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

.ipynb_checkpoints/README-checkpoint.md

@@ -0,0 +1,71 @@
+---
+library_name: transformers
+license: other
+base_model: Qwen3.5-9B
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+model-index:
+- name: qwen35_caption_galore
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# qwen35_caption_galore
+
+This model is a fine-tuned version of [/workspace/models/Qwen3.5-9B](https://huggingface.co//workspace/models/Qwen3.5-9B) on the my_caption dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+same as 9b-41, freeze lm_head + embed, every other step freeze bottom 16 lang
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- family_to_muon_lr = {
+        "language": _fallback(getattr(training_args, "language_muon_lr", 3e-5), language_lr),
+        "vision": _fallback(getattr(training_args, "vision_muon_lr", 3e-5), vision_lr),
+        "merger": _fallback(getattr(training_args, "merger_muon_lr", 3e-5), merger_lr),
+    }
+
+    family_to_adamw_lr = {
+        "language": _fallback(getattr(training_args, "language_adamw_lr", 3e-5), language_lr),
+        "vision": _fallback(getattr(training_args, "vision_adamw_lr", 3e-6), vision_lr),
+        "merger": _fallback(getattr(training_args, "merger_adamw_lr", 3e-5), merger_lr),
+    }
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- gradient_accumulation_steps: 32
+- total_train_batch_size: 64
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine_with_min_lr
+- lr_scheduler_warmup_steps: 0.05
+- num_epochs: 3
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 5.5.3
+- Pytorch 2.11.0+cu130
+- Datasets 4.0.0
+- Tokenizers 0.22.2

README.md

@@ -0,0 +1,71 @@
+---
+library_name: transformers
+license: other
+base_model: Qwen3.5-9B
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+model-index:
+- name: qwen35_caption_galore
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# qwen35_caption_galore
+
+This model is a fine-tuned version of [/workspace/models/Qwen3.5-9B](https://huggingface.co//workspace/models/Qwen3.5-9B) on the my_caption dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+same as 9b-41, freeze lm_head + embed, every other step freeze bottom 16 lang
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- family_to_muon_lr = {
+        "language": _fallback(getattr(training_args, "language_muon_lr", 3e-5), language_lr),
+        "vision": _fallback(getattr(training_args, "vision_muon_lr", 3e-5), vision_lr),
+        "merger": _fallback(getattr(training_args, "merger_muon_lr", 3e-5), merger_lr),
+    }
+
+    family_to_adamw_lr = {
+        "language": _fallback(getattr(training_args, "language_adamw_lr", 3e-5), language_lr),
+        "vision": _fallback(getattr(training_args, "vision_adamw_lr", 3e-6), vision_lr),
+        "merger": _fallback(getattr(training_args, "merger_adamw_lr", 3e-5), merger_lr),
+    }
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- gradient_accumulation_steps: 32
+- total_train_batch_size: 64
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine_with_min_lr
+- lr_scheduler_warmup_steps: 0.05
+- num_epochs: 3
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 5.5.3
+- Pytorch 2.11.0+cu130
+- Datasets 4.0.0
+- Tokenizers 0.22.2

all_results.json

@@ -0,0 +1,9 @@
+{
+    "effective_tokens_per_sec": 6380.59331163488,
+    "epoch": 3.0,
+    "total_flos": 3.0241141054567875e+18,
+    "train_loss": 0.8476084547685393,
+    "train_runtime": 9225.3576,
+    "train_samples_per_second": 4.891,
+    "train_steps_per_second": 0.076
+}

chat_template.jinja

@@ -0,0 +1,154 @@
+{%- set image_count = namespace(value=0) %}
+{%- set video_count = namespace(value=0) %}
+{%- macro render_content(content, do_vision_count, is_system_content=false) %}
+    {%- if content is string %}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping %}
+        {%- for item in content %}
+            {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain images.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set image_count.value = image_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Picture ' ~ image_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
+            {%- elif 'video' in item or item.type == 'video' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain videos.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set video_count.value = video_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Video ' ~ video_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
+            {%- elif 'text' in item %}
+                {{- item.text }}
+            {%- else %}
+                {{- raise_exception('Unexpected item type in content.') }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif content is none or content is undefined %}
+        {{- '' }}
+    {%- else %}
+        {{- raise_exception('Unexpected content type.') }}
+    {%- endif %}
+{%- endmacro %}
+{%- if not messages %}
+    {{- raise_exception('No messages provided.') }}
+{%- endif %}
+{%- if tools and tools is iterable and tools is not mapping %}
+    {{- '<|im_start|>system\n' }}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>" }}
+    {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '\n\n' + content }}
+        {%- endif %}
+    {%- endif %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" %}
+        {%- set content = render_content(message.content, false)|trim %}
+        {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
+            {%- set ns.multi_step_tool = false %}
+            {%- set ns.last_query_index = index %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if ns.multi_step_tool %}
+    {{- raise_exception('No user query found in messages.') }}
+{%- endif %}
+{%- for message in messages %}
+    {%- set content = render_content(message.content, true)|trim %}
+    {%- if message.role == "system" %}
+        {%- if not loop.first %}
+            {{- raise_exception('System message must be at the beginning.') }}
+        {%- endif %}
+    {%- elif message.role == "user" %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- set reasoning_content = reasoning_content|trim %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if tool_call.function is defined %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {%- if loop.first %}
+                    {%- if content|trim %}
+                        {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- else %}
+                        {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- endif %}
+                {%- else %}
+                    {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                {%- endif %}
+                {%- if tool_call.arguments is defined %}
+                    {%- for args_name, args_value in tool_call.arguments|items %}
+                        {{- '<parameter=' + args_name + '>\n' }}
+                        {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
+                        {{- args_value }}
+                        {{- '\n</parameter>\n' }}
+                    {%- endfor %}
+                {%- endif %}
+                {{- '</function>\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- raise_exception('Unexpected message role.') }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- else %}
+        {{- '<think>\n' }}
+    {%- endif %}
+{%- endif %}

config.json

@@ -0,0 +1,113 @@
+{
+  "architectures": [
+    "Qwen3_5ForConditionalGeneration"
+  ],
+  "dtype": "float32",
+  "eos_token_id": 248046,
+  "hidden_size": 4096,
+  "image_token_id": 248056,
+  "model_type": "qwen3_5",
+  "pad_token_id": 248044,
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "attn_output_gate": true,
+    "bos_token_id": null,
+    "dtype": "bfloat16",
+    "eos_token_id": 248044,
+    "full_attention_interval": 4,
+    "head_dim": 256,
+    "hidden_act": "silu",
+    "hidden_size": 4096,
+    "initializer_range": 0.02,
+    "intermediate_size": 12288,
+    "layer_types": [
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention"
+    ],
+    "linear_conv_kernel_dim": 4,
+    "linear_key_head_dim": 128,
+    "linear_num_key_heads": 16,
+    "linear_num_value_heads": 32,
+    "linear_value_head_dim": 128,
+    "mamba_ssm_dtype": "float32",
+    "max_position_embeddings": 262144,
+    "mlp_only_layers": [],
+    "model_type": "qwen3_5_text",
+    "mtp_num_hidden_layers": 1,
+    "mtp_use_dedicated_embeddings": false,
+    "num_attention_heads": 16,
+    "num_hidden_layers": 32,
+    "num_key_value_heads": 4,
+    "pad_token_id": null,
+    "partial_rotary_factor": 0.25,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "mrope_interleaved": true,
+      "mrope_section": [
+        11,
+        11,
+        10
+      ],
+      "partial_rotary_factor": 0.25,
+      "rope_theta": 10000000,
+      "rope_type": "default"
+    },
+    "tie_word_embeddings": false,
+    "use_cache": false,
+    "vocab_size": 248320
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.5.3",
+  "use_cache": false,
+  "video_token_id": 248057,
+  "vision_config": {
+    "deepstack_visual_indexes": [],
+    "depth": 27,
+    "dtype": "bfloat16",
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 1152,
+    "in_channels": 3,
+    "initializer_range": 0.02,
+    "intermediate_size": 4304,
+    "model_type": "qwen3_5",
+    "num_heads": 16,
+    "num_position_embeddings": 2304,
+    "out_hidden_size": 4096,
+    "patch_size": 16,
+    "spatial_merge_size": 2,
+    "temporal_patch_size": 2
+  },
+  "vision_end_token_id": 248054,
+  "vision_start_token_id": 248053
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": [
+    248046,
+    248044
+  ],
+  "pad_token_id": 248044,
+  "transformers_version": "5.5.3",
+  "use_cache": true
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81af939cfb6edf437ba9b7942cd032821373a045d0b2db72efed98c741c6d753
+size 37639354416

processor_config.json

@@ -0,0 +1,60 @@
+{
+  "image_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "Qwen2VLImageProcessor",
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "merge_size": 2,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "longest_edge": 16777216,
+      "shortest_edge": 65536
+    },
+    "temporal_patch_size": 2
+  },
+  "processor_class": "Qwen3VLProcessor",
+  "video_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "do_sample_frames": true,
+    "fps": 2,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "max_frames": 768,
+    "merge_size": 2,
+    "min_frames": 4,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "return_metadata": false,
+    "size": {
+      "longest_edge": 25165824,
+      "shortest_edge": 4096
+    },
+    "temporal_patch_size": 2,
+    "video_processor_type": "Qwen3VLVideoProcessor"
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4
+size 19989343

tokenizer_config.json

@@ -0,0 +1,33 @@
+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "is_local": true,
+  "model_max_length": 262144,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
+  "processor_class": "Qwen3VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "effective_tokens_per_sec": 6380.59331163488,
+    "epoch": 3.0,
+    "total_flos": 3.0241141054567875e+18,
+    "train_loss": 0.8476084547685393,
+    "train_runtime": 9225.3576,
+    "train_samples_per_second": 4.891,
+    "train_steps_per_second": 0.076
+}

trainer_log.jsonl

@@ -0,0 +1,353 @@
+{"current_steps": 2, "total_steps": 705, "loss": 4.115976333618164, "lr": 8.333333333333333e-07, "epoch": 0.00851063829787234, "percentage": 0.28, "elapsed_time": "0:00:22", "remaining_time": "2:10:45"}
+{"current_steps": 4, "total_steps": 705, "loss": 1.878818154335022, "lr": 2.4999999999999998e-06, "epoch": 0.01702127659574468, "percentage": 0.57, "elapsed_time": "0:00:47", "remaining_time": "2:19:53"}
+{"current_steps": 6, "total_steps": 705, "loss": 2.0509164333343506, "lr": 4.166666666666667e-06, "epoch": 0.02553191489361702, "percentage": 0.85, "elapsed_time": "0:01:18", "remaining_time": "2:32:56"}
+{"current_steps": 8, "total_steps": 705, "loss": 1.7545194625854492, "lr": 5.833333333333334e-06, "epoch": 0.03404255319148936, "percentage": 1.13, "elapsed_time": "0:01:36", "remaining_time": "2:20:28"}
+{"current_steps": 10, "total_steps": 705, "loss": 1.7690430879592896, "lr": 7.5e-06, "epoch": 0.0425531914893617, "percentage": 1.42, "elapsed_time": "0:02:09", "remaining_time": "2:30:29"}
+{"current_steps": 12, "total_steps": 705, "loss": 2.355275869369507, "lr": 9.166666666666668e-06, "epoch": 0.05106382978723404, "percentage": 1.7, "elapsed_time": "0:02:37", "remaining_time": "2:31:22"}
+{"current_steps": 14, "total_steps": 705, "loss": 1.876306176185608, "lr": 1.0833333333333334e-05, "epoch": 0.059574468085106386, "percentage": 1.99, "elapsed_time": "0:03:05", "remaining_time": "2:32:21"}
+{"current_steps": 16, "total_steps": 705, "loss": 1.785620927810669, "lr": 1.25e-05, "epoch": 0.06808510638297872, "percentage": 2.27, "elapsed_time": "0:03:33", "remaining_time": "2:32:57"}
+{"current_steps": 18, "total_steps": 705, "loss": 1.523728370666504, "lr": 1.4166666666666666e-05, "epoch": 0.07659574468085106, "percentage": 2.55, "elapsed_time": "0:03:57", "remaining_time": "2:30:57"}
+{"current_steps": 20, "total_steps": 705, "loss": 1.510640025138855, "lr": 1.5833333333333333e-05, "epoch": 0.0851063829787234, "percentage": 2.84, "elapsed_time": "0:04:26", "remaining_time": "2:32:24"}
+{"current_steps": 22, "total_steps": 705, "loss": 1.4279940128326416, "lr": 1.7500000000000002e-05, "epoch": 0.09361702127659574, "percentage": 3.12, "elapsed_time": "0:05:01", "remaining_time": "2:36:06"}
+{"current_steps": 24, "total_steps": 705, "loss": 1.1794553995132446, "lr": 1.9166666666666667e-05, "epoch": 0.10212765957446808, "percentage": 3.4, "elapsed_time": "0:05:20", "remaining_time": "2:31:20"}
+{"current_steps": 26, "total_steps": 705, "loss": 1.2579714059829712, "lr": 2.0833333333333333e-05, "epoch": 0.11063829787234042, "percentage": 3.69, "elapsed_time": "0:05:42", "remaining_time": "2:28:59"}
+{"current_steps": 28, "total_steps": 705, "loss": 0.9510668516159058, "lr": 2.25e-05, "epoch": 0.11914893617021277, "percentage": 3.97, "elapsed_time": "0:06:02", "remaining_time": "2:26:15"}
+{"current_steps": 30, "total_steps": 705, "loss": 1.656806230545044, "lr": 2.4166666666666667e-05, "epoch": 0.1276595744680851, "percentage": 4.26, "elapsed_time": "0:06:31", "remaining_time": "2:26:45"}
+{"current_steps": 32, "total_steps": 705, "loss": 0.683425784111023, "lr": 2.5833333333333336e-05, "epoch": 0.13617021276595745, "percentage": 4.54, "elapsed_time": "0:06:49", "remaining_time": "2:23:27"}
+{"current_steps": 34, "total_steps": 705, "loss": 1.2381809949874878, "lr": 2.75e-05, "epoch": 0.14468085106382977, "percentage": 4.82, "elapsed_time": "0:07:18", "remaining_time": "2:24:22"}
+{"current_steps": 36, "total_steps": 705, "loss": 1.422176480293274, "lr": 2.9166666666666666e-05, "epoch": 0.15319148936170213, "percentage": 5.11, "elapsed_time": "0:07:51", "remaining_time": "2:26:10"}
+{"current_steps": 38, "total_steps": 705, "loss": 1.2621259689331055, "lr": 2.9998734788806287e-05, "epoch": 0.16170212765957448, "percentage": 5.39, "elapsed_time": "0:08:18", "remaining_time": "2:25:52"}
+{"current_steps": 40, "total_steps": 705, "loss": 1.2012425661087036, "lr": 2.9988614605803806e-05, "epoch": 0.1702127659574468, "percentage": 5.67, "elapsed_time": "0:08:46", "remaining_time": "2:25:53"}
+{"current_steps": 42, "total_steps": 705, "loss": 1.2804166078567505, "lr": 2.99683822733885e-05, "epoch": 0.17872340425531916, "percentage": 5.96, "elapsed_time": "0:09:13", "remaining_time": "2:25:40"}
+{"current_steps": 44, "total_steps": 705, "loss": 1.1188151836395264, "lr": 2.9938053852362484e-05, "epoch": 0.18723404255319148, "percentage": 6.24, "elapsed_time": "0:09:39", "remaining_time": "2:25:01"}
+{"current_steps": 46, "total_steps": 705, "loss": 1.338016152381897, "lr": 2.989765341799095e-05, "epoch": 0.19574468085106383, "percentage": 6.52, "elapsed_time": "0:10:09", "remaining_time": "2:25:26"}
+{"current_steps": 48, "total_steps": 705, "loss": 0.8224107027053833, "lr": 2.9847213040890793e-05, "epoch": 0.20425531914893616, "percentage": 6.81, "elapsed_time": "0:10:28", "remaining_time": "2:23:22"}
+{"current_steps": 50, "total_steps": 705, "loss": 0.9487060308456421, "lr": 2.9786772761572335e-05, "epoch": 0.2127659574468085, "percentage": 7.09, "elapsed_time": "0:10:41", "remaining_time": "2:20:04"}
+{"current_steps": 52, "total_steps": 705, "loss": 1.3093581199645996, "lr": 2.9716380558654445e-05, "epoch": 0.22127659574468084, "percentage": 7.38, "elapsed_time": "0:11:16", "remaining_time": "2:21:38"}
+{"current_steps": 54, "total_steps": 705, "loss": 1.2255353927612305, "lr": 2.9636092310778195e-05, "epoch": 0.2297872340425532, "percentage": 7.66, "elapsed_time": "0:11:45", "remaining_time": "2:21:49"}
+{"current_steps": 56, "total_steps": 705, "loss": 0.8253700137138367, "lr": 2.954597175224938e-05, "epoch": 0.23829787234042554, "percentage": 7.94, "elapsed_time": "0:12:07", "remaining_time": "2:20:28"}
+{"current_steps": 58, "total_steps": 705, "loss": 1.0494465827941895, "lr": 2.9446090422445016e-05, "epoch": 0.24680851063829787, "percentage": 8.23, "elapsed_time": "0:12:36", "remaining_time": "2:20:37"}
+{"current_steps": 60, "total_steps": 705, "loss": 1.358299970626831, "lr": 2.9336527609024072e-05, "epoch": 0.2553191489361702, "percentage": 8.51, "elapsed_time": "0:13:08", "remaining_time": "2:21:15"}
+{"current_steps": 62, "total_steps": 705, "loss": 1.335315465927124, "lr": 2.9217370284987434e-05, "epoch": 0.26382978723404255, "percentage": 8.79, "elapsed_time": "0:13:41", "remaining_time": "2:22:04"}
+{"current_steps": 64, "total_steps": 705, "loss": 1.2662416696548462, "lr": 2.9088713039637117e-05, "epoch": 0.2723404255319149, "percentage": 9.08, "elapsed_time": "0:14:14", "remaining_time": "2:22:41"}
+{"current_steps": 66, "total_steps": 705, "loss": 1.2614431381225586, "lr": 2.8950658003489534e-05, "epoch": 0.28085106382978725, "percentage": 9.36, "elapsed_time": "0:14:37", "remaining_time": "2:21:37"}
+{"current_steps": 68, "total_steps": 705, "loss": 1.2736550569534302, "lr": 2.880331476720238e-05, "epoch": 0.28936170212765955, "percentage": 9.65, "elapsed_time": "0:15:12", "remaining_time": "2:22:25"}
+{"current_steps": 70, "total_steps": 705, "loss": 1.2623916864395142, "lr": 2.8646800294579517e-05, "epoch": 0.2978723404255319, "percentage": 9.93, "elapsed_time": "0:15:34", "remaining_time": "2:21:19"}
+{"current_steps": 72, "total_steps": 705, "loss": 1.2526882886886597, "lr": 2.848123882972295e-05, "epoch": 0.30638297872340425, "percentage": 10.21, "elapsed_time": "0:16:01", "remaining_time": "2:20:48"}
+{"current_steps": 74, "total_steps": 705, "loss": 0.9114877581596375, "lr": 2.8306761798405526e-05, "epoch": 0.3148936170212766, "percentage": 10.5, "elapsed_time": "0:16:23", "remaining_time": "2:19:48"}
+{"current_steps": 76, "total_steps": 705, "loss": 1.2244638204574585, "lr": 2.812350770374273e-05, "epoch": 0.32340425531914896, "percentage": 10.78, "elapsed_time": "0:16:50", "remaining_time": "2:19:19"}
+{"current_steps": 78, "total_steps": 705, "loss": 1.2725831270217896, "lr": 2.793162201624631e-05, "epoch": 0.33191489361702126, "percentage": 11.06, "elapsed_time": "0:17:20", "remaining_time": "2:19:26"}
+{"current_steps": 80, "total_steps": 705, "loss": 0.7258127927780151, "lr": 2.77312570583471e-05, "epoch": 0.3404255319148936, "percentage": 11.35, "elapsed_time": "0:17:36", "remaining_time": "2:17:33"}
+{"current_steps": 82, "total_steps": 705, "loss": 1.2518136501312256, "lr": 2.752257188347862e-05, "epoch": 0.34893617021276596, "percentage": 11.63, "elapsed_time": "0:17:58", "remaining_time": "2:16:30"}
+{"current_steps": 84, "total_steps": 705, "loss": 0.9943649172782898, "lr": 2.730573214981751e-05, "epoch": 0.3574468085106383, "percentage": 11.91, "elapsed_time": "0:18:22", "remaining_time": "2:15:48"}
+{"current_steps": 86, "total_steps": 705, "loss": 1.001371145248413, "lr": 2.7080909988780982e-05, "epoch": 0.3659574468085106, "percentage": 12.2, "elapsed_time": "0:18:47", "remaining_time": "2:15:14"}
+{"current_steps": 88, "total_steps": 705, "loss": 1.2457740306854248, "lr": 2.684828386838569e-05, "epoch": 0.37446808510638296, "percentage": 12.48, "elapsed_time": "0:19:13", "remaining_time": "2:14:45"}
+{"current_steps": 90, "total_steps": 705, "loss": 0.9049152135848999, "lr": 2.6608038451576528e-05, "epoch": 0.3829787234042553, "percentage": 12.77, "elapsed_time": "0:19:33", "remaining_time": "2:13:40"}
+{"current_steps": 92, "total_steps": 705, "loss": 1.0495070219039917, "lr": 2.636036444963769e-05, "epoch": 0.39148936170212767, "percentage": 13.05, "elapsed_time": "0:19:57", "remaining_time": "2:12:59"}
+{"current_steps": 94, "total_steps": 705, "loss": 1.163374900817871, "lr": 2.6105458470802563e-05, "epoch": 0.4, "percentage": 13.33, "elapsed_time": "0:20:25", "remaining_time": "2:12:48"}
+{"current_steps": 96, "total_steps": 705, "loss": 1.064250111579895, "lr": 2.5843522864182394e-05, "epoch": 0.4085106382978723, "percentage": 13.62, "elapsed_time": "0:20:53", "remaining_time": "2:12:30"}
+{"current_steps": 98, "total_steps": 705, "loss": 1.0862053632736206, "lr": 2.557476555913785e-05, "epoch": 0.41702127659574467, "percentage": 13.9, "elapsed_time": "0:21:17", "remaining_time": "2:11:55"}
+{"current_steps": 100, "total_steps": 705, "loss": 0.9672101736068726, "lr": 2.5299399900220803e-05, "epoch": 0.425531914893617, "percentage": 14.18, "elapsed_time": "0:21:42", "remaining_time": "2:11:20"}
+{"current_steps": 102, "total_steps": 705, "loss": 1.38407564163208, "lr": 2.5017644477817424e-05, "epoch": 0.4340425531914894, "percentage": 14.47, "elapsed_time": "0:22:09", "remaining_time": "2:10:57"}
+{"current_steps": 104, "total_steps": 705, "loss": 1.0887106657028198, "lr": 2.47297229546271e-05, "epoch": 0.4425531914893617, "percentage": 14.75, "elapsed_time": "0:22:33", "remaining_time": "2:10:20"}
+{"current_steps": 106, "total_steps": 705, "loss": 0.8722383975982666, "lr": 2.4435863888114814e-05, "epoch": 0.451063829787234, "percentage": 15.04, "elapsed_time": "0:22:54", "remaining_time": "2:09:24"}
+{"current_steps": 108, "total_steps": 705, "loss": 1.2673208713531494, "lr": 2.4136300549077976e-05, "epoch": 0.4595744680851064, "percentage": 15.32, "elapsed_time": "0:23:25", "remaining_time": "2:09:31"}
+{"current_steps": 110, "total_steps": 705, "loss": 1.2440522909164429, "lr": 2.3831270736471703e-05, "epoch": 0.46808510638297873, "percentage": 15.6, "elapsed_time": "0:23:56", "remaining_time": "2:09:30"}
+{"current_steps": 112, "total_steps": 705, "loss": 1.1419363021850586, "lr": 2.352101658863959e-05, "epoch": 0.4765957446808511, "percentage": 15.89, "elapsed_time": "0:24:20", "remaining_time": "2:08:54"}
+{"current_steps": 114, "total_steps": 705, "loss": 1.0633783340454102, "lr": 2.32057843910998e-05, "epoch": 0.4851063829787234, "percentage": 16.17, "elapsed_time": "0:24:50", "remaining_time": "2:08:45"}
+{"current_steps": 116, "total_steps": 705, "loss": 1.179775595664978, "lr": 2.288582438103903e-05, "epoch": 0.49361702127659574, "percentage": 16.45, "elapsed_time": "0:25:19", "remaining_time": "2:08:33"}
+{"current_steps": 118, "total_steps": 705, "loss": 1.2002534866333008, "lr": 2.256139054866955e-05, "epoch": 0.502127659574468, "percentage": 16.74, "elapsed_time": "0:25:52", "remaining_time": "2:08:44"}
+{"current_steps": 120, "total_steps": 705, "loss": 0.9621443748474121, "lr": 2.2232740435607067e-05, "epoch": 0.5106382978723404, "percentage": 17.02, "elapsed_time": "0:26:15", "remaining_time": "2:08:00"}
+{"current_steps": 122, "total_steps": 705, "loss": 1.253274917602539, "lr": 2.19001349304294e-05, "epoch": 0.5191489361702127, "percentage": 17.3, "elapsed_time": "0:26:45", "remaining_time": "2:07:52"}
+{"current_steps": 124, "total_steps": 705, "loss": 1.3183720111846924, "lr": 2.156383806157826e-05, "epoch": 0.5276595744680851, "percentage": 17.59, "elapsed_time": "0:27:15", "remaining_time": "2:07:41"}
+{"current_steps": 126, "total_steps": 705, "loss": 1.238662838935852, "lr": 2.1224116787768552e-05, "epoch": 0.5361702127659574, "percentage": 17.87, "elapsed_time": "0:27:47", "remaining_time": "2:07:42"}
+{"current_steps": 128, "total_steps": 705, "loss": 1.1723605394363403, "lr": 2.0881240786071588e-05, "epoch": 0.5446808510638298, "percentage": 18.16, "elapsed_time": "0:28:11", "remaining_time": "2:07:06"}
+{"current_steps": 130, "total_steps": 705, "loss": 1.021695852279663, "lr": 2.05354822378404e-05, "epoch": 0.5531914893617021, "percentage": 18.44, "elapsed_time": "0:28:35", "remaining_time": "2:06:29"}
+{"current_steps": 132, "total_steps": 705, "loss": 1.2407881021499634, "lr": 2.018711561264714e-05, "epoch": 0.5617021276595745, "percentage": 18.72, "elapsed_time": "0:29:09", "remaining_time": "2:06:36"}
+{"current_steps": 134, "total_steps": 705, "loss": 1.186881422996521, "lr": 1.9836417450403978e-05, "epoch": 0.5702127659574469, "percentage": 19.01, "elapsed_time": "0:29:40", "remaining_time": "2:06:25"}
+{"current_steps": 136, "total_steps": 705, "loss": 0.965923547744751, "lr": 1.9483666141840615e-05, "epoch": 0.5787234042553191, "percentage": 19.29, "elapsed_time": "0:30:09", "remaining_time": "2:06:11"}
+{"current_steps": 138, "total_steps": 705, "loss": 0.5913432836532593, "lr": 1.9129141707512508e-05, "epoch": 0.5872340425531914, "percentage": 19.57, "elapsed_time": "0:30:27", "remaining_time": "2:05:08"}
+{"current_steps": 140, "total_steps": 705, "loss": 1.1055043935775757, "lr": 1.8773125575515364e-05, "epoch": 0.5957446808510638, "percentage": 19.86, "elapsed_time": "0:30:55", "remaining_time": "2:04:47"}
+{"current_steps": 142, "total_steps": 705, "loss": 1.3598577976226807, "lr": 1.8415900358082268e-05, "epoch": 0.6042553191489362, "percentage": 20.14, "elapsed_time": "0:31:28", "remaining_time": "2:04:47"}
+{"current_steps": 144, "total_steps": 705, "loss": 0.9218302369117737, "lr": 1.805774962724083e-05, "epoch": 0.6127659574468085, "percentage": 20.43, "elapsed_time": "0:31:44", "remaining_time": "2:03:41"}
+{"current_steps": 146, "total_steps": 705, "loss": 1.1118155717849731, "lr": 1.7698957689708426e-05, "epoch": 0.6212765957446809, "percentage": 20.71, "elapsed_time": "0:32:09", "remaining_time": "2:03:07"}
+{"current_steps": 148, "total_steps": 705, "loss": 1.2171759605407715, "lr": 1.7339809361204252e-05, "epoch": 0.6297872340425532, "percentage": 20.99, "elapsed_time": "0:32:44", "remaining_time": "2:03:11"}
+{"current_steps": 150, "total_steps": 705, "loss": 1.1775646209716797, "lr": 1.6980589740357294e-05, "epoch": 0.6382978723404256, "percentage": 21.28, "elapsed_time": "0:33:13", "remaining_time": "2:02:57"}
+{"current_steps": 152, "total_steps": 705, "loss": 1.1114767789840698, "lr": 1.6621583982389707e-05, "epoch": 0.6468085106382979, "percentage": 21.56, "elapsed_time": "0:33:36", "remaining_time": "2:02:16"}
+{"current_steps": 154, "total_steps": 705, "loss": 0.7064566612243652, "lr": 1.6263077072755326e-05, "epoch": 0.6553191489361702, "percentage": 21.84, "elapsed_time": "0:33:55", "remaining_time": "2:01:23"}
+{"current_steps": 156, "total_steps": 705, "loss": 0.8802153468132019, "lr": 1.5905353600912898e-05, "epoch": 0.6638297872340425, "percentage": 22.13, "elapsed_time": "0:34:14", "remaining_time": "2:00:31"}
+{"current_steps": 158, "total_steps": 705, "loss": 0.9915045499801636, "lr": 1.5548697534413646e-05, "epoch": 0.6723404255319149, "percentage": 22.41, "elapsed_time": "0:34:32", "remaining_time": "1:59:34"}
+{"current_steps": 160, "total_steps": 705, "loss": 1.2188233137130737, "lr": 1.5193391993482582e-05, "epoch": 0.6808510638297872, "percentage": 22.7, "elapsed_time": "0:35:07", "remaining_time": "1:59:39"}
+{"current_steps": 162, "total_steps": 705, "loss": 1.0931644439697266, "lr": 1.4839719026272377e-05, "epoch": 0.6893617021276596, "percentage": 22.98, "elapsed_time": "0:35:29", "remaining_time": "1:58:58"}
+{"current_steps": 164, "total_steps": 705, "loss": 0.875884473323822, "lr": 1.4487959384968272e-05, "epoch": 0.6978723404255319, "percentage": 23.26, "elapsed_time": "0:35:50", "remaining_time": "1:58:14"}
+{"current_steps": 166, "total_steps": 705, "loss": 0.9931239485740662, "lr": 1.4138392302921813e-05, "epoch": 0.7063829787234043, "percentage": 23.55, "elapsed_time": "0:36:17", "remaining_time": "1:57:48"}
+{"current_steps": 168, "total_steps": 705, "loss": 1.1112878322601318, "lr": 1.3791295272990175e-05, "epoch": 0.7148936170212766, "percentage": 23.83, "elapsed_time": "0:36:41", "remaining_time": "1:57:15"}
+{"current_steps": 170, "total_steps": 705, "loss": 0.8394916653633118, "lr": 1.344694382725718e-05, "epoch": 0.723404255319149, "percentage": 24.11, "elapsed_time": "0:37:05", "remaining_time": "1:56:42"}
+{"current_steps": 172, "total_steps": 705, "loss": 0.8859961032867432, "lr": 1.3105611318310818e-05, "epoch": 0.7319148936170212, "percentage": 24.4, "elapsed_time": "0:37:25", "remaining_time": "1:55:58"}
+{"current_steps": 174, "total_steps": 705, "loss": 1.1839947700500488, "lr": 1.2767568702250844e-05, "epoch": 0.7404255319148936, "percentage": 24.68, "elapsed_time": "0:37:56", "remaining_time": "1:55:46"}
+{"current_steps": 176, "total_steps": 705, "loss": 0.6160851120948792, "lr": 1.2433084323598791e-05, "epoch": 0.7489361702127659, "percentage": 24.96, "elapsed_time": "0:38:11", "remaining_time": "1:54:48"}
+{"current_steps": 178, "total_steps": 705, "loss": 1.0064830780029297, "lr": 1.2102423702281116e-05, "epoch": 0.7574468085106383, "percentage": 25.25, "elapsed_time": "0:38:43", "remaining_time": "1:54:39"}
+{"current_steps": 180, "total_steps": 705, "loss": 1.0609067678451538, "lr": 1.1775849322854516e-05, "epoch": 0.7659574468085106, "percentage": 25.53, "elapsed_time": "0:39:09", "remaining_time": "1:54:13"}
+{"current_steps": 182, "total_steps": 705, "loss": 0.7533571720123291, "lr": 1.1453620426140795e-05, "epoch": 0.774468085106383, "percentage": 25.82, "elapsed_time": "0:39:31", "remaining_time": "1:53:34"}
+{"current_steps": 184, "total_steps": 705, "loss": 0.9635902643203735, "lr": 1.1135992803436695e-05, "epoch": 0.7829787234042553, "percentage": 26.1, "elapsed_time": "0:39:54", "remaining_time": "1:52:59"}
+{"current_steps": 186, "total_steps": 705, "loss": 0.9765074849128723, "lr": 1.0823218593461992e-05, "epoch": 0.7914893617021277, "percentage": 26.38, "elapsed_time": "0:40:13", "remaining_time": "1:52:15"}
+{"current_steps": 188, "total_steps": 705, "loss": 0.730889081954956, "lr": 1.0515546082207097e-05, "epoch": 0.8, "percentage": 26.67, "elapsed_time": "0:40:34", "remaining_time": "1:51:35"}
+{"current_steps": 190, "total_steps": 705, "loss": 0.966530442237854, "lr": 1.0213219505838983e-05, "epoch": 0.8085106382978723, "percentage": 26.95, "elapsed_time": "0:40:58", "remaining_time": "1:51:03"}
+{"current_steps": 192, "total_steps": 705, "loss": 1.3054033517837524, "lr": 9.91647885682201e-06, "epoch": 0.8170212765957446, "percentage": 27.23, "elapsed_time": "0:41:31", "remaining_time": "1:50:57"}
+{"current_steps": 194, "total_steps": 705, "loss": 1.2358617782592773, "lr": 9.625559693407413e-06, "epoch": 0.825531914893617, "percentage": 27.52, "elapsed_time": "0:42:05", "remaining_time": "1:50:51"}
+{"current_steps": 196, "total_steps": 705, "loss": 0.789035975933075, "lr": 9.340692952642789e-06, "epoch": 0.8340425531914893, "percentage": 27.8, "elapsed_time": "0:42:23", "remaining_time": "1:50:05"}
+{"current_steps": 198, "total_steps": 705, "loss": 1.2202699184417725, "lr": 9.062104767049956e-06, "epoch": 0.8425531914893617, "percentage": 28.09, "elapsed_time": "0:42:58", "remaining_time": "1:50:02"}
+{"current_steps": 200, "total_steps": 705, "loss": 1.005751132965088, "lr": 8.790016285116763e-06, "epoch": 0.851063829787234, "percentage": 28.37, "elapsed_time": "0:43:28", "remaining_time": "1:49:46"}
+{"current_steps": 202, "total_steps": 705, "loss": 0.9970273971557617, "lr": 8.524643495745306e-06, "epoch": 0.8595744680851064, "percentage": 28.65, "elapsed_time": "0:43:56", "remaining_time": "1:49:25"}
+{"current_steps": 204, "total_steps": 705, "loss": 1.1567330360412598, "lr": 8.26619705679589e-06, "epoch": 0.8680851063829788, "percentage": 28.94, "elapsed_time": "0:44:26", "remaining_time": "1:49:08"}
+{"current_steps": 206, "total_steps": 705, "loss": 1.1192888021469116, "lr": 8.014882127862923e-06, "epoch": 0.8765957446808511, "percentage": 29.22, "elapsed_time": "0:44:51", "remaining_time": "1:48:38"}
+{"current_steps": 208, "total_steps": 705, "loss": 0.9340039491653442, "lr": 7.770898207415416e-06, "epoch": 0.8851063829787233, "percentage": 29.5, "elapsed_time": "0:45:18", "remaining_time": "1:48:15"}
+{"current_steps": 210, "total_steps": 705, "loss": 1.2615220546722412, "lr": 7.534438974431351e-06, "epoch": 0.8936170212765957, "percentage": 29.79, "elapsed_time": "0:45:50", "remaining_time": "1:48:04"}
+{"current_steps": 212, "total_steps": 705, "loss": 0.7793570160865784, "lr": 7.305692134651742e-06, "epoch": 0.902127659574468, "percentage": 30.07, "elapsed_time": "0:46:13", "remaining_time": "1:47:30"}
+{"current_steps": 214, "total_steps": 705, "loss": 0.9480925798416138, "lr": 7.084839271576291e-06, "epoch": 0.9106382978723404, "percentage": 30.35, "elapsed_time": "0:46:39", "remaining_time": "1:47:02"}
+{"current_steps": 216, "total_steps": 705, "loss": 0.7321507334709167, "lr": 6.872055702319054e-06, "epoch": 0.9191489361702128, "percentage": 30.64, "elapsed_time": "0:47:00", "remaining_time": "1:46:25"}
+{"current_steps": 218, "total_steps": 705, "loss": 1.2155194282531738, "lr": 6.667510338438419e-06, "epoch": 0.9276595744680851, "percentage": 30.92, "elapsed_time": "0:47:35", "remaining_time": "1:46:19"}
+{"current_steps": 220, "total_steps": 705, "loss": 0.39672887325286865, "lr": 6.471365551852012e-06, "epoch": 0.9361702127659575, "percentage": 31.21, "elapsed_time": "0:47:47", "remaining_time": "1:45:21"}
+{"current_steps": 222, "total_steps": 705, "loss": 1.254830241203308, "lr": 6.2837770459428e-06, "epoch": 0.9446808510638298, "percentage": 31.49, "elapsed_time": "0:48:17", "remaining_time": "1:45:04"}
+{"current_steps": 224, "total_steps": 705, "loss": 1.138830304145813, "lr": 6.1048937319588676e-06, "epoch": 0.9531914893617022, "percentage": 31.77, "elapsed_time": "0:48:44", "remaining_time": "1:44:40"}
+{"current_steps": 226, "total_steps": 705, "loss": 0.8572854399681091, "lr": 5.9348576108049065e-06, "epoch": 0.9617021276595744, "percentage": 32.06, "elapsed_time": "0:49:01", "remaining_time": "1:43:55"}
+{"current_steps": 228, "total_steps": 705, "loss": 1.0662239789962769, "lr": 5.773803660319234e-06, "epoch": 0.9702127659574468, "percentage": 32.34, "elapsed_time": "0:49:28", "remaining_time": "1:43:30"}
+{"current_steps": 230, "total_steps": 705, "loss": 1.222013235092163, "lr": 5.621859728125884e-06, "epoch": 0.9787234042553191, "percentage": 32.62, "elapsed_time": "0:49:58", "remaining_time": "1:43:11"}
+{"current_steps": 232, "total_steps": 705, "loss": 1.0417675971984863, "lr": 5.479146430146781e-06, "epoch": 0.9872340425531915, "percentage": 32.91, "elapsed_time": "0:50:26", "remaining_time": "1:42:50"}
+{"current_steps": 234, "total_steps": 705, "loss": 0.9495846629142761, "lr": 5.345777054854579e-06, "epoch": 0.9957446808510638, "percentage": 33.19, "elapsed_time": "0:50:48", "remaining_time": "1:42:16"}
+{"current_steps": 236, "total_steps": 705, "loss": 0.843216061592102, "lr": 5.221857473342149e-06, "epoch": 1.004255319148936, "percentage": 33.48, "elapsed_time": "0:51:15", "remaining_time": "1:41:51"}
+{"current_steps": 238, "total_steps": 705, "loss": 0.9633752107620239, "lr": 5.10748605528015e-06, "epoch": 1.0127659574468084, "percentage": 33.76, "elapsed_time": "0:51:50", "remaining_time": "1:41:43"}
+{"current_steps": 240, "total_steps": 705, "loss": 0.6067866683006287, "lr": 5.002753590829349e-06, "epoch": 1.0212765957446808, "percentage": 34.04, "elapsed_time": "0:52:14", "remaining_time": "1:41:12"}
+{"current_steps": 242, "total_steps": 705, "loss": 0.9341376423835754, "lr": 4.9077432185697e-06, "epoch": 1.0297872340425531, "percentage": 34.33, "elapsed_time": "0:52:46", "remaining_time": "1:40:57"}
+{"current_steps": 244, "total_steps": 705, "loss": 1.046706199645996, "lr": 4.822530359503393e-06, "epoch": 1.0382978723404255, "percentage": 34.61, "elapsed_time": "0:53:16", "remaining_time": "1:40:40"}
+{"current_steps": 246, "total_steps": 705, "loss": 0.6613105535507202, "lr": 4.747182657184251e-06, "epoch": 1.0468085106382978, "percentage": 34.89, "elapsed_time": "0:53:35", "remaining_time": "1:39:59"}
+{"current_steps": 248, "total_steps": 705, "loss": 0.7598414421081543, "lr": 4.681759924021033e-06, "epoch": 1.0553191489361702, "percentage": 35.18, "elapsed_time": "0:54:04", "remaining_time": "1:39:39"}
+{"current_steps": 250, "total_steps": 705, "loss": 0.7334257364273071, "lr": 4.626314093797213e-06, "epoch": 1.0638297872340425, "percentage": 35.46, "elapsed_time": "0:54:29", "remaining_time": "1:39:11"}
+{"current_steps": 252, "total_steps": 705, "loss": 0.5062970519065857, "lr": 4.580889180444988e-06, "epoch": 1.0723404255319149, "percentage": 35.74, "elapsed_time": "0:54:47", "remaining_time": "1:38:30"}
+{"current_steps": 254, "total_steps": 705, "loss": 1.0599923133850098, "lr": 4.545521243106197e-06, "epoch": 1.0808510638297872, "percentage": 36.03, "elapsed_time": "0:55:22", "remaining_time": "1:38:20"}
+{"current_steps": 256, "total_steps": 705, "loss": 0.940389096736908, "lr": 4.520238357507899e-06, "epoch": 1.0893617021276596, "percentage": 36.31, "elapsed_time": "0:55:57", "remaining_time": "1:38:08"}
+{"current_steps": 258, "total_steps": 705, "loss": 1.0018244981765747, "lr": 4.505060593675342e-06, "epoch": 1.097872340425532, "percentage": 36.6, "elapsed_time": "0:56:27", "remaining_time": "1:37:48"}
+{"current_steps": 260, "total_steps": 705, "loss": 0.8150784969329834, "lr": 4.5e-06, "epoch": 1.1063829787234043, "percentage": 36.88, "elapsed_time": "0:56:54", "remaining_time": "1:37:24"}
+{"current_steps": 262, "total_steps": 705, "loss": 0.7950176000595093, "lr": 4.505060593675342e-06, "epoch": 1.1148936170212767, "percentage": 37.16, "elapsed_time": "0:57:21", "remaining_time": "1:36:58"}
+{"current_steps": 264, "total_steps": 705, "loss": 0.7621920704841614, "lr": 4.520238357507899e-06, "epoch": 1.123404255319149, "percentage": 37.45, "elapsed_time": "0:57:44", "remaining_time": "1:36:26"}
+{"current_steps": 266, "total_steps": 705, "loss": 0.8206462264060974, "lr": 4.545521243106197e-06, "epoch": 1.1319148936170214, "percentage": 37.73, "elapsed_time": "0:58:07", "remaining_time": "1:35:56"}
+{"current_steps": 268, "total_steps": 705, "loss": 0.9801955819129944, "lr": 4.580889180444988e-06, "epoch": 1.1404255319148937, "percentage": 38.01, "elapsed_time": "0:58:35", "remaining_time": "1:35:32"}
+{"current_steps": 270, "total_steps": 705, "loss": 0.8219439387321472, "lr": 4.6263140937972124e-06, "epoch": 1.148936170212766, "percentage": 38.3, "elapsed_time": "0:59:03", "remaining_time": "1:35:08"}
+{"current_steps": 272, "total_steps": 705, "loss": 0.6543676257133484, "lr": 4.6817599240210315e-06, "epoch": 1.1574468085106382, "percentage": 38.58, "elapsed_time": "0:59:26", "remaining_time": "1:34:38"}
+{"current_steps": 274, "total_steps": 705, "loss": 0.9012349247932434, "lr": 4.747182657184251e-06, "epoch": 1.1659574468085105, "percentage": 38.87, "elapsed_time": "0:59:53", "remaining_time": "1:34:12"}
+{"current_steps": 276, "total_steps": 705, "loss": 0.7129980325698853, "lr": 4.822530359503391e-06, "epoch": 1.174468085106383, "percentage": 39.15, "elapsed_time": "1:00:22", "remaining_time": "1:33:51"}
+{"current_steps": 278, "total_steps": 705, "loss": 0.5482591986656189, "lr": 4.9077432185697e-06, "epoch": 1.1829787234042553, "percentage": 39.43, "elapsed_time": "1:00:47", "remaining_time": "1:33:21"}
+{"current_steps": 280, "total_steps": 705, "loss": 0.5497639775276184, "lr": 5.002753590829348e-06, "epoch": 1.1914893617021276, "percentage": 39.72, "elapsed_time": "1:01:11", "remaining_time": "1:32:52"}
+{"current_steps": 282, "total_steps": 705, "loss": 0.6242255568504333, "lr": 5.1074860552801466e-06, "epoch": 1.2, "percentage": 40.0, "elapsed_time": "1:01:34", "remaining_time": "1:32:22"}
+{"current_steps": 284, "total_steps": 705, "loss": 0.693662703037262, "lr": 5.2218574733421455e-06, "epoch": 1.2085106382978723, "percentage": 40.28, "elapsed_time": "1:01:56", "remaining_time": "1:31:48"}
+{"current_steps": 286, "total_steps": 705, "loss": 0.9797627329826355, "lr": 5.345777054854578e-06, "epoch": 1.2170212765957447, "percentage": 40.57, "elapsed_time": "1:02:23", "remaining_time": "1:31:24"}
+{"current_steps": 288, "total_steps": 705, "loss": 0.9435557126998901, "lr": 5.479146430146783e-06, "epoch": 1.225531914893617, "percentage": 40.85, "elapsed_time": "1:02:51", "remaining_time": "1:31:00"}
+{"current_steps": 290, "total_steps": 705, "loss": 0.6460038423538208, "lr": 5.6218597281258834e-06, "epoch": 1.2340425531914894, "percentage": 41.13, "elapsed_time": "1:03:10", "remaining_time": "1:30:24"}
+{"current_steps": 292, "total_steps": 705, "loss": 0.8468654155731201, "lr": 5.773803660319232e-06, "epoch": 1.2425531914893617, "percentage": 41.42, "elapsed_time": "1:03:33", "remaining_time": "1:29:53"}
+{"current_steps": 294, "total_steps": 705, "loss": 0.8879528641700745, "lr": 5.934857610804904e-06, "epoch": 1.251063829787234, "percentage": 41.7, "elapsed_time": "1:04:08", "remaining_time": "1:29:40"}
+{"current_steps": 296, "total_steps": 705, "loss": 0.7949476838111877, "lr": 6.1048937319588676e-06, "epoch": 1.2595744680851064, "percentage": 41.99, "elapsed_time": "1:04:33", "remaining_time": "1:29:12"}
+{"current_steps": 298, "total_steps": 705, "loss": 0.8318912386894226, "lr": 6.283777045942798e-06, "epoch": 1.2680851063829788, "percentage": 42.27, "elapsed_time": "1:05:02", "remaining_time": "1:28:50"}
+{"current_steps": 300, "total_steps": 705, "loss": 1.0059137344360352, "lr": 6.471365551852011e-06, "epoch": 1.2765957446808511, "percentage": 42.55, "elapsed_time": "1:05:34", "remaining_time": "1:28:31"}
+{"current_steps": 302, "total_steps": 705, "loss": 0.39432206749916077, "lr": 6.66751033843842e-06, "epoch": 1.2851063829787235, "percentage": 42.84, "elapsed_time": "1:05:48", "remaining_time": "1:27:49"}
+{"current_steps": 304, "total_steps": 705, "loss": 0.9369863271713257, "lr": 6.872055702319048e-06, "epoch": 1.2936170212765958, "percentage": 43.12, "elapsed_time": "1:06:19", "remaining_time": "1:27:29"}
+{"current_steps": 306, "total_steps": 705, "loss": 0.9826689958572388, "lr": 7.084839271576289e-06, "epoch": 1.302127659574468, "percentage": 43.4, "elapsed_time": "1:06:47", "remaining_time": "1:27:06"}
+{"current_steps": 308, "total_steps": 705, "loss": 0.5573344826698303, "lr": 7.30569213465174e-06, "epoch": 1.3106382978723405, "percentage": 43.69, "elapsed_time": "1:07:08", "remaining_time": "1:26:32"}
+{"current_steps": 310, "total_steps": 705, "loss": 0.5633378624916077, "lr": 7.534438974431351e-06, "epoch": 1.3191489361702127, "percentage": 43.97, "elapsed_time": "1:07:28", "remaining_time": "1:25:58"}
+{"current_steps": 312, "total_steps": 705, "loss": 0.7375507354736328, "lr": 7.770898207415414e-06, "epoch": 1.327659574468085, "percentage": 44.26, "elapsed_time": "1:07:54", "remaining_time": "1:25:32"}
+{"current_steps": 314, "total_steps": 705, "loss": 0.4668198823928833, "lr": 8.014882127862923e-06, "epoch": 1.3361702127659574, "percentage": 44.54, "elapsed_time": "1:08:12", "remaining_time": "1:24:56"}
+{"current_steps": 316, "total_steps": 705, "loss": 0.7280799150466919, "lr": 8.266197056795886e-06, "epoch": 1.3446808510638297, "percentage": 44.82, "elapsed_time": "1:08:41", "remaining_time": "1:24:33"}
+{"current_steps": 318, "total_steps": 705, "loss": 0.5291551351547241, "lr": 8.5246434957453e-06, "epoch": 1.353191489361702, "percentage": 45.11, "elapsed_time": "1:09:01", "remaining_time": "1:24:00"}
+{"current_steps": 320, "total_steps": 705, "loss": 0.5350017547607422, "lr": 8.790016285116766e-06, "epoch": 1.3617021276595744, "percentage": 45.39, "elapsed_time": "1:09:21", "remaining_time": "1:23:26"}
+{"current_steps": 322, "total_steps": 705, "loss": 0.9615387916564941, "lr": 9.062104767049957e-06, "epoch": 1.3702127659574468, "percentage": 45.67, "elapsed_time": "1:09:53", "remaining_time": "1:23:07"}
+{"current_steps": 324, "total_steps": 705, "loss": 0.6444461345672607, "lr": 9.340692952642787e-06, "epoch": 1.3787234042553191, "percentage": 45.96, "elapsed_time": "1:10:14", "remaining_time": "1:22:36"}
+{"current_steps": 326, "total_steps": 705, "loss": 0.9194571375846863, "lr": 9.625559693407413e-06, "epoch": 1.3872340425531915, "percentage": 46.24, "elapsed_time": "1:10:46", "remaining_time": "1:22:16"}
+{"current_steps": 328, "total_steps": 705, "loss": 0.9841423034667969, "lr": 9.916478856822006e-06, "epoch": 1.3957446808510638, "percentage": 46.52, "elapsed_time": "1:11:12", "remaining_time": "1:21:50"}
+{"current_steps": 330, "total_steps": 705, "loss": 0.8577695488929749, "lr": 1.0213219505838985e-05, "epoch": 1.4042553191489362, "percentage": 46.81, "elapsed_time": "1:11:38", "remaining_time": "1:21:25"}
+{"current_steps": 332, "total_steps": 705, "loss": 0.8521637916564941, "lr": 1.0515546082207094e-05, "epoch": 1.4127659574468086, "percentage": 47.09, "elapsed_time": "1:12:06", "remaining_time": "1:21:01"}
+{"current_steps": 334, "total_steps": 705, "loss": 0.6609964966773987, "lr": 1.082321859346199e-05, "epoch": 1.421276595744681, "percentage": 47.38, "elapsed_time": "1:12:33", "remaining_time": "1:20:35"}
+{"current_steps": 336, "total_steps": 705, "loss": 0.5523483753204346, "lr": 1.1135992803436696e-05, "epoch": 1.4297872340425533, "percentage": 47.66, "elapsed_time": "1:12:55", "remaining_time": "1:20:05"}
+{"current_steps": 338, "total_steps": 705, "loss": 0.7419446110725403, "lr": 1.1453620426140791e-05, "epoch": 1.4382978723404256, "percentage": 47.94, "elapsed_time": "1:13:25", "remaining_time": "1:19:43"}
+{"current_steps": 340, "total_steps": 705, "loss": 0.7033047676086426, "lr": 1.1775849322854508e-05, "epoch": 1.4468085106382977, "percentage": 48.23, "elapsed_time": "1:13:50", "remaining_time": "1:19:16"}
+{"current_steps": 342, "total_steps": 705, "loss": 0.5836398005485535, "lr": 1.210242370228112e-05, "epoch": 1.4553191489361703, "percentage": 48.51, "elapsed_time": "1:14:14", "remaining_time": "1:18:48"}
+{"current_steps": 344, "total_steps": 705, "loss": 0.9612702131271362, "lr": 1.2433084323598791e-05, "epoch": 1.4638297872340424, "percentage": 48.79, "elapsed_time": "1:14:48", "remaining_time": "1:18:30"}
+{"current_steps": 346, "total_steps": 705, "loss": 0.9444953799247742, "lr": 1.2767568702250838e-05, "epoch": 1.472340425531915, "percentage": 49.08, "elapsed_time": "1:15:20", "remaining_time": "1:18:10"}
+{"current_steps": 348, "total_steps": 705, "loss": 0.8060452938079834, "lr": 1.3105611318310811e-05, "epoch": 1.4808510638297872, "percentage": 49.36, "elapsed_time": "1:15:46", "remaining_time": "1:17:44"}
+{"current_steps": 350, "total_steps": 705, "loss": 0.8150795698165894, "lr": 1.3446943827257184e-05, "epoch": 1.4893617021276595, "percentage": 49.65, "elapsed_time": "1:16:10", "remaining_time": "1:17:15"}
+{"current_steps": 352, "total_steps": 705, "loss": 0.3241785168647766, "lr": 1.3791295272990172e-05, "epoch": 1.4978723404255319, "percentage": 49.93, "elapsed_time": "1:16:27", "remaining_time": "1:16:40"}
+{"current_steps": 354, "total_steps": 705, "loss": 0.7416596412658691, "lr": 1.413839230292182e-05, "epoch": 1.5063829787234042, "percentage": 50.21, "elapsed_time": "1:16:50", "remaining_time": "1:16:11"}
+{"current_steps": 356, "total_steps": 705, "loss": 0.518079400062561, "lr": 1.4487959384968276e-05, "epoch": 1.5148936170212766, "percentage": 50.5, "elapsed_time": "1:17:16", "remaining_time": "1:15:45"}
+{"current_steps": 358, "total_steps": 705, "loss": 0.5670950412750244, "lr": 1.4839719026272373e-05, "epoch": 1.523404255319149, "percentage": 50.78, "elapsed_time": "1:17:39", "remaining_time": "1:15:16"}
+{"current_steps": 360, "total_steps": 705, "loss": 1.1480307579040527, "lr": 1.5193391993482579e-05, "epoch": 1.5319148936170213, "percentage": 51.06, "elapsed_time": "1:18:11", "remaining_time": "1:14:56"}
+{"current_steps": 362, "total_steps": 705, "loss": 0.5208263397216797, "lr": 1.5548697534413636e-05, "epoch": 1.5404255319148936, "percentage": 51.35, "elapsed_time": "1:18:29", "remaining_time": "1:14:22"}
+{"current_steps": 364, "total_steps": 705, "loss": 0.9330644607543945, "lr": 1.5905353600912894e-05, "epoch": 1.548936170212766, "percentage": 51.63, "elapsed_time": "1:18:58", "remaining_time": "1:13:59"}
+{"current_steps": 366, "total_steps": 705, "loss": 0.8360081911087036, "lr": 1.6263077072755323e-05, "epoch": 1.5574468085106383, "percentage": 51.91, "elapsed_time": "1:19:24", "remaining_time": "1:13:32"}
+{"current_steps": 368, "total_steps": 705, "loss": 0.7362462878227234, "lr": 1.6621583982389707e-05, "epoch": 1.5659574468085107, "percentage": 52.2, "elapsed_time": "1:19:43", "remaining_time": "1:13:00"}
+{"current_steps": 370, "total_steps": 705, "loss": 0.6744348406791687, "lr": 1.6980589740357294e-05, "epoch": 1.574468085106383, "percentage": 52.48, "elapsed_time": "1:20:10", "remaining_time": "1:12:35"}
+{"current_steps": 372, "total_steps": 705, "loss": 0.9976662397384644, "lr": 1.733980936120425e-05, "epoch": 1.5829787234042554, "percentage": 52.77, "elapsed_time": "1:20:46", "remaining_time": "1:12:18"}
+{"current_steps": 374, "total_steps": 705, "loss": 1.0515512228012085, "lr": 1.7698957689708416e-05, "epoch": 1.5914893617021275, "percentage": 53.05, "elapsed_time": "1:21:23", "remaining_time": "1:12:01"}
+{"current_steps": 376, "total_steps": 705, "loss": 0.7891988754272461, "lr": 1.805774962724083e-05, "epoch": 1.6, "percentage": 53.33, "elapsed_time": "1:21:50", "remaining_time": "1:11:36"}
+{"current_steps": 378, "total_steps": 705, "loss": 0.6114462018013, "lr": 1.8415900358082265e-05, "epoch": 1.6085106382978722, "percentage": 53.62, "elapsed_time": "1:22:14", "remaining_time": "1:11:09"}
+{"current_steps": 380, "total_steps": 705, "loss": 0.5973749756813049, "lr": 1.8773125575515357e-05, "epoch": 1.6170212765957448, "percentage": 53.9, "elapsed_time": "1:22:38", "remaining_time": "1:10:40"}
+{"current_steps": 382, "total_steps": 705, "loss": 0.7943100929260254, "lr": 1.912914170751251e-05, "epoch": 1.625531914893617, "percentage": 54.18, "elapsed_time": "1:23:07", "remaining_time": "1:10:16"}
+{"current_steps": 384, "total_steps": 705, "loss": 0.9919679760932922, "lr": 1.9483666141840612e-05, "epoch": 1.6340425531914895, "percentage": 54.47, "elapsed_time": "1:23:34", "remaining_time": "1:09:51"}
+{"current_steps": 386, "total_steps": 705, "loss": 0.8425494432449341, "lr": 1.9836417450403974e-05, "epoch": 1.6425531914893616, "percentage": 54.75, "elapsed_time": "1:23:54", "remaining_time": "1:09:20"}
+{"current_steps": 388, "total_steps": 705, "loss": 0.7412829399108887, "lr": 2.0187115612647133e-05, "epoch": 1.6510638297872342, "percentage": 55.04, "elapsed_time": "1:24:21", "remaining_time": "1:08:55"}
+{"current_steps": 390, "total_steps": 705, "loss": 0.7992886900901794, "lr": 2.05354822378404e-05, "epoch": 1.6595744680851063, "percentage": 55.32, "elapsed_time": "1:24:48", "remaining_time": "1:08:29"}
+{"current_steps": 392, "total_steps": 705, "loss": 0.909046471118927, "lr": 2.0881240786071595e-05, "epoch": 1.6680851063829787, "percentage": 55.6, "elapsed_time": "1:25:22", "remaining_time": "1:08:09"}
+{"current_steps": 394, "total_steps": 705, "loss": 0.6243407726287842, "lr": 2.122411678776854e-05, "epoch": 1.676595744680851, "percentage": 55.89, "elapsed_time": "1:25:44", "remaining_time": "1:07:40"}
+{"current_steps": 396, "total_steps": 705, "loss": 0.8719924688339233, "lr": 2.1563838061578258e-05, "epoch": 1.6851063829787234, "percentage": 56.17, "elapsed_time": "1:26:11", "remaining_time": "1:07:15"}
+{"current_steps": 398, "total_steps": 705, "loss": 1.0623129606246948, "lr": 2.1900134930429396e-05, "epoch": 1.6936170212765957, "percentage": 56.45, "elapsed_time": "1:26:43", "remaining_time": "1:06:53"}
+{"current_steps": 400, "total_steps": 705, "loss": 0.9607300162315369, "lr": 2.223274043560706e-05, "epoch": 1.702127659574468, "percentage": 56.74, "elapsed_time": "1:27:16", "remaining_time": "1:06:33"}
+{"current_steps": 402, "total_steps": 705, "loss": 0.4819542169570923, "lr": 2.256139054866954e-05, "epoch": 1.7106382978723405, "percentage": 57.02, "elapsed_time": "1:27:33", "remaining_time": "1:05:59"}
+{"current_steps": 404, "total_steps": 705, "loss": 0.6816413402557373, "lr": 2.2885824381039028e-05, "epoch": 1.7191489361702128, "percentage": 57.3, "elapsed_time": "1:27:54", "remaining_time": "1:05:29"}
+{"current_steps": 406, "total_steps": 705, "loss": 0.5322574377059937, "lr": 2.3205784391099808e-05, "epoch": 1.7276595744680852, "percentage": 57.59, "elapsed_time": "1:28:11", "remaining_time": "1:04:56"}
+{"current_steps": 408, "total_steps": 705, "loss": 0.8366644382476807, "lr": 2.352101658863958e-05, "epoch": 1.7361702127659573, "percentage": 57.87, "elapsed_time": "1:28:34", "remaining_time": "1:04:28"}
+{"current_steps": 410, "total_steps": 705, "loss": 0.8219179511070251, "lr": 2.3831270736471703e-05, "epoch": 1.7446808510638299, "percentage": 58.16, "elapsed_time": "1:28:59", "remaining_time": "1:04:01"}
+{"current_steps": 412, "total_steps": 705, "loss": 0.8943849802017212, "lr": 2.4136300549077973e-05, "epoch": 1.753191489361702, "percentage": 58.44, "elapsed_time": "1:29:29", "remaining_time": "1:03:38"}
+{"current_steps": 414, "total_steps": 705, "loss": 0.7499610781669617, "lr": 2.443586388811481e-05, "epoch": 1.7617021276595746, "percentage": 58.72, "elapsed_time": "1:29:54", "remaining_time": "1:03:12"}
+{"current_steps": 416, "total_steps": 705, "loss": 0.6121576428413391, "lr": 2.47297229546271e-05, "epoch": 1.7702127659574467, "percentage": 59.01, "elapsed_time": "1:30:15", "remaining_time": "1:02:42"}
+{"current_steps": 418, "total_steps": 705, "loss": 0.7867974638938904, "lr": 2.5017644477817418e-05, "epoch": 1.7787234042553193, "percentage": 59.29, "elapsed_time": "1:30:38", "remaining_time": "1:02:14"}
+{"current_steps": 420, "total_steps": 705, "loss": 0.6045423150062561, "lr": 2.5299399900220807e-05, "epoch": 1.7872340425531914, "percentage": 59.57, "elapsed_time": "1:31:05", "remaining_time": "1:01:48"}
+{"current_steps": 422, "total_steps": 705, "loss": 0.7998336553573608, "lr": 2.5574765559137848e-05, "epoch": 1.795744680851064, "percentage": 59.86, "elapsed_time": "1:31:29", "remaining_time": "1:01:21"}
+{"current_steps": 424, "total_steps": 705, "loss": 0.7926132678985596, "lr": 2.584352286418239e-05, "epoch": 1.804255319148936, "percentage": 60.14, "elapsed_time": "1:31:53", "remaining_time": "1:00:54"}
+{"current_steps": 426, "total_steps": 705, "loss": 0.8235700130462646, "lr": 2.6105458470802563e-05, "epoch": 1.8127659574468085, "percentage": 60.43, "elapsed_time": "1:32:20", "remaining_time": "1:00:28"}
+{"current_steps": 428, "total_steps": 705, "loss": 0.8320169448852539, "lr": 2.6360364449637683e-05, "epoch": 1.8212765957446808, "percentage": 60.71, "elapsed_time": "1:32:47", "remaining_time": "1:00:03"}
+{"current_steps": 430, "total_steps": 705, "loss": 0.8250552415847778, "lr": 2.6608038451576528e-05, "epoch": 1.8297872340425532, "percentage": 60.99, "elapsed_time": "1:33:05", "remaining_time": "0:59:32"}
+{"current_steps": 432, "total_steps": 705, "loss": 0.8021817207336426, "lr": 2.684828386838569e-05, "epoch": 1.8382978723404255, "percentage": 61.28, "elapsed_time": "1:33:31", "remaining_time": "0:59:05"}
+{"current_steps": 434, "total_steps": 705, "loss": 0.7879712581634521, "lr": 2.7080909988780982e-05, "epoch": 1.8468085106382979, "percentage": 61.56, "elapsed_time": "1:33:58", "remaining_time": "0:58:41"}
+{"current_steps": 436, "total_steps": 705, "loss": 0.7762024998664856, "lr": 2.7305732149817502e-05, "epoch": 1.8553191489361702, "percentage": 61.84, "elapsed_time": "1:34:23", "remaining_time": "0:58:14"}
+{"current_steps": 438, "total_steps": 705, "loss": 1.2395200729370117, "lr": 2.7522571883478617e-05, "epoch": 1.8638297872340426, "percentage": 62.13, "elapsed_time": "1:34:56", "remaining_time": "0:57:52"}
+{"current_steps": 440, "total_steps": 705, "loss": 0.8664835691452026, "lr": 2.77312570583471e-05, "epoch": 1.872340425531915, "percentage": 62.41, "elapsed_time": "1:35:21", "remaining_time": "0:57:25"}
+{"current_steps": 442, "total_steps": 705, "loss": 1.0493484735488892, "lr": 2.7931622016246304e-05, "epoch": 1.8808510638297873, "percentage": 62.7, "elapsed_time": "1:35:49", "remaining_time": "0:57:01"}
+{"current_steps": 444, "total_steps": 705, "loss": 0.8971328139305115, "lr": 2.8123507703742727e-05, "epoch": 1.8893617021276596, "percentage": 62.98, "elapsed_time": "1:36:20", "remaining_time": "0:56:38"}
+{"current_steps": 446, "total_steps": 705, "loss": 0.9821380376815796, "lr": 2.8306761798405522e-05, "epoch": 1.8978723404255318, "percentage": 63.26, "elapsed_time": "1:36:53", "remaining_time": "0:56:15"}
+{"current_steps": 448, "total_steps": 705, "loss": 1.0046873092651367, "lr": 2.848123882972295e-05, "epoch": 1.9063829787234043, "percentage": 63.55, "elapsed_time": "1:37:21", "remaining_time": "0:55:50"}
+{"current_steps": 450, "total_steps": 705, "loss": 0.647858738899231, "lr": 2.8646800294579514e-05, "epoch": 1.9148936170212765, "percentage": 63.83, "elapsed_time": "1:37:44", "remaining_time": "0:55:23"}
+{"current_steps": 452, "total_steps": 705, "loss": 0.5715224742889404, "lr": 2.8803314767202376e-05, "epoch": 1.923404255319149, "percentage": 64.11, "elapsed_time": "1:38:04", "remaining_time": "0:54:53"}
+{"current_steps": 454, "total_steps": 705, "loss": 0.8555135726928711, "lr": 2.8950658003489534e-05, "epoch": 1.9319148936170212, "percentage": 64.4, "elapsed_time": "1:38:33", "remaining_time": "0:54:29"}
+{"current_steps": 456, "total_steps": 705, "loss": 0.9127392768859863, "lr": 2.908871303963711e-05, "epoch": 1.9404255319148938, "percentage": 64.68, "elapsed_time": "1:39:03", "remaining_time": "0:54:05"}
+{"current_steps": 458, "total_steps": 705, "loss": 0.9905153512954712, "lr": 2.9217370284987434e-05, "epoch": 1.9489361702127659, "percentage": 64.96, "elapsed_time": "1:39:31", "remaining_time": "0:53:40"}
+{"current_steps": 460, "total_steps": 705, "loss": 1.022163987159729, "lr": 2.9336527609024075e-05, "epoch": 1.9574468085106385, "percentage": 65.25, "elapsed_time": "1:40:04", "remaining_time": "0:53:18"}
+{"current_steps": 462, "total_steps": 705, "loss": 0.8251804709434509, "lr": 2.9446090422445016e-05, "epoch": 1.9659574468085106, "percentage": 65.53, "elapsed_time": "1:40:31", "remaining_time": "0:52:52"}
+{"current_steps": 464, "total_steps": 705, "loss": 1.0136280059814453, "lr": 2.9545971752249376e-05, "epoch": 1.974468085106383, "percentage": 65.82, "elapsed_time": "1:40:59", "remaining_time": "0:52:27"}
+{"current_steps": 466, "total_steps": 705, "loss": 0.9567205905914307, "lr": 2.9636092310778195e-05, "epoch": 1.9829787234042553, "percentage": 66.1, "elapsed_time": "1:41:20", "remaining_time": "0:51:58"}
+{"current_steps": 468, "total_steps": 705, "loss": 0.6730895042419434, "lr": 2.9716380558654445e-05, "epoch": 1.9914893617021276, "percentage": 66.38, "elapsed_time": "1:41:41", "remaining_time": "0:51:29"}
+{"current_steps": 470, "total_steps": 705, "loss": 0.6498197317123413, "lr": 2.9786772761572338e-05, "epoch": 2.0, "percentage": 66.67, "elapsed_time": "1:42:00", "remaining_time": "0:51:00"}
+{"current_steps": 472, "total_steps": 705, "loss": 0.4679168164730072, "lr": 2.9847213040890793e-05, "epoch": 2.008510638297872, "percentage": 66.95, "elapsed_time": "1:42:19", "remaining_time": "0:50:30"}
+{"current_steps": 474, "total_steps": 705, "loss": 0.7698261737823486, "lr": 2.989765341799095e-05, "epoch": 2.0170212765957447, "percentage": 67.23, "elapsed_time": "1:42:53", "remaining_time": "0:50:08"}
+{"current_steps": 476, "total_steps": 705, "loss": 0.47750866413116455, "lr": 2.9938053852362484e-05, "epoch": 2.025531914893617, "percentage": 67.52, "elapsed_time": "1:43:17", "remaining_time": "0:49:41"}
+{"current_steps": 478, "total_steps": 705, "loss": 0.6370930075645447, "lr": 2.99683822733885e-05, "epoch": 2.0340425531914894, "percentage": 67.8, "elapsed_time": "1:43:42", "remaining_time": "0:49:15"}
+{"current_steps": 480, "total_steps": 705, "loss": 0.44695600867271423, "lr": 2.9988614605803806e-05, "epoch": 2.0425531914893615, "percentage": 68.09, "elapsed_time": "1:44:06", "remaining_time": "0:48:47"}
+{"current_steps": 482, "total_steps": 705, "loss": 0.4687855839729309, "lr": 2.9998734788806287e-05, "epoch": 2.051063829787234, "percentage": 68.37, "elapsed_time": "1:44:30", "remaining_time": "0:48:21"}
+{"current_steps": 484, "total_steps": 705, "loss": 0.28051936626434326, "lr": 2.9998734788806287e-05, "epoch": 2.0595744680851062, "percentage": 68.65, "elapsed_time": "1:44:42", "remaining_time": "0:47:48"}
+{"current_steps": 486, "total_steps": 705, "loss": 0.6579060554504395, "lr": 2.9988614605803806e-05, "epoch": 2.068085106382979, "percentage": 68.94, "elapsed_time": "1:45:06", "remaining_time": "0:47:21"}
+{"current_steps": 488, "total_steps": 705, "loss": 0.5791319608688354, "lr": 2.99683822733885e-05, "epoch": 2.076595744680851, "percentage": 69.22, "elapsed_time": "1:45:32", "remaining_time": "0:46:55"}
+{"current_steps": 490, "total_steps": 705, "loss": 0.5440673828125, "lr": 2.9938053852362484e-05, "epoch": 2.0851063829787235, "percentage": 69.5, "elapsed_time": "1:45:58", "remaining_time": "0:46:30"}
+{"current_steps": 492, "total_steps": 705, "loss": 0.7786204218864441, "lr": 2.989765341799095e-05, "epoch": 2.0936170212765957, "percentage": 69.79, "elapsed_time": "1:46:28", "remaining_time": "0:46:05"}
+{"current_steps": 494, "total_steps": 705, "loss": 0.6289861798286438, "lr": 2.9847213040890793e-05, "epoch": 2.1021276595744682, "percentage": 70.07, "elapsed_time": "1:47:03", "remaining_time": "0:45:43"}
+{"current_steps": 496, "total_steps": 705, "loss": 0.7049335241317749, "lr": 2.9786772761572338e-05, "epoch": 2.1106382978723404, "percentage": 70.35, "elapsed_time": "1:47:33", "remaining_time": "0:45:19"}
+{"current_steps": 498, "total_steps": 705, "loss": 0.9018527269363403, "lr": 2.9716380558654445e-05, "epoch": 2.119148936170213, "percentage": 70.64, "elapsed_time": "1:48:12", "remaining_time": "0:44:58"}
+{"current_steps": 500, "total_steps": 705, "loss": 0.6065503358840942, "lr": 2.9636092310778195e-05, "epoch": 2.127659574468085, "percentage": 70.92, "elapsed_time": "1:48:42", "remaining_time": "0:44:34"}
+{"current_steps": 502, "total_steps": 705, "loss": 0.5845508575439453, "lr": 2.9545971752249383e-05, "epoch": 2.1361702127659576, "percentage": 71.21, "elapsed_time": "1:49:02", "remaining_time": "0:44:05"}
+{"current_steps": 504, "total_steps": 705, "loss": 0.6952933669090271, "lr": 2.9446090422445016e-05, "epoch": 2.1446808510638298, "percentage": 71.49, "elapsed_time": "1:49:34", "remaining_time": "0:43:42"}
+{"current_steps": 506, "total_steps": 705, "loss": 0.7144219279289246, "lr": 2.933652760902408e-05, "epoch": 2.153191489361702, "percentage": 71.77, "elapsed_time": "1:50:06", "remaining_time": "0:43:18"}
+{"current_steps": 508, "total_steps": 705, "loss": 0.5674622654914856, "lr": 2.9217370284987434e-05, "epoch": 2.1617021276595745, "percentage": 72.06, "elapsed_time": "1:50:30", "remaining_time": "0:42:51"}
+{"current_steps": 510, "total_steps": 705, "loss": 0.28315988183021545, "lr": 2.908871303963712e-05, "epoch": 2.1702127659574466, "percentage": 72.34, "elapsed_time": "1:50:51", "remaining_time": "0:42:23"}
+{"current_steps": 512, "total_steps": 705, "loss": 0.430463582277298, "lr": 2.8950658003489534e-05, "epoch": 2.178723404255319, "percentage": 72.62, "elapsed_time": "1:51:14", "remaining_time": "0:41:56"}
+{"current_steps": 514, "total_steps": 705, "loss": 0.4114861488342285, "lr": 2.8803314767202387e-05, "epoch": 2.1872340425531913, "percentage": 72.91, "elapsed_time": "1:51:38", "remaining_time": "0:41:29"}
+{"current_steps": 516, "total_steps": 705, "loss": 0.5308094620704651, "lr": 2.864680029457952e-05, "epoch": 2.195744680851064, "percentage": 73.19, "elapsed_time": "1:52:09", "remaining_time": "0:41:05"}
+{"current_steps": 518, "total_steps": 705, "loss": 0.6915931701660156, "lr": 2.8481238829722946e-05, "epoch": 2.204255319148936, "percentage": 73.48, "elapsed_time": "1:52:38", "remaining_time": "0:40:39"}
+{"current_steps": 520, "total_steps": 705, "loss": 0.4150570333003998, "lr": 2.8306761798405533e-05, "epoch": 2.2127659574468086, "percentage": 73.76, "elapsed_time": "1:53:05", "remaining_time": "0:40:14"}
+{"current_steps": 522, "total_steps": 705, "loss": 0.36256328225135803, "lr": 2.812350770374273e-05, "epoch": 2.2212765957446807, "percentage": 74.04, "elapsed_time": "1:53:26", "remaining_time": "0:39:46"}
+{"current_steps": 524, "total_steps": 705, "loss": 0.37502366304397583, "lr": 2.793162201624631e-05, "epoch": 2.2297872340425533, "percentage": 74.33, "elapsed_time": "1:53:43", "remaining_time": "0:39:17"}
+{"current_steps": 526, "total_steps": 705, "loss": 0.7607256770133972, "lr": 2.7731257058347098e-05, "epoch": 2.2382978723404254, "percentage": 74.61, "elapsed_time": "1:54:14", "remaining_time": "0:38:52"}
+{"current_steps": 528, "total_steps": 705, "loss": 0.6797502040863037, "lr": 2.752257188347862e-05, "epoch": 2.246808510638298, "percentage": 74.89, "elapsed_time": "1:54:38", "remaining_time": "0:38:25"}
+{"current_steps": 530, "total_steps": 705, "loss": 0.7978914380073547, "lr": 2.7305732149817512e-05, "epoch": 2.25531914893617, "percentage": 75.18, "elapsed_time": "1:55:07", "remaining_time": "0:38:00"}
+{"current_steps": 532, "total_steps": 705, "loss": 0.5080232620239258, "lr": 2.708090998878098e-05, "epoch": 2.2638297872340427, "percentage": 75.46, "elapsed_time": "1:55:38", "remaining_time": "0:37:36"}
+{"current_steps": 534, "total_steps": 705, "loss": 0.2904958724975586, "lr": 2.68482838683857e-05, "epoch": 2.272340425531915, "percentage": 75.74, "elapsed_time": "1:56:02", "remaining_time": "0:37:09"}
+{"current_steps": 536, "total_steps": 705, "loss": 0.5988547205924988, "lr": 2.6608038451576528e-05, "epoch": 2.2808510638297874, "percentage": 76.03, "elapsed_time": "1:56:36", "remaining_time": "0:36:46"}
+{"current_steps": 538, "total_steps": 705, "loss": 0.8217217922210693, "lr": 2.6360364449637686e-05, "epoch": 2.2893617021276595, "percentage": 76.31, "elapsed_time": "1:57:11", "remaining_time": "0:36:22"}
+{"current_steps": 540, "total_steps": 705, "loss": 0.29454290866851807, "lr": 2.6105458470802563e-05, "epoch": 2.297872340425532, "percentage": 76.6, "elapsed_time": "1:57:30", "remaining_time": "0:35:54"}
+{"current_steps": 542, "total_steps": 705, "loss": 0.2661294937133789, "lr": 2.5843522864182397e-05, "epoch": 2.3063829787234043, "percentage": 76.88, "elapsed_time": "1:57:42", "remaining_time": "0:35:24"}
+{"current_steps": 544, "total_steps": 705, "loss": 0.4138932228088379, "lr": 2.557476555913786e-05, "epoch": 2.3148936170212764, "percentage": 77.16, "elapsed_time": "1:58:00", "remaining_time": "0:34:55"}
+{"current_steps": 546, "total_steps": 705, "loss": 0.47122570872306824, "lr": 2.52993999002208e-05, "epoch": 2.323404255319149, "percentage": 77.45, "elapsed_time": "1:58:22", "remaining_time": "0:34:28"}
+{"current_steps": 548, "total_steps": 705, "loss": 0.5940241813659668, "lr": 2.501764447781743e-05, "epoch": 2.331914893617021, "percentage": 77.73, "elapsed_time": "1:58:45", "remaining_time": "0:34:01"}
+{"current_steps": 550, "total_steps": 705, "loss": 0.443272203207016, "lr": 2.4729722954627106e-05, "epoch": 2.3404255319148937, "percentage": 78.01, "elapsed_time": "1:59:05", "remaining_time": "0:33:33"}
+{"current_steps": 552, "total_steps": 705, "loss": 0.612157940864563, "lr": 2.4435863888114814e-05, "epoch": 2.348936170212766, "percentage": 78.3, "elapsed_time": "1:59:33", "remaining_time": "0:33:08"}
+{"current_steps": 554, "total_steps": 705, "loss": 0.6884116530418396, "lr": 2.413630054907798e-05, "epoch": 2.3574468085106384, "percentage": 78.58, "elapsed_time": "2:00:01", "remaining_time": "0:32:42"}
+{"current_steps": 556, "total_steps": 705, "loss": 0.6459460854530334, "lr": 2.383127073647171e-05, "epoch": 2.3659574468085105, "percentage": 78.87, "elapsed_time": "2:00:31", "remaining_time": "0:32:17"}
+{"current_steps": 558, "total_steps": 705, "loss": 0.5485546588897705, "lr": 2.3521016588639598e-05, "epoch": 2.374468085106383, "percentage": 79.15, "elapsed_time": "2:00:54", "remaining_time": "0:31:51"}
+{"current_steps": 560, "total_steps": 705, "loss": 0.6871886253356934, "lr": 2.32057843910998e-05, "epoch": 2.382978723404255, "percentage": 79.43, "elapsed_time": "2:01:23", "remaining_time": "0:31:25"}
+{"current_steps": 562, "total_steps": 705, "loss": 0.5943357944488525, "lr": 2.2885824381039024e-05, "epoch": 2.391489361702128, "percentage": 79.72, "elapsed_time": "2:01:47", "remaining_time": "0:30:59"}
+{"current_steps": 564, "total_steps": 705, "loss": 0.44520270824432373, "lr": 2.2561390548669552e-05, "epoch": 2.4, "percentage": 80.0, "elapsed_time": "2:02:03", "remaining_time": "0:30:30"}
+{"current_steps": 566, "total_steps": 705, "loss": 0.4427604377269745, "lr": 2.2232740435607067e-05, "epoch": 2.4085106382978725, "percentage": 80.28, "elapsed_time": "2:02:25", "remaining_time": "0:30:04"}
+{"current_steps": 568, "total_steps": 705, "loss": 0.7208576798439026, "lr": 2.1900134930429403e-05, "epoch": 2.4170212765957446, "percentage": 80.57, "elapsed_time": "2:02:51", "remaining_time": "0:29:37"}
+{"current_steps": 570, "total_steps": 705, "loss": 0.6061448454856873, "lr": 2.1563838061578264e-05, "epoch": 2.425531914893617, "percentage": 80.85, "elapsed_time": "2:03:15", "remaining_time": "0:29:11"}
+{"current_steps": 572, "total_steps": 705, "loss": 0.48493388295173645, "lr": 2.122411678776855e-05, "epoch": 2.4340425531914893, "percentage": 81.13, "elapsed_time": "2:03:40", "remaining_time": "0:28:45"}
+{"current_steps": 574, "total_steps": 705, "loss": 0.6101267337799072, "lr": 2.0881240786071588e-05, "epoch": 2.4425531914893615, "percentage": 81.42, "elapsed_time": "2:04:04", "remaining_time": "0:28:18"}
+{"current_steps": 576, "total_steps": 705, "loss": 0.6204816102981567, "lr": 2.0535482237840398e-05, "epoch": 2.451063829787234, "percentage": 81.7, "elapsed_time": "2:04:28", "remaining_time": "0:27:52"}
+{"current_steps": 578, "total_steps": 705, "loss": 0.5858104825019836, "lr": 2.018711561264715e-05, "epoch": 2.4595744680851066, "percentage": 81.99, "elapsed_time": "2:04:56", "remaining_time": "0:27:27"}
+{"current_steps": 580, "total_steps": 705, "loss": 0.5342880487442017, "lr": 1.9836417450403978e-05, "epoch": 2.4680851063829787, "percentage": 82.27, "elapsed_time": "2:05:25", "remaining_time": "0:27:01"}
+{"current_steps": 582, "total_steps": 705, "loss": 0.6852924823760986, "lr": 1.948366614184062e-05, "epoch": 2.476595744680851, "percentage": 82.55, "elapsed_time": "2:05:49", "remaining_time": "0:26:35"}
+{"current_steps": 584, "total_steps": 705, "loss": 0.5905125737190247, "lr": 1.9129141707512514e-05, "epoch": 2.4851063829787234, "percentage": 82.84, "elapsed_time": "2:06:19", "remaining_time": "0:26:10"}
+{"current_steps": 586, "total_steps": 705, "loss": 0.6729872822761536, "lr": 1.877312557551536e-05, "epoch": 2.4936170212765956, "percentage": 83.12, "elapsed_time": "2:06:43", "remaining_time": "0:25:44"}
+{"current_steps": 588, "total_steps": 705, "loss": 0.7322168350219727, "lr": 1.841590035808227e-05, "epoch": 2.502127659574468, "percentage": 83.4, "elapsed_time": "2:07:09", "remaining_time": "0:25:18"}
+{"current_steps": 590, "total_steps": 705, "loss": 0.6285769939422607, "lr": 1.8057749627240822e-05, "epoch": 2.5106382978723403, "percentage": 83.69, "elapsed_time": "2:07:32", "remaining_time": "0:24:51"}
+{"current_steps": 592, "total_steps": 705, "loss": 0.766042172908783, "lr": 1.7698957689708436e-05, "epoch": 2.519148936170213, "percentage": 83.97, "elapsed_time": "2:08:07", "remaining_time": "0:24:27"}
+{"current_steps": 594, "total_steps": 705, "loss": 0.43586933612823486, "lr": 1.7339809361204255e-05, "epoch": 2.527659574468085, "percentage": 84.26, "elapsed_time": "2:08:32", "remaining_time": "0:24:01"}
+{"current_steps": 596, "total_steps": 705, "loss": 0.6923111081123352, "lr": 1.6980589740357287e-05, "epoch": 2.5361702127659576, "percentage": 84.54, "elapsed_time": "2:08:59", "remaining_time": "0:23:35"}
+{"current_steps": 598, "total_steps": 705, "loss": 0.47686851024627686, "lr": 1.6621583982389714e-05, "epoch": 2.5446808510638297, "percentage": 84.82, "elapsed_time": "2:09:23", "remaining_time": "0:23:09"}
+{"current_steps": 600, "total_steps": 705, "loss": 0.5013985633850098, "lr": 1.626307707275533e-05, "epoch": 2.5531914893617023, "percentage": 85.11, "elapsed_time": "2:09:52", "remaining_time": "0:22:43"}
+{"current_steps": 602, "total_steps": 705, "loss": 0.6609531044960022, "lr": 1.59053536009129e-05, "epoch": 2.5617021276595744, "percentage": 85.39, "elapsed_time": "2:10:23", "remaining_time": "0:22:18"}
+{"current_steps": 604, "total_steps": 705, "loss": 0.6927770972251892, "lr": 1.5548697534413653e-05, "epoch": 2.570212765957447, "percentage": 85.67, "elapsed_time": "2:10:51", "remaining_time": "0:21:52"}
+{"current_steps": 606, "total_steps": 705, "loss": 0.8361954689025879, "lr": 1.5193391993482573e-05, "epoch": 2.578723404255319, "percentage": 85.96, "elapsed_time": "2:11:21", "remaining_time": "0:21:27"}
+{"current_steps": 608, "total_steps": 705, "loss": 0.5158759951591492, "lr": 1.483971902627239e-05, "epoch": 2.5872340425531917, "percentage": 86.24, "elapsed_time": "2:11:42", "remaining_time": "0:21:00"}
+{"current_steps": 610, "total_steps": 705, "loss": 0.6141366362571716, "lr": 1.4487959384968272e-05, "epoch": 2.595744680851064, "percentage": 86.52, "elapsed_time": "2:12:08", "remaining_time": "0:20:34"}
+{"current_steps": 612, "total_steps": 705, "loss": 0.669286847114563, "lr": 1.4138392302921813e-05, "epoch": 2.604255319148936, "percentage": 86.81, "elapsed_time": "2:12:38", "remaining_time": "0:20:09"}
+{"current_steps": 614, "total_steps": 705, "loss": 0.8155961036682129, "lr": 1.3791295272990178e-05, "epoch": 2.6127659574468085, "percentage": 87.09, "elapsed_time": "2:13:10", "remaining_time": "0:19:44"}
+{"current_steps": 616, "total_steps": 705, "loss": 0.6486954689025879, "lr": 1.344694382725719e-05, "epoch": 2.621276595744681, "percentage": 87.38, "elapsed_time": "2:13:39", "remaining_time": "0:19:18"}
+{"current_steps": 618, "total_steps": 705, "loss": 0.7624613046646118, "lr": 1.3105611318310828e-05, "epoch": 2.629787234042553, "percentage": 87.66, "elapsed_time": "2:14:07", "remaining_time": "0:18:52"}
+{"current_steps": 620, "total_steps": 705, "loss": 0.44974949955940247, "lr": 1.2767568702250834e-05, "epoch": 2.6382978723404253, "percentage": 87.94, "elapsed_time": "2:14:31", "remaining_time": "0:18:26"}
+{"current_steps": 622, "total_steps": 705, "loss": 0.5666506886482239, "lr": 1.2433084323598806e-05, "epoch": 2.646808510638298, "percentage": 88.23, "elapsed_time": "2:14:54", "remaining_time": "0:18:00"}
+{"current_steps": 624, "total_steps": 705, "loss": 0.7800987362861633, "lr": 1.2102423702281114e-05, "epoch": 2.65531914893617, "percentage": 88.51, "elapsed_time": "2:15:24", "remaining_time": "0:17:34"}
+{"current_steps": 626, "total_steps": 705, "loss": 0.5514877438545227, "lr": 1.1775849322854515e-05, "epoch": 2.6638297872340426, "percentage": 88.79, "elapsed_time": "2:15:50", "remaining_time": "0:17:08"}
+{"current_steps": 628, "total_steps": 705, "loss": 0.554164469242096, "lr": 1.1453620426140796e-05, "epoch": 2.6723404255319148, "percentage": 89.08, "elapsed_time": "2:16:11", "remaining_time": "0:16:41"}
+{"current_steps": 630, "total_steps": 705, "loss": 0.5563373565673828, "lr": 1.1135992803436701e-05, "epoch": 2.6808510638297873, "percentage": 89.36, "elapsed_time": "2:16:33", "remaining_time": "0:16:15"}
+{"current_steps": 632, "total_steps": 705, "loss": 0.5627699494361877, "lr": 1.0823218593462006e-05, "epoch": 2.6893617021276595, "percentage": 89.65, "elapsed_time": "2:16:59", "remaining_time": "0:15:49"}
+{"current_steps": 634, "total_steps": 705, "loss": 0.698486328125, "lr": 1.0515546082207089e-05, "epoch": 2.697872340425532, "percentage": 89.93, "elapsed_time": "2:17:27", "remaining_time": "0:15:23"}
+{"current_steps": 636, "total_steps": 705, "loss": 0.7639659643173218, "lr": 1.0213219505839e-05, "epoch": 2.706382978723404, "percentage": 90.21, "elapsed_time": "2:18:00", "remaining_time": "0:14:58"}
+{"current_steps": 638, "total_steps": 705, "loss": 0.6158884763717651, "lr": 9.916478856822011e-06, "epoch": 2.7148936170212767, "percentage": 90.5, "elapsed_time": "2:18:30", "remaining_time": "0:14:32"}
+{"current_steps": 640, "total_steps": 705, "loss": 0.9152920842170715, "lr": 9.625559693407416e-06, "epoch": 2.723404255319149, "percentage": 90.78, "elapsed_time": "2:19:06", "remaining_time": "0:14:07"}
+{"current_steps": 642, "total_steps": 705, "loss": 0.8749684691429138, "lr": 9.34069295264279e-06, "epoch": 2.731914893617021, "percentage": 91.06, "elapsed_time": "2:19:40", "remaining_time": "0:13:42"}
+{"current_steps": 644, "total_steps": 705, "loss": 0.8344287872314453, "lr": 9.062104767049962e-06, "epoch": 2.7404255319148936, "percentage": 91.35, "elapsed_time": "2:20:12", "remaining_time": "0:13:16"}
+{"current_steps": 646, "total_steps": 705, "loss": 0.45629745721817017, "lr": 8.790016285116773e-06, "epoch": 2.748936170212766, "percentage": 91.63, "elapsed_time": "2:20:31", "remaining_time": "0:12:50"}
+{"current_steps": 648, "total_steps": 705, "loss": 0.3841492533683777, "lr": 8.5246434957453e-06, "epoch": 2.7574468085106383, "percentage": 91.91, "elapsed_time": "2:20:48", "remaining_time": "0:12:23"}
+{"current_steps": 650, "total_steps": 705, "loss": 0.4197019338607788, "lr": 8.266197056795886e-06, "epoch": 2.7659574468085104, "percentage": 92.2, "elapsed_time": "2:21:09", "remaining_time": "0:11:56"}
+{"current_steps": 652, "total_steps": 705, "loss": 0.6872882843017578, "lr": 8.014882127862926e-06, "epoch": 2.774468085106383, "percentage": 92.48, "elapsed_time": "2:21:36", "remaining_time": "0:11:30"}
+{"current_steps": 654, "total_steps": 705, "loss": 0.6434124112129211, "lr": 7.77089820741542e-06, "epoch": 2.7829787234042556, "percentage": 92.77, "elapsed_time": "2:22:09", "remaining_time": "0:11:05"}
+{"current_steps": 656, "total_steps": 705, "loss": 0.5578271150588989, "lr": 7.534438974431356e-06, "epoch": 2.7914893617021277, "percentage": 93.05, "elapsed_time": "2:22:32", "remaining_time": "0:10:38"}
+{"current_steps": 658, "total_steps": 705, "loss": 0.6800917983055115, "lr": 7.305692134651748e-06, "epoch": 2.8, "percentage": 93.33, "elapsed_time": "2:23:06", "remaining_time": "0:10:13"}
+{"current_steps": 660, "total_steps": 705, "loss": 0.4372290074825287, "lr": 7.0848392715763e-06, "epoch": 2.8085106382978724, "percentage": 93.62, "elapsed_time": "2:23:35", "remaining_time": "0:09:47"}
+{"current_steps": 662, "total_steps": 705, "loss": 0.3426898121833801, "lr": 6.872055702319048e-06, "epoch": 2.8170212765957445, "percentage": 93.9, "elapsed_time": "2:23:50", "remaining_time": "0:09:20"}
+{"current_steps": 664, "total_steps": 705, "loss": 0.5213098526000977, "lr": 6.66751033843842e-06, "epoch": 2.825531914893617, "percentage": 94.18, "elapsed_time": "2:24:14", "remaining_time": "0:08:54"}
+{"current_steps": 666, "total_steps": 705, "loss": 0.40890708565711975, "lr": 6.471365551852014e-06, "epoch": 2.8340425531914892, "percentage": 94.47, "elapsed_time": "2:24:31", "remaining_time": "0:08:27"}
+{"current_steps": 668, "total_steps": 705, "loss": 0.6171141266822815, "lr": 6.283777045942801e-06, "epoch": 2.842553191489362, "percentage": 94.75, "elapsed_time": "2:24:57", "remaining_time": "0:08:01"}
+{"current_steps": 670, "total_steps": 705, "loss": 0.79531329870224, "lr": 6.104893731958872e-06, "epoch": 2.851063829787234, "percentage": 95.04, "elapsed_time": "2:25:24", "remaining_time": "0:07:35"}
+{"current_steps": 672, "total_steps": 705, "loss": 0.6109257936477661, "lr": 5.934857610804912e-06, "epoch": 2.8595744680851065, "percentage": 95.32, "elapsed_time": "2:25:51", "remaining_time": "0:07:09"}
+{"current_steps": 674, "total_steps": 705, "loss": 0.7840122580528259, "lr": 5.773803660319231e-06, "epoch": 2.8680851063829786, "percentage": 95.6, "elapsed_time": "2:26:24", "remaining_time": "0:06:44"}
+{"current_steps": 676, "total_steps": 705, "loss": 0.4490019679069519, "lr": 5.6218597281258834e-06, "epoch": 2.876595744680851, "percentage": 95.89, "elapsed_time": "2:26:50", "remaining_time": "0:06:17"}
+{"current_steps": 678, "total_steps": 705, "loss": 0.5424228310585022, "lr": 5.479146430146783e-06, "epoch": 2.8851063829787233, "percentage": 96.17, "elapsed_time": "2:27:18", "remaining_time": "0:05:51"}
+{"current_steps": 680, "total_steps": 705, "loss": 0.5151770114898682, "lr": 5.3457770548545805e-06, "epoch": 2.8936170212765955, "percentage": 96.45, "elapsed_time": "2:27:47", "remaining_time": "0:05:26"}
+{"current_steps": 682, "total_steps": 705, "loss": 0.7383885979652405, "lr": 5.22185747334215e-06, "epoch": 2.902127659574468, "percentage": 96.74, "elapsed_time": "2:28:13", "remaining_time": "0:04:59"}
+{"current_steps": 684, "total_steps": 705, "loss": 0.4885507822036743, "lr": 5.107486055280145e-06, "epoch": 2.9106382978723406, "percentage": 97.02, "elapsed_time": "2:28:31", "remaining_time": "0:04:33"}
+{"current_steps": 686, "total_steps": 705, "loss": 0.29081299901008606, "lr": 5.002753590829352e-06, "epoch": 2.9191489361702128, "percentage": 97.3, "elapsed_time": "2:28:45", "remaining_time": "0:04:07"}
+{"current_steps": 688, "total_steps": 705, "loss": 0.436556339263916, "lr": 4.9077432185697e-06, "epoch": 2.927659574468085, "percentage": 97.59, "elapsed_time": "2:29:05", "remaining_time": "0:03:41"}
+{"current_steps": 690, "total_steps": 705, "loss": 0.759993314743042, "lr": 4.822530359503391e-06, "epoch": 2.9361702127659575, "percentage": 97.87, "elapsed_time": "2:29:37", "remaining_time": "0:03:15"}
+{"current_steps": 692, "total_steps": 705, "loss": 0.6529436111450195, "lr": 4.747182657184251e-06, "epoch": 2.94468085106383, "percentage": 98.16, "elapsed_time": "2:30:06", "remaining_time": "0:02:49"}
+{"current_steps": 694, "total_steps": 705, "loss": 0.6622499227523804, "lr": 4.681759924021033e-06, "epoch": 2.953191489361702, "percentage": 98.44, "elapsed_time": "2:30:37", "remaining_time": "0:02:23"}
+{"current_steps": 696, "total_steps": 705, "loss": 0.5456458926200867, "lr": 4.626314093797213e-06, "epoch": 2.9617021276595743, "percentage": 98.72, "elapsed_time": "2:31:05", "remaining_time": "0:01:57"}
+{"current_steps": 698, "total_steps": 705, "loss": 0.738301157951355, "lr": 4.580889180444988e-06, "epoch": 2.970212765957447, "percentage": 99.01, "elapsed_time": "2:31:41", "remaining_time": "0:01:31"}
+{"current_steps": 700, "total_steps": 705, "loss": 0.4914497435092926, "lr": 4.5455212431061985e-06, "epoch": 2.978723404255319, "percentage": 99.29, "elapsed_time": "2:32:05", "remaining_time": "0:01:05"}
+{"current_steps": 702, "total_steps": 705, "loss": 0.5938633680343628, "lr": 4.520238357507899e-06, "epoch": 2.9872340425531916, "percentage": 99.57, "elapsed_time": "2:32:30", "remaining_time": "0:00:39"}
+{"current_steps": 704, "total_steps": 705, "loss": 0.4166713058948517, "lr": 4.505060593675342e-06, "epoch": 2.9957446808510637, "percentage": 99.86, "elapsed_time": "2:32:55", "remaining_time": "0:00:13"}
+{"current_steps": 705, "total_steps": 705, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "2:33:45", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,2507 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 705,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00851063829787234,
+      "grad_norm": 0.9370151162147522,
+      "learning_rate": 8.333333333333333e-07,
+      "loss": 4.115976333618164,
+      "step": 2
+    },
+    {
+      "epoch": 0.01702127659574468,
+      "grad_norm": 0.41166678071022034,
+      "learning_rate": 2.4999999999999998e-06,
+      "loss": 1.878818154335022,
+      "step": 4
+    },
+    {
+      "epoch": 0.02553191489361702,
+      "grad_norm": 0.4669341444969177,
+      "learning_rate": 4.166666666666667e-06,
+      "loss": 2.0509164333343506,
+      "step": 6
+    },
+    {
+      "epoch": 0.03404255319148936,
+      "grad_norm": 0.6062728762626648,
+      "learning_rate": 5.833333333333334e-06,
+      "loss": 1.7545194625854492,
+      "step": 8
+    },
+    {
+      "epoch": 0.0425531914893617,
+      "grad_norm": 0.37637069821357727,
+      "learning_rate": 7.5e-06,
+      "loss": 1.7690430879592896,
+      "step": 10
+    },
+    {
+      "epoch": 0.05106382978723404,
+      "grad_norm": 2.065159797668457,
+      "learning_rate": 9.166666666666668e-06,
+      "loss": 2.355275869369507,
+      "step": 12
+    },
+    {
+      "epoch": 0.059574468085106386,
+      "grad_norm": 1.091937780380249,
+      "learning_rate": 1.0833333333333334e-05,
+      "loss": 1.876306176185608,
+      "step": 14
+    },
+    {
+      "epoch": 0.06808510638297872,
+      "grad_norm": 0.49564921855926514,
+      "learning_rate": 1.25e-05,
+      "loss": 1.785620927810669,
+      "step": 16
+    },
+    {
+      "epoch": 0.07659574468085106,
+      "grad_norm": 1.3353960514068604,
+      "learning_rate": 1.4166666666666666e-05,
+      "loss": 1.523728370666504,
+      "step": 18
+    },
+    {
+      "epoch": 0.0851063829787234,
+      "grad_norm": 0.30626556277275085,
+      "learning_rate": 1.5833333333333333e-05,
+      "loss": 1.510640025138855,
+      "step": 20
+    },
+    {
+      "epoch": 0.09361702127659574,
+      "grad_norm": 0.1502075493335724,
+      "learning_rate": 1.7500000000000002e-05,
+      "loss": 1.4279940128326416,
+      "step": 22
+    },
+    {
+      "epoch": 0.10212765957446808,
+      "grad_norm": 0.8587890863418579,
+      "learning_rate": 1.9166666666666667e-05,
+      "loss": 1.1794553995132446,
+      "step": 24
+    },
+    {
+      "epoch": 0.11063829787234042,
+      "grad_norm": 1.0492326021194458,
+      "learning_rate": 2.0833333333333333e-05,
+      "loss": 1.2579714059829712,
+      "step": 26
+    },
+    {
+      "epoch": 0.11914893617021277,
+      "grad_norm": 0.264726847410202,
+      "learning_rate": 2.25e-05,
+      "loss": 0.9510668516159058,
+      "step": 28
+    },
+    {
+      "epoch": 0.1276595744680851,
+      "grad_norm": 0.2859492301940918,
+      "learning_rate": 2.4166666666666667e-05,
+      "loss": 1.656806230545044,
+      "step": 30
+    },
+    {
+      "epoch": 0.13617021276595745,
+      "grad_norm": 1.1864591836929321,
+      "learning_rate": 2.5833333333333336e-05,
+      "loss": 0.683425784111023,
+      "step": 32
+    },
+    {
+      "epoch": 0.14468085106382977,
+      "grad_norm": 0.22737205028533936,
+      "learning_rate": 2.75e-05,
+      "loss": 1.2381809949874878,
+      "step": 34
+    },
+    {
+      "epoch": 0.15319148936170213,
+      "grad_norm": 0.32182246446609497,
+      "learning_rate": 2.9166666666666666e-05,
+      "loss": 1.422176480293274,
+      "step": 36
+    },
+    {
+      "epoch": 0.16170212765957448,
+      "grad_norm": 0.36561885476112366,
+      "learning_rate": 2.9998734788806287e-05,
+      "loss": 1.2621259689331055,
+      "step": 38
+    },
+    {
+      "epoch": 0.1702127659574468,
+      "grad_norm": 0.7920733690261841,
+      "learning_rate": 2.9988614605803806e-05,
+      "loss": 1.2012425661087036,
+      "step": 40
+    },
+    {
+      "epoch": 0.17872340425531916,
+      "grad_norm": 0.19826872646808624,
+      "learning_rate": 2.99683822733885e-05,
+      "loss": 1.2804166078567505,
+      "step": 42
+    },
+    {
+      "epoch": 0.18723404255319148,
+      "grad_norm": 0.5331127643585205,
+      "learning_rate": 2.9938053852362484e-05,
+      "loss": 1.1188151836395264,
+      "step": 44
+    },
+    {
+      "epoch": 0.19574468085106383,
+      "grad_norm": 0.22702525556087494,
+      "learning_rate": 2.989765341799095e-05,
+      "loss": 1.338016152381897,
+      "step": 46
+    },
+    {
+      "epoch": 0.20425531914893616,
+      "grad_norm": 1.6932013034820557,
+      "learning_rate": 2.9847213040890793e-05,
+      "loss": 0.8224107027053833,
+      "step": 48
+    },
+    {
+      "epoch": 0.2127659574468085,
+      "grad_norm": 0.49766504764556885,
+      "learning_rate": 2.9786772761572335e-05,
+      "loss": 0.9487060308456421,
+      "step": 50
+    },
+    {
+      "epoch": 0.22127659574468084,
+      "grad_norm": 0.20401130616664886,
+      "learning_rate": 2.9716380558654445e-05,
+      "loss": 1.3093581199645996,
+      "step": 52
+    },
+    {
+      "epoch": 0.2297872340425532,
+      "grad_norm": 0.4012207090854645,
+      "learning_rate": 2.9636092310778195e-05,
+      "loss": 1.2255353927612305,
+      "step": 54
+    },
+    {
+      "epoch": 0.23829787234042554,
+      "grad_norm": 0.8393444418907166,
+      "learning_rate": 2.954597175224938e-05,
+      "loss": 0.8253700137138367,
+      "step": 56
+    },
+    {
+      "epoch": 0.24680851063829787,
+      "grad_norm": 0.27147796750068665,
+      "learning_rate": 2.9446090422445016e-05,
+      "loss": 1.0494465827941895,
+      "step": 58
+    },
+    {
+      "epoch": 0.2553191489361702,
+      "grad_norm": 0.42597004771232605,
+      "learning_rate": 2.9336527609024072e-05,
+      "loss": 1.358299970626831,
+      "step": 60
+    },
+    {
+      "epoch": 0.26382978723404255,
+      "grad_norm": 0.20699705183506012,
+      "learning_rate": 2.9217370284987434e-05,
+      "loss": 1.335315465927124,
+      "step": 62
+    },
+    {
+      "epoch": 0.2723404255319149,
+      "grad_norm": 0.16389895975589752,
+      "learning_rate": 2.9088713039637117e-05,
+      "loss": 1.2662416696548462,
+      "step": 64
+    },
+    {
+      "epoch": 0.28085106382978725,
+      "grad_norm": 0.25043565034866333,
+      "learning_rate": 2.8950658003489534e-05,
+      "loss": 1.2614431381225586,
+      "step": 66
+    },
+    {
+      "epoch": 0.28936170212765955,
+      "grad_norm": 0.19853799045085907,
+      "learning_rate": 2.880331476720238e-05,
+      "loss": 1.2736550569534302,
+      "step": 68
+    },
+    {
+      "epoch": 0.2978723404255319,
+      "grad_norm": 2.3273322582244873,
+      "learning_rate": 2.8646800294579517e-05,
+      "loss": 1.2623916864395142,
+      "step": 70
+    },
+    {
+      "epoch": 0.30638297872340425,
+      "grad_norm": 0.25601401925086975,
+      "learning_rate": 2.848123882972295e-05,
+      "loss": 1.2526882886886597,
+      "step": 72
+    },
+    {
+      "epoch": 0.3148936170212766,
+      "grad_norm": 0.342012494802475,
+      "learning_rate": 2.8306761798405526e-05,
+      "loss": 0.9114877581596375,
+      "step": 74
+    },
+    {
+      "epoch": 0.32340425531914896,
+      "grad_norm": 0.5187206864356995,
+      "learning_rate": 2.812350770374273e-05,
+      "loss": 1.2244638204574585,
+      "step": 76
+    },
+    {
+      "epoch": 0.33191489361702126,
+      "grad_norm": 0.2703041434288025,
+      "learning_rate": 2.793162201624631e-05,
+      "loss": 1.2725831270217896,
+      "step": 78
+    },
+    {
+      "epoch": 0.3404255319148936,
+      "grad_norm": 0.1924225091934204,
+      "learning_rate": 2.77312570583471e-05,
+      "loss": 0.7258127927780151,
+      "step": 80
+    },
+    {
+      "epoch": 0.34893617021276596,
+      "grad_norm": 0.25936806201934814,
+      "learning_rate": 2.752257188347862e-05,
+      "loss": 1.2518136501312256,
+      "step": 82
+    },
+    {
+      "epoch": 0.3574468085106383,
+      "grad_norm": 0.1409609615802765,
+      "learning_rate": 2.730573214981751e-05,
+      "loss": 0.9943649172782898,
+      "step": 84
+    },
+    {
+      "epoch": 0.3659574468085106,
+      "grad_norm": 0.18737079203128815,
+      "learning_rate": 2.7080909988780982e-05,
+      "loss": 1.001371145248413,
+      "step": 86
+    },
+    {
+      "epoch": 0.37446808510638296,
+      "grad_norm": 0.512141227722168,
+      "learning_rate": 2.684828386838569e-05,
+      "loss": 1.2457740306854248,
+      "step": 88
+    },
+    {
+      "epoch": 0.3829787234042553,
+      "grad_norm": 0.7956476807594299,
+      "learning_rate": 2.6608038451576528e-05,
+      "loss": 0.9049152135848999,
+      "step": 90
+    },
+    {
+      "epoch": 0.39148936170212767,
+      "grad_norm": 0.17900533974170685,
+      "learning_rate": 2.636036444963769e-05,
+      "loss": 1.0495070219039917,
+      "step": 92
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.28203219175338745,
+      "learning_rate": 2.6105458470802563e-05,
+      "loss": 1.163374900817871,
+      "step": 94
+    },
+    {
+      "epoch": 0.4085106382978723,
+      "grad_norm": 0.2998636066913605,
+      "learning_rate": 2.5843522864182394e-05,
+      "loss": 1.064250111579895,
+      "step": 96
+    },
+    {
+      "epoch": 0.41702127659574467,
+      "grad_norm": 0.34955132007598877,
+      "learning_rate": 2.557476555913785e-05,
+      "loss": 1.0862053632736206,
+      "step": 98
+    },
+    {
+      "epoch": 0.425531914893617,
+      "grad_norm": 0.5751879215240479,
+      "learning_rate": 2.5299399900220803e-05,
+      "loss": 0.9672101736068726,
+      "step": 100
+    },
+    {
+      "epoch": 0.4340425531914894,
+      "grad_norm": 0.556722104549408,
+      "learning_rate": 2.5017644477817424e-05,
+      "loss": 1.38407564163208,
+      "step": 102
+    },
+    {
+      "epoch": 0.4425531914893617,
+      "grad_norm": 0.24831920862197876,
+      "learning_rate": 2.47297229546271e-05,
+      "loss": 1.0887106657028198,
+      "step": 104
+    },
+    {
+      "epoch": 0.451063829787234,
+      "grad_norm": 0.39581331610679626,
+      "learning_rate": 2.4435863888114814e-05,
+      "loss": 0.8722383975982666,
+      "step": 106
+    },
+    {
+      "epoch": 0.4595744680851064,
+      "grad_norm": 0.16548699140548706,
+      "learning_rate": 2.4136300549077976e-05,
+      "loss": 1.2673208713531494,
+      "step": 108
+    },
+    {
+      "epoch": 0.46808510638297873,
+      "grad_norm": 0.20999404788017273,
+      "learning_rate": 2.3831270736471703e-05,
+      "loss": 1.2440522909164429,
+      "step": 110
+    },
+    {
+      "epoch": 0.4765957446808511,
+      "grad_norm": 0.18370361626148224,
+      "learning_rate": 2.352101658863959e-05,
+      "loss": 1.1419363021850586,
+      "step": 112
+    },
+    {
+      "epoch": 0.4851063829787234,
+      "grad_norm": 0.24067164957523346,
+      "learning_rate": 2.32057843910998e-05,
+      "loss": 1.0633783340454102,
+      "step": 114
+    },
+    {
+      "epoch": 0.49361702127659574,
+      "grad_norm": 0.47072628140449524,
+      "learning_rate": 2.288582438103903e-05,
+      "loss": 1.179775595664978,
+      "step": 116
+    },
+    {
+      "epoch": 0.502127659574468,
+      "grad_norm": 0.2161823809146881,
+      "learning_rate": 2.256139054866955e-05,
+      "loss": 1.2002534866333008,
+      "step": 118
+    },
+    {
+      "epoch": 0.5106382978723404,
+      "grad_norm": 1.2126752138137817,
+      "learning_rate": 2.2232740435607067e-05,
+      "loss": 0.9621443748474121,
+      "step": 120
+    },
+    {
+      "epoch": 0.5191489361702127,
+      "grad_norm": 0.1435491442680359,
+      "learning_rate": 2.19001349304294e-05,
+      "loss": 1.253274917602539,
+      "step": 122
+    },
+    {
+      "epoch": 0.5276595744680851,
+      "grad_norm": 0.19416141510009766,
+      "learning_rate": 2.156383806157826e-05,
+      "loss": 1.3183720111846924,
+      "step": 124
+    },
+    {
+      "epoch": 0.5361702127659574,
+      "grad_norm": 0.14461560547351837,
+      "learning_rate": 2.1224116787768552e-05,
+      "loss": 1.238662838935852,
+      "step": 126
+    },
+    {
+      "epoch": 0.5446808510638298,
+      "grad_norm": 0.155701145529747,
+      "learning_rate": 2.0881240786071588e-05,
+      "loss": 1.1723605394363403,
+      "step": 128
+    },
+    {
+      "epoch": 0.5531914893617021,
+      "grad_norm": 0.17010867595672607,
+      "learning_rate": 2.05354822378404e-05,
+      "loss": 1.021695852279663,
+      "step": 130
+    },
+    {
+      "epoch": 0.5617021276595745,
+      "grad_norm": 0.17941494286060333,
+      "learning_rate": 2.018711561264714e-05,
+      "loss": 1.2407881021499634,
+      "step": 132
+    },
+    {
+      "epoch": 0.5702127659574469,
+      "grad_norm": 0.3246072828769684,
+      "learning_rate": 1.9836417450403978e-05,
+      "loss": 1.186881422996521,
+      "step": 134
+    },
+    {
+      "epoch": 0.5787234042553191,
+      "grad_norm": 0.32305997610092163,
+      "learning_rate": 1.9483666141840615e-05,
+      "loss": 0.965923547744751,
+      "step": 136
+    },
+    {
+      "epoch": 0.5872340425531914,
+      "grad_norm": 0.15171490609645844,
+      "learning_rate": 1.9129141707512508e-05,
+      "loss": 0.5913432836532593,
+      "step": 138
+    },
+    {
+      "epoch": 0.5957446808510638,
+      "grad_norm": 0.37349367141723633,
+      "learning_rate": 1.8773125575515364e-05,
+      "loss": 1.1055043935775757,
+      "step": 140
+    },
+    {
+      "epoch": 0.6042553191489362,
+      "grad_norm": 0.15140922367572784,
+      "learning_rate": 1.8415900358082268e-05,
+      "loss": 1.3598577976226807,
+      "step": 142
+    },
+    {
+      "epoch": 0.6127659574468085,
+      "grad_norm": 0.5196365118026733,
+      "learning_rate": 1.805774962724083e-05,
+      "loss": 0.9218302369117737,
+      "step": 144
+    },
+    {
+      "epoch": 0.6212765957446809,
+      "grad_norm": 0.15164780616760254,
+      "learning_rate": 1.7698957689708426e-05,
+      "loss": 1.1118155717849731,
+      "step": 146
+    },
+    {
+      "epoch": 0.6297872340425532,
+      "grad_norm": 0.2984839677810669,
+      "learning_rate": 1.7339809361204252e-05,
+      "loss": 1.2171759605407715,
+      "step": 148
+    },
+    {
+      "epoch": 0.6382978723404256,
+      "grad_norm": 0.28176164627075195,
+      "learning_rate": 1.6980589740357294e-05,
+      "loss": 1.1775646209716797,
+      "step": 150
+    },
+    {
+      "epoch": 0.6468085106382979,
+      "grad_norm": 0.19051885604858398,
+      "learning_rate": 1.6621583982389707e-05,
+      "loss": 1.1114767789840698,
+      "step": 152
+    },
+    {
+      "epoch": 0.6553191489361702,
+      "grad_norm": 0.28726738691329956,
+      "learning_rate": 1.6263077072755326e-05,
+      "loss": 0.7064566612243652,
+      "step": 154
+    },
+    {
+      "epoch": 0.6638297872340425,
+      "grad_norm": 0.3889693021774292,
+      "learning_rate": 1.5905353600912898e-05,
+      "loss": 0.8802153468132019,
+      "step": 156
+    },
+    {
+      "epoch": 0.6723404255319149,
+      "grad_norm": 0.28420913219451904,
+      "learning_rate": 1.5548697534413646e-05,
+      "loss": 0.9915045499801636,
+      "step": 158
+    },
+    {
+      "epoch": 0.6808510638297872,
+      "grad_norm": 0.2218930721282959,
+      "learning_rate": 1.5193391993482582e-05,
+      "loss": 1.2188233137130737,
+      "step": 160
+    },
+    {
+      "epoch": 0.6893617021276596,
+      "grad_norm": 0.1267762929201126,
+      "learning_rate": 1.4839719026272377e-05,
+      "loss": 1.0931644439697266,
+      "step": 162
+    },
+    {
+      "epoch": 0.6978723404255319,
+      "grad_norm": 0.20958174765110016,
+      "learning_rate": 1.4487959384968272e-05,
+      "loss": 0.875884473323822,
+      "step": 164
+    },
+    {
+      "epoch": 0.7063829787234043,
+      "grad_norm": 0.2075873166322708,
+      "learning_rate": 1.4138392302921813e-05,
+      "loss": 0.9931239485740662,
+      "step": 166
+    },
+    {
+      "epoch": 0.7148936170212766,
+      "grad_norm": 0.45486778020858765,
+      "learning_rate": 1.3791295272990175e-05,
+      "loss": 1.1112878322601318,
+      "step": 168
+    },
+    {
+      "epoch": 0.723404255319149,
+      "grad_norm": 0.12606321275234222,
+      "learning_rate": 1.344694382725718e-05,
+      "loss": 0.8394916653633118,
+      "step": 170
+    },
+    {
+      "epoch": 0.7319148936170212,
+      "grad_norm": 1.405964732170105,
+      "learning_rate": 1.3105611318310818e-05,
+      "loss": 0.8859961032867432,
+      "step": 172
+    },
+    {
+      "epoch": 0.7404255319148936,
+      "grad_norm": 0.24160844087600708,
+      "learning_rate": 1.2767568702250844e-05,
+      "loss": 1.1839947700500488,
+      "step": 174
+    },
+    {
+      "epoch": 0.7489361702127659,
+      "grad_norm": 0.28286510705947876,
+      "learning_rate": 1.2433084323598791e-05,
+      "loss": 0.6160851120948792,
+      "step": 176
+    },
+    {
+      "epoch": 0.7574468085106383,
+      "grad_norm": 0.36609017848968506,
+      "learning_rate": 1.2102423702281116e-05,
+      "loss": 1.0064830780029297,
+      "step": 178
+    },
+    {
+      "epoch": 0.7659574468085106,
+      "grad_norm": 0.25650474429130554,
+      "learning_rate": 1.1775849322854516e-05,
+      "loss": 1.0609067678451538,
+      "step": 180
+    },
+    {
+      "epoch": 0.774468085106383,
+      "grad_norm": 0.3248478174209595,
+      "learning_rate": 1.1453620426140795e-05,
+      "loss": 0.7533571720123291,
+      "step": 182
+    },
+    {
+      "epoch": 0.7829787234042553,
+      "grad_norm": 0.22153323888778687,
+      "learning_rate": 1.1135992803436695e-05,
+      "loss": 0.9635902643203735,
+      "step": 184
+    },
+    {
+      "epoch": 0.7914893617021277,
+      "grad_norm": 0.11977202445268631,
+      "learning_rate": 1.0823218593461992e-05,
+      "loss": 0.9765074849128723,
+      "step": 186
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.2166285663843155,
+      "learning_rate": 1.0515546082207097e-05,
+      "loss": 0.730889081954956,
+      "step": 188
+    },
+    {
+      "epoch": 0.8085106382978723,
+      "grad_norm": 0.6851885914802551,
+      "learning_rate": 1.0213219505838983e-05,
+      "loss": 0.966530442237854,
+      "step": 190
+    },
+    {
+      "epoch": 0.8170212765957446,
+      "grad_norm": 0.33713918924331665,
+      "learning_rate": 9.91647885682201e-06,
+      "loss": 1.3054033517837524,
+      "step": 192
+    },
+    {
+      "epoch": 0.825531914893617,
+      "grad_norm": 0.196068674325943,
+      "learning_rate": 9.625559693407413e-06,
+      "loss": 1.2358617782592773,
+      "step": 194
+    },
+    {
+      "epoch": 0.8340425531914893,
+      "grad_norm": 0.2631889283657074,
+      "learning_rate": 9.340692952642789e-06,
+      "loss": 0.789035975933075,
+      "step": 196
+    },
+    {
+      "epoch": 0.8425531914893617,
+      "grad_norm": 0.165288507938385,
+      "learning_rate": 9.062104767049956e-06,
+      "loss": 1.2202699184417725,
+      "step": 198
+    },
+    {
+      "epoch": 0.851063829787234,
+      "grad_norm": 0.4081096649169922,
+      "learning_rate": 8.790016285116763e-06,
+      "loss": 1.005751132965088,
+      "step": 200
+    },
+    {
+      "epoch": 0.8595744680851064,
+      "grad_norm": 0.16539311408996582,
+      "learning_rate": 8.524643495745306e-06,
+      "loss": 0.9970273971557617,
+      "step": 202
+    },
+    {
+      "epoch": 0.8680851063829788,
+      "grad_norm": 0.1629868745803833,
+      "learning_rate": 8.26619705679589e-06,
+      "loss": 1.1567330360412598,
+      "step": 204
+    },
+    {
+      "epoch": 0.8765957446808511,
+      "grad_norm": 0.13489580154418945,
+      "learning_rate": 8.014882127862923e-06,
+      "loss": 1.1192888021469116,
+      "step": 206
+    },
+    {
+      "epoch": 0.8851063829787233,
+      "grad_norm": 0.1189538910984993,
+      "learning_rate": 7.770898207415416e-06,
+      "loss": 0.9340039491653442,
+      "step": 208
+    },
+    {
+      "epoch": 0.8936170212765957,
+      "grad_norm": 0.12542982399463654,
+      "learning_rate": 7.534438974431351e-06,
+      "loss": 1.2615220546722412,
+      "step": 210
+    },
+    {
+      "epoch": 0.902127659574468,
+      "grad_norm": 0.2067514955997467,
+      "learning_rate": 7.305692134651742e-06,
+      "loss": 0.7793570160865784,
+      "step": 212
+    },
+    {
+      "epoch": 0.9106382978723404,
+      "grad_norm": 0.12745940685272217,
+      "learning_rate": 7.084839271576291e-06,
+      "loss": 0.9480925798416138,
+      "step": 214
+    },
+    {
+      "epoch": 0.9191489361702128,
+      "grad_norm": 0.14118105173110962,
+      "learning_rate": 6.872055702319054e-06,
+      "loss": 0.7321507334709167,
+      "step": 216
+    },
+    {
+      "epoch": 0.9276595744680851,
+      "grad_norm": 0.19066962599754333,
+      "learning_rate": 6.667510338438419e-06,
+      "loss": 1.2155194282531738,
+      "step": 218
+    },
+    {
+      "epoch": 0.9361702127659575,
+      "grad_norm": 0.2428826540708542,
+      "learning_rate": 6.471365551852012e-06,
+      "loss": 0.39672887325286865,
+      "step": 220
+    },
+    {
+      "epoch": 0.9446808510638298,
+      "grad_norm": 0.16802483797073364,
+      "learning_rate": 6.2837770459428e-06,
+      "loss": 1.254830241203308,
+      "step": 222
+    },
+    {
+      "epoch": 0.9531914893617022,
+      "grad_norm": 0.1394779235124588,
+      "learning_rate": 6.1048937319588676e-06,
+      "loss": 1.138830304145813,
+      "step": 224
+    },
+    {
+      "epoch": 0.9617021276595744,
+      "grad_norm": 0.2871001362800598,
+      "learning_rate": 5.9348576108049065e-06,
+      "loss": 0.8572854399681091,
+      "step": 226
+    },
+    {
+      "epoch": 0.9702127659574468,
+      "grad_norm": 0.1366124153137207,
+      "learning_rate": 5.773803660319234e-06,
+      "loss": 1.0662239789962769,
+      "step": 228
+    },
+    {
+      "epoch": 0.9787234042553191,
+      "grad_norm": 0.12555794417858124,
+      "learning_rate": 5.621859728125884e-06,
+      "loss": 1.222013235092163,
+      "step": 230
+    },
+    {
+      "epoch": 0.9872340425531915,
+      "grad_norm": 0.29969581961631775,
+      "learning_rate": 5.479146430146781e-06,
+      "loss": 1.0417675971984863,
+      "step": 232
+    },
+    {
+      "epoch": 0.9957446808510638,
+      "grad_norm": 0.4836138188838959,
+      "learning_rate": 5.345777054854579e-06,
+      "loss": 0.9495846629142761,
+      "step": 234
+    },
+    {
+      "epoch": 1.004255319148936,
+      "grad_norm": 0.28257113695144653,
+      "learning_rate": 5.221857473342149e-06,
+      "loss": 0.843216061592102,
+      "step": 236
+    },
+    {
+      "epoch": 1.0127659574468084,
+      "grad_norm": 0.1423388570547104,
+      "learning_rate": 5.10748605528015e-06,
+      "loss": 0.9633752107620239,
+      "step": 238
+    },
+    {
+      "epoch": 1.0212765957446808,
+      "grad_norm": 0.12806400656700134,
+      "learning_rate": 5.002753590829349e-06,
+      "loss": 0.6067866683006287,
+      "step": 240
+    },
+    {
+      "epoch": 1.0297872340425531,
+      "grad_norm": 0.243302121758461,
+      "learning_rate": 4.9077432185697e-06,
+      "loss": 0.9341376423835754,
+      "step": 242
+    },
+    {
+      "epoch": 1.0382978723404255,
+      "grad_norm": 0.22620677947998047,
+      "learning_rate": 4.822530359503393e-06,
+      "loss": 1.046706199645996,
+      "step": 244
+    },
+    {
+      "epoch": 1.0468085106382978,
+      "grad_norm": 0.1606007069349289,
+      "learning_rate": 4.747182657184251e-06,
+      "loss": 0.6613105535507202,
+      "step": 246
+    },
+    {
+      "epoch": 1.0553191489361702,
+      "grad_norm": 0.3230328857898712,
+      "learning_rate": 4.681759924021033e-06,
+      "loss": 0.7598414421081543,
+      "step": 248
+    },
+    {
+      "epoch": 1.0638297872340425,
+      "grad_norm": 0.13726186752319336,
+      "learning_rate": 4.626314093797213e-06,
+      "loss": 0.7334257364273071,
+      "step": 250
+    },
+    {
+      "epoch": 1.0723404255319149,
+      "grad_norm": 0.3747357726097107,
+      "learning_rate": 4.580889180444988e-06,
+      "loss": 0.5062970519065857,
+      "step": 252
+    },
+    {
+      "epoch": 1.0808510638297872,
+      "grad_norm": 0.15663602948188782,
+      "learning_rate": 4.545521243106197e-06,
+      "loss": 1.0599923133850098,
+      "step": 254
+    },
+    {
+      "epoch": 1.0893617021276596,
+      "grad_norm": 0.1843012273311615,
+      "learning_rate": 4.520238357507899e-06,
+      "loss": 0.940389096736908,
+      "step": 256
+    },
+    {
+      "epoch": 1.097872340425532,
+      "grad_norm": 0.22409912943840027,
+      "learning_rate": 4.505060593675342e-06,
+      "loss": 1.0018244981765747,
+      "step": 258
+    },
+    {
+      "epoch": 1.1063829787234043,
+      "grad_norm": 0.45610666275024414,
+      "learning_rate": 4.5e-06,
+      "loss": 0.8150784969329834,
+      "step": 260
+    },
+    {
+      "epoch": 1.1148936170212767,
+      "grad_norm": 0.19215510785579681,
+      "learning_rate": 4.505060593675342e-06,
+      "loss": 0.7950176000595093,
+      "step": 262
+    },
+    {
+      "epoch": 1.123404255319149,
+      "grad_norm": 0.23423391580581665,
+      "learning_rate": 4.520238357507899e-06,
+      "loss": 0.7621920704841614,
+      "step": 264
+    },
+    {
+      "epoch": 1.1319148936170214,
+      "grad_norm": 0.15685321390628815,
+      "learning_rate": 4.545521243106197e-06,
+      "loss": 0.8206462264060974,
+      "step": 266
+    },
+    {
+      "epoch": 1.1404255319148937,
+      "grad_norm": 0.4606432020664215,
+      "learning_rate": 4.580889180444988e-06,
+      "loss": 0.9801955819129944,
+      "step": 268
+    },
+    {
+      "epoch": 1.148936170212766,
+      "grad_norm": 0.21424169838428497,
+      "learning_rate": 4.6263140937972124e-06,
+      "loss": 0.8219439387321472,
+      "step": 270
+    },
+    {
+      "epoch": 1.1574468085106382,
+      "grad_norm": 0.2082989364862442,
+      "learning_rate": 4.6817599240210315e-06,
+      "loss": 0.6543676257133484,
+      "step": 272
+    },
+    {
+      "epoch": 1.1659574468085105,
+      "grad_norm": 0.1622384637594223,
+      "learning_rate": 4.747182657184251e-06,
+      "loss": 0.9012349247932434,
+      "step": 274
+    },
+    {
+      "epoch": 1.174468085106383,
+      "grad_norm": 0.13174618780612946,
+      "learning_rate": 4.822530359503391e-06,
+      "loss": 0.7129980325698853,
+      "step": 276
+    },
+    {
+      "epoch": 1.1829787234042553,
+      "grad_norm": 0.2709389925003052,
+      "learning_rate": 4.9077432185697e-06,
+      "loss": 0.5482591986656189,
+      "step": 278
+    },
+    {
+      "epoch": 1.1914893617021276,
+      "grad_norm": 0.2576906979084015,
+      "learning_rate": 5.002753590829348e-06,
+      "loss": 0.5497639775276184,
+      "step": 280
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 0.1526031494140625,
+      "learning_rate": 5.1074860552801466e-06,
+      "loss": 0.6242255568504333,
+      "step": 282
+    },
+    {
+      "epoch": 1.2085106382978723,
+      "grad_norm": 0.22073198854923248,
+      "learning_rate": 5.2218574733421455e-06,
+      "loss": 0.693662703037262,
+      "step": 284
+    },
+    {
+      "epoch": 1.2170212765957447,
+      "grad_norm": 0.22191239893436432,
+      "learning_rate": 5.345777054854578e-06,
+      "loss": 0.9797627329826355,
+      "step": 286
+    },
+    {
+      "epoch": 1.225531914893617,
+      "grad_norm": 0.17303584516048431,
+      "learning_rate": 5.479146430146783e-06,
+      "loss": 0.9435557126998901,
+      "step": 288
+    },
+    {
+      "epoch": 1.2340425531914894,
+      "grad_norm": 0.2218431681394577,
+      "learning_rate": 5.6218597281258834e-06,
+      "loss": 0.6460038423538208,
+      "step": 290
+    },
+    {
+      "epoch": 1.2425531914893617,
+      "grad_norm": 0.22332793474197388,
+      "learning_rate": 5.773803660319232e-06,
+      "loss": 0.8468654155731201,
+      "step": 292
+    },
+    {
+      "epoch": 1.251063829787234,
+      "grad_norm": 0.13609401881694794,
+      "learning_rate": 5.934857610804904e-06,
+      "loss": 0.8879528641700745,
+      "step": 294
+    },
+    {
+      "epoch": 1.2595744680851064,
+      "grad_norm": 0.18097440898418427,
+      "learning_rate": 6.1048937319588676e-06,
+      "loss": 0.7949476838111877,
+      "step": 296
+    },
+    {
+      "epoch": 1.2680851063829788,
+      "grad_norm": 0.3170534074306488,
+      "learning_rate": 6.283777045942798e-06,
+      "loss": 0.8318912386894226,
+      "step": 298
+    },
+    {
+      "epoch": 1.2765957446808511,
+      "grad_norm": 0.19770559668540955,
+      "learning_rate": 6.471365551852011e-06,
+      "loss": 1.0059137344360352,
+      "step": 300
+    },
+    {
+      "epoch": 1.2851063829787235,
+      "grad_norm": 0.14630961418151855,
+      "learning_rate": 6.66751033843842e-06,
+      "loss": 0.39432206749916077,
+      "step": 302
+    },
+    {
+      "epoch": 1.2936170212765958,
+      "grad_norm": 0.3450065553188324,
+      "learning_rate": 6.872055702319048e-06,
+      "loss": 0.9369863271713257,
+      "step": 304
+    },
+    {
+      "epoch": 1.302127659574468,
+      "grad_norm": 0.18490372598171234,
+      "learning_rate": 7.084839271576289e-06,
+      "loss": 0.9826689958572388,
+      "step": 306
+    },
+    {
+      "epoch": 1.3106382978723405,
+      "grad_norm": 0.1866351068019867,
+      "learning_rate": 7.30569213465174e-06,
+      "loss": 0.5573344826698303,
+      "step": 308
+    },
+    {
+      "epoch": 1.3191489361702127,
+      "grad_norm": 0.1976342350244522,
+      "learning_rate": 7.534438974431351e-06,
+      "loss": 0.5633378624916077,
+      "step": 310
+    },
+    {
+      "epoch": 1.327659574468085,
+      "grad_norm": 0.3162063956260681,
+      "learning_rate": 7.770898207415414e-06,
+      "loss": 0.7375507354736328,
+      "step": 312
+    },
+    {
+      "epoch": 1.3361702127659574,
+      "grad_norm": 0.2016829252243042,
+      "learning_rate": 8.014882127862923e-06,
+      "loss": 0.4668198823928833,
+      "step": 314
+    },
+    {
+      "epoch": 1.3446808510638297,
+      "grad_norm": 1.459976315498352,
+      "learning_rate": 8.266197056795886e-06,
+      "loss": 0.7280799150466919,
+      "step": 316
+    },
+    {
+      "epoch": 1.353191489361702,
+      "grad_norm": 0.3398215174674988,
+      "learning_rate": 8.5246434957453e-06,
+      "loss": 0.5291551351547241,
+      "step": 318
+    },
+    {
+      "epoch": 1.3617021276595744,
+      "grad_norm": 0.19973713159561157,
+      "learning_rate": 8.790016285116766e-06,
+      "loss": 0.5350017547607422,
+      "step": 320
+    },
+    {
+      "epoch": 1.3702127659574468,
+      "grad_norm": 0.14308671653270721,
+      "learning_rate": 9.062104767049957e-06,
+      "loss": 0.9615387916564941,
+      "step": 322
+    },
+    {
+      "epoch": 1.3787234042553191,
+      "grad_norm": 0.1939127892255783,
+      "learning_rate": 9.340692952642787e-06,
+      "loss": 0.6444461345672607,
+      "step": 324
+    },
+    {
+      "epoch": 1.3872340425531915,
+      "grad_norm": 0.18625043332576752,
+      "learning_rate": 9.625559693407413e-06,
+      "loss": 0.9194571375846863,
+      "step": 326
+    },
+    {
+      "epoch": 1.3957446808510638,
+      "grad_norm": 0.2951974868774414,
+      "learning_rate": 9.916478856822006e-06,
+      "loss": 0.9841423034667969,
+      "step": 328
+    },
+    {
+      "epoch": 1.4042553191489362,
+      "grad_norm": 0.16953378915786743,
+      "learning_rate": 1.0213219505838985e-05,
+      "loss": 0.8577695488929749,
+      "step": 330
+    },
+    {
+      "epoch": 1.4127659574468086,
+      "grad_norm": 0.2879262864589691,
+      "learning_rate": 1.0515546082207094e-05,
+      "loss": 0.8521637916564941,
+      "step": 332
+    },
+    {
+      "epoch": 1.421276595744681,
+      "grad_norm": 0.22187970578670502,
+      "learning_rate": 1.082321859346199e-05,
+      "loss": 0.6609964966773987,
+      "step": 334
+    },
+    {
+      "epoch": 1.4297872340425533,
+      "grad_norm": 0.19994378089904785,
+      "learning_rate": 1.1135992803436696e-05,
+      "loss": 0.5523483753204346,
+      "step": 336
+    },
+    {
+      "epoch": 1.4382978723404256,
+      "grad_norm": 0.19332869350910187,
+      "learning_rate": 1.1453620426140791e-05,
+      "loss": 0.7419446110725403,
+      "step": 338
+    },
+    {
+      "epoch": 1.4468085106382977,
+      "grad_norm": 0.17565295100212097,
+      "learning_rate": 1.1775849322854508e-05,
+      "loss": 0.7033047676086426,
+      "step": 340
+    },
+    {
+      "epoch": 1.4553191489361703,
+      "grad_norm": 0.3124133050441742,
+      "learning_rate": 1.210242370228112e-05,
+      "loss": 0.5836398005485535,
+      "step": 342
+    },
+    {
+      "epoch": 1.4638297872340424,
+      "grad_norm": 0.21754996478557587,
+      "learning_rate": 1.2433084323598791e-05,
+      "loss": 0.9612702131271362,
+      "step": 344
+    },
+    {
+      "epoch": 1.472340425531915,
+      "grad_norm": 0.21039743721485138,
+      "learning_rate": 1.2767568702250838e-05,
+      "loss": 0.9444953799247742,
+      "step": 346
+    },
+    {
+      "epoch": 1.4808510638297872,
+      "grad_norm": 0.42962566018104553,
+      "learning_rate": 1.3105611318310811e-05,
+      "loss": 0.8060452938079834,
+      "step": 348
+    },
+    {
+      "epoch": 1.4893617021276595,
+      "grad_norm": 0.19513703882694244,
+      "learning_rate": 1.3446943827257184e-05,
+      "loss": 0.8150795698165894,
+      "step": 350
+    },
+    {
+      "epoch": 1.4978723404255319,
+      "grad_norm": 0.2640067934989929,
+      "learning_rate": 1.3791295272990172e-05,
+      "loss": 0.3241785168647766,
+      "step": 352
+    },
+    {
+      "epoch": 1.5063829787234042,
+      "grad_norm": 0.25144851207733154,
+      "learning_rate": 1.413839230292182e-05,
+      "loss": 0.7416596412658691,
+      "step": 354
+    },
+    {
+      "epoch": 1.5148936170212766,
+      "grad_norm": 0.16774724423885345,
+      "learning_rate": 1.4487959384968276e-05,
+      "loss": 0.518079400062561,
+      "step": 356
+    },
+    {
+      "epoch": 1.523404255319149,
+      "grad_norm": 0.1518772393465042,
+      "learning_rate": 1.4839719026272373e-05,
+      "loss": 0.5670950412750244,
+      "step": 358
+    },
+    {
+      "epoch": 1.5319148936170213,
+      "grad_norm": 0.23323562741279602,
+      "learning_rate": 1.5193391993482579e-05,
+      "loss": 1.1480307579040527,
+      "step": 360
+    },
+    {
+      "epoch": 1.5404255319148936,
+      "grad_norm": 0.1399996131658554,
+      "learning_rate": 1.5548697534413636e-05,
+      "loss": 0.5208263397216797,
+      "step": 362
+    },
+    {
+      "epoch": 1.548936170212766,
+      "grad_norm": 0.19347982108592987,
+      "learning_rate": 1.5905353600912894e-05,
+      "loss": 0.9330644607543945,
+      "step": 364
+    },
+    {
+      "epoch": 1.5574468085106383,
+      "grad_norm": 0.15133146941661835,
+      "learning_rate": 1.6263077072755323e-05,
+      "loss": 0.8360081911087036,
+      "step": 366
+    },
+    {
+      "epoch": 1.5659574468085107,
+      "grad_norm": 0.18284079432487488,
+      "learning_rate": 1.6621583982389707e-05,
+      "loss": 0.7362462878227234,
+      "step": 368
+    },
+    {
+      "epoch": 1.574468085106383,
+      "grad_norm": 0.19372136890888214,
+      "learning_rate": 1.6980589740357294e-05,
+      "loss": 0.6744348406791687,
+      "step": 370
+    },
+    {
+      "epoch": 1.5829787234042554,
+      "grad_norm": 0.12256433069705963,
+      "learning_rate": 1.733980936120425e-05,
+      "loss": 0.9976662397384644,
+      "step": 372
+    },
+    {
+      "epoch": 1.5914893617021275,
+      "grad_norm": 0.20529648661613464,
+      "learning_rate": 1.7698957689708416e-05,
+      "loss": 1.0515512228012085,
+      "step": 374
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.16672298312187195,
+      "learning_rate": 1.805774962724083e-05,
+      "loss": 0.7891988754272461,
+      "step": 376
+    },
+    {
+      "epoch": 1.6085106382978722,
+      "grad_norm": 0.1434396356344223,
+      "learning_rate": 1.8415900358082265e-05,
+      "loss": 0.6114462018013,
+      "step": 378
+    },
+    {
+      "epoch": 1.6170212765957448,
+      "grad_norm": 0.16634748876094818,
+      "learning_rate": 1.8773125575515357e-05,
+      "loss": 0.5973749756813049,
+      "step": 380
+    },
+    {
+      "epoch": 1.625531914893617,
+      "grad_norm": 0.13240687549114227,
+      "learning_rate": 1.912914170751251e-05,
+      "loss": 0.7943100929260254,
+      "step": 382
+    },
+    {
+      "epoch": 1.6340425531914895,
+      "grad_norm": 0.12580837309360504,
+      "learning_rate": 1.9483666141840612e-05,
+      "loss": 0.9919679760932922,
+      "step": 384
+    },
+    {
+      "epoch": 1.6425531914893616,
+      "grad_norm": 0.35519301891326904,
+      "learning_rate": 1.9836417450403974e-05,
+      "loss": 0.8425494432449341,
+      "step": 386
+    },
+    {
+      "epoch": 1.6510638297872342,
+      "grad_norm": 0.2488897144794464,
+      "learning_rate": 2.0187115612647133e-05,
+      "loss": 0.7412829399108887,
+      "step": 388
+    },
+    {
+      "epoch": 1.6595744680851063,
+      "grad_norm": 0.5074354410171509,
+      "learning_rate": 2.05354822378404e-05,
+      "loss": 0.7992886900901794,
+      "step": 390
+    },
+    {
+      "epoch": 1.6680851063829787,
+      "grad_norm": 0.12573714554309845,
+      "learning_rate": 2.0881240786071595e-05,
+      "loss": 0.909046471118927,
+      "step": 392
+    },
+    {
+      "epoch": 1.676595744680851,
+      "grad_norm": 0.33457598090171814,
+      "learning_rate": 2.122411678776854e-05,
+      "loss": 0.6243407726287842,
+      "step": 394
+    },
+    {
+      "epoch": 1.6851063829787234,
+      "grad_norm": 0.15411563217639923,
+      "learning_rate": 2.1563838061578258e-05,
+      "loss": 0.8719924688339233,
+      "step": 396
+    },
+    {
+      "epoch": 1.6936170212765957,
+      "grad_norm": 0.13289184868335724,
+      "learning_rate": 2.1900134930429396e-05,
+      "loss": 1.0623129606246948,
+      "step": 398
+    },
+    {
+      "epoch": 1.702127659574468,
+      "grad_norm": 0.3402920663356781,
+      "learning_rate": 2.223274043560706e-05,
+      "loss": 0.9607300162315369,
+      "step": 400
+    },
+    {
+      "epoch": 1.7106382978723405,
+      "grad_norm": 0.20696532726287842,
+      "learning_rate": 2.256139054866954e-05,
+      "loss": 0.4819542169570923,
+      "step": 402
+    },
+    {
+      "epoch": 1.7191489361702128,
+      "grad_norm": 0.1925085484981537,
+      "learning_rate": 2.2885824381039028e-05,
+      "loss": 0.6816413402557373,
+      "step": 404
+    },
+    {
+      "epoch": 1.7276595744680852,
+      "grad_norm": 0.5462839603424072,
+      "learning_rate": 2.3205784391099808e-05,
+      "loss": 0.5322574377059937,
+      "step": 406
+    },
+    {
+      "epoch": 1.7361702127659573,
+      "grad_norm": 0.12952132523059845,
+      "learning_rate": 2.352101658863958e-05,
+      "loss": 0.8366644382476807,
+      "step": 408
+    },
+    {
+      "epoch": 1.7446808510638299,
+      "grad_norm": 0.15898458659648895,
+      "learning_rate": 2.3831270736471703e-05,
+      "loss": 0.8219179511070251,
+      "step": 410
+    },
+    {
+      "epoch": 1.753191489361702,
+      "grad_norm": 0.10134760290384293,
+      "learning_rate": 2.4136300549077973e-05,
+      "loss": 0.8943849802017212,
+      "step": 412
+    },
+    {
+      "epoch": 1.7617021276595746,
+      "grad_norm": 0.21209417283535004,
+      "learning_rate": 2.443586388811481e-05,
+      "loss": 0.7499610781669617,
+      "step": 414
+    },
+    {
+      "epoch": 1.7702127659574467,
+      "grad_norm": 0.47646665573120117,
+      "learning_rate": 2.47297229546271e-05,
+      "loss": 0.6121576428413391,
+      "step": 416
+    },
+    {
+      "epoch": 1.7787234042553193,
+      "grad_norm": 0.26675277948379517,
+      "learning_rate": 2.5017644477817418e-05,
+      "loss": 0.7867974638938904,
+      "step": 418
+    },
+    {
+      "epoch": 1.7872340425531914,
+      "grad_norm": 0.840756356716156,
+      "learning_rate": 2.5299399900220807e-05,
+      "loss": 0.6045423150062561,
+      "step": 420
+    },
+    {
+      "epoch": 1.795744680851064,
+      "grad_norm": 0.400618314743042,
+      "learning_rate": 2.5574765559137848e-05,
+      "loss": 0.7998336553573608,
+      "step": 422
+    },
+    {
+      "epoch": 1.804255319148936,
+      "grad_norm": 0.15446501970291138,
+      "learning_rate": 2.584352286418239e-05,
+      "loss": 0.7926132678985596,
+      "step": 424
+    },
+    {
+      "epoch": 1.8127659574468085,
+      "grad_norm": 0.15063929557800293,
+      "learning_rate": 2.6105458470802563e-05,
+      "loss": 0.8235700130462646,
+      "step": 426
+    },
+    {
+      "epoch": 1.8212765957446808,
+      "grad_norm": 0.11289381980895996,
+      "learning_rate": 2.6360364449637683e-05,
+      "loss": 0.8320169448852539,
+      "step": 428
+    },
+    {
+      "epoch": 1.8297872340425532,
+      "grad_norm": 0.14190740883350372,
+      "learning_rate": 2.6608038451576528e-05,
+      "loss": 0.8250552415847778,
+      "step": 430
+    },
+    {
+      "epoch": 1.8382978723404255,
+      "grad_norm": 0.3685653507709503,
+      "learning_rate": 2.684828386838569e-05,
+      "loss": 0.8021817207336426,
+      "step": 432
+    },
+    {
+      "epoch": 1.8468085106382979,
+      "grad_norm": 0.12853652238845825,
+      "learning_rate": 2.7080909988780982e-05,
+      "loss": 0.7879712581634521,
+      "step": 434
+    },
+    {
+      "epoch": 1.8553191489361702,
+      "grad_norm": 0.3018054962158203,
+      "learning_rate": 2.7305732149817502e-05,
+      "loss": 0.7762024998664856,
+      "step": 436
+    },
+    {
+      "epoch": 1.8638297872340426,
+      "grad_norm": 0.14753182232379913,
+      "learning_rate": 2.7522571883478617e-05,
+      "loss": 1.2395200729370117,
+      "step": 438
+    },
+    {
+      "epoch": 1.872340425531915,
+      "grad_norm": 0.1134040355682373,
+      "learning_rate": 2.77312570583471e-05,
+      "loss": 0.8664835691452026,
+      "step": 440
+    },
+    {
+      "epoch": 1.8808510638297873,
+      "grad_norm": 0.19325752556324005,
+      "learning_rate": 2.7931622016246304e-05,
+      "loss": 1.0493484735488892,
+      "step": 442
+    },
+    {
+      "epoch": 1.8893617021276596,
+      "grad_norm": 0.15366853773593903,
+      "learning_rate": 2.8123507703742727e-05,
+      "loss": 0.8971328139305115,
+      "step": 444
+    },
+    {
+      "epoch": 1.8978723404255318,
+      "grad_norm": 0.1713661104440689,
+      "learning_rate": 2.8306761798405522e-05,
+      "loss": 0.9821380376815796,
+      "step": 446
+    },
+    {
+      "epoch": 1.9063829787234043,
+      "grad_norm": 0.4014803469181061,
+      "learning_rate": 2.848123882972295e-05,
+      "loss": 1.0046873092651367,
+      "step": 448
+    },
+    {
+      "epoch": 1.9148936170212765,
+      "grad_norm": 0.1893383413553238,
+      "learning_rate": 2.8646800294579514e-05,
+      "loss": 0.647858738899231,
+      "step": 450
+    },
+    {
+      "epoch": 1.923404255319149,
+      "grad_norm": 0.5622705221176147,
+      "learning_rate": 2.8803314767202376e-05,
+      "loss": 0.5715224742889404,
+      "step": 452
+    },
+    {
+      "epoch": 1.9319148936170212,
+      "grad_norm": 0.14517684280872345,
+      "learning_rate": 2.8950658003489534e-05,
+      "loss": 0.8555135726928711,
+      "step": 454
+    },
+    {
+      "epoch": 1.9404255319148938,
+      "grad_norm": 0.16155880689620972,
+      "learning_rate": 2.908871303963711e-05,
+      "loss": 0.9127392768859863,
+      "step": 456
+    },
+    {
+      "epoch": 1.9489361702127659,
+      "grad_norm": 0.12455891817808151,
+      "learning_rate": 2.9217370284987434e-05,
+      "loss": 0.9905153512954712,
+      "step": 458
+    },
+    {
+      "epoch": 1.9574468085106385,
+      "grad_norm": 0.1255619376897812,
+      "learning_rate": 2.9336527609024075e-05,
+      "loss": 1.022163987159729,
+      "step": 460
+    },
+    {
+      "epoch": 1.9659574468085106,
+      "grad_norm": 0.38456299901008606,
+      "learning_rate": 2.9446090422445016e-05,
+      "loss": 0.8251804709434509,
+      "step": 462
+    },
+    {
+      "epoch": 1.974468085106383,
+      "grad_norm": 0.12075914442539215,
+      "learning_rate": 2.9545971752249376e-05,
+      "loss": 1.0136280059814453,
+      "step": 464
+    },
+    {
+      "epoch": 1.9829787234042553,
+      "grad_norm": 0.20235367119312286,
+      "learning_rate": 2.9636092310778195e-05,
+      "loss": 0.9567205905914307,
+      "step": 466
+    },
+    {
+      "epoch": 1.9914893617021276,
+      "grad_norm": 0.542694091796875,
+      "learning_rate": 2.9716380558654445e-05,
+      "loss": 0.6730895042419434,
+      "step": 468
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.11073193699121475,
+      "learning_rate": 2.9786772761572338e-05,
+      "loss": 0.6498197317123413,
+      "step": 470
+    },
+    {
+      "epoch": 2.008510638297872,
+      "grad_norm": 0.11644601076841354,
+      "learning_rate": 2.9847213040890793e-05,
+      "loss": 0.4679168164730072,
+      "step": 472
+    },
+    {
+      "epoch": 2.0170212765957447,
+      "grad_norm": 0.14129982888698578,
+      "learning_rate": 2.989765341799095e-05,
+      "loss": 0.7698261737823486,
+      "step": 474
+    },
+    {
+      "epoch": 2.025531914893617,
+      "grad_norm": 0.13531845808029175,
+      "learning_rate": 2.9938053852362484e-05,
+      "loss": 0.47750866413116455,
+      "step": 476
+    },
+    {
+      "epoch": 2.0340425531914894,
+      "grad_norm": 0.37192943692207336,
+      "learning_rate": 2.99683822733885e-05,
+      "loss": 0.6370930075645447,
+      "step": 478
+    },
+    {
+      "epoch": 2.0425531914893615,
+      "grad_norm": 0.13162164390087128,
+      "learning_rate": 2.9988614605803806e-05,
+      "loss": 0.44695600867271423,
+      "step": 480
+    },
+    {
+      "epoch": 2.051063829787234,
+      "grad_norm": 0.20202530920505524,
+      "learning_rate": 2.9998734788806287e-05,
+      "loss": 0.4687855839729309,
+      "step": 482
+    },
+    {
+      "epoch": 2.0595744680851062,
+      "grad_norm": 0.09361886233091354,
+      "learning_rate": 2.9998734788806287e-05,
+      "loss": 0.28051936626434326,
+      "step": 484
+    },
+    {
+      "epoch": 2.068085106382979,
+      "grad_norm": 0.1879085749387741,
+      "learning_rate": 2.9988614605803806e-05,
+      "loss": 0.6579060554504395,
+      "step": 486
+    },
+    {
+      "epoch": 2.076595744680851,
+      "grad_norm": 0.24100597202777863,
+      "learning_rate": 2.99683822733885e-05,
+      "loss": 0.5791319608688354,
+      "step": 488
+    },
+    {
+      "epoch": 2.0851063829787235,
+      "grad_norm": 0.15436111390590668,
+      "learning_rate": 2.9938053852362484e-05,
+      "loss": 0.5440673828125,
+      "step": 490
+    },
+    {
+      "epoch": 2.0936170212765957,
+      "grad_norm": 0.1665167361497879,
+      "learning_rate": 2.989765341799095e-05,
+      "loss": 0.7786204218864441,
+      "step": 492
+    },
+    {
+      "epoch": 2.1021276595744682,
+      "grad_norm": 0.12963709235191345,
+      "learning_rate": 2.9847213040890793e-05,
+      "loss": 0.6289861798286438,
+      "step": 494
+    },
+    {
+      "epoch": 2.1106382978723404,
+      "grad_norm": 0.23455171287059784,
+      "learning_rate": 2.9786772761572338e-05,
+      "loss": 0.7049335241317749,
+      "step": 496
+    },
+    {
+      "epoch": 2.119148936170213,
+      "grad_norm": 0.10971709340810776,
+      "learning_rate": 2.9716380558654445e-05,
+      "loss": 0.9018527269363403,
+      "step": 498
+    },
+    {
+      "epoch": 2.127659574468085,
+      "grad_norm": 0.3122045695781708,
+      "learning_rate": 2.9636092310778195e-05,
+      "loss": 0.6065503358840942,
+      "step": 500
+    },
+    {
+      "epoch": 2.1361702127659576,
+      "grad_norm": 0.15959368646144867,
+      "learning_rate": 2.9545971752249383e-05,
+      "loss": 0.5845508575439453,
+      "step": 502
+    },
+    {
+      "epoch": 2.1446808510638298,
+      "grad_norm": 0.12507954239845276,
+      "learning_rate": 2.9446090422445016e-05,
+      "loss": 0.6952933669090271,
+      "step": 504
+    },
+    {
+      "epoch": 2.153191489361702,
+      "grad_norm": 0.12080808728933334,
+      "learning_rate": 2.933652760902408e-05,
+      "loss": 0.7144219279289246,
+      "step": 506
+    },
+    {
+      "epoch": 2.1617021276595745,
+      "grad_norm": 0.3896176815032959,
+      "learning_rate": 2.9217370284987434e-05,
+      "loss": 0.5674622654914856,
+      "step": 508
+    },
+    {
+      "epoch": 2.1702127659574466,
+      "grad_norm": 0.14054062962532043,
+      "learning_rate": 2.908871303963712e-05,
+      "loss": 0.28315988183021545,
+      "step": 510
+    },
+    {
+      "epoch": 2.178723404255319,
+      "grad_norm": 0.1403239667415619,
+      "learning_rate": 2.8950658003489534e-05,
+      "loss": 0.430463582277298,
+      "step": 512
+    },
+    {
+      "epoch": 2.1872340425531913,
+      "grad_norm": 0.14137110114097595,
+      "learning_rate": 2.8803314767202387e-05,
+      "loss": 0.4114861488342285,
+      "step": 514
+    },
+    {
+      "epoch": 2.195744680851064,
+      "grad_norm": 0.1280188113451004,
+      "learning_rate": 2.864680029457952e-05,
+      "loss": 0.5308094620704651,
+      "step": 516
+    },
+    {
+      "epoch": 2.204255319148936,
+      "grad_norm": 0.25649434328079224,
+      "learning_rate": 2.8481238829722946e-05,
+      "loss": 0.6915931701660156,
+      "step": 518
+    },
+    {
+      "epoch": 2.2127659574468086,
+      "grad_norm": 0.10003609210252762,
+      "learning_rate": 2.8306761798405533e-05,
+      "loss": 0.4150570333003998,
+      "step": 520
+    },
+    {
+      "epoch": 2.2212765957446807,
+      "grad_norm": 0.15974430739879608,
+      "learning_rate": 2.812350770374273e-05,
+      "loss": 0.36256328225135803,
+      "step": 522
+    },
+    {
+      "epoch": 2.2297872340425533,
+      "grad_norm": 0.38845136761665344,
+      "learning_rate": 2.793162201624631e-05,
+      "loss": 0.37502366304397583,
+      "step": 524
+    },
+    {
+      "epoch": 2.2382978723404254,
+      "grad_norm": 0.1723078191280365,
+      "learning_rate": 2.7731257058347098e-05,
+      "loss": 0.7607256770133972,
+      "step": 526
+    },
+    {
+      "epoch": 2.246808510638298,
+      "grad_norm": 0.255744993686676,
+      "learning_rate": 2.752257188347862e-05,
+      "loss": 0.6797502040863037,
+      "step": 528
+    },
+    {
+      "epoch": 2.25531914893617,
+      "grad_norm": 0.3702438771724701,
+      "learning_rate": 2.7305732149817512e-05,
+      "loss": 0.7978914380073547,
+      "step": 530
+    },
+    {
+      "epoch": 2.2638297872340427,
+      "grad_norm": 0.26327913999557495,
+      "learning_rate": 2.708090998878098e-05,
+      "loss": 0.5080232620239258,
+      "step": 532
+    },
+    {
+      "epoch": 2.272340425531915,
+      "grad_norm": 0.14510440826416016,
+      "learning_rate": 2.68482838683857e-05,
+      "loss": 0.2904958724975586,
+      "step": 534
+    },
+    {
+      "epoch": 2.2808510638297874,
+      "grad_norm": 0.1330595761537552,
+      "learning_rate": 2.6608038451576528e-05,
+      "loss": 0.5988547205924988,
+      "step": 536
+    },
+    {
+      "epoch": 2.2893617021276595,
+      "grad_norm": 0.15609091520309448,
+      "learning_rate": 2.6360364449637686e-05,
+      "loss": 0.8217217922210693,
+      "step": 538
+    },
+    {
+      "epoch": 2.297872340425532,
+      "grad_norm": 0.27816882729530334,
+      "learning_rate": 2.6105458470802563e-05,
+      "loss": 0.29454290866851807,
+      "step": 540
+    },
+    {
+      "epoch": 2.3063829787234043,
+      "grad_norm": 0.16807164251804352,
+      "learning_rate": 2.5843522864182397e-05,
+      "loss": 0.2661294937133789,
+      "step": 542
+    },
+    {
+      "epoch": 2.3148936170212764,
+      "grad_norm": 0.2340618073940277,
+      "learning_rate": 2.557476555913786e-05,
+      "loss": 0.4138932228088379,
+      "step": 544
+    },
+    {
+      "epoch": 2.323404255319149,
+      "grad_norm": 0.12209529429674149,
+      "learning_rate": 2.52993999002208e-05,
+      "loss": 0.47122570872306824,
+      "step": 546
+    },
+    {
+      "epoch": 2.331914893617021,
+      "grad_norm": 0.22724129259586334,
+      "learning_rate": 2.501764447781743e-05,
+      "loss": 0.5940241813659668,
+      "step": 548
+    },
+    {
+      "epoch": 2.3404255319148937,
+      "grad_norm": 0.10988523811101913,
+      "learning_rate": 2.4729722954627106e-05,
+      "loss": 0.443272203207016,
+      "step": 550
+    },
+    {
+      "epoch": 2.348936170212766,
+      "grad_norm": 0.3907153904438019,
+      "learning_rate": 2.4435863888114814e-05,
+      "loss": 0.612157940864563,
+      "step": 552
+    },
+    {
+      "epoch": 2.3574468085106384,
+      "grad_norm": 0.15928088128566742,
+      "learning_rate": 2.413630054907798e-05,
+      "loss": 0.6884116530418396,
+      "step": 554
+    },
+    {
+      "epoch": 2.3659574468085105,
+      "grad_norm": 0.12084610015153885,
+      "learning_rate": 2.383127073647171e-05,
+      "loss": 0.6459460854530334,
+      "step": 556
+    },
+    {
+      "epoch": 2.374468085106383,
+      "grad_norm": 0.14340610802173615,
+      "learning_rate": 2.3521016588639598e-05,
+      "loss": 0.5485546588897705,
+      "step": 558
+    },
+    {
+      "epoch": 2.382978723404255,
+      "grad_norm": 0.1330912858247757,
+      "learning_rate": 2.32057843910998e-05,
+      "loss": 0.6871886253356934,
+      "step": 560
+    },
+    {
+      "epoch": 2.391489361702128,
+      "grad_norm": 0.1461075246334076,
+      "learning_rate": 2.2885824381039024e-05,
+      "loss": 0.5943357944488525,
+      "step": 562
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 0.4632057547569275,
+      "learning_rate": 2.2561390548669552e-05,
+      "loss": 0.44520270824432373,
+      "step": 564
+    },
+    {
+      "epoch": 2.4085106382978725,
+      "grad_norm": 0.16760534048080444,
+      "learning_rate": 2.2232740435607067e-05,
+      "loss": 0.4427604377269745,
+      "step": 566
+    },
+    {
+      "epoch": 2.4170212765957446,
+      "grad_norm": 0.17341409623622894,
+      "learning_rate": 2.1900134930429403e-05,
+      "loss": 0.7208576798439026,
+      "step": 568
+    },
+    {
+      "epoch": 2.425531914893617,
+      "grad_norm": 0.12514737248420715,
+      "learning_rate": 2.1563838061578264e-05,
+      "loss": 0.6061448454856873,
+      "step": 570
+    },
+    {
+      "epoch": 2.4340425531914893,
+      "grad_norm": 0.275291383266449,
+      "learning_rate": 2.122411678776855e-05,
+      "loss": 0.48493388295173645,
+      "step": 572
+    },
+    {
+      "epoch": 2.4425531914893615,
+      "grad_norm": 0.35213977098464966,
+      "learning_rate": 2.0881240786071588e-05,
+      "loss": 0.6101267337799072,
+      "step": 574
+    },
+    {
+      "epoch": 2.451063829787234,
+      "grad_norm": 0.3173239827156067,
+      "learning_rate": 2.0535482237840398e-05,
+      "loss": 0.6204816102981567,
+      "step": 576
+    },
+    {
+      "epoch": 2.4595744680851066,
+      "grad_norm": 0.12788061797618866,
+      "learning_rate": 2.018711561264715e-05,
+      "loss": 0.5858104825019836,
+      "step": 578
+    },
+    {
+      "epoch": 2.4680851063829787,
+      "grad_norm": 0.21373364329338074,
+      "learning_rate": 1.9836417450403978e-05,
+      "loss": 0.5342880487442017,
+      "step": 580
+    },
+    {
+      "epoch": 2.476595744680851,
+      "grad_norm": 0.1782151162624359,
+      "learning_rate": 1.948366614184062e-05,
+      "loss": 0.6852924823760986,
+      "step": 582
+    },
+    {
+      "epoch": 2.4851063829787234,
+      "grad_norm": 0.12118978798389435,
+      "learning_rate": 1.9129141707512514e-05,
+      "loss": 0.5905125737190247,
+      "step": 584
+    },
+    {
+      "epoch": 2.4936170212765956,
+      "grad_norm": 0.6266365647315979,
+      "learning_rate": 1.877312557551536e-05,
+      "loss": 0.6729872822761536,
+      "step": 586
+    },
+    {
+      "epoch": 2.502127659574468,
+      "grad_norm": 0.13392919301986694,
+      "learning_rate": 1.841590035808227e-05,
+      "loss": 0.7322168350219727,
+      "step": 588
+    },
+    {
+      "epoch": 2.5106382978723403,
+      "grad_norm": 0.27286240458488464,
+      "learning_rate": 1.8057749627240822e-05,
+      "loss": 0.6285769939422607,
+      "step": 590
+    },
+    {
+      "epoch": 2.519148936170213,
+      "grad_norm": 0.10404688119888306,
+      "learning_rate": 1.7698957689708436e-05,
+      "loss": 0.766042172908783,
+      "step": 592
+    },
+    {
+      "epoch": 2.527659574468085,
+      "grad_norm": 0.11238724738359451,
+      "learning_rate": 1.7339809361204255e-05,
+      "loss": 0.43586933612823486,
+      "step": 594
+    },
+    {
+      "epoch": 2.5361702127659576,
+      "grad_norm": 0.12042172253131866,
+      "learning_rate": 1.6980589740357287e-05,
+      "loss": 0.6923111081123352,
+      "step": 596
+    },
+    {
+      "epoch": 2.5446808510638297,
+      "grad_norm": 0.2841528058052063,
+      "learning_rate": 1.6621583982389714e-05,
+      "loss": 0.47686851024627686,
+      "step": 598
+    },
+    {
+      "epoch": 2.5531914893617023,
+      "grad_norm": 0.11969015002250671,
+      "learning_rate": 1.626307707275533e-05,
+      "loss": 0.5013985633850098,
+      "step": 600
+    },
+    {
+      "epoch": 2.5617021276595744,
+      "grad_norm": 0.09876365959644318,
+      "learning_rate": 1.59053536009129e-05,
+      "loss": 0.6609531044960022,
+      "step": 602
+    },
+    {
+      "epoch": 2.570212765957447,
+      "grad_norm": 0.4649478495121002,
+      "learning_rate": 1.5548697534413653e-05,
+      "loss": 0.6927770972251892,
+      "step": 604
+    },
+    {
+      "epoch": 2.578723404255319,
+      "grad_norm": 0.1714295893907547,
+      "learning_rate": 1.5193391993482573e-05,
+      "loss": 0.8361954689025879,
+      "step": 606
+    },
+    {
+      "epoch": 2.5872340425531917,
+      "grad_norm": 0.15755438804626465,
+      "learning_rate": 1.483971902627239e-05,
+      "loss": 0.5158759951591492,
+      "step": 608
+    },
+    {
+      "epoch": 2.595744680851064,
+      "grad_norm": 0.11576741188764572,
+      "learning_rate": 1.4487959384968272e-05,
+      "loss": 0.6141366362571716,
+      "step": 610
+    },
+    {
+      "epoch": 2.604255319148936,
+      "grad_norm": 0.2397107034921646,
+      "learning_rate": 1.4138392302921813e-05,
+      "loss": 0.669286847114563,
+      "step": 612
+    },
+    {
+      "epoch": 2.6127659574468085,
+      "grad_norm": 0.17677916586399078,
+      "learning_rate": 1.3791295272990178e-05,
+      "loss": 0.8155961036682129,
+      "step": 614
+    },
+    {
+      "epoch": 2.621276595744681,
+      "grad_norm": 0.10534735023975372,
+      "learning_rate": 1.344694382725719e-05,
+      "loss": 0.6486954689025879,
+      "step": 616
+    },
+    {
+      "epoch": 2.629787234042553,
+      "grad_norm": 0.17969125509262085,
+      "learning_rate": 1.3105611318310828e-05,
+      "loss": 0.7624613046646118,
+      "step": 618
+    },
+    {
+      "epoch": 2.6382978723404253,
+      "grad_norm": 0.10477960854768753,
+      "learning_rate": 1.2767568702250834e-05,
+      "loss": 0.44974949955940247,
+      "step": 620
+    },
+    {
+      "epoch": 2.646808510638298,
+      "grad_norm": 0.41400146484375,
+      "learning_rate": 1.2433084323598806e-05,
+      "loss": 0.5666506886482239,
+      "step": 622
+    },
+    {
+      "epoch": 2.65531914893617,
+      "grad_norm": 0.12478289753198624,
+      "learning_rate": 1.2102423702281114e-05,
+      "loss": 0.7800987362861633,
+      "step": 624
+    },
+    {
+      "epoch": 2.6638297872340426,
+      "grad_norm": 0.16413263976573944,
+      "learning_rate": 1.1775849322854515e-05,
+      "loss": 0.5514877438545227,
+      "step": 626
+    },
+    {
+      "epoch": 2.6723404255319148,
+      "grad_norm": 0.09944586455821991,
+      "learning_rate": 1.1453620426140796e-05,
+      "loss": 0.554164469242096,
+      "step": 628
+    },
+    {
+      "epoch": 2.6808510638297873,
+      "grad_norm": 0.2592844069004059,
+      "learning_rate": 1.1135992803436701e-05,
+      "loss": 0.5563373565673828,
+      "step": 630
+    },
+    {
+      "epoch": 2.6893617021276595,
+      "grad_norm": 0.2102140486240387,
+      "learning_rate": 1.0823218593462006e-05,
+      "loss": 0.5627699494361877,
+      "step": 632
+    },
+    {
+      "epoch": 2.697872340425532,
+      "grad_norm": 0.1483510136604309,
+      "learning_rate": 1.0515546082207089e-05,
+      "loss": 0.698486328125,
+      "step": 634
+    },
+    {
+      "epoch": 2.706382978723404,
+      "grad_norm": 0.11229369789361954,
+      "learning_rate": 1.0213219505839e-05,
+      "loss": 0.7639659643173218,
+      "step": 636
+    },
+    {
+      "epoch": 2.7148936170212767,
+      "grad_norm": 0.10941766947507858,
+      "learning_rate": 9.916478856822011e-06,
+      "loss": 0.6158884763717651,
+      "step": 638
+    },
+    {
+      "epoch": 2.723404255319149,
+      "grad_norm": 0.1750131994485855,
+      "learning_rate": 9.625559693407416e-06,
+      "loss": 0.9152920842170715,
+      "step": 640
+    },
+    {
+      "epoch": 2.731914893617021,
+      "grad_norm": 0.18939398229122162,
+      "learning_rate": 9.34069295264279e-06,
+      "loss": 0.8749684691429138,
+      "step": 642
+    },
+    {
+      "epoch": 2.7404255319148936,
+      "grad_norm": 0.0939684808254242,
+      "learning_rate": 9.062104767049962e-06,
+      "loss": 0.8344287872314453,
+      "step": 644
+    },
+    {
+      "epoch": 2.748936170212766,
+      "grad_norm": 0.10787810385227203,
+      "learning_rate": 8.790016285116773e-06,
+      "loss": 0.45629745721817017,
+      "step": 646
+    },
+    {
+      "epoch": 2.7574468085106383,
+      "grad_norm": 0.10236497223377228,
+      "learning_rate": 8.5246434957453e-06,
+      "loss": 0.3841492533683777,
+      "step": 648
+    },
+    {
+      "epoch": 2.7659574468085104,
+      "grad_norm": 0.31643790006637573,
+      "learning_rate": 8.266197056795886e-06,
+      "loss": 0.4197019338607788,
+      "step": 650
+    },
+    {
+      "epoch": 2.774468085106383,
+      "grad_norm": 0.15869994461536407,
+      "learning_rate": 8.014882127862926e-06,
+      "loss": 0.6872882843017578,
+      "step": 652
+    },
+    {
+      "epoch": 2.7829787234042556,
+      "grad_norm": 0.15020446479320526,
+      "learning_rate": 7.77089820741542e-06,
+      "loss": 0.6434124112129211,
+      "step": 654
+    },
+    {
+      "epoch": 2.7914893617021277,
+      "grad_norm": 0.13434921205043793,
+      "learning_rate": 7.534438974431356e-06,
+      "loss": 0.5578271150588989,
+      "step": 656
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 0.1459910273551941,
+      "learning_rate": 7.305692134651748e-06,
+      "loss": 0.6800917983055115,
+      "step": 658
+    },
+    {
+      "epoch": 2.8085106382978724,
+      "grad_norm": 0.13331423699855804,
+      "learning_rate": 7.0848392715763e-06,
+      "loss": 0.4372290074825287,
+      "step": 660
+    },
+    {
+      "epoch": 2.8170212765957445,
+      "grad_norm": 0.11432936787605286,
+      "learning_rate": 6.872055702319048e-06,
+      "loss": 0.3426898121833801,
+      "step": 662
+    },
+    {
+      "epoch": 2.825531914893617,
+      "grad_norm": 0.47585824131965637,
+      "learning_rate": 6.66751033843842e-06,
+      "loss": 0.5213098526000977,
+      "step": 664
+    },
+    {
+      "epoch": 2.8340425531914892,
+      "grad_norm": 0.23138658702373505,
+      "learning_rate": 6.471365551852014e-06,
+      "loss": 0.40890708565711975,
+      "step": 666
+    },
+    {
+      "epoch": 2.842553191489362,
+      "grad_norm": 0.1170782670378685,
+      "learning_rate": 6.283777045942801e-06,
+      "loss": 0.6171141266822815,
+      "step": 668
+    },
+    {
+      "epoch": 2.851063829787234,
+      "grad_norm": 0.2987634837627411,
+      "learning_rate": 6.104893731958872e-06,
+      "loss": 0.79531329870224,
+      "step": 670
+    },
+    {
+      "epoch": 2.8595744680851065,
+      "grad_norm": 0.16172188520431519,
+      "learning_rate": 5.934857610804912e-06,
+      "loss": 0.6109257936477661,
+      "step": 672
+    },
+    {
+      "epoch": 2.8680851063829786,
+      "grad_norm": 0.19967441260814667,
+      "learning_rate": 5.773803660319231e-06,
+      "loss": 0.7840122580528259,
+      "step": 674
+    },
+    {
+      "epoch": 2.876595744680851,
+      "grad_norm": 0.09976833313703537,
+      "learning_rate": 5.6218597281258834e-06,
+      "loss": 0.4490019679069519,
+      "step": 676
+    },
+    {
+      "epoch": 2.8851063829787233,
+      "grad_norm": 0.1243966668844223,
+      "learning_rate": 5.479146430146783e-06,
+      "loss": 0.5424228310585022,
+      "step": 678
+    },
+    {
+      "epoch": 2.8936170212765955,
+      "grad_norm": 0.17045721411705017,
+      "learning_rate": 5.3457770548545805e-06,
+      "loss": 0.5151770114898682,
+      "step": 680
+    },
+    {
+      "epoch": 2.902127659574468,
+      "grad_norm": 0.1732950657606125,
+      "learning_rate": 5.22185747334215e-06,
+      "loss": 0.7383885979652405,
+      "step": 682
+    },
+    {
+      "epoch": 2.9106382978723406,
+      "grad_norm": 0.21424034237861633,
+      "learning_rate": 5.107486055280145e-06,
+      "loss": 0.4885507822036743,
+      "step": 684
+    },
+    {
+      "epoch": 2.9191489361702128,
+      "grad_norm": 1.2460668087005615,
+      "learning_rate": 5.002753590829352e-06,
+      "loss": 0.29081299901008606,
+      "step": 686
+    },
+    {
+      "epoch": 2.927659574468085,
+      "grad_norm": 0.11700358986854553,
+      "learning_rate": 4.9077432185697e-06,
+      "loss": 0.436556339263916,
+      "step": 688
+    },
+    {
+      "epoch": 2.9361702127659575,
+      "grad_norm": 0.2099040299654007,
+      "learning_rate": 4.822530359503391e-06,
+      "loss": 0.759993314743042,
+      "step": 690
+    },
+    {
+      "epoch": 2.94468085106383,
+      "grad_norm": 0.12797491252422333,
+      "learning_rate": 4.747182657184251e-06,
+      "loss": 0.6529436111450195,
+      "step": 692
+    },
+    {
+      "epoch": 2.953191489361702,
+      "grad_norm": 0.122190460562706,
+      "learning_rate": 4.681759924021033e-06,
+      "loss": 0.6622499227523804,
+      "step": 694
+    },
+    {
+      "epoch": 2.9617021276595743,
+      "grad_norm": 0.19574476778507233,
+      "learning_rate": 4.626314093797213e-06,
+      "loss": 0.5456458926200867,
+      "step": 696
+    },
+    {
+      "epoch": 2.970212765957447,
+      "grad_norm": 0.15566527843475342,
+      "learning_rate": 4.580889180444988e-06,
+      "loss": 0.738301157951355,
+      "step": 698
+    },
+    {
+      "epoch": 2.978723404255319,
+      "grad_norm": 0.13139484822750092,
+      "learning_rate": 4.5455212431061985e-06,
+      "loss": 0.4914497435092926,
+      "step": 700
+    },
+    {
+      "epoch": 2.9872340425531916,
+      "grad_norm": 0.24261365830898285,
+      "learning_rate": 4.520238357507899e-06,
+      "loss": 0.5938633680343628,
+      "step": 702
+    },
+    {
+      "epoch": 2.9957446808510637,
+      "grad_norm": 0.09112333506345749,
+      "learning_rate": 4.505060593675342e-06,
+      "loss": 0.4166713058948517,
+      "step": 704
+    },
+    {
+      "epoch": 3.0,
+      "step": 705,
+      "total_flos": 3.0241141054567875e+18,
+      "train_loss": 0.8476084547685393,
+      "train_runtime": 9225.3576,
+      "train_samples_per_second": 4.891,
+      "train_steps_per_second": 0.076
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 705,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 99999,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.0241141054567875e+18,
+  "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:d7c0102a845490f60d3bdd954ffe891534663159c5e3d320ac07aed7c3e4fa97
+size 5649