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 +1,69 @@
+---
+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
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- family_to_muon_lr = {
+        "language": _fallback(getattr(training_args, "language_muon_lr", 2e-5), language_lr),
+        "vision": _fallback(getattr(training_args, "vision_muon_lr", 2e-5), vision_lr),
+        "merger": _fallback(getattr(training_args, "merger_muon_lr", 6e-5), merger_lr),
+    }
+
+    family_to_adamw_lr = {
+        "language": _fallback(getattr(training_args, "language_adamw_lr", 1e-6), language_lr),
+        "vision": _fallback(getattr(training_args, "vision_adamw_lr", 1e-6), vision_lr),
+        "merger": _fallback(getattr(training_args, "merger_adamw_lr", 1e-6), merger_lr),
+    }
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- gradient_accumulation_steps: 60
+- total_train_batch_size: 120
+- 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: 4
+
+### 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 +1,69 @@
+---
+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
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- family_to_muon_lr = {
+        "language": _fallback(getattr(training_args, "language_muon_lr", 2e-5), language_lr),
+        "vision": _fallback(getattr(training_args, "vision_muon_lr", 2e-5), vision_lr),
+        "merger": _fallback(getattr(training_args, "merger_muon_lr", 6e-5), merger_lr),
+    }
+
+    family_to_adamw_lr = {
+        "language": _fallback(getattr(training_args, "language_adamw_lr", 1e-6), language_lr),
+        "vision": _fallback(getattr(training_args, "vision_adamw_lr", 1e-6), vision_lr),
+        "merger": _fallback(getattr(training_args, "merger_adamw_lr", 1e-6), merger_lr),
+    }
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- gradient_accumulation_steps: 60
+- total_train_batch_size: 120
+- 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: 4
+
+### 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": 4731.714930074944,
+    "epoch": 4.0,
+    "total_flos": 3.800554784105169e+18,
+    "train_loss": 0.9838281438538903,
+    "train_runtime": 15715.7329,
+    "train_samples_per_second": 4.062,
+    "train_steps_per_second": 0.034
+}

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": "bfloat16",
+  "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 sha256:98c216186dae88c6acc1d26858b3bd5fee67b0d88e642fa9e2714b68de771705
+size 20859273368

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": 4731.714930074944,
+    "epoch": 4.0,
+    "total_flos": 3.800554784105169e+18,
+    "train_loss": 0.9838281438538903,
+    "train_runtime": 15715.7329,
+    "train_samples_per_second": 4.062,
+    "train_steps_per_second": 0.034
+}

trainer_log.jsonl

@@ -0,0 +1,267 @@
+{"current_steps": 2, "total_steps": 532, "loss": 1.9719613790512085, "lr": 7.407407407407407e-07, "epoch": 0.015037593984962405, "percentage": 0.38, "elapsed_time": "0:01:17", "remaining_time": "5:40:14"}
+{"current_steps": 4, "total_steps": 532, "loss": 2.2840397357940674, "lr": 2.222222222222222e-06, "epoch": 0.03007518796992481, "percentage": 0.75, "elapsed_time": "0:02:11", "remaining_time": "4:48:37"}
+{"current_steps": 6, "total_steps": 532, "loss": 2.0337564945220947, "lr": 3.7037037037037037e-06, "epoch": 0.045112781954887216, "percentage": 1.13, "elapsed_time": "0:03:32", "remaining_time": "5:10:22"}
+{"current_steps": 8, "total_steps": 532, "loss": 1.938320517539978, "lr": 5.185185185185185e-06, "epoch": 0.06015037593984962, "percentage": 1.5, "elapsed_time": "0:04:44", "remaining_time": "5:10:50"}
+{"current_steps": 10, "total_steps": 532, "loss": 1.9521509408950806, "lr": 6.666666666666667e-06, "epoch": 0.07518796992481203, "percentage": 1.88, "elapsed_time": "0:05:48", "remaining_time": "5:02:54"}
+{"current_steps": 12, "total_steps": 532, "loss": 2.3038036823272705, "lr": 8.148148148148148e-06, "epoch": 0.09022556390977443, "percentage": 2.26, "elapsed_time": "0:06:40", "remaining_time": "4:49:03"}
+{"current_steps": 14, "total_steps": 532, "loss": 1.9031141996383667, "lr": 9.62962962962963e-06, "epoch": 0.10526315789473684, "percentage": 2.63, "elapsed_time": "0:07:55", "remaining_time": "4:53:18"}
+{"current_steps": 16, "total_steps": 532, "loss": 2.545740842819214, "lr": 1.1111111111111113e-05, "epoch": 0.12030075187969924, "percentage": 3.01, "elapsed_time": "0:09:01", "remaining_time": "4:51:00"}
+{"current_steps": 18, "total_steps": 532, "loss": 3.2497336864471436, "lr": 1.2592592592592593e-05, "epoch": 0.13533834586466165, "percentage": 3.38, "elapsed_time": "0:09:41", "remaining_time": "4:36:30"}
+{"current_steps": 20, "total_steps": 532, "loss": 1.7776570320129395, "lr": 1.4074074074074075e-05, "epoch": 0.15037593984962405, "percentage": 3.76, "elapsed_time": "0:10:35", "remaining_time": "4:30:57"}
+{"current_steps": 22, "total_steps": 532, "loss": 1.9716706275939941, "lr": 1.555555555555556e-05, "epoch": 0.16541353383458646, "percentage": 4.14, "elapsed_time": "0:11:20", "remaining_time": "4:22:55"}
+{"current_steps": 24, "total_steps": 532, "loss": 1.9381436109542847, "lr": 1.7037037037037038e-05, "epoch": 0.18045112781954886, "percentage": 4.51, "elapsed_time": "0:12:25", "remaining_time": "4:23:01"}
+{"current_steps": 26, "total_steps": 532, "loss": 1.9333921670913696, "lr": 1.851851851851852e-05, "epoch": 0.19548872180451127, "percentage": 4.89, "elapsed_time": "0:13:26", "remaining_time": "4:21:42"}
+{"current_steps": 28, "total_steps": 532, "loss": 1.7134058475494385, "lr": 2e-05, "epoch": 0.21052631578947367, "percentage": 5.26, "elapsed_time": "0:14:15", "remaining_time": "4:16:30"}
+{"current_steps": 30, "total_steps": 532, "loss": 2.0251219272613525, "lr": 1.9994079505294254e-05, "epoch": 0.22556390977443608, "percentage": 5.64, "elapsed_time": "0:14:54", "remaining_time": "4:09:25"}
+{"current_steps": 32, "total_steps": 532, "loss": 1.578896164894104, "lr": 1.9976326268767035e-05, "epoch": 0.24060150375939848, "percentage": 6.02, "elapsed_time": "0:16:04", "remaining_time": "4:11:07"}
+{"current_steps": 34, "total_steps": 532, "loss": 1.5274699926376343, "lr": 1.994676502169901e-05, "epoch": 0.2556390977443609, "percentage": 6.39, "elapsed_time": "0:17:10", "remaining_time": "4:11:38"}
+{"current_steps": 36, "total_steps": 532, "loss": 1.510546326637268, "lr": 1.9905436944609424e-05, "epoch": 0.2706766917293233, "percentage": 6.77, "elapsed_time": "0:18:18", "remaining_time": "4:12:13"}
+{"current_steps": 38, "total_steps": 532, "loss": 2.110567331314087, "lr": 1.9852399609889242e-05, "epoch": 0.2857142857142857, "percentage": 7.14, "elapsed_time": "0:19:15", "remaining_time": "4:10:17"}
+{"current_steps": 40, "total_steps": 532, "loss": 1.1862285137176514, "lr": 1.9787726901599502e-05, "epoch": 0.3007518796992481, "percentage": 7.52, "elapsed_time": "0:20:12", "remaining_time": "4:08:27"}
+{"current_steps": 42, "total_steps": 532, "loss": 1.440342903137207, "lr": 1.9711508912546566e-05, "epoch": 0.3157894736842105, "percentage": 7.89, "elapsed_time": "0:21:02", "remaining_time": "4:05:24"}
+{"current_steps": 44, "total_steps": 532, "loss": 1.0962785482406616, "lr": 1.9623851818777652e-05, "epoch": 0.3308270676691729, "percentage": 8.27, "elapsed_time": "0:21:53", "remaining_time": "4:02:45"}
+{"current_steps": 46, "total_steps": 532, "loss": 1.493391513824463, "lr": 1.9524877731671482e-05, "epoch": 0.3458646616541353, "percentage": 8.65, "elapsed_time": "0:23:14", "remaining_time": "4:05:33"}
+{"current_steps": 48, "total_steps": 532, "loss": 1.2433573007583618, "lr": 1.941472452783011e-05, "epoch": 0.3609022556390977, "percentage": 9.02, "elapsed_time": "0:24:09", "remaining_time": "4:03:30"}
+{"current_steps": 50, "total_steps": 532, "loss": 1.1390293836593628, "lr": 1.9293545657008865e-05, "epoch": 0.37593984962406013, "percentage": 9.4, "elapsed_time": "0:25:06", "remaining_time": "4:02:01"}
+{"current_steps": 52, "total_steps": 532, "loss": 1.0903499126434326, "lr": 1.9161509928352017e-05, "epoch": 0.39097744360902253, "percentage": 9.77, "elapsed_time": "0:26:10", "remaining_time": "4:01:40"}
+{"current_steps": 54, "total_steps": 532, "loss": 1.1544872522354126, "lr": 1.901880127523192e-05, "epoch": 0.40601503759398494, "percentage": 10.15, "elapsed_time": "0:27:14", "remaining_time": "4:01:09"}
+{"current_steps": 56, "total_steps": 532, "loss": 1.0822100639343262, "lr": 1.886561849901922e-05, "epoch": 0.42105263157894735, "percentage": 10.53, "elapsed_time": "0:28:13", "remaining_time": "3:59:58"}
+{"current_steps": 58, "total_steps": 532, "loss": 1.0416043996810913, "lr": 1.870217499214111e-05, "epoch": 0.43609022556390975, "percentage": 10.9, "elapsed_time": "0:29:06", "remaining_time": "3:57:56"}
+{"current_steps": 60, "total_steps": 532, "loss": 1.1815505027770996, "lr": 1.8528698440813397e-05, "epoch": 0.45112781954887216, "percentage": 11.28, "elapsed_time": "0:30:05", "remaining_time": "3:56:40"}
+{"current_steps": 62, "total_steps": 532, "loss": 0.9047210812568665, "lr": 1.8345430507860478e-05, "epoch": 0.46616541353383456, "percentage": 11.65, "elapsed_time": "0:30:58", "remaining_time": "3:54:46"}
+{"current_steps": 64, "total_steps": 532, "loss": 1.201892614364624, "lr": 1.8152626496065128e-05, "epoch": 0.48120300751879697, "percentage": 12.03, "elapsed_time": "0:32:01", "remaining_time": "3:54:10"}
+{"current_steps": 66, "total_steps": 532, "loss": 1.1795772314071655, "lr": 1.7950554992517014e-05, "epoch": 0.49624060150375937, "percentage": 12.41, "elapsed_time": "0:33:10", "remaining_time": "3:54:15"}
+{"current_steps": 68, "total_steps": 532, "loss": 1.3534270524978638, "lr": 1.7739497494455412e-05, "epoch": 0.5112781954887218, "percentage": 12.78, "elapsed_time": "0:34:15", "remaining_time": "3:53:46"}
+{"current_steps": 70, "total_steps": 532, "loss": 1.125345230102539, "lr": 1.7519748017127354e-05, "epoch": 0.5263157894736842, "percentage": 13.16, "elapsed_time": "0:35:08", "remaining_time": "3:51:59"}
+{"current_steps": 72, "total_steps": 532, "loss": 0.8090606331825256, "lr": 1.729161268420746e-05, "epoch": 0.5413533834586466, "percentage": 13.53, "elapsed_time": "0:35:57", "remaining_time": "3:49:44"}
+{"current_steps": 74, "total_steps": 532, "loss": 0.9213717579841614, "lr": 1.7055409301350013e-05, "epoch": 0.556390977443609, "percentage": 13.91, "elapsed_time": "0:36:53", "remaining_time": "3:48:17"}
+{"current_steps": 76, "total_steps": 532, "loss": 1.0821505784988403, "lr": 1.681146691346742e-05, "epoch": 0.5714285714285714, "percentage": 14.29, "elapsed_time": "0:37:48", "remaining_time": "3:46:48"}
+{"current_steps": 78, "total_steps": 532, "loss": 1.300316333770752, "lr": 1.6560125346351663e-05, "epoch": 0.5864661654135338, "percentage": 14.66, "elapsed_time": "0:38:57", "remaining_time": "3:46:48"}
+{"current_steps": 80, "total_steps": 532, "loss": 0.6484270691871643, "lr": 1.6301734733277442e-05, "epoch": 0.6015037593984962, "percentage": 15.04, "elapsed_time": "0:39:28", "remaining_time": "3:43:03"}
+{"current_steps": 82, "total_steps": 532, "loss": 1.2577356100082397, "lr": 1.603665502724633e-05, "epoch": 0.6165413533834586, "percentage": 15.41, "elapsed_time": "0:40:42", "remaining_time": "3:43:24"}
+{"current_steps": 84, "total_steps": 532, "loss": 1.3543009757995605, "lr": 1.576525549955156e-05, "epoch": 0.631578947368421, "percentage": 15.79, "elapsed_time": "0:42:02", "remaining_time": "3:44:12"}
+{"current_steps": 86, "total_steps": 532, "loss": 0.9327285885810852, "lr": 1.548791422536178e-05, "epoch": 0.6466165413533834, "percentage": 16.17, "elapsed_time": "0:42:46", "remaining_time": "3:41:50"}
+{"current_steps": 88, "total_steps": 532, "loss": 1.1237722635269165, "lr": 1.5205017557040656e-05, "epoch": 0.6616541353383458, "percentage": 16.54, "elapsed_time": "0:43:41", "remaining_time": "3:40:24"}
+{"current_steps": 90, "total_steps": 532, "loss": 1.1964070796966553, "lr": 1.4916959585935732e-05, "epoch": 0.6766917293233082, "percentage": 16.92, "elapsed_time": "0:44:42", "remaining_time": "3:39:34"}
+{"current_steps": 92, "total_steps": 532, "loss": 1.0308165550231934, "lr": 1.4624141593386507e-05, "epoch": 0.6917293233082706, "percentage": 17.29, "elapsed_time": "0:45:29", "remaining_time": "3:37:35"}
+{"current_steps": 94, "total_steps": 532, "loss": 0.9982426762580872, "lr": 1.4326971491716427e-05, "epoch": 0.706766917293233, "percentage": 17.67, "elapsed_time": "0:46:28", "remaining_time": "3:36:34"}
+{"current_steps": 96, "total_steps": 532, "loss": 1.3940727710723877, "lr": 1.402586325598752e-05, "epoch": 0.7218045112781954, "percentage": 18.05, "elapsed_time": "0:47:46", "remaining_time": "3:36:59"}
+{"current_steps": 98, "total_steps": 532, "loss": 1.1801196336746216, "lr": 1.3721236347309314e-05, "epoch": 0.7368421052631579, "percentage": 18.42, "elapsed_time": "0:49:01", "remaining_time": "3:37:05"}
+{"current_steps": 100, "total_steps": 532, "loss": 0.6430416703224182, "lr": 1.3413515128505363e-05, "epoch": 0.7518796992481203, "percentage": 18.8, "elapsed_time": "0:49:43", "remaining_time": "3:34:48"}
+{"current_steps": 102, "total_steps": 532, "loss": 1.3783133029937744, "lr": 1.3103128272951363e-05, "epoch": 0.7669172932330827, "percentage": 19.17, "elapsed_time": "0:51:01", "remaining_time": "3:35:04"}
+{"current_steps": 104, "total_steps": 532, "loss": 0.9889219403266907, "lr": 1.2790508167408509e-05, "epoch": 0.7819548872180451, "percentage": 19.55, "elapsed_time": "0:51:48", "remaining_time": "3:33:12"}
+{"current_steps": 106, "total_steps": 532, "loss": 0.6871194243431091, "lr": 1.2476090309683804e-05, "epoch": 0.7969924812030075, "percentage": 19.92, "elapsed_time": "0:52:37", "remaining_time": "3:31:30"}
+{"current_steps": 108, "total_steps": 532, "loss": 1.0068978071212769, "lr": 1.2160312701956553e-05, "epoch": 0.8120300751879699, "percentage": 20.3, "elapsed_time": "0:53:30", "remaining_time": "3:30:02"}
+{"current_steps": 110, "total_steps": 532, "loss": 0.8988245725631714, "lr": 1.1843615240616111e-05, "epoch": 0.8270676691729323, "percentage": 20.68, "elapsed_time": "0:54:34", "remaining_time": "3:29:21"}
+{"current_steps": 112, "total_steps": 532, "loss": 0.7688661813735962, "lr": 1.1526439103460874e-05, "epoch": 0.8421052631578947, "percentage": 21.05, "elapsed_time": "0:55:19", "remaining_time": "3:27:26"}
+{"current_steps": 114, "total_steps": 532, "loss": 1.0024021863937378, "lr": 1.120922613511221e-05, "epoch": 0.8571428571428571, "percentage": 21.43, "elapsed_time": "0:56:10", "remaining_time": "3:25:58"}
+{"current_steps": 116, "total_steps": 532, "loss": 1.4191375970840454, "lr": 1.0892418231499461e-05, "epoch": 0.8721804511278195, "percentage": 21.8, "elapsed_time": "0:57:27", "remaining_time": "3:26:02"}
+{"current_steps": 118, "total_steps": 532, "loss": 1.073761224746704, "lr": 1.057645672427347e-05, "epoch": 0.8872180451127819, "percentage": 22.18, "elapsed_time": "0:58:28", "remaining_time": "3:25:08"}
+{"current_steps": 120, "total_steps": 532, "loss": 0.9056495428085327, "lr": 1.0261781766006174e-05, "epoch": 0.9022556390977443, "percentage": 22.56, "elapsed_time": "0:59:24", "remaining_time": "3:23:58"}
+{"current_steps": 122, "total_steps": 532, "loss": 0.8672894835472107, "lr": 9.948831717032738e-06, "epoch": 0.9172932330827067, "percentage": 22.93, "elapsed_time": "1:00:10", "remaining_time": "3:22:14"}
+{"current_steps": 124, "total_steps": 532, "loss": 0.827739417552948, "lr": 9.638042534790373e-06, "epoch": 0.9323308270676691, "percentage": 23.31, "elapsed_time": "1:00:56", "remaining_time": "3:20:32"}
+{"current_steps": 126, "total_steps": 532, "loss": 0.7799423933029175, "lr": 9.329847166504497e-06, "epoch": 0.9473684210526315, "percentage": 23.68, "elapsed_time": "1:01:50", "remaining_time": "3:19:17"}
+{"current_steps": 128, "total_steps": 532, "loss": 1.0653791427612305, "lr": 9.024674946068357e-06, "epoch": 0.9624060150375939, "percentage": 24.06, "elapsed_time": "1:02:56", "remaining_time": "3:18:38"}
+{"current_steps": 130, "total_steps": 532, "loss": 0.9135006666183472, "lr": 8.722950995956172e-06, "epoch": 0.9774436090225563, "percentage": 24.44, "elapsed_time": "1:03:56", "remaining_time": "3:17:43"}
+{"current_steps": 132, "total_steps": 532, "loss": 1.0402815341949463, "lr": 8.425095635003053e-06, "epoch": 0.9924812030075187, "percentage": 24.81, "elapsed_time": "1:04:50", "remaining_time": "3:16:28"}
+{"current_steps": 134, "total_steps": 532, "loss": 0.7629735469818115, "lr": 8.13152379287667e-06, "epoch": 1.0075187969924813, "percentage": 25.19, "elapsed_time": "1:05:38", "remaining_time": "3:14:56"}
+{"current_steps": 136, "total_steps": 532, "loss": 0.9513287544250488, "lr": 7.842644432056336e-06, "epoch": 1.0225563909774436, "percentage": 25.56, "elapsed_time": "1:06:42", "remaining_time": "3:14:13"}
+{"current_steps": 138, "total_steps": 532, "loss": 0.6946667432785034, "lr": 7.55885997812472e-06, "epoch": 1.037593984962406, "percentage": 25.94, "elapsed_time": "1:07:30", "remaining_time": "3:12:43"}
+{"current_steps": 140, "total_steps": 532, "loss": 0.7876754403114319, "lr": 7.280565759165833e-06, "epoch": 1.0526315789473684, "percentage": 26.32, "elapsed_time": "1:08:11", "remaining_time": "3:10:57"}
+{"current_steps": 142, "total_steps": 532, "loss": 0.8406305909156799, "lr": 7.008149455050264e-06, "epoch": 1.0676691729323309, "percentage": 26.69, "elapsed_time": "1:09:10", "remaining_time": "3:09:59"}
+{"current_steps": 144, "total_steps": 532, "loss": 1.0424906015396118, "lr": 6.741990557374784e-06, "epoch": 1.0827067669172932, "percentage": 27.07, "elapsed_time": "1:10:08", "remaining_time": "3:08:59"}
+{"current_steps": 146, "total_steps": 532, "loss": 1.0722497701644897, "lr": 6.4824598408087015e-06, "epoch": 1.0977443609022557, "percentage": 27.44, "elapsed_time": "1:11:18", "remaining_time": "3:08:32"}
+{"current_steps": 148, "total_steps": 532, "loss": 1.0312786102294922, "lr": 6.229918846583414e-06, "epoch": 1.112781954887218, "percentage": 27.82, "elapsed_time": "1:12:28", "remaining_time": "3:08:02"}
+{"current_steps": 150, "total_steps": 532, "loss": 1.0075746774673462, "lr": 5.984719378844628e-06, "epoch": 1.1278195488721805, "percentage": 28.2, "elapsed_time": "1:13:14", "remaining_time": "3:06:31"}
+{"current_steps": 152, "total_steps": 532, "loss": 1.0427347421646118, "lr": 5.7472030145689604e-06, "epoch": 1.1428571428571428, "percentage": 28.57, "elapsed_time": "1:14:29", "remaining_time": "3:06:14"}
+{"current_steps": 154, "total_steps": 532, "loss": 1.0927042961120605, "lr": 5.51770062772752e-06, "epoch": 1.1578947368421053, "percentage": 28.95, "elapsed_time": "1:15:37", "remaining_time": "3:05:38"}
+{"current_steps": 156, "total_steps": 532, "loss": 0.9705473780632019, "lr": 5.296531928359431e-06, "epoch": 1.1729323308270676, "percentage": 29.32, "elapsed_time": "1:16:28", "remaining_time": "3:04:20"}
+{"current_steps": 158, "total_steps": 532, "loss": 1.0172467231750488, "lr": 5.084005017197318e-06, "epoch": 1.1879699248120301, "percentage": 29.7, "elapsed_time": "1:17:34", "remaining_time": "3:03:37"}
+{"current_steps": 160, "total_steps": 532, "loss": 0.5409541726112366, "lr": 4.8804159564652665e-06, "epoch": 1.2030075187969924, "percentage": 30.08, "elapsed_time": "1:18:20", "remaining_time": "3:02:07"}
+{"current_steps": 162, "total_steps": 532, "loss": 0.9430153965950012, "lr": 4.686048357447095e-06, "epoch": 1.218045112781955, "percentage": 30.45, "elapsed_time": "1:19:08", "remaining_time": "3:00:45"}
+{"current_steps": 164, "total_steps": 532, "loss": 1.178081750869751, "lr": 4.501172985399498e-06, "epoch": 1.2330827067669172, "percentage": 30.83, "elapsed_time": "1:20:18", "remaining_time": "3:00:12"}
+{"current_steps": 166, "total_steps": 532, "loss": 0.8844167590141296, "lr": 4.326047382360457e-06, "epoch": 1.2481203007518797, "percentage": 31.2, "elapsed_time": "1:21:17", "remaining_time": "2:59:13"}
+{"current_steps": 168, "total_steps": 532, "loss": 0.7063813209533691, "lr": 4.160915508378359e-06, "epoch": 1.263157894736842, "percentage": 31.58, "elapsed_time": "1:22:08", "remaining_time": "2:57:57"}
+{"current_steps": 170, "total_steps": 532, "loss": 0.7762787938117981, "lr": 4.006007401661596e-06, "epoch": 1.2781954887218046, "percentage": 31.95, "elapsed_time": "1:23:02", "remaining_time": "2:56:48"}
+{"current_steps": 172, "total_steps": 532, "loss": 0.937335193157196, "lr": 3.861538858122092e-06, "epoch": 1.2932330827067668, "percentage": 32.33, "elapsed_time": "1:23:57", "remaining_time": "2:55:43"}
+{"current_steps": 174, "total_steps": 532, "loss": 0.9747655987739563, "lr": 3.727711130759182e-06, "epoch": 1.3082706766917294, "percentage": 32.71, "elapsed_time": "1:25:02", "remaining_time": "2:54:58"}
+{"current_steps": 176, "total_steps": 532, "loss": 0.7746855020523071, "lr": 3.6047106493025923e-06, "epoch": 1.3233082706766917, "percentage": 33.08, "elapsed_time": "1:25:48", "remaining_time": "2:53:34"}
+{"current_steps": 178, "total_steps": 532, "loss": 0.6926825642585754, "lr": 3.492708760505093e-06, "epoch": 1.3383458646616542, "percentage": 33.46, "elapsed_time": "1:26:37", "remaining_time": "2:52:15"}
+{"current_steps": 180, "total_steps": 532, "loss": 1.2502151727676392, "lr": 3.3918614894466045e-06, "epoch": 1.3533834586466165, "percentage": 33.83, "elapsed_time": "1:27:52", "remaining_time": "2:51:50"}
+{"current_steps": 182, "total_steps": 532, "loss": 0.9960780143737793, "lr": 3.3023093221822746e-06, "epoch": 1.368421052631579, "percentage": 34.21, "elapsed_time": "1:28:53", "remaining_time": "2:50:56"}
+{"current_steps": 184, "total_steps": 532, "loss": 0.8326720595359802, "lr": 3.224177010037323e-06, "epoch": 1.3834586466165413, "percentage": 34.59, "elapsed_time": "1:29:39", "remaining_time": "2:49:33"}
+{"current_steps": 186, "total_steps": 532, "loss": 1.1920455694198608, "lr": 3.1575733958212563e-06, "epoch": 1.3984962406015038, "percentage": 34.96, "elapsed_time": "1:30:57", "remaining_time": "2:49:12"}
+{"current_steps": 188, "total_steps": 532, "loss": 0.5979896783828735, "lr": 3.1025912622035687e-06, "epoch": 1.413533834586466, "percentage": 35.34, "elapsed_time": "1:31:35", "remaining_time": "2:47:36"}
+{"current_steps": 190, "total_steps": 532, "loss": 0.969947099685669, "lr": 3.0593072024621396e-06, "epoch": 1.4285714285714286, "percentage": 35.71, "elapsed_time": "1:32:39", "remaining_time": "2:46:47"}
+{"current_steps": 192, "total_steps": 532, "loss": 0.7010709643363953, "lr": 3.0277815137843917e-06, "epoch": 1.443609022556391, "percentage": 36.09, "elapsed_time": "1:33:28", "remaining_time": "2:45:31"}
+{"current_steps": 194, "total_steps": 532, "loss": 0.7660905718803406, "lr": 3.008058113269836e-06, "epoch": 1.4586466165413534, "percentage": 36.47, "elapsed_time": "1:34:13", "remaining_time": "2:44:09"}
+{"current_steps": 196, "total_steps": 532, "loss": 0.9958880543708801, "lr": 3.0001644767510154e-06, "epoch": 1.4736842105263157, "percentage": 36.84, "elapsed_time": "1:35:19", "remaining_time": "2:43:24"}
+{"current_steps": 198, "total_steps": 532, "loss": 0.9368724226951599, "lr": 3.0041116005181016e-06, "epoch": 1.4887218045112782, "percentage": 37.22, "elapsed_time": "1:36:24", "remaining_time": "2:42:36"}
+{"current_steps": 200, "total_steps": 532, "loss": 0.9826479554176331, "lr": 3.0198939860004202e-06, "epoch": 1.5037593984962405, "percentage": 37.59, "elapsed_time": "1:37:30", "remaining_time": "2:41:51"}
+{"current_steps": 202, "total_steps": 532, "loss": 1.3954254388809204, "lr": 3.0474896474262772e-06, "epoch": 1.518796992481203, "percentage": 37.97, "elapsed_time": "1:38:49", "remaining_time": "2:41:26"}
+{"current_steps": 204, "total_steps": 532, "loss": 1.0462982654571533, "lr": 3.08686014245041e-06, "epoch": 1.5338345864661656, "percentage": 38.35, "elapsed_time": "1:39:56", "remaining_time": "2:40:40"}
+{"current_steps": 206, "total_steps": 532, "loss": 0.8277729153633118, "lr": 3.1379506257063825e-06, "epoch": 1.5488721804511278, "percentage": 38.72, "elapsed_time": "1:40:53", "remaining_time": "2:39:39"}
+{"current_steps": 208, "total_steps": 532, "loss": 0.985795259475708, "lr": 3.20068992520934e-06, "epoch": 1.5639097744360901, "percentage": 39.1, "elapsed_time": "1:41:50", "remaining_time": "2:38:38"}
+{"current_steps": 210, "total_steps": 532, "loss": 0.908364474773407, "lr": 3.274990641502683e-06, "epoch": 1.5789473684210527, "percentage": 39.47, "elapsed_time": "1:42:44", "remaining_time": "2:37:32"}
+{"current_steps": 212, "total_steps": 532, "loss": 0.6370347142219543, "lr": 3.3607492694105405e-06, "epoch": 1.5939849624060152, "percentage": 39.85, "elapsed_time": "1:43:31", "remaining_time": "2:36:16"}
+{"current_steps": 214, "total_steps": 532, "loss": 0.9584491848945618, "lr": 3.457846342226442e-06, "epoch": 1.6090225563909775, "percentage": 40.23, "elapsed_time": "1:44:31", "remaining_time": "2:35:19"}
+{"current_steps": 216, "total_steps": 532, "loss": 1.0774601697921753, "lr": 3.5661465981373183e-06, "epoch": 1.6240601503759398, "percentage": 40.6, "elapsed_time": "1:45:30", "remaining_time": "2:34:20"}
+{"current_steps": 218, "total_steps": 532, "loss": 0.7987947463989258, "lr": 3.6854991686509906e-06, "epoch": 1.6390977443609023, "percentage": 40.98, "elapsed_time": "1:46:24", "remaining_time": "2:33:16"}
+{"current_steps": 220, "total_steps": 532, "loss": 0.8292507529258728, "lr": 3.815737788764674e-06, "epoch": 1.6541353383458648, "percentage": 41.35, "elapsed_time": "1:47:24", "remaining_time": "2:32:18"}
+{"current_steps": 222, "total_steps": 532, "loss": 0.9586336612701416, "lr": 3.956681028581693e-06, "epoch": 1.669172932330827, "percentage": 41.73, "elapsed_time": "1:48:23", "remaining_time": "2:31:21"}
+{"current_steps": 224, "total_steps": 532, "loss": 0.8150299191474915, "lr": 4.108132546053779e-06, "epoch": 1.6842105263157894, "percentage": 42.11, "elapsed_time": "1:49:08", "remaining_time": "2:30:04"}
+{"current_steps": 226, "total_steps": 532, "loss": 1.0946331024169922, "lr": 4.269881360496842e-06, "epoch": 1.699248120300752, "percentage": 42.48, "elapsed_time": "1:50:06", "remaining_time": "2:29:05"}
+{"current_steps": 228, "total_steps": 532, "loss": 0.9061790108680725, "lr": 4.441702146499222e-06, "epoch": 1.7142857142857144, "percentage": 42.86, "elapsed_time": "1:51:03", "remaining_time": "2:28:04"}
+{"current_steps": 230, "total_steps": 532, "loss": 1.0011441707611084, "lr": 4.623355547812946e-06, "epoch": 1.7293233082706767, "percentage": 43.23, "elapsed_time": "1:52:01", "remaining_time": "2:27:05"}
+{"current_steps": 232, "total_steps": 532, "loss": 1.1533119678497314, "lr": 4.814588510790782e-06, "epoch": 1.744360902255639, "percentage": 43.61, "elapsed_time": "1:53:13", "remaining_time": "2:26:25"}
+{"current_steps": 234, "total_steps": 532, "loss": 1.1073795557022095, "lr": 5.01513463690452e-06, "epoch": 1.7593984962406015, "percentage": 43.98, "elapsed_time": "1:54:13", "remaining_time": "2:25:28"}
+{"current_steps": 236, "total_steps": 532, "loss": 1.148139476776123, "lr": 5.224714553853478e-06, "epoch": 1.774436090225564, "percentage": 44.36, "elapsed_time": "1:55:34", "remaining_time": "2:24:57"}
+{"current_steps": 238, "total_steps": 532, "loss": 0.6271846294403076, "lr": 5.443036304746191e-06, "epoch": 1.7894736842105263, "percentage": 44.74, "elapsed_time": "1:56:17", "remaining_time": "2:23:39"}
+{"current_steps": 240, "total_steps": 532, "loss": 1.0237879753112793, "lr": 5.66979575481317e-06, "epoch": 1.8045112781954886, "percentage": 45.11, "elapsed_time": "1:57:10", "remaining_time": "2:22:33"}
+{"current_steps": 242, "total_steps": 532, "loss": 0.7585715055465698, "lr": 5.904677015084159e-06, "epoch": 1.8195488721804511, "percentage": 45.49, "elapsed_time": "1:58:03", "remaining_time": "2:21:28"}
+{"current_steps": 244, "total_steps": 532, "loss": 1.1932705640792847, "lr": 6.147352882439652e-06, "epoch": 1.8345864661654137, "percentage": 45.86, "elapsed_time": "1:59:04", "remaining_time": "2:20:32"}
+{"current_steps": 246, "total_steps": 532, "loss": 1.1568275690078735, "lr": 6.397485295423669e-06, "epoch": 1.849624060150376, "percentage": 46.24, "elapsed_time": "2:00:04", "remaining_time": "2:19:35"}
+{"current_steps": 248, "total_steps": 532, "loss": 0.9782670736312866, "lr": 6.6547258051828426e-06, "epoch": 1.8646616541353382, "percentage": 46.62, "elapsed_time": "2:00:54", "remaining_time": "2:18:27"}
+{"current_steps": 250, "total_steps": 532, "loss": 1.2261128425598145, "lr": 6.918716060875743e-06, "epoch": 1.8796992481203008, "percentage": 46.99, "elapsed_time": "2:02:01", "remaining_time": "2:17:39"}
+{"current_steps": 252, "total_steps": 532, "loss": 1.0725328922271729, "lr": 7.1890883088761885e-06, "epoch": 1.8947368421052633, "percentage": 47.37, "elapsed_time": "2:03:01", "remaining_time": "2:16:41"}
+{"current_steps": 254, "total_steps": 532, "loss": 1.2271314859390259, "lr": 7.4654659050752845e-06, "epoch": 1.9097744360902256, "percentage": 47.74, "elapsed_time": "2:04:21", "remaining_time": "2:16:07"}
+{"current_steps": 256, "total_steps": 532, "loss": 0.8813698291778564, "lr": 7.747463839568292e-06, "epoch": 1.9248120300751879, "percentage": 48.12, "elapsed_time": "2:05:09", "remaining_time": "2:14:56"}
+{"current_steps": 258, "total_steps": 532, "loss": 0.5630529522895813, "lr": 8.034689272995649e-06, "epoch": 1.9398496240601504, "percentage": 48.5, "elapsed_time": "2:05:41", "remaining_time": "2:13:29"}
+{"current_steps": 260, "total_steps": 532, "loss": 1.0115076303482056, "lr": 8.32674208379076e-06, "epoch": 1.954887218045113, "percentage": 48.87, "elapsed_time": "2:06:46", "remaining_time": "2:12:37"}
+{"current_steps": 262, "total_steps": 532, "loss": 0.9677861332893372, "lr": 8.623215425572433e-06, "epoch": 1.9699248120300752, "percentage": 49.25, "elapsed_time": "2:08:50", "remaining_time": "2:12:46"}
+{"current_steps": 264, "total_steps": 532, "loss": 0.9606061577796936, "lr": 8.923696293905279e-06, "epoch": 1.9849624060150375, "percentage": 49.62, "elapsed_time": "2:09:43", "remaining_time": "2:11:41"}
+{"current_steps": 266, "total_steps": 532, "loss": 0.8462581038475037, "lr": 9.227766101638668e-06, "epoch": 2.0, "percentage": 50.0, "elapsed_time": "2:10:38", "remaining_time": "2:10:38"}
+{"current_steps": 268, "total_steps": 532, "loss": 0.5168899893760681, "lr": 9.535001262022755e-06, "epoch": 2.0150375939849625, "percentage": 50.38, "elapsed_time": "2:11:24", "remaining_time": "2:09:26"}
+{"current_steps": 270, "total_steps": 532, "loss": 0.725158154964447, "lr": 9.844973778789198e-06, "epoch": 2.030075187969925, "percentage": 50.75, "elapsed_time": "2:12:16", "remaining_time": "2:08:21"}
+{"current_steps": 272, "total_steps": 532, "loss": 0.5184513330459595, "lr": 1.0157251842374587e-05, "epoch": 2.045112781954887, "percentage": 51.13, "elapsed_time": "2:12:54", "remaining_time": "2:07:02"}
+{"current_steps": 274, "total_steps": 532, "loss": 1.0087833404541016, "lr": 1.0471400431456062e-05, "epoch": 2.0601503759398496, "percentage": 51.5, "elapsed_time": "2:14:03", "remaining_time": "2:06:14"}
+{"current_steps": 276, "total_steps": 532, "loss": 0.9181981682777405, "lr": 1.078698191896105e-05, "epoch": 2.075187969924812, "percentage": 51.88, "elapsed_time": "2:14:59", "remaining_time": "2:05:12"}
+{"current_steps": 278, "total_steps": 532, "loss": 0.867568850517273, "lr": 1.1103556681706949e-05, "epoch": 2.090225563909774, "percentage": 52.26, "elapsed_time": "2:15:52", "remaining_time": "2:04:08"}
+{"current_steps": 280, "total_steps": 532, "loss": 0.8937767744064331, "lr": 1.1420683712821577e-05, "epoch": 2.1052631578947367, "percentage": 52.63, "elapsed_time": "2:16:54", "remaining_time": "2:03:13"}
+{"current_steps": 282, "total_steps": 532, "loss": 1.0325863361358643, "lr": 1.1737921236091097e-05, "epoch": 2.1203007518796992, "percentage": 53.01, "elapsed_time": "2:17:56", "remaining_time": "2:02:17"}
+{"current_steps": 284, "total_steps": 532, "loss": 0.6633681654930115, "lr": 1.2054827321379756e-05, "epoch": 2.1353383458646618, "percentage": 53.38, "elapsed_time": "2:18:45", "remaining_time": "2:01:10"}
+{"current_steps": 286, "total_steps": 532, "loss": 0.998541533946991, "lr": 1.237096050026399e-05, "epoch": 2.1503759398496243, "percentage": 53.76, "elapsed_time": "2:19:51", "remaining_time": "2:00:17"}
+{"current_steps": 288, "total_steps": 532, "loss": 1.1486636400222778, "lr": 1.2685880381023343e-05, "epoch": 2.1654135338345863, "percentage": 54.14, "elapsed_time": "2:20:53", "remaining_time": "1:59:22"}
+{"current_steps": 290, "total_steps": 532, "loss": 1.1767593622207642, "lr": 1.2999148262131527e-05, "epoch": 2.180451127819549, "percentage": 54.51, "elapsed_time": "2:22:04", "remaining_time": "1:58:33"}
+{"current_steps": 292, "total_steps": 532, "loss": 0.8024111986160278, "lr": 1.3310327743392861e-05, "epoch": 2.1954887218045114, "percentage": 54.89, "elapsed_time": "2:22:55", "remaining_time": "1:57:28"}
+{"current_steps": 294, "total_steps": 532, "loss": 0.9202947616577148, "lr": 1.3618985333872927e-05, "epoch": 2.2105263157894735, "percentage": 55.26, "elapsed_time": "2:23:50", "remaining_time": "1:56:26"}
+{"current_steps": 296, "total_steps": 532, "loss": 0.3760508596897125, "lr": 1.3924691055776338e-05, "epoch": 2.225563909774436, "percentage": 55.64, "elapsed_time": "2:24:16", "remaining_time": "1:55:02"}
+{"current_steps": 298, "total_steps": 532, "loss": 1.2139897346496582, "lr": 1.4227019043430623e-05, "epoch": 2.2406015037593985, "percentage": 56.02, "elapsed_time": "2:25:38", "remaining_time": "1:54:21"}
+{"current_steps": 300, "total_steps": 532, "loss": 1.1467747688293457, "lr": 1.4525548136541599e-05, "epoch": 2.255639097744361, "percentage": 56.39, "elapsed_time": "2:26:50", "remaining_time": "1:53:33"}
+{"current_steps": 302, "total_steps": 532, "loss": 1.152048110961914, "lr": 1.4819862466893891e-05, "epoch": 2.2706766917293235, "percentage": 56.77, "elapsed_time": "2:28:00", "remaining_time": "1:52:43"}
+{"current_steps": 304, "total_steps": 532, "loss": 0.8999449610710144, "lr": 1.5109552037679353e-05, "epoch": 2.2857142857142856, "percentage": 57.14, "elapsed_time": "2:29:03", "remaining_time": "1:51:47"}
+{"current_steps": 306, "total_steps": 532, "loss": 0.8530235290527344, "lr": 1.539421329464618e-05, "epoch": 2.300751879699248, "percentage": 57.52, "elapsed_time": "2:29:47", "remaining_time": "1:50:38"}
+{"current_steps": 308, "total_steps": 532, "loss": 1.1707788705825806, "lr": 1.5673449688273246e-05, "epoch": 2.3157894736842106, "percentage": 57.89, "elapsed_time": "2:31:08", "remaining_time": "1:49:55"}
+{"current_steps": 310, "total_steps": 532, "loss": 0.543804407119751, "lr": 1.5946872226186394e-05, "epoch": 2.3308270676691727, "percentage": 58.27, "elapsed_time": "2:31:58", "remaining_time": "1:48:49"}
+{"current_steps": 312, "total_steps": 532, "loss": 0.9259825944900513, "lr": 1.621410001504725e-05, "epoch": 2.345864661654135, "percentage": 58.65, "elapsed_time": "2:33:01", "remaining_time": "1:47:54"}
+{"current_steps": 314, "total_steps": 532, "loss": 0.6412667036056519, "lr": 1.6474760791159528e-05, "epoch": 2.3609022556390977, "percentage": 59.02, "elapsed_time": "2:33:42", "remaining_time": "1:46:43"}
+{"current_steps": 316, "total_steps": 532, "loss": 1.0553992986679077, "lr": 1.672849143905383e-05, "epoch": 2.3759398496240602, "percentage": 59.4, "elapsed_time": "2:34:53", "remaining_time": "1:45:52"}
+{"current_steps": 318, "total_steps": 532, "loss": 0.9918773174285889, "lr": 1.697493849732838e-05, "epoch": 2.3909774436090228, "percentage": 59.77, "elapsed_time": "2:35:55", "remaining_time": "1:44:56"}
+{"current_steps": 320, "total_steps": 532, "loss": 1.183601975440979, "lr": 1.7213758651041114e-05, "epoch": 2.406015037593985, "percentage": 60.15, "elapsed_time": "2:37:04", "remaining_time": "1:44:03"}
+{"current_steps": 322, "total_steps": 532, "loss": 1.2674732208251953, "lr": 1.7444619209967163e-05, "epoch": 2.4210526315789473, "percentage": 60.53, "elapsed_time": "2:38:14", "remaining_time": "1:43:12"}
+{"current_steps": 324, "total_steps": 532, "loss": 0.9618882536888123, "lr": 1.766719857205551e-05, "epoch": 2.43609022556391, "percentage": 60.9, "elapsed_time": "2:39:21", "remaining_time": "1:42:18"}
+{"current_steps": 326, "total_steps": 532, "loss": 1.1368170976638794, "lr": 1.7881186671439107e-05, "epoch": 2.451127819548872, "percentage": 61.28, "elapsed_time": "2:40:22", "remaining_time": "1:41:20"}
+{"current_steps": 328, "total_steps": 532, "loss": 0.9204502701759338, "lr": 1.8086285410374563e-05, "epoch": 2.4661654135338344, "percentage": 61.65, "elapsed_time": "2:41:21", "remaining_time": "1:40:21"}
+{"current_steps": 330, "total_steps": 532, "loss": 0.7325559258460999, "lr": 1.8282209074509393e-05, "epoch": 2.481203007518797, "percentage": 62.03, "elapsed_time": "2:42:03", "remaining_time": "1:39:11"}
+{"current_steps": 332, "total_steps": 532, "loss": 1.1228058338165283, "lr": 1.8468684730898602e-05, "epoch": 2.4962406015037595, "percentage": 62.41, "elapsed_time": "2:43:13", "remaining_time": "1:38:19"}
+{"current_steps": 334, "total_steps": 532, "loss": 0.6551952958106995, "lr": 1.864545260821598e-05, "epoch": 2.511278195488722, "percentage": 62.78, "elapsed_time": "2:43:57", "remaining_time": "1:37:11"}
+{"current_steps": 336, "total_steps": 532, "loss": 0.7751361727714539, "lr": 1.8812266458630505e-05, "epoch": 2.526315789473684, "percentage": 63.16, "elapsed_time": "2:44:43", "remaining_time": "1:36:05"}
+{"current_steps": 338, "total_steps": 532, "loss": 0.7552028298377991, "lr": 1.8968893900843786e-05, "epoch": 2.5413533834586466, "percentage": 63.53, "elapsed_time": "2:45:38", "remaining_time": "1:35:04"}
+{"current_steps": 340, "total_steps": 532, "loss": 1.1345906257629395, "lr": 1.9115116743810537e-05, "epoch": 2.556390977443609, "percentage": 63.91, "elapsed_time": "2:46:38", "remaining_time": "1:34:06"}
+{"current_steps": 342, "total_steps": 532, "loss": 1.0457109212875366, "lr": 1.9250731290691326e-05, "epoch": 2.571428571428571, "percentage": 64.29, "elapsed_time": "2:47:44", "remaining_time": "1:33:11"}
+{"current_steps": 344, "total_steps": 532, "loss": 0.8526303768157959, "lr": 1.9375548622613952e-05, "epoch": 2.5864661654135337, "percentage": 64.66, "elapsed_time": "2:48:47", "remaining_time": "1:32:14"}
+{"current_steps": 346, "total_steps": 532, "loss": 0.6485392451286316, "lr": 1.9489394861848333e-05, "epoch": 2.601503759398496, "percentage": 65.04, "elapsed_time": "2:49:42", "remaining_time": "1:31:13"}
+{"current_steps": 348, "total_steps": 532, "loss": 0.8217464685440063, "lr": 1.9592111414028205e-05, "epoch": 2.6165413533834587, "percentage": 65.41, "elapsed_time": "2:50:32", "remaining_time": "1:30:10"}
+{"current_steps": 350, "total_steps": 532, "loss": 0.9268519282341003, "lr": 1.9683555189082178e-05, "epoch": 2.6315789473684212, "percentage": 65.79, "elapsed_time": "2:51:20", "remaining_time": "1:29:05"}
+{"current_steps": 352, "total_steps": 532, "loss": 0.7672562599182129, "lr": 1.9763598800566478e-05, "epoch": 2.6466165413533833, "percentage": 66.17, "elapsed_time": "2:52:15", "remaining_time": "1:28:05"}
+{"current_steps": 354, "total_steps": 532, "loss": 1.0508689880371094, "lr": 1.983213074312156e-05, "epoch": 2.661654135338346, "percentage": 66.54, "elapsed_time": "2:53:35", "remaining_time": "1:27:17"}
+{"current_steps": 356, "total_steps": 532, "loss": 0.8092398047447205, "lr": 1.9889055547805527e-05, "epoch": 2.6766917293233083, "percentage": 66.92, "elapsed_time": "2:54:35", "remaining_time": "1:26:18"}
+{"current_steps": 358, "total_steps": 532, "loss": 0.4492420554161072, "lr": 1.993429391508785e-05, "epoch": 2.6917293233082704, "percentage": 67.29, "elapsed_time": "2:55:16", "remaining_time": "1:25:11"}
+{"current_steps": 360, "total_steps": 532, "loss": 1.1782996654510498, "lr": 1.9967782825318183e-05, "epoch": 2.706766917293233, "percentage": 67.67, "elapsed_time": "2:56:30", "remaining_time": "1:24:19"}
+{"current_steps": 362, "total_steps": 532, "loss": 0.6154137253761292, "lr": 1.998947562651635e-05, "epoch": 2.7218045112781954, "percentage": 68.05, "elapsed_time": "2:57:03", "remaining_time": "1:23:09"}
+{"current_steps": 364, "total_steps": 532, "loss": 0.7517685890197754, "lr": 1.999934209936126e-05, "epoch": 2.736842105263158, "percentage": 68.42, "elapsed_time": "2:57:59", "remaining_time": "1:22:09"}
+{"current_steps": 366, "total_steps": 532, "loss": 0.8893219232559204, "lr": 1.9997368499288147e-05, "epoch": 2.7518796992481205, "percentage": 68.8, "elapsed_time": "2:58:52", "remaining_time": "1:21:07"}
+{"current_steps": 368, "total_steps": 532, "loss": 0.7814352512359619, "lr": 1.9983557575635553e-05, "epoch": 2.7669172932330826, "percentage": 69.17, "elapsed_time": "2:59:44", "remaining_time": "1:20:06"}
+{"current_steps": 370, "total_steps": 532, "loss": 0.8266798853874207, "lr": 1.9957928567815327e-05, "epoch": 2.781954887218045, "percentage": 69.55, "elapsed_time": "3:00:44", "remaining_time": "1:19:08"}
+{"current_steps": 372, "total_steps": 532, "loss": 0.9625144004821777, "lr": 1.9920517178511025e-05, "epoch": 2.7969924812030076, "percentage": 69.92, "elapsed_time": "3:01:53", "remaining_time": "1:18:14"}
+{"current_steps": 374, "total_steps": 532, "loss": 0.7290459275245667, "lr": 1.9871375523942014e-05, "epoch": 2.8120300751879697, "percentage": 70.3, "elapsed_time": "3:02:40", "remaining_time": "1:17:10"}
+{"current_steps": 376, "total_steps": 532, "loss": 1.0641855001449585, "lr": 1.981057206126258e-05, "epoch": 2.827067669172932, "percentage": 70.68, "elapsed_time": "3:03:49", "remaining_time": "1:16:16"}
+{"current_steps": 378, "total_steps": 532, "loss": 1.1699275970458984, "lr": 1.9738191493197164e-05, "epoch": 2.8421052631578947, "percentage": 71.05, "elapsed_time": "3:05:14", "remaining_time": "1:15:28"}
+{"current_steps": 380, "total_steps": 532, "loss": 0.8162869811058044, "lr": 1.96543346500446e-05, "epoch": 2.857142857142857, "percentage": 71.43, "elapsed_time": "3:06:03", "remaining_time": "1:14:25"}
+{"current_steps": 382, "total_steps": 532, "loss": 0.9681864380836487, "lr": 1.9559118349215685e-05, "epoch": 2.8721804511278197, "percentage": 71.8, "elapsed_time": "3:07:11", "remaining_time": "1:13:30"}
+{"current_steps": 384, "total_steps": 532, "loss": 0.9782270789146423, "lr": 1.9452675232499833e-05, "epoch": 2.887218045112782, "percentage": 72.18, "elapsed_time": "3:08:04", "remaining_time": "1:12:29"}
+{"current_steps": 386, "total_steps": 532, "loss": 0.7416080832481384, "lr": 1.9335153581287384e-05, "epoch": 2.9022556390977443, "percentage": 72.56, "elapsed_time": "3:08:52", "remaining_time": "1:11:26"}
+{"current_steps": 388, "total_steps": 532, "loss": 0.9780222773551941, "lr": 1.920671711000512e-05, "epoch": 2.917293233082707, "percentage": 72.93, "elapsed_time": "3:09:53", "remaining_time": "1:10:28"}
+{"current_steps": 390, "total_steps": 532, "loss": 0.8454363942146301, "lr": 1.906754473805263e-05, "epoch": 2.932330827067669, "percentage": 73.31, "elapsed_time": "3:10:52", "remaining_time": "1:09:29"}
+{"current_steps": 392, "total_steps": 532, "loss": 0.974253237247467, "lr": 1.8917830340557287e-05, "epoch": 2.9473684210526314, "percentage": 73.68, "elapsed_time": "3:11:53", "remaining_time": "1:08:31"}
+{"current_steps": 394, "total_steps": 532, "loss": 0.767073392868042, "lr": 1.875778247829503e-05, "epoch": 2.962406015037594, "percentage": 74.06, "elapsed_time": "3:12:49", "remaining_time": "1:07:32"}
+{"current_steps": 396, "total_steps": 532, "loss": 1.0176156759262085, "lr": 1.8587624107153192e-05, "epoch": 2.9774436090225564, "percentage": 74.44, "elapsed_time": "3:14:01", "remaining_time": "1:06:38"}
+{"current_steps": 398, "total_steps": 532, "loss": 0.8344214558601379, "lr": 1.840759226754013e-05, "epoch": 2.992481203007519, "percentage": 74.81, "elapsed_time": "3:14:47", "remaining_time": "1:05:35"}
+{"current_steps": 400, "total_steps": 532, "loss": 0.9761697053909302, "lr": 1.8217937754174265e-05, "epoch": 3.007518796992481, "percentage": 75.19, "elapsed_time": "3:15:50", "remaining_time": "1:04:37"}
+{"current_steps": 402, "total_steps": 532, "loss": 0.7501745223999023, "lr": 1.8018924766712608e-05, "epoch": 3.0225563909774436, "percentage": 75.56, "elapsed_time": "3:16:37", "remaining_time": "1:03:35"}
+{"current_steps": 404, "total_steps": 532, "loss": 0.9392253160476685, "lr": 1.7810830541705425e-05, "epoch": 3.037593984962406, "percentage": 75.94, "elapsed_time": "3:17:34", "remaining_time": "1:02:35"}
+{"current_steps": 406, "total_steps": 532, "loss": 0.6723920702934265, "lr": 1.7593944966389767e-05, "epoch": 3.0526315789473686, "percentage": 76.32, "elapsed_time": "3:18:22", "remaining_time": "1:01:33"}
+{"current_steps": 408, "total_steps": 532, "loss": 0.5313879251480103, "lr": 1.7368570174859855e-05, "epoch": 3.0676691729323307, "percentage": 76.69, "elapsed_time": "3:19:07", "remaining_time": "1:00:30"}
+{"current_steps": 410, "total_steps": 532, "loss": 0.8502162098884583, "lr": 1.713502012717686e-05, "epoch": 3.082706766917293, "percentage": 77.07, "elapsed_time": "3:20:15", "remaining_time": "0:59:35"}
+{"current_steps": 412, "total_steps": 532, "loss": 0.6262786388397217, "lr": 1.6893620172004526e-05, "epoch": 3.0977443609022557, "percentage": 77.44, "elapsed_time": "3:20:49", "remaining_time": "0:58:29"}
+{"current_steps": 414, "total_steps": 532, "loss": 0.9044028520584106, "lr": 1.664470659337968e-05, "epoch": 3.112781954887218, "percentage": 77.82, "elapsed_time": "3:21:53", "remaining_time": "0:57:32"}
+{"current_steps": 416, "total_steps": 532, "loss": 0.4721546471118927, "lr": 1.638862614224926e-05, "epoch": 3.1278195488721803, "percentage": 78.2, "elapsed_time": "3:22:31", "remaining_time": "0:56:28"}
+{"current_steps": 418, "total_steps": 532, "loss": 0.7782120108604431, "lr": 1.6125735553426228e-05, "epoch": 3.142857142857143, "percentage": 78.57, "elapsed_time": "3:23:28", "remaining_time": "0:55:29"}
+{"current_steps": 420, "total_steps": 532, "loss": 0.8657242059707642, "lr": 1.5856401048637505e-05, "epoch": 3.1578947368421053, "percentage": 78.95, "elapsed_time": "3:24:33", "remaining_time": "0:54:32"}
+{"current_steps": 422, "total_steps": 532, "loss": 0.7349554896354675, "lr": 1.5580997826355945e-05, "epoch": 3.172932330827068, "percentage": 79.32, "elapsed_time": "3:25:39", "remaining_time": "0:53:36"}
+{"current_steps": 424, "total_steps": 532, "loss": 0.8061299324035645, "lr": 1.5299909539127293e-05, "epoch": 3.18796992481203, "percentage": 79.7, "elapsed_time": "3:26:36", "remaining_time": "0:52:37"}
+{"current_steps": 426, "total_steps": 532, "loss": 0.7771633267402649, "lr": 1.5013527759120128e-05, "epoch": 3.2030075187969924, "percentage": 80.08, "elapsed_time": "3:27:42", "remaining_time": "0:51:41"}
+{"current_steps": 428, "total_steps": 532, "loss": 0.7236590385437012, "lr": 1.4722251432643222e-05, "epoch": 3.218045112781955, "percentage": 80.45, "elapsed_time": "3:28:40", "remaining_time": "0:50:42"}
+{"current_steps": 430, "total_steps": 532, "loss": 0.5557109713554382, "lr": 1.4426486324390432e-05, "epoch": 3.2330827067669174, "percentage": 80.83, "elapsed_time": "3:29:24", "remaining_time": "0:49:40"}
+{"current_steps": 432, "total_steps": 532, "loss": 0.5296133160591125, "lr": 1.4126644452187057e-05, "epoch": 3.2481203007518795, "percentage": 81.2, "elapsed_time": "3:30:11", "remaining_time": "0:48:39"}
+{"current_steps": 434, "total_steps": 532, "loss": 0.5751173496246338, "lr": 1.382314351302531e-05, "epoch": 3.263157894736842, "percentage": 81.58, "elapsed_time": "3:30:54", "remaining_time": "0:47:37"}
+{"current_steps": 436, "total_steps": 532, "loss": 0.5346217751502991, "lr": 1.351640630118838e-05, "epoch": 3.2781954887218046, "percentage": 81.95, "elapsed_time": "3:31:34", "remaining_time": "0:46:35"}
+{"current_steps": 438, "total_steps": 532, "loss": 0.5648679137229919, "lr": 1.3206860119273606e-05, "epoch": 3.293233082706767, "percentage": 82.33, "elapsed_time": "3:32:16", "remaining_time": "0:45:33"}
+{"current_steps": 440, "total_steps": 532, "loss": 1.0638585090637207, "lr": 1.2894936182935407e-05, "epoch": 3.308270676691729, "percentage": 82.71, "elapsed_time": "3:33:32", "remaining_time": "0:44:38"}
+{"current_steps": 442, "total_steps": 532, "loss": 0.452185720205307, "lr": 1.2581069020176986e-05, "epoch": 3.3233082706766917, "percentage": 83.08, "elapsed_time": "3:34:19", "remaining_time": "0:43:38"}
+{"current_steps": 444, "total_steps": 532, "loss": 0.7373262047767639, "lr": 1.2265695866027849e-05, "epoch": 3.338345864661654, "percentage": 83.46, "elapsed_time": "3:35:17", "remaining_time": "0:42:40"}
+{"current_steps": 446, "total_steps": 532, "loss": 0.5868238806724548, "lr": 1.1949256053450243e-05, "epoch": 3.3533834586466167, "percentage": 83.83, "elapsed_time": "3:36:02", "remaining_time": "0:41:39"}
+{"current_steps": 448, "total_steps": 532, "loss": 0.8645040392875671, "lr": 1.1632190401322987e-05, "epoch": 3.3684210526315788, "percentage": 84.21, "elapsed_time": "3:37:07", "remaining_time": "0:40:42"}
+{"current_steps": 450, "total_steps": 532, "loss": 0.9512546062469482, "lr": 1.1314940600355459e-05, "epoch": 3.3834586466165413, "percentage": 84.59, "elapsed_time": "3:38:16", "remaining_time": "0:39:46"}
+{"current_steps": 452, "total_steps": 532, "loss": 0.9128929376602173, "lr": 1.099794859778695e-05, "epoch": 3.398496240601504, "percentage": 84.96, "elapsed_time": "3:39:26", "remaining_time": "0:38:50"}
+{"current_steps": 454, "total_steps": 532, "loss": 0.9643846750259399, "lr": 1.0681655981728753e-05, "epoch": 3.4135338345864663, "percentage": 85.34, "elapsed_time": "3:40:35", "remaining_time": "0:37:53"}
+{"current_steps": 456, "total_steps": 532, "loss": 0.8578605055809021, "lr": 1.0366503366006477e-05, "epoch": 3.4285714285714284, "percentage": 85.71, "elapsed_time": "3:41:37", "remaining_time": "0:36:56"}
+{"current_steps": 458, "total_steps": 532, "loss": 0.5702735781669617, "lr": 1.0052929776359614e-05, "epoch": 3.443609022556391, "percentage": 86.09, "elapsed_time": "3:42:27", "remaining_time": "0:35:56"}
+{"current_steps": 460, "total_steps": 532, "loss": 0.9215094447135925, "lr": 9.741372038853437e-06, "epoch": 3.4586466165413534, "percentage": 86.47, "elapsed_time": "3:43:23", "remaining_time": "0:34:57"}
+{"current_steps": 462, "total_steps": 532, "loss": 0.8250299096107483, "lr": 9.432264171355117e-06, "epoch": 3.473684210526316, "percentage": 86.84, "elapsed_time": "3:44:20", "remaining_time": "0:33:59"}
+{"current_steps": 464, "total_steps": 532, "loss": 0.8027570843696594, "lr": 9.126036778921812e-06, "epoch": 3.488721804511278, "percentage": 87.22, "elapsed_time": "3:45:28", "remaining_time": "0:33:02"}
+{"current_steps": 466, "total_steps": 532, "loss": 0.8936192989349365, "lr": 8.8231164539431e-06, "epoch": 3.5037593984962405, "percentage": 87.59, "elapsed_time": "3:46:31", "remaining_time": "0:32:05"}
+{"current_steps": 468, "total_steps": 532, "loss": 0.6892407536506653, "lr": 8.52392518187319e-06, "epoch": 3.518796992481203, "percentage": 87.97, "elapsed_time": "3:47:31", "remaining_time": "0:31:06"}
+{"current_steps": 470, "total_steps": 532, "loss": 0.7898232340812683, "lr": 8.228879753380818e-06, "epoch": 3.5338345864661656, "percentage": 88.35, "elapsed_time": "3:48:31", "remaining_time": "0:30:08"}
+{"current_steps": 472, "total_steps": 532, "loss": 0.7635357975959778, "lr": 7.938391183735972e-06, "epoch": 3.548872180451128, "percentage": 88.72, "elapsed_time": "3:49:34", "remaining_time": "0:29:11"}
+{"current_steps": 474, "total_steps": 532, "loss": 0.8505274653434753, "lr": 7.65286414024186e-06, "epoch": 3.56390977443609, "percentage": 89.1, "elapsed_time": "3:50:41", "remaining_time": "0:28:13"}
+{"current_steps": 476, "total_steps": 532, "loss": 0.38889795541763306, "lr": 7.3726963785101094e-06, "epoch": 3.5789473684210527, "percentage": 89.47, "elapsed_time": "3:51:15", "remaining_time": "0:27:12"}
+{"current_steps": 478, "total_steps": 532, "loss": 0.365261048078537, "lr": 7.098278188364236e-06, "epoch": 3.593984962406015, "percentage": 89.85, "elapsed_time": "3:51:49", "remaining_time": "0:26:11"}
+{"current_steps": 480, "total_steps": 532, "loss": 0.9008171558380127, "lr": 6.829991850143528e-06, "epoch": 3.6090225563909772, "percentage": 90.23, "elapsed_time": "3:52:54", "remaining_time": "0:25:13"}
+{"current_steps": 482, "total_steps": 532, "loss": 0.900292158126831, "lr": 6.5682111021644835e-06, "epoch": 3.6240601503759398, "percentage": 90.6, "elapsed_time": "3:54:08", "remaining_time": "0:24:17"}
+{"current_steps": 484, "total_steps": 532, "loss": 0.7757176160812378, "lr": 6.313300620081777e-06, "epoch": 3.6390977443609023, "percentage": 90.98, "elapsed_time": "3:55:08", "remaining_time": "0:23:19"}
+{"current_steps": 486, "total_steps": 532, "loss": 0.3761276602745056, "lr": 6.065615508874151e-06, "epoch": 3.654135338345865, "percentage": 91.35, "elapsed_time": "3:55:39", "remaining_time": "0:22:18"}
+{"current_steps": 488, "total_steps": 532, "loss": 0.7676253318786621, "lr": 5.82550080816269e-06, "epoch": 3.6691729323308273, "percentage": 91.73, "elapsed_time": "3:56:35", "remaining_time": "0:21:19"}
+{"current_steps": 490, "total_steps": 532, "loss": 0.9202339053153992, "lr": 5.593291011550776e-06, "epoch": 3.6842105263157894, "percentage": 92.11, "elapsed_time": "3:57:50", "remaining_time": "0:20:23"}
+{"current_steps": 492, "total_steps": 532, "loss": 0.970383882522583, "lr": 5.3693096006552025e-06, "epoch": 3.699248120300752, "percentage": 92.48, "elapsed_time": "3:58:50", "remaining_time": "0:19:25"}
+{"current_steps": 494, "total_steps": 532, "loss": 0.48905929923057556, "lr": 5.15386859447772e-06, "epoch": 3.7142857142857144, "percentage": 92.86, "elapsed_time": "3:59:32", "remaining_time": "0:18:25"}
+{"current_steps": 496, "total_steps": 532, "loss": 0.9869436025619507, "lr": 4.947268114744566e-06, "epoch": 3.7293233082706765, "percentage": 93.23, "elapsed_time": "4:00:42", "remaining_time": "0:17:28"}
+{"current_steps": 498, "total_steps": 532, "loss": 1.0997637510299683, "lr": 4.749795967819596e-06, "epoch": 3.744360902255639, "percentage": 93.61, "elapsed_time": "4:02:04", "remaining_time": "0:16:31"}
+{"current_steps": 500, "total_steps": 532, "loss": 0.9353739023208618, "lr": 4.561727243773504e-06, "epoch": 3.7593984962406015, "percentage": 93.98, "elapsed_time": "4:03:15", "remaining_time": "0:15:34"}
+{"current_steps": 502, "total_steps": 532, "loss": 0.7048070430755615, "lr": 4.383323933167479e-06, "epoch": 3.774436090225564, "percentage": 94.36, "elapsed_time": "4:04:07", "remaining_time": "0:14:35"}
+{"current_steps": 504, "total_steps": 532, "loss": 0.6421667337417603, "lr": 4.214834562085255e-06, "epoch": 3.7894736842105265, "percentage": 94.74, "elapsed_time": "4:05:03", "remaining_time": "0:13:36"}
+{"current_steps": 506, "total_steps": 532, "loss": 0.5478267669677734, "lr": 4.056493845922015e-06, "epoch": 3.8045112781954886, "percentage": 95.11, "elapsed_time": "4:05:54", "remaining_time": "0:12:38"}
+{"current_steps": 508, "total_steps": 532, "loss": 0.8992588520050049, "lr": 3.908522362412276e-06, "epoch": 3.819548872180451, "percentage": 95.49, "elapsed_time": "4:07:05", "remaining_time": "0:11:40"}
+{"current_steps": 510, "total_steps": 532, "loss": 0.591274082660675, "lr": 3.7711262443524467e-06, "epoch": 3.8345864661654137, "percentage": 95.86, "elapsed_time": "4:07:59", "remaining_time": "0:10:41"}
+{"current_steps": 512, "total_steps": 532, "loss": 0.771586000919342, "lr": 3.6444968924459545e-06, "epoch": 3.8496240601503757, "percentage": 96.24, "elapsed_time": "4:09:12", "remaining_time": "0:09:44"}
+{"current_steps": 514, "total_steps": 532, "loss": 0.7341738939285278, "lr": 3.5288107086710545e-06, "epoch": 3.8646616541353382, "percentage": 96.62, "elapsed_time": "4:10:08", "remaining_time": "0:08:45"}
+{"current_steps": 516, "total_steps": 532, "loss": 1.0383778810501099, "lr": 3.4242288505426928e-06, "epoch": 3.8796992481203008, "percentage": 96.99, "elapsed_time": "4:11:27", "remaining_time": "0:07:47"}
+{"current_steps": 518, "total_steps": 532, "loss": 0.5621640682220459, "lr": 3.3308970066108413e-06, "epoch": 3.8947368421052633, "percentage": 97.37, "elapsed_time": "4:12:13", "remaining_time": "0:06:49"}
+{"current_steps": 520, "total_steps": 532, "loss": 0.9413120150566101, "lr": 3.2489451935079473e-06, "epoch": 3.909774436090226, "percentage": 97.74, "elapsed_time": "4:13:20", "remaining_time": "0:05:50"}
+{"current_steps": 522, "total_steps": 532, "loss": 1.0549259185791016, "lr": 3.1784875748282608e-06, "epoch": 3.924812030075188, "percentage": 98.12, "elapsed_time": "4:15:39", "remaining_time": "0:04:53"}
+{"current_steps": 524, "total_steps": 532, "loss": 1.0559005737304688, "lr": 3.119622302091404e-06, "epoch": 3.9398496240601504, "percentage": 98.5, "elapsed_time": "4:16:50", "remaining_time": "0:03:55"}
+{"current_steps": 526, "total_steps": 532, "loss": 0.9401967525482178, "lr": 3.072431378011632e-06, "epoch": 3.954887218045113, "percentage": 98.87, "elapsed_time": "4:18:08", "remaining_time": "0:02:56"}
+{"current_steps": 528, "total_steps": 532, "loss": 0.6218282580375671, "lr": 3.0369805422633432e-06, "epoch": 3.969924812030075, "percentage": 99.25, "elapsed_time": "4:19:07", "remaining_time": "0:01:57"}
+{"current_steps": 530, "total_steps": 532, "loss": 0.5278748869895935, "lr": 3.0133191799019103e-06, "epoch": 3.9849624060150375, "percentage": 99.62, "elapsed_time": "4:19:54", "remaining_time": "0:00:58"}
+{"current_steps": 532, "total_steps": 532, "loss": 0.6851533651351929, "lr": 3.001480252567475e-06, "epoch": 4.0, "percentage": 100.0, "elapsed_time": "4:20:53", "remaining_time": "0:00:00"}
+{"current_steps": 532, "total_steps": 532, "epoch": 4.0, "percentage": 100.0, "elapsed_time": "4:21:55", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,1905 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 532,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.015037593984962405,
+      "grad_norm": 0.4758685231208801,
+      "learning_rate": 7.407407407407407e-07,
+      "loss": 1.9719613790512085,
+      "step": 2
+    },
+    {
+      "epoch": 0.03007518796992481,
+      "grad_norm": 1.2873609066009521,
+      "learning_rate": 2.222222222222222e-06,
+      "loss": 2.2840397357940674,
+      "step": 4
+    },
+    {
+      "epoch": 0.045112781954887216,
+      "grad_norm": 0.4226410984992981,
+      "learning_rate": 3.7037037037037037e-06,
+      "loss": 2.0337564945220947,
+      "step": 6
+    },
+    {
+      "epoch": 0.06015037593984962,
+      "grad_norm": 0.20389820635318756,
+      "learning_rate": 5.185185185185185e-06,
+      "loss": 1.938320517539978,
+      "step": 8
+    },
+    {
+      "epoch": 0.07518796992481203,
+      "grad_norm": 1.1121097803115845,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 1.9521509408950806,
+      "step": 10
+    },
+    {
+      "epoch": 0.09022556390977443,
+      "grad_norm": 1.0452852249145508,
+      "learning_rate": 8.148148148148148e-06,
+      "loss": 2.3038036823272705,
+      "step": 12
+    },
+    {
+      "epoch": 0.10526315789473684,
+      "grad_norm": 0.17127464711666107,
+      "learning_rate": 9.62962962962963e-06,
+      "loss": 1.9031141996383667,
+      "step": 14
+    },
+    {
+      "epoch": 0.12030075187969924,
+      "grad_norm": 1.0378482341766357,
+      "learning_rate": 1.1111111111111113e-05,
+      "loss": 2.545740842819214,
+      "step": 16
+    },
+    {
+      "epoch": 0.13533834586466165,
+      "grad_norm": 1.3770403861999512,
+      "learning_rate": 1.2592592592592593e-05,
+      "loss": 3.2497336864471436,
+      "step": 18
+    },
+    {
+      "epoch": 0.15037593984962405,
+      "grad_norm": 0.6125630140304565,
+      "learning_rate": 1.4074074074074075e-05,
+      "loss": 1.7776570320129395,
+      "step": 20
+    },
+    {
+      "epoch": 0.16541353383458646,
+      "grad_norm": 0.6343645453453064,
+      "learning_rate": 1.555555555555556e-05,
+      "loss": 1.9716706275939941,
+      "step": 22
+    },
+    {
+      "epoch": 0.18045112781954886,
+      "grad_norm": 0.6397086977958679,
+      "learning_rate": 1.7037037037037038e-05,
+      "loss": 1.9381436109542847,
+      "step": 24
+    },
+    {
+      "epoch": 0.19548872180451127,
+      "grad_norm": 1.0100289583206177,
+      "learning_rate": 1.851851851851852e-05,
+      "loss": 1.9333921670913696,
+      "step": 26
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 0.6244422793388367,
+      "learning_rate": 2e-05,
+      "loss": 1.7134058475494385,
+      "step": 28
+    },
+    {
+      "epoch": 0.22556390977443608,
+      "grad_norm": 1.4532567262649536,
+      "learning_rate": 1.9994079505294254e-05,
+      "loss": 2.0251219272613525,
+      "step": 30
+    },
+    {
+      "epoch": 0.24060150375939848,
+      "grad_norm": 0.5191428661346436,
+      "learning_rate": 1.9976326268767035e-05,
+      "loss": 1.578896164894104,
+      "step": 32
+    },
+    {
+      "epoch": 0.2556390977443609,
+      "grad_norm": 0.2822546064853668,
+      "learning_rate": 1.994676502169901e-05,
+      "loss": 1.5274699926376343,
+      "step": 34
+    },
+    {
+      "epoch": 0.2706766917293233,
+      "grad_norm": 0.26378244161605835,
+      "learning_rate": 1.9905436944609424e-05,
+      "loss": 1.510546326637268,
+      "step": 36
+    },
+    {
+      "epoch": 0.2857142857142857,
+      "grad_norm": 2.3758437633514404,
+      "learning_rate": 1.9852399609889242e-05,
+      "loss": 2.110567331314087,
+      "step": 38
+    },
+    {
+      "epoch": 0.3007518796992481,
+      "grad_norm": 0.419758677482605,
+      "learning_rate": 1.9787726901599502e-05,
+      "loss": 1.1862285137176514,
+      "step": 40
+    },
+    {
+      "epoch": 0.3157894736842105,
+      "grad_norm": 0.24496647715568542,
+      "learning_rate": 1.9711508912546566e-05,
+      "loss": 1.440342903137207,
+      "step": 42
+    },
+    {
+      "epoch": 0.3308270676691729,
+      "grad_norm": 0.3002743721008301,
+      "learning_rate": 1.9623851818777652e-05,
+      "loss": 1.0962785482406616,
+      "step": 44
+    },
+    {
+      "epoch": 0.3458646616541353,
+      "grad_norm": 0.11880263686180115,
+      "learning_rate": 1.9524877731671482e-05,
+      "loss": 1.493391513824463,
+      "step": 46
+    },
+    {
+      "epoch": 0.3609022556390977,
+      "grad_norm": 1.7871476411819458,
+      "learning_rate": 1.941472452783011e-05,
+      "loss": 1.2433573007583618,
+      "step": 48
+    },
+    {
+      "epoch": 0.37593984962406013,
+      "grad_norm": 0.2921432852745056,
+      "learning_rate": 1.9293545657008865e-05,
+      "loss": 1.1390293836593628,
+      "step": 50
+    },
+    {
+      "epoch": 0.39097744360902253,
+      "grad_norm": 0.12673601508140564,
+      "learning_rate": 1.9161509928352017e-05,
+      "loss": 1.0903499126434326,
+      "step": 52
+    },
+    {
+      "epoch": 0.40601503759398494,
+      "grad_norm": 0.1648157387971878,
+      "learning_rate": 1.901880127523192e-05,
+      "loss": 1.1544872522354126,
+      "step": 54
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.08962542563676834,
+      "learning_rate": 1.886561849901922e-05,
+      "loss": 1.0822100639343262,
+      "step": 56
+    },
+    {
+      "epoch": 0.43609022556390975,
+      "grad_norm": 0.724937379360199,
+      "learning_rate": 1.870217499214111e-05,
+      "loss": 1.0416043996810913,
+      "step": 58
+    },
+    {
+      "epoch": 0.45112781954887216,
+      "grad_norm": 0.19841401278972626,
+      "learning_rate": 1.8528698440813397e-05,
+      "loss": 1.1815505027770996,
+      "step": 60
+    },
+    {
+      "epoch": 0.46616541353383456,
+      "grad_norm": 0.2616511285305023,
+      "learning_rate": 1.8345430507860478e-05,
+      "loss": 0.9047210812568665,
+      "step": 62
+    },
+    {
+      "epoch": 0.48120300751879697,
+      "grad_norm": 0.22353343665599823,
+      "learning_rate": 1.8152626496065128e-05,
+      "loss": 1.201892614364624,
+      "step": 64
+    },
+    {
+      "epoch": 0.49624060150375937,
+      "grad_norm": 0.6634237766265869,
+      "learning_rate": 1.7950554992517014e-05,
+      "loss": 1.1795772314071655,
+      "step": 66
+    },
+    {
+      "epoch": 0.5112781954887218,
+      "grad_norm": 0.13962750136852264,
+      "learning_rate": 1.7739497494455412e-05,
+      "loss": 1.3534270524978638,
+      "step": 68
+    },
+    {
+      "epoch": 0.5263157894736842,
+      "grad_norm": 0.665572464466095,
+      "learning_rate": 1.7519748017127354e-05,
+      "loss": 1.125345230102539,
+      "step": 70
+    },
+    {
+      "epoch": 0.5413533834586466,
+      "grad_norm": 0.179820716381073,
+      "learning_rate": 1.729161268420746e-05,
+      "loss": 0.8090606331825256,
+      "step": 72
+    },
+    {
+      "epoch": 0.556390977443609,
+      "grad_norm": 0.11726067215204239,
+      "learning_rate": 1.7055409301350013e-05,
+      "loss": 0.9213717579841614,
+      "step": 74
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.1627010852098465,
+      "learning_rate": 1.681146691346742e-05,
+      "loss": 1.0821505784988403,
+      "step": 76
+    },
+    {
+      "epoch": 0.5864661654135338,
+      "grad_norm": 0.1170893982052803,
+      "learning_rate": 1.6560125346351663e-05,
+      "loss": 1.300316333770752,
+      "step": 78
+    },
+    {
+      "epoch": 0.6015037593984962,
+      "grad_norm": 0.5334263443946838,
+      "learning_rate": 1.6301734733277442e-05,
+      "loss": 0.6484270691871643,
+      "step": 80
+    },
+    {
+      "epoch": 0.6165413533834586,
+      "grad_norm": 0.101778045296669,
+      "learning_rate": 1.603665502724633e-05,
+      "loss": 1.2577356100082397,
+      "step": 82
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.1670590043067932,
+      "learning_rate": 1.576525549955156e-05,
+      "loss": 1.3543009757995605,
+      "step": 84
+    },
+    {
+      "epoch": 0.6466165413533834,
+      "grad_norm": 0.20087628066539764,
+      "learning_rate": 1.548791422536178e-05,
+      "loss": 0.9327285885810852,
+      "step": 86
+    },
+    {
+      "epoch": 0.6616541353383458,
+      "grad_norm": 0.23011328279972076,
+      "learning_rate": 1.5205017557040656e-05,
+      "loss": 1.1237722635269165,
+      "step": 88
+    },
+    {
+      "epoch": 0.6766917293233082,
+      "grad_norm": 0.159201517701149,
+      "learning_rate": 1.4916959585935732e-05,
+      "loss": 1.1964070796966553,
+      "step": 90
+    },
+    {
+      "epoch": 0.6917293233082706,
+      "grad_norm": 0.44444212317466736,
+      "learning_rate": 1.4624141593386507e-05,
+      "loss": 1.0308165550231934,
+      "step": 92
+    },
+    {
+      "epoch": 0.706766917293233,
+      "grad_norm": 0.22338902950286865,
+      "learning_rate": 1.4326971491716427e-05,
+      "loss": 0.9982426762580872,
+      "step": 94
+    },
+    {
+      "epoch": 0.7218045112781954,
+      "grad_norm": 0.2096806913614273,
+      "learning_rate": 1.402586325598752e-05,
+      "loss": 1.3940727710723877,
+      "step": 96
+    },
+    {
+      "epoch": 0.7368421052631579,
+      "grad_norm": 0.21466873586177826,
+      "learning_rate": 1.3721236347309314e-05,
+      "loss": 1.1801196336746216,
+      "step": 98
+    },
+    {
+      "epoch": 0.7518796992481203,
+      "grad_norm": 0.30590999126434326,
+      "learning_rate": 1.3413515128505363e-05,
+      "loss": 0.6430416703224182,
+      "step": 100
+    },
+    {
+      "epoch": 0.7669172932330827,
+      "grad_norm": 0.13401509821414948,
+      "learning_rate": 1.3103128272951363e-05,
+      "loss": 1.3783133029937744,
+      "step": 102
+    },
+    {
+      "epoch": 0.7819548872180451,
+      "grad_norm": 0.12506185472011566,
+      "learning_rate": 1.2790508167408509e-05,
+      "loss": 0.9889219403266907,
+      "step": 104
+    },
+    {
+      "epoch": 0.7969924812030075,
+      "grad_norm": 0.22044184803962708,
+      "learning_rate": 1.2476090309683804e-05,
+      "loss": 0.6871194243431091,
+      "step": 106
+    },
+    {
+      "epoch": 0.8120300751879699,
+      "grad_norm": 0.18642327189445496,
+      "learning_rate": 1.2160312701956553e-05,
+      "loss": 1.0068978071212769,
+      "step": 108
+    },
+    {
+      "epoch": 0.8270676691729323,
+      "grad_norm": 0.5397107601165771,
+      "learning_rate": 1.1843615240616111e-05,
+      "loss": 0.8988245725631714,
+      "step": 110
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.23860491812229156,
+      "learning_rate": 1.1526439103460874e-05,
+      "loss": 0.7688661813735962,
+      "step": 112
+    },
+    {
+      "epoch": 0.8571428571428571,
+      "grad_norm": 0.1502920240163803,
+      "learning_rate": 1.120922613511221e-05,
+      "loss": 1.0024021863937378,
+      "step": 114
+    },
+    {
+      "epoch": 0.8721804511278195,
+      "grad_norm": 0.16608496010303497,
+      "learning_rate": 1.0892418231499461e-05,
+      "loss": 1.4191375970840454,
+      "step": 116
+    },
+    {
+      "epoch": 0.8872180451127819,
+      "grad_norm": 0.17546993494033813,
+      "learning_rate": 1.057645672427347e-05,
+      "loss": 1.073761224746704,
+      "step": 118
+    },
+    {
+      "epoch": 0.9022556390977443,
+      "grad_norm": 5.0469841957092285,
+      "learning_rate": 1.0261781766006174e-05,
+      "loss": 0.9056495428085327,
+      "step": 120
+    },
+    {
+      "epoch": 0.9172932330827067,
+      "grad_norm": 0.36213457584381104,
+      "learning_rate": 9.948831717032738e-06,
+      "loss": 0.8672894835472107,
+      "step": 122
+    },
+    {
+      "epoch": 0.9323308270676691,
+      "grad_norm": 0.4218904972076416,
+      "learning_rate": 9.638042534790373e-06,
+      "loss": 0.827739417552948,
+      "step": 124
+    },
+    {
+      "epoch": 0.9473684210526315,
+      "grad_norm": 0.0983240008354187,
+      "learning_rate": 9.329847166504497e-06,
+      "loss": 0.7799423933029175,
+      "step": 126
+    },
+    {
+      "epoch": 0.9624060150375939,
+      "grad_norm": 0.14850644767284393,
+      "learning_rate": 9.024674946068357e-06,
+      "loss": 1.0653791427612305,
+      "step": 128
+    },
+    {
+      "epoch": 0.9774436090225563,
+      "grad_norm": 0.27504855394363403,
+      "learning_rate": 8.722950995956172e-06,
+      "loss": 0.9135006666183472,
+      "step": 130
+    },
+    {
+      "epoch": 0.9924812030075187,
+      "grad_norm": 0.27222368121147156,
+      "learning_rate": 8.425095635003053e-06,
+      "loss": 1.0402815341949463,
+      "step": 132
+    },
+    {
+      "epoch": 1.0075187969924813,
+      "grad_norm": 0.1408149003982544,
+      "learning_rate": 8.13152379287667e-06,
+      "loss": 0.7629735469818115,
+      "step": 134
+    },
+    {
+      "epoch": 1.0225563909774436,
+      "grad_norm": 0.15156933665275574,
+      "learning_rate": 7.842644432056336e-06,
+      "loss": 0.9513287544250488,
+      "step": 136
+    },
+    {
+      "epoch": 1.037593984962406,
+      "grad_norm": 0.9048015475273132,
+      "learning_rate": 7.55885997812472e-06,
+      "loss": 0.6946667432785034,
+      "step": 138
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 0.24624674022197723,
+      "learning_rate": 7.280565759165833e-06,
+      "loss": 0.7876754403114319,
+      "step": 140
+    },
+    {
+      "epoch": 1.0676691729323309,
+      "grad_norm": 0.09362401068210602,
+      "learning_rate": 7.008149455050264e-06,
+      "loss": 0.8406305909156799,
+      "step": 142
+    },
+    {
+      "epoch": 1.0827067669172932,
+      "grad_norm": 0.17641493678092957,
+      "learning_rate": 6.741990557374784e-06,
+      "loss": 1.0424906015396118,
+      "step": 144
+    },
+    {
+      "epoch": 1.0977443609022557,
+      "grad_norm": 0.12022742629051208,
+      "learning_rate": 6.4824598408087015e-06,
+      "loss": 1.0722497701644897,
+      "step": 146
+    },
+    {
+      "epoch": 1.112781954887218,
+      "grad_norm": 0.08780567348003387,
+      "learning_rate": 6.229918846583414e-06,
+      "loss": 1.0312786102294922,
+      "step": 148
+    },
+    {
+      "epoch": 1.1278195488721805,
+      "grad_norm": 0.3304075598716736,
+      "learning_rate": 5.984719378844628e-06,
+      "loss": 1.0075746774673462,
+      "step": 150
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 0.10900267213582993,
+      "learning_rate": 5.7472030145689604e-06,
+      "loss": 1.0427347421646118,
+      "step": 152
+    },
+    {
+      "epoch": 1.1578947368421053,
+      "grad_norm": 0.09693319350481033,
+      "learning_rate": 5.51770062772752e-06,
+      "loss": 1.0927042961120605,
+      "step": 154
+    },
+    {
+      "epoch": 1.1729323308270676,
+      "grad_norm": 0.12883000075817108,
+      "learning_rate": 5.296531928359431e-06,
+      "loss": 0.9705473780632019,
+      "step": 156
+    },
+    {
+      "epoch": 1.1879699248120301,
+      "grad_norm": 0.11416517943143845,
+      "learning_rate": 5.084005017197318e-06,
+      "loss": 1.0172467231750488,
+      "step": 158
+    },
+    {
+      "epoch": 1.2030075187969924,
+      "grad_norm": 0.10797861963510513,
+      "learning_rate": 4.8804159564652665e-06,
+      "loss": 0.5409541726112366,
+      "step": 160
+    },
+    {
+      "epoch": 1.218045112781955,
+      "grad_norm": 0.24861538410186768,
+      "learning_rate": 4.686048357447095e-06,
+      "loss": 0.9430153965950012,
+      "step": 162
+    },
+    {
+      "epoch": 1.2330827067669172,
+      "grad_norm": 0.14524255692958832,
+      "learning_rate": 4.501172985399498e-06,
+      "loss": 1.178081750869751,
+      "step": 164
+    },
+    {
+      "epoch": 1.2481203007518797,
+      "grad_norm": 0.2940176725387573,
+      "learning_rate": 4.326047382360457e-06,
+      "loss": 0.8844167590141296,
+      "step": 166
+    },
+    {
+      "epoch": 1.263157894736842,
+      "grad_norm": 0.1306038349866867,
+      "learning_rate": 4.160915508378359e-06,
+      "loss": 0.7063813209533691,
+      "step": 168
+    },
+    {
+      "epoch": 1.2781954887218046,
+      "grad_norm": 0.15677547454833984,
+      "learning_rate": 4.006007401661596e-06,
+      "loss": 0.7762787938117981,
+      "step": 170
+    },
+    {
+      "epoch": 1.2932330827067668,
+      "grad_norm": 0.14480236172676086,
+      "learning_rate": 3.861538858122092e-06,
+      "loss": 0.937335193157196,
+      "step": 172
+    },
+    {
+      "epoch": 1.3082706766917294,
+      "grad_norm": 0.15142542123794556,
+      "learning_rate": 3.727711130759182e-06,
+      "loss": 0.9747655987739563,
+      "step": 174
+    },
+    {
+      "epoch": 1.3233082706766917,
+      "grad_norm": 0.18726477026939392,
+      "learning_rate": 3.6047106493025923e-06,
+      "loss": 0.7746855020523071,
+      "step": 176
+    },
+    {
+      "epoch": 1.3383458646616542,
+      "grad_norm": 2.189209461212158,
+      "learning_rate": 3.492708760505093e-06,
+      "loss": 0.6926825642585754,
+      "step": 178
+    },
+    {
+      "epoch": 1.3533834586466165,
+      "grad_norm": 0.12323262542486191,
+      "learning_rate": 3.3918614894466045e-06,
+      "loss": 1.2502151727676392,
+      "step": 180
+    },
+    {
+      "epoch": 1.368421052631579,
+      "grad_norm": 0.1458672434091568,
+      "learning_rate": 3.3023093221822746e-06,
+      "loss": 0.9960780143737793,
+      "step": 182
+    },
+    {
+      "epoch": 1.3834586466165413,
+      "grad_norm": 0.1072113886475563,
+      "learning_rate": 3.224177010037323e-06,
+      "loss": 0.8326720595359802,
+      "step": 184
+    },
+    {
+      "epoch": 1.3984962406015038,
+      "grad_norm": 0.09054487943649292,
+      "learning_rate": 3.1575733958212563e-06,
+      "loss": 1.1920455694198608,
+      "step": 186
+    },
+    {
+      "epoch": 1.413533834586466,
+      "grad_norm": 0.13850678503513336,
+      "learning_rate": 3.1025912622035687e-06,
+      "loss": 0.5979896783828735,
+      "step": 188
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.2317809760570526,
+      "learning_rate": 3.0593072024621396e-06,
+      "loss": 0.969947099685669,
+      "step": 190
+    },
+    {
+      "epoch": 1.443609022556391,
+      "grad_norm": 0.29049012064933777,
+      "learning_rate": 3.0277815137843917e-06,
+      "loss": 0.7010709643363953,
+      "step": 192
+    },
+    {
+      "epoch": 1.4586466165413534,
+      "grad_norm": 0.17280922830104828,
+      "learning_rate": 3.008058113269836e-06,
+      "loss": 0.7660905718803406,
+      "step": 194
+    },
+    {
+      "epoch": 1.4736842105263157,
+      "grad_norm": 0.12446308135986328,
+      "learning_rate": 3.0001644767510154e-06,
+      "loss": 0.9958880543708801,
+      "step": 196
+    },
+    {
+      "epoch": 1.4887218045112782,
+      "grad_norm": 0.1736506223678589,
+      "learning_rate": 3.0041116005181016e-06,
+      "loss": 0.9368724226951599,
+      "step": 198
+    },
+    {
+      "epoch": 1.5037593984962405,
+      "grad_norm": 0.15870751440525055,
+      "learning_rate": 3.0198939860004202e-06,
+      "loss": 0.9826479554176331,
+      "step": 200
+    },
+    {
+      "epoch": 1.518796992481203,
+      "grad_norm": 0.15492001175880432,
+      "learning_rate": 3.0474896474262772e-06,
+      "loss": 1.3954254388809204,
+      "step": 202
+    },
+    {
+      "epoch": 1.5338345864661656,
+      "grad_norm": 0.1334741860628128,
+      "learning_rate": 3.08686014245041e-06,
+      "loss": 1.0462982654571533,
+      "step": 204
+    },
+    {
+      "epoch": 1.5488721804511278,
+      "grad_norm": 0.26873454451560974,
+      "learning_rate": 3.1379506257063825e-06,
+      "loss": 0.8277729153633118,
+      "step": 206
+    },
+    {
+      "epoch": 1.5639097744360901,
+      "grad_norm": 0.16168057918548584,
+      "learning_rate": 3.20068992520934e-06,
+      "loss": 0.985795259475708,
+      "step": 208
+    },
+    {
+      "epoch": 1.5789473684210527,
+      "grad_norm": 0.1431337147951126,
+      "learning_rate": 3.274990641502683e-06,
+      "loss": 0.908364474773407,
+      "step": 210
+    },
+    {
+      "epoch": 1.5939849624060152,
+      "grad_norm": 0.2742830514907837,
+      "learning_rate": 3.3607492694105405e-06,
+      "loss": 0.6370347142219543,
+      "step": 212
+    },
+    {
+      "epoch": 1.6090225563909775,
+      "grad_norm": 0.14003214240074158,
+      "learning_rate": 3.457846342226442e-06,
+      "loss": 0.9584491848945618,
+      "step": 214
+    },
+    {
+      "epoch": 1.6240601503759398,
+      "grad_norm": 0.29483214020729065,
+      "learning_rate": 3.5661465981373183e-06,
+      "loss": 1.0774601697921753,
+      "step": 216
+    },
+    {
+      "epoch": 1.6390977443609023,
+      "grad_norm": 0.18227490782737732,
+      "learning_rate": 3.6854991686509906e-06,
+      "loss": 0.7987947463989258,
+      "step": 218
+    },
+    {
+      "epoch": 1.6541353383458648,
+      "grad_norm": 0.20908379554748535,
+      "learning_rate": 3.815737788764674e-06,
+      "loss": 0.8292507529258728,
+      "step": 220
+    },
+    {
+      "epoch": 1.669172932330827,
+      "grad_norm": 0.23883360624313354,
+      "learning_rate": 3.956681028581693e-06,
+      "loss": 0.9586336612701416,
+      "step": 222
+    },
+    {
+      "epoch": 1.6842105263157894,
+      "grad_norm": 0.3358386158943176,
+      "learning_rate": 4.108132546053779e-06,
+      "loss": 0.8150299191474915,
+      "step": 224
+    },
+    {
+      "epoch": 1.699248120300752,
+      "grad_norm": 0.298909068107605,
+      "learning_rate": 4.269881360496842e-06,
+      "loss": 1.0946331024169922,
+      "step": 226
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 0.24479779601097107,
+      "learning_rate": 4.441702146499222e-06,
+      "loss": 0.9061790108680725,
+      "step": 228
+    },
+    {
+      "epoch": 1.7293233082706767,
+      "grad_norm": 0.2154252976179123,
+      "learning_rate": 4.623355547812946e-06,
+      "loss": 1.0011441707611084,
+      "step": 230
+    },
+    {
+      "epoch": 1.744360902255639,
+      "grad_norm": 0.1481187492609024,
+      "learning_rate": 4.814588510790782e-06,
+      "loss": 1.1533119678497314,
+      "step": 232
+    },
+    {
+      "epoch": 1.7593984962406015,
+      "grad_norm": 0.10200405865907669,
+      "learning_rate": 5.01513463690452e-06,
+      "loss": 1.1073795557022095,
+      "step": 234
+    },
+    {
+      "epoch": 1.774436090225564,
+      "grad_norm": 0.19221612811088562,
+      "learning_rate": 5.224714553853478e-06,
+      "loss": 1.148139476776123,
+      "step": 236
+    },
+    {
+      "epoch": 1.7894736842105263,
+      "grad_norm": 0.1456657350063324,
+      "learning_rate": 5.443036304746191e-06,
+      "loss": 0.6271846294403076,
+      "step": 238
+    },
+    {
+      "epoch": 1.8045112781954886,
+      "grad_norm": 0.3602541983127594,
+      "learning_rate": 5.66979575481317e-06,
+      "loss": 1.0237879753112793,
+      "step": 240
+    },
+    {
+      "epoch": 1.8195488721804511,
+      "grad_norm": 0.14290253818035126,
+      "learning_rate": 5.904677015084159e-06,
+      "loss": 0.7585715055465698,
+      "step": 242
+    },
+    {
+      "epoch": 1.8345864661654137,
+      "grad_norm": 0.15334200859069824,
+      "learning_rate": 6.147352882439652e-06,
+      "loss": 1.1932705640792847,
+      "step": 244
+    },
+    {
+      "epoch": 1.849624060150376,
+      "grad_norm": 0.09177304059267044,
+      "learning_rate": 6.397485295423669e-06,
+      "loss": 1.1568275690078735,
+      "step": 246
+    },
+    {
+      "epoch": 1.8646616541353382,
+      "grad_norm": 0.08284302055835724,
+      "learning_rate": 6.6547258051828426e-06,
+      "loss": 0.9782670736312866,
+      "step": 248
+    },
+    {
+      "epoch": 1.8796992481203008,
+      "grad_norm": 0.12089983373880386,
+      "learning_rate": 6.918716060875743e-06,
+      "loss": 1.2261128425598145,
+      "step": 250
+    },
+    {
+      "epoch": 1.8947368421052633,
+      "grad_norm": 0.1200575977563858,
+      "learning_rate": 7.1890883088761885e-06,
+      "loss": 1.0725328922271729,
+      "step": 252
+    },
+    {
+      "epoch": 1.9097744360902256,
+      "grad_norm": 0.2572805881500244,
+      "learning_rate": 7.4654659050752845e-06,
+      "loss": 1.2271314859390259,
+      "step": 254
+    },
+    {
+      "epoch": 1.9248120300751879,
+      "grad_norm": 0.2713569104671478,
+      "learning_rate": 7.747463839568292e-06,
+      "loss": 0.8813698291778564,
+      "step": 256
+    },
+    {
+      "epoch": 1.9398496240601504,
+      "grad_norm": 0.5769055485725403,
+      "learning_rate": 8.034689272995649e-06,
+      "loss": 0.5630529522895813,
+      "step": 258
+    },
+    {
+      "epoch": 1.954887218045113,
+      "grad_norm": 0.13255445659160614,
+      "learning_rate": 8.32674208379076e-06,
+      "loss": 1.0115076303482056,
+      "step": 260
+    },
+    {
+      "epoch": 1.9699248120300752,
+      "grad_norm": 0.15804584324359894,
+      "learning_rate": 8.623215425572433e-06,
+      "loss": 0.9677861332893372,
+      "step": 262
+    },
+    {
+      "epoch": 1.9849624060150375,
+      "grad_norm": 0.09761196374893188,
+      "learning_rate": 8.923696293905279e-06,
+      "loss": 0.9606061577796936,
+      "step": 264
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.24307098984718323,
+      "learning_rate": 9.227766101638668e-06,
+      "loss": 0.8462581038475037,
+      "step": 266
+    },
+    {
+      "epoch": 2.0150375939849625,
+      "grad_norm": 0.15874817967414856,
+      "learning_rate": 9.535001262022755e-06,
+      "loss": 0.5168899893760681,
+      "step": 268
+    },
+    {
+      "epoch": 2.030075187969925,
+      "grad_norm": 0.09884876012802124,
+      "learning_rate": 9.844973778789198e-06,
+      "loss": 0.725158154964447,
+      "step": 270
+    },
+    {
+      "epoch": 2.045112781954887,
+      "grad_norm": 0.11747175455093384,
+      "learning_rate": 1.0157251842374587e-05,
+      "loss": 0.5184513330459595,
+      "step": 272
+    },
+    {
+      "epoch": 2.0601503759398496,
+      "grad_norm": 0.17868469655513763,
+      "learning_rate": 1.0471400431456062e-05,
+      "loss": 1.0087833404541016,
+      "step": 274
+    },
+    {
+      "epoch": 2.075187969924812,
+      "grad_norm": 0.12505443394184113,
+      "learning_rate": 1.078698191896105e-05,
+      "loss": 0.9181981682777405,
+      "step": 276
+    },
+    {
+      "epoch": 2.090225563909774,
+      "grad_norm": 0.45129311084747314,
+      "learning_rate": 1.1103556681706949e-05,
+      "loss": 0.867568850517273,
+      "step": 278
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 0.38053497672080994,
+      "learning_rate": 1.1420683712821577e-05,
+      "loss": 0.8937767744064331,
+      "step": 280
+    },
+    {
+      "epoch": 2.1203007518796992,
+      "grad_norm": 0.2623603045940399,
+      "learning_rate": 1.1737921236091097e-05,
+      "loss": 1.0325863361358643,
+      "step": 282
+    },
+    {
+      "epoch": 2.1353383458646618,
+      "grad_norm": 0.13423830270767212,
+      "learning_rate": 1.2054827321379756e-05,
+      "loss": 0.6633681654930115,
+      "step": 284
+    },
+    {
+      "epoch": 2.1503759398496243,
+      "grad_norm": 0.20472559332847595,
+      "learning_rate": 1.237096050026399e-05,
+      "loss": 0.998541533946991,
+      "step": 286
+    },
+    {
+      "epoch": 2.1654135338345863,
+      "grad_norm": 0.28264734148979187,
+      "learning_rate": 1.2685880381023343e-05,
+      "loss": 1.1486636400222778,
+      "step": 288
+    },
+    {
+      "epoch": 2.180451127819549,
+      "grad_norm": 0.11129236966371536,
+      "learning_rate": 1.2999148262131527e-05,
+      "loss": 1.1767593622207642,
+      "step": 290
+    },
+    {
+      "epoch": 2.1954887218045114,
+      "grad_norm": 0.17632298171520233,
+      "learning_rate": 1.3310327743392861e-05,
+      "loss": 0.8024111986160278,
+      "step": 292
+    },
+    {
+      "epoch": 2.2105263157894735,
+      "grad_norm": 0.1285456418991089,
+      "learning_rate": 1.3618985333872927e-05,
+      "loss": 0.9202947616577148,
+      "step": 294
+    },
+    {
+      "epoch": 2.225563909774436,
+      "grad_norm": 0.2532704770565033,
+      "learning_rate": 1.3924691055776338e-05,
+      "loss": 0.3760508596897125,
+      "step": 296
+    },
+    {
+      "epoch": 2.2406015037593985,
+      "grad_norm": 0.18166831135749817,
+      "learning_rate": 1.4227019043430623e-05,
+      "loss": 1.2139897346496582,
+      "step": 298
+    },
+    {
+      "epoch": 2.255639097744361,
+      "grad_norm": 0.31214964389801025,
+      "learning_rate": 1.4525548136541599e-05,
+      "loss": 1.1467747688293457,
+      "step": 300
+    },
+    {
+      "epoch": 2.2706766917293235,
+      "grad_norm": 0.2208021879196167,
+      "learning_rate": 1.4819862466893891e-05,
+      "loss": 1.152048110961914,
+      "step": 302
+    },
+    {
+      "epoch": 2.2857142857142856,
+      "grad_norm": 0.28248584270477295,
+      "learning_rate": 1.5109552037679353e-05,
+      "loss": 0.8999449610710144,
+      "step": 304
+    },
+    {
+      "epoch": 2.300751879699248,
+      "grad_norm": 0.17761224508285522,
+      "learning_rate": 1.539421329464618e-05,
+      "loss": 0.8530235290527344,
+      "step": 306
+    },
+    {
+      "epoch": 2.3157894736842106,
+      "grad_norm": 0.12931574881076813,
+      "learning_rate": 1.5673449688273246e-05,
+      "loss": 1.1707788705825806,
+      "step": 308
+    },
+    {
+      "epoch": 2.3308270676691727,
+      "grad_norm": 0.45592963695526123,
+      "learning_rate": 1.5946872226186394e-05,
+      "loss": 0.543804407119751,
+      "step": 310
+    },
+    {
+      "epoch": 2.345864661654135,
+      "grad_norm": 0.13944678008556366,
+      "learning_rate": 1.621410001504725e-05,
+      "loss": 0.9259825944900513,
+      "step": 312
+    },
+    {
+      "epoch": 2.3609022556390977,
+      "grad_norm": 0.6154989004135132,
+      "learning_rate": 1.6474760791159528e-05,
+      "loss": 0.6412667036056519,
+      "step": 314
+    },
+    {
+      "epoch": 2.3759398496240602,
+      "grad_norm": 0.19801302254199982,
+      "learning_rate": 1.672849143905383e-05,
+      "loss": 1.0553992986679077,
+      "step": 316
+    },
+    {
+      "epoch": 2.3909774436090228,
+      "grad_norm": 0.16842691600322723,
+      "learning_rate": 1.697493849732838e-05,
+      "loss": 0.9918773174285889,
+      "step": 318
+    },
+    {
+      "epoch": 2.406015037593985,
+      "grad_norm": 0.17690639197826385,
+      "learning_rate": 1.7213758651041114e-05,
+      "loss": 1.183601975440979,
+      "step": 320
+    },
+    {
+      "epoch": 2.4210526315789473,
+      "grad_norm": 0.1753586232662201,
+      "learning_rate": 1.7444619209967163e-05,
+      "loss": 1.2674732208251953,
+      "step": 322
+    },
+    {
+      "epoch": 2.43609022556391,
+      "grad_norm": 0.1980062872171402,
+      "learning_rate": 1.766719857205551e-05,
+      "loss": 0.9618882536888123,
+      "step": 324
+    },
+    {
+      "epoch": 2.451127819548872,
+      "grad_norm": 0.14773985743522644,
+      "learning_rate": 1.7881186671439107e-05,
+      "loss": 1.1368170976638794,
+      "step": 326
+    },
+    {
+      "epoch": 2.4661654135338344,
+      "grad_norm": 0.3491007685661316,
+      "learning_rate": 1.8086285410374563e-05,
+      "loss": 0.9204502701759338,
+      "step": 328
+    },
+    {
+      "epoch": 2.481203007518797,
+      "grad_norm": 0.41794508695602417,
+      "learning_rate": 1.8282209074509393e-05,
+      "loss": 0.7325559258460999,
+      "step": 330
+    },
+    {
+      "epoch": 2.4962406015037595,
+      "grad_norm": 0.1503211408853531,
+      "learning_rate": 1.8468684730898602e-05,
+      "loss": 1.1228058338165283,
+      "step": 332
+    },
+    {
+      "epoch": 2.511278195488722,
+      "grad_norm": 0.29104840755462646,
+      "learning_rate": 1.864545260821598e-05,
+      "loss": 0.6551952958106995,
+      "step": 334
+    },
+    {
+      "epoch": 2.526315789473684,
+      "grad_norm": 0.17387054860591888,
+      "learning_rate": 1.8812266458630505e-05,
+      "loss": 0.7751361727714539,
+      "step": 336
+    },
+    {
+      "epoch": 2.5413533834586466,
+      "grad_norm": 0.7814794778823853,
+      "learning_rate": 1.8968893900843786e-05,
+      "loss": 0.7552028298377991,
+      "step": 338
+    },
+    {
+      "epoch": 2.556390977443609,
+      "grad_norm": 0.15030670166015625,
+      "learning_rate": 1.9115116743810537e-05,
+      "loss": 1.1345906257629395,
+      "step": 340
+    },
+    {
+      "epoch": 2.571428571428571,
+      "grad_norm": 0.2455630898475647,
+      "learning_rate": 1.9250731290691326e-05,
+      "loss": 1.0457109212875366,
+      "step": 342
+    },
+    {
+      "epoch": 2.5864661654135337,
+      "grad_norm": 0.17742854356765747,
+      "learning_rate": 1.9375548622613952e-05,
+      "loss": 0.8526303768157959,
+      "step": 344
+    },
+    {
+      "epoch": 2.601503759398496,
+      "grad_norm": 0.35299667716026306,
+      "learning_rate": 1.9489394861848333e-05,
+      "loss": 0.6485392451286316,
+      "step": 346
+    },
+    {
+      "epoch": 2.6165413533834587,
+      "grad_norm": 0.24292704463005066,
+      "learning_rate": 1.9592111414028205e-05,
+      "loss": 0.8217464685440063,
+      "step": 348
+    },
+    {
+      "epoch": 2.6315789473684212,
+      "grad_norm": 0.11688588559627533,
+      "learning_rate": 1.9683555189082178e-05,
+      "loss": 0.9268519282341003,
+      "step": 350
+    },
+    {
+      "epoch": 2.6466165413533833,
+      "grad_norm": 0.8352806568145752,
+      "learning_rate": 1.9763598800566478e-05,
+      "loss": 0.7672562599182129,
+      "step": 352
+    },
+    {
+      "epoch": 2.661654135338346,
+      "grad_norm": 0.10471988469362259,
+      "learning_rate": 1.983213074312156e-05,
+      "loss": 1.0508689880371094,
+      "step": 354
+    },
+    {
+      "epoch": 2.6766917293233083,
+      "grad_norm": 0.9734071493148804,
+      "learning_rate": 1.9889055547805527e-05,
+      "loss": 0.8092398047447205,
+      "step": 356
+    },
+    {
+      "epoch": 2.6917293233082704,
+      "grad_norm": 0.22569192945957184,
+      "learning_rate": 1.993429391508785e-05,
+      "loss": 0.4492420554161072,
+      "step": 358
+    },
+    {
+      "epoch": 2.706766917293233,
+      "grad_norm": 0.19527536630630493,
+      "learning_rate": 1.9967782825318183e-05,
+      "loss": 1.1782996654510498,
+      "step": 360
+    },
+    {
+      "epoch": 2.7218045112781954,
+      "grad_norm": 0.4661547839641571,
+      "learning_rate": 1.998947562651635e-05,
+      "loss": 0.6154137253761292,
+      "step": 362
+    },
+    {
+      "epoch": 2.736842105263158,
+      "grad_norm": 0.2177896499633789,
+      "learning_rate": 1.999934209936126e-05,
+      "loss": 0.7517685890197754,
+      "step": 364
+    },
+    {
+      "epoch": 2.7518796992481205,
+      "grad_norm": 0.18744942545890808,
+      "learning_rate": 1.9997368499288147e-05,
+      "loss": 0.8893219232559204,
+      "step": 366
+    },
+    {
+      "epoch": 2.7669172932330826,
+      "grad_norm": 0.12195390462875366,
+      "learning_rate": 1.9983557575635553e-05,
+      "loss": 0.7814352512359619,
+      "step": 368
+    },
+    {
+      "epoch": 2.781954887218045,
+      "grad_norm": 0.20601901412010193,
+      "learning_rate": 1.9957928567815327e-05,
+      "loss": 0.8266798853874207,
+      "step": 370
+    },
+    {
+      "epoch": 2.7969924812030076,
+      "grad_norm": 0.24167893826961517,
+      "learning_rate": 1.9920517178511025e-05,
+      "loss": 0.9625144004821777,
+      "step": 372
+    },
+    {
+      "epoch": 2.8120300751879697,
+      "grad_norm": 0.20055215060710907,
+      "learning_rate": 1.9871375523942014e-05,
+      "loss": 0.7290459275245667,
+      "step": 374
+    },
+    {
+      "epoch": 2.827067669172932,
+      "grad_norm": 0.349552720785141,
+      "learning_rate": 1.981057206126258e-05,
+      "loss": 1.0641855001449585,
+      "step": 376
+    },
+    {
+      "epoch": 2.8421052631578947,
+      "grad_norm": 0.19066953659057617,
+      "learning_rate": 1.9738191493197164e-05,
+      "loss": 1.1699275970458984,
+      "step": 378
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.4483891725540161,
+      "learning_rate": 1.96543346500446e-05,
+      "loss": 0.8162869811058044,
+      "step": 380
+    },
+    {
+      "epoch": 2.8721804511278197,
+      "grad_norm": 0.3727665841579437,
+      "learning_rate": 1.9559118349215685e-05,
+      "loss": 0.9681864380836487,
+      "step": 382
+    },
+    {
+      "epoch": 2.887218045112782,
+      "grad_norm": 0.8629833459854126,
+      "learning_rate": 1.9452675232499833e-05,
+      "loss": 0.9782270789146423,
+      "step": 384
+    },
+    {
+      "epoch": 2.9022556390977443,
+      "grad_norm": 0.40549924969673157,
+      "learning_rate": 1.9335153581287384e-05,
+      "loss": 0.7416080832481384,
+      "step": 386
+    },
+    {
+      "epoch": 2.917293233082707,
+      "grad_norm": 0.2615848481655121,
+      "learning_rate": 1.920671711000512e-05,
+      "loss": 0.9780222773551941,
+      "step": 388
+    },
+    {
+      "epoch": 2.932330827067669,
+      "grad_norm": 0.1308208554983139,
+      "learning_rate": 1.906754473805263e-05,
+      "loss": 0.8454363942146301,
+      "step": 390
+    },
+    {
+      "epoch": 2.9473684210526314,
+      "grad_norm": 0.19276592135429382,
+      "learning_rate": 1.8917830340557287e-05,
+      "loss": 0.974253237247467,
+      "step": 392
+    },
+    {
+      "epoch": 2.962406015037594,
+      "grad_norm": 0.40342700481414795,
+      "learning_rate": 1.875778247829503e-05,
+      "loss": 0.767073392868042,
+      "step": 394
+    },
+    {
+      "epoch": 2.9774436090225564,
+      "grad_norm": 0.28507354855537415,
+      "learning_rate": 1.8587624107153192e-05,
+      "loss": 1.0176156759262085,
+      "step": 396
+    },
+    {
+      "epoch": 2.992481203007519,
+      "grad_norm": 0.08241955190896988,
+      "learning_rate": 1.840759226754013e-05,
+      "loss": 0.8344214558601379,
+      "step": 398
+    },
+    {
+      "epoch": 3.007518796992481,
+      "grad_norm": 0.11373747140169144,
+      "learning_rate": 1.8217937754174265e-05,
+      "loss": 0.9761697053909302,
+      "step": 400
+    },
+    {
+      "epoch": 3.0225563909774436,
+      "grad_norm": 0.34416520595550537,
+      "learning_rate": 1.8018924766712608e-05,
+      "loss": 0.7501745223999023,
+      "step": 402
+    },
+    {
+      "epoch": 3.037593984962406,
+      "grad_norm": 0.7743138074874878,
+      "learning_rate": 1.7810830541705425e-05,
+      "loss": 0.9392253160476685,
+      "step": 404
+    },
+    {
+      "epoch": 3.0526315789473686,
+      "grad_norm": 0.2988702356815338,
+      "learning_rate": 1.7593944966389767e-05,
+      "loss": 0.6723920702934265,
+      "step": 406
+    },
+    {
+      "epoch": 3.0676691729323307,
+      "grad_norm": 0.1014496460556984,
+      "learning_rate": 1.7368570174859855e-05,
+      "loss": 0.5313879251480103,
+      "step": 408
+    },
+    {
+      "epoch": 3.082706766917293,
+      "grad_norm": 0.1569093018770218,
+      "learning_rate": 1.713502012717686e-05,
+      "loss": 0.8502162098884583,
+      "step": 410
+    },
+    {
+      "epoch": 3.0977443609022557,
+      "grad_norm": 0.4344463348388672,
+      "learning_rate": 1.6893620172004526e-05,
+      "loss": 0.6262786388397217,
+      "step": 412
+    },
+    {
+      "epoch": 3.112781954887218,
+      "grad_norm": 0.1802617460489273,
+      "learning_rate": 1.664470659337968e-05,
+      "loss": 0.9044028520584106,
+      "step": 414
+    },
+    {
+      "epoch": 3.1278195488721803,
+      "grad_norm": 0.09171602874994278,
+      "learning_rate": 1.638862614224926e-05,
+      "loss": 0.4721546471118927,
+      "step": 416
+    },
+    {
+      "epoch": 3.142857142857143,
+      "grad_norm": 0.21735042333602905,
+      "learning_rate": 1.6125735553426228e-05,
+      "loss": 0.7782120108604431,
+      "step": 418
+    },
+    {
+      "epoch": 3.1578947368421053,
+      "grad_norm": 0.10799870640039444,
+      "learning_rate": 1.5856401048637505e-05,
+      "loss": 0.8657242059707642,
+      "step": 420
+    },
+    {
+      "epoch": 3.172932330827068,
+      "grad_norm": 0.2299298644065857,
+      "learning_rate": 1.5580997826355945e-05,
+      "loss": 0.7349554896354675,
+      "step": 422
+    },
+    {
+      "epoch": 3.18796992481203,
+      "grad_norm": 0.1193917915225029,
+      "learning_rate": 1.5299909539127293e-05,
+      "loss": 0.8061299324035645,
+      "step": 424
+    },
+    {
+      "epoch": 3.2030075187969924,
+      "grad_norm": 0.313683420419693,
+      "learning_rate": 1.5013527759120128e-05,
+      "loss": 0.7771633267402649,
+      "step": 426
+    },
+    {
+      "epoch": 3.218045112781955,
+      "grad_norm": 0.28184592723846436,
+      "learning_rate": 1.4722251432643222e-05,
+      "loss": 0.7236590385437012,
+      "step": 428
+    },
+    {
+      "epoch": 3.2330827067669174,
+      "grad_norm": 0.2527337968349457,
+      "learning_rate": 1.4426486324390432e-05,
+      "loss": 0.5557109713554382,
+      "step": 430
+    },
+    {
+      "epoch": 3.2481203007518795,
+      "grad_norm": 0.14354459941387177,
+      "learning_rate": 1.4126644452187057e-05,
+      "loss": 0.5296133160591125,
+      "step": 432
+    },
+    {
+      "epoch": 3.263157894736842,
+      "grad_norm": 0.45321938395500183,
+      "learning_rate": 1.382314351302531e-05,
+      "loss": 0.5751173496246338,
+      "step": 434
+    },
+    {
+      "epoch": 3.2781954887218046,
+      "grad_norm": 0.5287338495254517,
+      "learning_rate": 1.351640630118838e-05,
+      "loss": 0.5346217751502991,
+      "step": 436
+    },
+    {
+      "epoch": 3.293233082706767,
+      "grad_norm": 0.07787490636110306,
+      "learning_rate": 1.3206860119273606e-05,
+      "loss": 0.5648679137229919,
+      "step": 438
+    },
+    {
+      "epoch": 3.308270676691729,
+      "grad_norm": 0.13844254612922668,
+      "learning_rate": 1.2894936182935407e-05,
+      "loss": 1.0638585090637207,
+      "step": 440
+    },
+    {
+      "epoch": 3.3233082706766917,
+      "grad_norm": 0.19562360644340515,
+      "learning_rate": 1.2581069020176986e-05,
+      "loss": 0.452185720205307,
+      "step": 442
+    },
+    {
+      "epoch": 3.338345864661654,
+      "grad_norm": 0.3099493980407715,
+      "learning_rate": 1.2265695866027849e-05,
+      "loss": 0.7373262047767639,
+      "step": 444
+    },
+    {
+      "epoch": 3.3533834586466167,
+      "grad_norm": 0.2395428866147995,
+      "learning_rate": 1.1949256053450243e-05,
+      "loss": 0.5868238806724548,
+      "step": 446
+    },
+    {
+      "epoch": 3.3684210526315788,
+      "grad_norm": 0.1869363635778427,
+      "learning_rate": 1.1632190401322987e-05,
+      "loss": 0.8645040392875671,
+      "step": 448
+    },
+    {
+      "epoch": 3.3834586466165413,
+      "grad_norm": 0.37590962648391724,
+      "learning_rate": 1.1314940600355459e-05,
+      "loss": 0.9512546062469482,
+      "step": 450
+    },
+    {
+      "epoch": 3.398496240601504,
+      "grad_norm": 0.2025744765996933,
+      "learning_rate": 1.099794859778695e-05,
+      "loss": 0.9128929376602173,
+      "step": 452
+    },
+    {
+      "epoch": 3.4135338345864663,
+      "grad_norm": 0.17627839744091034,
+      "learning_rate": 1.0681655981728753e-05,
+      "loss": 0.9643846750259399,
+      "step": 454
+    },
+    {
+      "epoch": 3.4285714285714284,
+      "grad_norm": 1.0564072132110596,
+      "learning_rate": 1.0366503366006477e-05,
+      "loss": 0.8578605055809021,
+      "step": 456
+    },
+    {
+      "epoch": 3.443609022556391,
+      "grad_norm": 0.3315916061401367,
+      "learning_rate": 1.0052929776359614e-05,
+      "loss": 0.5702735781669617,
+      "step": 458
+    },
+    {
+      "epoch": 3.4586466165413534,
+      "grad_norm": 0.1655089259147644,
+      "learning_rate": 9.741372038853437e-06,
+      "loss": 0.9215094447135925,
+      "step": 460
+    },
+    {
+      "epoch": 3.473684210526316,
+      "grad_norm": 0.26911604404449463,
+      "learning_rate": 9.432264171355117e-06,
+      "loss": 0.8250299096107483,
+      "step": 462
+    },
+    {
+      "epoch": 3.488721804511278,
+      "grad_norm": 0.22876925766468048,
+      "learning_rate": 9.126036778921812e-06,
+      "loss": 0.8027570843696594,
+      "step": 464
+    },
+    {
+      "epoch": 3.5037593984962405,
+      "grad_norm": 0.172451913356781,
+      "learning_rate": 8.8231164539431e-06,
+      "loss": 0.8936192989349365,
+      "step": 466
+    },
+    {
+      "epoch": 3.518796992481203,
+      "grad_norm": 0.2068740874528885,
+      "learning_rate": 8.52392518187319e-06,
+      "loss": 0.6892407536506653,
+      "step": 468
+    },
+    {
+      "epoch": 3.5338345864661656,
+      "grad_norm": 0.18001121282577515,
+      "learning_rate": 8.228879753380818e-06,
+      "loss": 0.7898232340812683,
+      "step": 470
+    },
+    {
+      "epoch": 3.548872180451128,
+      "grad_norm": 0.19327786564826965,
+      "learning_rate": 7.938391183735972e-06,
+      "loss": 0.7635357975959778,
+      "step": 472
+    },
+    {
+      "epoch": 3.56390977443609,
+      "grad_norm": 0.20077842473983765,
+      "learning_rate": 7.65286414024186e-06,
+      "loss": 0.8505274653434753,
+      "step": 474
+    },
+    {
+      "epoch": 3.5789473684210527,
+      "grad_norm": 0.19398260116577148,
+      "learning_rate": 7.3726963785101094e-06,
+      "loss": 0.38889795541763306,
+      "step": 476
+    },
+    {
+      "epoch": 3.593984962406015,
+      "grad_norm": 0.14551198482513428,
+      "learning_rate": 7.098278188364236e-06,
+      "loss": 0.365261048078537,
+      "step": 478
+    },
+    {
+      "epoch": 3.6090225563909772,
+      "grad_norm": 0.17524459958076477,
+      "learning_rate": 6.829991850143528e-06,
+      "loss": 0.9008171558380127,
+      "step": 480
+    },
+    {
+      "epoch": 3.6240601503759398,
+      "grad_norm": 0.3653651475906372,
+      "learning_rate": 6.5682111021644835e-06,
+      "loss": 0.900292158126831,
+      "step": 482
+    },
+    {
+      "epoch": 3.6390977443609023,
+      "grad_norm": 0.19287094473838806,
+      "learning_rate": 6.313300620081777e-06,
+      "loss": 0.7757176160812378,
+      "step": 484
+    },
+    {
+      "epoch": 3.654135338345865,
+      "grad_norm": 0.20128026604652405,
+      "learning_rate": 6.065615508874151e-06,
+      "loss": 0.3761276602745056,
+      "step": 486
+    },
+    {
+      "epoch": 3.6691729323308273,
+      "grad_norm": 0.1511046588420868,
+      "learning_rate": 5.82550080816269e-06,
+      "loss": 0.7676253318786621,
+      "step": 488
+    },
+    {
+      "epoch": 3.6842105263157894,
+      "grad_norm": 0.13384103775024414,
+      "learning_rate": 5.593291011550776e-06,
+      "loss": 0.9202339053153992,
+      "step": 490
+    },
+    {
+      "epoch": 3.699248120300752,
+      "grad_norm": 0.4986729025840759,
+      "learning_rate": 5.3693096006552025e-06,
+      "loss": 0.970383882522583,
+      "step": 492
+    },
+    {
+      "epoch": 3.7142857142857144,
+      "grad_norm": 0.5382466316223145,
+      "learning_rate": 5.15386859447772e-06,
+      "loss": 0.48905929923057556,
+      "step": 494
+    },
+    {
+      "epoch": 3.7293233082706765,
+      "grad_norm": 0.14699164032936096,
+      "learning_rate": 4.947268114744566e-06,
+      "loss": 0.9869436025619507,
+      "step": 496
+    },
+    {
+      "epoch": 3.744360902255639,
+      "grad_norm": 0.17605887353420258,
+      "learning_rate": 4.749795967819596e-06,
+      "loss": 1.0997637510299683,
+      "step": 498
+    },
+    {
+      "epoch": 3.7593984962406015,
+      "grad_norm": 0.24531900882720947,
+      "learning_rate": 4.561727243773504e-06,
+      "loss": 0.9353739023208618,
+      "step": 500
+    },
+    {
+      "epoch": 3.774436090225564,
+      "grad_norm": 0.23771578073501587,
+      "learning_rate": 4.383323933167479e-06,
+      "loss": 0.7048070430755615,
+      "step": 502
+    },
+    {
+      "epoch": 3.7894736842105265,
+      "grad_norm": 0.1848079115152359,
+      "learning_rate": 4.214834562085255e-06,
+      "loss": 0.6421667337417603,
+      "step": 504
+    },
+    {
+      "epoch": 3.8045112781954886,
+      "grad_norm": 1.46511971950531,
+      "learning_rate": 4.056493845922015e-06,
+      "loss": 0.5478267669677734,
+      "step": 506
+    },
+    {
+      "epoch": 3.819548872180451,
+      "grad_norm": 0.1415521800518036,
+      "learning_rate": 3.908522362412276e-06,
+      "loss": 0.8992588520050049,
+      "step": 508
+    },
+    {
+      "epoch": 3.8345864661654137,
+      "grad_norm": 0.6491537690162659,
+      "learning_rate": 3.7711262443524467e-06,
+      "loss": 0.591274082660675,
+      "step": 510
+    },
+    {
+      "epoch": 3.8496240601503757,
+      "grad_norm": 0.2289588451385498,
+      "learning_rate": 3.6444968924459545e-06,
+      "loss": 0.771586000919342,
+      "step": 512
+    },
+    {
+      "epoch": 3.8646616541353382,
+      "grad_norm": 0.1354115605354309,
+      "learning_rate": 3.5288107086710545e-06,
+      "loss": 0.7341738939285278,
+      "step": 514
+    },
+    {
+      "epoch": 3.8796992481203008,
+      "grad_norm": 0.1334567666053772,
+      "learning_rate": 3.4242288505426928e-06,
+      "loss": 1.0383778810501099,
+      "step": 516
+    },
+    {
+      "epoch": 3.8947368421052633,
+      "grad_norm": 0.17545144259929657,
+      "learning_rate": 3.3308970066108413e-06,
+      "loss": 0.5621640682220459,
+      "step": 518
+    },
+    {
+      "epoch": 3.909774436090226,
+      "grad_norm": 0.09879021346569061,
+      "learning_rate": 3.2489451935079473e-06,
+      "loss": 0.9413120150566101,
+      "step": 520
+    },
+    {
+      "epoch": 3.924812030075188,
+      "grad_norm": 0.2358972579240799,
+      "learning_rate": 3.1784875748282608e-06,
+      "loss": 1.0549259185791016,
+      "step": 522
+    },
+    {
+      "epoch": 3.9398496240601504,
+      "grad_norm": 0.1280510276556015,
+      "learning_rate": 3.119622302091404e-06,
+      "loss": 1.0559005737304688,
+      "step": 524
+    },
+    {
+      "epoch": 3.954887218045113,
+      "grad_norm": 0.24243058264255524,
+      "learning_rate": 3.072431378011632e-06,
+      "loss": 0.9401967525482178,
+      "step": 526
+    },
+    {
+      "epoch": 3.969924812030075,
+      "grad_norm": 0.1816936582326889,
+      "learning_rate": 3.0369805422633432e-06,
+      "loss": 0.6218282580375671,
+      "step": 528
+    },
+    {
+      "epoch": 3.9849624060150375,
+      "grad_norm": 0.4231378734111786,
+      "learning_rate": 3.0133191799019103e-06,
+      "loss": 0.5278748869895935,
+      "step": 530
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.3513203263282776,
+      "learning_rate": 3.001480252567475e-06,
+      "loss": 0.6851533651351929,
+      "step": 532
+    },
+    {
+      "epoch": 4.0,
+      "step": 532,
+      "total_flos": 3.800554784105169e+18,
+      "train_loss": 0.9838281438538903,
+      "train_runtime": 15715.7329,
+      "train_samples_per_second": 4.062,
+      "train_steps_per_second": 0.034
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 532,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 260,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.800554784105169e+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 sha256:2570955afd4eade3ef4eb2917f6645ea424aa7a13c2d2638bb3f503a277a73a7
+size 5649