step 4400 | loss 0.5551

.gitattributes

@@ -54,3 +54,4 @@ checkpoint-3600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoint-3800/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoint-4000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoint-4200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-4400/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoint-4400/README.md

@@ -0,0 +1,210 @@
+---
+base_model: unsloth/qwen3-8b-unsloth-bnb-4bit
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:unsloth/qwen3-8b-unsloth-bnb-4bit
+- lora
+- sft
+- transformers
+- trl
+- unsloth
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.19.1

checkpoint-4400/adapter_config.json

@@ -0,0 +1,52 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3ForCausalLM",
+    "parent_library": "transformers.models.qwen3.modeling_qwen3",
+    "unsloth_fixed": true
+  },
+  "base_model_name_or_path": "unsloth/qwen3-8b-unsloth-bnb-4bit",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate_proj",
+    "down_proj",
+    "up_proj",
+    "o_proj",
+    "v_proj",
+    "q_proj",
+    "k_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoint-4400/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ba5ce334d97fa66595e390df4a38239bda767f8c63d85e8f28d16e66f713ffb
+size 349243752

checkpoint-4400/added_tokens.json

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

checkpoint-4400/chat_template.jinja

@@ -0,0 +1,97 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for forward_message in messages %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- set message = messages[index] %}
+    {%- set tool_start = '<tool_response>' %}
+    {%- set tool_start_length = tool_start|length %}
+    {%- set start_of_message = message.content[:tool_start_length] %}
+    {%- set tool_end = '</tool_response>' %}
+    {%- set tool_end_length = tool_end|length %}
+    {%- set start_pos = (message.content|length) - tool_end_length %}
+    {%- if start_pos < 0 %}
+        {%- set start_pos = 0 %}
+    {%- endif %}
+    {%- set end_of_message = message.content[start_pos:] %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(start_of_message == tool_start and end_of_message == tool_end) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = (message.content.split('</think>')|last).lstrip('\n') %}
+                {%- set reasoning_content = (message.content.split('</think>')|first).rstrip('\n') %}
+                {%- set reasoning_content = (reasoning_content.split('<think>')|last).lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- 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' }}
+    {%- endif %}
+{%- endif %}

checkpoint-4400/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f593a84a821e9590ab3d808512b7b7ac61bda5958e4b3a1f9b7769a8688888a
+size 179225907

checkpoint-4400/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8e2011629d8bed3ef560fa11175cac55684c4e12a72634bb24abf767b6c7399
+size 14645

checkpoint-4400/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa01c0faab1a39c11ceba075e1e73b81a5689cfd1ac0d27ee7fece150d320be6
+size 1383

checkpoint-4400/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c81dfdbd6517168869fbab5b82b19023de7bd0aea005af4358cde78a6511dc2
+size 1465

checkpoint-4400/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|vision_pad|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-4400/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae1a036a9837df9caeebb840d09d80e8feef0f6d2bae982970d1ad34f5946aff
+size 11422753

checkpoint-4400/tokenizer_config.json

@@ -0,0 +1,240 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 40960,
+  "pad_token": "<|vision_pad|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-4400/trainer_state.json

@@ -0,0 +1,3114 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.975312129258234,
+  "eval_steps": 500,
+  "global_step": 4400,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.009039037342523022,
+      "grad_norm": 7.855499744415283,
+      "learning_rate": 6.000000000000001e-07,
+      "loss": 1.4181,
+      "step": 10
+    },
+    {
+      "epoch": 0.018078074685046044,
+      "grad_norm": 3.3778610229492188,
+      "learning_rate": 1.2666666666666669e-06,
+      "loss": 1.2494,
+      "step": 20
+    },
+    {
+      "epoch": 0.027117112027569064,
+      "grad_norm": 2.6225311756134033,
+      "learning_rate": 1.9333333333333336e-06,
+      "loss": 1.0602,
+      "step": 30
+    },
+    {
+      "epoch": 0.03615614937009209,
+      "grad_norm": 1.594939112663269,
+      "learning_rate": 2.6e-06,
+      "loss": 0.9123,
+      "step": 40
+    },
+    {
+      "epoch": 0.045195186712615104,
+      "grad_norm": 0.9766483902931213,
+      "learning_rate": 3.266666666666667e-06,
+      "loss": 0.7841,
+      "step": 50
+    },
+    {
+      "epoch": 0.05423422405513813,
+      "grad_norm": 0.43368658423423767,
+      "learning_rate": 3.9333333333333335e-06,
+      "loss": 0.6803,
+      "step": 60
+    },
+    {
+      "epoch": 0.06327326139766115,
+      "grad_norm": 0.49014410376548767,
+      "learning_rate": 4.600000000000001e-06,
+      "loss": 0.5921,
+      "step": 70
+    },
+    {
+      "epoch": 0.07231229874018417,
+      "grad_norm": 0.46368488669395447,
+      "learning_rate": 5.2666666666666665e-06,
+      "loss": 0.6214,
+      "step": 80
+    },
+    {
+      "epoch": 0.0813513360827072,
+      "grad_norm": 0.46477940678596497,
+      "learning_rate": 5.933333333333335e-06,
+      "loss": 0.6157,
+      "step": 90
+    },
+    {
+      "epoch": 0.09039037342523021,
+      "grad_norm": 0.4853752553462982,
+      "learning_rate": 6.600000000000001e-06,
+      "loss": 0.6628,
+      "step": 100
+    },
+    {
+      "epoch": 0.09942941076775323,
+      "grad_norm": 0.36940082907676697,
+      "learning_rate": 7.266666666666668e-06,
+      "loss": 0.5775,
+      "step": 110
+    },
+    {
+      "epoch": 0.10846844811027626,
+      "grad_norm": 0.34785982966423035,
+      "learning_rate": 7.933333333333334e-06,
+      "loss": 0.6343,
+      "step": 120
+    },
+    {
+      "epoch": 0.11750748545279928,
+      "grad_norm": 0.4787227511405945,
+      "learning_rate": 8.6e-06,
+      "loss": 0.6016,
+      "step": 130
+    },
+    {
+      "epoch": 0.1265465227953223,
+      "grad_norm": 0.4643513262271881,
+      "learning_rate": 9.266666666666667e-06,
+      "loss": 0.6388,
+      "step": 140
+    },
+    {
+      "epoch": 0.1355855601378453,
+      "grad_norm": 0.384676456451416,
+      "learning_rate": 9.933333333333334e-06,
+      "loss": 0.5895,
+      "step": 150
+    },
+    {
+      "epoch": 0.14462459748036835,
+      "grad_norm": 0.4324859380722046,
+      "learning_rate": 9.99991503499922e-06,
+      "loss": 0.5942,
+      "step": 160
+    },
+    {
+      "epoch": 0.15366363482289136,
+      "grad_norm": 0.41830965876579285,
+      "learning_rate": 9.99962133253095e-06,
+      "loss": 0.6121,
+      "step": 170
+    },
+    {
+      "epoch": 0.1627026721654144,
+      "grad_norm": 0.3795354962348938,
+      "learning_rate": 9.999117855964797e-06,
+      "loss": 0.6041,
+      "step": 180
+    },
+    {
+      "epoch": 0.1717417095079374,
+      "grad_norm": 0.28756389021873474,
+      "learning_rate": 9.998404626425627e-06,
+      "loss": 0.5529,
+      "step": 190
+    },
+    {
+      "epoch": 0.18078074685046042,
+      "grad_norm": 0.40732458233833313,
+      "learning_rate": 9.997481673839125e-06,
+      "loss": 0.6321,
+      "step": 200
+    },
+    {
+      "epoch": 0.18981978419298345,
+      "grad_norm": 0.3699369430541992,
+      "learning_rate": 9.996349036930533e-06,
+      "loss": 0.5886,
+      "step": 210
+    },
+    {
+      "epoch": 0.19885882153550646,
+      "grad_norm": 0.4334062337875366,
+      "learning_rate": 9.995006763223028e-06,
+      "loss": 0.646,
+      "step": 220
+    },
+    {
+      "epoch": 0.2078978588780295,
+      "grad_norm": 0.3638211786746979,
+      "learning_rate": 9.993454909035724e-06,
+      "loss": 0.5996,
+      "step": 230
+    },
+    {
+      "epoch": 0.2169368962205525,
+      "grad_norm": 0.377408891916275,
+      "learning_rate": 9.991693539481317e-06,
+      "loss": 0.5999,
+      "step": 240
+    },
+    {
+      "epoch": 0.22597593356307552,
+      "grad_norm": 0.4337885081768036,
+      "learning_rate": 9.989722728463345e-06,
+      "loss": 0.6233,
+      "step": 250
+    },
+    {
+      "epoch": 0.23501497090559856,
+      "grad_norm": 0.35624244809150696,
+      "learning_rate": 9.98754255867309e-06,
+      "loss": 0.6279,
+      "step": 260
+    },
+    {
+      "epoch": 0.24405400824812157,
+      "grad_norm": 0.385580837726593,
+      "learning_rate": 9.985153121586111e-06,
+      "loss": 0.5896,
+      "step": 270
+    },
+    {
+      "epoch": 0.2530930455906446,
+      "grad_norm": 0.36690962314605713,
+      "learning_rate": 9.982554517458403e-06,
+      "loss": 0.6212,
+      "step": 280
+    },
+    {
+      "epoch": 0.26213208293316764,
+      "grad_norm": 0.3190907835960388,
+      "learning_rate": 9.979746855322192e-06,
+      "loss": 0.6019,
+      "step": 290
+    },
+    {
+      "epoch": 0.2711711202756906,
+      "grad_norm": 0.33014318346977234,
+      "learning_rate": 9.976730252981354e-06,
+      "loss": 0.5906,
+      "step": 300
+    },
+    {
+      "epoch": 0.28021015761821366,
+      "grad_norm": 0.4516271650791168,
+      "learning_rate": 9.973504837006487e-06,
+      "loss": 0.612,
+      "step": 310
+    },
+    {
+      "epoch": 0.2892491949607367,
+      "grad_norm": 0.3498607873916626,
+      "learning_rate": 9.97007074272958e-06,
+      "loss": 0.5591,
+      "step": 320
+    },
+    {
+      "epoch": 0.2982882323032597,
+      "grad_norm": 0.4618026316165924,
+      "learning_rate": 9.966428114238353e-06,
+      "loss": 0.5808,
+      "step": 330
+    },
+    {
+      "epoch": 0.3073272696457827,
+      "grad_norm": 0.35378628969192505,
+      "learning_rate": 9.962577104370206e-06,
+      "loss": 0.5983,
+      "step": 340
+    },
+    {
+      "epoch": 0.31636630698830576,
+      "grad_norm": 0.3605310916900635,
+      "learning_rate": 9.958517874705793e-06,
+      "loss": 0.5898,
+      "step": 350
+    },
+    {
+      "epoch": 0.3254053443308288,
+      "grad_norm": 0.41437646746635437,
+      "learning_rate": 9.954250595562267e-06,
+      "loss": 0.5746,
+      "step": 360
+    },
+    {
+      "epoch": 0.3344443816733518,
+      "grad_norm": 0.3715222477912903,
+      "learning_rate": 9.949775445986112e-06,
+      "loss": 0.5994,
+      "step": 370
+    },
+    {
+      "epoch": 0.3434834190158748,
+      "grad_norm": 0.41986992955207825,
+      "learning_rate": 9.945092613745642e-06,
+      "loss": 0.5492,
+      "step": 380
+    },
+    {
+      "epoch": 0.35252245635839785,
+      "grad_norm": 0.3996647000312805,
+      "learning_rate": 9.940202295323116e-06,
+      "loss": 0.6066,
+      "step": 390
+    },
+    {
+      "epoch": 0.36156149370092083,
+      "grad_norm": 0.3808385729789734,
+      "learning_rate": 9.935104695906506e-06,
+      "loss": 0.5628,
+      "step": 400
+    },
+    {
+      "epoch": 0.37060053104344387,
+      "grad_norm": 0.3926442563533783,
+      "learning_rate": 9.92980002938087e-06,
+      "loss": 0.6087,
+      "step": 410
+    },
+    {
+      "epoch": 0.3796395683859669,
+      "grad_norm": 0.3604278862476349,
+      "learning_rate": 9.924288518319394e-06,
+      "loss": 0.5661,
+      "step": 420
+    },
+    {
+      "epoch": 0.3886786057284899,
+      "grad_norm": 0.36008164286613464,
+      "learning_rate": 9.918570393974041e-06,
+      "loss": 0.5948,
+      "step": 430
+    },
+    {
+      "epoch": 0.3977176430710129,
+      "grad_norm": 0.3590993285179138,
+      "learning_rate": 9.912645896265858e-06,
+      "loss": 0.5692,
+      "step": 440
+    },
+    {
+      "epoch": 0.40675668041353596,
+      "grad_norm": 0.3519396483898163,
+      "learning_rate": 9.906515273774903e-06,
+      "loss": 0.5289,
+      "step": 450
+    },
+    {
+      "epoch": 0.415795717756059,
+      "grad_norm": 0.33698147535324097,
+      "learning_rate": 9.900178783729817e-06,
+      "loss": 0.5643,
+      "step": 460
+    },
+    {
+      "epoch": 0.424834755098582,
+      "grad_norm": 0.41117480397224426,
+      "learning_rate": 9.89363669199703e-06,
+      "loss": 0.5875,
+      "step": 470
+    },
+    {
+      "epoch": 0.433873792441105,
+      "grad_norm": 0.3583988547325134,
+      "learning_rate": 9.886889273069614e-06,
+      "loss": 0.5494,
+      "step": 480
+    },
+    {
+      "epoch": 0.44291282978362806,
+      "grad_norm": 0.3181304633617401,
+      "learning_rate": 9.879936810055746e-06,
+      "loss": 0.5652,
+      "step": 490
+    },
+    {
+      "epoch": 0.45195186712615104,
+      "grad_norm": 0.3814559280872345,
+      "learning_rate": 9.872779594666856e-06,
+      "loss": 0.6046,
+      "step": 500
+    },
+    {
+      "epoch": 0.4609909044686741,
+      "grad_norm": 0.33179372549057007,
+      "learning_rate": 9.865417927205363e-06,
+      "loss": 0.6156,
+      "step": 510
+    },
+    {
+      "epoch": 0.4700299418111971,
+      "grad_norm": 0.4139130115509033,
+      "learning_rate": 9.857852116552094e-06,
+      "loss": 0.5993,
+      "step": 520
+    },
+    {
+      "epoch": 0.47906897915372015,
+      "grad_norm": 0.32068437337875366,
+      "learning_rate": 9.850082480153306e-06,
+      "loss": 0.603,
+      "step": 530
+    },
+    {
+      "epoch": 0.48810801649624314,
+      "grad_norm": 0.9201164841651917,
+      "learning_rate": 9.842109344007386e-06,
+      "loss": 0.6036,
+      "step": 540
+    },
+    {
+      "epoch": 0.4971470538387662,
+      "grad_norm": 0.42034047842025757,
+      "learning_rate": 9.833933042651156e-06,
+      "loss": 0.5678,
+      "step": 550
+    },
+    {
+      "epoch": 0.5061860911812892,
+      "grad_norm": 0.3811807632446289,
+      "learning_rate": 9.825553919145845e-06,
+      "loss": 0.5596,
+      "step": 560
+    },
+    {
+      "epoch": 0.5152251285238122,
+      "grad_norm": 0.36168864369392395,
+      "learning_rate": 9.816972325062694e-06,
+      "loss": 0.5906,
+      "step": 570
+    },
+    {
+      "epoch": 0.5242641658663353,
+      "grad_norm": 0.3304130733013153,
+      "learning_rate": 9.808188620468204e-06,
+      "loss": 0.5398,
+      "step": 580
+    },
+    {
+      "epoch": 0.5333032032088583,
+      "grad_norm": 0.32785892486572266,
+      "learning_rate": 9.799203173909028e-06,
+      "loss": 0.5982,
+      "step": 590
+    },
+    {
+      "epoch": 0.5423422405513812,
+      "grad_norm": 0.39726918935775757,
+      "learning_rate": 9.790016362396506e-06,
+      "loss": 0.5561,
+      "step": 600
+    },
+    {
+      "epoch": 0.5513812778939043,
+      "grad_norm": 0.3174471855163574,
+      "learning_rate": 9.780628571390853e-06,
+      "loss": 0.5268,
+      "step": 610
+    },
+    {
+      "epoch": 0.5604203152364273,
+      "grad_norm": 0.4243045151233673,
+      "learning_rate": 9.771040194784973e-06,
+      "loss": 0.5954,
+      "step": 620
+    },
+    {
+      "epoch": 0.5694593525789503,
+      "grad_norm": 0.35231855511665344,
+      "learning_rate": 9.761251634887949e-06,
+      "loss": 0.5711,
+      "step": 630
+    },
+    {
+      "epoch": 0.5784983899214734,
+      "grad_norm": 0.349280446767807,
+      "learning_rate": 9.751263302408146e-06,
+      "loss": 0.6263,
+      "step": 640
+    },
+    {
+      "epoch": 0.5875374272639964,
+      "grad_norm": 0.7063365578651428,
+      "learning_rate": 9.741075616435995e-06,
+      "loss": 0.6093,
+      "step": 650
+    },
+    {
+      "epoch": 0.5965764646065194,
+      "grad_norm": 0.3803994059562683,
+      "learning_rate": 9.730689004426392e-06,
+      "loss": 0.5552,
+      "step": 660
+    },
+    {
+      "epoch": 0.6056155019490425,
+      "grad_norm": 0.36839503049850464,
+      "learning_rate": 9.720103902180776e-06,
+      "loss": 0.5987,
+      "step": 670
+    },
+    {
+      "epoch": 0.6146545392915654,
+      "grad_norm": 0.39339619874954224,
+      "learning_rate": 9.709320753828837e-06,
+      "loss": 0.5621,
+      "step": 680
+    },
+    {
+      "epoch": 0.6236935766340884,
+      "grad_norm": 0.35199737548828125,
+      "learning_rate": 9.698340011809883e-06,
+      "loss": 0.5732,
+      "step": 690
+    },
+    {
+      "epoch": 0.6327326139766115,
+      "grad_norm": 0.37193241715431213,
+      "learning_rate": 9.687162136853858e-06,
+      "loss": 0.5534,
+      "step": 700
+    },
+    {
+      "epoch": 0.6417716513191345,
+      "grad_norm": 0.3605504035949707,
+      "learning_rate": 9.675787597962007e-06,
+      "loss": 0.608,
+      "step": 710
+    },
+    {
+      "epoch": 0.6508106886616576,
+      "grad_norm": 0.3748184144496918,
+      "learning_rate": 9.664216872387202e-06,
+      "loss": 0.5903,
+      "step": 720
+    },
+    {
+      "epoch": 0.6598497260041806,
+      "grad_norm": 0.3527565896511078,
+      "learning_rate": 9.652450445613913e-06,
+      "loss": 0.5836,
+      "step": 730
+    },
+    {
+      "epoch": 0.6688887633467036,
+      "grad_norm": 0.34412819147109985,
+      "learning_rate": 9.64048881133784e-06,
+      "loss": 0.5672,
+      "step": 740
+    },
+    {
+      "epoch": 0.6779278006892266,
+      "grad_norm": 0.3731626868247986,
+      "learning_rate": 9.628332471445206e-06,
+      "loss": 0.6236,
+      "step": 750
+    },
+    {
+      "epoch": 0.6869668380317496,
+      "grad_norm": 0.35358232259750366,
+      "learning_rate": 9.615981935991683e-06,
+      "loss": 0.5535,
+      "step": 760
+    },
+    {
+      "epoch": 0.6960058753742726,
+      "grad_norm": 0.3264493942260742,
+      "learning_rate": 9.603437723181002e-06,
+      "loss": 0.6,
+      "step": 770
+    },
+    {
+      "epoch": 0.7050449127167957,
+      "grad_norm": 0.3995952308177948,
+      "learning_rate": 9.59070035934321e-06,
+      "loss": 0.5739,
+      "step": 780
+    },
+    {
+      "epoch": 0.7140839500593187,
+      "grad_norm": 0.4088019132614136,
+      "learning_rate": 9.577770378912584e-06,
+      "loss": 0.5463,
+      "step": 790
+    },
+    {
+      "epoch": 0.7231229874018417,
+      "grad_norm": 0.36028265953063965,
+      "learning_rate": 9.564648324405206e-06,
+      "loss": 0.6317,
+      "step": 800
+    },
+    {
+      "epoch": 0.7321620247443648,
+      "grad_norm": 0.36535605788230896,
+      "learning_rate": 9.551334746396203e-06,
+      "loss": 0.5675,
+      "step": 810
+    },
+    {
+      "epoch": 0.7412010620868877,
+      "grad_norm": 0.339760959148407,
+      "learning_rate": 9.537830203496642e-06,
+      "loss": 0.5587,
+      "step": 820
+    },
+    {
+      "epoch": 0.7502400994294107,
+      "grad_norm": 0.36886465549468994,
+      "learning_rate": 9.524135262330098e-06,
+      "loss": 0.5954,
+      "step": 830
+    },
+    {
+      "epoch": 0.7592791367719338,
+      "grad_norm": 0.36694473028182983,
+      "learning_rate": 9.510250497508873e-06,
+      "loss": 0.5626,
+      "step": 840
+    },
+    {
+      "epoch": 0.7683181741144568,
+      "grad_norm": 0.3380376100540161,
+      "learning_rate": 9.496176491609893e-06,
+      "loss": 0.6203,
+      "step": 850
+    },
+    {
+      "epoch": 0.7773572114569798,
+      "grad_norm": 0.33642667531967163,
+      "learning_rate": 9.481913835150256e-06,
+      "loss": 0.5856,
+      "step": 860
+    },
+    {
+      "epoch": 0.7863962487995029,
+      "grad_norm": 0.43253782391548157,
+      "learning_rate": 9.467463126562461e-06,
+      "loss": 0.586,
+      "step": 870
+    },
+    {
+      "epoch": 0.7954352861420259,
+      "grad_norm": 0.35205066204071045,
+      "learning_rate": 9.452824972169298e-06,
+      "loss": 0.581,
+      "step": 880
+    },
+    {
+      "epoch": 0.804474323484549,
+      "grad_norm": 0.33954545855522156,
+      "learning_rate": 9.437999986158413e-06,
+      "loss": 0.5961,
+      "step": 890
+    },
+    {
+      "epoch": 0.8135133608270719,
+      "grad_norm": 0.29890018701553345,
+      "learning_rate": 9.422988790556524e-06,
+      "loss": 0.6118,
+      "step": 900
+    },
+    {
+      "epoch": 0.8225523981695949,
+      "grad_norm": 0.3720478415489197,
+      "learning_rate": 9.407792015203331e-06,
+      "loss": 0.5773,
+      "step": 910
+    },
+    {
+      "epoch": 0.831591435512118,
+      "grad_norm": 0.48966363072395325,
+      "learning_rate": 9.392410297725099e-06,
+      "loss": 0.5739,
+      "step": 920
+    },
+    {
+      "epoch": 0.840630472854641,
+      "grad_norm": 0.33022603392601013,
+      "learning_rate": 9.37684428350788e-06,
+      "loss": 0.5572,
+      "step": 930
+    },
+    {
+      "epoch": 0.849669510197164,
+      "grad_norm": 0.3322959840297699,
+      "learning_rate": 9.36109462567046e-06,
+      "loss": 0.5884,
+      "step": 940
+    },
+    {
+      "epoch": 0.8587085475396871,
+      "grad_norm": 0.38541945815086365,
+      "learning_rate": 9.345161985036937e-06,
+      "loss": 0.5485,
+      "step": 950
+    },
+    {
+      "epoch": 0.86774758488221,
+      "grad_norm": 0.3497406244277954,
+      "learning_rate": 9.329047030109e-06,
+      "loss": 0.5774,
+      "step": 960
+    },
+    {
+      "epoch": 0.876786622224733,
+      "grad_norm": 0.3379673361778259,
+      "learning_rate": 9.312750437037886e-06,
+      "loss": 0.5682,
+      "step": 970
+    },
+    {
+      "epoch": 0.8858256595672561,
+      "grad_norm": 0.3425869047641754,
+      "learning_rate": 9.296272889595997e-06,
+      "loss": 0.597,
+      "step": 980
+    },
+    {
+      "epoch": 0.8948646969097791,
+      "grad_norm": 0.3111202120780945,
+      "learning_rate": 9.279615079148227e-06,
+      "loss": 0.5629,
+      "step": 990
+    },
+    {
+      "epoch": 0.9039037342523021,
+      "grad_norm": 0.36501139402389526,
+      "learning_rate": 9.262777704622939e-06,
+      "loss": 0.5763,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9129427715948252,
+      "grad_norm": 0.4016779661178589,
+      "learning_rate": 9.245761472482646e-06,
+      "loss": 0.6107,
+      "step": 1010
+    },
+    {
+      "epoch": 0.9219818089373482,
+      "grad_norm": 0.3700155019760132,
+      "learning_rate": 9.228567096694366e-06,
+      "loss": 0.5753,
+      "step": 1020
+    },
+    {
+      "epoch": 0.9310208462798711,
+      "grad_norm": 0.32044124603271484,
+      "learning_rate": 9.211195298699674e-06,
+      "loss": 0.5664,
+      "step": 1030
+    },
+    {
+      "epoch": 0.9400598836223942,
+      "grad_norm": 0.3247808516025543,
+      "learning_rate": 9.193646807384421e-06,
+      "loss": 0.5559,
+      "step": 1040
+    },
+    {
+      "epoch": 0.9490989209649172,
+      "grad_norm": 0.37978535890579224,
+      "learning_rate": 9.175922359048156e-06,
+      "loss": 0.6026,
+      "step": 1050
+    },
+    {
+      "epoch": 0.9581379583074403,
+      "grad_norm": 0.37707069516181946,
+      "learning_rate": 9.15802269737323e-06,
+      "loss": 0.5683,
+      "step": 1060
+    },
+    {
+      "epoch": 0.9671769956499633,
+      "grad_norm": 0.37162524461746216,
+      "learning_rate": 9.139948573393602e-06,
+      "loss": 0.622,
+      "step": 1070
+    },
+    {
+      "epoch": 0.9762160329924863,
+      "grad_norm": 0.35211777687072754,
+      "learning_rate": 9.121700745463312e-06,
+      "loss": 0.5754,
+      "step": 1080
+    },
+    {
+      "epoch": 0.9852550703350094,
+      "grad_norm": 0.35246849060058594,
+      "learning_rate": 9.103279979224676e-06,
+      "loss": 0.6038,
+      "step": 1090
+    },
+    {
+      "epoch": 0.9942941076775323,
+      "grad_norm": 0.30493950843811035,
+      "learning_rate": 9.084687047576149e-06,
+      "loss": 0.5873,
+      "step": 1100
+    },
+    {
+      "epoch": 1.0027117112027568,
+      "grad_norm": 0.37721627950668335,
+      "learning_rate": 9.065922730639906e-06,
+      "loss": 0.5726,
+      "step": 1110
+    },
+    {
+      "epoch": 1.01175074854528,
+      "grad_norm": 0.4114892780780792,
+      "learning_rate": 9.046987815729108e-06,
+      "loss": 0.547,
+      "step": 1120
+    },
+    {
+      "epoch": 1.020789785887803,
+      "grad_norm": 0.3268738090991974,
+      "learning_rate": 9.027883097314859e-06,
+      "loss": 0.5896,
+      "step": 1130
+    },
+    {
+      "epoch": 1.0298288232303259,
+      "grad_norm": 0.38347479701042175,
+      "learning_rate": 9.008609376992879e-06,
+      "loss": 0.5496,
+      "step": 1140
+    },
+    {
+      "epoch": 1.038867860572849,
+      "grad_norm": 0.4380169212818146,
+      "learning_rate": 8.989167463449874e-06,
+      "loss": 0.5468,
+      "step": 1150
+    },
+    {
+      "epoch": 1.047906897915372,
+      "grad_norm": 0.3695785105228424,
+      "learning_rate": 8.9695581724296e-06,
+      "loss": 0.5636,
+      "step": 1160
+    },
+    {
+      "epoch": 1.056945935257895,
+      "grad_norm": 0.31363677978515625,
+      "learning_rate": 8.94978232669863e-06,
+      "loss": 0.5774,
+      "step": 1170
+    },
+    {
+      "epoch": 1.065984972600418,
+      "grad_norm": 0.32090502977371216,
+      "learning_rate": 8.929840756011847e-06,
+      "loss": 0.5819,
+      "step": 1180
+    },
+    {
+      "epoch": 1.0750240099429411,
+      "grad_norm": 0.2986926734447479,
+      "learning_rate": 8.909734297077618e-06,
+      "loss": 0.5565,
+      "step": 1190
+    },
+    {
+      "epoch": 1.084063047285464,
+      "grad_norm": 0.3923097252845764,
+      "learning_rate": 8.889463793522687e-06,
+      "loss": 0.5971,
+      "step": 1200
+    },
+    {
+      "epoch": 1.093102084627987,
+      "grad_norm": 0.45070353150367737,
+      "learning_rate": 8.86903009585679e-06,
+      "loss": 0.5481,
+      "step": 1210
+    },
+    {
+      "epoch": 1.1021411219705102,
+      "grad_norm": 0.42040202021598816,
+      "learning_rate": 8.848434061436954e-06,
+      "loss": 0.5468,
+      "step": 1220
+    },
+    {
+      "epoch": 1.111180159313033,
+      "grad_norm": 0.45057106018066406,
+      "learning_rate": 8.827676554431534e-06,
+      "loss": 0.6003,
+      "step": 1230
+    },
+    {
+      "epoch": 1.1202191966555561,
+      "grad_norm": 0.34813055396080017,
+      "learning_rate": 8.806758445783954e-06,
+      "loss": 0.5673,
+      "step": 1240
+    },
+    {
+      "epoch": 1.1292582339980792,
+      "grad_norm": 0.39887797832489014,
+      "learning_rate": 8.785680613176153e-06,
+      "loss": 0.5783,
+      "step": 1250
+    },
+    {
+      "epoch": 1.138297271340602,
+      "grad_norm": 0.3870787024497986,
+      "learning_rate": 8.764443940991776e-06,
+      "loss": 0.5533,
+      "step": 1260
+    },
+    {
+      "epoch": 1.1473363086831252,
+      "grad_norm": 0.355211466550827,
+      "learning_rate": 8.743049320279053e-06,
+      "loss": 0.605,
+      "step": 1270
+    },
+    {
+      "epoch": 1.1563753460256483,
+      "grad_norm": 0.335464209318161,
+      "learning_rate": 8.721497648713423e-06,
+      "loss": 0.5681,
+      "step": 1280
+    },
+    {
+      "epoch": 1.1654143833681714,
+      "grad_norm": 0.411471962928772,
+      "learning_rate": 8.699789830559856e-06,
+      "loss": 0.5502,
+      "step": 1290
+    },
+    {
+      "epoch": 1.1744534207106943,
+      "grad_norm": 0.2810121178627014,
+      "learning_rate": 8.677926776634933e-06,
+      "loss": 0.5689,
+      "step": 1300
+    },
+    {
+      "epoch": 1.1834924580532173,
+      "grad_norm": 0.44342437386512756,
+      "learning_rate": 8.655909404268603e-06,
+      "loss": 0.5694,
+      "step": 1310
+    },
+    {
+      "epoch": 1.1925314953957404,
+      "grad_norm": 0.40787559747695923,
+      "learning_rate": 8.633738637265716e-06,
+      "loss": 0.6138,
+      "step": 1320
+    },
+    {
+      "epoch": 1.2015705327382633,
+      "grad_norm": 0.40878427028656006,
+      "learning_rate": 8.611415405867251e-06,
+      "loss": 0.5663,
+      "step": 1330
+    },
+    {
+      "epoch": 1.2106095700807864,
+      "grad_norm": 0.4044298231601715,
+      "learning_rate": 8.588940646711293e-06,
+      "loss": 0.5703,
+      "step": 1340
+    },
+    {
+      "epoch": 1.2196486074233095,
+      "grad_norm": 0.38636693358421326,
+      "learning_rate": 8.566315302793717e-06,
+      "loss": 0.5878,
+      "step": 1350
+    },
+    {
+      "epoch": 1.2286876447658324,
+      "grad_norm": 0.4851875901222229,
+      "learning_rate": 8.543540323428646e-06,
+      "loss": 0.5393,
+      "step": 1360
+    },
+    {
+      "epoch": 1.2377266821083555,
+      "grad_norm": 0.3599594235420227,
+      "learning_rate": 8.520616664208597e-06,
+      "loss": 0.5637,
+      "step": 1370
+    },
+    {
+      "epoch": 1.2467657194508786,
+      "grad_norm": 0.393518328666687,
+      "learning_rate": 8.4975452869644e-06,
+      "loss": 0.5846,
+      "step": 1380
+    },
+    {
+      "epoch": 1.2558047567934014,
+      "grad_norm": 0.4419288635253906,
+      "learning_rate": 8.474327159724837e-06,
+      "loss": 0.5647,
+      "step": 1390
+    },
+    {
+      "epoch": 1.2648437941359245,
+      "grad_norm": 0.46883851289749146,
+      "learning_rate": 8.450963256676026e-06,
+      "loss": 0.5978,
+      "step": 1400
+    },
+    {
+      "epoch": 1.2738828314784476,
+      "grad_norm": 0.3780512511730194,
+      "learning_rate": 8.427454558120545e-06,
+      "loss": 0.5254,
+      "step": 1410
+    },
+    {
+      "epoch": 1.2829218688209707,
+      "grad_norm": 0.39592650532722473,
+      "learning_rate": 8.403802050436302e-06,
+      "loss": 0.6072,
+      "step": 1420
+    },
+    {
+      "epoch": 1.2919609061634936,
+      "grad_norm": 0.34795117378234863,
+      "learning_rate": 8.380006726035152e-06,
+      "loss": 0.5118,
+      "step": 1430
+    },
+    {
+      "epoch": 1.3009999435060167,
+      "grad_norm": 0.3898769021034241,
+      "learning_rate": 8.356069583321245e-06,
+      "loss": 0.5666,
+      "step": 1440
+    },
+    {
+      "epoch": 1.3100389808485398,
+      "grad_norm": 0.4135303199291229,
+      "learning_rate": 8.33199162664915e-06,
+      "loss": 0.5661,
+      "step": 1450
+    },
+    {
+      "epoch": 1.3190780181910626,
+      "grad_norm": 0.4047456979751587,
+      "learning_rate": 8.307773866281705e-06,
+      "loss": 0.5509,
+      "step": 1460
+    },
+    {
+      "epoch": 1.3281170555335857,
+      "grad_norm": 0.3716682493686676,
+      "learning_rate": 8.283417318347632e-06,
+      "loss": 0.5456,
+      "step": 1470
+    },
+    {
+      "epoch": 1.3371560928761088,
+      "grad_norm": 0.3757040500640869,
+      "learning_rate": 8.258923004798902e-06,
+      "loss": 0.5614,
+      "step": 1480
+    },
+    {
+      "epoch": 1.3461951302186317,
+      "grad_norm": 0.3806832432746887,
+      "learning_rate": 8.234291953367855e-06,
+      "loss": 0.5157,
+      "step": 1490
+    },
+    {
+      "epoch": 1.3552341675611548,
+      "grad_norm": 0.34731006622314453,
+      "learning_rate": 8.209525197524073e-06,
+      "loss": 0.5892,
+      "step": 1500
+    },
+    {
+      "epoch": 1.3642732049036779,
+      "grad_norm": 0.38066595792770386,
+      "learning_rate": 8.184623776431036e-06,
+      "loss": 0.5416,
+      "step": 1510
+    },
+    {
+      "epoch": 1.3733122422462007,
+      "grad_norm": 0.3750848174095154,
+      "learning_rate": 8.159588734902495e-06,
+      "loss": 0.5464,
+      "step": 1520
+    },
+    {
+      "epoch": 1.3823512795887238,
+      "grad_norm": 0.42894601821899414,
+      "learning_rate": 8.134421123358653e-06,
+      "loss": 0.5601,
+      "step": 1530
+    },
+    {
+      "epoch": 1.391390316931247,
+      "grad_norm": 0.40111929178237915,
+      "learning_rate": 8.109121997782088e-06,
+      "loss": 0.5573,
+      "step": 1540
+    },
+    {
+      "epoch": 1.4004293542737698,
+      "grad_norm": 0.4170910120010376,
+      "learning_rate": 8.083692419673436e-06,
+      "loss": 0.53,
+      "step": 1550
+    },
+    {
+      "epoch": 1.409468391616293,
+      "grad_norm": 0.37612515687942505,
+      "learning_rate": 8.058133456006867e-06,
+      "loss": 0.4952,
+      "step": 1560
+    },
+    {
+      "epoch": 1.418507428958816,
+      "grad_norm": 0.42817550897598267,
+      "learning_rate": 8.032446179185304e-06,
+      "loss": 0.5558,
+      "step": 1570
+    },
+    {
+      "epoch": 1.4275464663013389,
+      "grad_norm": 0.39047130942344666,
+      "learning_rate": 8.00663166699544e-06,
+      "loss": 0.5615,
+      "step": 1580
+    },
+    {
+      "epoch": 1.436585503643862,
+      "grad_norm": 0.4020339846611023,
+      "learning_rate": 7.980691002562505e-06,
+      "loss": 0.5043,
+      "step": 1590
+    },
+    {
+      "epoch": 1.445624540986385,
+      "grad_norm": 0.3834945559501648,
+      "learning_rate": 7.954625274304822e-06,
+      "loss": 0.5715,
+      "step": 1600
+    },
+    {
+      "epoch": 1.454663578328908,
+      "grad_norm": 0.4076308310031891,
+      "learning_rate": 7.928435575888146e-06,
+      "loss": 0.5898,
+      "step": 1610
+    },
+    {
+      "epoch": 1.463702615671431,
+      "grad_norm": 0.3430909514427185,
+      "learning_rate": 7.902123006179767e-06,
+      "loss": 0.5598,
+      "step": 1620
+    },
+    {
+      "epoch": 1.472741653013954,
+      "grad_norm": 0.4031376242637634,
+      "learning_rate": 7.875688669202416e-06,
+      "loss": 0.5884,
+      "step": 1630
+    },
+    {
+      "epoch": 1.481780690356477,
+      "grad_norm": 0.44467347860336304,
+      "learning_rate": 7.849133674087926e-06,
+      "loss": 0.5741,
+      "step": 1640
+    },
+    {
+      "epoch": 1.490819727699,
+      "grad_norm": 0.37968936562538147,
+      "learning_rate": 7.822459135030709e-06,
+      "loss": 0.5536,
+      "step": 1650
+    },
+    {
+      "epoch": 1.4998587650415232,
+      "grad_norm": 0.5008074045181274,
+      "learning_rate": 7.795666171241001e-06,
+      "loss": 0.5338,
+      "step": 1660
+    },
+    {
+      "epoch": 1.508897802384046,
+      "grad_norm": 0.37119391560554504,
+      "learning_rate": 7.7687559068979e-06,
+      "loss": 0.5386,
+      "step": 1670
+    },
+    {
+      "epoch": 1.5179368397265691,
+      "grad_norm": 0.4206055998802185,
+      "learning_rate": 7.741729471102204e-06,
+      "loss": 0.54,
+      "step": 1680
+    },
+    {
+      "epoch": 1.5269758770690922,
+      "grad_norm": 0.4098663330078125,
+      "learning_rate": 7.71458799782903e-06,
+      "loss": 0.5443,
+      "step": 1690
+    },
+    {
+      "epoch": 1.536014914411615,
+      "grad_norm": 0.41614359617233276,
+      "learning_rate": 7.687332625880237e-06,
+      "loss": 0.5714,
+      "step": 1700
+    },
+    {
+      "epoch": 1.5450539517541382,
+      "grad_norm": 0.429938942193985,
+      "learning_rate": 7.659964498836645e-06,
+      "loss": 0.5209,
+      "step": 1710
+    },
+    {
+      "epoch": 1.5540929890966613,
+      "grad_norm": 0.3705986440181732,
+      "learning_rate": 7.632484765010052e-06,
+      "loss": 0.5164,
+      "step": 1720
+    },
+    {
+      "epoch": 1.5631320264391841,
+      "grad_norm": 0.3999808728694916,
+      "learning_rate": 7.6048945773950536e-06,
+      "loss": 0.5602,
+      "step": 1730
+    },
+    {
+      "epoch": 1.5721710637817072,
+      "grad_norm": 0.4704653322696686,
+      "learning_rate": 7.577195093620664e-06,
+      "loss": 0.5639,
+      "step": 1740
+    },
+    {
+      "epoch": 1.5812101011242303,
+      "grad_norm": 0.40082478523254395,
+      "learning_rate": 7.549387475901746e-06,
+      "loss": 0.5766,
+      "step": 1750
+    },
+    {
+      "epoch": 1.5902491384667532,
+      "grad_norm": 0.4126080274581909,
+      "learning_rate": 7.521472890990247e-06,
+      "loss": 0.5779,
+      "step": 1760
+    },
+    {
+      "epoch": 1.5992881758092763,
+      "grad_norm": 0.39301300048828125,
+      "learning_rate": 7.493452510126241e-06,
+      "loss": 0.5263,
+      "step": 1770
+    },
+    {
+      "epoch": 1.6083272131517994,
+      "grad_norm": 0.4135958254337311,
+      "learning_rate": 7.465327508988791e-06,
+      "loss": 0.5831,
+      "step": 1780
+    },
+    {
+      "epoch": 1.6173662504943223,
+      "grad_norm": 0.4341680109500885,
+      "learning_rate": 7.437099067646616e-06,
+      "loss": 0.5506,
+      "step": 1790
+    },
+    {
+      "epoch": 1.6264052878368453,
+      "grad_norm": 0.43189555406570435,
+      "learning_rate": 7.408768370508577e-06,
+      "loss": 0.5403,
+      "step": 1800
+    },
+    {
+      "epoch": 1.6354443251793684,
+      "grad_norm": 0.3924044966697693,
+      "learning_rate": 7.3803366062739875e-06,
+      "loss": 0.5378,
+      "step": 1810
+    },
+    {
+      "epoch": 1.6444833625218913,
+      "grad_norm": 0.44657233357429504,
+      "learning_rate": 7.3518049678827316e-06,
+      "loss": 0.5927,
+      "step": 1820
+    },
+    {
+      "epoch": 1.6535223998644144,
+      "grad_norm": 0.3682084083557129,
+      "learning_rate": 7.323174652465213e-06,
+      "loss": 0.5395,
+      "step": 1830
+    },
+    {
+      "epoch": 1.6625614372069375,
+      "grad_norm": 0.4883032739162445,
+      "learning_rate": 7.2944468612921235e-06,
+      "loss": 0.565,
+      "step": 1840
+    },
+    {
+      "epoch": 1.6716004745494604,
+      "grad_norm": 0.47772058844566345,
+      "learning_rate": 7.2656227997240494e-06,
+      "loss": 0.5589,
+      "step": 1850
+    },
+    {
+      "epoch": 1.6806395118919835,
+      "grad_norm": 0.419160932302475,
+      "learning_rate": 7.236703677160881e-06,
+      "loss": 0.5703,
+      "step": 1860
+    },
+    {
+      "epoch": 1.6896785492345066,
+      "grad_norm": 0.4483725428581238,
+      "learning_rate": 7.207690706991084e-06,
+      "loss": 0.5528,
+      "step": 1870
+    },
+    {
+      "epoch": 1.6987175865770294,
+      "grad_norm": 0.36597567796707153,
+      "learning_rate": 7.178585106540778e-06,
+      "loss": 0.5522,
+      "step": 1880
+    },
+    {
+      "epoch": 1.7077566239195525,
+      "grad_norm": 0.39857447147369385,
+      "learning_rate": 7.149388097022668e-06,
+      "loss": 0.5597,
+      "step": 1890
+    },
+    {
+      "epoch": 1.7167956612620756,
+      "grad_norm": 0.4677294194698334,
+      "learning_rate": 7.120100903484797e-06,
+      "loss": 0.5495,
+      "step": 1900
+    },
+    {
+      "epoch": 1.7258346986045985,
+      "grad_norm": 0.3514731824398041,
+      "learning_rate": 7.090724754759146e-06,
+      "loss": 0.5952,
+      "step": 1910
+    },
+    {
+      "epoch": 1.7348737359471218,
+      "grad_norm": 0.44050273299217224,
+      "learning_rate": 7.061260883410086e-06,
+      "loss": 0.5154,
+      "step": 1920
+    },
+    {
+      "epoch": 1.7439127732896447,
+      "grad_norm": 0.4586310386657715,
+      "learning_rate": 7.031710525682645e-06,
+      "loss": 0.5421,
+      "step": 1930
+    },
+    {
+      "epoch": 1.7529518106321675,
+      "grad_norm": 0.4656844139099121,
+      "learning_rate": 7.002074921450652e-06,
+      "loss": 0.5591,
+      "step": 1940
+    },
+    {
+      "epoch": 1.7619908479746909,
+      "grad_norm": 0.4101289212703705,
+      "learning_rate": 6.9723553141647045e-06,
+      "loss": 0.5589,
+      "step": 1950
+    },
+    {
+      "epoch": 1.7710298853172137,
+      "grad_norm": 0.4236315190792084,
+      "learning_rate": 6.942552950800003e-06,
+      "loss": 0.5853,
+      "step": 1960
+    },
+    {
+      "epoch": 1.7800689226597366,
+      "grad_norm": 0.3590008318424225,
+      "learning_rate": 6.912669081804027e-06,
+      "loss": 0.5627,
+      "step": 1970
+    },
+    {
+      "epoch": 1.78910796000226,
+      "grad_norm": 0.3790905177593231,
+      "learning_rate": 6.8827049610440685e-06,
+      "loss": 0.555,
+      "step": 1980
+    },
+    {
+      "epoch": 1.7981469973447828,
+      "grad_norm": 0.436749130487442,
+      "learning_rate": 6.85266184575462e-06,
+      "loss": 0.5676,
+      "step": 1990
+    },
+    {
+      "epoch": 1.8071860346873057,
+      "grad_norm": 0.455223023891449,
+      "learning_rate": 6.822540996484631e-06,
+      "loss": 0.5975,
+      "step": 2000
+    },
+    {
+      "epoch": 1.816225072029829,
+      "grad_norm": 0.4379204511642456,
+      "learning_rate": 6.792343677044611e-06,
+      "loss": 0.5476,
+      "step": 2010
+    },
+    {
+      "epoch": 1.8252641093723518,
+      "grad_norm": 0.36188870668411255,
+      "learning_rate": 6.7620711544536024e-06,
+      "loss": 0.5546,
+      "step": 2020
+    },
+    {
+      "epoch": 1.8343031467148747,
+      "grad_norm": 0.37291303277015686,
+      "learning_rate": 6.731724698886023e-06,
+      "loss": 0.5642,
+      "step": 2030
+    },
+    {
+      "epoch": 1.843342184057398,
+      "grad_norm": 0.5276045203208923,
+      "learning_rate": 6.701305583618374e-06,
+      "loss": 0.5251,
+      "step": 2040
+    },
+    {
+      "epoch": 1.852381221399921,
+      "grad_norm": 0.320058673620224,
+      "learning_rate": 6.670815084975808e-06,
+      "loss": 0.5369,
+      "step": 2050
+    },
+    {
+      "epoch": 1.8614202587424438,
+      "grad_norm": 0.4910373091697693,
+      "learning_rate": 6.640254482278581e-06,
+      "loss": 0.54,
+      "step": 2060
+    },
+    {
+      "epoch": 1.870459296084967,
+      "grad_norm": 0.43907907605171204,
+      "learning_rate": 6.6096250577883725e-06,
+      "loss": 0.5842,
+      "step": 2070
+    },
+    {
+      "epoch": 1.87949833342749,
+      "grad_norm": 0.4701830744743347,
+      "learning_rate": 6.578928096654494e-06,
+      "loss": 0.5584,
+      "step": 2080
+    },
+    {
+      "epoch": 1.888537370770013,
+      "grad_norm": 0.4494197964668274,
+      "learning_rate": 6.548164886859955e-06,
+      "loss": 0.5499,
+      "step": 2090
+    },
+    {
+      "epoch": 1.8975764081125361,
+      "grad_norm": 0.4171752631664276,
+      "learning_rate": 6.5173367191674205e-06,
+      "loss": 0.5439,
+      "step": 2100
+    },
+    {
+      "epoch": 1.906615445455059,
+      "grad_norm": 0.4559502899646759,
+      "learning_rate": 6.48644488706507e-06,
+      "loss": 0.5512,
+      "step": 2110
+    },
+    {
+      "epoch": 1.915654482797582,
+      "grad_norm": 0.4155573546886444,
+      "learning_rate": 6.455490686712307e-06,
+      "loss": 0.5816,
+      "step": 2120
+    },
+    {
+      "epoch": 1.9246935201401052,
+      "grad_norm": 0.3807109594345093,
+      "learning_rate": 6.424475416885382e-06,
+      "loss": 0.5405,
+      "step": 2130
+    },
+    {
+      "epoch": 1.933732557482628,
+      "grad_norm": 0.4987061619758606,
+      "learning_rate": 6.393400378922901e-06,
+      "loss": 0.5548,
+      "step": 2140
+    },
+    {
+      "epoch": 1.9427715948251512,
+      "grad_norm": 0.47996246814727783,
+      "learning_rate": 6.362266876671217e-06,
+      "loss": 0.5279,
+      "step": 2150
+    },
+    {
+      "epoch": 1.9518106321676743,
+      "grad_norm": 0.4458663761615753,
+      "learning_rate": 6.331076216429735e-06,
+      "loss": 0.5769,
+      "step": 2160
+    },
+    {
+      "epoch": 1.9608496695101971,
+      "grad_norm": 0.4673076570034027,
+      "learning_rate": 6.299829706896088e-06,
+      "loss": 0.5679,
+      "step": 2170
+    },
+    {
+      "epoch": 1.9698887068527202,
+      "grad_norm": 0.5307784676551819,
+      "learning_rate": 6.268528659111234e-06,
+      "loss": 0.5807,
+      "step": 2180
+    },
+    {
+      "epoch": 1.9789277441952433,
+      "grad_norm": 0.36635711789131165,
+      "learning_rate": 6.237174386404448e-06,
+      "loss": 0.5632,
+      "step": 2190
+    },
+    {
+      "epoch": 1.9879667815377662,
+      "grad_norm": 0.46293821930885315,
+      "learning_rate": 6.205768204338221e-06,
+      "loss": 0.5731,
+      "step": 2200
+    },
+    {
+      "epoch": 1.9970058188802893,
+      "grad_norm": 0.3912748694419861,
+      "learning_rate": 6.17431143065305e-06,
+      "loss": 0.5652,
+      "step": 2210
+    },
+    {
+      "epoch": 2.0054234224055136,
+      "grad_norm": 0.4155929386615753,
+      "learning_rate": 6.142805385212156e-06,
+      "loss": 0.5301,
+      "step": 2220
+    },
+    {
+      "epoch": 2.014462459748037,
+      "grad_norm": 0.5124784111976624,
+      "learning_rate": 6.111251389946105e-06,
+      "loss": 0.5091,
+      "step": 2230
+    },
+    {
+      "epoch": 2.02350149709056,
+      "grad_norm": 0.49529412388801575,
+      "learning_rate": 6.079650768797346e-06,
+      "loss": 0.5545,
+      "step": 2240
+    },
+    {
+      "epoch": 2.0325405344330827,
+      "grad_norm": 0.5367194414138794,
+      "learning_rate": 6.048004847664649e-06,
+      "loss": 0.5467,
+      "step": 2250
+    },
+    {
+      "epoch": 2.041579571775606,
+      "grad_norm": 0.406827449798584,
+      "learning_rate": 6.016314954347484e-06,
+      "loss": 0.5464,
+      "step": 2260
+    },
+    {
+      "epoch": 2.050618609118129,
+      "grad_norm": 0.49716565012931824,
+      "learning_rate": 5.984582418490309e-06,
+      "loss": 0.5598,
+      "step": 2270
+    },
+    {
+      "epoch": 2.0596576464606517,
+      "grad_norm": 0.7402800917625427,
+      "learning_rate": 5.952808571526775e-06,
+      "loss": 0.5291,
+      "step": 2280
+    },
+    {
+      "epoch": 2.068696683803175,
+      "grad_norm": 0.360857218503952,
+      "learning_rate": 5.92099474662386e-06,
+      "loss": 0.5452,
+      "step": 2290
+    },
+    {
+      "epoch": 2.077735721145698,
+      "grad_norm": 0.4300866425037384,
+      "learning_rate": 5.889142278625943e-06,
+      "loss": 0.5743,
+      "step": 2300
+    },
+    {
+      "epoch": 2.086774758488221,
+      "grad_norm": 0.5264549851417542,
+      "learning_rate": 5.857252503998788e-06,
+      "loss": 0.5354,
+      "step": 2310
+    },
+    {
+      "epoch": 2.095813795830744,
+      "grad_norm": 0.4711679220199585,
+      "learning_rate": 5.825326760773469e-06,
+      "loss": 0.5225,
+      "step": 2320
+    },
+    {
+      "epoch": 2.104852833173267,
+      "grad_norm": 0.3973844349384308,
+      "learning_rate": 5.793366388490228e-06,
+      "loss": 0.566,
+      "step": 2330
+    },
+    {
+      "epoch": 2.11389187051579,
+      "grad_norm": 0.4139045774936676,
+      "learning_rate": 5.7613727281422784e-06,
+      "loss": 0.573,
+      "step": 2340
+    },
+    {
+      "epoch": 2.122930907858313,
+      "grad_norm": 0.48233065009117126,
+      "learning_rate": 5.729347122119533e-06,
+      "loss": 0.5191,
+      "step": 2350
+    },
+    {
+      "epoch": 2.131969945200836,
+      "grad_norm": 0.4828956425189972,
+      "learning_rate": 5.697290914152277e-06,
+      "loss": 0.5471,
+      "step": 2360
+    },
+    {
+      "epoch": 2.141008982543359,
+      "grad_norm": 0.5164515376091003,
+      "learning_rate": 5.665205449254797e-06,
+      "loss": 0.5526,
+      "step": 2370
+    },
+    {
+      "epoch": 2.1500480198858822,
+      "grad_norm": 0.410300076007843,
+      "learning_rate": 5.6330920736689365e-06,
+      "loss": 0.5362,
+      "step": 2380
+    },
+    {
+      "epoch": 2.159087057228405,
+      "grad_norm": 0.41370904445648193,
+      "learning_rate": 5.600952134807621e-06,
+      "loss": 0.5238,
+      "step": 2390
+    },
+    {
+      "epoch": 2.168126094570928,
+      "grad_norm": 0.4590449333190918,
+      "learning_rate": 5.568786981198315e-06,
+      "loss": 0.541,
+      "step": 2400
+    },
+    {
+      "epoch": 2.1771651319134513,
+      "grad_norm": 0.43501394987106323,
+      "learning_rate": 5.536597962426441e-06,
+      "loss": 0.5608,
+      "step": 2410
+    },
+    {
+      "epoch": 2.186204169255974,
+      "grad_norm": 0.42458391189575195,
+      "learning_rate": 5.504386429078762e-06,
+      "loss": 0.5378,
+      "step": 2420
+    },
+    {
+      "epoch": 2.195243206598497,
+      "grad_norm": 0.5655559301376343,
+      "learning_rate": 5.4721537326867025e-06,
+      "loss": 0.5568,
+      "step": 2430
+    },
+    {
+      "epoch": 2.2042822439410203,
+      "grad_norm": 0.42385461926460266,
+      "learning_rate": 5.439901225669649e-06,
+      "loss": 0.5929,
+      "step": 2440
+    },
+    {
+      "epoch": 2.213321281283543,
+      "grad_norm": 0.4631207585334778,
+      "learning_rate": 5.407630261278197e-06,
+      "loss": 0.6009,
+      "step": 2450
+    },
+    {
+      "epoch": 2.222360318626066,
+      "grad_norm": 0.42976030707359314,
+      "learning_rate": 5.375342193537385e-06,
+      "loss": 0.5515,
+      "step": 2460
+    },
+    {
+      "epoch": 2.2313993559685894,
+      "grad_norm": 0.5118324160575867,
+      "learning_rate": 5.343038377189867e-06,
+      "loss": 0.5943,
+      "step": 2470
+    },
+    {
+      "epoch": 2.2404383933111123,
+      "grad_norm": 0.40487727522850037,
+      "learning_rate": 5.310720167639076e-06,
+      "loss": 0.5499,
+      "step": 2480
+    },
+    {
+      "epoch": 2.2494774306536356,
+      "grad_norm": 0.4322616159915924,
+      "learning_rate": 5.2783889208923625e-06,
+      "loss": 0.5327,
+      "step": 2490
+    },
+    {
+      "epoch": 2.2585164679961585,
+      "grad_norm": 0.5547094941139221,
+      "learning_rate": 5.246045993504083e-06,
+      "loss": 0.5524,
+      "step": 2500
+    },
+    {
+      "epoch": 2.2675555053386813,
+      "grad_norm": 0.5066604018211365,
+      "learning_rate": 5.213692742518699e-06,
+      "loss": 0.5559,
+      "step": 2510
+    },
+    {
+      "epoch": 2.276594542681204,
+      "grad_norm": 0.4938390851020813,
+      "learning_rate": 5.181330525413822e-06,
+      "loss": 0.5323,
+      "step": 2520
+    },
+    {
+      "epoch": 2.2856335800237275,
+      "grad_norm": 0.5223654508590698,
+      "learning_rate": 5.148960700043264e-06,
+      "loss": 0.5615,
+      "step": 2530
+    },
+    {
+      "epoch": 2.2946726173662504,
+      "grad_norm": 0.4902248978614807,
+      "learning_rate": 5.116584624580072e-06,
+      "loss": 0.5367,
+      "step": 2540
+    },
+    {
+      "epoch": 2.3037116547087737,
+      "grad_norm": 0.5171300172805786,
+      "learning_rate": 5.0842036574595285e-06,
+      "loss": 0.5222,
+      "step": 2550
+    },
+    {
+      "epoch": 2.3127506920512966,
+      "grad_norm": 0.5555601119995117,
+      "learning_rate": 5.051819157322161e-06,
+      "loss": 0.5305,
+      "step": 2560
+    },
+    {
+      "epoch": 2.3217897293938194,
+      "grad_norm": 0.4432179629802704,
+      "learning_rate": 5.019432482956734e-06,
+      "loss": 0.5452,
+      "step": 2570
+    },
+    {
+      "epoch": 2.3308287667363428,
+      "grad_norm": 0.5064467787742615,
+      "learning_rate": 4.987044993243244e-06,
+      "loss": 0.5854,
+      "step": 2580
+    },
+    {
+      "epoch": 2.3398678040788656,
+      "grad_norm": 0.5878727436065674,
+      "learning_rate": 4.95465804709589e-06,
+      "loss": 0.5305,
+      "step": 2590
+    },
+    {
+      "epoch": 2.3489068414213885,
+      "grad_norm": 0.5292733311653137,
+      "learning_rate": 4.9222730034060675e-06,
+      "loss": 0.531,
+      "step": 2600
+    },
+    {
+      "epoch": 2.357945878763912,
+      "grad_norm": 0.5251352190971375,
+      "learning_rate": 4.889891220985351e-06,
+      "loss": 0.5485,
+      "step": 2610
+    },
+    {
+      "epoch": 2.3669849161064347,
+      "grad_norm": 0.5175120234489441,
+      "learning_rate": 4.857514058508478e-06,
+      "loss": 0.5639,
+      "step": 2620
+    },
+    {
+      "epoch": 2.3760239534489576,
+      "grad_norm": 0.509757399559021,
+      "learning_rate": 4.825142874456339e-06,
+      "loss": 0.5301,
+      "step": 2630
+    },
+    {
+      "epoch": 2.385062990791481,
+      "grad_norm": 0.5946202874183655,
+      "learning_rate": 4.792779027058985e-06,
+      "loss": 0.5803,
+      "step": 2640
+    },
+    {
+      "epoch": 2.3941020281340037,
+      "grad_norm": 0.5153007507324219,
+      "learning_rate": 4.760423874238633e-06,
+      "loss": 0.5415,
+      "step": 2650
+    },
+    {
+      "epoch": 2.4031410654765266,
+      "grad_norm": 0.38492918014526367,
+      "learning_rate": 4.728078773552695e-06,
+      "loss": 0.529,
+      "step": 2660
+    },
+    {
+      "epoch": 2.41218010281905,
+      "grad_norm": 0.43594977259635925,
+      "learning_rate": 4.6957450821368175e-06,
+      "loss": 0.5374,
+      "step": 2670
+    },
+    {
+      "epoch": 2.421219140161573,
+      "grad_norm": 0.5468082427978516,
+      "learning_rate": 4.66342415664793e-06,
+      "loss": 0.5973,
+      "step": 2680
+    },
+    {
+      "epoch": 2.4302581775040957,
+      "grad_norm": 0.43409329652786255,
+      "learning_rate": 4.631117353207339e-06,
+      "loss": 0.5392,
+      "step": 2690
+    },
+    {
+      "epoch": 2.439297214846619,
+      "grad_norm": 0.4601250886917114,
+      "learning_rate": 4.598826027343809e-06,
+      "loss": 0.5624,
+      "step": 2700
+    },
+    {
+      "epoch": 2.448336252189142,
+      "grad_norm": 0.38829344511032104,
+      "learning_rate": 4.5665515339367014e-06,
+      "loss": 0.5317,
+      "step": 2710
+    },
+    {
+      "epoch": 2.4573752895316647,
+      "grad_norm": 0.48692500591278076,
+      "learning_rate": 4.53429522715912e-06,
+      "loss": 0.5487,
+      "step": 2720
+    },
+    {
+      "epoch": 2.466414326874188,
+      "grad_norm": 0.42683878540992737,
+      "learning_rate": 4.502058460421094e-06,
+      "loss": 0.5496,
+      "step": 2730
+    },
+    {
+      "epoch": 2.475453364216711,
+      "grad_norm": 0.5086000561714172,
+      "learning_rate": 4.469842586312794e-06,
+      "loss": 0.5314,
+      "step": 2740
+    },
+    {
+      "epoch": 2.484492401559234,
+      "grad_norm": 0.4432428777217865,
+      "learning_rate": 4.437648956547772e-06,
+      "loss": 0.5344,
+      "step": 2750
+    },
+    {
+      "epoch": 2.493531438901757,
+      "grad_norm": 0.5344327688217163,
+      "learning_rate": 4.40547892190626e-06,
+      "loss": 0.5155,
+      "step": 2760
+    },
+    {
+      "epoch": 2.50257047624428,
+      "grad_norm": 0.4258926808834076,
+      "learning_rate": 4.373333832178478e-06,
+      "loss": 0.5452,
+      "step": 2770
+    },
+    {
+      "epoch": 2.511609513586803,
+      "grad_norm": 0.45202168822288513,
+      "learning_rate": 4.3412150361080144e-06,
+      "loss": 0.5504,
+      "step": 2780
+    },
+    {
+      "epoch": 2.520648550929326,
+      "grad_norm": 0.5143294930458069,
+      "learning_rate": 4.309123881335222e-06,
+      "loss": 0.5682,
+      "step": 2790
+    },
+    {
+      "epoch": 2.529687588271849,
+      "grad_norm": 0.43196791410446167,
+      "learning_rate": 4.277061714340683e-06,
+      "loss": 0.5193,
+      "step": 2800
+    },
+    {
+      "epoch": 2.5387266256143723,
+      "grad_norm": 0.46495088934898376,
+      "learning_rate": 4.245029880388713e-06,
+      "loss": 0.5151,
+      "step": 2810
+    },
+    {
+      "epoch": 2.547765662956895,
+      "grad_norm": 0.41115784645080566,
+      "learning_rate": 4.2130297234709085e-06,
+      "loss": 0.4851,
+      "step": 2820
+    },
+    {
+      "epoch": 2.556804700299418,
+      "grad_norm": 0.5914024710655212,
+      "learning_rate": 4.181062586249765e-06,
+      "loss": 0.4954,
+      "step": 2830
+    },
+    {
+      "epoch": 2.5658437376419414,
+      "grad_norm": 0.46192359924316406,
+      "learning_rate": 4.149129810002335e-06,
+      "loss": 0.5406,
+      "step": 2840
+    },
+    {
+      "epoch": 2.5748827749844643,
+      "grad_norm": 0.409563809633255,
+      "learning_rate": 4.117232734563959e-06,
+      "loss": 0.5722,
+      "step": 2850
+    },
+    {
+      "epoch": 2.583921812326987,
+      "grad_norm": 0.5417585968971252,
+      "learning_rate": 4.085372698272036e-06,
+      "loss": 0.5412,
+      "step": 2860
+    },
+    {
+      "epoch": 2.5929608496695105,
+      "grad_norm": 0.5342581272125244,
+      "learning_rate": 4.05355103790988e-06,
+      "loss": 0.5368,
+      "step": 2870
+    },
+    {
+      "epoch": 2.6019998870120333,
+      "grad_norm": 0.5177707076072693,
+      "learning_rate": 4.021769088650628e-06,
+      "loss": 0.541,
+      "step": 2880
+    },
+    {
+      "epoch": 2.611038924354556,
+      "grad_norm": 0.5150489807128906,
+      "learning_rate": 3.990028184001218e-06,
+      "loss": 0.5214,
+      "step": 2890
+    },
+    {
+      "epoch": 2.6200779616970795,
+      "grad_norm": 0.5407333970069885,
+      "learning_rate": 3.9583296557464355e-06,
+      "loss": 0.5229,
+      "step": 2900
+    },
+    {
+      "epoch": 2.6291169990396024,
+      "grad_norm": 0.41024380922317505,
+      "learning_rate": 3.926674833893045e-06,
+      "loss": 0.5145,
+      "step": 2910
+    },
+    {
+      "epoch": 2.6381560363821253,
+      "grad_norm": 0.5214430689811707,
+      "learning_rate": 3.895065046613972e-06,
+      "loss": 0.5464,
+      "step": 2920
+    },
+    {
+      "epoch": 2.6471950737246486,
+      "grad_norm": 0.43542003631591797,
+      "learning_rate": 3.863501620192581e-06,
+      "loss": 0.5492,
+      "step": 2930
+    },
+    {
+      "epoch": 2.6562341110671714,
+      "grad_norm": 0.5365663766860962,
+      "learning_rate": 3.831985878967032e-06,
+      "loss": 0.539,
+      "step": 2940
+    },
+    {
+      "epoch": 2.6652731484096943,
+      "grad_norm": 0.5596737861633301,
+      "learning_rate": 3.800519145274705e-06,
+      "loss": 0.5185,
+      "step": 2950
+    },
+    {
+      "epoch": 2.6743121857522176,
+      "grad_norm": 0.5031512379646301,
+      "learning_rate": 3.769102739396724e-06,
+      "loss": 0.5608,
+      "step": 2960
+    },
+    {
+      "epoch": 2.6833512230947405,
+      "grad_norm": 0.4453997313976288,
+      "learning_rate": 3.7377379795025616e-06,
+      "loss": 0.5217,
+      "step": 2970
+    },
+    {
+      "epoch": 2.6923902604372634,
+      "grad_norm": 0.5238685011863708,
+      "learning_rate": 3.7064261815947232e-06,
+      "loss": 0.562,
+      "step": 2980
+    },
+    {
+      "epoch": 2.7014292977797867,
+      "grad_norm": 0.5626428723335266,
+      "learning_rate": 3.6751686594535373e-06,
+      "loss": 0.5179,
+      "step": 2990
+    },
+    {
+      "epoch": 2.7104683351223096,
+      "grad_norm": 0.4801914691925049,
+      "learning_rate": 3.643966724582034e-06,
+      "loss": 0.5322,
+      "step": 3000
+    },
+    {
+      "epoch": 2.7195073724648324,
+      "grad_norm": 0.4691397547721863,
+      "learning_rate": 3.612821686150906e-06,
+      "loss": 0.5352,
+      "step": 3010
+    },
+    {
+      "epoch": 2.7285464098073557,
+      "grad_norm": 0.39714473485946655,
+      "learning_rate": 3.581734850943591e-06,
+      "loss": 0.5515,
+      "step": 3020
+    },
+    {
+      "epoch": 2.7375854471498786,
+      "grad_norm": 0.508211612701416,
+      "learning_rate": 3.5507075233014303e-06,
+      "loss": 0.4986,
+      "step": 3030
+    },
+    {
+      "epoch": 2.7466244844924015,
+      "grad_norm": 0.451964795589447,
+      "learning_rate": 3.519741005068954e-06,
+      "loss": 0.5519,
+      "step": 3040
+    },
+    {
+      "epoch": 2.755663521834925,
+      "grad_norm": 0.3844453990459442,
+      "learning_rate": 3.4888365955392434e-06,
+      "loss": 0.5099,
+      "step": 3050
+    },
+    {
+      "epoch": 2.7647025591774477,
+      "grad_norm": 0.43304121494293213,
+      "learning_rate": 3.4579955913994323e-06,
+      "loss": 0.5689,
+      "step": 3060
+    },
+    {
+      "epoch": 2.7737415965199705,
+      "grad_norm": 0.4876834750175476,
+      "learning_rate": 3.4272192866762832e-06,
+      "loss": 0.5447,
+      "step": 3070
+    },
+    {
+      "epoch": 2.782780633862494,
+      "grad_norm": 0.5928559303283691,
+      "learning_rate": 3.396508972681902e-06,
+      "loss": 0.5345,
+      "step": 3080
+    },
+    {
+      "epoch": 2.7918196712050167,
+      "grad_norm": 0.5259817242622375,
+      "learning_rate": 3.3658659379595615e-06,
+      "loss": 0.5296,
+      "step": 3090
+    },
+    {
+      "epoch": 2.8008587085475396,
+      "grad_norm": 0.5222415924072266,
+      "learning_rate": 3.3352914682296243e-06,
+      "loss": 0.5723,
+      "step": 3100
+    },
+    {
+      "epoch": 2.809897745890063,
+      "grad_norm": 0.5005621910095215,
+      "learning_rate": 3.3047868463356047e-06,
+      "loss": 0.5577,
+      "step": 3110
+    },
+    {
+      "epoch": 2.818936783232586,
+      "grad_norm": 0.48413482308387756,
+      "learning_rate": 3.274353352190345e-06,
+      "loss": 0.5125,
+      "step": 3120
+    },
+    {
+      "epoch": 2.8279758205751087,
+      "grad_norm": 0.6147083044052124,
+      "learning_rate": 3.2439922627223064e-06,
+      "loss": 0.5494,
+      "step": 3130
+    },
+    {
+      "epoch": 2.837014857917632,
+      "grad_norm": 0.5478202104568481,
+      "learning_rate": 3.213704851821995e-06,
+      "loss": 0.5259,
+      "step": 3140
+    },
+    {
+      "epoch": 2.846053895260155,
+      "grad_norm": 0.5824897289276123,
+      "learning_rate": 3.183492390288515e-06,
+      "loss": 0.5246,
+      "step": 3150
+    },
+    {
+      "epoch": 2.8550929326026777,
+      "grad_norm": 0.5393964648246765,
+      "learning_rate": 3.153356145776242e-06,
+      "loss": 0.5153,
+      "step": 3160
+    },
+    {
+      "epoch": 2.864131969945201,
+      "grad_norm": 0.5183600783348083,
+      "learning_rate": 3.123297382741637e-06,
+      "loss": 0.5306,
+      "step": 3170
+    },
+    {
+      "epoch": 2.873171007287724,
+      "grad_norm": 0.4243277907371521,
+      "learning_rate": 3.093317362390197e-06,
+      "loss": 0.5245,
+      "step": 3180
+    },
+    {
+      "epoch": 2.8822100446302468,
+      "grad_norm": 0.4832552671432495,
+      "learning_rate": 3.0634173426235325e-06,
+      "loss": 0.5658,
+      "step": 3190
+    },
+    {
+      "epoch": 2.89124908197277,
+      "grad_norm": 0.45082056522369385,
+      "learning_rate": 3.033598577986586e-06,
+      "loss": 0.5257,
+      "step": 3200
+    },
+    {
+      "epoch": 2.900288119315293,
+      "grad_norm": 0.5070288181304932,
+      "learning_rate": 3.003862319615005e-06,
+      "loss": 0.5207,
+      "step": 3210
+    },
+    {
+      "epoch": 2.909327156657816,
+      "grad_norm": 0.588970422744751,
+      "learning_rate": 2.9742098151826347e-06,
+      "loss": 0.5255,
+      "step": 3220
+    },
+    {
+      "epoch": 2.918366194000339,
+      "grad_norm": 0.47716692090034485,
+      "learning_rate": 2.9446423088491695e-06,
+      "loss": 0.5754,
+      "step": 3230
+    },
+    {
+      "epoch": 2.927405231342862,
+      "grad_norm": 0.48788127303123474,
+      "learning_rate": 2.915161041207961e-06,
+      "loss": 0.5193,
+      "step": 3240
+    },
+    {
+      "epoch": 2.936444268685385,
+      "grad_norm": 0.4768235683441162,
+      "learning_rate": 2.885767249233954e-06,
+      "loss": 0.5149,
+      "step": 3250
+    },
+    {
+      "epoch": 2.945483306027908,
+      "grad_norm": 0.544245183467865,
+      "learning_rate": 2.8564621662317875e-06,
+      "loss": 0.5454,
+      "step": 3260
+    },
+    {
+      "epoch": 2.954522343370431,
+      "grad_norm": 0.544935941696167,
+      "learning_rate": 2.8272470217840544e-06,
+      "loss": 0.5235,
+      "step": 3270
+    },
+    {
+      "epoch": 2.963561380712954,
+      "grad_norm": 0.5481933951377869,
+      "learning_rate": 2.798123041699701e-06,
+      "loss": 0.5611,
+      "step": 3280
+    },
+    {
+      "epoch": 2.9726004180554773,
+      "grad_norm": 0.5460256338119507,
+      "learning_rate": 2.769091447962606e-06,
+      "loss": 0.5661,
+      "step": 3290
+    },
+    {
+      "epoch": 2.981639455398,
+      "grad_norm": 0.5495103001594543,
+      "learning_rate": 2.7401534586802958e-06,
+      "loss": 0.5294,
+      "step": 3300
+    },
+    {
+      "epoch": 2.990678492740523,
+      "grad_norm": 0.42724740505218506,
+      "learning_rate": 2.7113102880328444e-06,
+      "loss": 0.5549,
+      "step": 3310
+    },
+    {
+      "epoch": 2.9997175300830463,
+      "grad_norm": 0.507023274898529,
+      "learning_rate": 2.682563146221926e-06,
+      "loss": 0.5328,
+      "step": 3320
+    },
+    {
+      "epoch": 3.008135133608271,
+      "grad_norm": 0.5757578015327454,
+      "learning_rate": 2.653913239420034e-06,
+      "loss": 0.496,
+      "step": 3330
+    },
+    {
+      "epoch": 3.0171741709507938,
+      "grad_norm": 0.5573206543922424,
+      "learning_rate": 2.6253617697198798e-06,
+      "loss": 0.5132,
+      "step": 3340
+    },
+    {
+      "epoch": 3.0262132082933166,
+      "grad_norm": 0.49787354469299316,
+      "learning_rate": 2.5969099350839467e-06,
+      "loss": 0.5238,
+      "step": 3350
+    },
+    {
+      "epoch": 3.03525224563584,
+      "grad_norm": 0.494858056306839,
+      "learning_rate": 2.568558929294236e-06,
+      "loss": 0.5221,
+      "step": 3360
+    },
+    {
+      "epoch": 3.044291282978363,
+      "grad_norm": 0.6178960800170898,
+      "learning_rate": 2.5403099419021663e-06,
+      "loss": 0.5875,
+      "step": 3370
+    },
+    {
+      "epoch": 3.0533303203208857,
+      "grad_norm": 0.6497880220413208,
+      "learning_rate": 2.512164158178675e-06,
+      "loss": 0.5037,
+      "step": 3380
+    },
+    {
+      "epoch": 3.062369357663409,
+      "grad_norm": 0.4282669126987457,
+      "learning_rate": 2.484122759064475e-06,
+      "loss": 0.568,
+      "step": 3390
+    },
+    {
+      "epoch": 3.071408395005932,
+      "grad_norm": 0.655299961566925,
+      "learning_rate": 2.4561869211205126e-06,
+      "loss": 0.4809,
+      "step": 3400
+    },
+    {
+      "epoch": 3.0804474323484548,
+      "grad_norm": 0.4744868576526642,
+      "learning_rate": 2.4283578164785967e-06,
+      "loss": 0.5336,
+      "step": 3410
+    },
+    {
+      "epoch": 3.089486469690978,
+      "grad_norm": 0.7016565203666687,
+      "learning_rate": 2.4006366127922225e-06,
+      "loss": 0.5241,
+      "step": 3420
+    },
+    {
+      "epoch": 3.098525507033501,
+      "grad_norm": 0.5095173120498657,
+      "learning_rate": 2.373024473187575e-06,
+      "loss": 0.5674,
+      "step": 3430
+    },
+    {
+      "epoch": 3.107564544376024,
+      "grad_norm": 0.5802739262580872,
+      "learning_rate": 2.345522556214735e-06,
+      "loss": 0.5556,
+      "step": 3440
+    },
+    {
+      "epoch": 3.116603581718547,
+      "grad_norm": 0.5532498359680176,
+      "learning_rate": 2.318132015799055e-06,
+      "loss": 0.5481,
+      "step": 3450
+    },
+    {
+      "epoch": 3.12564261906107,
+      "grad_norm": 0.4916612207889557,
+      "learning_rate": 2.290854001192753e-06,
+      "loss": 0.5088,
+      "step": 3460
+    },
+    {
+      "epoch": 3.134681656403593,
+      "grad_norm": 0.568167507648468,
+      "learning_rate": 2.2636896569266907e-06,
+      "loss": 0.5879,
+      "step": 3470
+    },
+    {
+      "epoch": 3.143720693746116,
+      "grad_norm": 0.510736882686615,
+      "learning_rate": 2.236640122762348e-06,
+      "loss": 0.5488,
+      "step": 3480
+    },
+    {
+      "epoch": 3.152759731088639,
+      "grad_norm": 0.590149462223053,
+      "learning_rate": 2.209706533644003e-06,
+      "loss": 0.5622,
+      "step": 3490
+    },
+    {
+      "epoch": 3.161798768431162,
+      "grad_norm": 0.44575035572052,
+      "learning_rate": 2.1828900196511127e-06,
+      "loss": 0.5368,
+      "step": 3500
+    },
+    {
+      "epoch": 3.1708378057736852,
+      "grad_norm": 0.47225379943847656,
+      "learning_rate": 2.156191705950898e-06,
+      "loss": 0.505,
+      "step": 3510
+    },
+    {
+      "epoch": 3.179876843116208,
+      "grad_norm": 0.44492092728614807,
+      "learning_rate": 2.129612712751128e-06,
+      "loss": 0.5193,
+      "step": 3520
+    },
+    {
+      "epoch": 3.188915880458731,
+      "grad_norm": 0.5899855494499207,
+      "learning_rate": 2.1031541552531276e-06,
+      "loss": 0.5688,
+      "step": 3530
+    },
+    {
+      "epoch": 3.1979549178012543,
+      "grad_norm": 0.5109498500823975,
+      "learning_rate": 2.0768171436049774e-06,
+      "loss": 0.5325,
+      "step": 3540
+    },
+    {
+      "epoch": 3.206993955143777,
+      "grad_norm": 0.5188099145889282,
+      "learning_rate": 2.050602782854938e-06,
+      "loss": 0.5223,
+      "step": 3550
+    },
+    {
+      "epoch": 3.2160329924863,
+      "grad_norm": 0.6767019033432007,
+      "learning_rate": 2.0245121729050843e-06,
+      "loss": 0.5317,
+      "step": 3560
+    },
+    {
+      "epoch": 3.2250720298288234,
+      "grad_norm": 0.4775676727294922,
+      "learning_rate": 1.9985464084651564e-06,
+      "loss": 0.5353,
+      "step": 3570
+    },
+    {
+      "epoch": 3.234111067171346,
+      "grad_norm": 0.40599289536476135,
+      "learning_rate": 1.9727065790066247e-06,
+      "loss": 0.5614,
+      "step": 3580
+    },
+    {
+      "epoch": 3.243150104513869,
+      "grad_norm": 0.5476986765861511,
+      "learning_rate": 1.9469937687169804e-06,
+      "loss": 0.5705,
+      "step": 3590
+    },
+    {
+      "epoch": 3.2521891418563924,
+      "grad_norm": 0.560183584690094,
+      "learning_rate": 1.921409056454249e-06,
+      "loss": 0.5265,
+      "step": 3600
+    },
+    {
+      "epoch": 3.2612281791989153,
+      "grad_norm": 0.4706651270389557,
+      "learning_rate": 1.895953515701715e-06,
+      "loss": 0.5013,
+      "step": 3610
+    },
+    {
+      "epoch": 3.270267216541438,
+      "grad_norm": 0.5632408261299133,
+      "learning_rate": 1.8706282145228855e-06,
+      "loss": 0.5301,
+      "step": 3620
+    },
+    {
+      "epoch": 3.2793062538839615,
+      "grad_norm": 0.5416795015335083,
+      "learning_rate": 1.845434215516676e-06,
+      "loss": 0.5168,
+      "step": 3630
+    },
+    {
+      "epoch": 3.2883452912264843,
+      "grad_norm": 0.4724695086479187,
+      "learning_rate": 1.8203725757728268e-06,
+      "loss": 0.5382,
+      "step": 3640
+    },
+    {
+      "epoch": 3.2973843285690076,
+      "grad_norm": 0.5644287467002869,
+      "learning_rate": 1.7954443468275484e-06,
+      "loss": 0.5359,
+      "step": 3650
+    },
+    {
+      "epoch": 3.3064233659115305,
+      "grad_norm": 0.5141358375549316,
+      "learning_rate": 1.770650574619399e-06,
+      "loss": 0.5127,
+      "step": 3660
+    },
+    {
+      "epoch": 3.3154624032540534,
+      "grad_norm": 0.4922308027744293,
+      "learning_rate": 1.745992299445406e-06,
+      "loss": 0.5277,
+      "step": 3670
+    },
+    {
+      "epoch": 3.3245014405965767,
+      "grad_norm": 0.5326086282730103,
+      "learning_rate": 1.721470555917411e-06,
+      "loss": 0.566,
+      "step": 3680
+    },
+    {
+      "epoch": 3.3335404779390996,
+      "grad_norm": 0.5815884470939636,
+      "learning_rate": 1.6970863729186588e-06,
+      "loss": 0.5228,
+      "step": 3690
+    },
+    {
+      "epoch": 3.3425795152816224,
+      "grad_norm": 0.4859824776649475,
+      "learning_rate": 1.6728407735606317e-06,
+      "loss": 0.5281,
+      "step": 3700
+    },
+    {
+      "epoch": 3.3516185526241458,
+      "grad_norm": 0.49963539838790894,
+      "learning_rate": 1.6487347751401194e-06,
+      "loss": 0.4916,
+      "step": 3710
+    },
+    {
+      "epoch": 3.3606575899666686,
+      "grad_norm": 0.49004191160202026,
+      "learning_rate": 1.624769389096537e-06,
+      "loss": 0.5883,
+      "step": 3720
+    },
+    {
+      "epoch": 3.3696966273091915,
+      "grad_norm": 0.46009162068367004,
+      "learning_rate": 1.6009456209694846e-06,
+      "loss": 0.5545,
+      "step": 3730
+    },
+    {
+      "epoch": 3.378735664651715,
+      "grad_norm": 0.7347127795219421,
+      "learning_rate": 1.5772644703565564e-06,
+      "loss": 0.5151,
+      "step": 3740
+    },
+    {
+      "epoch": 3.3877747019942377,
+      "grad_norm": 0.46795326471328735,
+      "learning_rate": 1.5537269308714075e-06,
+      "loss": 0.5196,
+      "step": 3750
+    },
+    {
+      "epoch": 3.3968137393367606,
+      "grad_norm": 0.5418402552604675,
+      "learning_rate": 1.530333990102052e-06,
+      "loss": 0.5819,
+      "step": 3760
+    },
+    {
+      "epoch": 3.405852776679284,
+      "grad_norm": 0.6440554261207581,
+      "learning_rate": 1.5070866295694336e-06,
+      "loss": 0.5574,
+      "step": 3770
+    },
+    {
+      "epoch": 3.4148918140218067,
+      "grad_norm": 0.526903510093689,
+      "learning_rate": 1.4839858246862398e-06,
+      "loss": 0.5386,
+      "step": 3780
+    },
+    {
+      "epoch": 3.4239308513643296,
+      "grad_norm": 0.5219286680221558,
+      "learning_rate": 1.4610325447159779e-06,
+      "loss": 0.5306,
+      "step": 3790
+    },
+    {
+      "epoch": 3.432969888706853,
+      "grad_norm": 0.5307949185371399,
+      "learning_rate": 1.4382277527323042e-06,
+      "loss": 0.5655,
+      "step": 3800
+    },
+    {
+      "epoch": 3.442008926049376,
+      "grad_norm": 0.5093571543693542,
+      "learning_rate": 1.4155724055786152e-06,
+      "loss": 0.5522,
+      "step": 3810
+    },
+    {
+      "epoch": 3.4510479633918987,
+      "grad_norm": 0.48783156275749207,
+      "learning_rate": 1.393067453827907e-06,
+      "loss": 0.5441,
+      "step": 3820
+    },
+    {
+      "epoch": 3.460087000734422,
+      "grad_norm": 0.654633641242981,
+      "learning_rate": 1.3707138417428794e-06,
+      "loss": 0.5332,
+      "step": 3830
+    },
+    {
+      "epoch": 3.469126038076945,
+      "grad_norm": 0.5997902154922485,
+      "learning_rate": 1.3485125072363282e-06,
+      "loss": 0.5724,
+      "step": 3840
+    },
+    {
+      "epoch": 3.4781650754194677,
+      "grad_norm": 0.6153718829154968,
+      "learning_rate": 1.3264643818317836e-06,
+      "loss": 0.5575,
+      "step": 3850
+    },
+    {
+      "epoch": 3.487204112761991,
+      "grad_norm": 0.5330713391304016,
+      "learning_rate": 1.3045703906244284e-06,
+      "loss": 0.5536,
+      "step": 3860
+    },
+    {
+      "epoch": 3.496243150104514,
+      "grad_norm": 0.4775387942790985,
+      "learning_rate": 1.2828314522422857e-06,
+      "loss": 0.5644,
+      "step": 3870
+    },
+    {
+      "epoch": 3.505282187447037,
+      "grad_norm": 0.563755989074707,
+      "learning_rate": 1.2612484788076707e-06,
+      "loss": 0.5561,
+      "step": 3880
+    },
+    {
+      "epoch": 3.51432122478956,
+      "grad_norm": 0.4650101661682129,
+      "learning_rate": 1.2398223758989225e-06,
+      "loss": 0.5158,
+      "step": 3890
+    },
+    {
+      "epoch": 3.523360262132083,
+      "grad_norm": 0.6213396787643433,
+      "learning_rate": 1.218554042512406e-06,
+      "loss": 0.537,
+      "step": 3900
+    },
+    {
+      "epoch": 3.532399299474606,
+      "grad_norm": 0.6274598240852356,
+      "learning_rate": 1.1974443710247963e-06,
+      "loss": 0.5315,
+      "step": 3910
+    },
+    {
+      "epoch": 3.541438336817129,
+      "grad_norm": 0.43882155418395996,
+      "learning_rate": 1.1764942471556295e-06,
+      "loss": 0.5551,
+      "step": 3920
+    },
+    {
+      "epoch": 3.550477374159652,
+      "grad_norm": 0.5064101219177246,
+      "learning_rate": 1.155704549930144e-06,
+      "loss": 0.5399,
+      "step": 3930
+    },
+    {
+      "epoch": 3.559516411502175,
+      "grad_norm": 0.47592440247535706,
+      "learning_rate": 1.1350761516423976e-06,
+      "loss": 0.5028,
+      "step": 3940
+    },
+    {
+      "epoch": 3.568555448844698,
+      "grad_norm": 0.4444693922996521,
+      "learning_rate": 1.1146099178186681e-06,
+      "loss": 0.5154,
+      "step": 3950
+    },
+    {
+      "epoch": 3.577594486187221,
+      "grad_norm": 0.42478910088539124,
+      "learning_rate": 1.0943067071811364e-06,
+      "loss": 0.4756,
+      "step": 3960
+    },
+    {
+      "epoch": 3.586633523529744,
+      "grad_norm": 0.6155830025672913,
+      "learning_rate": 1.0741673716118595e-06,
+      "loss": 0.5483,
+      "step": 3970
+    },
+    {
+      "epoch": 3.5956725608722673,
+      "grad_norm": 0.5128077268600464,
+      "learning_rate": 1.0541927561170206e-06,
+      "loss": 0.5201,
+      "step": 3980
+    },
+    {
+      "epoch": 3.60471159821479,
+      "grad_norm": 0.5960771441459656,
+      "learning_rate": 1.0343836987914834e-06,
+      "loss": 0.5433,
+      "step": 3990
+    },
+    {
+      "epoch": 3.613750635557313,
+      "grad_norm": 0.4781569540500641,
+      "learning_rate": 1.0147410307836185e-06,
+      "loss": 0.5325,
+      "step": 4000
+    },
+    {
+      "epoch": 3.6227896728998363,
+      "grad_norm": 0.5904795527458191,
+      "learning_rate": 9.952655762604346e-07,
+      "loss": 0.4982,
+      "step": 4010
+    },
+    {
+      "epoch": 3.631828710242359,
+      "grad_norm": 0.5607661008834839,
+      "learning_rate": 9.759581523729989e-07,
+      "loss": 0.5037,
+      "step": 4020
+    },
+    {
+      "epoch": 3.640867747584882,
+      "grad_norm": 0.42881935834884644,
+      "learning_rate": 9.568195692221482e-07,
+      "loss": 0.5484,
+      "step": 4030
+    },
+    {
+      "epoch": 3.6499067849274054,
+      "grad_norm": 0.5094344019889832,
+      "learning_rate": 9.37850629824501e-07,
+      "loss": 0.5372,
+      "step": 4040
+    },
+    {
+      "epoch": 3.6589458222699283,
+      "grad_norm": 0.4958745837211609,
+      "learning_rate": 9.190521300787625e-07,
+      "loss": 0.5067,
+      "step": 4050
+    },
+    {
+      "epoch": 3.667984859612451,
+      "grad_norm": 0.5339038968086243,
+      "learning_rate": 9.00424858732335e-07,
+      "loss": 0.5022,
+      "step": 4060
+    },
+    {
+      "epoch": 3.6770238969549744,
+      "grad_norm": 0.6184698343276978,
+      "learning_rate": 8.819695973482156e-07,
+      "loss": 0.5726,
+      "step": 4070
+    },
+    {
+      "epoch": 3.6860629342974973,
+      "grad_norm": 0.4936215877532959,
+      "learning_rate": 8.636871202722108e-07,
+      "loss": 0.5099,
+      "step": 4080
+    },
+    {
+      "epoch": 3.69510197164002,
+      "grad_norm": 0.40462467074394226,
+      "learning_rate": 8.455781946004415e-07,
+      "loss": 0.5423,
+      "step": 4090
+    },
+    {
+      "epoch": 3.7041410089825435,
+      "grad_norm": 0.5905640721321106,
+      "learning_rate": 8.276435801471616e-07,
+      "loss": 0.5029,
+      "step": 4100
+    },
+    {
+      "epoch": 3.7131800463250664,
+      "grad_norm": 0.49841251969337463,
+      "learning_rate": 8.098840294128718e-07,
+      "loss": 0.4849,
+      "step": 4110
+    },
+    {
+      "epoch": 3.7222190836675892,
+      "grad_norm": 0.562427282333374,
+      "learning_rate": 7.923002875527558e-07,
+      "loss": 0.5926,
+      "step": 4120
+    },
+    {
+      "epoch": 3.7312581210101126,
+      "grad_norm": 0.5971468091011047,
+      "learning_rate": 7.748930923454012e-07,
+      "loss": 0.5041,
+      "step": 4130
+    },
+    {
+      "epoch": 3.7402971583526354,
+      "grad_norm": 0.5000337958335876,
+      "learning_rate": 7.5766317416186e-07,
+      "loss": 0.5017,
+      "step": 4140
+    },
+    {
+      "epoch": 3.7493361956951583,
+      "grad_norm": 0.49958381056785583,
+      "learning_rate": 7.406112559349888e-07,
+      "loss": 0.5317,
+      "step": 4150
+    },
+    {
+      "epoch": 3.7583752330376816,
+      "grad_norm": 0.5545982718467712,
+      "learning_rate": 7.237380531291254e-07,
+      "loss": 0.5,
+      "step": 4160
+    },
+    {
+      "epoch": 3.7674142703802045,
+      "grad_norm": 0.5407012104988098,
+      "learning_rate": 7.070442737100658e-07,
+      "loss": 0.5603,
+      "step": 4170
+    },
+    {
+      "epoch": 3.7764533077227274,
+      "grad_norm": 0.6255388855934143,
+      "learning_rate": 6.905306181153604e-07,
+      "loss": 0.5383,
+      "step": 4180
+    },
+    {
+      "epoch": 3.7854923450652507,
+      "grad_norm": 0.4727570116519928,
+      "learning_rate": 6.741977792249238e-07,
+      "loss": 0.5117,
+      "step": 4190
+    },
+    {
+      "epoch": 3.7945313824077735,
+      "grad_norm": 0.602323055267334,
+      "learning_rate": 6.580464423319638e-07,
+      "loss": 0.5272,
+      "step": 4200
+    },
+    {
+      "epoch": 3.8035704197502964,
+      "grad_norm": 0.5532727837562561,
+      "learning_rate": 6.420772851142309e-07,
+      "loss": 0.5496,
+      "step": 4210
+    },
+    {
+      "epoch": 3.8126094570928197,
+      "grad_norm": 0.47623321413993835,
+      "learning_rate": 6.262909776055776e-07,
+      "loss": 0.5413,
+      "step": 4220
+    },
+    {
+      "epoch": 3.8216484944353426,
+      "grad_norm": 0.6107942461967468,
+      "learning_rate": 6.106881821678501e-07,
+      "loss": 0.5263,
+      "step": 4230
+    },
+    {
+      "epoch": 3.8306875317778655,
+      "grad_norm": 0.5319830775260925,
+      "learning_rate": 5.952695534630959e-07,
+      "loss": 0.5198,
+      "step": 4240
+    },
+    {
+      "epoch": 3.839726569120389,
+      "grad_norm": 0.560538649559021,
+      "learning_rate": 5.800357384260952e-07,
+      "loss": 0.4993,
+      "step": 4250
+    },
+    {
+      "epoch": 3.8487656064629117,
+      "grad_norm": 0.5307684540748596,
+      "learning_rate": 5.649873762372154e-07,
+      "loss": 0.5176,
+      "step": 4260
+    },
+    {
+      "epoch": 3.8578046438054345,
+      "grad_norm": 0.48246127367019653,
+      "learning_rate": 5.501250982955975e-07,
+      "loss": 0.4891,
+      "step": 4270
+    },
+    {
+      "epoch": 3.866843681147958,
+      "grad_norm": 0.48443603515625,
+      "learning_rate": 5.354495281926563e-07,
+      "loss": 0.5309,
+      "step": 4280
+    },
+    {
+      "epoch": 3.8758827184904807,
+      "grad_norm": 0.4793282449245453,
+      "learning_rate": 5.20961281685921e-07,
+      "loss": 0.5261,
+      "step": 4290
+    },
+    {
+      "epoch": 3.8849217558330036,
+      "grad_norm": 0.6097034811973572,
+      "learning_rate": 5.066609666731992e-07,
+      "loss": 0.5466,
+      "step": 4300
+    },
+    {
+      "epoch": 3.893960793175527,
+      "grad_norm": 0.4524785280227661,
+      "learning_rate": 4.92549183167067e-07,
+      "loss": 0.5186,
+      "step": 4310
+    },
+    {
+      "epoch": 3.9029998305180498,
+      "grad_norm": 0.6819131374359131,
+      "learning_rate": 4.78626523269698e-07,
+      "loss": 0.5112,
+      "step": 4320
+    },
+    {
+      "epoch": 3.9120388678605726,
+      "grad_norm": 0.5640683174133301,
+      "learning_rate": 4.648935711480168e-07,
+      "loss": 0.5359,
+      "step": 4330
+    },
+    {
+      "epoch": 3.921077905203096,
+      "grad_norm": 0.5294342041015625,
+      "learning_rate": 4.5135090300919006e-07,
+      "loss": 0.5497,
+      "step": 4340
+    },
+    {
+      "epoch": 3.930116942545619,
+      "grad_norm": 0.5066547393798828,
+      "learning_rate": 4.3799908707645e-07,
+      "loss": 0.505,
+      "step": 4350
+    },
+    {
+      "epoch": 3.9391559798881417,
+      "grad_norm": 0.4618836045265198,
+      "learning_rate": 4.2483868356525126e-07,
+      "loss": 0.5105,
+      "step": 4360
+    },
+    {
+      "epoch": 3.948195017230665,
+      "grad_norm": 0.5426579713821411,
+      "learning_rate": 4.1187024465976887e-07,
+      "loss": 0.5353,
+      "step": 4370
+    },
+    {
+      "epoch": 3.957234054573188,
+      "grad_norm": 0.5433206558227539,
+      "learning_rate": 3.990943144897258e-07,
+      "loss": 0.5172,
+      "step": 4380
+    },
+    {
+      "epoch": 3.9662730919157108,
+      "grad_norm": 0.4879183769226074,
+      "learning_rate": 3.865114291075639e-07,
+      "loss": 0.5549,
+      "step": 4390
+    },
+    {
+      "epoch": 3.975312129258234,
+      "grad_norm": 0.6213659644126892,
+      "learning_rate": 3.741221164659531e-07,
+      "loss": 0.5551,
+      "step": 4400
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 5000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.397837458934827e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-4400/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b10335d3e12c67c825f0eecf99c175d30115fd5cd7f73e1ab9c2d93693e6c9f2
+size 6353