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:b1efd4f290b5b91fb72d5873f1d77cbc69dd43a993222c7f1513ad3ab01cd7a5
+size 4997184968

model-00002-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa6e3fa7cb9fc03746884f4d3550fcdcbf05072fcb5849ff85bd0009eb0e848b
+size 4997741608

model-00003-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec5b50d55a63e7c71c8887c74ca9cdab69819f0d8e290b28593a0825e9029387
+size 4997742208

model-00004-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4c7e6e73e3e31a649e7e173ca0037c25b1588244fb94b4e70122c2e18bcf9f6
+size 4997743184

model-00005-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de06482521655933971f5fe2555422707b1e2b74939e3fdd993fa9fc5006a1eb
+size 4997743184

model-00006-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16fa817f4831a0ae36acf2c85517ce81e1bcce8020f1f9b445e083ca7885c7ba
+size 4997743184

model-00007-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:212a304bd61d6501344c6cd0afccd0aafbfd1a1c6b2e6773a74be10da9e768cc
+size 4997743184

model-00008-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a91d882fae6048cb06565cbbea1ac43dfb16315d2b5404ef154a5b4b5311cd1
+size 4997743184

model-00009-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b902f94ba942aa1111d699b11dcf7dac39967a41040d2c9a29a38285758d45c
+size 4997743184

model-00010-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf4666206817216eaadee888739f3ed88959646959caf10456390706335a0477
+size 4997743184

model-00011-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0f0df16eff1010f540ef20b0ee1b55f14ebdc5d287f71c6b10bc2cd8913f734
+size 4997743184

model-00012-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb1ed9e32cfb3919e4daf7a7bfefd3029ae0d8f89d9d5b0289762c6973a8dc04
+size 4997743184

model-00013-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42882d983f94c89459124b010ccb895cb3e304d33e6d7be098121d0fa3a69fe1
+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": 2.256555232517452,
+      "learning_rate": 0.0,
+      "loss": 0.8483,
+      "step": 1
+    },
+    {
+      "epoch": 0.03508771929824561,
+      "grad_norm": 2.205270245564155,
+      "learning_rate": 6.666666666666667e-07,
+      "loss": 0.8765,
+      "step": 2
+    },
+    {
+      "epoch": 0.05263157894736842,
+      "grad_norm": 2.361805594728836,
+      "learning_rate": 1.3333333333333334e-06,
+      "loss": 1.0209,
+      "step": 3
+    },
+    {
+      "epoch": 0.07017543859649122,
+      "grad_norm": 2.481606951567156,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.062,
+      "step": 4
+    },
+    {
+      "epoch": 0.08771929824561403,
+      "grad_norm": 1.923229005445868,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 0.8564,
+      "step": 5
+    },
+    {
+      "epoch": 0.10526315789473684,
+      "grad_norm": 2.1103374249725833,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 1.0082,
+      "step": 6
+    },
+    {
+      "epoch": 0.12280701754385964,
+      "grad_norm": 1.9672661474560547,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.9978,
+      "step": 7
+    },
+    {
+      "epoch": 0.14035087719298245,
+      "grad_norm": 1.810080319596374,
+      "learning_rate": 4.666666666666667e-06,
+      "loss": 0.9468,
+      "step": 8
+    },
+    {
+      "epoch": 0.15789473684210525,
+      "grad_norm": 2.2501314657852776,
+      "learning_rate": 5.333333333333334e-06,
+      "loss": 0.9614,
+      "step": 9
+    },
+    {
+      "epoch": 0.17543859649122806,
+      "grad_norm": 1.3827469830753207,
+      "learning_rate": 6e-06,
+      "loss": 0.9873,
+      "step": 10
+    },
+    {
+      "epoch": 0.19298245614035087,
+      "grad_norm": 1.5862884873258083,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 1.0305,
+      "step": 11
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 1.57454716689858,
+      "learning_rate": 7.333333333333333e-06,
+      "loss": 0.9006,
+      "step": 12
+    },
+    {
+      "epoch": 0.22807017543859648,
+      "grad_norm": 1.5638780668313408,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.936,
+      "step": 13
+    },
+    {
+      "epoch": 0.24561403508771928,
+      "grad_norm": 1.5588053318143216,
+      "learning_rate": 8.666666666666668e-06,
+      "loss": 0.9131,
+      "step": 14
+    },
+    {
+      "epoch": 0.2631578947368421,
+      "grad_norm": 1.2968184983553472,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.9716,
+      "step": 15
+    },
+    {
+      "epoch": 0.2807017543859649,
+      "grad_norm": 1.3213006637878912,
+      "learning_rate": 1e-05,
+      "loss": 0.922,
+      "step": 16
+    },
+    {
+      "epoch": 0.2982456140350877,
+      "grad_norm": 1.4807190056139585,
+      "learning_rate": 9.999661540018812e-06,
+      "loss": 1.0189,
+      "step": 17
+    },
+    {
+      "epoch": 0.3157894736842105,
+      "grad_norm": 1.7897441321489662,
+      "learning_rate": 9.99864620589731e-06,
+      "loss": 0.903,
+      "step": 18
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 1.1554720466501127,
+      "learning_rate": 9.99695413509548e-06,
+      "loss": 0.9231,
+      "step": 19
+    },
+    {
+      "epoch": 0.3508771929824561,
+      "grad_norm": 1.0430048166465928,
+      "learning_rate": 9.994585556692624e-06,
+      "loss": 0.9584,
+      "step": 20
+    },
+    {
+      "epoch": 0.3684210526315789,
+      "grad_norm": 0.8734173643171832,
+      "learning_rate": 9.991540791356342e-06,
+      "loss": 0.8954,
+      "step": 21
+    },
+    {
+      "epoch": 0.38596491228070173,
+      "grad_norm": 0.8505715098348454,
+      "learning_rate": 9.987820251299121e-06,
+      "loss": 0.7634,
+      "step": 22
+    },
+    {
+      "epoch": 0.40350877192982454,
+      "grad_norm": 0.932431362896424,
+      "learning_rate": 9.98342444022253e-06,
+      "loss": 0.8972,
+      "step": 23
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.994313791348145,
+      "learning_rate": 9.978353953249023e-06,
+      "loss": 0.7951,
+      "step": 24
+    },
+    {
+      "epoch": 0.43859649122807015,
+      "grad_norm": 0.8497211408850255,
+      "learning_rate": 9.972609476841368e-06,
+      "loss": 0.9234,
+      "step": 25
+    },
+    {
+      "epoch": 0.45614035087719296,
+      "grad_norm": 0.9190618422149534,
+      "learning_rate": 9.966191788709716e-06,
+      "loss": 0.8877,
+      "step": 26
+    },
+    {
+      "epoch": 0.47368421052631576,
+      "grad_norm": 0.8415994406297493,
+      "learning_rate": 9.959101757706308e-06,
+      "loss": 0.8086,
+      "step": 27
+    },
+    {
+      "epoch": 0.49122807017543857,
+      "grad_norm": 0.742782382735312,
+      "learning_rate": 9.951340343707852e-06,
+      "loss": 0.8769,
+      "step": 28
+    },
+    {
+      "epoch": 0.5087719298245614,
+      "grad_norm": 0.6160964124153906,
+      "learning_rate": 9.942908597485558e-06,
+      "loss": 0.8978,
+      "step": 29
+    },
+    {
+      "epoch": 0.5263157894736842,
+      "grad_norm": 0.5976609614605858,
+      "learning_rate": 9.933807660562898e-06,
+      "loss": 0.7837,
+      "step": 30
+    },
+    {
+      "epoch": 0.543859649122807,
+      "grad_norm": 0.5560568208246192,
+      "learning_rate": 9.924038765061042e-06,
+      "loss": 0.8396,
+      "step": 31
+    },
+    {
+      "epoch": 0.5614035087719298,
+      "grad_norm": 0.5836679513268156,
+      "learning_rate": 9.913603233532067e-06,
+      "loss": 0.7701,
+      "step": 32
+    },
+    {
+      "epoch": 0.5789473684210527,
+      "grad_norm": 0.7645333976719451,
+      "learning_rate": 9.902502478779897e-06,
+      "loss": 0.8688,
+      "step": 33
+    },
+    {
+      "epoch": 0.5964912280701754,
+      "grad_norm": 0.7612685741039715,
+      "learning_rate": 9.890738003669029e-06,
+      "loss": 0.9675,
+      "step": 34
+    },
+    {
+      "epoch": 0.6140350877192983,
+      "grad_norm": 0.6672122472693547,
+      "learning_rate": 9.878311400921072e-06,
+      "loss": 0.8122,
+      "step": 35
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.6198318849807769,
+      "learning_rate": 9.86522435289912e-06,
+      "loss": 0.8276,
+      "step": 36
+    },
+    {
+      "epoch": 0.6491228070175439,
+      "grad_norm": 0.5915273774198798,
+      "learning_rate": 9.851478631379982e-06,
+      "loss": 0.862,
+      "step": 37
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.7597701986857002,
+      "learning_rate": 9.83707609731432e-06,
+      "loss": 0.8755,
+      "step": 38
+    },
+    {
+      "epoch": 0.6842105263157895,
+      "grad_norm": 0.6165755211119401,
+      "learning_rate": 9.822018700574696e-06,
+      "loss": 0.898,
+      "step": 39
+    },
+    {
+      "epoch": 0.7017543859649122,
+      "grad_norm": 0.546722974197266,
+      "learning_rate": 9.806308479691595e-06,
+      "loss": 0.8637,
+      "step": 40
+    },
+    {
+      "epoch": 0.7192982456140351,
+      "grad_norm": 0.5624327518167074,
+      "learning_rate": 9.789947561577445e-06,
+      "loss": 0.8685,
+      "step": 41
+    },
+    {
+      "epoch": 0.7368421052631579,
+      "grad_norm": 0.5747062370239466,
+      "learning_rate": 9.77293816123866e-06,
+      "loss": 0.8039,
+      "step": 42
+    },
+    {
+      "epoch": 0.7543859649122807,
+      "grad_norm": 0.5540954371595526,
+      "learning_rate": 9.755282581475769e-06,
+      "loss": 0.8365,
+      "step": 43
+    },
+    {
+      "epoch": 0.7719298245614035,
+      "grad_norm": 0.5678372679311994,
+      "learning_rate": 9.736983212571646e-06,
+      "loss": 0.7528,
+      "step": 44
+    },
+    {
+      "epoch": 0.7894736842105263,
+      "grad_norm": 0.5498130339645014,
+      "learning_rate": 9.718042531967918e-06,
+      "loss": 0.8692,
+      "step": 45
+    },
+    {
+      "epoch": 0.8070175438596491,
+      "grad_norm": 0.723915088763488,
+      "learning_rate": 9.698463103929542e-06,
+      "loss": 0.8465,
+      "step": 46
+    },
+    {
+      "epoch": 0.8245614035087719,
+      "grad_norm": 0.5693533968278255,
+      "learning_rate": 9.678247579197658e-06,
+      "loss": 0.8505,
+      "step": 47
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.444847109914399,
+      "learning_rate": 9.657398694630713e-06,
+      "loss": 0.8346,
+      "step": 48
+    },
+    {
+      "epoch": 0.8596491228070176,
+      "grad_norm": 0.5481527250136425,
+      "learning_rate": 9.635919272833938e-06,
+      "loss": 0.9264,
+      "step": 49
+    },
+    {
+      "epoch": 0.8771929824561403,
+      "grad_norm": 0.4850156760944371,
+      "learning_rate": 9.613812221777212e-06,
+      "loss": 0.6667,
+      "step": 50
+    },
+    {
+      "epoch": 0.8947368421052632,
+      "grad_norm": 0.5196718609108957,
+      "learning_rate": 9.591080534401371e-06,
+      "loss": 0.8957,
+      "step": 51
+    },
+    {
+      "epoch": 0.9122807017543859,
+      "grad_norm": 0.5280436158388971,
+      "learning_rate": 9.567727288213005e-06,
+      "loss": 0.8639,
+      "step": 52
+    },
+    {
+      "epoch": 0.9298245614035088,
+      "grad_norm": 0.5997684565130753,
+      "learning_rate": 9.543755644867823e-06,
+      "loss": 0.8164,
+      "step": 53
+    },
+    {
+      "epoch": 0.9473684210526315,
+      "grad_norm": 0.5103541862955409,
+      "learning_rate": 9.519168849742603e-06,
+      "loss": 0.7713,
+      "step": 54
+    },
+    {
+      "epoch": 0.9649122807017544,
+      "grad_norm": 0.5148641741360179,
+      "learning_rate": 9.493970231495836e-06,
+      "loss": 0.8101,
+      "step": 55
+    },
+    {
+      "epoch": 0.9824561403508771,
+      "grad_norm": 0.46692510221358374,
+      "learning_rate": 9.468163201617063e-06,
+      "loss": 0.8216,
+      "step": 56
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.666709235115788,
+      "learning_rate": 9.441751253965022e-06,
+      "loss": 0.7334,
+      "step": 57
+    },
+    {
+      "epoch": 1.0175438596491229,
+      "grad_norm": 0.5936450769507504,
+      "learning_rate": 9.414737964294636e-06,
+      "loss": 0.8429,
+      "step": 58
+    },
+    {
+      "epoch": 1.0350877192982457,
+      "grad_norm": 0.5402515184573428,
+      "learning_rate": 9.38712698977291e-06,
+      "loss": 0.7318,
+      "step": 59
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 0.9071602417358279,
+      "learning_rate": 9.358922068483813e-06,
+      "loss": 0.8178,
+      "step": 60
+    },
+    {
+      "epoch": 1.0701754385964912,
+      "grad_norm": 0.5181420637496472,
+      "learning_rate": 9.330127018922195e-06,
+      "loss": 0.6926,
+      "step": 61
+    },
+    {
+      "epoch": 1.087719298245614,
+      "grad_norm": 0.516304628920244,
+      "learning_rate": 9.30074573947683e-06,
+      "loss": 0.7764,
+      "step": 62
+    },
+    {
+      "epoch": 1.1052631578947367,
+      "grad_norm": 0.6092031389895063,
+      "learning_rate": 9.27078220790263e-06,
+      "loss": 0.804,
+      "step": 63
+    },
+    {
+      "epoch": 1.1228070175438596,
+      "grad_norm": 0.5739882837391556,
+      "learning_rate": 9.24024048078213e-06,
+      "loss": 0.7961,
+      "step": 64
+    },
+    {
+      "epoch": 1.1403508771929824,
+      "grad_norm": 0.4850959044650596,
+      "learning_rate": 9.209124692976287e-06,
+      "loss": 0.719,
+      "step": 65
+    },
+    {
+      "epoch": 1.1578947368421053,
+      "grad_norm": 0.5269909739571422,
+      "learning_rate": 9.177439057064684e-06,
+      "loss": 0.7205,
+      "step": 66
+    },
+    {
+      "epoch": 1.1754385964912282,
+      "grad_norm": 0.5314922201966987,
+      "learning_rate": 9.145187862775208e-06,
+      "loss": 0.7181,
+      "step": 67
+    },
+    {
+      "epoch": 1.1929824561403508,
+      "grad_norm": 0.4899793932301242,
+      "learning_rate": 9.112375476403313e-06,
+      "loss": 0.7646,
+      "step": 68
+    },
+    {
+      "epoch": 1.2105263157894737,
+      "grad_norm": 0.5480170596915535,
+      "learning_rate": 9.079006340220862e-06,
+      "loss": 0.8132,
+      "step": 69
+    },
+    {
+      "epoch": 1.2280701754385965,
+      "grad_norm": 0.5344703713427396,
+      "learning_rate": 9.045084971874738e-06,
+      "loss": 0.7621,
+      "step": 70
+    },
+    {
+      "epoch": 1.2456140350877192,
+      "grad_norm": 0.5142366314984944,
+      "learning_rate": 9.01061596377522e-06,
+      "loss": 0.8121,
+      "step": 71
+    },
+    {
+      "epoch": 1.263157894736842,
+      "grad_norm": 0.5877062225708843,
+      "learning_rate": 8.97560398247424e-06,
+      "loss": 0.7764,
+      "step": 72
+    },
+    {
+      "epoch": 1.280701754385965,
+      "grad_norm": 0.5275593499501555,
+      "learning_rate": 8.94005376803361e-06,
+      "loss": 0.7774,
+      "step": 73
+    },
+    {
+      "epoch": 1.2982456140350878,
+      "grad_norm": 0.49685789720182055,
+      "learning_rate": 8.903970133383297e-06,
+      "loss": 0.6981,
+      "step": 74
+    },
+    {
+      "epoch": 1.3157894736842106,
+      "grad_norm": 0.475321792037016,
+      "learning_rate": 8.867357963669821e-06,
+      "loss": 0.8256,
+      "step": 75
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.4843848014144311,
+      "learning_rate": 8.83022221559489e-06,
+      "loss": 0.7196,
+      "step": 76
+    },
+    {
+      "epoch": 1.3508771929824561,
+      "grad_norm": 1.496794429410156,
+      "learning_rate": 8.792567916744346e-06,
+      "loss": 0.6912,
+      "step": 77
+    },
+    {
+      "epoch": 1.368421052631579,
+      "grad_norm": 0.6153325721376482,
+      "learning_rate": 8.754400164907496e-06,
+      "loss": 0.811,
+      "step": 78
+    },
+    {
+      "epoch": 1.3859649122807016,
+      "grad_norm": 0.6721942051923293,
+      "learning_rate": 8.715724127386971e-06,
+      "loss": 0.7933,
+      "step": 79
+    },
+    {
+      "epoch": 1.4035087719298245,
+      "grad_norm": 0.5030720500009336,
+      "learning_rate": 8.676545040299145e-06,
+      "loss": 0.743,
+      "step": 80
+    },
+    {
+      "epoch": 1.4210526315789473,
+      "grad_norm": 0.5538931248913758,
+      "learning_rate": 8.636868207865244e-06,
+      "loss": 0.767,
+      "step": 81
+    },
+    {
+      "epoch": 1.4385964912280702,
+      "grad_norm": 0.5337323807177417,
+      "learning_rate": 8.596699001693257e-06,
+      "loss": 0.7524,
+      "step": 82
+    },
+    {
+      "epoch": 1.456140350877193,
+      "grad_norm": 0.5308647219753817,
+      "learning_rate": 8.556042860050686e-06,
+      "loss": 0.6872,
+      "step": 83
+    },
+    {
+      "epoch": 1.4736842105263157,
+      "grad_norm": 0.47526293846704515,
+      "learning_rate": 8.51490528712831e-06,
+      "loss": 0.6151,
+      "step": 84
+    },
+    {
+      "epoch": 1.4912280701754386,
+      "grad_norm": 0.5167817753474783,
+      "learning_rate": 8.473291852294986e-06,
+      "loss": 0.7717,
+      "step": 85
+    },
+    {
+      "epoch": 1.5087719298245614,
+      "grad_norm": 0.581779409307219,
+      "learning_rate": 8.43120818934367e-06,
+      "loss": 0.7539,
+      "step": 86
+    },
+    {
+      "epoch": 1.526315789473684,
+      "grad_norm": 0.46750024633090104,
+      "learning_rate": 8.388659995728662e-06,
+      "loss": 0.8867,
+      "step": 87
+    },
+    {
+      "epoch": 1.543859649122807,
+      "grad_norm": 0.7775428447931367,
+      "learning_rate": 8.345653031794292e-06,
+      "loss": 0.7715,
+      "step": 88
+    },
+    {
+      "epoch": 1.5614035087719298,
+      "grad_norm": 0.45420868791461705,
+      "learning_rate": 8.302193119995038e-06,
+      "loss": 0.8097,
+      "step": 89
+    },
+    {
+      "epoch": 1.5789473684210527,
+      "grad_norm": 0.5734514921428913,
+      "learning_rate": 8.258286144107277e-06,
+      "loss": 0.7128,
+      "step": 90
+    },
+    {
+      "epoch": 1.5964912280701755,
+      "grad_norm": 0.5283496931762547,
+      "learning_rate": 8.213938048432697e-06,
+      "loss": 0.7471,
+      "step": 91
+    },
+    {
+      "epoch": 1.6140350877192984,
+      "grad_norm": 0.9000918883180169,
+      "learning_rate": 8.16915483699355e-06,
+      "loss": 0.7585,
+      "step": 92
+    },
+    {
+      "epoch": 1.631578947368421,
+      "grad_norm": 0.42473651420290276,
+      "learning_rate": 8.123942572719801e-06,
+      "loss": 0.6968,
+      "step": 93
+    },
+    {
+      "epoch": 1.6491228070175439,
+      "grad_norm": 0.4566134251259284,
+      "learning_rate": 8.078307376628292e-06,
+      "loss": 0.703,
+      "step": 94
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.5418186278604009,
+      "learning_rate": 8.032255426994069e-06,
+      "loss": 0.8007,
+      "step": 95
+    },
+    {
+      "epoch": 1.6842105263157894,
+      "grad_norm": 0.5733519308896541,
+      "learning_rate": 7.985792958513932e-06,
+      "loss": 0.7076,
+      "step": 96
+    },
+    {
+      "epoch": 1.7017543859649122,
+      "grad_norm": 0.5094945171963646,
+      "learning_rate": 7.938926261462366e-06,
+      "loss": 0.8732,
+      "step": 97
+    },
+    {
+      "epoch": 1.719298245614035,
+      "grad_norm": 0.4620218725432511,
+      "learning_rate": 7.891661680839932e-06,
+      "loss": 0.7513,
+      "step": 98
+    },
+    {
+      "epoch": 1.736842105263158,
+      "grad_norm": 1.7222136001410835,
+      "learning_rate": 7.84400561551426e-06,
+      "loss": 0.7866,
+      "step": 99
+    },
+    {
+      "epoch": 1.7543859649122808,
+      "grad_norm": 0.485528858744073,
+      "learning_rate": 7.795964517353734e-06,
+      "loss": 0.7314,
+      "step": 100
+    },
+    {
+      "epoch": 1.7719298245614035,
+      "grad_norm": 0.5534357225495664,
+      "learning_rate": 7.747544890354031e-06,
+      "loss": 0.8004,
+      "step": 101
+    },
+    {
+      "epoch": 1.7894736842105263,
+      "grad_norm": 0.4372744352751169,
+      "learning_rate": 7.698753289757565e-06,
+      "loss": 0.7979,
+      "step": 102
+    },
+    {
+      "epoch": 1.807017543859649,
+      "grad_norm": 0.5308017997290465,
+      "learning_rate": 7.649596321166024e-06,
+      "loss": 0.8179,
+      "step": 103
+    },
+    {
+      "epoch": 1.8245614035087718,
+      "grad_norm": 0.4838905935160564,
+      "learning_rate": 7.600080639646077e-06,
+      "loss": 0.6699,
+      "step": 104
+    },
+    {
+      "epoch": 1.8421052631578947,
+      "grad_norm": 0.4911437146733649,
+      "learning_rate": 7.550212948828377e-06,
+      "loss": 0.6439,
+      "step": 105
+    },
+    {
+      "epoch": 1.8596491228070176,
+      "grad_norm": 0.6771983179775685,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 0.8343,
+      "step": 106
+    },
+    {
+      "epoch": 1.8771929824561404,
+      "grad_norm": 0.5801932125189808,
+      "learning_rate": 7.449448591190436e-06,
+      "loss": 0.7647,
+      "step": 107
+    },
+    {
+      "epoch": 1.8947368421052633,
+      "grad_norm": 0.4957701647602047,
+      "learning_rate": 7.398565566251232e-06,
+      "loss": 0.786,
+      "step": 108
+    },
+    {
+      "epoch": 1.912280701754386,
+      "grad_norm": 0.5454933024553883,
+      "learning_rate": 7.347357813929455e-06,
+      "loss": 0.7003,
+      "step": 109
+    },
+    {
+      "epoch": 1.9298245614035088,
+      "grad_norm": 0.5365025981976435,
+      "learning_rate": 7.295832266935059e-06,
+      "loss": 0.8062,
+      "step": 110
+    },
+    {
+      "epoch": 1.9473684210526314,
+      "grad_norm": 0.4783223450378368,
+      "learning_rate": 7.243995901002312e-06,
+      "loss": 0.8047,
+      "step": 111
+    },
+    {
+      "epoch": 1.9649122807017543,
+      "grad_norm": 0.5409575330711582,
+      "learning_rate": 7.191855733945388e-06,
+      "loss": 0.7754,
+      "step": 112
+    },
+    {
+      "epoch": 1.9824561403508771,
+      "grad_norm": 0.5248042146794984,
+      "learning_rate": 7.1394188247082715e-06,
+      "loss": 0.8502,
+      "step": 113
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6120193413387582,
+      "learning_rate": 7.08669227240909e-06,
+      "loss": 0.7556,
+      "step": 114
+    },
+    {
+      "epoch": 2.017543859649123,
+      "grad_norm": 0.7361501613223079,
+      "learning_rate": 7.033683215379002e-06,
+      "loss": 0.6339,
+      "step": 115
+    },
+    {
+      "epoch": 2.0350877192982457,
+      "grad_norm": 0.48612283438006465,
+      "learning_rate": 6.980398830195785e-06,
+      "loss": 0.6751,
+      "step": 116
+    },
+    {
+      "epoch": 2.0526315789473686,
+      "grad_norm": 0.5765386286901503,
+      "learning_rate": 6.9268463307122425e-06,
+      "loss": 0.8533,
+      "step": 117
+    },
+    {
+      "epoch": 2.0701754385964914,
+      "grad_norm": 0.9475302064536593,
+      "learning_rate": 6.873032967079562e-06,
+      "loss": 0.7152,
+      "step": 118
+    },
+    {
+      "epoch": 2.087719298245614,
+      "grad_norm": 0.6816783523524951,
+      "learning_rate": 6.818966024765758e-06,
+      "loss": 0.695,
+      "step": 119
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 0.5184803763146616,
+      "learning_rate": 6.7646528235693445e-06,
+      "loss": 0.6766,
+      "step": 120
+    },
+    {
+      "epoch": 2.1228070175438596,
+      "grad_norm": 0.6330699863361179,
+      "learning_rate": 6.710100716628345e-06,
+      "loss": 0.6686,
+      "step": 121
+    },
+    {
+      "epoch": 2.1403508771929824,
+      "grad_norm": 0.7947009956832388,
+      "learning_rate": 6.655317089424791e-06,
+      "loss": 0.6232,
+      "step": 122
+    },
+    {
+      "epoch": 2.1578947368421053,
+      "grad_norm": 0.5853439741927821,
+      "learning_rate": 6.600309358784858e-06,
+      "loss": 0.7714,
+      "step": 123
+    },
+    {
+      "epoch": 2.175438596491228,
+      "grad_norm": 0.5106179636862447,
+      "learning_rate": 6.545084971874738e-06,
+      "loss": 0.6707,
+      "step": 124
+    },
+    {
+      "epoch": 2.192982456140351,
+      "grad_norm": 0.7480135311239193,
+      "learning_rate": 6.48965140519241e-06,
+      "loss": 0.7725,
+      "step": 125
+    },
+    {
+      "epoch": 2.2105263157894735,
+      "grad_norm": 0.6928475269900484,
+      "learning_rate": 6.434016163555452e-06,
+      "loss": 0.6781,
+      "step": 126
+    },
+    {
+      "epoch": 2.2280701754385963,
+      "grad_norm": 0.5490952113614971,
+      "learning_rate": 6.378186779084996e-06,
+      "loss": 0.6257,
+      "step": 127
+    },
+    {
+      "epoch": 2.245614035087719,
+      "grad_norm": 0.502579039315677,
+      "learning_rate": 6.322170810186013e-06,
+      "loss": 0.6496,
+      "step": 128
+    },
+    {
+      "epoch": 2.263157894736842,
+      "grad_norm": 0.6003478292628378,
+      "learning_rate": 6.26597584052401e-06,
+      "loss": 0.6962,
+      "step": 129
+    },
+    {
+      "epoch": 2.280701754385965,
+      "grad_norm": 0.6601767127278643,
+      "learning_rate": 6.209609477998339e-06,
+      "loss": 0.708,
+      "step": 130
+    },
+    {
+      "epoch": 2.2982456140350878,
+      "grad_norm": 0.622000203409357,
+      "learning_rate": 6.153079353712201e-06,
+      "loss": 0.6878,
+      "step": 131
+    },
+    {
+      "epoch": 2.3157894736842106,
+      "grad_norm": 0.48240577345877544,
+      "learning_rate": 6.0963931209395165e-06,
+      "loss": 0.6491,
+      "step": 132
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.9911130443752747,
+      "learning_rate": 6.039558454088796e-06,
+      "loss": 0.7466,
+      "step": 133
+    },
+    {
+      "epoch": 2.3508771929824563,
+      "grad_norm": 0.6134313110673182,
+      "learning_rate": 5.982583047664151e-06,
+      "loss": 0.6581,
+      "step": 134
+    },
+    {
+      "epoch": 2.3684210526315788,
+      "grad_norm": 0.5552942229576594,
+      "learning_rate": 5.925474615223573e-06,
+      "loss": 0.6779,
+      "step": 135
+    },
+    {
+      "epoch": 2.3859649122807016,
+      "grad_norm": 0.47921474129820735,
+      "learning_rate": 5.8682408883346535e-06,
+      "loss": 0.6726,
+      "step": 136
+    },
+    {
+      "epoch": 2.4035087719298245,
+      "grad_norm": 0.5975125549416256,
+      "learning_rate": 5.810889615527839e-06,
+      "loss": 0.7775,
+      "step": 137
+    },
+    {
+      "epoch": 2.4210526315789473,
+      "grad_norm": 0.5709618574106423,
+      "learning_rate": 5.753428561247416e-06,
+      "loss": 0.764,
+      "step": 138
+    },
+    {
+      "epoch": 2.43859649122807,
+      "grad_norm": 0.5574328468376977,
+      "learning_rate": 5.695865504800328e-06,
+      "loss": 0.7204,
+      "step": 139
+    },
+    {
+      "epoch": 2.456140350877193,
+      "grad_norm": 0.46286312725472867,
+      "learning_rate": 5.638208239302975e-06,
+      "loss": 0.678,
+      "step": 140
+    },
+    {
+      "epoch": 2.473684210526316,
+      "grad_norm": 0.5615944253216454,
+      "learning_rate": 5.5804645706261515e-06,
+      "loss": 0.7242,
+      "step": 141
+    },
+    {
+      "epoch": 2.4912280701754383,
+      "grad_norm": 0.9894641600501489,
+      "learning_rate": 5.522642316338268e-06,
+      "loss": 0.729,
+      "step": 142
+    },
+    {
+      "epoch": 2.5087719298245617,
+      "grad_norm": 0.5603266778988845,
+      "learning_rate": 5.464749304646963e-06,
+      "loss": 0.5743,
+      "step": 143
+    },
+    {
+      "epoch": 2.526315789473684,
+      "grad_norm": 0.47275353321816904,
+      "learning_rate": 5.406793373339292e-06,
+      "loss": 0.6641,
+      "step": 144
+    },
+    {
+      "epoch": 2.543859649122807,
+      "grad_norm": 0.5313523615891593,
+      "learning_rate": 5.348782368720627e-06,
+      "loss": 0.6757,
+      "step": 145
+    },
+    {
+      "epoch": 2.56140350877193,
+      "grad_norm": 0.6501600310164274,
+      "learning_rate": 5.290724144552379e-06,
+      "loss": 0.7383,
+      "step": 146
+    },
+    {
+      "epoch": 2.5789473684210527,
+      "grad_norm": 0.46564036792105784,
+      "learning_rate": 5.232626560988735e-06,
+      "loss": 0.6764,
+      "step": 147
+    },
+    {
+      "epoch": 2.5964912280701755,
+      "grad_norm": 0.4857631471866125,
+      "learning_rate": 5.174497483512506e-06,
+      "loss": 0.6983,
+      "step": 148
+    },
+    {
+      "epoch": 2.6140350877192984,
+      "grad_norm": 0.5750741038794321,
+      "learning_rate": 5.116344781870282e-06,
+      "loss": 0.6706,
+      "step": 149
+    },
+    {
+      "epoch": 2.6315789473684212,
+      "grad_norm": 1.6297882197304314,
+      "learning_rate": 5.0581763290069865e-06,
+      "loss": 0.7617,
+      "step": 150
+    },
+    {
+      "epoch": 2.6491228070175437,
+      "grad_norm": 0.4420256741935745,
+      "learning_rate": 5e-06,
+      "loss": 0.7365,
+      "step": 151
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.5012952805138378,
+      "learning_rate": 4.941823670993016e-06,
+      "loss": 0.7342,
+      "step": 152
+    },
+    {
+      "epoch": 2.6842105263157894,
+      "grad_norm": 0.6822847257044459,
+      "learning_rate": 4.883655218129719e-06,
+      "loss": 0.7296,
+      "step": 153
+    },
+    {
+      "epoch": 2.7017543859649122,
+      "grad_norm": 0.5496663789125772,
+      "learning_rate": 4.825502516487497e-06,
+      "loss": 0.725,
+      "step": 154
+    },
+    {
+      "epoch": 2.719298245614035,
+      "grad_norm": 0.5454870345489141,
+      "learning_rate": 4.767373439011267e-06,
+      "loss": 0.7378,
+      "step": 155
+    },
+    {
+      "epoch": 2.736842105263158,
+      "grad_norm": 0.44579908141785257,
+      "learning_rate": 4.7092758554476215e-06,
+      "loss": 0.6801,
+      "step": 156
+    },
+    {
+      "epoch": 2.754385964912281,
+      "grad_norm": 0.757828243425537,
+      "learning_rate": 4.651217631279374e-06,
+      "loss": 0.6939,
+      "step": 157
+    },
+    {
+      "epoch": 2.7719298245614032,
+      "grad_norm": 0.48905049348260005,
+      "learning_rate": 4.59320662666071e-06,
+      "loss": 0.6066,
+      "step": 158
+    },
+    {
+      "epoch": 2.7894736842105265,
+      "grad_norm": 0.4532664219320891,
+      "learning_rate": 4.53525069535304e-06,
+      "loss": 0.671,
+      "step": 159
+    },
+    {
+      "epoch": 2.807017543859649,
+      "grad_norm": 0.5530789682100729,
+      "learning_rate": 4.477357683661734e-06,
+      "loss": 0.5758,
+      "step": 160
+    },
+    {
+      "epoch": 2.824561403508772,
+      "grad_norm": 0.5711848277941134,
+      "learning_rate": 4.4195354293738484e-06,
+      "loss": 0.6817,
+      "step": 161
+    },
+    {
+      "epoch": 2.8421052631578947,
+      "grad_norm": 0.6208551083175806,
+      "learning_rate": 4.361791760697027e-06,
+      "loss": 0.7113,
+      "step": 162
+    },
+    {
+      "epoch": 2.8596491228070176,
+      "grad_norm": 0.511755626688625,
+      "learning_rate": 4.304134495199675e-06,
+      "loss": 0.7203,
+      "step": 163
+    },
+    {
+      "epoch": 2.8771929824561404,
+      "grad_norm": 0.46452370179196634,
+      "learning_rate": 4.246571438752585e-06,
+      "loss": 0.6899,
+      "step": 164
+    },
+    {
+      "epoch": 2.8947368421052633,
+      "grad_norm": 0.5480920925753996,
+      "learning_rate": 4.189110384472164e-06,
+      "loss": 0.7566,
+      "step": 165
+    },
+    {
+      "epoch": 2.912280701754386,
+      "grad_norm": 0.4190056151730369,
+      "learning_rate": 4.131759111665349e-06,
+      "loss": 0.6886,
+      "step": 166
+    },
+    {
+      "epoch": 2.9298245614035086,
+      "grad_norm": 0.536995320205688,
+      "learning_rate": 4.074525384776428e-06,
+      "loss": 0.6293,
+      "step": 167
+    },
+    {
+      "epoch": 2.9473684210526314,
+      "grad_norm": 0.4726163557044103,
+      "learning_rate": 4.017416952335849e-06,
+      "loss": 0.671,
+      "step": 168
+    },
+    {
+      "epoch": 2.9649122807017543,
+      "grad_norm": 0.44445215399973664,
+      "learning_rate": 3.960441545911205e-06,
+      "loss": 0.6999,
+      "step": 169
+    },
+    {
+      "epoch": 2.982456140350877,
+      "grad_norm": 0.48733313855001437,
+      "learning_rate": 3.903606879060483e-06,
+      "loss": 0.7132,
+      "step": 170
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.5797155302178888,
+      "learning_rate": 3.8469206462878e-06,
+      "loss": 0.5377,
+      "step": 171
+    },
+    {
+      "epoch": 3.017543859649123,
+      "grad_norm": 0.951443811222221,
+      "learning_rate": 3.790390522001662e-06,
+      "loss": 0.567,
+      "step": 172
+    },
+    {
+      "epoch": 3.0350877192982457,
+      "grad_norm": 0.45525450271547346,
+      "learning_rate": 3.7340241594759917e-06,
+      "loss": 0.6849,
+      "step": 173
+    },
+    {
+      "epoch": 3.0526315789473686,
+      "grad_norm": 0.8177901081675041,
+      "learning_rate": 3.6778291898139907e-06,
+      "loss": 0.7146,
+      "step": 174
+    },
+    {
+      "epoch": 3.0701754385964914,
+      "grad_norm": 0.9928390665504743,
+      "learning_rate": 3.6218132209150047e-06,
+      "loss": 0.7018,
+      "step": 175
+    },
+    {
+      "epoch": 3.087719298245614,
+      "grad_norm": 0.7228688831907634,
+      "learning_rate": 3.5659838364445505e-06,
+      "loss": 0.624,
+      "step": 176
+    },
+    {
+      "epoch": 3.1052631578947367,
+      "grad_norm": 0.6174779546831684,
+      "learning_rate": 3.51034859480759e-06,
+      "loss": 0.7395,
+      "step": 177
+    },
+    {
+      "epoch": 3.1228070175438596,
+      "grad_norm": 0.6226996841281692,
+      "learning_rate": 3.4549150281252635e-06,
+      "loss": 0.7003,
+      "step": 178
+    },
+    {
+      "epoch": 3.1403508771929824,
+      "grad_norm": 0.5887562473638587,
+      "learning_rate": 3.399690641215142e-06,
+      "loss": 0.6011,
+      "step": 179
+    },
+    {
+      "epoch": 3.1578947368421053,
+      "grad_norm": 0.728834446750838,
+      "learning_rate": 3.3446829105752103e-06,
+      "loss": 0.6046,
+      "step": 180
+    },
+    {
+      "epoch": 3.175438596491228,
+      "grad_norm": 0.5275311311641863,
+      "learning_rate": 3.289899283371657e-06,
+      "loss": 0.6471,
+      "step": 181
+    },
+    {
+      "epoch": 3.192982456140351,
+      "grad_norm": 0.5089171902455958,
+      "learning_rate": 3.2353471764306567e-06,
+      "loss": 0.7815,
+      "step": 182
+    },
+    {
+      "epoch": 3.2105263157894735,
+      "grad_norm": 0.537647180699857,
+      "learning_rate": 3.1810339752342446e-06,
+      "loss": 0.6874,
+      "step": 183
+    },
+    {
+      "epoch": 3.2280701754385963,
+      "grad_norm": 0.6647720242563314,
+      "learning_rate": 3.12696703292044e-06,
+      "loss": 0.6274,
+      "step": 184
+    },
+    {
+      "epoch": 3.245614035087719,
+      "grad_norm": 0.44182806117914913,
+      "learning_rate": 3.0731536692877596e-06,
+      "loss": 0.6159,
+      "step": 185
+    },
+    {
+      "epoch": 3.263157894736842,
+      "grad_norm": 0.4697771931849346,
+      "learning_rate": 3.019601169804216e-06,
+      "loss": 0.5482,
+      "step": 186
+    },
+    {
+      "epoch": 3.280701754385965,
+      "grad_norm": 0.5057846695657313,
+      "learning_rate": 2.966316784621e-06,
+      "loss": 0.6389,
+      "step": 187
+    },
+    {
+      "epoch": 3.2982456140350878,
+      "grad_norm": 0.4981217521813649,
+      "learning_rate": 2.9133077275909112e-06,
+      "loss": 0.6984,
+      "step": 188
+    },
+    {
+      "epoch": 3.3157894736842106,
+      "grad_norm": 0.6032222024600464,
+      "learning_rate": 2.86058117529173e-06,
+      "loss": 0.6099,
+      "step": 189
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.5359461718523733,
+      "learning_rate": 2.8081442660546126e-06,
+      "loss": 0.6224,
+      "step": 190
+    },
+    {
+      "epoch": 3.3508771929824563,
+      "grad_norm": 0.6305910650525369,
+      "learning_rate": 2.7560040989976894e-06,
+      "loss": 0.6562,
+      "step": 191
+    },
+    {
+      "epoch": 3.3684210526315788,
+      "grad_norm": 0.4770030802073729,
+      "learning_rate": 2.7041677330649408e-06,
+      "loss": 0.5513,
+      "step": 192
+    },
+    {
+      "epoch": 3.3859649122807016,
+      "grad_norm": 0.4671558368322152,
+      "learning_rate": 2.6526421860705474e-06,
+      "loss": 0.5698,
+      "step": 193
+    },
+    {
+      "epoch": 3.4035087719298245,
+      "grad_norm": 0.5252716158458746,
+      "learning_rate": 2.601434433748771e-06,
+      "loss": 0.6031,
+      "step": 194
+    },
+    {
+      "epoch": 3.4210526315789473,
+      "grad_norm": 0.7889622234379826,
+      "learning_rate": 2.550551408809566e-06,
+      "loss": 0.5293,
+      "step": 195
+    },
+    {
+      "epoch": 3.43859649122807,
+      "grad_norm": 0.5428784098508198,
+      "learning_rate": 2.5000000000000015e-06,
+      "loss": 0.6531,
+      "step": 196
+    },
+    {
+      "epoch": 3.456140350877193,
+      "grad_norm": 0.651537815082855,
+      "learning_rate": 2.4497870511716237e-06,
+      "loss": 0.4875,
+      "step": 197
+    },
+    {
+      "epoch": 3.473684210526316,
+      "grad_norm": 0.5015067058140525,
+      "learning_rate": 2.3999193603539234e-06,
+      "loss": 0.6016,
+      "step": 198
+    },
+    {
+      "epoch": 3.4912280701754383,
+      "grad_norm": 0.7734024188313369,
+      "learning_rate": 2.3504036788339763e-06,
+      "loss": 0.6394,
+      "step": 199
+    },
+    {
+      "epoch": 3.5087719298245617,
+      "grad_norm": 0.42961016005530156,
+      "learning_rate": 2.3012467102424373e-06,
+      "loss": 0.5889,
+      "step": 200
+    },
+    {
+      "epoch": 3.526315789473684,
+      "grad_norm": 0.5053174047946184,
+      "learning_rate": 2.2524551096459703e-06,
+      "loss": 0.6452,
+      "step": 201
+    },
+    {
+      "epoch": 3.543859649122807,
+      "grad_norm": 0.5318977971324598,
+      "learning_rate": 2.204035482646267e-06,
+      "loss": 0.5222,
+      "step": 202
+    },
+    {
+      "epoch": 3.56140350877193,
+      "grad_norm": 0.44457932305265335,
+      "learning_rate": 2.155994384485742e-06,
+      "loss": 0.7199,
+      "step": 203
+    },
+    {
+      "epoch": 3.5789473684210527,
+      "grad_norm": 0.582870701262501,
+      "learning_rate": 2.1083383191600676e-06,
+      "loss": 0.5879,
+      "step": 204
+    },
+    {
+      "epoch": 3.5964912280701755,
+      "grad_norm": 0.6555427165212202,
+      "learning_rate": 2.061073738537635e-06,
+      "loss": 0.602,
+      "step": 205
+    },
+    {
+      "epoch": 3.6140350877192984,
+      "grad_norm": 0.43931124439831437,
+      "learning_rate": 2.0142070414860704e-06,
+      "loss": 0.5298,
+      "step": 206
+    },
+    {
+      "epoch": 3.6315789473684212,
+      "grad_norm": 0.7062973948469753,
+      "learning_rate": 1.9677445730059348e-06,
+      "loss": 0.547,
+      "step": 207
+    },
+    {
+      "epoch": 3.6491228070175437,
+      "grad_norm": 0.5893044725212433,
+      "learning_rate": 1.9216926233717087e-06,
+      "loss": 0.5784,
+      "step": 208
+    },
+    {
+      "epoch": 3.6666666666666665,
+      "grad_norm": 0.5974603326910514,
+      "learning_rate": 1.8760574272802002e-06,
+      "loss": 0.641,
+      "step": 209
+    },
+    {
+      "epoch": 3.6842105263157894,
+      "grad_norm": 0.5058416720418838,
+      "learning_rate": 1.8308451630064484e-06,
+      "loss": 0.5701,
+      "step": 210
+    },
+    {
+      "epoch": 3.7017543859649122,
+      "grad_norm": 0.5650829549605466,
+      "learning_rate": 1.7860619515673034e-06,
+      "loss": 0.6571,
+      "step": 211
+    },
+    {
+      "epoch": 3.719298245614035,
+      "grad_norm": 0.5469272063794115,
+      "learning_rate": 1.7417138558927244e-06,
+      "loss": 0.6276,
+      "step": 212
+    },
+    {
+      "epoch": 3.736842105263158,
+      "grad_norm": 0.4826931903471456,
+      "learning_rate": 1.6978068800049624e-06,
+      "loss": 0.682,
+      "step": 213
+    },
+    {
+      "epoch": 3.754385964912281,
+      "grad_norm": 0.4851146813863607,
+      "learning_rate": 1.6543469682057105e-06,
+      "loss": 0.6868,
+      "step": 214
+    },
+    {
+      "epoch": 3.7719298245614032,
+      "grad_norm": 0.49586663141817217,
+      "learning_rate": 1.611340004271339e-06,
+      "loss": 0.7239,
+      "step": 215
+    },
+    {
+      "epoch": 3.7894736842105265,
+      "grad_norm": 0.5015190881089485,
+      "learning_rate": 1.5687918106563326e-06,
+      "loss": 0.6226,
+      "step": 216
+    },
+    {
+      "epoch": 3.807017543859649,
+      "grad_norm": 0.5062327753778796,
+      "learning_rate": 1.5267081477050132e-06,
+      "loss": 0.6694,
+      "step": 217
+    },
+    {
+      "epoch": 3.824561403508772,
+      "grad_norm": 0.5008115199773022,
+      "learning_rate": 1.4850947128716914e-06,
+      "loss": 0.689,
+      "step": 218
+    },
+    {
+      "epoch": 3.8421052631578947,
+      "grad_norm": 0.5553537886745609,
+      "learning_rate": 1.4439571399493146e-06,
+      "loss": 0.6508,
+      "step": 219
+    },
+    {
+      "epoch": 3.8596491228070176,
+      "grad_norm": 0.4668069189007763,
+      "learning_rate": 1.4033009983067454e-06,
+      "loss": 0.5546,
+      "step": 220
+    },
+    {
+      "epoch": 3.8771929824561404,
+      "grad_norm": 0.4567357945927226,
+      "learning_rate": 1.3631317921347564e-06,
+      "loss": 0.644,
+      "step": 221
+    },
+    {
+      "epoch": 3.8947368421052633,
+      "grad_norm": 0.5290790455225494,
+      "learning_rate": 1.3234549597008572e-06,
+      "loss": 0.635,
+      "step": 222
+    },
+    {
+      "epoch": 3.912280701754386,
+      "grad_norm": 0.4166384021171554,
+      "learning_rate": 1.2842758726130283e-06,
+      "loss": 0.5688,
+      "step": 223
+    },
+    {
+      "epoch": 3.9298245614035086,
+      "grad_norm": 0.5892522945424487,
+      "learning_rate": 1.2455998350925042e-06,
+      "loss": 0.5812,
+      "step": 224
+    },
+    {
+      "epoch": 3.9473684210526314,
+      "grad_norm": 0.5598685146911576,
+      "learning_rate": 1.2074320832556558e-06,
+      "loss": 0.6853,
+      "step": 225
+    },
+    {
+      "epoch": 3.9649122807017543,
+      "grad_norm": 0.7169986788709015,
+      "learning_rate": 1.1697777844051105e-06,
+      "loss": 0.6564,
+      "step": 226
+    },
+    {
+      "epoch": 3.982456140350877,
+      "grad_norm": 0.5722383429808436,
+      "learning_rate": 1.132642036330181e-06,
+      "loss": 0.6786,
+      "step": 227
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.48666587302012654,
+      "learning_rate": 1.096029866616704e-06,
+      "loss": 0.5785,
+      "step": 228
+    },
+    {
+      "epoch": 4.017543859649122,
+      "grad_norm": 0.7435784129153119,
+      "learning_rate": 1.0599462319663906e-06,
+      "loss": 0.6433,
+      "step": 229
+    },
+    {
+      "epoch": 4.035087719298246,
+      "grad_norm": 1.0356006627086014,
+      "learning_rate": 1.0243960175257605e-06,
+      "loss": 0.5484,
+      "step": 230
+    },
+    {
+      "epoch": 4.052631578947368,
+      "grad_norm": 0.7627046458225869,
+      "learning_rate": 9.893840362247809e-07,
+      "loss": 0.5862,
+      "step": 231
+    },
+    {
+      "epoch": 4.0701754385964914,
+      "grad_norm": 0.6645890987296554,
+      "learning_rate": 9.549150281252633e-07,
+      "loss": 0.5672,
+      "step": 232
+    },
+    {
+      "epoch": 4.087719298245614,
+      "grad_norm": 0.4548556793537139,
+      "learning_rate": 9.209936597791407e-07,
+      "loss": 0.6211,
+      "step": 233
+    },
+    {
+      "epoch": 4.105263157894737,
+      "grad_norm": 0.5214713901344799,
+      "learning_rate": 8.876245235966884e-07,
+      "loss": 0.6172,
+      "step": 234
+    },
+    {
+      "epoch": 4.12280701754386,
+      "grad_norm": 0.43248317251943286,
+      "learning_rate": 8.54812137224792e-07,
+      "loss": 0.55,
+      "step": 235
+    },
+    {
+      "epoch": 4.140350877192983,
+      "grad_norm": 0.5603353559501666,
+      "learning_rate": 8.225609429353187e-07,
+      "loss": 0.6468,
+      "step": 236
+    },
+    {
+      "epoch": 4.157894736842105,
+      "grad_norm": 0.7517506161215499,
+      "learning_rate": 7.908753070237124e-07,
+      "loss": 0.5081,
+      "step": 237
+    },
+    {
+      "epoch": 4.175438596491228,
+      "grad_norm": 0.8092578214552126,
+      "learning_rate": 7.597595192178702e-07,
+      "loss": 0.5973,
+      "step": 238
+    },
+    {
+      "epoch": 4.192982456140351,
+      "grad_norm": 0.7279009942606227,
+      "learning_rate": 7.292177920973726e-07,
+      "loss": 0.5821,
+      "step": 239
+    },
+    {
+      "epoch": 4.2105263157894735,
+      "grad_norm": 0.7900698471233691,
+      "learning_rate": 6.992542605231739e-07,
+      "loss": 0.5058,
+      "step": 240
+    },
+    {
+      "epoch": 4.228070175438597,
+      "grad_norm": 0.7737353628103142,
+      "learning_rate": 6.698729810778065e-07,
+      "loss": 0.582,
+      "step": 241
+    },
+    {
+      "epoch": 4.245614035087719,
+      "grad_norm": 0.8108843822776718,
+      "learning_rate": 6.410779315161885e-07,
+      "loss": 0.6187,
+      "step": 242
+    },
+    {
+      "epoch": 4.2631578947368425,
+      "grad_norm": 0.5723814584171666,
+      "learning_rate": 6.128730102270897e-07,
+      "loss": 0.5287,
+      "step": 243
+    },
+    {
+      "epoch": 4.280701754385965,
+      "grad_norm": 0.5001459342659815,
+      "learning_rate": 5.852620357053651e-07,
+      "loss": 0.6459,
+      "step": 244
+    },
+    {
+      "epoch": 4.298245614035087,
+      "grad_norm": 0.47155393221062974,
+      "learning_rate": 5.582487460349806e-07,
+      "loss": 0.6372,
+      "step": 245
+    },
+    {
+      "epoch": 4.315789473684211,
+      "grad_norm": 0.554418769345169,
+      "learning_rate": 5.318367983829393e-07,
+      "loss": 0.6895,
+      "step": 246
+    },
+    {
+      "epoch": 4.333333333333333,
+      "grad_norm": 0.5335078794513084,
+      "learning_rate": 5.06029768504166e-07,
+      "loss": 0.639,
+      "step": 247
+    },
+    {
+      "epoch": 4.350877192982456,
+      "grad_norm": 0.5298439982382793,
+      "learning_rate": 4.808311502573976e-07,
+      "loss": 0.5899,
+      "step": 248
+    },
+    {
+      "epoch": 4.368421052631579,
+      "grad_norm": 0.47198058164725953,
+      "learning_rate": 4.562443551321788e-07,
+      "loss": 0.4881,
+      "step": 249
+    },
+    {
+      "epoch": 4.385964912280702,
+      "grad_norm": 0.4784036950372624,
+      "learning_rate": 4.322727117869951e-07,
+      "loss": 0.6719,
+      "step": 250
+    },
+    {
+      "epoch": 4.4035087719298245,
+      "grad_norm": 0.5589165084742584,
+      "learning_rate": 4.089194655986306e-07,
+      "loss": 0.6555,
+      "step": 251
+    },
+    {
+      "epoch": 4.421052631578947,
+      "grad_norm": 0.4812066133910625,
+      "learning_rate": 3.8618777822278854e-07,
+      "loss": 0.5409,
+      "step": 252
+    },
+    {
+      "epoch": 4.43859649122807,
+      "grad_norm": 0.4596965504525194,
+      "learning_rate": 3.6408072716606346e-07,
+      "loss": 0.5954,
+      "step": 253
+    },
+    {
+      "epoch": 4.456140350877193,
+      "grad_norm": 0.4681771152127897,
+      "learning_rate": 3.426013053692878e-07,
+      "loss": 0.6311,
+      "step": 254
+    },
+    {
+      "epoch": 4.473684210526316,
+      "grad_norm": 0.5199381160834282,
+      "learning_rate": 3.2175242080234314e-07,
+      "loss": 0.7112,
+      "step": 255
+    },
+    {
+      "epoch": 4.491228070175438,
+      "grad_norm": 0.5029053126087911,
+      "learning_rate": 3.015368960704584e-07,
+      "loss": 0.6537,
+      "step": 256
+    },
+    {
+      "epoch": 4.508771929824562,
+      "grad_norm": 0.5176529669307418,
+      "learning_rate": 2.819574680320825e-07,
+      "loss": 0.6114,
+      "step": 257
+    },
+    {
+      "epoch": 4.526315789473684,
+      "grad_norm": 0.4642569050924163,
+      "learning_rate": 2.63016787428354e-07,
+      "loss": 0.5526,
+      "step": 258
+    },
+    {
+      "epoch": 4.543859649122807,
+      "grad_norm": 0.44052565968207763,
+      "learning_rate": 2.447174185242324e-07,
+      "loss": 0.6841,
+      "step": 259
+    },
+    {
+      "epoch": 4.56140350877193,
+      "grad_norm": 0.7045609106541972,
+      "learning_rate": 2.2706183876134047e-07,
+      "loss": 0.476,
+      "step": 260
+    },
+    {
+      "epoch": 4.578947368421053,
+      "grad_norm": 0.4704145373326146,
+      "learning_rate": 2.1005243842255552e-07,
+      "loss": 0.5372,
+      "step": 261
+    },
+    {
+      "epoch": 4.5964912280701755,
+      "grad_norm": 0.4622901813653495,
+      "learning_rate": 1.9369152030840553e-07,
+      "loss": 0.5344,
+      "step": 262
+    },
+    {
+      "epoch": 4.614035087719298,
+      "grad_norm": 0.4194593586978816,
+      "learning_rate": 1.779812994253055e-07,
+      "loss": 0.4811,
+      "step": 263
+    },
+    {
+      "epoch": 4.631578947368421,
+      "grad_norm": 0.7771717715817167,
+      "learning_rate": 1.6292390268568103e-07,
+      "loss": 0.5521,
+      "step": 264
+    },
+    {
+      "epoch": 4.649122807017544,
+      "grad_norm": 0.4356364649272727,
+      "learning_rate": 1.4852136862001766e-07,
+      "loss": 0.6213,
+      "step": 265
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.5786691196246273,
+      "learning_rate": 1.3477564710088097e-07,
+      "loss": 0.5663,
+      "step": 266
+    },
+    {
+      "epoch": 4.684210526315789,
+      "grad_norm": 0.5349279026451833,
+      "learning_rate": 1.2168859907892904e-07,
+      "loss": 0.5871,
+      "step": 267
+    },
+    {
+      "epoch": 4.701754385964913,
+      "grad_norm": 0.5281858345459043,
+      "learning_rate": 1.0926199633097156e-07,
+      "loss": 0.4862,
+      "step": 268
+    },
+    {
+      "epoch": 4.719298245614035,
+      "grad_norm": 0.5252740732608391,
+      "learning_rate": 9.749752122010347e-08,
+      "loss": 0.6213,
+      "step": 269
+    },
+    {
+      "epoch": 4.7368421052631575,
+      "grad_norm": 0.46986416996786,
+      "learning_rate": 8.639676646793382e-08,
+      "loss": 0.586,
+      "step": 270
+    },
+    {
+      "epoch": 4.754385964912281,
+      "grad_norm": 0.4489284779100406,
+      "learning_rate": 7.59612349389599e-08,
+      "loss": 0.6004,
+      "step": 271
+    },
+    {
+      "epoch": 4.771929824561403,
+      "grad_norm": 0.4336128174067588,
+      "learning_rate": 6.61923394371039e-08,
+      "loss": 0.7142,
+      "step": 272
+    },
+    {
+      "epoch": 4.7894736842105265,
+      "grad_norm": 0.8819435935998887,
+      "learning_rate": 5.709140251444201e-08,
+      "loss": 0.6063,
+      "step": 273
+    },
+    {
+      "epoch": 4.807017543859649,
+      "grad_norm": 0.5222963070085169,
+      "learning_rate": 4.865965629214819e-08,
+      "loss": 0.5778,
+      "step": 274
+    },
+    {
+      "epoch": 4.824561403508772,
+      "grad_norm": 0.44526342094736737,
+      "learning_rate": 4.0898242293691546e-08,
+      "loss": 0.5451,
+      "step": 275
+    },
+    {
+      "epoch": 4.842105263157895,
+      "grad_norm": 0.4587835893789319,
+      "learning_rate": 3.3808211290284886e-08,
+      "loss": 0.5918,
+      "step": 276
+    },
+    {
+      "epoch": 4.859649122807017,
+      "grad_norm": 0.5280122551661669,
+      "learning_rate": 2.7390523158633552e-08,
+      "loss": 0.6036,
+      "step": 277
+    },
+    {
+      "epoch": 4.87719298245614,
+      "grad_norm": 0.40310412351316227,
+      "learning_rate": 2.1646046750978255e-08,
+      "loss": 0.6156,
+      "step": 278
+    },
+    {
+      "epoch": 4.894736842105263,
+      "grad_norm": 0.4140303452886482,
+      "learning_rate": 1.657555977746972e-08,
+      "loss": 0.5783,
+      "step": 279
+    },
+    {
+      "epoch": 4.912280701754386,
+      "grad_norm": 0.5264167472323803,
+      "learning_rate": 1.2179748700879013e-08,
+      "loss": 0.6176,
+      "step": 280
+    },
+    {
+      "epoch": 4.9298245614035086,
+      "grad_norm": 0.5865644351649915,
+      "learning_rate": 8.459208643659122e-09,
+      "loss": 0.6039,
+      "step": 281
+    },
+    {
+      "epoch": 4.947368421052632,
+      "grad_norm": 0.6854618019142784,
+      "learning_rate": 5.414443307377171e-09,
+      "loss": 0.7034,
+      "step": 282
+    },
+    {
+      "epoch": 4.964912280701754,
+      "grad_norm": 0.4109875295185457,
+      "learning_rate": 3.0458649045211897e-09,
+      "loss": 0.5541,
+      "step": 283
+    },
+    {
+      "epoch": 4.982456140350877,
+      "grad_norm": 0.4211169634724555,
+      "learning_rate": 1.3537941026914302e-09,
+      "loss": 0.6294,
+      "step": 284
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.4347156685717635,
+      "learning_rate": 3.384599811889766e-10,
+      "loss": 0.6349,
+      "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 sha256:500a0c487f3ed048130f19df6cf26d355285f45b7f94f5d1f95a37b6bd9ace10
+size 8081