6729,
+      "step": 227
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.49136096649026906,
+      "learning_rate": 1.096029866616704e-06,
+      "loss": 0.5668,
+      "step": 228
+    },
+    {
+      "epoch": 4.017543859649122,
+      "grad_norm": 0.7231296994044895,
+      "learning_rate": 1.0599462319663906e-06,
+      "loss": 0.6329,
+      "step": 229
+    },
+    {
+      "epoch": 4.035087719298246,
+      "grad_norm": 1.0152116023186752,
+      "learning_rate": 1.0243960175257605e-06,
+      "loss": 0.5392,
+      "step": 230
+    },
+    {
+      "epoch": 4.052631578947368,
+      "grad_norm": 0.7560784344122596,
+      "learning_rate": 9.893840362247809e-07,
+      "loss": 0.5747,
+      "step": 231
+    },
+    {
+      "epoch": 4.0701754385964914,
+      "grad_norm": 0.6515700428214378,
+      "learning_rate": 9.549150281252633e-07,
+      "loss": 0.5582,
+      "step": 232
+    },
+    {
+      "epoch": 4.087719298245614,
+      "grad_norm": 0.45313908437035877,
+      "learning_rate": 9.209936597791407e-07,
+      "loss": 0.611,
+      "step": 233
+    },
+    {
+      "epoch": 4.105263157894737,
+      "grad_norm": 0.5370761928868943,
+      "learning_rate": 8.876245235966884e-07,
+      "loss": 0.607,
+      "step": 234
+    },
+    {
+      "epoch": 4.12280701754386,
+      "grad_norm": 0.43662222669206097,
+      "learning_rate": 8.54812137224792e-07,
+      "loss": 0.539,
+      "step": 235
+    },
+    {
+      "epoch": 4.140350877192983,
+      "grad_norm": 0.5046687968972682,
+      "learning_rate": 8.225609429353187e-07,
+      "loss": 0.6358,
+      "step": 236
+    },
+    {
+      "epoch": 4.157894736842105,
+      "grad_norm": 0.8052049999207681,
+      "learning_rate": 7.908753070237124e-07,
+      "loss": 0.4973,
+      "step": 237
+    },
+    {
+      "epoch": 4.175438596491228,
+      "grad_norm": 0.7693801835280304,
+      "learning_rate": 7.597595192178702e-07,
+      "loss": 0.5875,
+      "step": 238
+    },
+    {
+      "epoch": 4.192982456140351,
+      "grad_norm": 0.7221268272445959,
+      "learning_rate": 7.292177920973726e-07,
+      "loss": 0.5731,
+      "step": 239
+    },
+    {
+      "epoch": 4.2105263157894735,
+      "grad_norm": 0.8028148149327845,
+      "learning_rate": 6.992542605231739e-07,
+      "loss": 0.4967,
+      "step": 240
+    },
+    {
+      "epoch": 4.228070175438597,
+      "grad_norm": 0.47167171297763894,
+      "learning_rate": 6.698729810778065e-07,
+      "loss": 0.5715,
+      "step": 241
+    },
+    {
+      "epoch": 4.245614035087719,
+      "grad_norm": 0.5752392103057814,
+      "learning_rate": 6.410779315161885e-07,
+      "loss": 0.6086,
+      "step": 242
+    },
+    {
+      "epoch": 4.2631578947368425,
+      "grad_norm": 0.5906124825417852,
+      "learning_rate": 6.128730102270897e-07,
+      "loss": 0.5164,
+      "step": 243
+    },
+    {
+      "epoch": 4.280701754385965,
+      "grad_norm": 0.48280703799925595,
+      "learning_rate": 5.852620357053651e-07,
+      "loss": 0.6317,
+      "step": 244
+    },
+    {
+      "epoch": 4.298245614035087,
+      "grad_norm": 0.5637472304873896,
+      "learning_rate": 5.582487460349806e-07,
+      "loss": 0.6276,
+      "step": 245
+    },
+    {
+      "epoch": 4.315789473684211,
+      "grad_norm": 0.5762745465731932,
+      "learning_rate": 5.318367983829393e-07,
+      "loss": 0.6798,
+      "step": 246
+    },
+    {
+      "epoch": 4.333333333333333,
+      "grad_norm": 0.49370745341167194,
+      "learning_rate": 5.06029768504166e-07,
+      "loss": 0.628,
+      "step": 247
+    },
+    {
+      "epoch": 4.350877192982456,
+      "grad_norm": 0.49385098919261067,
+      "learning_rate": 4.808311502573976e-07,
+      "loss": 0.5806,
+      "step": 248
+    },
+    {
+      "epoch": 4.368421052631579,
+      "grad_norm": 0.4634445843058031,
+      "learning_rate": 4.562443551321788e-07,
+      "loss": 0.4806,
+      "step": 249
+    },
+    {
+      "epoch": 4.385964912280702,
+      "grad_norm": 0.4694066505533855,
+      "learning_rate": 4.322727117869951e-07,
+      "loss": 0.6639,
+      "step": 250
+    },
+    {
+      "epoch": 4.4035087719298245,
+      "grad_norm": 0.5133736584085855,
+      "learning_rate": 4.089194655986306e-07,
+      "loss": 0.6468,
+      "step": 251
+    },
+    {
+      "epoch": 4.421052631578947,
+      "grad_norm": 0.41817510526418245,
+      "learning_rate": 3.8618777822278854e-07,
+      "loss": 0.5321,
+      "step": 252
+    },
+    {
+      "epoch": 4.43859649122807,
+      "grad_norm": 0.5258679818299432,
+      "learning_rate": 3.6408072716606346e-07,
+      "loss": 0.5853,
+      "step": 253
+    },
+    {
+      "epoch": 4.456140350877193,
+      "grad_norm": 0.5010151986806652,
+      "learning_rate": 3.426013053692878e-07,
+      "loss": 0.6217,
+      "step": 254
+    },
+    {
+      "epoch": 4.473684210526316,
+      "grad_norm": 1.3240412775379553,
+      "learning_rate": 3.2175242080234314e-07,
+      "loss": 0.7006,
+      "step": 255
+    },
+    {
+      "epoch": 4.491228070175438,
+      "grad_norm": 0.4769147229082815,
+      "learning_rate": 3.015368960704584e-07,
+      "loss": 0.646,
+      "step": 256
+    },
+    {
+      "epoch": 4.508771929824562,
+      "grad_norm": 0.4407281001364891,
+      "learning_rate": 2.819574680320825e-07,
+      "loss": 0.6046,
+      "step": 257
+    },
+    {
+      "epoch": 4.526315789473684,
+      "grad_norm": 0.4598786522354988,
+      "learning_rate": 2.63016787428354e-07,
+      "loss": 0.5428,
+      "step": 258
+    },
+    {
+      "epoch": 4.543859649122807,
+      "grad_norm": 0.41674257024456035,
+      "learning_rate": 2.447174185242324e-07,
+      "loss": 0.6752,
+      "step": 259
+    },
+    {
+      "epoch": 4.56140350877193,
+      "grad_norm": 0.5369949713818419,
+      "learning_rate": 2.2706183876134047e-07,
+      "loss": 0.4643,
+      "step": 260
+    },
+    {
+      "epoch": 4.578947368421053,
+      "grad_norm": 0.5613046441983963,
+      "learning_rate": 2.1005243842255552e-07,
+      "loss": 0.5297,
+      "step": 261
+    },
+    {
+      "epoch": 4.5964912280701755,
+      "grad_norm": 0.43892104451673636,
+      "learning_rate": 1.9369152030840553e-07,
+      "loss": 0.5259,
+      "step": 262
+    },
+    {
+      "epoch": 4.614035087719298,
+      "grad_norm": 0.44720438149887654,
+      "learning_rate": 1.779812994253055e-07,
+      "loss": 0.4691,
+      "step": 263
+    },
+    {
+      "epoch": 4.631578947368421,
+      "grad_norm": 0.5499894560289246,
+      "learning_rate": 1.6292390268568103e-07,
+      "loss": 0.5457,
+      "step": 264
+    },
+    {
+      "epoch": 4.649122807017544,
+      "grad_norm": 0.4317789123558038,
+      "learning_rate": 1.4852136862001766e-07,
+      "loss": 0.6124,
+      "step": 265
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.4859803553610326,
+      "learning_rate": 1.3477564710088097e-07,
+      "loss": 0.5553,
+      "step": 266
+    },
+    {
+      "epoch": 4.684210526315789,
+      "grad_norm": 0.50380337358526,
+      "learning_rate": 1.2168859907892904e-07,
+      "loss": 0.5771,
+      "step": 267
+    },
+    {
+      "epoch": 4.701754385964913,
+      "grad_norm": 0.500528985151703,
+      "learning_rate": 1.0926199633097156e-07,
+      "loss": 0.4774,
+      "step": 268
+    },
+    {
+      "epoch": 4.719298245614035,
+      "grad_norm": 0.46397116813925426,
+      "learning_rate": 9.749752122010347e-08,
+      "loss": 0.6158,
+      "step": 269
+    },
+    {
+      "epoch": 4.7368421052631575,
+      "grad_norm": 0.4666174144563547,
+      "learning_rate": 8.639676646793382e-08,
+      "loss": 0.5777,
+      "step": 270
+    },
+    {
+      "epoch": 4.754385964912281,
+      "grad_norm": 0.4514233976151472,
+      "learning_rate": 7.59612349389599e-08,
+      "loss": 0.592,
+      "step": 271
+    },
+    {
+      "epoch": 4.771929824561403,
+      "grad_norm": 0.42960681268228296,
+      "learning_rate": 6.61923394371039e-08,
+      "loss": 0.7048,
+      "step": 272
+    },
+    {
+      "epoch": 4.7894736842105265,
+      "grad_norm": 0.4600055946113908,
+      "learning_rate": 5.709140251444201e-08,
+      "loss": 0.598,
+      "step": 273
+    },
+    {
+      "epoch": 4.807017543859649,
+      "grad_norm": 0.5153423305042425,
+      "learning_rate": 4.865965629214819e-08,
+      "loss": 0.5696,
+      "step": 274
+    },
+    {
+      "epoch": 4.824561403508772,
+      "grad_norm": 0.466711193283292,
+      "learning_rate": 4.0898242293691546e-08,
+      "loss": 0.5346,
+      "step": 275
+    },
+    {
+      "epoch": 4.842105263157895,
+      "grad_norm": 0.4612852905824852,
+      "learning_rate": 3.3808211290284886e-08,
+      "loss": 0.5819,
+      "step": 276
+    },
+    {
+      "epoch": 4.859649122807017,
+      "grad_norm": 0.5693360179602536,
+      "learning_rate": 2.7390523158633552e-08,
+      "loss": 0.5941,
+      "step": 277
+    },
+    {
+      "epoch": 4.87719298245614,
+      "grad_norm": 0.39865215194796066,
+      "learning_rate": 2.1646046750978255e-08,
+      "loss": 0.6095,
+      "step": 278
+    },
+    {
+      "epoch": 4.894736842105263,
+      "grad_norm": 0.40135313928244487,
+      "learning_rate": 1.657555977746972e-08,
+      "loss": 0.5701,
+      "step": 279
+    },
+    {
+      "epoch": 4.912280701754386,
+      "grad_norm": 0.5644171195119998,
+      "learning_rate": 1.2179748700879013e-08,
+      "loss": 0.6043,
+      "step": 280
+    },
+    {
+      "epoch": 4.9298245614035086,
+      "grad_norm": 0.47303740467191896,
+      "learning_rate": 8.459208643659122e-09,
+      "loss": 0.5963,
+      "step": 281
+    },
+    {
+      "epoch": 4.947368421052632,
+      "grad_norm": 0.6183026707892192,
+      "learning_rate": 5.414443307377171e-09,
+      "loss": 0.6918,
+      "step": 282
+    },
+    {
+      "epoch": 4.964912280701754,
+      "grad_norm": 0.4007786407111042,
+      "learning_rate": 3.0458649045211897e-09,
+      "loss": 0.544,
+      "step": 283
+    },
+    {
+      "epoch": 4.982456140350877,
+      "grad_norm": 0.4178834314449404,
+      "learning_rate": 1.3537941026914302e-09,
+      "loss": 0.6197,
+      "step": 284
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.43535643209515534,
+      "learning_rate": 3.384599811889766e-10,
+      "loss": 0.625,
+      "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 sha256:3edaf61c128494a6aa01c196934ffa844211797641021ad8a5d37a8c03e1ddc3
+size 8081