diff --git a/.gitattributes b/.gitattributes
index 2c70b4a4c7a5edc6e2694ca124fc22ed2d8a3bde..42840a62ddfa1d2fced8ca2aba8d024304dbe874 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -5029,3 +5029,12 @@ Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_
 Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.1-num-702-sd-0/checkpoint-752/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.1-num-702-sd-0/checkpoint-903/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.1-num-702-sd-0/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/README.md b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# 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 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "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": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ec1ef8d47b111af6b72ef86a5e420ba55ab736f6
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18d3065462d28b00e94897958b444dda35b3dd73c28df887bc7c6cfe0ea65800
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/README.md b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# 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 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "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": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e27047f51527938d40ac0db632858ab8c40f6335
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88cc8a0bc46271dd7b4c992c6f7d21a4b68a961d125b50bb54dc061d1ecd4cef
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..81a8cd298ea89ea511d69aa8264de6dd53db9161
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1355012e59d01cca6936cab9109c2731c32dc7c95d4fd18cca06b60d907884d5
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..b223759d202d46badee5d55abac6365814f4eb2d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26b99e05be79bd0f1b9ff6d769fd825f009768e072fbd1fcea8e5a6d49a34ede
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3f69df0e45093107ef86f42d9b7bfb2e3333706a
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8073b3be73946b85bc167aa6d6e91a7136066a38112f637602354d6a08b16a98
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..3c4fd1b46ecfd47235c7688033b7e3d52d77ceab
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/trainer_state.json
@@ -0,0 +1,7727 @@
+{
+  "best_metric": 1.1534006595611572,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727",
+  "epoch": 7.997066373303997,
+  "eval_steps": 10,
+  "global_step": 10904,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007334066740007334,
+      "grad_norm": 0.47521963715553284,
+      "learning_rate": 0.0002,
+      "loss": 1.9722,
+      "step": 10
+    },
+    {
+      "epoch": 0.014668133480014669,
+      "grad_norm": 0.5395162105560303,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 20
+    },
+    {
+      "epoch": 0.022002200220022004,
+      "grad_norm": 0.4305780231952667,
+      "learning_rate": 0.0002,
+      "loss": 1.4202,
+      "step": 30
+    },
+    {
+      "epoch": 0.029336266960029337,
+      "grad_norm": 0.6938246488571167,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 40
+    },
+    {
+      "epoch": 0.03667033370003667,
+      "grad_norm": 1.5133819580078125,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 50
+    },
+    {
+      "epoch": 0.04400440044004401,
+      "grad_norm": 0.9173883199691772,
+      "learning_rate": 0.0002,
+      "loss": 1.3132,
+      "step": 60
+    },
+    {
+      "epoch": 0.05133846718005134,
+      "grad_norm": 0.4619861841201782,
+      "learning_rate": 0.0002,
+      "loss": 1.2844,
+      "step": 70
+    },
+    {
+      "epoch": 0.058672533920058674,
+      "grad_norm": 0.46118637919425964,
+      "learning_rate": 0.0002,
+      "loss": 1.2108,
+      "step": 80
+    },
+    {
+      "epoch": 0.066006600660066,
+      "grad_norm": 0.4468648135662079,
+      "learning_rate": 0.0002,
+      "loss": 1.3441,
+      "step": 90
+    },
+    {
+      "epoch": 0.07334066740007333,
+      "grad_norm": 0.46123769879341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1863,
+      "step": 100
+    },
+    {
+      "epoch": 0.08067473414008068,
+      "grad_norm": 0.4859139025211334,
+      "learning_rate": 0.0002,
+      "loss": 1.2772,
+      "step": 110
+    },
+    {
+      "epoch": 0.08800880088008801,
+      "grad_norm": 0.4384922385215759,
+      "learning_rate": 0.0002,
+      "loss": 1.2087,
+      "step": 120
+    },
+    {
+      "epoch": 0.09534286762009535,
+      "grad_norm": 0.39519360661506653,
+      "learning_rate": 0.0002,
+      "loss": 1.2927,
+      "step": 130
+    },
+    {
+      "epoch": 0.10267693436010268,
+      "grad_norm": 0.4049859344959259,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 140
+    },
+    {
+      "epoch": 0.11001100110011001,
+      "grad_norm": 0.4605638086795807,
+      "learning_rate": 0.0002,
+      "loss": 1.293,
+      "step": 150
+    },
+    {
+      "epoch": 0.11734506784011735,
+      "grad_norm": 0.4201928377151489,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 160
+    },
+    {
+      "epoch": 0.12467913458012468,
+      "grad_norm": 0.5367777347564697,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 170
+    },
+    {
+      "epoch": 0.132013201320132,
+      "grad_norm": 0.41752299666404724,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 180
+    },
+    {
+      "epoch": 0.13934726806013933,
+      "grad_norm": 0.31597763299942017,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 190
+    },
+    {
+      "epoch": 0.14668133480014667,
+      "grad_norm": 0.7468788623809814,
+      "learning_rate": 0.0002,
+      "loss": 1.2441,
+      "step": 200
+    },
+    {
+      "epoch": 0.15401540154015403,
+      "grad_norm": 0.3403034508228302,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 210
+    },
+    {
+      "epoch": 0.16134946828016136,
+      "grad_norm": 0.34240293502807617,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 220
+    },
+    {
+      "epoch": 0.1686835350201687,
+      "grad_norm": 0.356158971786499,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 230
+    },
+    {
+      "epoch": 0.17601760176017603,
+      "grad_norm": 0.3448857367038727,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 240
+    },
+    {
+      "epoch": 0.18335166850018336,
+      "grad_norm": 0.3475699722766876,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 250
+    },
+    {
+      "epoch": 0.1906857352401907,
+      "grad_norm": 0.2770358622074127,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 260
+    },
+    {
+      "epoch": 0.19801980198019803,
+      "grad_norm": 0.4310270845890045,
+      "learning_rate": 0.0002,
+      "loss": 1.2238,
+      "step": 270
+    },
+    {
+      "epoch": 0.20535386872020536,
+      "grad_norm": 0.335041880607605,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 280
+    },
+    {
+      "epoch": 0.2126879354602127,
+      "grad_norm": 0.3420602083206177,
+      "learning_rate": 0.0002,
+      "loss": 1.0959,
+      "step": 290
+    },
+    {
+      "epoch": 0.22002200220022003,
+      "grad_norm": 0.325001060962677,
+      "learning_rate": 0.0002,
+      "loss": 1.1232,
+      "step": 300
+    },
+    {
+      "epoch": 0.22735606894022736,
+      "grad_norm": 0.3027827739715576,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 310
+    },
+    {
+      "epoch": 0.2346901356802347,
+      "grad_norm": 0.435550719499588,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 320
+    },
+    {
+      "epoch": 0.24202420242024203,
+      "grad_norm": 0.3884522616863251,
+      "learning_rate": 0.0002,
+      "loss": 1.2045,
+      "step": 330
+    },
+    {
+      "epoch": 0.24935826916024936,
+      "grad_norm": 0.7736002206802368,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 340
+    },
+    {
+      "epoch": 0.2566923359002567,
+      "grad_norm": 0.35052821040153503,
+      "learning_rate": 0.0002,
+      "loss": 1.3606,
+      "step": 350
+    },
+    {
+      "epoch": 0.264026402640264,
+      "grad_norm": 0.3311890959739685,
+      "learning_rate": 0.0002,
+      "loss": 1.2129,
+      "step": 360
+    },
+    {
+      "epoch": 0.27136046938027136,
+      "grad_norm": 0.7473500370979309,
+      "learning_rate": 0.0002,
+      "loss": 1.2219,
+      "step": 370
+    },
+    {
+      "epoch": 0.27869453612027867,
+      "grad_norm": 0.3681875765323639,
+      "learning_rate": 0.0002,
+      "loss": 1.2712,
+      "step": 380
+    },
+    {
+      "epoch": 0.28602860286028603,
+      "grad_norm": 0.3764737844467163,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 390
+    },
+    {
+      "epoch": 0.29336266960029334,
+      "grad_norm": 0.4243989586830139,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 400
+    },
+    {
+      "epoch": 0.3006967363403007,
+      "grad_norm": 0.2658531963825226,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 410
+    },
+    {
+      "epoch": 0.30803080308030806,
+      "grad_norm": 0.3436793386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.1622,
+      "step": 420
+    },
+    {
+      "epoch": 0.31536486982031536,
+      "grad_norm": 0.5101129412651062,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 430
+    },
+    {
+      "epoch": 0.3226989365603227,
+      "grad_norm": 0.3319750726222992,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 440
+    },
+    {
+      "epoch": 0.33003300330033003,
+      "grad_norm": 0.385148286819458,
+      "learning_rate": 0.0002,
+      "loss": 1.1804,
+      "step": 450
+    },
+    {
+      "epoch": 0.3373670700403374,
+      "grad_norm": 0.3477935791015625,
+      "learning_rate": 0.0002,
+      "loss": 1.1808,
+      "step": 460
+    },
+    {
+      "epoch": 0.3447011367803447,
+      "grad_norm": 0.29748716950416565,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 470
+    },
+    {
+      "epoch": 0.35203520352035206,
+      "grad_norm": 0.34083324670791626,
+      "learning_rate": 0.0002,
+      "loss": 1.19,
+      "step": 480
+    },
+    {
+      "epoch": 0.35936927026035936,
+      "grad_norm": 0.36904552578926086,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 490
+    },
+    {
+      "epoch": 0.3667033370003667,
+      "grad_norm": 0.315483033657074,
+      "learning_rate": 0.0002,
+      "loss": 1.2223,
+      "step": 500
+    },
+    {
+      "epoch": 0.37403740374037403,
+      "grad_norm": 0.44897955656051636,
+      "learning_rate": 0.0002,
+      "loss": 1.1461,
+      "step": 510
+    },
+    {
+      "epoch": 0.3813714704803814,
+      "grad_norm": 0.3160701394081116,
+      "learning_rate": 0.0002,
+      "loss": 1.3035,
+      "step": 520
+    },
+    {
+      "epoch": 0.3887055372203887,
+      "grad_norm": 0.29584741592407227,
+      "learning_rate": 0.0002,
+      "loss": 1.3197,
+      "step": 530
+    },
+    {
+      "epoch": 0.39603960396039606,
+      "grad_norm": 0.5430002808570862,
+      "learning_rate": 0.0002,
+      "loss": 1.2983,
+      "step": 540
+    },
+    {
+      "epoch": 0.40337367070040336,
+      "grad_norm": 0.2908070683479309,
+      "learning_rate": 0.0002,
+      "loss": 1.2459,
+      "step": 550
+    },
+    {
+      "epoch": 0.4107077374404107,
+      "grad_norm": 0.35066530108451843,
+      "learning_rate": 0.0002,
+      "loss": 1.2384,
+      "step": 560
+    },
+    {
+      "epoch": 0.41804180418041803,
+      "grad_norm": 0.37588003277778625,
+      "learning_rate": 0.0002,
+      "loss": 1.1784,
+      "step": 570
+    },
+    {
+      "epoch": 0.4253758709204254,
+      "grad_norm": 0.3112126886844635,
+      "learning_rate": 0.0002,
+      "loss": 1.2334,
+      "step": 580
+    },
+    {
+      "epoch": 0.4327099376604327,
+      "grad_norm": 0.35577139258384705,
+      "learning_rate": 0.0002,
+      "loss": 1.1439,
+      "step": 590
+    },
+    {
+      "epoch": 0.44004400440044006,
+      "grad_norm": 0.31706422567367554,
+      "learning_rate": 0.0002,
+      "loss": 1.184,
+      "step": 600
+    },
+    {
+      "epoch": 0.44737807114044736,
+      "grad_norm": 0.3249092102050781,
+      "learning_rate": 0.0002,
+      "loss": 1.2081,
+      "step": 610
+    },
+    {
+      "epoch": 0.4547121378804547,
+      "grad_norm": 0.3842705488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.0824,
+      "step": 620
+    },
+    {
+      "epoch": 0.46204620462046203,
+      "grad_norm": 0.390991747379303,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 630
+    },
+    {
+      "epoch": 0.4693802713604694,
+      "grad_norm": 0.27532413601875305,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 640
+    },
+    {
+      "epoch": 0.4767143381004767,
+      "grad_norm": 0.31412816047668457,
+      "learning_rate": 0.0002,
+      "loss": 1.1058,
+      "step": 650
+    },
+    {
+      "epoch": 0.48404840484048406,
+      "grad_norm": 0.32117101550102234,
+      "learning_rate": 0.0002,
+      "loss": 1.1312,
+      "step": 660
+    },
+    {
+      "epoch": 0.49138247158049136,
+      "grad_norm": 0.3810010254383087,
+      "learning_rate": 0.0002,
+      "loss": 1.2423,
+      "step": 670
+    },
+    {
+      "epoch": 0.4987165383204987,
+      "grad_norm": 0.36289164423942566,
+      "learning_rate": 0.0002,
+      "loss": 1.1978,
+      "step": 680
+    },
+    {
+      "epoch": 0.506050605060506,
+      "grad_norm": 0.34458720684051514,
+      "learning_rate": 0.0002,
+      "loss": 1.2034,
+      "step": 690
+    },
+    {
+      "epoch": 0.5133846718005134,
+      "grad_norm": 0.32844600081443787,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 700
+    },
+    {
+      "epoch": 0.5207187385405208,
+      "grad_norm": 0.3144175708293915,
+      "learning_rate": 0.0002,
+      "loss": 1.0807,
+      "step": 710
+    },
+    {
+      "epoch": 0.528052805280528,
+      "grad_norm": 0.3898887634277344,
+      "learning_rate": 0.0002,
+      "loss": 1.1952,
+      "step": 720
+    },
+    {
+      "epoch": 0.5353868720205354,
+      "grad_norm": 1.3220758438110352,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 730
+    },
+    {
+      "epoch": 0.5427209387605427,
+      "grad_norm": 0.3635874390602112,
+      "learning_rate": 0.0002,
+      "loss": 1.227,
+      "step": 740
+    },
+    {
+      "epoch": 0.5500550055005501,
+      "grad_norm": 0.3138217628002167,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 750
+    },
+    {
+      "epoch": 0.5573890722405573,
+      "grad_norm": 0.4063207805156708,
+      "learning_rate": 0.0002,
+      "loss": 1.1516,
+      "step": 760
+    },
+    {
+      "epoch": 0.5647231389805647,
+      "grad_norm": 0.3926219940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 770
+    },
+    {
+      "epoch": 0.5720572057205721,
+      "grad_norm": 0.31954652070999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1726,
+      "step": 780
+    },
+    {
+      "epoch": 0.5793912724605794,
+      "grad_norm": 0.4248711168766022,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 790
+    },
+    {
+      "epoch": 0.5867253392005867,
+      "grad_norm": 0.643004834651947,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 800
+    },
+    {
+      "epoch": 0.594059405940594,
+      "grad_norm": 0.3479592800140381,
+      "learning_rate": 0.0002,
+      "loss": 1.1793,
+      "step": 810
+    },
+    {
+      "epoch": 0.6013934726806014,
+      "grad_norm": 0.4684754014015198,
+      "learning_rate": 0.0002,
+      "loss": 1.2426,
+      "step": 820
+    },
+    {
+      "epoch": 0.6087275394206088,
+      "grad_norm": 0.3739790916442871,
+      "learning_rate": 0.0002,
+      "loss": 1.2002,
+      "step": 830
+    },
+    {
+      "epoch": 0.6160616061606161,
+      "grad_norm": 0.40884748101234436,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 840
+    },
+    {
+      "epoch": 0.6233956729006234,
+      "grad_norm": 0.9722164273262024,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 850
+    },
+    {
+      "epoch": 0.6307297396406307,
+      "grad_norm": 0.42992347478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.3069,
+      "step": 860
+    },
+    {
+      "epoch": 0.6380638063806381,
+      "grad_norm": 0.36654195189476013,
+      "learning_rate": 0.0002,
+      "loss": 1.1339,
+      "step": 870
+    },
+    {
+      "epoch": 0.6453978731206454,
+      "grad_norm": 0.4113832116127014,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 880
+    },
+    {
+      "epoch": 0.6527319398606527,
+      "grad_norm": 0.2948838770389557,
+      "learning_rate": 0.0002,
+      "loss": 1.2163,
+      "step": 890
+    },
+    {
+      "epoch": 0.6600660066006601,
+      "grad_norm": 0.38330280780792236,
+      "learning_rate": 0.0002,
+      "loss": 1.1081,
+      "step": 900
+    },
+    {
+      "epoch": 0.6674000733406674,
+      "grad_norm": 0.4428867697715759,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 910
+    },
+    {
+      "epoch": 0.6747341400806748,
+      "grad_norm": 0.23659265041351318,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 920
+    },
+    {
+      "epoch": 0.682068206820682,
+      "grad_norm": 0.323685884475708,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 930
+    },
+    {
+      "epoch": 0.6894022735606894,
+      "grad_norm": 0.39157727360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 940
+    },
+    {
+      "epoch": 0.6967363403006968,
+      "grad_norm": 0.27189481258392334,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 950
+    },
+    {
+      "epoch": 0.7040704070407041,
+      "grad_norm": 0.529883861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.1033,
+      "step": 960
+    },
+    {
+      "epoch": 0.7114044737807114,
+      "grad_norm": 0.34758689999580383,
+      "learning_rate": 0.0002,
+      "loss": 1.139,
+      "step": 970
+    },
+    {
+      "epoch": 0.7187385405207187,
+      "grad_norm": 0.831749439239502,
+      "learning_rate": 0.0002,
+      "loss": 1.2197,
+      "step": 980
+    },
+    {
+      "epoch": 0.7260726072607261,
+      "grad_norm": 0.4438304007053375,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 990
+    },
+    {
+      "epoch": 0.7334066740007334,
+      "grad_norm": 0.33840006589889526,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7407407407407407,
+      "grad_norm": 0.3454797863960266,
+      "learning_rate": 0.0002,
+      "loss": 1.254,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7480748074807481,
+      "grad_norm": 0.38999441266059875,
+      "learning_rate": 0.0002,
+      "loss": 1.106,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7554088742207554,
+      "grad_norm": 0.2829911708831787,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7627429409607628,
+      "grad_norm": 0.36918163299560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 1040
+    },
+    {
+      "epoch": 0.77007700770077,
+      "grad_norm": 0.3415680229663849,
+      "learning_rate": 0.0002,
+      "loss": 1.3028,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7774110744407774,
+      "grad_norm": 0.2974182963371277,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7847451411807848,
+      "grad_norm": 0.3880919814109802,
+      "learning_rate": 0.0002,
+      "loss": 1.194,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7920792079207921,
+      "grad_norm": 0.33503302931785583,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7994132746607994,
+      "grad_norm": 0.3728407025337219,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8067473414008067,
+      "grad_norm": 0.3509373664855957,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8140814081408141,
+      "grad_norm": 0.42228564620018005,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8214154748808215,
+      "grad_norm": 0.313467800617218,
+      "learning_rate": 0.0002,
+      "loss": 1.1788,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8287495416208287,
+      "grad_norm": 0.3378850817680359,
+      "learning_rate": 0.0002,
+      "loss": 1.1971,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8360836083608361,
+      "grad_norm": 0.43200382590293884,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8434176751008434,
+      "grad_norm": 0.3309599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.3203,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8507517418408508,
+      "grad_norm": 0.3526846170425415,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1160
+    },
+    {
+      "epoch": 0.858085808580858,
+      "grad_norm": 1.2722247838974,
+      "learning_rate": 0.0002,
+      "loss": 1.0851,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8654198753208654,
+      "grad_norm": 0.34142059087753296,
+      "learning_rate": 0.0002,
+      "loss": 1.0785,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8727539420608728,
+      "grad_norm": 0.3805823028087616,
+      "learning_rate": 0.0002,
+      "loss": 1.2187,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8800880088008801,
+      "grad_norm": 0.3931232690811157,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8874220755408874,
+      "grad_norm": 0.2937372624874115,
+      "learning_rate": 0.0002,
+      "loss": 1.0948,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8947561422808947,
+      "grad_norm": 0.3757196366786957,
+      "learning_rate": 0.0002,
+      "loss": 1.1228,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9020902090209021,
+      "grad_norm": 0.3502705991268158,
+      "learning_rate": 0.0002,
+      "loss": 1.1222,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9094242757609095,
+      "grad_norm": 0.32758915424346924,
+      "learning_rate": 0.0002,
+      "loss": 1.2242,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9167583425009168,
+      "grad_norm": 0.37199416756629944,
+      "learning_rate": 0.0002,
+      "loss": 1.215,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9240924092409241,
+      "grad_norm": 0.3551490604877472,
+      "learning_rate": 0.0002,
+      "loss": 1.1225,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9314264759809314,
+      "grad_norm": 0.2859550714492798,
+      "learning_rate": 0.0002,
+      "loss": 1.1966,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9387605427209388,
+      "grad_norm": 0.427990585565567,
+      "learning_rate": 0.0002,
+      "loss": 1.2186,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9460946094609461,
+      "grad_norm": 0.33717992901802063,
+      "learning_rate": 0.0002,
+      "loss": 1.2848,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9534286762009534,
+      "grad_norm": 0.30225634574890137,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9607627429409608,
+      "grad_norm": 0.385821133852005,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9680968096809681,
+      "grad_norm": 0.35278066992759705,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9754308764209755,
+      "grad_norm": 0.49987098574638367,
+      "learning_rate": 0.0002,
+      "loss": 1.1071,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9827649431609827,
+      "grad_norm": 0.3842747211456299,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9900990099009901,
+      "grad_norm": 0.6274653673171997,
+      "learning_rate": 0.0002,
+      "loss": 1.0862,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9974330766409975,
+      "grad_norm": 0.5239808559417725,
+      "learning_rate": 0.0002,
+      "loss": 1.124,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9996332966629996,
+      "eval_loss": 1.1822267770767212,
+      "eval_runtime": 32.7389,
+      "eval_samples_per_second": 13.165,
+      "eval_steps_per_second": 1.649,
+      "step": 1363
+    },
+    {
+      "epoch": 1.0047671433810048,
+      "grad_norm": 0.45311301946640015,
+      "learning_rate": 0.0002,
+      "loss": 1.096,
+      "step": 1370
+    },
+    {
+      "epoch": 1.012101210121012,
+      "grad_norm": 0.29685574769973755,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0194352768610195,
+      "grad_norm": 0.3290937840938568,
+      "learning_rate": 0.0002,
+      "loss": 1.0302,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0267693436010268,
+      "grad_norm": 0.3801758587360382,
+      "learning_rate": 0.0002,
+      "loss": 1.0295,
+      "step": 1400
+    },
+    {
+      "epoch": 1.034103410341034,
+      "grad_norm": 0.794174313545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0414374770810415,
+      "grad_norm": 0.3854154646396637,
+      "learning_rate": 0.0002,
+      "loss": 1.2232,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0487715438210488,
+      "grad_norm": 0.32702451944351196,
+      "learning_rate": 0.0002,
+      "loss": 1.0652,
+      "step": 1430
+    },
+    {
+      "epoch": 1.056105610561056,
+      "grad_norm": 0.7815203666687012,
+      "learning_rate": 0.0002,
+      "loss": 1.1144,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0634396773010635,
+      "grad_norm": 0.3087436854839325,
+      "learning_rate": 0.0002,
+      "loss": 1.1316,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0707737440410707,
+      "grad_norm": 0.3847602903842926,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0781078107810782,
+      "grad_norm": 0.3693031370639801,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0854418775210855,
+      "grad_norm": 0.4111202359199524,
+      "learning_rate": 0.0002,
+      "loss": 1.0995,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0927759442610927,
+      "grad_norm": 0.41452381014823914,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1001100110011002,
+      "grad_norm": 0.3336445093154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1068,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1074440777411074,
+      "grad_norm": 0.3923407793045044,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1147781444811147,
+      "grad_norm": 0.46215683221817017,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1221122112211221,
+      "grad_norm": 0.3592156767845154,
+      "learning_rate": 0.0002,
+      "loss": 1.1133,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1294462779611294,
+      "grad_norm": 0.361110657453537,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1367803447011369,
+      "grad_norm": 0.5317131280899048,
+      "learning_rate": 0.0002,
+      "loss": 1.1553,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1441144114411441,
+      "grad_norm": 0.3882388174533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1514484781811514,
+      "grad_norm": 0.3259428143501282,
+      "learning_rate": 0.0002,
+      "loss": 1.0805,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1587825449211588,
+      "grad_norm": 0.410935640335083,
+      "learning_rate": 0.0002,
+      "loss": 1.1819,
+      "step": 1580
+    },
+    {
+      "epoch": 1.166116611661166,
+      "grad_norm": 0.44940185546875,
+      "learning_rate": 0.0002,
+      "loss": 1.1143,
+      "step": 1590
+    },
+    {
+      "epoch": 1.1734506784011733,
+      "grad_norm": 0.5106484293937683,
+      "learning_rate": 0.0002,
+      "loss": 1.0334,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1807847451411808,
+      "grad_norm": 0.6603665947914124,
+      "learning_rate": 0.0002,
+      "loss": 1.2376,
+      "step": 1610
+    },
+    {
+      "epoch": 1.188118811881188,
+      "grad_norm": 0.4799964129924774,
+      "learning_rate": 0.0002,
+      "loss": 1.1227,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1954528786211955,
+      "grad_norm": 0.4389883279800415,
+      "learning_rate": 0.0002,
+      "loss": 1.1191,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2027869453612028,
+      "grad_norm": 0.4188813269138336,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 1640
+    },
+    {
+      "epoch": 1.21012101210121,
+      "grad_norm": 0.7132157683372498,
+      "learning_rate": 0.0002,
+      "loss": 1.0605,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2174550788412175,
+      "grad_norm": 0.507480263710022,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2247891455812248,
+      "grad_norm": 0.9452332854270935,
+      "learning_rate": 0.0002,
+      "loss": 0.9948,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2321232123212322,
+      "grad_norm": 0.4121614992618561,
+      "learning_rate": 0.0002,
+      "loss": 1.0228,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2394572790612395,
+      "grad_norm": 0.34230247139930725,
+      "learning_rate": 0.0002,
+      "loss": 1.0366,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2467913458012467,
+      "grad_norm": 0.4026208817958832,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2541254125412542,
+      "grad_norm": 0.46673697233200073,
+      "learning_rate": 0.0002,
+      "loss": 1.0206,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2614594792812615,
+      "grad_norm": 0.38349825143814087,
+      "learning_rate": 0.0002,
+      "loss": 1.0827,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2687935460212687,
+      "grad_norm": 0.4049997627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.0356,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2761276127612762,
+      "grad_norm": 0.3417615294456482,
+      "learning_rate": 0.0002,
+      "loss": 0.9504,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2834616795012834,
+      "grad_norm": 0.4277614951133728,
+      "learning_rate": 0.0002,
+      "loss": 1.094,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2907957462412907,
+      "grad_norm": 0.5864202976226807,
+      "learning_rate": 0.0002,
+      "loss": 0.9938,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2981298129812981,
+      "grad_norm": 0.7097493410110474,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3054638797213054,
+      "grad_norm": 0.3145381212234497,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3127979464613129,
+      "grad_norm": 0.5116165280342102,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3201320132013201,
+      "grad_norm": 0.7469736337661743,
+      "learning_rate": 0.0002,
+      "loss": 1.0765,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3274660799413276,
+      "grad_norm": 0.32272255420684814,
+      "learning_rate": 0.0002,
+      "loss": 1.0663,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3348001466813348,
+      "grad_norm": 0.3534623086452484,
+      "learning_rate": 0.0002,
+      "loss": 0.9887,
+      "step": 1820
+    },
+    {
+      "epoch": 1.342134213421342,
+      "grad_norm": 0.36127907037734985,
+      "learning_rate": 0.0002,
+      "loss": 1.1628,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3494682801613496,
+      "grad_norm": 0.4072401523590088,
+      "learning_rate": 0.0002,
+      "loss": 1.0972,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3568023469013568,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.1267,
+      "step": 1850
+    },
+    {
+      "epoch": 1.364136413641364,
+      "grad_norm": 0.412883460521698,
+      "learning_rate": 0.0002,
+      "loss": 1.0173,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3714704803813715,
+      "grad_norm": 0.3735875189304352,
+      "learning_rate": 0.0002,
+      "loss": 1.0265,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3788045471213788,
+      "grad_norm": 0.39158159494400024,
+      "learning_rate": 0.0002,
+      "loss": 1.1061,
+      "step": 1880
+    },
+    {
+      "epoch": 1.386138613861386,
+      "grad_norm": 0.44431769847869873,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3934726806013935,
+      "grad_norm": 0.37772801518440247,
+      "learning_rate": 0.0002,
+      "loss": 1.0216,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4008067473414008,
+      "grad_norm": 0.4056641757488251,
+      "learning_rate": 0.0002,
+      "loss": 1.0674,
+      "step": 1910
+    },
+    {
+      "epoch": 1.408140814081408,
+      "grad_norm": 0.41612377762794495,
+      "learning_rate": 0.0002,
+      "loss": 1.0256,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4154748808214155,
+      "grad_norm": 0.41153013706207275,
+      "learning_rate": 0.0002,
+      "loss": 1.0467,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4228089475614227,
+      "grad_norm": 0.387845516204834,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4301430143014302,
+      "grad_norm": 0.3809587061405182,
+      "learning_rate": 0.0002,
+      "loss": 1.1094,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4374770810414375,
+      "grad_norm": 0.3625726103782654,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1960
+    },
+    {
+      "epoch": 1.444811147781445,
+      "grad_norm": 0.5294290781021118,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4521452145214522,
+      "grad_norm": 0.39975494146347046,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4594792812614594,
+      "grad_norm": 0.4181167185306549,
+      "learning_rate": 0.0002,
+      "loss": 0.9704,
+      "step": 1990
+    },
+    {
+      "epoch": 1.466813348001467,
+      "grad_norm": 0.42001503705978394,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4741474147414741,
+      "grad_norm": 0.4877578616142273,
+      "learning_rate": 0.0002,
+      "loss": 1.1266,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4814814814814814,
+      "grad_norm": 0.4050969183444977,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4888155482214889,
+      "grad_norm": 0.39068883657455444,
+      "learning_rate": 0.0002,
+      "loss": 1.0562,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4961496149614961,
+      "grad_norm": 0.421282559633255,
+      "learning_rate": 0.0002,
+      "loss": 1.0464,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5034836817015034,
+      "grad_norm": 0.47092297673225403,
+      "learning_rate": 0.0002,
+      "loss": 1.0532,
+      "step": 2050
+    },
+    {
+      "epoch": 1.5108177484415108,
+      "grad_norm": 0.39688974618911743,
+      "learning_rate": 0.0002,
+      "loss": 0.9348,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5181518151815183,
+      "grad_norm": 0.5529879331588745,
+      "learning_rate": 0.0002,
+      "loss": 1.08,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5254858819215253,
+      "grad_norm": 0.4879782199859619,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5328199486615328,
+      "grad_norm": 0.5517361164093018,
+      "learning_rate": 0.0002,
+      "loss": 1.0432,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5401540154015403,
+      "grad_norm": 0.44015637040138245,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5474880821415475,
+      "grad_norm": 0.5435167551040649,
+      "learning_rate": 0.0002,
+      "loss": 1.1873,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5548221488815548,
+      "grad_norm": 0.5714033246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1076,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5621562156215623,
+      "grad_norm": 0.31732529401779175,
+      "learning_rate": 0.0002,
+      "loss": 1.1107,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5694902823615695,
+      "grad_norm": 0.49068278074264526,
+      "learning_rate": 0.0002,
+      "loss": 1.0817,
+      "step": 2140
+    },
+    {
+      "epoch": 1.5768243491015768,
+      "grad_norm": 0.46851542592048645,
+      "learning_rate": 0.0002,
+      "loss": 1.0254,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5841584158415842,
+      "grad_norm": 0.5083092451095581,
+      "learning_rate": 0.0002,
+      "loss": 1.0623,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5914924825815915,
+      "grad_norm": 0.9822936058044434,
+      "learning_rate": 0.0002,
+      "loss": 1.0603,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5988265493215987,
+      "grad_norm": 0.4575989246368408,
+      "learning_rate": 0.0002,
+      "loss": 0.9986,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6061606160616062,
+      "grad_norm": 0.47444286942481995,
+      "learning_rate": 0.0002,
+      "loss": 1.1292,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6134946828016135,
+      "grad_norm": 0.7208226919174194,
+      "learning_rate": 0.0002,
+      "loss": 1.0136,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6208287495416207,
+      "grad_norm": 0.43791481852531433,
+      "learning_rate": 0.0002,
+      "loss": 1.15,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6281628162816282,
+      "grad_norm": 0.5245792865753174,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6354968830216357,
+      "grad_norm": 0.39289429783821106,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6428309497616427,
+      "grad_norm": 0.6106135845184326,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6501650165016502,
+      "grad_norm": 0.3722580671310425,
+      "learning_rate": 0.0002,
+      "loss": 1.0129,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6574990832416576,
+      "grad_norm": 0.3649403750896454,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6648331499816649,
+      "grad_norm": 0.46514248847961426,
+      "learning_rate": 0.0002,
+      "loss": 1.0037,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6721672167216721,
+      "grad_norm": 0.42034927010536194,
+      "learning_rate": 0.0002,
+      "loss": 1.0022,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6795012834616796,
+      "grad_norm": 0.45202910900115967,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6868353502016868,
+      "grad_norm": 0.36257603764533997,
+      "learning_rate": 0.0002,
+      "loss": 1.0866,
+      "step": 2300
+    },
+    {
+      "epoch": 1.694169416941694,
+      "grad_norm": 0.6340323090553284,
+      "learning_rate": 0.0002,
+      "loss": 1.0973,
+      "step": 2310
+    },
+    {
+      "epoch": 1.7015034836817016,
+      "grad_norm": 0.4352878928184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0615,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7088375504217088,
+      "grad_norm": 0.45029792189598083,
+      "learning_rate": 0.0002,
+      "loss": 1.0629,
+      "step": 2330
+    },
+    {
+      "epoch": 1.716171617161716,
+      "grad_norm": 0.3891315758228302,
+      "learning_rate": 0.0002,
+      "loss": 0.9621,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7235056839017235,
+      "grad_norm": 0.35180050134658813,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7308397506417308,
+      "grad_norm": 0.42367449402809143,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 2360
+    },
+    {
+      "epoch": 1.738173817381738,
+      "grad_norm": 0.4553675353527069,
+      "learning_rate": 0.0002,
+      "loss": 1.0376,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7455078841217455,
+      "grad_norm": 0.5944654941558838,
+      "learning_rate": 0.0002,
+      "loss": 1.1467,
+      "step": 2380
+    },
+    {
+      "epoch": 1.752841950861753,
+      "grad_norm": 0.3479664623737335,
+      "learning_rate": 0.0002,
+      "loss": 1.0548,
+      "step": 2390
+    },
+    {
+      "epoch": 1.76017601760176,
+      "grad_norm": 0.3585502505302429,
+      "learning_rate": 0.0002,
+      "loss": 1.0798,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7675100843417675,
+      "grad_norm": 0.4263346493244171,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 2410
+    },
+    {
+      "epoch": 1.774844151081775,
+      "grad_norm": 0.5476409196853638,
+      "learning_rate": 0.0002,
+      "loss": 1.054,
+      "step": 2420
+    },
+    {
+      "epoch": 1.7821782178217822,
+      "grad_norm": 0.3694186508655548,
+      "learning_rate": 0.0002,
+      "loss": 1.1615,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7895122845617895,
+      "grad_norm": 0.9185658693313599,
+      "learning_rate": 0.0002,
+      "loss": 1.1343,
+      "step": 2440
+    },
+    {
+      "epoch": 1.796846351301797,
+      "grad_norm": 0.7171908020973206,
+      "learning_rate": 0.0002,
+      "loss": 1.0764,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8041804180418042,
+      "grad_norm": 0.550658643245697,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 2460
+    },
+    {
+      "epoch": 1.8115144847818114,
+      "grad_norm": 0.4075568914413452,
+      "learning_rate": 0.0002,
+      "loss": 0.9975,
+      "step": 2470
+    },
+    {
+      "epoch": 1.818848551521819,
+      "grad_norm": 0.3790127635002136,
+      "learning_rate": 0.0002,
+      "loss": 1.0935,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8261826182618262,
+      "grad_norm": 0.3576384484767914,
+      "learning_rate": 0.0002,
+      "loss": 0.9839,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8335166850018334,
+      "grad_norm": 0.3919370770454407,
+      "learning_rate": 0.0002,
+      "loss": 1.1369,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8408507517418409,
+      "grad_norm": 0.485083669424057,
+      "learning_rate": 0.0002,
+      "loss": 0.9985,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8481848184818483,
+      "grad_norm": 0.4564347565174103,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8555188852218554,
+      "grad_norm": 0.3613106608390808,
+      "learning_rate": 0.0002,
+      "loss": 1.0944,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8628529519618628,
+      "grad_norm": 0.39600759744644165,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8701870187018703,
+      "grad_norm": 1.123499870300293,
+      "learning_rate": 0.0002,
+      "loss": 0.9453,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8775210854418776,
+      "grad_norm": 0.4612680673599243,
+      "learning_rate": 0.0002,
+      "loss": 1.0635,
+      "step": 2560
+    },
+    {
+      "epoch": 1.8848551521818848,
+      "grad_norm": 0.42745399475097656,
+      "learning_rate": 0.0002,
+      "loss": 1.0087,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8921892189218923,
+      "grad_norm": 0.4055580198764801,
+      "learning_rate": 0.0002,
+      "loss": 1.0102,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8995232856618995,
+      "grad_norm": 0.44174644351005554,
+      "learning_rate": 0.0002,
+      "loss": 1.0177,
+      "step": 2590
+    },
+    {
+      "epoch": 1.9068573524019068,
+      "grad_norm": 1.0228385925292969,
+      "learning_rate": 0.0002,
+      "loss": 0.9886,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9141914191419143,
+      "grad_norm": 0.3496396243572235,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9215254858819215,
+      "grad_norm": 0.4191173017024994,
+      "learning_rate": 0.0002,
+      "loss": 1.0955,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9288595526219288,
+      "grad_norm": 0.6778554916381836,
+      "learning_rate": 0.0002,
+      "loss": 1.0943,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9361936193619362,
+      "grad_norm": 0.41992834210395813,
+      "learning_rate": 0.0002,
+      "loss": 1.0594,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9435276861019435,
+      "grad_norm": 0.8760401010513306,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9508617528419507,
+      "grad_norm": 0.44049209356307983,
+      "learning_rate": 0.0002,
+      "loss": 1.0379,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9581958195819582,
+      "grad_norm": 0.5651928782463074,
+      "learning_rate": 0.0002,
+      "loss": 1.1008,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9655298863219657,
+      "grad_norm": 0.5292727947235107,
+      "learning_rate": 0.0002,
+      "loss": 1.1317,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9728639530619727,
+      "grad_norm": 0.6012240648269653,
+      "learning_rate": 0.0002,
+      "loss": 1.1328,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9801980198019802,
+      "grad_norm": 0.3945149779319763,
+      "learning_rate": 0.0002,
+      "loss": 1.0683,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9875320865419877,
+      "grad_norm": 0.5732627511024475,
+      "learning_rate": 0.0002,
+      "loss": 1.0155,
+      "step": 2710
+    },
+    {
+      "epoch": 1.994866153281995,
+      "grad_norm": 0.3963361084461212,
+      "learning_rate": 0.0002,
+      "loss": 0.9857,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1534006595611572,
+      "eval_runtime": 32.7541,
+      "eval_samples_per_second": 13.159,
+      "eval_steps_per_second": 1.649,
+      "step": 2727
+    },
+    {
+      "epoch": 2.002200220022002,
+      "grad_norm": 0.48628315329551697,
+      "learning_rate": 0.0002,
+      "loss": 0.9624,
+      "step": 2730
+    },
+    {
+      "epoch": 2.0095342867620096,
+      "grad_norm": 0.413875013589859,
+      "learning_rate": 0.0002,
+      "loss": 0.9603,
+      "step": 2740
+    },
+    {
+      "epoch": 2.0168683535020167,
+      "grad_norm": 0.4988735616207123,
+      "learning_rate": 0.0002,
+      "loss": 0.965,
+      "step": 2750
+    },
+    {
+      "epoch": 2.024202420242024,
+      "grad_norm": 0.5634812712669373,
+      "learning_rate": 0.0002,
+      "loss": 0.9677,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0315364869820316,
+      "grad_norm": 0.48302653431892395,
+      "learning_rate": 0.0002,
+      "loss": 0.9547,
+      "step": 2770
+    },
+    {
+      "epoch": 2.038870553722039,
+      "grad_norm": 0.49914175271987915,
+      "learning_rate": 0.0002,
+      "loss": 0.9346,
+      "step": 2780
+    },
+    {
+      "epoch": 2.046204620462046,
+      "grad_norm": 1.14039945602417,
+      "learning_rate": 0.0002,
+      "loss": 0.904,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0535386872020536,
+      "grad_norm": 0.6359720826148987,
+      "learning_rate": 0.0002,
+      "loss": 0.9588,
+      "step": 2800
+    },
+    {
+      "epoch": 2.060872753942061,
+      "grad_norm": 0.4589158296585083,
+      "learning_rate": 0.0002,
+      "loss": 0.9031,
+      "step": 2810
+    },
+    {
+      "epoch": 2.068206820682068,
+      "grad_norm": 0.46255481243133545,
+      "learning_rate": 0.0002,
+      "loss": 0.9438,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0755408874220755,
+      "grad_norm": 0.6232137680053711,
+      "learning_rate": 0.0002,
+      "loss": 0.9464,
+      "step": 2830
+    },
+    {
+      "epoch": 2.082874954162083,
+      "grad_norm": 0.41042178869247437,
+      "learning_rate": 0.0002,
+      "loss": 0.8978,
+      "step": 2840
+    },
+    {
+      "epoch": 2.09020902090209,
+      "grad_norm": 0.5334428548812866,
+      "learning_rate": 0.0002,
+      "loss": 0.8516,
+      "step": 2850
+    },
+    {
+      "epoch": 2.0975430876420975,
+      "grad_norm": 0.8270058631896973,
+      "learning_rate": 0.0002,
+      "loss": 0.9313,
+      "step": 2860
+    },
+    {
+      "epoch": 2.104877154382105,
+      "grad_norm": 0.6624533534049988,
+      "learning_rate": 0.0002,
+      "loss": 1.0064,
+      "step": 2870
+    },
+    {
+      "epoch": 2.112211221122112,
+      "grad_norm": 0.5448863506317139,
+      "learning_rate": 0.0002,
+      "loss": 0.9196,
+      "step": 2880
+    },
+    {
+      "epoch": 2.1195452878621195,
+      "grad_norm": 0.621482789516449,
+      "learning_rate": 0.0002,
+      "loss": 0.887,
+      "step": 2890
+    },
+    {
+      "epoch": 2.126879354602127,
+      "grad_norm": 0.4556255340576172,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1342134213421344,
+      "grad_norm": 0.4620579183101654,
+      "learning_rate": 0.0002,
+      "loss": 0.9323,
+      "step": 2910
+    },
+    {
+      "epoch": 2.1415474880821415,
+      "grad_norm": 0.9602415561676025,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 2920
+    },
+    {
+      "epoch": 2.148881554822149,
+      "grad_norm": 0.587943971157074,
+      "learning_rate": 0.0002,
+      "loss": 0.8826,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1562156215621564,
+      "grad_norm": 0.5121372938156128,
+      "learning_rate": 0.0002,
+      "loss": 0.971,
+      "step": 2940
+    },
+    {
+      "epoch": 2.1635496883021634,
+      "grad_norm": 0.49424484372138977,
+      "learning_rate": 0.0002,
+      "loss": 0.8751,
+      "step": 2950
+    },
+    {
+      "epoch": 2.170883755042171,
+      "grad_norm": 0.6312560439109802,
+      "learning_rate": 0.0002,
+      "loss": 0.8674,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1782178217821784,
+      "grad_norm": 0.5235576629638672,
+      "learning_rate": 0.0002,
+      "loss": 0.9791,
+      "step": 2970
+    },
+    {
+      "epoch": 2.1855518885221854,
+      "grad_norm": 0.5868439674377441,
+      "learning_rate": 0.0002,
+      "loss": 0.9706,
+      "step": 2980
+    },
+    {
+      "epoch": 2.192885955262193,
+      "grad_norm": 0.42302873730659485,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 2990
+    },
+    {
+      "epoch": 2.2002200220022003,
+      "grad_norm": 0.5097725987434387,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 3000
+    },
+    {
+      "epoch": 2.2075540887422074,
+      "grad_norm": 0.5091572403907776,
+      "learning_rate": 0.0002,
+      "loss": 0.9239,
+      "step": 3010
+    },
+    {
+      "epoch": 2.214888155482215,
+      "grad_norm": 0.49433162808418274,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 3020
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.5577368140220642,
+      "learning_rate": 0.0002,
+      "loss": 0.9734,
+      "step": 3030
+    },
+    {
+      "epoch": 2.2295562889622293,
+      "grad_norm": 0.6177583932876587,
+      "learning_rate": 0.0002,
+      "loss": 0.9033,
+      "step": 3040
+    },
+    {
+      "epoch": 2.236890355702237,
+      "grad_norm": 0.5256719589233398,
+      "learning_rate": 0.0002,
+      "loss": 0.9882,
+      "step": 3050
+    },
+    {
+      "epoch": 2.2442244224422443,
+      "grad_norm": 0.5001118183135986,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2515584891822513,
+      "grad_norm": 0.5721249580383301,
+      "learning_rate": 0.0002,
+      "loss": 0.8718,
+      "step": 3070
+    },
+    {
+      "epoch": 2.258892555922259,
+      "grad_norm": 0.5325384140014648,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 3080
+    },
+    {
+      "epoch": 2.2662266226622663,
+      "grad_norm": 0.5719189047813416,
+      "learning_rate": 0.0002,
+      "loss": 0.9843,
+      "step": 3090
+    },
+    {
+      "epoch": 2.2735606894022737,
+      "grad_norm": 0.6337835788726807,
+      "learning_rate": 0.0002,
+      "loss": 0.8633,
+      "step": 3100
+    },
+    {
+      "epoch": 2.2808947561422808,
+      "grad_norm": 0.5381836891174316,
+      "learning_rate": 0.0002,
+      "loss": 0.9962,
+      "step": 3110
+    },
+    {
+      "epoch": 2.2882288228822882,
+      "grad_norm": 0.5408531427383423,
+      "learning_rate": 0.0002,
+      "loss": 0.8265,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2955628896222957,
+      "grad_norm": 0.43705281615257263,
+      "learning_rate": 0.0002,
+      "loss": 1.0325,
+      "step": 3130
+    },
+    {
+      "epoch": 2.3028969563623027,
+      "grad_norm": 0.6454030275344849,
+      "learning_rate": 0.0002,
+      "loss": 0.9388,
+      "step": 3140
+    },
+    {
+      "epoch": 2.31023102310231,
+      "grad_norm": 0.686030387878418,
+      "learning_rate": 0.0002,
+      "loss": 0.954,
+      "step": 3150
+    },
+    {
+      "epoch": 2.3175650898423177,
+      "grad_norm": 0.5123633146286011,
+      "learning_rate": 0.0002,
+      "loss": 0.9403,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3248991565823247,
+      "grad_norm": 0.842506468296051,
+      "learning_rate": 0.0002,
+      "loss": 0.8834,
+      "step": 3170
+    },
+    {
+      "epoch": 2.332233223322332,
+      "grad_norm": 0.5193818807601929,
+      "learning_rate": 0.0002,
+      "loss": 1.0497,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3395672900623397,
+      "grad_norm": 0.5634409189224243,
+      "learning_rate": 0.0002,
+      "loss": 0.9473,
+      "step": 3190
+    },
+    {
+      "epoch": 2.3469013568023467,
+      "grad_norm": 0.6475534439086914,
+      "learning_rate": 0.0002,
+      "loss": 0.8499,
+      "step": 3200
+    },
+    {
+      "epoch": 2.354235423542354,
+      "grad_norm": 1.1503914594650269,
+      "learning_rate": 0.0002,
+      "loss": 0.874,
+      "step": 3210
+    },
+    {
+      "epoch": 2.3615694902823616,
+      "grad_norm": 0.7234905362129211,
+      "learning_rate": 0.0002,
+      "loss": 0.9762,
+      "step": 3220
+    },
+    {
+      "epoch": 2.368903557022369,
+      "grad_norm": 0.664903461933136,
+      "learning_rate": 0.0002,
+      "loss": 0.9007,
+      "step": 3230
+    },
+    {
+      "epoch": 2.376237623762376,
+      "grad_norm": 0.5453006625175476,
+      "learning_rate": 0.0002,
+      "loss": 0.9987,
+      "step": 3240
+    },
+    {
+      "epoch": 2.3835716905023836,
+      "grad_norm": 0.6256654262542725,
+      "learning_rate": 0.0002,
+      "loss": 0.9742,
+      "step": 3250
+    },
+    {
+      "epoch": 2.390905757242391,
+      "grad_norm": 0.5166565179824829,
+      "learning_rate": 0.0002,
+      "loss": 0.9922,
+      "step": 3260
+    },
+    {
+      "epoch": 2.398239823982398,
+      "grad_norm": 0.5699098110198975,
+      "learning_rate": 0.0002,
+      "loss": 0.927,
+      "step": 3270
+    },
+    {
+      "epoch": 2.4055738907224056,
+      "grad_norm": 0.4472540020942688,
+      "learning_rate": 0.0002,
+      "loss": 0.8878,
+      "step": 3280
+    },
+    {
+      "epoch": 2.412907957462413,
+      "grad_norm": 0.6790403127670288,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 3290
+    },
+    {
+      "epoch": 2.42024202420242,
+      "grad_norm": 0.5182185173034668,
+      "learning_rate": 0.0002,
+      "loss": 0.972,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4275760909424275,
+      "grad_norm": 0.564647912979126,
+      "learning_rate": 0.0002,
+      "loss": 0.9775,
+      "step": 3310
+    },
+    {
+      "epoch": 2.434910157682435,
+      "grad_norm": 0.5625313520431519,
+      "learning_rate": 0.0002,
+      "loss": 1.072,
+      "step": 3320
+    },
+    {
+      "epoch": 2.442244224422442,
+      "grad_norm": 0.7496559619903564,
+      "learning_rate": 0.0002,
+      "loss": 0.8798,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4495782911624495,
+      "grad_norm": 0.4779128134250641,
+      "learning_rate": 0.0002,
+      "loss": 0.868,
+      "step": 3340
+    },
+    {
+      "epoch": 2.456912357902457,
+      "grad_norm": 0.578093409538269,
+      "learning_rate": 0.0002,
+      "loss": 1.0316,
+      "step": 3350
+    },
+    {
+      "epoch": 2.4642464246424645,
+      "grad_norm": 0.5456080436706543,
+      "learning_rate": 0.0002,
+      "loss": 0.9282,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4715804913824715,
+      "grad_norm": 0.4769273102283478,
+      "learning_rate": 0.0002,
+      "loss": 0.8409,
+      "step": 3370
+    },
+    {
+      "epoch": 2.478914558122479,
+      "grad_norm": 0.5608189702033997,
+      "learning_rate": 0.0002,
+      "loss": 0.9312,
+      "step": 3380
+    },
+    {
+      "epoch": 2.4862486248624864,
+      "grad_norm": 0.5590165853500366,
+      "learning_rate": 0.0002,
+      "loss": 0.9934,
+      "step": 3390
+    },
+    {
+      "epoch": 2.4935826916024935,
+      "grad_norm": 0.801306962966919,
+      "learning_rate": 0.0002,
+      "loss": 1.025,
+      "step": 3400
+    },
+    {
+      "epoch": 2.500916758342501,
+      "grad_norm": 0.6045624613761902,
+      "learning_rate": 0.0002,
+      "loss": 0.9049,
+      "step": 3410
+    },
+    {
+      "epoch": 2.5082508250825084,
+      "grad_norm": 0.5735858082771301,
+      "learning_rate": 0.0002,
+      "loss": 0.944,
+      "step": 3420
+    },
+    {
+      "epoch": 2.5155848918225154,
+      "grad_norm": 0.6827309131622314,
+      "learning_rate": 0.0002,
+      "loss": 0.9846,
+      "step": 3430
+    },
+    {
+      "epoch": 2.522918958562523,
+      "grad_norm": 0.5702602863311768,
+      "learning_rate": 0.0002,
+      "loss": 0.9789,
+      "step": 3440
+    },
+    {
+      "epoch": 2.5302530253025304,
+      "grad_norm": 0.6674721240997314,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5375870920425374,
+      "grad_norm": 0.5635907649993896,
+      "learning_rate": 0.0002,
+      "loss": 0.914,
+      "step": 3460
+    },
+    {
+      "epoch": 2.544921158782545,
+      "grad_norm": 0.42737770080566406,
+      "learning_rate": 0.0002,
+      "loss": 0.8398,
+      "step": 3470
+    },
+    {
+      "epoch": 2.5522552255225524,
+      "grad_norm": 0.6720691919326782,
+      "learning_rate": 0.0002,
+      "loss": 0.9474,
+      "step": 3480
+    },
+    {
+      "epoch": 2.55958929226256,
+      "grad_norm": 0.8917084336280823,
+      "learning_rate": 0.0002,
+      "loss": 0.8637,
+      "step": 3490
+    },
+    {
+      "epoch": 2.566923359002567,
+      "grad_norm": 0.5134549140930176,
+      "learning_rate": 0.0002,
+      "loss": 0.9257,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5742574257425743,
+      "grad_norm": 0.4951367974281311,
+      "learning_rate": 0.0002,
+      "loss": 0.9362,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5815914924825814,
+      "grad_norm": 0.9438204765319824,
+      "learning_rate": 0.0002,
+      "loss": 0.9184,
+      "step": 3520
+    },
+    {
+      "epoch": 2.588925559222589,
+      "grad_norm": 0.6024714708328247,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5962596259625963,
+      "grad_norm": 0.5248535871505737,
+      "learning_rate": 0.0002,
+      "loss": 0.9298,
+      "step": 3540
+    },
+    {
+      "epoch": 2.6035936927026038,
+      "grad_norm": 0.8677568435668945,
+      "learning_rate": 0.0002,
+      "loss": 0.941,
+      "step": 3550
+    },
+    {
+      "epoch": 2.610927759442611,
+      "grad_norm": 0.82008296251297,
+      "learning_rate": 0.0002,
+      "loss": 0.9253,
+      "step": 3560
+    },
+    {
+      "epoch": 2.6182618261826183,
+      "grad_norm": 0.4724634885787964,
+      "learning_rate": 0.0002,
+      "loss": 0.8429,
+      "step": 3570
+    },
+    {
+      "epoch": 2.6255958929226257,
+      "grad_norm": 0.5434244275093079,
+      "learning_rate": 0.0002,
+      "loss": 0.9058,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6329299596626328,
+      "grad_norm": 0.4948740005493164,
+      "learning_rate": 0.0002,
+      "loss": 0.9379,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6402640264026402,
+      "grad_norm": 0.42109328508377075,
+      "learning_rate": 0.0002,
+      "loss": 0.8718,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6475980931426477,
+      "grad_norm": 0.7979786396026611,
+      "learning_rate": 0.0002,
+      "loss": 0.9809,
+      "step": 3610
+    },
+    {
+      "epoch": 2.654932159882655,
+      "grad_norm": 0.6345919370651245,
+      "learning_rate": 0.0002,
+      "loss": 0.9229,
+      "step": 3620
+    },
+    {
+      "epoch": 2.662266226622662,
+      "grad_norm": 0.4971671402454376,
+      "learning_rate": 0.0002,
+      "loss": 0.8506,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6696002933626697,
+      "grad_norm": 0.6467748284339905,
+      "learning_rate": 0.0002,
+      "loss": 0.8054,
+      "step": 3640
+    },
+    {
+      "epoch": 2.6769343601026767,
+      "grad_norm": 0.4240160286426544,
+      "learning_rate": 0.0002,
+      "loss": 0.9277,
+      "step": 3650
+    },
+    {
+      "epoch": 2.684268426842684,
+      "grad_norm": 0.5179754495620728,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 3660
+    },
+    {
+      "epoch": 2.6916024935826917,
+      "grad_norm": 0.754012405872345,
+      "learning_rate": 0.0002,
+      "loss": 0.9221,
+      "step": 3670
+    },
+    {
+      "epoch": 2.698936560322699,
+      "grad_norm": 0.5141299962997437,
+      "learning_rate": 0.0002,
+      "loss": 0.9194,
+      "step": 3680
+    },
+    {
+      "epoch": 2.706270627062706,
+      "grad_norm": 0.5737819075584412,
+      "learning_rate": 0.0002,
+      "loss": 0.9495,
+      "step": 3690
+    },
+    {
+      "epoch": 2.7136046938027136,
+      "grad_norm": 0.5887577533721924,
+      "learning_rate": 0.0002,
+      "loss": 1.0162,
+      "step": 3700
+    },
+    {
+      "epoch": 2.720938760542721,
+      "grad_norm": 0.6740471720695496,
+      "learning_rate": 0.0002,
+      "loss": 0.9169,
+      "step": 3710
+    },
+    {
+      "epoch": 2.728272827282728,
+      "grad_norm": 0.5879453420639038,
+      "learning_rate": 0.0002,
+      "loss": 0.9297,
+      "step": 3720
+    },
+    {
+      "epoch": 2.7356068940227356,
+      "grad_norm": 0.4858354926109314,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 3730
+    },
+    {
+      "epoch": 2.742940960762743,
+      "grad_norm": 0.5489001870155334,
+      "learning_rate": 0.0002,
+      "loss": 0.9308,
+      "step": 3740
+    },
+    {
+      "epoch": 2.7502750275027505,
+      "grad_norm": 0.8187092542648315,
+      "learning_rate": 0.0002,
+      "loss": 0.894,
+      "step": 3750
+    },
+    {
+      "epoch": 2.7576090942427576,
+      "grad_norm": 0.5666626691818237,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 3760
+    },
+    {
+      "epoch": 2.764943160982765,
+      "grad_norm": 0.5377066135406494,
+      "learning_rate": 0.0002,
+      "loss": 1.0059,
+      "step": 3770
+    },
+    {
+      "epoch": 2.772277227722772,
+      "grad_norm": 0.566330075263977,
+      "learning_rate": 0.0002,
+      "loss": 0.9132,
+      "step": 3780
+    },
+    {
+      "epoch": 2.7796112944627795,
+      "grad_norm": 0.5522832870483398,
+      "learning_rate": 0.0002,
+      "loss": 0.9415,
+      "step": 3790
+    },
+    {
+      "epoch": 2.786945361202787,
+      "grad_norm": 0.5668695569038391,
+      "learning_rate": 0.0002,
+      "loss": 0.8816,
+      "step": 3800
+    },
+    {
+      "epoch": 2.7942794279427945,
+      "grad_norm": 0.7566602826118469,
+      "learning_rate": 0.0002,
+      "loss": 0.8885,
+      "step": 3810
+    },
+    {
+      "epoch": 2.8016134946828015,
+      "grad_norm": 0.5603684782981873,
+      "learning_rate": 0.0002,
+      "loss": 0.8598,
+      "step": 3820
+    },
+    {
+      "epoch": 2.808947561422809,
+      "grad_norm": 0.49122217297554016,
+      "learning_rate": 0.0002,
+      "loss": 0.9602,
+      "step": 3830
+    },
+    {
+      "epoch": 2.816281628162816,
+      "grad_norm": 0.6798251867294312,
+      "learning_rate": 0.0002,
+      "loss": 0.9738,
+      "step": 3840
+    },
+    {
+      "epoch": 2.8236156949028235,
+      "grad_norm": 0.6097991466522217,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 3850
+    },
+    {
+      "epoch": 2.830949761642831,
+      "grad_norm": 0.6675726175308228,
+      "learning_rate": 0.0002,
+      "loss": 0.8672,
+      "step": 3860
+    },
+    {
+      "epoch": 2.8382838283828384,
+      "grad_norm": 0.9223952889442444,
+      "learning_rate": 0.0002,
+      "loss": 0.9324,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8456178951228455,
+      "grad_norm": 0.6020799875259399,
+      "learning_rate": 0.0002,
+      "loss": 0.8767,
+      "step": 3880
+    },
+    {
+      "epoch": 2.852951961862853,
+      "grad_norm": 0.5206381678581238,
+      "learning_rate": 0.0002,
+      "loss": 0.9148,
+      "step": 3890
+    },
+    {
+      "epoch": 2.8602860286028604,
+      "grad_norm": 0.6268777251243591,
+      "learning_rate": 0.0002,
+      "loss": 0.9479,
+      "step": 3900
+    },
+    {
+      "epoch": 2.8676200953428674,
+      "grad_norm": 1.1583497524261475,
+      "learning_rate": 0.0002,
+      "loss": 0.9409,
+      "step": 3910
+    },
+    {
+      "epoch": 2.874954162082875,
+      "grad_norm": 0.7263903021812439,
+      "learning_rate": 0.0002,
+      "loss": 0.895,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8822882288228824,
+      "grad_norm": 0.5369910001754761,
+      "learning_rate": 0.0002,
+      "loss": 0.8786,
+      "step": 3930
+    },
+    {
+      "epoch": 2.88962229556289,
+      "grad_norm": 0.7298350930213928,
+      "learning_rate": 0.0002,
+      "loss": 1.0015,
+      "step": 3940
+    },
+    {
+      "epoch": 2.896956362302897,
+      "grad_norm": 0.577012836933136,
+      "learning_rate": 0.0002,
+      "loss": 0.979,
+      "step": 3950
+    },
+    {
+      "epoch": 2.9042904290429044,
+      "grad_norm": 0.5859594345092773,
+      "learning_rate": 0.0002,
+      "loss": 0.9716,
+      "step": 3960
+    },
+    {
+      "epoch": 2.9116244957829114,
+      "grad_norm": 0.47176122665405273,
+      "learning_rate": 0.0002,
+      "loss": 0.8772,
+      "step": 3970
+    },
+    {
+      "epoch": 2.918958562522919,
+      "grad_norm": 0.9699620604515076,
+      "learning_rate": 0.0002,
+      "loss": 0.8997,
+      "step": 3980
+    },
+    {
+      "epoch": 2.9262926292629263,
+      "grad_norm": 0.7908747792243958,
+      "learning_rate": 0.0002,
+      "loss": 0.9057,
+      "step": 3990
+    },
+    {
+      "epoch": 2.933626696002934,
+      "grad_norm": 0.5777379274368286,
+      "learning_rate": 0.0002,
+      "loss": 0.9462,
+      "step": 4000
+    },
+    {
+      "epoch": 2.940960762742941,
+      "grad_norm": 0.599288284778595,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9482948294829483,
+      "grad_norm": 0.5232274532318115,
+      "learning_rate": 0.0002,
+      "loss": 0.9812,
+      "step": 4020
+    },
+    {
+      "epoch": 2.9556288962229558,
+      "grad_norm": 0.6395137310028076,
+      "learning_rate": 0.0002,
+      "loss": 0.96,
+      "step": 4030
+    },
+    {
+      "epoch": 2.962962962962963,
+      "grad_norm": 0.589260458946228,
+      "learning_rate": 0.0002,
+      "loss": 0.9813,
+      "step": 4040
+    },
+    {
+      "epoch": 2.9702970297029703,
+      "grad_norm": 0.5699581503868103,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 4050
+    },
+    {
+      "epoch": 2.9776310964429777,
+      "grad_norm": 0.528468132019043,
+      "learning_rate": 0.0002,
+      "loss": 0.9585,
+      "step": 4060
+    },
+    {
+      "epoch": 2.984965163182985,
+      "grad_norm": 0.4804670512676239,
+      "learning_rate": 0.0002,
+      "loss": 0.9164,
+      "step": 4070
+    },
+    {
+      "epoch": 2.9922992299229922,
+      "grad_norm": 1.1918889284133911,
+      "learning_rate": 0.0002,
+      "loss": 0.9771,
+      "step": 4080
+    },
+    {
+      "epoch": 2.9996332966629997,
+      "grad_norm": 0.5479103326797485,
+      "learning_rate": 0.0002,
+      "loss": 0.9178,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9996332966629997,
+      "eval_loss": 1.1642853021621704,
+      "eval_runtime": 32.7511,
+      "eval_samples_per_second": 13.16,
+      "eval_steps_per_second": 1.649,
+      "step": 4090
+    },
+    {
+      "epoch": 3.006967363403007,
+      "grad_norm": 0.7430027723312378,
+      "learning_rate": 0.0002,
+      "loss": 0.7981,
+      "step": 4100
+    },
+    {
+      "epoch": 3.014301430143014,
+      "grad_norm": 0.6293647289276123,
+      "learning_rate": 0.0002,
+      "loss": 0.7871,
+      "step": 4110
+    },
+    {
+      "epoch": 3.0216354968830217,
+      "grad_norm": 0.6191329956054688,
+      "learning_rate": 0.0002,
+      "loss": 0.78,
+      "step": 4120
+    },
+    {
+      "epoch": 3.028969563623029,
+      "grad_norm": 0.7959313988685608,
+      "learning_rate": 0.0002,
+      "loss": 0.7618,
+      "step": 4130
+    },
+    {
+      "epoch": 3.036303630363036,
+      "grad_norm": 0.5956351161003113,
+      "learning_rate": 0.0002,
+      "loss": 0.8039,
+      "step": 4140
+    },
+    {
+      "epoch": 3.0436376971030437,
+      "grad_norm": 0.670383632183075,
+      "learning_rate": 0.0002,
+      "loss": 0.7477,
+      "step": 4150
+    },
+    {
+      "epoch": 3.050971763843051,
+      "grad_norm": 0.6414518356323242,
+      "learning_rate": 0.0002,
+      "loss": 0.7984,
+      "step": 4160
+    },
+    {
+      "epoch": 3.058305830583058,
+      "grad_norm": 0.7928852438926697,
+      "learning_rate": 0.0002,
+      "loss": 0.7369,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0656398973230656,
+      "grad_norm": 0.6211121082305908,
+      "learning_rate": 0.0002,
+      "loss": 0.7914,
+      "step": 4180
+    },
+    {
+      "epoch": 3.072973964063073,
+      "grad_norm": 0.6237057447433472,
+      "learning_rate": 0.0002,
+      "loss": 0.7365,
+      "step": 4190
+    },
+    {
+      "epoch": 3.08030803080308,
+      "grad_norm": 0.6522233486175537,
+      "learning_rate": 0.0002,
+      "loss": 0.702,
+      "step": 4200
+    },
+    {
+      "epoch": 3.0876420975430876,
+      "grad_norm": 0.9396848678588867,
+      "learning_rate": 0.0002,
+      "loss": 0.7646,
+      "step": 4210
+    },
+    {
+      "epoch": 3.094976164283095,
+      "grad_norm": 0.8003010749816895,
+      "learning_rate": 0.0002,
+      "loss": 0.7559,
+      "step": 4220
+    },
+    {
+      "epoch": 3.102310231023102,
+      "grad_norm": 0.6733810305595398,
+      "learning_rate": 0.0002,
+      "loss": 0.711,
+      "step": 4230
+    },
+    {
+      "epoch": 3.1096442977631096,
+      "grad_norm": 0.6365828514099121,
+      "learning_rate": 0.0002,
+      "loss": 0.696,
+      "step": 4240
+    },
+    {
+      "epoch": 3.116978364503117,
+      "grad_norm": 1.0805548429489136,
+      "learning_rate": 0.0002,
+      "loss": 0.8362,
+      "step": 4250
+    },
+    {
+      "epoch": 3.1243124312431245,
+      "grad_norm": 0.7262141108512878,
+      "learning_rate": 0.0002,
+      "loss": 0.7651,
+      "step": 4260
+    },
+    {
+      "epoch": 3.1316464979831315,
+      "grad_norm": 0.5500539541244507,
+      "learning_rate": 0.0002,
+      "loss": 0.7304,
+      "step": 4270
+    },
+    {
+      "epoch": 3.138980564723139,
+      "grad_norm": 0.793912947177887,
+      "learning_rate": 0.0002,
+      "loss": 0.7721,
+      "step": 4280
+    },
+    {
+      "epoch": 3.1463146314631465,
+      "grad_norm": 1.2540518045425415,
+      "learning_rate": 0.0002,
+      "loss": 0.7708,
+      "step": 4290
+    },
+    {
+      "epoch": 3.1536486982031535,
+      "grad_norm": 0.7020077705383301,
+      "learning_rate": 0.0002,
+      "loss": 0.782,
+      "step": 4300
+    },
+    {
+      "epoch": 3.160982764943161,
+      "grad_norm": 0.5111123323440552,
+      "learning_rate": 0.0002,
+      "loss": 0.7253,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1683168316831685,
+      "grad_norm": 0.7172090411186218,
+      "learning_rate": 0.0002,
+      "loss": 0.8159,
+      "step": 4320
+    },
+    {
+      "epoch": 3.1756508984231755,
+      "grad_norm": 0.6343168616294861,
+      "learning_rate": 0.0002,
+      "loss": 0.6962,
+      "step": 4330
+    },
+    {
+      "epoch": 3.182984965163183,
+      "grad_norm": 0.9563672542572021,
+      "learning_rate": 0.0002,
+      "loss": 0.7938,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1903190319031904,
+      "grad_norm": 1.0225574970245361,
+      "learning_rate": 0.0002,
+      "loss": 0.7385,
+      "step": 4350
+    },
+    {
+      "epoch": 3.1976530986431975,
+      "grad_norm": 1.1633386611938477,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 4360
+    },
+    {
+      "epoch": 3.204987165383205,
+      "grad_norm": 0.8915148973464966,
+      "learning_rate": 0.0002,
+      "loss": 0.7259,
+      "step": 4370
+    },
+    {
+      "epoch": 3.2123212321232124,
+      "grad_norm": 0.9156812429428101,
+      "learning_rate": 0.0002,
+      "loss": 0.8061,
+      "step": 4380
+    },
+    {
+      "epoch": 3.21965529886322,
+      "grad_norm": 0.6363258957862854,
+      "learning_rate": 0.0002,
+      "loss": 0.8189,
+      "step": 4390
+    },
+    {
+      "epoch": 3.226989365603227,
+      "grad_norm": 0.579099178314209,
+      "learning_rate": 0.0002,
+      "loss": 0.7996,
+      "step": 4400
+    },
+    {
+      "epoch": 3.2343234323432344,
+      "grad_norm": 0.8778146505355835,
+      "learning_rate": 0.0002,
+      "loss": 0.8592,
+      "step": 4410
+    },
+    {
+      "epoch": 3.241657499083242,
+      "grad_norm": 0.8356770873069763,
+      "learning_rate": 0.0002,
+      "loss": 0.8281,
+      "step": 4420
+    },
+    {
+      "epoch": 3.248991565823249,
+      "grad_norm": 0.702032208442688,
+      "learning_rate": 0.0002,
+      "loss": 0.8484,
+      "step": 4430
+    },
+    {
+      "epoch": 3.2563256325632564,
+      "grad_norm": 0.6386539340019226,
+      "learning_rate": 0.0002,
+      "loss": 0.7227,
+      "step": 4440
+    },
+    {
+      "epoch": 3.263659699303264,
+      "grad_norm": 0.7008408904075623,
+      "learning_rate": 0.0002,
+      "loss": 0.8374,
+      "step": 4450
+    },
+    {
+      "epoch": 3.270993766043271,
+      "grad_norm": 0.9556332230567932,
+      "learning_rate": 0.0002,
+      "loss": 0.7572,
+      "step": 4460
+    },
+    {
+      "epoch": 3.2783278327832783,
+      "grad_norm": 0.5667835474014282,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 4470
+    },
+    {
+      "epoch": 3.285661899523286,
+      "grad_norm": 0.8239172697067261,
+      "learning_rate": 0.0002,
+      "loss": 0.8152,
+      "step": 4480
+    },
+    {
+      "epoch": 3.292995966263293,
+      "grad_norm": 0.7045050859451294,
+      "learning_rate": 0.0002,
+      "loss": 0.756,
+      "step": 4490
+    },
+    {
+      "epoch": 3.3003300330033003,
+      "grad_norm": 0.7131434082984924,
+      "learning_rate": 0.0002,
+      "loss": 0.7655,
+      "step": 4500
+    },
+    {
+      "epoch": 3.3076640997433078,
+      "grad_norm": 0.6924910545349121,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 4510
+    },
+    {
+      "epoch": 3.3149981664833152,
+      "grad_norm": 0.8945356607437134,
+      "learning_rate": 0.0002,
+      "loss": 0.736,
+      "step": 4520
+    },
+    {
+      "epoch": 3.3223322332233223,
+      "grad_norm": 0.6546903252601624,
+      "learning_rate": 0.0002,
+      "loss": 0.7575,
+      "step": 4530
+    },
+    {
+      "epoch": 3.3296662999633297,
+      "grad_norm": 0.8206679224967957,
+      "learning_rate": 0.0002,
+      "loss": 0.7893,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3370003667033368,
+      "grad_norm": 0.6482203602790833,
+      "learning_rate": 0.0002,
+      "loss": 0.7502,
+      "step": 4550
+    },
+    {
+      "epoch": 3.3443344334433442,
+      "grad_norm": 0.7558760046958923,
+      "learning_rate": 0.0002,
+      "loss": 0.8172,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3516685001833517,
+      "grad_norm": 0.7794756889343262,
+      "learning_rate": 0.0002,
+      "loss": 0.744,
+      "step": 4570
+    },
+    {
+      "epoch": 3.359002566923359,
+      "grad_norm": 0.7382805943489075,
+      "learning_rate": 0.0002,
+      "loss": 0.7385,
+      "step": 4580
+    },
+    {
+      "epoch": 3.366336633663366,
+      "grad_norm": 0.5912511944770813,
+      "learning_rate": 0.0002,
+      "loss": 0.8511,
+      "step": 4590
+    },
+    {
+      "epoch": 3.3736707004033737,
+      "grad_norm": 0.7444885969161987,
+      "learning_rate": 0.0002,
+      "loss": 0.8272,
+      "step": 4600
+    },
+    {
+      "epoch": 3.381004767143381,
+      "grad_norm": 0.7354922890663147,
+      "learning_rate": 0.0002,
+      "loss": 0.7927,
+      "step": 4610
+    },
+    {
+      "epoch": 3.388338833883388,
+      "grad_norm": 0.7685934901237488,
+      "learning_rate": 0.0002,
+      "loss": 0.7183,
+      "step": 4620
+    },
+    {
+      "epoch": 3.3956729006233957,
+      "grad_norm": 0.61041259765625,
+      "learning_rate": 0.0002,
+      "loss": 0.7436,
+      "step": 4630
+    },
+    {
+      "epoch": 3.403006967363403,
+      "grad_norm": 0.6820451021194458,
+      "learning_rate": 0.0002,
+      "loss": 0.7661,
+      "step": 4640
+    },
+    {
+      "epoch": 3.41034103410341,
+      "grad_norm": 0.5819534063339233,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 4650
+    },
+    {
+      "epoch": 3.4176751008434176,
+      "grad_norm": 0.705410897731781,
+      "learning_rate": 0.0002,
+      "loss": 0.7314,
+      "step": 4660
+    },
+    {
+      "epoch": 3.425009167583425,
+      "grad_norm": 0.8052892088890076,
+      "learning_rate": 0.0002,
+      "loss": 0.7901,
+      "step": 4670
+    },
+    {
+      "epoch": 3.432343234323432,
+      "grad_norm": 0.7746483087539673,
+      "learning_rate": 0.0002,
+      "loss": 0.7298,
+      "step": 4680
+    },
+    {
+      "epoch": 3.4396773010634396,
+      "grad_norm": 0.7713689804077148,
+      "learning_rate": 0.0002,
+      "loss": 0.7976,
+      "step": 4690
+    },
+    {
+      "epoch": 3.447011367803447,
+      "grad_norm": 0.810371994972229,
+      "learning_rate": 0.0002,
+      "loss": 0.7427,
+      "step": 4700
+    },
+    {
+      "epoch": 3.4543454345434546,
+      "grad_norm": 0.7702969312667847,
+      "learning_rate": 0.0002,
+      "loss": 0.7594,
+      "step": 4710
+    },
+    {
+      "epoch": 3.4616795012834616,
+      "grad_norm": 0.7069268822669983,
+      "learning_rate": 0.0002,
+      "loss": 0.7957,
+      "step": 4720
+    },
+    {
+      "epoch": 3.469013568023469,
+      "grad_norm": 0.7640359401702881,
+      "learning_rate": 0.0002,
+      "loss": 0.8199,
+      "step": 4730
+    },
+    {
+      "epoch": 3.4763476347634765,
+      "grad_norm": 0.8661707639694214,
+      "learning_rate": 0.0002,
+      "loss": 0.6875,
+      "step": 4740
+    },
+    {
+      "epoch": 3.4836817015034836,
+      "grad_norm": 0.9970282912254333,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 4750
+    },
+    {
+      "epoch": 3.491015768243491,
+      "grad_norm": 0.5824355483055115,
+      "learning_rate": 0.0002,
+      "loss": 0.8462,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4983498349834985,
+      "grad_norm": 1.3072649240493774,
+      "learning_rate": 0.0002,
+      "loss": 0.851,
+      "step": 4770
+    },
+    {
+      "epoch": 3.5056839017235055,
+      "grad_norm": 0.873978316783905,
+      "learning_rate": 0.0002,
+      "loss": 0.9101,
+      "step": 4780
+    },
+    {
+      "epoch": 3.513017968463513,
+      "grad_norm": 0.5526657104492188,
+      "learning_rate": 0.0002,
+      "loss": 0.7403,
+      "step": 4790
+    },
+    {
+      "epoch": 3.5203520352035205,
+      "grad_norm": 0.790894627571106,
+      "learning_rate": 0.0002,
+      "loss": 0.7921,
+      "step": 4800
+    },
+    {
+      "epoch": 3.5276861019435275,
+      "grad_norm": 0.8119630217552185,
+      "learning_rate": 0.0002,
+      "loss": 0.831,
+      "step": 4810
+    },
+    {
+      "epoch": 3.535020168683535,
+      "grad_norm": 0.633212149143219,
+      "learning_rate": 0.0002,
+      "loss": 0.7351,
+      "step": 4820
+    },
+    {
+      "epoch": 3.5423542354235424,
+      "grad_norm": 0.703029990196228,
+      "learning_rate": 0.0002,
+      "loss": 0.8505,
+      "step": 4830
+    },
+    {
+      "epoch": 3.54968830216355,
+      "grad_norm": 0.7603771686553955,
+      "learning_rate": 0.0002,
+      "loss": 0.7204,
+      "step": 4840
+    },
+    {
+      "epoch": 3.557022368903557,
+      "grad_norm": 0.6260480880737305,
+      "learning_rate": 0.0002,
+      "loss": 0.8868,
+      "step": 4850
+    },
+    {
+      "epoch": 3.5643564356435644,
+      "grad_norm": 0.8203664422035217,
+      "learning_rate": 0.0002,
+      "loss": 0.8137,
+      "step": 4860
+    },
+    {
+      "epoch": 3.5716905023835714,
+      "grad_norm": 0.7793813347816467,
+      "learning_rate": 0.0002,
+      "loss": 0.8821,
+      "step": 4870
+    },
+    {
+      "epoch": 3.579024569123579,
+      "grad_norm": 0.7667397260665894,
+      "learning_rate": 0.0002,
+      "loss": 0.8164,
+      "step": 4880
+    },
+    {
+      "epoch": 3.5863586358635864,
+      "grad_norm": 0.8198829889297485,
+      "learning_rate": 0.0002,
+      "loss": 0.7597,
+      "step": 4890
+    },
+    {
+      "epoch": 3.593692702603594,
+      "grad_norm": 0.7689233422279358,
+      "learning_rate": 0.0002,
+      "loss": 0.7027,
+      "step": 4900
+    },
+    {
+      "epoch": 3.601026769343601,
+      "grad_norm": 0.7870983481407166,
+      "learning_rate": 0.0002,
+      "loss": 0.804,
+      "step": 4910
+    },
+    {
+      "epoch": 3.6083608360836084,
+      "grad_norm": 0.8133853077888489,
+      "learning_rate": 0.0002,
+      "loss": 0.8269,
+      "step": 4920
+    },
+    {
+      "epoch": 3.615694902823616,
+      "grad_norm": 1.308401346206665,
+      "learning_rate": 0.0002,
+      "loss": 0.8515,
+      "step": 4930
+    },
+    {
+      "epoch": 3.623028969563623,
+      "grad_norm": 0.7131121754646301,
+      "learning_rate": 0.0002,
+      "loss": 0.8494,
+      "step": 4940
+    },
+    {
+      "epoch": 3.6303630363036303,
+      "grad_norm": 0.6825910210609436,
+      "learning_rate": 0.0002,
+      "loss": 0.7235,
+      "step": 4950
+    },
+    {
+      "epoch": 3.637697103043638,
+      "grad_norm": 0.7254678606987,
+      "learning_rate": 0.0002,
+      "loss": 0.7824,
+      "step": 4960
+    },
+    {
+      "epoch": 3.6450311697836453,
+      "grad_norm": 0.8045085072517395,
+      "learning_rate": 0.0002,
+      "loss": 0.7983,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6523652365236523,
+      "grad_norm": 0.6991777420043945,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6596993032636598,
+      "grad_norm": 0.7804713249206543,
+      "learning_rate": 0.0002,
+      "loss": 0.7806,
+      "step": 4990
+    },
+    {
+      "epoch": 3.667033370003667,
+      "grad_norm": 0.8525708317756653,
+      "learning_rate": 0.0002,
+      "loss": 0.8402,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6743674367436743,
+      "grad_norm": 0.7959994673728943,
+      "learning_rate": 0.0002,
+      "loss": 0.8496,
+      "step": 5010
+    },
+    {
+      "epoch": 3.6817015034836817,
+      "grad_norm": 0.8103628158569336,
+      "learning_rate": 0.0002,
+      "loss": 0.8022,
+      "step": 5020
+    },
+    {
+      "epoch": 3.689035570223689,
+      "grad_norm": 0.7517836093902588,
+      "learning_rate": 0.0002,
+      "loss": 0.7376,
+      "step": 5030
+    },
+    {
+      "epoch": 3.6963696369636962,
+      "grad_norm": 0.6878514289855957,
+      "learning_rate": 0.0002,
+      "loss": 0.8375,
+      "step": 5040
+    },
+    {
+      "epoch": 3.7037037037037037,
+      "grad_norm": 1.2371820211410522,
+      "learning_rate": 0.0002,
+      "loss": 0.7998,
+      "step": 5050
+    },
+    {
+      "epoch": 3.711037770443711,
+      "grad_norm": 0.6567103862762451,
+      "learning_rate": 0.0002,
+      "loss": 0.6941,
+      "step": 5060
+    },
+    {
+      "epoch": 3.718371837183718,
+      "grad_norm": 1.1254922151565552,
+      "learning_rate": 0.0002,
+      "loss": 0.8465,
+      "step": 5070
+    },
+    {
+      "epoch": 3.7257059039237257,
+      "grad_norm": 0.6796132326126099,
+      "learning_rate": 0.0002,
+      "loss": 0.8365,
+      "step": 5080
+    },
+    {
+      "epoch": 3.733039970663733,
+      "grad_norm": 0.7285300493240356,
+      "learning_rate": 0.0002,
+      "loss": 0.7818,
+      "step": 5090
+    },
+    {
+      "epoch": 3.7403740374037406,
+      "grad_norm": 0.8931500911712646,
+      "learning_rate": 0.0002,
+      "loss": 0.8581,
+      "step": 5100
+    },
+    {
+      "epoch": 3.7477081041437477,
+      "grad_norm": 0.6256856918334961,
+      "learning_rate": 0.0002,
+      "loss": 0.8181,
+      "step": 5110
+    },
+    {
+      "epoch": 3.755042170883755,
+      "grad_norm": 0.79310142993927,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 5120
+    },
+    {
+      "epoch": 3.762376237623762,
+      "grad_norm": 0.6594041585922241,
+      "learning_rate": 0.0002,
+      "loss": 0.8235,
+      "step": 5130
+    },
+    {
+      "epoch": 3.7697103043637696,
+      "grad_norm": 0.7029327750205994,
+      "learning_rate": 0.0002,
+      "loss": 0.6925,
+      "step": 5140
+    },
+    {
+      "epoch": 3.777044371103777,
+      "grad_norm": 0.5880070328712463,
+      "learning_rate": 0.0002,
+      "loss": 0.7457,
+      "step": 5150
+    },
+    {
+      "epoch": 3.7843784378437846,
+      "grad_norm": 0.7578945159912109,
+      "learning_rate": 0.0002,
+      "loss": 0.8716,
+      "step": 5160
+    },
+    {
+      "epoch": 3.7917125045837916,
+      "grad_norm": 0.8276378512382507,
+      "learning_rate": 0.0002,
+      "loss": 0.8819,
+      "step": 5170
+    },
+    {
+      "epoch": 3.799046571323799,
+      "grad_norm": 0.7627953886985779,
+      "learning_rate": 0.0002,
+      "loss": 0.7559,
+      "step": 5180
+    },
+    {
+      "epoch": 3.806380638063806,
+      "grad_norm": 0.8169086575508118,
+      "learning_rate": 0.0002,
+      "loss": 0.7665,
+      "step": 5190
+    },
+    {
+      "epoch": 3.8137147048038136,
+      "grad_norm": 0.6605030298233032,
+      "learning_rate": 0.0002,
+      "loss": 0.761,
+      "step": 5200
+    },
+    {
+      "epoch": 3.821048771543821,
+      "grad_norm": 0.5837286114692688,
+      "learning_rate": 0.0002,
+      "loss": 0.8804,
+      "step": 5210
+    },
+    {
+      "epoch": 3.8283828382838285,
+      "grad_norm": 1.2422157526016235,
+      "learning_rate": 0.0002,
+      "loss": 0.8369,
+      "step": 5220
+    },
+    {
+      "epoch": 3.8357169050238356,
+      "grad_norm": 0.6589220762252808,
+      "learning_rate": 0.0002,
+      "loss": 0.8431,
+      "step": 5230
+    },
+    {
+      "epoch": 3.843050971763843,
+      "grad_norm": 0.8567556142807007,
+      "learning_rate": 0.0002,
+      "loss": 0.7686,
+      "step": 5240
+    },
+    {
+      "epoch": 3.8503850385038505,
+      "grad_norm": 0.6490627527236938,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8577191052438575,
+      "grad_norm": 0.620232880115509,
+      "learning_rate": 0.0002,
+      "loss": 0.7386,
+      "step": 5260
+    },
+    {
+      "epoch": 3.865053171983865,
+      "grad_norm": 0.7685128450393677,
+      "learning_rate": 0.0002,
+      "loss": 0.9192,
+      "step": 5270
+    },
+    {
+      "epoch": 3.8723872387238725,
+      "grad_norm": 0.8113296627998352,
+      "learning_rate": 0.0002,
+      "loss": 0.872,
+      "step": 5280
+    },
+    {
+      "epoch": 3.87972130546388,
+      "grad_norm": 0.8092675805091858,
+      "learning_rate": 0.0002,
+      "loss": 0.7156,
+      "step": 5290
+    },
+    {
+      "epoch": 3.887055372203887,
+      "grad_norm": 0.583570122718811,
+      "learning_rate": 0.0002,
+      "loss": 0.7325,
+      "step": 5300
+    },
+    {
+      "epoch": 3.8943894389438944,
+      "grad_norm": 1.712363600730896,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 5310
+    },
+    {
+      "epoch": 3.9017235056839015,
+      "grad_norm": 0.6673534512519836,
+      "learning_rate": 0.0002,
+      "loss": 0.7537,
+      "step": 5320
+    },
+    {
+      "epoch": 3.909057572423909,
+      "grad_norm": 1.9770312309265137,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 5330
+    },
+    {
+      "epoch": 3.9163916391639164,
+      "grad_norm": 0.6430999636650085,
+      "learning_rate": 0.0002,
+      "loss": 0.8793,
+      "step": 5340
+    },
+    {
+      "epoch": 3.923725705903924,
+      "grad_norm": 1.0159571170806885,
+      "learning_rate": 0.0002,
+      "loss": 0.839,
+      "step": 5350
+    },
+    {
+      "epoch": 3.931059772643931,
+      "grad_norm": 0.8607584834098816,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 5360
+    },
+    {
+      "epoch": 3.9383938393839384,
+      "grad_norm": 0.6967900991439819,
+      "learning_rate": 0.0002,
+      "loss": 0.7261,
+      "step": 5370
+    },
+    {
+      "epoch": 3.945727906123946,
+      "grad_norm": 0.7683077454566956,
+      "learning_rate": 0.0002,
+      "loss": 0.8456,
+      "step": 5380
+    },
+    {
+      "epoch": 3.953061972863953,
+      "grad_norm": 0.6805762648582458,
+      "learning_rate": 0.0002,
+      "loss": 0.7682,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9603960396039604,
+      "grad_norm": 0.7033619284629822,
+      "learning_rate": 0.0002,
+      "loss": 0.7746,
+      "step": 5400
+    },
+    {
+      "epoch": 3.967730106343968,
+      "grad_norm": 0.966112494468689,
+      "learning_rate": 0.0002,
+      "loss": 0.8393,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9750641730839753,
+      "grad_norm": 0.8467881083488464,
+      "learning_rate": 0.0002,
+      "loss": 0.8316,
+      "step": 5420
+    },
+    {
+      "epoch": 3.9823982398239823,
+      "grad_norm": 0.8005317449569702,
+      "learning_rate": 0.0002,
+      "loss": 0.8084,
+      "step": 5430
+    },
+    {
+      "epoch": 3.98973230656399,
+      "grad_norm": 1.1615241765975952,
+      "learning_rate": 0.0002,
+      "loss": 0.7168,
+      "step": 5440
+    },
+    {
+      "epoch": 3.997066373303997,
+      "grad_norm": 0.6121614575386047,
+      "learning_rate": 0.0002,
+      "loss": 0.8263,
+      "step": 5450
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.1834222078323364,
+      "eval_runtime": 32.7569,
+      "eval_samples_per_second": 13.158,
+      "eval_steps_per_second": 1.649,
+      "step": 5454
+    },
+    {
+      "epoch": 4.004400440044004,
+      "grad_norm": 0.6055727005004883,
+      "learning_rate": 0.0002,
+      "loss": 0.7267,
+      "step": 5460
+    },
+    {
+      "epoch": 4.011734506784012,
+      "grad_norm": 0.8232647180557251,
+      "learning_rate": 0.0002,
+      "loss": 0.5766,
+      "step": 5470
+    },
+    {
+      "epoch": 4.019068573524019,
+      "grad_norm": 0.7739192247390747,
+      "learning_rate": 0.0002,
+      "loss": 0.6489,
+      "step": 5480
+    },
+    {
+      "epoch": 4.026402640264027,
+      "grad_norm": 0.6264950633049011,
+      "learning_rate": 0.0002,
+      "loss": 0.5978,
+      "step": 5490
+    },
+    {
+      "epoch": 4.033736707004033,
+      "grad_norm": 1.4798702001571655,
+      "learning_rate": 0.0002,
+      "loss": 0.6392,
+      "step": 5500
+    },
+    {
+      "epoch": 4.041070773744041,
+      "grad_norm": 0.9538470506668091,
+      "learning_rate": 0.0002,
+      "loss": 0.6143,
+      "step": 5510
+    },
+    {
+      "epoch": 4.048404840484048,
+      "grad_norm": 0.834561288356781,
+      "learning_rate": 0.0002,
+      "loss": 0.6056,
+      "step": 5520
+    },
+    {
+      "epoch": 4.055738907224056,
+      "grad_norm": 0.6407850384712219,
+      "learning_rate": 0.0002,
+      "loss": 0.6077,
+      "step": 5530
+    },
+    {
+      "epoch": 4.063072973964063,
+      "grad_norm": 0.9035961627960205,
+      "learning_rate": 0.0002,
+      "loss": 0.6733,
+      "step": 5540
+    },
+    {
+      "epoch": 4.070407040704071,
+      "grad_norm": 0.842812716960907,
+      "learning_rate": 0.0002,
+      "loss": 0.5854,
+      "step": 5550
+    },
+    {
+      "epoch": 4.077741107444078,
+      "grad_norm": 0.8197882175445557,
+      "learning_rate": 0.0002,
+      "loss": 0.654,
+      "step": 5560
+    },
+    {
+      "epoch": 4.085075174184085,
+      "grad_norm": 0.8652673959732056,
+      "learning_rate": 0.0002,
+      "loss": 0.5919,
+      "step": 5570
+    },
+    {
+      "epoch": 4.092409240924092,
+      "grad_norm": 0.8048318028450012,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 5580
+    },
+    {
+      "epoch": 4.0997433076641,
+      "grad_norm": 0.9604969024658203,
+      "learning_rate": 0.0002,
+      "loss": 0.6487,
+      "step": 5590
+    },
+    {
+      "epoch": 4.107077374404107,
+      "grad_norm": 1.244756817817688,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 5600
+    },
+    {
+      "epoch": 4.114411441144115,
+      "grad_norm": 0.7975269556045532,
+      "learning_rate": 0.0002,
+      "loss": 0.6489,
+      "step": 5610
+    },
+    {
+      "epoch": 4.121745507884122,
+      "grad_norm": 0.6130099296569824,
+      "learning_rate": 0.0002,
+      "loss": 0.6445,
+      "step": 5620
+    },
+    {
+      "epoch": 4.129079574624129,
+      "grad_norm": 0.7793202996253967,
+      "learning_rate": 0.0002,
+      "loss": 0.6024,
+      "step": 5630
+    },
+    {
+      "epoch": 4.136413641364136,
+      "grad_norm": 1.187238335609436,
+      "learning_rate": 0.0002,
+      "loss": 0.5723,
+      "step": 5640
+    },
+    {
+      "epoch": 4.143747708104144,
+      "grad_norm": 0.8450375199317932,
+      "learning_rate": 0.0002,
+      "loss": 0.6385,
+      "step": 5650
+    },
+    {
+      "epoch": 4.151081774844151,
+      "grad_norm": 0.9006940126419067,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 5660
+    },
+    {
+      "epoch": 4.158415841584159,
+      "grad_norm": 0.9447154998779297,
+      "learning_rate": 0.0002,
+      "loss": 0.6179,
+      "step": 5670
+    },
+    {
+      "epoch": 4.165749908324166,
+      "grad_norm": 0.798032283782959,
+      "learning_rate": 0.0002,
+      "loss": 0.6476,
+      "step": 5680
+    },
+    {
+      "epoch": 4.1730839750641735,
+      "grad_norm": 0.65578693151474,
+      "learning_rate": 0.0002,
+      "loss": 0.6666,
+      "step": 5690
+    },
+    {
+      "epoch": 4.18041804180418,
+      "grad_norm": 1.0864700078964233,
+      "learning_rate": 0.0002,
+      "loss": 0.701,
+      "step": 5700
+    },
+    {
+      "epoch": 4.187752108544188,
+      "grad_norm": 0.7344121932983398,
+      "learning_rate": 0.0002,
+      "loss": 0.6895,
+      "step": 5710
+    },
+    {
+      "epoch": 4.195086175284195,
+      "grad_norm": 0.9722456932067871,
+      "learning_rate": 0.0002,
+      "loss": 0.6659,
+      "step": 5720
+    },
+    {
+      "epoch": 4.2024202420242025,
+      "grad_norm": 1.263814926147461,
+      "learning_rate": 0.0002,
+      "loss": 0.6887,
+      "step": 5730
+    },
+    {
+      "epoch": 4.20975430876421,
+      "grad_norm": 0.9622581005096436,
+      "learning_rate": 0.0002,
+      "loss": 0.608,
+      "step": 5740
+    },
+    {
+      "epoch": 4.2170883755042174,
+      "grad_norm": 0.8497143387794495,
+      "learning_rate": 0.0002,
+      "loss": 0.6221,
+      "step": 5750
+    },
+    {
+      "epoch": 4.224422442244224,
+      "grad_norm": 0.8248446583747864,
+      "learning_rate": 0.0002,
+      "loss": 0.6322,
+      "step": 5760
+    },
+    {
+      "epoch": 4.2317565089842315,
+      "grad_norm": 1.2544798851013184,
+      "learning_rate": 0.0002,
+      "loss": 0.6045,
+      "step": 5770
+    },
+    {
+      "epoch": 4.239090575724239,
+      "grad_norm": 0.8224676251411438,
+      "learning_rate": 0.0002,
+      "loss": 0.641,
+      "step": 5780
+    },
+    {
+      "epoch": 4.2464246424642464,
+      "grad_norm": 0.8924877047538757,
+      "learning_rate": 0.0002,
+      "loss": 0.6399,
+      "step": 5790
+    },
+    {
+      "epoch": 4.253758709204254,
+      "grad_norm": 0.8545848727226257,
+      "learning_rate": 0.0002,
+      "loss": 0.6845,
+      "step": 5800
+    },
+    {
+      "epoch": 4.261092775944261,
+      "grad_norm": 0.8081067800521851,
+      "learning_rate": 0.0002,
+      "loss": 0.6669,
+      "step": 5810
+    },
+    {
+      "epoch": 4.268426842684269,
+      "grad_norm": 0.7111002802848816,
+      "learning_rate": 0.0002,
+      "loss": 0.6149,
+      "step": 5820
+    },
+    {
+      "epoch": 4.2757609094242754,
+      "grad_norm": 0.8696979880332947,
+      "learning_rate": 0.0002,
+      "loss": 0.6343,
+      "step": 5830
+    },
+    {
+      "epoch": 4.283094976164283,
+      "grad_norm": 0.821401834487915,
+      "learning_rate": 0.0002,
+      "loss": 0.6384,
+      "step": 5840
+    },
+    {
+      "epoch": 4.29042904290429,
+      "grad_norm": 0.888908326625824,
+      "learning_rate": 0.0002,
+      "loss": 0.6912,
+      "step": 5850
+    },
+    {
+      "epoch": 4.297763109644298,
+      "grad_norm": 1.9380123615264893,
+      "learning_rate": 0.0002,
+      "loss": 0.6061,
+      "step": 5860
+    },
+    {
+      "epoch": 4.305097176384305,
+      "grad_norm": 1.121774435043335,
+      "learning_rate": 0.0002,
+      "loss": 0.6766,
+      "step": 5870
+    },
+    {
+      "epoch": 4.312431243124313,
+      "grad_norm": 0.9238282442092896,
+      "learning_rate": 0.0002,
+      "loss": 0.7205,
+      "step": 5880
+    },
+    {
+      "epoch": 4.319765309864319,
+      "grad_norm": 0.7321620583534241,
+      "learning_rate": 0.0002,
+      "loss": 0.6351,
+      "step": 5890
+    },
+    {
+      "epoch": 4.327099376604327,
+      "grad_norm": 0.8739548325538635,
+      "learning_rate": 0.0002,
+      "loss": 0.6404,
+      "step": 5900
+    },
+    {
+      "epoch": 4.334433443344334,
+      "grad_norm": 0.9686012268066406,
+      "learning_rate": 0.0002,
+      "loss": 0.5892,
+      "step": 5910
+    },
+    {
+      "epoch": 4.341767510084342,
+      "grad_norm": 0.9033839106559753,
+      "learning_rate": 0.0002,
+      "loss": 0.641,
+      "step": 5920
+    },
+    {
+      "epoch": 4.349101576824349,
+      "grad_norm": 0.8131115436553955,
+      "learning_rate": 0.0002,
+      "loss": 0.6456,
+      "step": 5930
+    },
+    {
+      "epoch": 4.356435643564357,
+      "grad_norm": 0.8942412734031677,
+      "learning_rate": 0.0002,
+      "loss": 0.5826,
+      "step": 5940
+    },
+    {
+      "epoch": 4.363769710304364,
+      "grad_norm": 0.8439112901687622,
+      "learning_rate": 0.0002,
+      "loss": 0.7336,
+      "step": 5950
+    },
+    {
+      "epoch": 4.371103777044371,
+      "grad_norm": 0.9176713228225708,
+      "learning_rate": 0.0002,
+      "loss": 0.6537,
+      "step": 5960
+    },
+    {
+      "epoch": 4.378437843784378,
+      "grad_norm": 0.6799634695053101,
+      "learning_rate": 0.0002,
+      "loss": 0.6792,
+      "step": 5970
+    },
+    {
+      "epoch": 4.385771910524386,
+      "grad_norm": 1.0435824394226074,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 5980
+    },
+    {
+      "epoch": 4.393105977264393,
+      "grad_norm": 0.997937798500061,
+      "learning_rate": 0.0002,
+      "loss": 0.68,
+      "step": 5990
+    },
+    {
+      "epoch": 4.400440044004401,
+      "grad_norm": 1.0308842658996582,
+      "learning_rate": 0.0002,
+      "loss": 0.6604,
+      "step": 6000
+    },
+    {
+      "epoch": 4.407774110744408,
+      "grad_norm": 1.3683775663375854,
+      "learning_rate": 0.0002,
+      "loss": 0.6402,
+      "step": 6010
+    },
+    {
+      "epoch": 4.415108177484415,
+      "grad_norm": 0.7569534182548523,
+      "learning_rate": 0.0002,
+      "loss": 0.7027,
+      "step": 6020
+    },
+    {
+      "epoch": 4.422442244224422,
+      "grad_norm": 1.089978575706482,
+      "learning_rate": 0.0002,
+      "loss": 0.5949,
+      "step": 6030
+    },
+    {
+      "epoch": 4.42977631096443,
+      "grad_norm": 0.7522459626197815,
+      "learning_rate": 0.0002,
+      "loss": 0.6353,
+      "step": 6040
+    },
+    {
+      "epoch": 4.437110377704437,
+      "grad_norm": 0.6709823608398438,
+      "learning_rate": 0.0002,
+      "loss": 0.5852,
+      "step": 6050
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.6992089748382568,
+      "learning_rate": 0.0002,
+      "loss": 0.6718,
+      "step": 6060
+    },
+    {
+      "epoch": 4.451778511184452,
+      "grad_norm": 1.0182931423187256,
+      "learning_rate": 0.0002,
+      "loss": 0.6933,
+      "step": 6070
+    },
+    {
+      "epoch": 4.459112577924459,
+      "grad_norm": 1.0685160160064697,
+      "learning_rate": 0.0002,
+      "loss": 0.6255,
+      "step": 6080
+    },
+    {
+      "epoch": 4.466446644664466,
+      "grad_norm": 0.8295124769210815,
+      "learning_rate": 0.0002,
+      "loss": 0.6086,
+      "step": 6090
+    },
+    {
+      "epoch": 4.473780711404474,
+      "grad_norm": 1.1862998008728027,
+      "learning_rate": 0.0002,
+      "loss": 0.6359,
+      "step": 6100
+    },
+    {
+      "epoch": 4.481114778144481,
+      "grad_norm": 0.7400273084640503,
+      "learning_rate": 0.0002,
+      "loss": 0.638,
+      "step": 6110
+    },
+    {
+      "epoch": 4.488448844884489,
+      "grad_norm": 0.7098417282104492,
+      "learning_rate": 0.0002,
+      "loss": 0.6854,
+      "step": 6120
+    },
+    {
+      "epoch": 4.495782911624496,
+      "grad_norm": 0.9745053648948669,
+      "learning_rate": 0.0002,
+      "loss": 0.6976,
+      "step": 6130
+    },
+    {
+      "epoch": 4.503116978364503,
+      "grad_norm": 0.8638797998428345,
+      "learning_rate": 0.0002,
+      "loss": 0.605,
+      "step": 6140
+    },
+    {
+      "epoch": 4.51045104510451,
+      "grad_norm": 0.8291046619415283,
+      "learning_rate": 0.0002,
+      "loss": 0.6491,
+      "step": 6150
+    },
+    {
+      "epoch": 4.517785111844518,
+      "grad_norm": 1.0301737785339355,
+      "learning_rate": 0.0002,
+      "loss": 0.6457,
+      "step": 6160
+    },
+    {
+      "epoch": 4.525119178584525,
+      "grad_norm": 1.1996512413024902,
+      "learning_rate": 0.0002,
+      "loss": 0.6742,
+      "step": 6170
+    },
+    {
+      "epoch": 4.5324532453245325,
+      "grad_norm": 1.151038408279419,
+      "learning_rate": 0.0002,
+      "loss": 0.6484,
+      "step": 6180
+    },
+    {
+      "epoch": 4.53978731206454,
+      "grad_norm": 0.8385201096534729,
+      "learning_rate": 0.0002,
+      "loss": 0.668,
+      "step": 6190
+    },
+    {
+      "epoch": 4.5471213788045475,
+      "grad_norm": 0.8969188332557678,
+      "learning_rate": 0.0002,
+      "loss": 0.6381,
+      "step": 6200
+    },
+    {
+      "epoch": 4.554455445544555,
+      "grad_norm": 1.60659658908844,
+      "learning_rate": 0.0002,
+      "loss": 0.7141,
+      "step": 6210
+    },
+    {
+      "epoch": 4.5617895122845615,
+      "grad_norm": 0.9356731176376343,
+      "learning_rate": 0.0002,
+      "loss": 0.6388,
+      "step": 6220
+    },
+    {
+      "epoch": 4.569123579024569,
+      "grad_norm": 0.95856773853302,
+      "learning_rate": 0.0002,
+      "loss": 0.7393,
+      "step": 6230
+    },
+    {
+      "epoch": 4.5764576457645765,
+      "grad_norm": 1.1162524223327637,
+      "learning_rate": 0.0002,
+      "loss": 0.6554,
+      "step": 6240
+    },
+    {
+      "epoch": 4.583791712504584,
+      "grad_norm": 0.8809238076210022,
+      "learning_rate": 0.0002,
+      "loss": 0.6012,
+      "step": 6250
+    },
+    {
+      "epoch": 4.591125779244591,
+      "grad_norm": 0.890738844871521,
+      "learning_rate": 0.0002,
+      "loss": 0.648,
+      "step": 6260
+    },
+    {
+      "epoch": 4.598459845984598,
+      "grad_norm": 0.918684720993042,
+      "learning_rate": 0.0002,
+      "loss": 0.6663,
+      "step": 6270
+    },
+    {
+      "epoch": 4.6057939127246055,
+      "grad_norm": 0.8156296610832214,
+      "learning_rate": 0.0002,
+      "loss": 0.5992,
+      "step": 6280
+    },
+    {
+      "epoch": 4.613127979464613,
+      "grad_norm": 1.046634316444397,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 6290
+    },
+    {
+      "epoch": 4.62046204620462,
+      "grad_norm": 0.7725525498390198,
+      "learning_rate": 0.0002,
+      "loss": 0.7023,
+      "step": 6300
+    },
+    {
+      "epoch": 4.627796112944628,
+      "grad_norm": 0.9992046356201172,
+      "learning_rate": 0.0002,
+      "loss": 0.6414,
+      "step": 6310
+    },
+    {
+      "epoch": 4.635130179684635,
+      "grad_norm": 0.8480095267295837,
+      "learning_rate": 0.0002,
+      "loss": 0.6201,
+      "step": 6320
+    },
+    {
+      "epoch": 4.642464246424643,
+      "grad_norm": 0.7061955332756042,
+      "learning_rate": 0.0002,
+      "loss": 0.6869,
+      "step": 6330
+    },
+    {
+      "epoch": 4.649798313164649,
+      "grad_norm": 1.0354212522506714,
+      "learning_rate": 0.0002,
+      "loss": 0.6828,
+      "step": 6340
+    },
+    {
+      "epoch": 4.657132379904657,
+      "grad_norm": 1.0081377029418945,
+      "learning_rate": 0.0002,
+      "loss": 0.6651,
+      "step": 6350
+    },
+    {
+      "epoch": 4.664466446644664,
+      "grad_norm": 1.2904249429702759,
+      "learning_rate": 0.0002,
+      "loss": 0.726,
+      "step": 6360
+    },
+    {
+      "epoch": 4.671800513384672,
+      "grad_norm": 0.9248910546302795,
+      "learning_rate": 0.0002,
+      "loss": 0.7148,
+      "step": 6370
+    },
+    {
+      "epoch": 4.679134580124679,
+      "grad_norm": 0.9907804131507874,
+      "learning_rate": 0.0002,
+      "loss": 0.6961,
+      "step": 6380
+    },
+    {
+      "epoch": 4.686468646864687,
+      "grad_norm": 1.201143741607666,
+      "learning_rate": 0.0002,
+      "loss": 0.6163,
+      "step": 6390
+    },
+    {
+      "epoch": 4.693802713604693,
+      "grad_norm": 0.8709394335746765,
+      "learning_rate": 0.0002,
+      "loss": 0.6762,
+      "step": 6400
+    },
+    {
+      "epoch": 4.701136780344701,
+      "grad_norm": 0.7468608021736145,
+      "learning_rate": 0.0002,
+      "loss": 0.7217,
+      "step": 6410
+    },
+    {
+      "epoch": 4.708470847084708,
+      "grad_norm": 0.8607903718948364,
+      "learning_rate": 0.0002,
+      "loss": 0.6548,
+      "step": 6420
+    },
+    {
+      "epoch": 4.715804913824716,
+      "grad_norm": 0.9840512871742249,
+      "learning_rate": 0.0002,
+      "loss": 0.6449,
+      "step": 6430
+    },
+    {
+      "epoch": 4.723138980564723,
+      "grad_norm": 0.8328204154968262,
+      "learning_rate": 0.0002,
+      "loss": 0.685,
+      "step": 6440
+    },
+    {
+      "epoch": 4.730473047304731,
+      "grad_norm": 0.924505352973938,
+      "learning_rate": 0.0002,
+      "loss": 0.697,
+      "step": 6450
+    },
+    {
+      "epoch": 4.737807114044738,
+      "grad_norm": 0.8897685408592224,
+      "learning_rate": 0.0002,
+      "loss": 0.7422,
+      "step": 6460
+    },
+    {
+      "epoch": 4.745141180784745,
+      "grad_norm": 0.9605024456977844,
+      "learning_rate": 0.0002,
+      "loss": 0.6842,
+      "step": 6470
+    },
+    {
+      "epoch": 4.752475247524752,
+      "grad_norm": 0.8150759935379028,
+      "learning_rate": 0.0002,
+      "loss": 0.6488,
+      "step": 6480
+    },
+    {
+      "epoch": 4.75980931426476,
+      "grad_norm": 0.8128412961959839,
+      "learning_rate": 0.0002,
+      "loss": 0.6606,
+      "step": 6490
+    },
+    {
+      "epoch": 4.767143381004767,
+      "grad_norm": 0.7381404638290405,
+      "learning_rate": 0.0002,
+      "loss": 0.6729,
+      "step": 6500
+    },
+    {
+      "epoch": 4.774477447744775,
+      "grad_norm": 1.0565853118896484,
+      "learning_rate": 0.0002,
+      "loss": 0.6713,
+      "step": 6510
+    },
+    {
+      "epoch": 4.781811514484782,
+      "grad_norm": 0.9298134446144104,
+      "learning_rate": 0.0002,
+      "loss": 0.6496,
+      "step": 6520
+    },
+    {
+      "epoch": 4.789145581224789,
+      "grad_norm": 1.0145525932312012,
+      "learning_rate": 0.0002,
+      "loss": 0.7279,
+      "step": 6530
+    },
+    {
+      "epoch": 4.796479647964796,
+      "grad_norm": 0.92259681224823,
+      "learning_rate": 0.0002,
+      "loss": 0.5986,
+      "step": 6540
+    },
+    {
+      "epoch": 4.803813714704804,
+      "grad_norm": 0.7881024479866028,
+      "learning_rate": 0.0002,
+      "loss": 0.63,
+      "step": 6550
+    },
+    {
+      "epoch": 4.811147781444811,
+      "grad_norm": 1.4935206174850464,
+      "learning_rate": 0.0002,
+      "loss": 0.7134,
+      "step": 6560
+    },
+    {
+      "epoch": 4.818481848184819,
+      "grad_norm": 0.8612369298934937,
+      "learning_rate": 0.0002,
+      "loss": 0.6695,
+      "step": 6570
+    },
+    {
+      "epoch": 4.825815914924826,
+      "grad_norm": 1.0118653774261475,
+      "learning_rate": 0.0002,
+      "loss": 0.779,
+      "step": 6580
+    },
+    {
+      "epoch": 4.833149981664834,
+      "grad_norm": 1.1303809881210327,
+      "learning_rate": 0.0002,
+      "loss": 0.6991,
+      "step": 6590
+    },
+    {
+      "epoch": 4.84048404840484,
+      "grad_norm": 0.9112492203712463,
+      "learning_rate": 0.0002,
+      "loss": 0.7887,
+      "step": 6600
+    },
+    {
+      "epoch": 4.847818115144848,
+      "grad_norm": 0.864762544631958,
+      "learning_rate": 0.0002,
+      "loss": 0.7699,
+      "step": 6610
+    },
+    {
+      "epoch": 4.855152181884855,
+      "grad_norm": 0.9090572595596313,
+      "learning_rate": 0.0002,
+      "loss": 0.7347,
+      "step": 6620
+    },
+    {
+      "epoch": 4.862486248624863,
+      "grad_norm": 1.014953374862671,
+      "learning_rate": 0.0002,
+      "loss": 0.6608,
+      "step": 6630
+    },
+    {
+      "epoch": 4.86982031536487,
+      "grad_norm": 1.0702149868011475,
+      "learning_rate": 0.0002,
+      "loss": 0.6429,
+      "step": 6640
+    },
+    {
+      "epoch": 4.8771543821048775,
+      "grad_norm": 1.002135157585144,
+      "learning_rate": 0.0002,
+      "loss": 0.6943,
+      "step": 6650
+    },
+    {
+      "epoch": 4.884488448844884,
+      "grad_norm": 0.862545907497406,
+      "learning_rate": 0.0002,
+      "loss": 0.7225,
+      "step": 6660
+    },
+    {
+      "epoch": 4.891822515584892,
+      "grad_norm": 0.7302131056785583,
+      "learning_rate": 0.0002,
+      "loss": 0.6206,
+      "step": 6670
+    },
+    {
+      "epoch": 4.899156582324899,
+      "grad_norm": 0.8380730152130127,
+      "learning_rate": 0.0002,
+      "loss": 0.7175,
+      "step": 6680
+    },
+    {
+      "epoch": 4.9064906490649065,
+      "grad_norm": 0.7956018447875977,
+      "learning_rate": 0.0002,
+      "loss": 0.645,
+      "step": 6690
+    },
+    {
+      "epoch": 4.913824715804914,
+      "grad_norm": 0.6717583537101746,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 6700
+    },
+    {
+      "epoch": 4.9211587825449215,
+      "grad_norm": 1.09099280834198,
+      "learning_rate": 0.0002,
+      "loss": 0.6942,
+      "step": 6710
+    },
+    {
+      "epoch": 4.928492849284929,
+      "grad_norm": 0.8589889407157898,
+      "learning_rate": 0.0002,
+      "loss": 0.7533,
+      "step": 6720
+    },
+    {
+      "epoch": 4.9358269160249355,
+      "grad_norm": 1.0046314001083374,
+      "learning_rate": 0.0002,
+      "loss": 0.66,
+      "step": 6730
+    },
+    {
+      "epoch": 4.943160982764943,
+      "grad_norm": 0.8559659123420715,
+      "learning_rate": 0.0002,
+      "loss": 0.6864,
+      "step": 6740
+    },
+    {
+      "epoch": 4.9504950495049505,
+      "grad_norm": 0.8588525652885437,
+      "learning_rate": 0.0002,
+      "loss": 0.6847,
+      "step": 6750
+    },
+    {
+      "epoch": 4.957829116244958,
+      "grad_norm": 0.9192708134651184,
+      "learning_rate": 0.0002,
+      "loss": 0.6428,
+      "step": 6760
+    },
+    {
+      "epoch": 4.965163182984965,
+      "grad_norm": 1.051398754119873,
+      "learning_rate": 0.0002,
+      "loss": 0.6873,
+      "step": 6770
+    },
+    {
+      "epoch": 4.972497249724973,
+      "grad_norm": 0.9111362099647522,
+      "learning_rate": 0.0002,
+      "loss": 0.7249,
+      "step": 6780
+    },
+    {
+      "epoch": 4.9798313164649795,
+      "grad_norm": 0.7305638194084167,
+      "learning_rate": 0.0002,
+      "loss": 0.7613,
+      "step": 6790
+    },
+    {
+      "epoch": 4.987165383204987,
+      "grad_norm": 1.118837594985962,
+      "learning_rate": 0.0002,
+      "loss": 0.6747,
+      "step": 6800
+    },
+    {
+      "epoch": 4.994499449944994,
+      "grad_norm": 0.9075239300727844,
+      "learning_rate": 0.0002,
+      "loss": 0.6412,
+      "step": 6810
+    },
+    {
+      "epoch": 4.999633296662999,
+      "eval_loss": 1.2361247539520264,
+      "eval_runtime": 32.7325,
+      "eval_samples_per_second": 13.167,
+      "eval_steps_per_second": 1.65,
+      "step": 6817
+    },
+    {
+      "epoch": 5.001833516685002,
+      "grad_norm": 1.0541315078735352,
+      "learning_rate": 0.0002,
+      "loss": 0.7091,
+      "step": 6820
+    },
+    {
+      "epoch": 5.009167583425009,
+      "grad_norm": 0.9750140905380249,
+      "learning_rate": 0.0002,
+      "loss": 0.4882,
+      "step": 6830
+    },
+    {
+      "epoch": 5.016501650165017,
+      "grad_norm": 0.931838870048523,
+      "learning_rate": 0.0002,
+      "loss": 0.6022,
+      "step": 6840
+    },
+    {
+      "epoch": 5.023835716905023,
+      "grad_norm": 1.110278844833374,
+      "learning_rate": 0.0002,
+      "loss": 0.5194,
+      "step": 6850
+    },
+    {
+      "epoch": 5.031169783645031,
+      "grad_norm": 1.0670180320739746,
+      "learning_rate": 0.0002,
+      "loss": 0.4676,
+      "step": 6860
+    },
+    {
+      "epoch": 5.038503850385038,
+      "grad_norm": 0.8762092590332031,
+      "learning_rate": 0.0002,
+      "loss": 0.4374,
+      "step": 6870
+    },
+    {
+      "epoch": 5.045837917125046,
+      "grad_norm": 1.1169432401657104,
+      "learning_rate": 0.0002,
+      "loss": 0.505,
+      "step": 6880
+    },
+    {
+      "epoch": 5.053171983865053,
+      "grad_norm": 1.005491018295288,
+      "learning_rate": 0.0002,
+      "loss": 0.5114,
+      "step": 6890
+    },
+    {
+      "epoch": 5.060506050605061,
+      "grad_norm": 1.1751841306686401,
+      "learning_rate": 0.0002,
+      "loss": 0.5221,
+      "step": 6900
+    },
+    {
+      "epoch": 5.067840117345068,
+      "grad_norm": 0.8501367568969727,
+      "learning_rate": 0.0002,
+      "loss": 0.451,
+      "step": 6910
+    },
+    {
+      "epoch": 5.075174184085075,
+      "grad_norm": 0.9795131683349609,
+      "learning_rate": 0.0002,
+      "loss": 0.5292,
+      "step": 6920
+    },
+    {
+      "epoch": 5.082508250825082,
+      "grad_norm": 0.8929879665374756,
+      "learning_rate": 0.0002,
+      "loss": 0.5234,
+      "step": 6930
+    },
+    {
+      "epoch": 5.08984231756509,
+      "grad_norm": 1.0156651735305786,
+      "learning_rate": 0.0002,
+      "loss": 0.5378,
+      "step": 6940
+    },
+    {
+      "epoch": 5.097176384305097,
+      "grad_norm": 1.0974335670471191,
+      "learning_rate": 0.0002,
+      "loss": 0.5241,
+      "step": 6950
+    },
+    {
+      "epoch": 5.104510451045105,
+      "grad_norm": 1.7015666961669922,
+      "learning_rate": 0.0002,
+      "loss": 0.5705,
+      "step": 6960
+    },
+    {
+      "epoch": 5.111844517785112,
+      "grad_norm": 1.0343226194381714,
+      "learning_rate": 0.0002,
+      "loss": 0.523,
+      "step": 6970
+    },
+    {
+      "epoch": 5.119178584525119,
+      "grad_norm": 1.3072983026504517,
+      "learning_rate": 0.0002,
+      "loss": 0.4616,
+      "step": 6980
+    },
+    {
+      "epoch": 5.126512651265126,
+      "grad_norm": 1.038986086845398,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 6990
+    },
+    {
+      "epoch": 5.133846718005134,
+      "grad_norm": 0.8638386130332947,
+      "learning_rate": 0.0002,
+      "loss": 0.4616,
+      "step": 7000
+    },
+    {
+      "epoch": 5.141180784745141,
+      "grad_norm": 0.8326523900032043,
+      "learning_rate": 0.0002,
+      "loss": 0.5294,
+      "step": 7010
+    },
+    {
+      "epoch": 5.148514851485149,
+      "grad_norm": 1.0976895093917847,
+      "learning_rate": 0.0002,
+      "loss": 0.5021,
+      "step": 7020
+    },
+    {
+      "epoch": 5.155848918225156,
+      "grad_norm": 1.0077873468399048,
+      "learning_rate": 0.0002,
+      "loss": 0.4677,
+      "step": 7030
+    },
+    {
+      "epoch": 5.163182984965164,
+      "grad_norm": 1.0662257671356201,
+      "learning_rate": 0.0002,
+      "loss": 0.5262,
+      "step": 7040
+    },
+    {
+      "epoch": 5.17051705170517,
+      "grad_norm": 1.206271767616272,
+      "learning_rate": 0.0002,
+      "loss": 0.5484,
+      "step": 7050
+    },
+    {
+      "epoch": 5.177851118445178,
+      "grad_norm": 1.1990262269973755,
+      "learning_rate": 0.0002,
+      "loss": 0.4817,
+      "step": 7060
+    },
+    {
+      "epoch": 5.185185185185185,
+      "grad_norm": 1.0207163095474243,
+      "learning_rate": 0.0002,
+      "loss": 0.6048,
+      "step": 7070
+    },
+    {
+      "epoch": 5.192519251925193,
+      "grad_norm": 1.2783987522125244,
+      "learning_rate": 0.0002,
+      "loss": 0.4816,
+      "step": 7080
+    },
+    {
+      "epoch": 5.1998533186652,
+      "grad_norm": 1.1592512130737305,
+      "learning_rate": 0.0002,
+      "loss": 0.5322,
+      "step": 7090
+    },
+    {
+      "epoch": 5.2071873854052075,
+      "grad_norm": 1.1053160429000854,
+      "learning_rate": 0.0002,
+      "loss": 0.5472,
+      "step": 7100
+    },
+    {
+      "epoch": 5.214521452145214,
+      "grad_norm": 1.1925510168075562,
+      "learning_rate": 0.0002,
+      "loss": 0.4986,
+      "step": 7110
+    },
+    {
+      "epoch": 5.221855518885222,
+      "grad_norm": 1.0714877843856812,
+      "learning_rate": 0.0002,
+      "loss": 0.5065,
+      "step": 7120
+    },
+    {
+      "epoch": 5.229189585625229,
+      "grad_norm": 0.9451011419296265,
+      "learning_rate": 0.0002,
+      "loss": 0.5209,
+      "step": 7130
+    },
+    {
+      "epoch": 5.2365236523652365,
+      "grad_norm": 1.03838050365448,
+      "learning_rate": 0.0002,
+      "loss": 0.5298,
+      "step": 7140
+    },
+    {
+      "epoch": 5.243857719105244,
+      "grad_norm": 0.9204146265983582,
+      "learning_rate": 0.0002,
+      "loss": 0.4848,
+      "step": 7150
+    },
+    {
+      "epoch": 5.2511917858452515,
+      "grad_norm": 1.0142229795455933,
+      "learning_rate": 0.0002,
+      "loss": 0.5164,
+      "step": 7160
+    },
+    {
+      "epoch": 5.258525852585258,
+      "grad_norm": 1.4432005882263184,
+      "learning_rate": 0.0002,
+      "loss": 0.5092,
+      "step": 7170
+    },
+    {
+      "epoch": 5.2658599193252655,
+      "grad_norm": 1.1239633560180664,
+      "learning_rate": 0.0002,
+      "loss": 0.5133,
+      "step": 7180
+    },
+    {
+      "epoch": 5.273193986065273,
+      "grad_norm": 0.7012821435928345,
+      "learning_rate": 0.0002,
+      "loss": 0.4969,
+      "step": 7190
+    },
+    {
+      "epoch": 5.2805280528052805,
+      "grad_norm": 1.3499128818511963,
+      "learning_rate": 0.0002,
+      "loss": 0.5466,
+      "step": 7200
+    },
+    {
+      "epoch": 5.287862119545288,
+      "grad_norm": 0.9498730897903442,
+      "learning_rate": 0.0002,
+      "loss": 0.5282,
+      "step": 7210
+    },
+    {
+      "epoch": 5.295196186285295,
+      "grad_norm": 0.9552369117736816,
+      "learning_rate": 0.0002,
+      "loss": 0.5051,
+      "step": 7220
+    },
+    {
+      "epoch": 5.302530253025303,
+      "grad_norm": 0.7610348463058472,
+      "learning_rate": 0.0002,
+      "loss": 0.5329,
+      "step": 7230
+    },
+    {
+      "epoch": 5.3098643197653095,
+      "grad_norm": 1.0314512252807617,
+      "learning_rate": 0.0002,
+      "loss": 0.468,
+      "step": 7240
+    },
+    {
+      "epoch": 5.317198386505317,
+      "grad_norm": 1.0534334182739258,
+      "learning_rate": 0.0002,
+      "loss": 0.5367,
+      "step": 7250
+    },
+    {
+      "epoch": 5.324532453245324,
+      "grad_norm": 1.2553406953811646,
+      "learning_rate": 0.0002,
+      "loss": 0.5491,
+      "step": 7260
+    },
+    {
+      "epoch": 5.331866519985332,
+      "grad_norm": 0.7061691880226135,
+      "learning_rate": 0.0002,
+      "loss": 0.5218,
+      "step": 7270
+    },
+    {
+      "epoch": 5.339200586725339,
+      "grad_norm": 0.9652578830718994,
+      "learning_rate": 0.0002,
+      "loss": 0.5625,
+      "step": 7280
+    },
+    {
+      "epoch": 5.346534653465347,
+      "grad_norm": 1.114788293838501,
+      "learning_rate": 0.0002,
+      "loss": 0.5608,
+      "step": 7290
+    },
+    {
+      "epoch": 5.353868720205353,
+      "grad_norm": 1.0940049886703491,
+      "learning_rate": 0.0002,
+      "loss": 0.578,
+      "step": 7300
+    },
+    {
+      "epoch": 5.361202786945361,
+      "grad_norm": 1.0151008367538452,
+      "learning_rate": 0.0002,
+      "loss": 0.5256,
+      "step": 7310
+    },
+    {
+      "epoch": 5.368536853685368,
+      "grad_norm": 1.0369552373886108,
+      "learning_rate": 0.0002,
+      "loss": 0.5377,
+      "step": 7320
+    },
+    {
+      "epoch": 5.375870920425376,
+      "grad_norm": 0.8489866256713867,
+      "learning_rate": 0.0002,
+      "loss": 0.5028,
+      "step": 7330
+    },
+    {
+      "epoch": 5.383204987165383,
+      "grad_norm": 1.1031713485717773,
+      "learning_rate": 0.0002,
+      "loss": 0.5937,
+      "step": 7340
+    },
+    {
+      "epoch": 5.390539053905391,
+      "grad_norm": 0.9094716310501099,
+      "learning_rate": 0.0002,
+      "loss": 0.5355,
+      "step": 7350
+    },
+    {
+      "epoch": 5.397873120645398,
+      "grad_norm": 0.9530431032180786,
+      "learning_rate": 0.0002,
+      "loss": 0.5406,
+      "step": 7360
+    },
+    {
+      "epoch": 5.405207187385405,
+      "grad_norm": 0.9633604884147644,
+      "learning_rate": 0.0002,
+      "loss": 0.529,
+      "step": 7370
+    },
+    {
+      "epoch": 5.412541254125412,
+      "grad_norm": 0.9541662335395813,
+      "learning_rate": 0.0002,
+      "loss": 0.5315,
+      "step": 7380
+    },
+    {
+      "epoch": 5.41987532086542,
+      "grad_norm": 1.0459771156311035,
+      "learning_rate": 0.0002,
+      "loss": 0.6774,
+      "step": 7390
+    },
+    {
+      "epoch": 5.427209387605427,
+      "grad_norm": 1.027388334274292,
+      "learning_rate": 0.0002,
+      "loss": 0.5737,
+      "step": 7400
+    },
+    {
+      "epoch": 5.434543454345435,
+      "grad_norm": 0.7267653346061707,
+      "learning_rate": 0.0002,
+      "loss": 0.556,
+      "step": 7410
+    },
+    {
+      "epoch": 5.441877521085442,
+      "grad_norm": 1.020142674446106,
+      "learning_rate": 0.0002,
+      "loss": 0.4581,
+      "step": 7420
+    },
+    {
+      "epoch": 5.449211587825449,
+      "grad_norm": 1.044754147529602,
+      "learning_rate": 0.0002,
+      "loss": 0.4853,
+      "step": 7430
+    },
+    {
+      "epoch": 5.456545654565456,
+      "grad_norm": 1.5476195812225342,
+      "learning_rate": 0.0002,
+      "loss": 0.5666,
+      "step": 7440
+    },
+    {
+      "epoch": 5.463879721305464,
+      "grad_norm": 0.9879506826400757,
+      "learning_rate": 0.0002,
+      "loss": 0.5302,
+      "step": 7450
+    },
+    {
+      "epoch": 5.471213788045471,
+      "grad_norm": 1.2562980651855469,
+      "learning_rate": 0.0002,
+      "loss": 0.591,
+      "step": 7460
+    },
+    {
+      "epoch": 5.478547854785479,
+      "grad_norm": 1.3051384687423706,
+      "learning_rate": 0.0002,
+      "loss": 0.5188,
+      "step": 7470
+    },
+    {
+      "epoch": 5.485881921525486,
+      "grad_norm": 1.0511597394943237,
+      "learning_rate": 0.0002,
+      "loss": 0.5658,
+      "step": 7480
+    },
+    {
+      "epoch": 5.493215988265494,
+      "grad_norm": 1.0380817651748657,
+      "learning_rate": 0.0002,
+      "loss": 0.6327,
+      "step": 7490
+    },
+    {
+      "epoch": 5.5005500550055,
+      "grad_norm": 1.170274257659912,
+      "learning_rate": 0.0002,
+      "loss": 0.5356,
+      "step": 7500
+    },
+    {
+      "epoch": 5.507884121745508,
+      "grad_norm": 1.3356517553329468,
+      "learning_rate": 0.0002,
+      "loss": 0.5405,
+      "step": 7510
+    },
+    {
+      "epoch": 5.515218188485515,
+      "grad_norm": 1.0727124214172363,
+      "learning_rate": 0.0002,
+      "loss": 0.5305,
+      "step": 7520
+    },
+    {
+      "epoch": 5.522552255225523,
+      "grad_norm": 1.0110199451446533,
+      "learning_rate": 0.0002,
+      "loss": 0.5543,
+      "step": 7530
+    },
+    {
+      "epoch": 5.52988632196553,
+      "grad_norm": 1.3086743354797363,
+      "learning_rate": 0.0002,
+      "loss": 0.5962,
+      "step": 7540
+    },
+    {
+      "epoch": 5.537220388705538,
+      "grad_norm": 1.1904916763305664,
+      "learning_rate": 0.0002,
+      "loss": 0.5512,
+      "step": 7550
+    },
+    {
+      "epoch": 5.544554455445544,
+      "grad_norm": 0.9466280937194824,
+      "learning_rate": 0.0002,
+      "loss": 0.5915,
+      "step": 7560
+    },
+    {
+      "epoch": 5.551888522185552,
+      "grad_norm": 1.1237901449203491,
+      "learning_rate": 0.0002,
+      "loss": 0.5573,
+      "step": 7570
+    },
+    {
+      "epoch": 5.559222588925559,
+      "grad_norm": 0.9590660333633423,
+      "learning_rate": 0.0002,
+      "loss": 0.5383,
+      "step": 7580
+    },
+    {
+      "epoch": 5.566556655665567,
+      "grad_norm": 1.0890778303146362,
+      "learning_rate": 0.0002,
+      "loss": 0.5594,
+      "step": 7590
+    },
+    {
+      "epoch": 5.573890722405574,
+      "grad_norm": 0.7206931114196777,
+      "learning_rate": 0.0002,
+      "loss": 0.5698,
+      "step": 7600
+    },
+    {
+      "epoch": 5.5812247891455815,
+      "grad_norm": 1.2884514331817627,
+      "learning_rate": 0.0002,
+      "loss": 0.5511,
+      "step": 7610
+    },
+    {
+      "epoch": 5.588558855885589,
+      "grad_norm": 0.7798039317131042,
+      "learning_rate": 0.0002,
+      "loss": 0.5279,
+      "step": 7620
+    },
+    {
+      "epoch": 5.595892922625596,
+      "grad_norm": 1.166046142578125,
+      "learning_rate": 0.0002,
+      "loss": 0.4847,
+      "step": 7630
+    },
+    {
+      "epoch": 5.603226989365603,
+      "grad_norm": 1.0150201320648193,
+      "learning_rate": 0.0002,
+      "loss": 0.5821,
+      "step": 7640
+    },
+    {
+      "epoch": 5.6105610561056105,
+      "grad_norm": 1.0449682474136353,
+      "learning_rate": 0.0002,
+      "loss": 0.5296,
+      "step": 7650
+    },
+    {
+      "epoch": 5.617895122845618,
+      "grad_norm": 0.9310530424118042,
+      "learning_rate": 0.0002,
+      "loss": 0.5431,
+      "step": 7660
+    },
+    {
+      "epoch": 5.6252291895856255,
+      "grad_norm": 0.9117933511734009,
+      "learning_rate": 0.0002,
+      "loss": 0.5234,
+      "step": 7670
+    },
+    {
+      "epoch": 5.632563256325633,
+      "grad_norm": 1.1475164890289307,
+      "learning_rate": 0.0002,
+      "loss": 0.5807,
+      "step": 7680
+    },
+    {
+      "epoch": 5.6398973230656395,
+      "grad_norm": 1.066809058189392,
+      "learning_rate": 0.0002,
+      "loss": 0.5816,
+      "step": 7690
+    },
+    {
+      "epoch": 5.647231389805647,
+      "grad_norm": 1.2834991216659546,
+      "learning_rate": 0.0002,
+      "loss": 0.551,
+      "step": 7700
+    },
+    {
+      "epoch": 5.6545654565456545,
+      "grad_norm": 1.2245112657546997,
+      "learning_rate": 0.0002,
+      "loss": 0.5914,
+      "step": 7710
+    },
+    {
+      "epoch": 5.661899523285662,
+      "grad_norm": 1.1424106359481812,
+      "learning_rate": 0.0002,
+      "loss": 0.5552,
+      "step": 7720
+    },
+    {
+      "epoch": 5.669233590025669,
+      "grad_norm": 1.0673892498016357,
+      "learning_rate": 0.0002,
+      "loss": 0.559,
+      "step": 7730
+    },
+    {
+      "epoch": 5.676567656765677,
+      "grad_norm": 1.4312121868133545,
+      "learning_rate": 0.0002,
+      "loss": 0.544,
+      "step": 7740
+    },
+    {
+      "epoch": 5.683901723505684,
+      "grad_norm": 0.9976982474327087,
+      "learning_rate": 0.0002,
+      "loss": 0.5576,
+      "step": 7750
+    },
+    {
+      "epoch": 5.691235790245691,
+      "grad_norm": 0.9464678168296814,
+      "learning_rate": 0.0002,
+      "loss": 0.4855,
+      "step": 7760
+    },
+    {
+      "epoch": 5.698569856985698,
+      "grad_norm": 1.010995626449585,
+      "learning_rate": 0.0002,
+      "loss": 0.5363,
+      "step": 7770
+    },
+    {
+      "epoch": 5.705903923725706,
+      "grad_norm": 1.3787750005722046,
+      "learning_rate": 0.0002,
+      "loss": 0.5873,
+      "step": 7780
+    },
+    {
+      "epoch": 5.713237990465713,
+      "grad_norm": 1.020922303199768,
+      "learning_rate": 0.0002,
+      "loss": 0.6234,
+      "step": 7790
+    },
+    {
+      "epoch": 5.720572057205721,
+      "grad_norm": 0.9748636484146118,
+      "learning_rate": 0.0002,
+      "loss": 0.5337,
+      "step": 7800
+    },
+    {
+      "epoch": 5.727906123945728,
+      "grad_norm": 1.3077744245529175,
+      "learning_rate": 0.0002,
+      "loss": 0.5507,
+      "step": 7810
+    },
+    {
+      "epoch": 5.735240190685735,
+      "grad_norm": 1.4770057201385498,
+      "learning_rate": 0.0002,
+      "loss": 0.558,
+      "step": 7820
+    },
+    {
+      "epoch": 5.742574257425742,
+      "grad_norm": 1.6349090337753296,
+      "learning_rate": 0.0002,
+      "loss": 0.5571,
+      "step": 7830
+    },
+    {
+      "epoch": 5.74990832416575,
+      "grad_norm": 0.9818630814552307,
+      "learning_rate": 0.0002,
+      "loss": 0.5056,
+      "step": 7840
+    },
+    {
+      "epoch": 5.757242390905757,
+      "grad_norm": 0.9659715890884399,
+      "learning_rate": 0.0002,
+      "loss": 0.5495,
+      "step": 7850
+    },
+    {
+      "epoch": 5.764576457645765,
+      "grad_norm": 0.9269950985908508,
+      "learning_rate": 0.0002,
+      "loss": 0.5628,
+      "step": 7860
+    },
+    {
+      "epoch": 5.771910524385772,
+      "grad_norm": 1.0099073648452759,
+      "learning_rate": 0.0002,
+      "loss": 0.5594,
+      "step": 7870
+    },
+    {
+      "epoch": 5.77924459112578,
+      "grad_norm": 0.9123615026473999,
+      "learning_rate": 0.0002,
+      "loss": 0.5912,
+      "step": 7880
+    },
+    {
+      "epoch": 5.786578657865786,
+      "grad_norm": 1.1542246341705322,
+      "learning_rate": 0.0002,
+      "loss": 0.6054,
+      "step": 7890
+    },
+    {
+      "epoch": 5.793912724605794,
+      "grad_norm": 1.0792022943496704,
+      "learning_rate": 0.0002,
+      "loss": 0.5829,
+      "step": 7900
+    },
+    {
+      "epoch": 5.801246791345801,
+      "grad_norm": 0.95615553855896,
+      "learning_rate": 0.0002,
+      "loss": 0.504,
+      "step": 7910
+    },
+    {
+      "epoch": 5.808580858085809,
+      "grad_norm": 1.2471332550048828,
+      "learning_rate": 0.0002,
+      "loss": 0.5918,
+      "step": 7920
+    },
+    {
+      "epoch": 5.815914924825816,
+      "grad_norm": 1.0189851522445679,
+      "learning_rate": 0.0002,
+      "loss": 0.5719,
+      "step": 7930
+    },
+    {
+      "epoch": 5.823248991565823,
+      "grad_norm": 1.3309742212295532,
+      "learning_rate": 0.0002,
+      "loss": 0.5958,
+      "step": 7940
+    },
+    {
+      "epoch": 5.83058305830583,
+      "grad_norm": 1.2930549383163452,
+      "learning_rate": 0.0002,
+      "loss": 0.6255,
+      "step": 7950
+    },
+    {
+      "epoch": 5.837917125045838,
+      "grad_norm": 0.8216308951377869,
+      "learning_rate": 0.0002,
+      "loss": 0.5301,
+      "step": 7960
+    },
+    {
+      "epoch": 5.845251191785845,
+      "grad_norm": 1.1205775737762451,
+      "learning_rate": 0.0002,
+      "loss": 0.5397,
+      "step": 7970
+    },
+    {
+      "epoch": 5.852585258525853,
+      "grad_norm": 0.851298451423645,
+      "learning_rate": 0.0002,
+      "loss": 0.5903,
+      "step": 7980
+    },
+    {
+      "epoch": 5.85991932526586,
+      "grad_norm": 0.8797095417976379,
+      "learning_rate": 0.0002,
+      "loss": 0.5981,
+      "step": 7990
+    },
+    {
+      "epoch": 5.867253392005868,
+      "grad_norm": 1.5784614086151123,
+      "learning_rate": 0.0002,
+      "loss": 0.6106,
+      "step": 8000
+    },
+    {
+      "epoch": 5.874587458745875,
+      "grad_norm": 1.1531187295913696,
+      "learning_rate": 0.0002,
+      "loss": 0.5956,
+      "step": 8010
+    },
+    {
+      "epoch": 5.881921525485882,
+      "grad_norm": 1.2469146251678467,
+      "learning_rate": 0.0002,
+      "loss": 0.6289,
+      "step": 8020
+    },
+    {
+      "epoch": 5.889255592225889,
+      "grad_norm": 1.0784350633621216,
+      "learning_rate": 0.0002,
+      "loss": 0.5827,
+      "step": 8030
+    },
+    {
+      "epoch": 5.896589658965897,
+      "grad_norm": 1.1311599016189575,
+      "learning_rate": 0.0002,
+      "loss": 0.6339,
+      "step": 8040
+    },
+    {
+      "epoch": 5.903923725705904,
+      "grad_norm": 0.9654512405395508,
+      "learning_rate": 0.0002,
+      "loss": 0.5815,
+      "step": 8050
+    },
+    {
+      "epoch": 5.9112577924459115,
+      "grad_norm": 1.3288270235061646,
+      "learning_rate": 0.0002,
+      "loss": 0.6198,
+      "step": 8060
+    },
+    {
+      "epoch": 5.918591859185918,
+      "grad_norm": 1.12800931930542,
+      "learning_rate": 0.0002,
+      "loss": 0.6515,
+      "step": 8070
+    },
+    {
+      "epoch": 5.925925925925926,
+      "grad_norm": 0.9449917674064636,
+      "learning_rate": 0.0002,
+      "loss": 0.5684,
+      "step": 8080
+    },
+    {
+      "epoch": 5.933259992665933,
+      "grad_norm": 1.1532357931137085,
+      "learning_rate": 0.0002,
+      "loss": 0.6063,
+      "step": 8090
+    },
+    {
+      "epoch": 5.9405940594059405,
+      "grad_norm": 1.2211151123046875,
+      "learning_rate": 0.0002,
+      "loss": 0.5318,
+      "step": 8100
+    },
+    {
+      "epoch": 5.947928126145948,
+      "grad_norm": 1.3459105491638184,
+      "learning_rate": 0.0002,
+      "loss": 0.6512,
+      "step": 8110
+    },
+    {
+      "epoch": 5.9552621928859555,
+      "grad_norm": 1.251999855041504,
+      "learning_rate": 0.0002,
+      "loss": 0.5952,
+      "step": 8120
+    },
+    {
+      "epoch": 5.962596259625963,
+      "grad_norm": 1.5682506561279297,
+      "learning_rate": 0.0002,
+      "loss": 0.6203,
+      "step": 8130
+    },
+    {
+      "epoch": 5.9699303263659695,
+      "grad_norm": 0.926075279712677,
+      "learning_rate": 0.0002,
+      "loss": 0.6253,
+      "step": 8140
+    },
+    {
+      "epoch": 5.977264393105977,
+      "grad_norm": 0.9622511863708496,
+      "learning_rate": 0.0002,
+      "loss": 0.5545,
+      "step": 8150
+    },
+    {
+      "epoch": 5.9845984598459845,
+      "grad_norm": 0.9633373618125916,
+      "learning_rate": 0.0002,
+      "loss": 0.5518,
+      "step": 8160
+    },
+    {
+      "epoch": 5.991932526585992,
+      "grad_norm": 0.8960476517677307,
+      "learning_rate": 0.0002,
+      "loss": 0.5831,
+      "step": 8170
+    },
+    {
+      "epoch": 5.999266593325999,
+      "grad_norm": 0.9372805953025818,
+      "learning_rate": 0.0002,
+      "loss": 0.5442,
+      "step": 8180
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.3233846426010132,
+      "eval_runtime": 32.7419,
+      "eval_samples_per_second": 13.164,
+      "eval_steps_per_second": 1.649,
+      "step": 8181
+    },
+    {
+      "epoch": 6.006600660066007,
+      "grad_norm": 1.1900787353515625,
+      "learning_rate": 0.0002,
+      "loss": 0.4644,
+      "step": 8190
+    },
+    {
+      "epoch": 6.013934726806014,
+      "grad_norm": 1.1448326110839844,
+      "learning_rate": 0.0002,
+      "loss": 0.4509,
+      "step": 8200
+    },
+    {
+      "epoch": 6.021268793546021,
+      "grad_norm": 1.1848368644714355,
+      "learning_rate": 0.0002,
+      "loss": 0.3667,
+      "step": 8210
+    },
+    {
+      "epoch": 6.028602860286028,
+      "grad_norm": 1.2315572500228882,
+      "learning_rate": 0.0002,
+      "loss": 0.4315,
+      "step": 8220
+    },
+    {
+      "epoch": 6.035936927026036,
+      "grad_norm": 1.2214244604110718,
+      "learning_rate": 0.0002,
+      "loss": 0.3541,
+      "step": 8230
+    },
+    {
+      "epoch": 6.043270993766043,
+      "grad_norm": 0.9455513954162598,
+      "learning_rate": 0.0002,
+      "loss": 0.4025,
+      "step": 8240
+    },
+    {
+      "epoch": 6.050605060506051,
+      "grad_norm": 0.9574248790740967,
+      "learning_rate": 0.0002,
+      "loss": 0.4448,
+      "step": 8250
+    },
+    {
+      "epoch": 6.057939127246058,
+      "grad_norm": 1.1022400856018066,
+      "learning_rate": 0.0002,
+      "loss": 0.4271,
+      "step": 8260
+    },
+    {
+      "epoch": 6.065273193986065,
+      "grad_norm": 0.9555122256278992,
+      "learning_rate": 0.0002,
+      "loss": 0.3603,
+      "step": 8270
+    },
+    {
+      "epoch": 6.072607260726072,
+      "grad_norm": 1.1956106424331665,
+      "learning_rate": 0.0002,
+      "loss": 0.4324,
+      "step": 8280
+    },
+    {
+      "epoch": 6.07994132746608,
+      "grad_norm": 1.3110876083374023,
+      "learning_rate": 0.0002,
+      "loss": 0.3924,
+      "step": 8290
+    },
+    {
+      "epoch": 6.087275394206087,
+      "grad_norm": 1.1293374300003052,
+      "learning_rate": 0.0002,
+      "loss": 0.3664,
+      "step": 8300
+    },
+    {
+      "epoch": 6.094609460946095,
+      "grad_norm": 0.9176164269447327,
+      "learning_rate": 0.0002,
+      "loss": 0.385,
+      "step": 8310
+    },
+    {
+      "epoch": 6.101943527686102,
+      "grad_norm": 0.9751231670379639,
+      "learning_rate": 0.0002,
+      "loss": 0.4142,
+      "step": 8320
+    },
+    {
+      "epoch": 6.109277594426109,
+      "grad_norm": 1.0536044836044312,
+      "learning_rate": 0.0002,
+      "loss": 0.4356,
+      "step": 8330
+    },
+    {
+      "epoch": 6.116611661166116,
+      "grad_norm": 1.289342999458313,
+      "learning_rate": 0.0002,
+      "loss": 0.409,
+      "step": 8340
+    },
+    {
+      "epoch": 6.123945727906124,
+      "grad_norm": 1.1773661375045776,
+      "learning_rate": 0.0002,
+      "loss": 0.4121,
+      "step": 8350
+    },
+    {
+      "epoch": 6.131279794646131,
+      "grad_norm": 1.2450661659240723,
+      "learning_rate": 0.0002,
+      "loss": 0.4499,
+      "step": 8360
+    },
+    {
+      "epoch": 6.138613861386139,
+      "grad_norm": 1.3965914249420166,
+      "learning_rate": 0.0002,
+      "loss": 0.4467,
+      "step": 8370
+    },
+    {
+      "epoch": 6.145947928126146,
+      "grad_norm": 1.3530808687210083,
+      "learning_rate": 0.0002,
+      "loss": 0.4024,
+      "step": 8380
+    },
+    {
+      "epoch": 6.153281994866154,
+      "grad_norm": 1.296276330947876,
+      "learning_rate": 0.0002,
+      "loss": 0.4658,
+      "step": 8390
+    },
+    {
+      "epoch": 6.16061606160616,
+      "grad_norm": 0.9759053587913513,
+      "learning_rate": 0.0002,
+      "loss": 0.5073,
+      "step": 8400
+    },
+    {
+      "epoch": 6.167950128346168,
+      "grad_norm": 1.2110707759857178,
+      "learning_rate": 0.0002,
+      "loss": 0.4718,
+      "step": 8410
+    },
+    {
+      "epoch": 6.175284195086175,
+      "grad_norm": 1.312226414680481,
+      "learning_rate": 0.0002,
+      "loss": 0.4453,
+      "step": 8420
+    },
+    {
+      "epoch": 6.182618261826183,
+      "grad_norm": 1.1696736812591553,
+      "learning_rate": 0.0002,
+      "loss": 0.4183,
+      "step": 8430
+    },
+    {
+      "epoch": 6.18995232856619,
+      "grad_norm": 1.260304570198059,
+      "learning_rate": 0.0002,
+      "loss": 0.4546,
+      "step": 8440
+    },
+    {
+      "epoch": 6.197286395306198,
+      "grad_norm": 1.472961187362671,
+      "learning_rate": 0.0002,
+      "loss": 0.4137,
+      "step": 8450
+    },
+    {
+      "epoch": 6.204620462046204,
+      "grad_norm": 1.3618475198745728,
+      "learning_rate": 0.0002,
+      "loss": 0.42,
+      "step": 8460
+    },
+    {
+      "epoch": 6.211954528786212,
+      "grad_norm": 1.2544318437576294,
+      "learning_rate": 0.0002,
+      "loss": 0.415,
+      "step": 8470
+    },
+    {
+      "epoch": 6.219288595526219,
+      "grad_norm": 1.205898642539978,
+      "learning_rate": 0.0002,
+      "loss": 0.3907,
+      "step": 8480
+    },
+    {
+      "epoch": 6.226622662266227,
+      "grad_norm": 0.9984724521636963,
+      "learning_rate": 0.0002,
+      "loss": 0.4431,
+      "step": 8490
+    },
+    {
+      "epoch": 6.233956729006234,
+      "grad_norm": 1.3184109926223755,
+      "learning_rate": 0.0002,
+      "loss": 0.4768,
+      "step": 8500
+    },
+    {
+      "epoch": 6.241290795746242,
+      "grad_norm": 1.135520100593567,
+      "learning_rate": 0.0002,
+      "loss": 0.3859,
+      "step": 8510
+    },
+    {
+      "epoch": 6.248624862486249,
+      "grad_norm": 1.4528400897979736,
+      "learning_rate": 0.0002,
+      "loss": 0.4159,
+      "step": 8520
+    },
+    {
+      "epoch": 6.255958929226256,
+      "grad_norm": 1.1222716569900513,
+      "learning_rate": 0.0002,
+      "loss": 0.4347,
+      "step": 8530
+    },
+    {
+      "epoch": 6.263292995966263,
+      "grad_norm": 1.7878046035766602,
+      "learning_rate": 0.0002,
+      "loss": 0.4581,
+      "step": 8540
+    },
+    {
+      "epoch": 6.270627062706271,
+      "grad_norm": 0.9789481163024902,
+      "learning_rate": 0.0002,
+      "loss": 0.4298,
+      "step": 8550
+    },
+    {
+      "epoch": 6.277961129446278,
+      "grad_norm": 1.151977300643921,
+      "learning_rate": 0.0002,
+      "loss": 0.4316,
+      "step": 8560
+    },
+    {
+      "epoch": 6.2852951961862855,
+      "grad_norm": 1.389968752861023,
+      "learning_rate": 0.0002,
+      "loss": 0.428,
+      "step": 8570
+    },
+    {
+      "epoch": 6.292629262926293,
+      "grad_norm": 0.884211003780365,
+      "learning_rate": 0.0002,
+      "loss": 0.3903,
+      "step": 8580
+    },
+    {
+      "epoch": 6.2999633296663,
+      "grad_norm": 1.3604296445846558,
+      "learning_rate": 0.0002,
+      "loss": 0.4611,
+      "step": 8590
+    },
+    {
+      "epoch": 6.307297396406307,
+      "grad_norm": 1.1845694780349731,
+      "learning_rate": 0.0002,
+      "loss": 0.4183,
+      "step": 8600
+    },
+    {
+      "epoch": 6.3146314631463145,
+      "grad_norm": 1.3231550455093384,
+      "learning_rate": 0.0002,
+      "loss": 0.472,
+      "step": 8610
+    },
+    {
+      "epoch": 6.321965529886322,
+      "grad_norm": 0.9546721577644348,
+      "learning_rate": 0.0002,
+      "loss": 0.3922,
+      "step": 8620
+    },
+    {
+      "epoch": 6.3292995966263295,
+      "grad_norm": 1.2329787015914917,
+      "learning_rate": 0.0002,
+      "loss": 0.4395,
+      "step": 8630
+    },
+    {
+      "epoch": 6.336633663366337,
+      "grad_norm": 1.0240199565887451,
+      "learning_rate": 0.0002,
+      "loss": 0.4344,
+      "step": 8640
+    },
+    {
+      "epoch": 6.343967730106344,
+      "grad_norm": 1.1866962909698486,
+      "learning_rate": 0.0002,
+      "loss": 0.4529,
+      "step": 8650
+    },
+    {
+      "epoch": 6.351301796846351,
+      "grad_norm": 1.2819687128067017,
+      "learning_rate": 0.0002,
+      "loss": 0.4575,
+      "step": 8660
+    },
+    {
+      "epoch": 6.3586358635863585,
+      "grad_norm": 0.9654944539070129,
+      "learning_rate": 0.0002,
+      "loss": 0.455,
+      "step": 8670
+    },
+    {
+      "epoch": 6.365969930326366,
+      "grad_norm": 0.9443874955177307,
+      "learning_rate": 0.0002,
+      "loss": 0.4739,
+      "step": 8680
+    },
+    {
+      "epoch": 6.373303997066373,
+      "grad_norm": 1.2914115190505981,
+      "learning_rate": 0.0002,
+      "loss": 0.435,
+      "step": 8690
+    },
+    {
+      "epoch": 6.380638063806381,
+      "grad_norm": 1.4558709859848022,
+      "learning_rate": 0.0002,
+      "loss": 0.4392,
+      "step": 8700
+    },
+    {
+      "epoch": 6.387972130546388,
+      "grad_norm": 1.3255952596664429,
+      "learning_rate": 0.0002,
+      "loss": 0.4398,
+      "step": 8710
+    },
+    {
+      "epoch": 6.395306197286395,
+      "grad_norm": 1.348742961883545,
+      "learning_rate": 0.0002,
+      "loss": 0.4451,
+      "step": 8720
+    },
+    {
+      "epoch": 6.402640264026402,
+      "grad_norm": 1.0096025466918945,
+      "learning_rate": 0.0002,
+      "loss": 0.41,
+      "step": 8730
+    },
+    {
+      "epoch": 6.40997433076641,
+      "grad_norm": 1.1720590591430664,
+      "learning_rate": 0.0002,
+      "loss": 0.4459,
+      "step": 8740
+    },
+    {
+      "epoch": 6.417308397506417,
+      "grad_norm": 1.1803077459335327,
+      "learning_rate": 0.0002,
+      "loss": 0.5059,
+      "step": 8750
+    },
+    {
+      "epoch": 6.424642464246425,
+      "grad_norm": 1.3649998903274536,
+      "learning_rate": 0.0002,
+      "loss": 0.4539,
+      "step": 8760
+    },
+    {
+      "epoch": 6.431976530986432,
+      "grad_norm": 1.1503992080688477,
+      "learning_rate": 0.0002,
+      "loss": 0.4171,
+      "step": 8770
+    },
+    {
+      "epoch": 6.43931059772644,
+      "grad_norm": 1.1537176370620728,
+      "learning_rate": 0.0002,
+      "loss": 0.488,
+      "step": 8780
+    },
+    {
+      "epoch": 6.446644664466446,
+      "grad_norm": 0.9743003845214844,
+      "learning_rate": 0.0002,
+      "loss": 0.4167,
+      "step": 8790
+    },
+    {
+      "epoch": 6.453978731206454,
+      "grad_norm": 0.9097744822502136,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 8800
+    },
+    {
+      "epoch": 6.461312797946461,
+      "grad_norm": 2.0174002647399902,
+      "learning_rate": 0.0002,
+      "loss": 0.4809,
+      "step": 8810
+    },
+    {
+      "epoch": 6.468646864686469,
+      "grad_norm": 1.0809309482574463,
+      "learning_rate": 0.0002,
+      "loss": 0.4879,
+      "step": 8820
+    },
+    {
+      "epoch": 6.475980931426476,
+      "grad_norm": 1.100294828414917,
+      "learning_rate": 0.0002,
+      "loss": 0.4235,
+      "step": 8830
+    },
+    {
+      "epoch": 6.483314998166484,
+      "grad_norm": 1.3707489967346191,
+      "learning_rate": 0.0002,
+      "loss": 0.4251,
+      "step": 8840
+    },
+    {
+      "epoch": 6.49064906490649,
+      "grad_norm": 1.1304761171340942,
+      "learning_rate": 0.0002,
+      "loss": 0.4533,
+      "step": 8850
+    },
+    {
+      "epoch": 6.497983131646498,
+      "grad_norm": 1.2171573638916016,
+      "learning_rate": 0.0002,
+      "loss": 0.4596,
+      "step": 8860
+    },
+    {
+      "epoch": 6.505317198386505,
+      "grad_norm": 1.0452901124954224,
+      "learning_rate": 0.0002,
+      "loss": 0.4694,
+      "step": 8870
+    },
+    {
+      "epoch": 6.512651265126513,
+      "grad_norm": 1.197298526763916,
+      "learning_rate": 0.0002,
+      "loss": 0.4855,
+      "step": 8880
+    },
+    {
+      "epoch": 6.51998533186652,
+      "grad_norm": 0.9179880619049072,
+      "learning_rate": 0.0002,
+      "loss": 0.4167,
+      "step": 8890
+    },
+    {
+      "epoch": 6.527319398606528,
+      "grad_norm": 1.415079951286316,
+      "learning_rate": 0.0002,
+      "loss": 0.445,
+      "step": 8900
+    },
+    {
+      "epoch": 6.534653465346535,
+      "grad_norm": 1.1032487154006958,
+      "learning_rate": 0.0002,
+      "loss": 0.424,
+      "step": 8910
+    },
+    {
+      "epoch": 6.541987532086542,
+      "grad_norm": 1.2295007705688477,
+      "learning_rate": 0.0002,
+      "loss": 0.4496,
+      "step": 8920
+    },
+    {
+      "epoch": 6.549321598826549,
+      "grad_norm": 1.4223219156265259,
+      "learning_rate": 0.0002,
+      "loss": 0.4755,
+      "step": 8930
+    },
+    {
+      "epoch": 6.556655665566557,
+      "grad_norm": 1.2785786390304565,
+      "learning_rate": 0.0002,
+      "loss": 0.4597,
+      "step": 8940
+    },
+    {
+      "epoch": 6.563989732306564,
+      "grad_norm": 1.3514775037765503,
+      "learning_rate": 0.0002,
+      "loss": 0.4651,
+      "step": 8950
+    },
+    {
+      "epoch": 6.571323799046572,
+      "grad_norm": 1.107937216758728,
+      "learning_rate": 0.0002,
+      "loss": 0.4961,
+      "step": 8960
+    },
+    {
+      "epoch": 6.578657865786578,
+      "grad_norm": 1.2839902639389038,
+      "learning_rate": 0.0002,
+      "loss": 0.4954,
+      "step": 8970
+    },
+    {
+      "epoch": 6.585991932526586,
+      "grad_norm": 0.9793244004249573,
+      "learning_rate": 0.0002,
+      "loss": 0.4207,
+      "step": 8980
+    },
+    {
+      "epoch": 6.593325999266593,
+      "grad_norm": 1.3403126001358032,
+      "learning_rate": 0.0002,
+      "loss": 0.4989,
+      "step": 8990
+    },
+    {
+      "epoch": 6.600660066006601,
+      "grad_norm": 1.2612813711166382,
+      "learning_rate": 0.0002,
+      "loss": 0.465,
+      "step": 9000
+    },
+    {
+      "epoch": 6.607994132746608,
+      "grad_norm": 1.4347625970840454,
+      "learning_rate": 0.0002,
+      "loss": 0.4589,
+      "step": 9010
+    },
+    {
+      "epoch": 6.6153281994866155,
+      "grad_norm": 1.225921869277954,
+      "learning_rate": 0.0002,
+      "loss": 0.4864,
+      "step": 9020
+    },
+    {
+      "epoch": 6.622662266226623,
+      "grad_norm": 1.033644676208496,
+      "learning_rate": 0.0002,
+      "loss": 0.4364,
+      "step": 9030
+    },
+    {
+      "epoch": 6.6299963329666305,
+      "grad_norm": 1.1791894435882568,
+      "learning_rate": 0.0002,
+      "loss": 0.4698,
+      "step": 9040
+    },
+    {
+      "epoch": 6.637330399706637,
+      "grad_norm": 1.0968137979507446,
+      "learning_rate": 0.0002,
+      "loss": 0.4908,
+      "step": 9050
+    },
+    {
+      "epoch": 6.6446644664466445,
+      "grad_norm": 1.5639140605926514,
+      "learning_rate": 0.0002,
+      "loss": 0.4346,
+      "step": 9060
+    },
+    {
+      "epoch": 6.651998533186652,
+      "grad_norm": 1.4158905744552612,
+      "learning_rate": 0.0002,
+      "loss": 0.4627,
+      "step": 9070
+    },
+    {
+      "epoch": 6.6593325999266595,
+      "grad_norm": 1.2120254039764404,
+      "learning_rate": 0.0002,
+      "loss": 0.4619,
+      "step": 9080
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.1866531372070312,
+      "learning_rate": 0.0002,
+      "loss": 0.4564,
+      "step": 9090
+    },
+    {
+      "epoch": 6.6740007334066735,
+      "grad_norm": 1.2704026699066162,
+      "learning_rate": 0.0002,
+      "loss": 0.5175,
+      "step": 9100
+    },
+    {
+      "epoch": 6.681334800146681,
+      "grad_norm": 1.1878353357315063,
+      "learning_rate": 0.0002,
+      "loss": 0.4859,
+      "step": 9110
+    },
+    {
+      "epoch": 6.6886688668866885,
+      "grad_norm": 1.193995475769043,
+      "learning_rate": 0.0002,
+      "loss": 0.4657,
+      "step": 9120
+    },
+    {
+      "epoch": 6.696002933626696,
+      "grad_norm": 1.2927545309066772,
+      "learning_rate": 0.0002,
+      "loss": 0.4939,
+      "step": 9130
+    },
+    {
+      "epoch": 6.703337000366703,
+      "grad_norm": 1.0770703554153442,
+      "learning_rate": 0.0002,
+      "loss": 0.4157,
+      "step": 9140
+    },
+    {
+      "epoch": 6.710671067106711,
+      "grad_norm": 1.2200851440429688,
+      "learning_rate": 0.0002,
+      "loss": 0.4571,
+      "step": 9150
+    },
+    {
+      "epoch": 6.718005133846718,
+      "grad_norm": 1.293891191482544,
+      "learning_rate": 0.0002,
+      "loss": 0.4605,
+      "step": 9160
+    },
+    {
+      "epoch": 6.725339200586725,
+      "grad_norm": 1.9376052618026733,
+      "learning_rate": 0.0002,
+      "loss": 0.5328,
+      "step": 9170
+    },
+    {
+      "epoch": 6.732673267326732,
+      "grad_norm": 1.0353254079818726,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 9180
+    },
+    {
+      "epoch": 6.74000733406674,
+      "grad_norm": 1.1274057626724243,
+      "learning_rate": 0.0002,
+      "loss": 0.5668,
+      "step": 9190
+    },
+    {
+      "epoch": 6.747341400806747,
+      "grad_norm": 1.3344064950942993,
+      "learning_rate": 0.0002,
+      "loss": 0.4486,
+      "step": 9200
+    },
+    {
+      "epoch": 6.754675467546755,
+      "grad_norm": 1.303621768951416,
+      "learning_rate": 0.0002,
+      "loss": 0.49,
+      "step": 9210
+    },
+    {
+      "epoch": 6.762009534286762,
+      "grad_norm": 1.2327780723571777,
+      "learning_rate": 0.0002,
+      "loss": 0.5059,
+      "step": 9220
+    },
+    {
+      "epoch": 6.769343601026769,
+      "grad_norm": 1.3513109683990479,
+      "learning_rate": 0.0002,
+      "loss": 0.486,
+      "step": 9230
+    },
+    {
+      "epoch": 6.776677667766776,
+      "grad_norm": 1.4762850999832153,
+      "learning_rate": 0.0002,
+      "loss": 0.5254,
+      "step": 9240
+    },
+    {
+      "epoch": 6.784011734506784,
+      "grad_norm": 1.0967189073562622,
+      "learning_rate": 0.0002,
+      "loss": 0.4181,
+      "step": 9250
+    },
+    {
+      "epoch": 6.791345801246791,
+      "grad_norm": 0.933936357498169,
+      "learning_rate": 0.0002,
+      "loss": 0.4862,
+      "step": 9260
+    },
+    {
+      "epoch": 6.798679867986799,
+      "grad_norm": 1.065553903579712,
+      "learning_rate": 0.0002,
+      "loss": 0.4667,
+      "step": 9270
+    },
+    {
+      "epoch": 6.806013934726806,
+      "grad_norm": 1.2044163942337036,
+      "learning_rate": 0.0002,
+      "loss": 0.5164,
+      "step": 9280
+    },
+    {
+      "epoch": 6.813348001466814,
+      "grad_norm": 1.404137134552002,
+      "learning_rate": 0.0002,
+      "loss": 0.4648,
+      "step": 9290
+    },
+    {
+      "epoch": 6.82068206820682,
+      "grad_norm": 1.4005582332611084,
+      "learning_rate": 0.0002,
+      "loss": 0.4442,
+      "step": 9300
+    },
+    {
+      "epoch": 6.828016134946828,
+      "grad_norm": 1.1771104335784912,
+      "learning_rate": 0.0002,
+      "loss": 0.459,
+      "step": 9310
+    },
+    {
+      "epoch": 6.835350201686835,
+      "grad_norm": 1.191933035850525,
+      "learning_rate": 0.0002,
+      "loss": 0.5059,
+      "step": 9320
+    },
+    {
+      "epoch": 6.842684268426843,
+      "grad_norm": 1.3395432233810425,
+      "learning_rate": 0.0002,
+      "loss": 0.4733,
+      "step": 9330
+    },
+    {
+      "epoch": 6.85001833516685,
+      "grad_norm": 1.4145503044128418,
+      "learning_rate": 0.0002,
+      "loss": 0.4882,
+      "step": 9340
+    },
+    {
+      "epoch": 6.857352401906858,
+      "grad_norm": 1.1128839254379272,
+      "learning_rate": 0.0002,
+      "loss": 0.4872,
+      "step": 9350
+    },
+    {
+      "epoch": 6.864686468646864,
+      "grad_norm": 1.0771174430847168,
+      "learning_rate": 0.0002,
+      "loss": 0.4909,
+      "step": 9360
+    },
+    {
+      "epoch": 6.872020535386872,
+      "grad_norm": 1.1089814901351929,
+      "learning_rate": 0.0002,
+      "loss": 0.4739,
+      "step": 9370
+    },
+    {
+      "epoch": 6.879354602126879,
+      "grad_norm": 1.078444004058838,
+      "learning_rate": 0.0002,
+      "loss": 0.4854,
+      "step": 9380
+    },
+    {
+      "epoch": 6.886688668866887,
+      "grad_norm": 1.3676636219024658,
+      "learning_rate": 0.0002,
+      "loss": 0.4904,
+      "step": 9390
+    },
+    {
+      "epoch": 6.894022735606894,
+      "grad_norm": 0.8973749876022339,
+      "learning_rate": 0.0002,
+      "loss": 0.4854,
+      "step": 9400
+    },
+    {
+      "epoch": 6.901356802346902,
+      "grad_norm": 1.141552448272705,
+      "learning_rate": 0.0002,
+      "loss": 0.4274,
+      "step": 9410
+    },
+    {
+      "epoch": 6.908690869086909,
+      "grad_norm": 0.8345359563827515,
+      "learning_rate": 0.0002,
+      "loss": 0.4972,
+      "step": 9420
+    },
+    {
+      "epoch": 6.916024935826916,
+      "grad_norm": 1.1602197885513306,
+      "learning_rate": 0.0002,
+      "loss": 0.5218,
+      "step": 9430
+    },
+    {
+      "epoch": 6.923359002566923,
+      "grad_norm": 1.275466799736023,
+      "learning_rate": 0.0002,
+      "loss": 0.4911,
+      "step": 9440
+    },
+    {
+      "epoch": 6.930693069306931,
+      "grad_norm": 0.9186071157455444,
+      "learning_rate": 0.0002,
+      "loss": 0.4904,
+      "step": 9450
+    },
+    {
+      "epoch": 6.938027136046938,
+      "grad_norm": 0.9069198966026306,
+      "learning_rate": 0.0002,
+      "loss": 0.4604,
+      "step": 9460
+    },
+    {
+      "epoch": 6.945361202786946,
+      "grad_norm": 1.2331899404525757,
+      "learning_rate": 0.0002,
+      "loss": 0.4363,
+      "step": 9470
+    },
+    {
+      "epoch": 6.952695269526953,
+      "grad_norm": 0.8685150742530823,
+      "learning_rate": 0.0002,
+      "loss": 0.4815,
+      "step": 9480
+    },
+    {
+      "epoch": 6.96002933626696,
+      "grad_norm": 1.4067939519882202,
+      "learning_rate": 0.0002,
+      "loss": 0.4424,
+      "step": 9490
+    },
+    {
+      "epoch": 6.967363403006967,
+      "grad_norm": 1.1864029169082642,
+      "learning_rate": 0.0002,
+      "loss": 0.5089,
+      "step": 9500
+    },
+    {
+      "epoch": 6.974697469746975,
+      "grad_norm": 1.3697725534439087,
+      "learning_rate": 0.0002,
+      "loss": 0.4906,
+      "step": 9510
+    },
+    {
+      "epoch": 6.982031536486982,
+      "grad_norm": 1.1632893085479736,
+      "learning_rate": 0.0002,
+      "loss": 0.4797,
+      "step": 9520
+    },
+    {
+      "epoch": 6.9893656032269895,
+      "grad_norm": 1.1447268724441528,
+      "learning_rate": 0.0002,
+      "loss": 0.4526,
+      "step": 9530
+    },
+    {
+      "epoch": 6.996699669966997,
+      "grad_norm": 1.5017213821411133,
+      "learning_rate": 0.0002,
+      "loss": 0.4627,
+      "step": 9540
+    },
+    {
+      "epoch": 6.999633296662999,
+      "eval_loss": 1.4178194999694824,
+      "eval_runtime": 32.7488,
+      "eval_samples_per_second": 13.161,
+      "eval_steps_per_second": 1.649,
+      "step": 9544
+    },
+    {
+      "epoch": 7.0040337367070045,
+      "grad_norm": 1.110981822013855,
+      "learning_rate": 0.0002,
+      "loss": 0.4396,
+      "step": 9550
+    },
+    {
+      "epoch": 7.011367803447011,
+      "grad_norm": 1.2793253660202026,
+      "learning_rate": 0.0002,
+      "loss": 0.3475,
+      "step": 9560
+    },
+    {
+      "epoch": 7.0187018701870185,
+      "grad_norm": 1.1258823871612549,
+      "learning_rate": 0.0002,
+      "loss": 0.4022,
+      "step": 9570
+    },
+    {
+      "epoch": 7.026035936927026,
+      "grad_norm": 1.410486102104187,
+      "learning_rate": 0.0002,
+      "loss": 0.3008,
+      "step": 9580
+    },
+    {
+      "epoch": 7.0333700036670335,
+      "grad_norm": 1.2088500261306763,
+      "learning_rate": 0.0002,
+      "loss": 0.3716,
+      "step": 9590
+    },
+    {
+      "epoch": 7.040704070407041,
+      "grad_norm": 0.8303650617599487,
+      "learning_rate": 0.0002,
+      "loss": 0.3379,
+      "step": 9600
+    },
+    {
+      "epoch": 7.048038137147048,
+      "grad_norm": 0.9813525080680847,
+      "learning_rate": 0.0002,
+      "loss": 0.3376,
+      "step": 9610
+    },
+    {
+      "epoch": 7.055372203887055,
+      "grad_norm": 0.9679017066955566,
+      "learning_rate": 0.0002,
+      "loss": 0.3202,
+      "step": 9620
+    },
+    {
+      "epoch": 7.0627062706270625,
+      "grad_norm": 1.1532220840454102,
+      "learning_rate": 0.0002,
+      "loss": 0.3287,
+      "step": 9630
+    },
+    {
+      "epoch": 7.07004033736707,
+      "grad_norm": 1.312053918838501,
+      "learning_rate": 0.0002,
+      "loss": 0.3639,
+      "step": 9640
+    },
+    {
+      "epoch": 7.077374404107077,
+      "grad_norm": 1.0594364404678345,
+      "learning_rate": 0.0002,
+      "loss": 0.3278,
+      "step": 9650
+    },
+    {
+      "epoch": 7.084708470847085,
+      "grad_norm": 1.545080542564392,
+      "learning_rate": 0.0002,
+      "loss": 0.3259,
+      "step": 9660
+    },
+    {
+      "epoch": 7.092042537587092,
+      "grad_norm": 1.1748381853103638,
+      "learning_rate": 0.0002,
+      "loss": 0.328,
+      "step": 9670
+    },
+    {
+      "epoch": 7.0993766043271,
+      "grad_norm": 1.6107453107833862,
+      "learning_rate": 0.0002,
+      "loss": 0.3313,
+      "step": 9680
+    },
+    {
+      "epoch": 7.106710671067106,
+      "grad_norm": 0.9478244185447693,
+      "learning_rate": 0.0002,
+      "loss": 0.3469,
+      "step": 9690
+    },
+    {
+      "epoch": 7.114044737807114,
+      "grad_norm": 1.508410930633545,
+      "learning_rate": 0.0002,
+      "loss": 0.3289,
+      "step": 9700
+    },
+    {
+      "epoch": 7.121378804547121,
+      "grad_norm": 1.3175169229507446,
+      "learning_rate": 0.0002,
+      "loss": 0.3077,
+      "step": 9710
+    },
+    {
+      "epoch": 7.128712871287129,
+      "grad_norm": 1.2631924152374268,
+      "learning_rate": 0.0002,
+      "loss": 0.3241,
+      "step": 9720
+    },
+    {
+      "epoch": 7.136046938027136,
+      "grad_norm": 1.0640755891799927,
+      "learning_rate": 0.0002,
+      "loss": 0.3806,
+      "step": 9730
+    },
+    {
+      "epoch": 7.143381004767144,
+      "grad_norm": 1.247279405593872,
+      "learning_rate": 0.0002,
+      "loss": 0.3418,
+      "step": 9740
+    },
+    {
+      "epoch": 7.15071507150715,
+      "grad_norm": 1.2538974285125732,
+      "learning_rate": 0.0002,
+      "loss": 0.3385,
+      "step": 9750
+    },
+    {
+      "epoch": 7.158049138247158,
+      "grad_norm": 1.3157252073287964,
+      "learning_rate": 0.0002,
+      "loss": 0.3445,
+      "step": 9760
+    },
+    {
+      "epoch": 7.165383204987165,
+      "grad_norm": 1.5254220962524414,
+      "learning_rate": 0.0002,
+      "loss": 0.3518,
+      "step": 9770
+    },
+    {
+      "epoch": 7.172717271727173,
+      "grad_norm": 1.0063719749450684,
+      "learning_rate": 0.0002,
+      "loss": 0.3575,
+      "step": 9780
+    },
+    {
+      "epoch": 7.18005133846718,
+      "grad_norm": 0.8030351996421814,
+      "learning_rate": 0.0002,
+      "loss": 0.3701,
+      "step": 9790
+    },
+    {
+      "epoch": 7.187385405207188,
+      "grad_norm": 1.2086257934570312,
+      "learning_rate": 0.0002,
+      "loss": 0.3645,
+      "step": 9800
+    },
+    {
+      "epoch": 7.194719471947194,
+      "grad_norm": 1.7020413875579834,
+      "learning_rate": 0.0002,
+      "loss": 0.354,
+      "step": 9810
+    },
+    {
+      "epoch": 7.202053538687202,
+      "grad_norm": 1.2517976760864258,
+      "learning_rate": 0.0002,
+      "loss": 0.3242,
+      "step": 9820
+    },
+    {
+      "epoch": 7.209387605427209,
+      "grad_norm": 1.330505132675171,
+      "learning_rate": 0.0002,
+      "loss": 0.3503,
+      "step": 9830
+    },
+    {
+      "epoch": 7.216721672167217,
+      "grad_norm": 1.1273366212844849,
+      "learning_rate": 0.0002,
+      "loss": 0.3349,
+      "step": 9840
+    },
+    {
+      "epoch": 7.224055738907224,
+      "grad_norm": 1.3738148212432861,
+      "learning_rate": 0.0002,
+      "loss": 0.3562,
+      "step": 9850
+    },
+    {
+      "epoch": 7.231389805647232,
+      "grad_norm": 1.2162928581237793,
+      "learning_rate": 0.0002,
+      "loss": 0.3622,
+      "step": 9860
+    },
+    {
+      "epoch": 7.238723872387239,
+      "grad_norm": 1.743969440460205,
+      "learning_rate": 0.0002,
+      "loss": 0.3913,
+      "step": 9870
+    },
+    {
+      "epoch": 7.246057939127246,
+      "grad_norm": 1.5357484817504883,
+      "learning_rate": 0.0002,
+      "loss": 0.3855,
+      "step": 9880
+    },
+    {
+      "epoch": 7.253392005867253,
+      "grad_norm": 1.342976450920105,
+      "learning_rate": 0.0002,
+      "loss": 0.3556,
+      "step": 9890
+    },
+    {
+      "epoch": 7.260726072607261,
+      "grad_norm": 1.428523302078247,
+      "learning_rate": 0.0002,
+      "loss": 0.3787,
+      "step": 9900
+    },
+    {
+      "epoch": 7.268060139347268,
+      "grad_norm": 1.5631695985794067,
+      "learning_rate": 0.0002,
+      "loss": 0.343,
+      "step": 9910
+    },
+    {
+      "epoch": 7.275394206087276,
+      "grad_norm": 1.192564606666565,
+      "learning_rate": 0.0002,
+      "loss": 0.3292,
+      "step": 9920
+    },
+    {
+      "epoch": 7.282728272827283,
+      "grad_norm": 1.1428006887435913,
+      "learning_rate": 0.0002,
+      "loss": 0.324,
+      "step": 9930
+    },
+    {
+      "epoch": 7.29006233956729,
+      "grad_norm": 1.1959515810012817,
+      "learning_rate": 0.0002,
+      "loss": 0.2868,
+      "step": 9940
+    },
+    {
+      "epoch": 7.297396406307297,
+      "grad_norm": 0.984326958656311,
+      "learning_rate": 0.0002,
+      "loss": 0.3623,
+      "step": 9950
+    },
+    {
+      "epoch": 7.304730473047305,
+      "grad_norm": 1.1258848905563354,
+      "learning_rate": 0.0002,
+      "loss": 0.3442,
+      "step": 9960
+    },
+    {
+      "epoch": 7.312064539787312,
+      "grad_norm": 1.2445521354675293,
+      "learning_rate": 0.0002,
+      "loss": 0.3682,
+      "step": 9970
+    },
+    {
+      "epoch": 7.3193986065273196,
+      "grad_norm": 1.066351294517517,
+      "learning_rate": 0.0002,
+      "loss": 0.4032,
+      "step": 9980
+    },
+    {
+      "epoch": 7.326732673267327,
+      "grad_norm": 1.3111763000488281,
+      "learning_rate": 0.0002,
+      "loss": 0.3689,
+      "step": 9990
+    },
+    {
+      "epoch": 7.334066740007334,
+      "grad_norm": 1.5113508701324463,
+      "learning_rate": 0.0002,
+      "loss": 0.4251,
+      "step": 10000
+    },
+    {
+      "epoch": 7.341400806747341,
+      "grad_norm": 1.2499724626541138,
+      "learning_rate": 0.0002,
+      "loss": 0.3668,
+      "step": 10010
+    },
+    {
+      "epoch": 7.3487348734873486,
+      "grad_norm": 1.1003477573394775,
+      "learning_rate": 0.0002,
+      "loss": 0.392,
+      "step": 10020
+    },
+    {
+      "epoch": 7.356068940227356,
+      "grad_norm": 1.4911425113677979,
+      "learning_rate": 0.0002,
+      "loss": 0.4045,
+      "step": 10030
+    },
+    {
+      "epoch": 7.3634030069673635,
+      "grad_norm": 1.291712999343872,
+      "learning_rate": 0.0002,
+      "loss": 0.3892,
+      "step": 10040
+    },
+    {
+      "epoch": 7.370737073707371,
+      "grad_norm": 1.4001942873001099,
+      "learning_rate": 0.0002,
+      "loss": 0.3817,
+      "step": 10050
+    },
+    {
+      "epoch": 7.378071140447378,
+      "grad_norm": 2.015535593032837,
+      "learning_rate": 0.0002,
+      "loss": 0.404,
+      "step": 10060
+    },
+    {
+      "epoch": 7.385405207187385,
+      "grad_norm": 1.3355735540390015,
+      "learning_rate": 0.0002,
+      "loss": 0.3758,
+      "step": 10070
+    },
+    {
+      "epoch": 7.3927392739273925,
+      "grad_norm": 1.1258678436279297,
+      "learning_rate": 0.0002,
+      "loss": 0.3764,
+      "step": 10080
+    },
+    {
+      "epoch": 7.4000733406674,
+      "grad_norm": 1.3883707523345947,
+      "learning_rate": 0.0002,
+      "loss": 0.3827,
+      "step": 10090
+    },
+    {
+      "epoch": 7.407407407407407,
+      "grad_norm": 1.144474744796753,
+      "learning_rate": 0.0002,
+      "loss": 0.3623,
+      "step": 10100
+    },
+    {
+      "epoch": 7.414741474147415,
+      "grad_norm": 1.636843204498291,
+      "learning_rate": 0.0002,
+      "loss": 0.4024,
+      "step": 10110
+    },
+    {
+      "epoch": 7.422075540887422,
+      "grad_norm": 1.6167247295379639,
+      "learning_rate": 0.0002,
+      "loss": 0.3924,
+      "step": 10120
+    },
+    {
+      "epoch": 7.429409607627429,
+      "grad_norm": 1.3800078630447388,
+      "learning_rate": 0.0002,
+      "loss": 0.4233,
+      "step": 10130
+    },
+    {
+      "epoch": 7.436743674367436,
+      "grad_norm": 1.2631969451904297,
+      "learning_rate": 0.0002,
+      "loss": 0.3859,
+      "step": 10140
+    },
+    {
+      "epoch": 7.444077741107444,
+      "grad_norm": 1.32834792137146,
+      "learning_rate": 0.0002,
+      "loss": 0.3523,
+      "step": 10150
+    },
+    {
+      "epoch": 7.451411807847451,
+      "grad_norm": 1.370316982269287,
+      "learning_rate": 0.0002,
+      "loss": 0.3945,
+      "step": 10160
+    },
+    {
+      "epoch": 7.458745874587459,
+      "grad_norm": 1.6096234321594238,
+      "learning_rate": 0.0002,
+      "loss": 0.3695,
+      "step": 10170
+    },
+    {
+      "epoch": 7.466079941327466,
+      "grad_norm": 1.3638662099838257,
+      "learning_rate": 0.0002,
+      "loss": 0.3378,
+      "step": 10180
+    },
+    {
+      "epoch": 7.473414008067474,
+      "grad_norm": 1.3508107662200928,
+      "learning_rate": 0.0002,
+      "loss": 0.4015,
+      "step": 10190
+    },
+    {
+      "epoch": 7.48074807480748,
+      "grad_norm": 1.5599194765090942,
+      "learning_rate": 0.0002,
+      "loss": 0.4169,
+      "step": 10200
+    },
+    {
+      "epoch": 7.488082141547488,
+      "grad_norm": 1.4922538995742798,
+      "learning_rate": 0.0002,
+      "loss": 0.4071,
+      "step": 10210
+    },
+    {
+      "epoch": 7.495416208287495,
+      "grad_norm": 1.485437273979187,
+      "learning_rate": 0.0002,
+      "loss": 0.419,
+      "step": 10220
+    },
+    {
+      "epoch": 7.502750275027503,
+      "grad_norm": 0.9040785431861877,
+      "learning_rate": 0.0002,
+      "loss": 0.382,
+      "step": 10230
+    },
+    {
+      "epoch": 7.51008434176751,
+      "grad_norm": 1.2453011274337769,
+      "learning_rate": 0.0002,
+      "loss": 0.3437,
+      "step": 10240
+    },
+    {
+      "epoch": 7.517418408507518,
+      "grad_norm": 1.4167460203170776,
+      "learning_rate": 0.0002,
+      "loss": 0.4063,
+      "step": 10250
+    },
+    {
+      "epoch": 7.524752475247524,
+      "grad_norm": 1.396972894668579,
+      "learning_rate": 0.0002,
+      "loss": 0.402,
+      "step": 10260
+    },
+    {
+      "epoch": 7.532086541987532,
+      "grad_norm": 1.384286880493164,
+      "learning_rate": 0.0002,
+      "loss": 0.3658,
+      "step": 10270
+    },
+    {
+      "epoch": 7.539420608727539,
+      "grad_norm": 1.478095531463623,
+      "learning_rate": 0.0002,
+      "loss": 0.3813,
+      "step": 10280
+    },
+    {
+      "epoch": 7.546754675467547,
+      "grad_norm": 1.2642205953598022,
+      "learning_rate": 0.0002,
+      "loss": 0.3813,
+      "step": 10290
+    },
+    {
+      "epoch": 7.554088742207554,
+      "grad_norm": 1.1110541820526123,
+      "learning_rate": 0.0002,
+      "loss": 0.4084,
+      "step": 10300
+    },
+    {
+      "epoch": 7.561422808947562,
+      "grad_norm": 1.1147890090942383,
+      "learning_rate": 0.0002,
+      "loss": 0.418,
+      "step": 10310
+    },
+    {
+      "epoch": 7.568756875687569,
+      "grad_norm": 1.5677998065948486,
+      "learning_rate": 0.0002,
+      "loss": 0.4148,
+      "step": 10320
+    },
+    {
+      "epoch": 7.576090942427576,
+      "grad_norm": 0.9347636699676514,
+      "learning_rate": 0.0002,
+      "loss": 0.3675,
+      "step": 10330
+    },
+    {
+      "epoch": 7.583425009167583,
+      "grad_norm": 1.1126737594604492,
+      "learning_rate": 0.0002,
+      "loss": 0.3534,
+      "step": 10340
+    },
+    {
+      "epoch": 7.590759075907591,
+      "grad_norm": 1.462611198425293,
+      "learning_rate": 0.0002,
+      "loss": 0.418,
+      "step": 10350
+    },
+    {
+      "epoch": 7.598093142647598,
+      "grad_norm": 0.9907522201538086,
+      "learning_rate": 0.0002,
+      "loss": 0.3755,
+      "step": 10360
+    },
+    {
+      "epoch": 7.605427209387606,
+      "grad_norm": 1.306152582168579,
+      "learning_rate": 0.0002,
+      "loss": 0.4133,
+      "step": 10370
+    },
+    {
+      "epoch": 7.612761276127613,
+      "grad_norm": 1.11135995388031,
+      "learning_rate": 0.0002,
+      "loss": 0.3644,
+      "step": 10380
+    },
+    {
+      "epoch": 7.62009534286762,
+      "grad_norm": 1.0825806856155396,
+      "learning_rate": 0.0002,
+      "loss": 0.3659,
+      "step": 10390
+    },
+    {
+      "epoch": 7.627429409607627,
+      "grad_norm": 1.5346975326538086,
+      "learning_rate": 0.0002,
+      "loss": 0.3952,
+      "step": 10400
+    },
+    {
+      "epoch": 7.634763476347635,
+      "grad_norm": 1.5885388851165771,
+      "learning_rate": 0.0002,
+      "loss": 0.3807,
+      "step": 10410
+    },
+    {
+      "epoch": 7.642097543087642,
+      "grad_norm": 1.130261778831482,
+      "learning_rate": 0.0002,
+      "loss": 0.3751,
+      "step": 10420
+    },
+    {
+      "epoch": 7.64943160982765,
+      "grad_norm": 1.2318342924118042,
+      "learning_rate": 0.0002,
+      "loss": 0.3598,
+      "step": 10430
+    },
+    {
+      "epoch": 7.656765676567657,
+      "grad_norm": 1.07103431224823,
+      "learning_rate": 0.0002,
+      "loss": 0.3823,
+      "step": 10440
+    },
+    {
+      "epoch": 7.6640997433076645,
+      "grad_norm": 1.3836923837661743,
+      "learning_rate": 0.0002,
+      "loss": 0.4707,
+      "step": 10450
+    },
+    {
+      "epoch": 7.671433810047671,
+      "grad_norm": 1.2110271453857422,
+      "learning_rate": 0.0002,
+      "loss": 0.3829,
+      "step": 10460
+    },
+    {
+      "epoch": 7.678767876787679,
+      "grad_norm": 1.2304844856262207,
+      "learning_rate": 0.0002,
+      "loss": 0.3747,
+      "step": 10470
+    },
+    {
+      "epoch": 7.686101943527686,
+      "grad_norm": 1.3444706201553345,
+      "learning_rate": 0.0002,
+      "loss": 0.3761,
+      "step": 10480
+    },
+    {
+      "epoch": 7.6934360102676935,
+      "grad_norm": 1.151705026626587,
+      "learning_rate": 0.0002,
+      "loss": 0.3772,
+      "step": 10490
+    },
+    {
+      "epoch": 7.700770077007701,
+      "grad_norm": 1.4373983144760132,
+      "learning_rate": 0.0002,
+      "loss": 0.3351,
+      "step": 10500
+    },
+    {
+      "epoch": 7.7081041437477085,
+      "grad_norm": 1.5898514986038208,
+      "learning_rate": 0.0002,
+      "loss": 0.389,
+      "step": 10510
+    },
+    {
+      "epoch": 7.715438210487715,
+      "grad_norm": 1.3767389059066772,
+      "learning_rate": 0.0002,
+      "loss": 0.3679,
+      "step": 10520
+    },
+    {
+      "epoch": 7.7227722772277225,
+      "grad_norm": 1.3730027675628662,
+      "learning_rate": 0.0002,
+      "loss": 0.4023,
+      "step": 10530
+    },
+    {
+      "epoch": 7.73010634396773,
+      "grad_norm": 1.3917304277420044,
+      "learning_rate": 0.0002,
+      "loss": 0.4291,
+      "step": 10540
+    },
+    {
+      "epoch": 7.7374404107077375,
+      "grad_norm": 1.230663776397705,
+      "learning_rate": 0.0002,
+      "loss": 0.4378,
+      "step": 10550
+    },
+    {
+      "epoch": 7.744774477447745,
+      "grad_norm": 0.9922441244125366,
+      "learning_rate": 0.0002,
+      "loss": 0.4289,
+      "step": 10560
+    },
+    {
+      "epoch": 7.752108544187752,
+      "grad_norm": 1.4576551914215088,
+      "learning_rate": 0.0002,
+      "loss": 0.4118,
+      "step": 10570
+    },
+    {
+      "epoch": 7.75944261092776,
+      "grad_norm": 1.061685562133789,
+      "learning_rate": 0.0002,
+      "loss": 0.3793,
+      "step": 10580
+    },
+    {
+      "epoch": 7.7667766776677665,
+      "grad_norm": 1.1523276567459106,
+      "learning_rate": 0.0002,
+      "loss": 0.3748,
+      "step": 10590
+    },
+    {
+      "epoch": 7.774110744407774,
+      "grad_norm": 1.3917267322540283,
+      "learning_rate": 0.0002,
+      "loss": 0.3982,
+      "step": 10600
+    },
+    {
+      "epoch": 7.781444811147781,
+      "grad_norm": 1.1314283609390259,
+      "learning_rate": 0.0002,
+      "loss": 0.4305,
+      "step": 10610
+    },
+    {
+      "epoch": 7.788778877887789,
+      "grad_norm": 1.624324083328247,
+      "learning_rate": 0.0002,
+      "loss": 0.4244,
+      "step": 10620
+    },
+    {
+      "epoch": 7.796112944627796,
+      "grad_norm": 1.5369168519973755,
+      "learning_rate": 0.0002,
+      "loss": 0.4129,
+      "step": 10630
+    },
+    {
+      "epoch": 7.803447011367804,
+      "grad_norm": 1.082222819328308,
+      "learning_rate": 0.0002,
+      "loss": 0.3661,
+      "step": 10640
+    },
+    {
+      "epoch": 7.81078107810781,
+      "grad_norm": 1.281540870666504,
+      "learning_rate": 0.0002,
+      "loss": 0.4341,
+      "step": 10650
+    },
+    {
+      "epoch": 7.818115144847818,
+      "grad_norm": 1.1889171600341797,
+      "learning_rate": 0.0002,
+      "loss": 0.3939,
+      "step": 10660
+    },
+    {
+      "epoch": 7.825449211587825,
+      "grad_norm": 1.5310896635055542,
+      "learning_rate": 0.0002,
+      "loss": 0.3923,
+      "step": 10670
+    },
+    {
+      "epoch": 7.832783278327833,
+      "grad_norm": 1.6724708080291748,
+      "learning_rate": 0.0002,
+      "loss": 0.4148,
+      "step": 10680
+    },
+    {
+      "epoch": 7.84011734506784,
+      "grad_norm": 1.3742409944534302,
+      "learning_rate": 0.0002,
+      "loss": 0.363,
+      "step": 10690
+    },
+    {
+      "epoch": 7.847451411807848,
+      "grad_norm": 1.2421947717666626,
+      "learning_rate": 0.0002,
+      "loss": 0.3599,
+      "step": 10700
+    },
+    {
+      "epoch": 7.854785478547855,
+      "grad_norm": 1.0233848094940186,
+      "learning_rate": 0.0002,
+      "loss": 0.3829,
+      "step": 10710
+    },
+    {
+      "epoch": 7.862119545287862,
+      "grad_norm": 1.640842080116272,
+      "learning_rate": 0.0002,
+      "loss": 0.4265,
+      "step": 10720
+    },
+    {
+      "epoch": 7.869453612027869,
+      "grad_norm": 1.3571926355361938,
+      "learning_rate": 0.0002,
+      "loss": 0.4257,
+      "step": 10730
+    },
+    {
+      "epoch": 7.876787678767877,
+      "grad_norm": 1.459564208984375,
+      "learning_rate": 0.0002,
+      "loss": 0.4454,
+      "step": 10740
+    },
+    {
+      "epoch": 7.884121745507884,
+      "grad_norm": 0.9202831387519836,
+      "learning_rate": 0.0002,
+      "loss": 0.3903,
+      "step": 10750
+    },
+    {
+      "epoch": 7.891455812247892,
+      "grad_norm": 1.3509176969528198,
+      "learning_rate": 0.0002,
+      "loss": 0.4149,
+      "step": 10760
+    },
+    {
+      "epoch": 7.898789878987898,
+      "grad_norm": 1.5858603715896606,
+      "learning_rate": 0.0002,
+      "loss": 0.4001,
+      "step": 10770
+    },
+    {
+      "epoch": 7.906123945727906,
+      "grad_norm": 1.2391952276229858,
+      "learning_rate": 0.0002,
+      "loss": 0.3753,
+      "step": 10780
+    },
+    {
+      "epoch": 7.913458012467913,
+      "grad_norm": 1.3442552089691162,
+      "learning_rate": 0.0002,
+      "loss": 0.4085,
+      "step": 10790
+    },
+    {
+      "epoch": 7.920792079207921,
+      "grad_norm": 1.7327884435653687,
+      "learning_rate": 0.0002,
+      "loss": 0.4377,
+      "step": 10800
+    },
+    {
+      "epoch": 7.928126145947928,
+      "grad_norm": 1.4246922731399536,
+      "learning_rate": 0.0002,
+      "loss": 0.376,
+      "step": 10810
+    },
+    {
+      "epoch": 7.935460212687936,
+      "grad_norm": 1.4421411752700806,
+      "learning_rate": 0.0002,
+      "loss": 0.4158,
+      "step": 10820
+    },
+    {
+      "epoch": 7.942794279427943,
+      "grad_norm": 1.3445014953613281,
+      "learning_rate": 0.0002,
+      "loss": 0.4084,
+      "step": 10830
+    },
+    {
+      "epoch": 7.950128346167951,
+      "grad_norm": 1.2219295501708984,
+      "learning_rate": 0.0002,
+      "loss": 0.3986,
+      "step": 10840
+    },
+    {
+      "epoch": 7.957462412907957,
+      "grad_norm": 1.241843342781067,
+      "learning_rate": 0.0002,
+      "loss": 0.428,
+      "step": 10850
+    },
+    {
+      "epoch": 7.964796479647965,
+      "grad_norm": 0.9814007878303528,
+      "learning_rate": 0.0002,
+      "loss": 0.3776,
+      "step": 10860
+    },
+    {
+      "epoch": 7.972130546387972,
+      "grad_norm": 1.4015462398529053,
+      "learning_rate": 0.0002,
+      "loss": 0.4866,
+      "step": 10870
+    },
+    {
+      "epoch": 7.97946461312798,
+      "grad_norm": 1.4638406038284302,
+      "learning_rate": 0.0002,
+      "loss": 0.437,
+      "step": 10880
+    },
+    {
+      "epoch": 7.986798679867987,
+      "grad_norm": 1.585194706916809,
+      "learning_rate": 0.0002,
+      "loss": 0.4191,
+      "step": 10890
+    },
+    {
+      "epoch": 7.994132746607994,
+      "grad_norm": 1.197031855583191,
+      "learning_rate": 0.0002,
+      "loss": 0.3969,
+      "step": 10900
+    },
+    {
+      "epoch": 7.997066373303997,
+      "eval_loss": 1.499578595161438,
+      "eval_runtime": 32.7458,
+      "eval_samples_per_second": 13.162,
+      "eval_steps_per_second": 1.649,
+      "step": 10904
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10904,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.0461270495513805e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..2868cff7027115396e695775cacd838522aca295
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-10904/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b12b6f6817632087b5a5e37d744e25312b96e839de5005320b96bc0c2473c41f
+size 5624
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/README.md b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# 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 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "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": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3c36ecbbb47676521c72eea7028470f3e1d4b672
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb3e68c859e7f949d61c2b851f8413912f0844f23f0d7451d5d00de039c04417
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..830ccdecc75c527c4aae4cce853e39ceda3b15f1
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8db681f7fd9dfa85d55512999836db447ef2a457bd7b27f0648bbbc56412b607
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..70edec11e7459d3c7aa5439be9a2b2056c65a9ad
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e8c1f88129afe081291492ecda26f8bde107365cff0da0a4ab1c6f1ccb89efa
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..381516e5ded3f51b9d3cf89c73c0cdf6a851cc0d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec792ccb2845954d3473c14cd59e6ae800f3539bd2b48bbf23c7aa849803b8fa
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..a565e5777c2ddd8662b5f94ca0ccbbcccf6f4366
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/trainer_state.json
@@ -0,0 +1,993 @@
+{
+  "best_metric": 1.1822267770767212,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363",
+  "epoch": 0.9996332966629996,
+  "eval_steps": 10,
+  "global_step": 1363,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007334066740007334,
+      "grad_norm": 0.47521963715553284,
+      "learning_rate": 0.0002,
+      "loss": 1.9722,
+      "step": 10
+    },
+    {
+      "epoch": 0.014668133480014669,
+      "grad_norm": 0.5395162105560303,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 20
+    },
+    {
+      "epoch": 0.022002200220022004,
+      "grad_norm": 0.4305780231952667,
+      "learning_rate": 0.0002,
+      "loss": 1.4202,
+      "step": 30
+    },
+    {
+      "epoch": 0.029336266960029337,
+      "grad_norm": 0.6938246488571167,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 40
+    },
+    {
+      "epoch": 0.03667033370003667,
+      "grad_norm": 1.5133819580078125,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 50
+    },
+    {
+      "epoch": 0.04400440044004401,
+      "grad_norm": 0.9173883199691772,
+      "learning_rate": 0.0002,
+      "loss": 1.3132,
+      "step": 60
+    },
+    {
+      "epoch": 0.05133846718005134,
+      "grad_norm": 0.4619861841201782,
+      "learning_rate": 0.0002,
+      "loss": 1.2844,
+      "step": 70
+    },
+    {
+      "epoch": 0.058672533920058674,
+      "grad_norm": 0.46118637919425964,
+      "learning_rate": 0.0002,
+      "loss": 1.2108,
+      "step": 80
+    },
+    {
+      "epoch": 0.066006600660066,
+      "grad_norm": 0.4468648135662079,
+      "learning_rate": 0.0002,
+      "loss": 1.3441,
+      "step": 90
+    },
+    {
+      "epoch": 0.07334066740007333,
+      "grad_norm": 0.46123769879341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1863,
+      "step": 100
+    },
+    {
+      "epoch": 0.08067473414008068,
+      "grad_norm": 0.4859139025211334,
+      "learning_rate": 0.0002,
+      "loss": 1.2772,
+      "step": 110
+    },
+    {
+      "epoch": 0.08800880088008801,
+      "grad_norm": 0.4384922385215759,
+      "learning_rate": 0.0002,
+      "loss": 1.2087,
+      "step": 120
+    },
+    {
+      "epoch": 0.09534286762009535,
+      "grad_norm": 0.39519360661506653,
+      "learning_rate": 0.0002,
+      "loss": 1.2927,
+      "step": 130
+    },
+    {
+      "epoch": 0.10267693436010268,
+      "grad_norm": 0.4049859344959259,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 140
+    },
+    {
+      "epoch": 0.11001100110011001,
+      "grad_norm": 0.4605638086795807,
+      "learning_rate": 0.0002,
+      "loss": 1.293,
+      "step": 150
+    },
+    {
+      "epoch": 0.11734506784011735,
+      "grad_norm": 0.4201928377151489,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 160
+    },
+    {
+      "epoch": 0.12467913458012468,
+      "grad_norm": 0.5367777347564697,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 170
+    },
+    {
+      "epoch": 0.132013201320132,
+      "grad_norm": 0.41752299666404724,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 180
+    },
+    {
+      "epoch": 0.13934726806013933,
+      "grad_norm": 0.31597763299942017,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 190
+    },
+    {
+      "epoch": 0.14668133480014667,
+      "grad_norm": 0.7468788623809814,
+      "learning_rate": 0.0002,
+      "loss": 1.2441,
+      "step": 200
+    },
+    {
+      "epoch": 0.15401540154015403,
+      "grad_norm": 0.3403034508228302,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 210
+    },
+    {
+      "epoch": 0.16134946828016136,
+      "grad_norm": 0.34240293502807617,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 220
+    },
+    {
+      "epoch": 0.1686835350201687,
+      "grad_norm": 0.356158971786499,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 230
+    },
+    {
+      "epoch": 0.17601760176017603,
+      "grad_norm": 0.3448857367038727,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 240
+    },
+    {
+      "epoch": 0.18335166850018336,
+      "grad_norm": 0.3475699722766876,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 250
+    },
+    {
+      "epoch": 0.1906857352401907,
+      "grad_norm": 0.2770358622074127,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 260
+    },
+    {
+      "epoch": 0.19801980198019803,
+      "grad_norm": 0.4310270845890045,
+      "learning_rate": 0.0002,
+      "loss": 1.2238,
+      "step": 270
+    },
+    {
+      "epoch": 0.20535386872020536,
+      "grad_norm": 0.335041880607605,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 280
+    },
+    {
+      "epoch": 0.2126879354602127,
+      "grad_norm": 0.3420602083206177,
+      "learning_rate": 0.0002,
+      "loss": 1.0959,
+      "step": 290
+    },
+    {
+      "epoch": 0.22002200220022003,
+      "grad_norm": 0.325001060962677,
+      "learning_rate": 0.0002,
+      "loss": 1.1232,
+      "step": 300
+    },
+    {
+      "epoch": 0.22735606894022736,
+      "grad_norm": 0.3027827739715576,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 310
+    },
+    {
+      "epoch": 0.2346901356802347,
+      "grad_norm": 0.435550719499588,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 320
+    },
+    {
+      "epoch": 0.24202420242024203,
+      "grad_norm": 0.3884522616863251,
+      "learning_rate": 0.0002,
+      "loss": 1.2045,
+      "step": 330
+    },
+    {
+      "epoch": 0.24935826916024936,
+      "grad_norm": 0.7736002206802368,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 340
+    },
+    {
+      "epoch": 0.2566923359002567,
+      "grad_norm": 0.35052821040153503,
+      "learning_rate": 0.0002,
+      "loss": 1.3606,
+      "step": 350
+    },
+    {
+      "epoch": 0.264026402640264,
+      "grad_norm": 0.3311890959739685,
+      "learning_rate": 0.0002,
+      "loss": 1.2129,
+      "step": 360
+    },
+    {
+      "epoch": 0.27136046938027136,
+      "grad_norm": 0.7473500370979309,
+      "learning_rate": 0.0002,
+      "loss": 1.2219,
+      "step": 370
+    },
+    {
+      "epoch": 0.27869453612027867,
+      "grad_norm": 0.3681875765323639,
+      "learning_rate": 0.0002,
+      "loss": 1.2712,
+      "step": 380
+    },
+    {
+      "epoch": 0.28602860286028603,
+      "grad_norm": 0.3764737844467163,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 390
+    },
+    {
+      "epoch": 0.29336266960029334,
+      "grad_norm": 0.4243989586830139,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 400
+    },
+    {
+      "epoch": 0.3006967363403007,
+      "grad_norm": 0.2658531963825226,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 410
+    },
+    {
+      "epoch": 0.30803080308030806,
+      "grad_norm": 0.3436793386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.1622,
+      "step": 420
+    },
+    {
+      "epoch": 0.31536486982031536,
+      "grad_norm": 0.5101129412651062,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 430
+    },
+    {
+      "epoch": 0.3226989365603227,
+      "grad_norm": 0.3319750726222992,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 440
+    },
+    {
+      "epoch": 0.33003300330033003,
+      "grad_norm": 0.385148286819458,
+      "learning_rate": 0.0002,
+      "loss": 1.1804,
+      "step": 450
+    },
+    {
+      "epoch": 0.3373670700403374,
+      "grad_norm": 0.3477935791015625,
+      "learning_rate": 0.0002,
+      "loss": 1.1808,
+      "step": 460
+    },
+    {
+      "epoch": 0.3447011367803447,
+      "grad_norm": 0.29748716950416565,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 470
+    },
+    {
+      "epoch": 0.35203520352035206,
+      "grad_norm": 0.34083324670791626,
+      "learning_rate": 0.0002,
+      "loss": 1.19,
+      "step": 480
+    },
+    {
+      "epoch": 0.35936927026035936,
+      "grad_norm": 0.36904552578926086,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 490
+    },
+    {
+      "epoch": 0.3667033370003667,
+      "grad_norm": 0.315483033657074,
+      "learning_rate": 0.0002,
+      "loss": 1.2223,
+      "step": 500
+    },
+    {
+      "epoch": 0.37403740374037403,
+      "grad_norm": 0.44897955656051636,
+      "learning_rate": 0.0002,
+      "loss": 1.1461,
+      "step": 510
+    },
+    {
+      "epoch": 0.3813714704803814,
+      "grad_norm": 0.3160701394081116,
+      "learning_rate": 0.0002,
+      "loss": 1.3035,
+      "step": 520
+    },
+    {
+      "epoch": 0.3887055372203887,
+      "grad_norm": 0.29584741592407227,
+      "learning_rate": 0.0002,
+      "loss": 1.3197,
+      "step": 530
+    },
+    {
+      "epoch": 0.39603960396039606,
+      "grad_norm": 0.5430002808570862,
+      "learning_rate": 0.0002,
+      "loss": 1.2983,
+      "step": 540
+    },
+    {
+      "epoch": 0.40337367070040336,
+      "grad_norm": 0.2908070683479309,
+      "learning_rate": 0.0002,
+      "loss": 1.2459,
+      "step": 550
+    },
+    {
+      "epoch": 0.4107077374404107,
+      "grad_norm": 0.35066530108451843,
+      "learning_rate": 0.0002,
+      "loss": 1.2384,
+      "step": 560
+    },
+    {
+      "epoch": 0.41804180418041803,
+      "grad_norm": 0.37588003277778625,
+      "learning_rate": 0.0002,
+      "loss": 1.1784,
+      "step": 570
+    },
+    {
+      "epoch": 0.4253758709204254,
+      "grad_norm": 0.3112126886844635,
+      "learning_rate": 0.0002,
+      "loss": 1.2334,
+      "step": 580
+    },
+    {
+      "epoch": 0.4327099376604327,
+      "grad_norm": 0.35577139258384705,
+      "learning_rate": 0.0002,
+      "loss": 1.1439,
+      "step": 590
+    },
+    {
+      "epoch": 0.44004400440044006,
+      "grad_norm": 0.31706422567367554,
+      "learning_rate": 0.0002,
+      "loss": 1.184,
+      "step": 600
+    },
+    {
+      "epoch": 0.44737807114044736,
+      "grad_norm": 0.3249092102050781,
+      "learning_rate": 0.0002,
+      "loss": 1.2081,
+      "step": 610
+    },
+    {
+      "epoch": 0.4547121378804547,
+      "grad_norm": 0.3842705488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.0824,
+      "step": 620
+    },
+    {
+      "epoch": 0.46204620462046203,
+      "grad_norm": 0.390991747379303,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 630
+    },
+    {
+      "epoch": 0.4693802713604694,
+      "grad_norm": 0.27532413601875305,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 640
+    },
+    {
+      "epoch": 0.4767143381004767,
+      "grad_norm": 0.31412816047668457,
+      "learning_rate": 0.0002,
+      "loss": 1.1058,
+      "step": 650
+    },
+    {
+      "epoch": 0.48404840484048406,
+      "grad_norm": 0.32117101550102234,
+      "learning_rate": 0.0002,
+      "loss": 1.1312,
+      "step": 660
+    },
+    {
+      "epoch": 0.49138247158049136,
+      "grad_norm": 0.3810010254383087,
+      "learning_rate": 0.0002,
+      "loss": 1.2423,
+      "step": 670
+    },
+    {
+      "epoch": 0.4987165383204987,
+      "grad_norm": 0.36289164423942566,
+      "learning_rate": 0.0002,
+      "loss": 1.1978,
+      "step": 680
+    },
+    {
+      "epoch": 0.506050605060506,
+      "grad_norm": 0.34458720684051514,
+      "learning_rate": 0.0002,
+      "loss": 1.2034,
+      "step": 690
+    },
+    {
+      "epoch": 0.5133846718005134,
+      "grad_norm": 0.32844600081443787,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 700
+    },
+    {
+      "epoch": 0.5207187385405208,
+      "grad_norm": 0.3144175708293915,
+      "learning_rate": 0.0002,
+      "loss": 1.0807,
+      "step": 710
+    },
+    {
+      "epoch": 0.528052805280528,
+      "grad_norm": 0.3898887634277344,
+      "learning_rate": 0.0002,
+      "loss": 1.1952,
+      "step": 720
+    },
+    {
+      "epoch": 0.5353868720205354,
+      "grad_norm": 1.3220758438110352,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 730
+    },
+    {
+      "epoch": 0.5427209387605427,
+      "grad_norm": 0.3635874390602112,
+      "learning_rate": 0.0002,
+      "loss": 1.227,
+      "step": 740
+    },
+    {
+      "epoch": 0.5500550055005501,
+      "grad_norm": 0.3138217628002167,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 750
+    },
+    {
+      "epoch": 0.5573890722405573,
+      "grad_norm": 0.4063207805156708,
+      "learning_rate": 0.0002,
+      "loss": 1.1516,
+      "step": 760
+    },
+    {
+      "epoch": 0.5647231389805647,
+      "grad_norm": 0.3926219940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 770
+    },
+    {
+      "epoch": 0.5720572057205721,
+      "grad_norm": 0.31954652070999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1726,
+      "step": 780
+    },
+    {
+      "epoch": 0.5793912724605794,
+      "grad_norm": 0.4248711168766022,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 790
+    },
+    {
+      "epoch": 0.5867253392005867,
+      "grad_norm": 0.643004834651947,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 800
+    },
+    {
+      "epoch": 0.594059405940594,
+      "grad_norm": 0.3479592800140381,
+      "learning_rate": 0.0002,
+      "loss": 1.1793,
+      "step": 810
+    },
+    {
+      "epoch": 0.6013934726806014,
+      "grad_norm": 0.4684754014015198,
+      "learning_rate": 0.0002,
+      "loss": 1.2426,
+      "step": 820
+    },
+    {
+      "epoch": 0.6087275394206088,
+      "grad_norm": 0.3739790916442871,
+      "learning_rate": 0.0002,
+      "loss": 1.2002,
+      "step": 830
+    },
+    {
+      "epoch": 0.6160616061606161,
+      "grad_norm": 0.40884748101234436,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 840
+    },
+    {
+      "epoch": 0.6233956729006234,
+      "grad_norm": 0.9722164273262024,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 850
+    },
+    {
+      "epoch": 0.6307297396406307,
+      "grad_norm": 0.42992347478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.3069,
+      "step": 860
+    },
+    {
+      "epoch": 0.6380638063806381,
+      "grad_norm": 0.36654195189476013,
+      "learning_rate": 0.0002,
+      "loss": 1.1339,
+      "step": 870
+    },
+    {
+      "epoch": 0.6453978731206454,
+      "grad_norm": 0.4113832116127014,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 880
+    },
+    {
+      "epoch": 0.6527319398606527,
+      "grad_norm": 0.2948838770389557,
+      "learning_rate": 0.0002,
+      "loss": 1.2163,
+      "step": 890
+    },
+    {
+      "epoch": 0.6600660066006601,
+      "grad_norm": 0.38330280780792236,
+      "learning_rate": 0.0002,
+      "loss": 1.1081,
+      "step": 900
+    },
+    {
+      "epoch": 0.6674000733406674,
+      "grad_norm": 0.4428867697715759,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 910
+    },
+    {
+      "epoch": 0.6747341400806748,
+      "grad_norm": 0.23659265041351318,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 920
+    },
+    {
+      "epoch": 0.682068206820682,
+      "grad_norm": 0.323685884475708,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 930
+    },
+    {
+      "epoch": 0.6894022735606894,
+      "grad_norm": 0.39157727360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 940
+    },
+    {
+      "epoch": 0.6967363403006968,
+      "grad_norm": 0.27189481258392334,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 950
+    },
+    {
+      "epoch": 0.7040704070407041,
+      "grad_norm": 0.529883861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.1033,
+      "step": 960
+    },
+    {
+      "epoch": 0.7114044737807114,
+      "grad_norm": 0.34758689999580383,
+      "learning_rate": 0.0002,
+      "loss": 1.139,
+      "step": 970
+    },
+    {
+      "epoch": 0.7187385405207187,
+      "grad_norm": 0.831749439239502,
+      "learning_rate": 0.0002,
+      "loss": 1.2197,
+      "step": 980
+    },
+    {
+      "epoch": 0.7260726072607261,
+      "grad_norm": 0.4438304007053375,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 990
+    },
+    {
+      "epoch": 0.7334066740007334,
+      "grad_norm": 0.33840006589889526,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7407407407407407,
+      "grad_norm": 0.3454797863960266,
+      "learning_rate": 0.0002,
+      "loss": 1.254,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7480748074807481,
+      "grad_norm": 0.38999441266059875,
+      "learning_rate": 0.0002,
+      "loss": 1.106,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7554088742207554,
+      "grad_norm": 0.2829911708831787,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7627429409607628,
+      "grad_norm": 0.36918163299560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 1040
+    },
+    {
+      "epoch": 0.77007700770077,
+      "grad_norm": 0.3415680229663849,
+      "learning_rate": 0.0002,
+      "loss": 1.3028,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7774110744407774,
+      "grad_norm": 0.2974182963371277,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7847451411807848,
+      "grad_norm": 0.3880919814109802,
+      "learning_rate": 0.0002,
+      "loss": 1.194,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7920792079207921,
+      "grad_norm": 0.33503302931785583,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7994132746607994,
+      "grad_norm": 0.3728407025337219,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8067473414008067,
+      "grad_norm": 0.3509373664855957,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8140814081408141,
+      "grad_norm": 0.42228564620018005,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8214154748808215,
+      "grad_norm": 0.313467800617218,
+      "learning_rate": 0.0002,
+      "loss": 1.1788,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8287495416208287,
+      "grad_norm": 0.3378850817680359,
+      "learning_rate": 0.0002,
+      "loss": 1.1971,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8360836083608361,
+      "grad_norm": 0.43200382590293884,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8434176751008434,
+      "grad_norm": 0.3309599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.3203,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8507517418408508,
+      "grad_norm": 0.3526846170425415,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1160
+    },
+    {
+      "epoch": 0.858085808580858,
+      "grad_norm": 1.2722247838974,
+      "learning_rate": 0.0002,
+      "loss": 1.0851,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8654198753208654,
+      "grad_norm": 0.34142059087753296,
+      "learning_rate": 0.0002,
+      "loss": 1.0785,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8727539420608728,
+      "grad_norm": 0.3805823028087616,
+      "learning_rate": 0.0002,
+      "loss": 1.2187,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8800880088008801,
+      "grad_norm": 0.3931232690811157,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8874220755408874,
+      "grad_norm": 0.2937372624874115,
+      "learning_rate": 0.0002,
+      "loss": 1.0948,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8947561422808947,
+      "grad_norm": 0.3757196366786957,
+      "learning_rate": 0.0002,
+      "loss": 1.1228,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9020902090209021,
+      "grad_norm": 0.3502705991268158,
+      "learning_rate": 0.0002,
+      "loss": 1.1222,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9094242757609095,
+      "grad_norm": 0.32758915424346924,
+      "learning_rate": 0.0002,
+      "loss": 1.2242,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9167583425009168,
+      "grad_norm": 0.37199416756629944,
+      "learning_rate": 0.0002,
+      "loss": 1.215,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9240924092409241,
+      "grad_norm": 0.3551490604877472,
+      "learning_rate": 0.0002,
+      "loss": 1.1225,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9314264759809314,
+      "grad_norm": 0.2859550714492798,
+      "learning_rate": 0.0002,
+      "loss": 1.1966,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9387605427209388,
+      "grad_norm": 0.427990585565567,
+      "learning_rate": 0.0002,
+      "loss": 1.2186,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9460946094609461,
+      "grad_norm": 0.33717992901802063,
+      "learning_rate": 0.0002,
+      "loss": 1.2848,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9534286762009534,
+      "grad_norm": 0.30225634574890137,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9607627429409608,
+      "grad_norm": 0.385821133852005,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9680968096809681,
+      "grad_norm": 0.35278066992759705,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9754308764209755,
+      "grad_norm": 0.49987098574638367,
+      "learning_rate": 0.0002,
+      "loss": 1.1071,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9827649431609827,
+      "grad_norm": 0.3842747211456299,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9900990099009901,
+      "grad_norm": 0.6274653673171997,
+      "learning_rate": 0.0002,
+      "loss": 1.0862,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9974330766409975,
+      "grad_norm": 0.5239808559417725,
+      "learning_rate": 0.0002,
+      "loss": 1.124,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9996332966629996,
+      "eval_loss": 1.1822267770767212,
+      "eval_runtime": 32.7389,
+      "eval_samples_per_second": 13.165,
+      "eval_steps_per_second": 1.649,
+      "step": 1363
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10904,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "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": 6.309972699984691e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..2868cff7027115396e695775cacd838522aca295
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b12b6f6817632087b5a5e37d744e25312b96e839de5005320b96bc0c2473c41f
+size 5624
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/README.md b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# 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 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "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": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ec1ef8d47b111af6b72ef86a5e420ba55ab736f6
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18d3065462d28b00e94897958b444dda35b3dd73c28df887bc7c6cfe0ea65800
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fa3b50e0d499bce41ed2192265ed261d64261b2f
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3192e65fe12343970eafb01e10d7a8147b440f12a84b3ec976ed4c80429062b5
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..76249ea6e890a3dc0c443940bcaaa512d966c49e
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc6c2826db31d92be03894f62dd34cd3704f8d386851d4da63f91a3f2227279f
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3f5c8f6520190e4e417e0a861746661e0e651c80
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b53869add167718298cc75c10e199544e4e266f7a2d65e010562c09bd1ea51d7
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..9b431e9b5c4a3fff5d9580eaa26b9cdf08e2c648
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/trainer_state.json
@@ -0,0 +1,1953 @@
+{
+  "best_metric": 1.1534006595611572,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 2727,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007334066740007334,
+      "grad_norm": 0.47521963715553284,
+      "learning_rate": 0.0002,
+      "loss": 1.9722,
+      "step": 10
+    },
+    {
+      "epoch": 0.014668133480014669,
+      "grad_norm": 0.5395162105560303,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 20
+    },
+    {
+      "epoch": 0.022002200220022004,
+      "grad_norm": 0.4305780231952667,
+      "learning_rate": 0.0002,
+      "loss": 1.4202,
+      "step": 30
+    },
+    {
+      "epoch": 0.029336266960029337,
+      "grad_norm": 0.6938246488571167,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 40
+    },
+    {
+      "epoch": 0.03667033370003667,
+      "grad_norm": 1.5133819580078125,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 50
+    },
+    {
+      "epoch": 0.04400440044004401,
+      "grad_norm": 0.9173883199691772,
+      "learning_rate": 0.0002,
+      "loss": 1.3132,
+      "step": 60
+    },
+    {
+      "epoch": 0.05133846718005134,
+      "grad_norm": 0.4619861841201782,
+      "learning_rate": 0.0002,
+      "loss": 1.2844,
+      "step": 70
+    },
+    {
+      "epoch": 0.058672533920058674,
+      "grad_norm": 0.46118637919425964,
+      "learning_rate": 0.0002,
+      "loss": 1.2108,
+      "step": 80
+    },
+    {
+      "epoch": 0.066006600660066,
+      "grad_norm": 0.4468648135662079,
+      "learning_rate": 0.0002,
+      "loss": 1.3441,
+      "step": 90
+    },
+    {
+      "epoch": 0.07334066740007333,
+      "grad_norm": 0.46123769879341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1863,
+      "step": 100
+    },
+    {
+      "epoch": 0.08067473414008068,
+      "grad_norm": 0.4859139025211334,
+      "learning_rate": 0.0002,
+      "loss": 1.2772,
+      "step": 110
+    },
+    {
+      "epoch": 0.08800880088008801,
+      "grad_norm": 0.4384922385215759,
+      "learning_rate": 0.0002,
+      "loss": 1.2087,
+      "step": 120
+    },
+    {
+      "epoch": 0.09534286762009535,
+      "grad_norm": 0.39519360661506653,
+      "learning_rate": 0.0002,
+      "loss": 1.2927,
+      "step": 130
+    },
+    {
+      "epoch": 0.10267693436010268,
+      "grad_norm": 0.4049859344959259,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 140
+    },
+    {
+      "epoch": 0.11001100110011001,
+      "grad_norm": 0.4605638086795807,
+      "learning_rate": 0.0002,
+      "loss": 1.293,
+      "step": 150
+    },
+    {
+      "epoch": 0.11734506784011735,
+      "grad_norm": 0.4201928377151489,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 160
+    },
+    {
+      "epoch": 0.12467913458012468,
+      "grad_norm": 0.5367777347564697,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 170
+    },
+    {
+      "epoch": 0.132013201320132,
+      "grad_norm": 0.41752299666404724,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 180
+    },
+    {
+      "epoch": 0.13934726806013933,
+      "grad_norm": 0.31597763299942017,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 190
+    },
+    {
+      "epoch": 0.14668133480014667,
+      "grad_norm": 0.7468788623809814,
+      "learning_rate": 0.0002,
+      "loss": 1.2441,
+      "step": 200
+    },
+    {
+      "epoch": 0.15401540154015403,
+      "grad_norm": 0.3403034508228302,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 210
+    },
+    {
+      "epoch": 0.16134946828016136,
+      "grad_norm": 0.34240293502807617,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 220
+    },
+    {
+      "epoch": 0.1686835350201687,
+      "grad_norm": 0.356158971786499,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 230
+    },
+    {
+      "epoch": 0.17601760176017603,
+      "grad_norm": 0.3448857367038727,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 240
+    },
+    {
+      "epoch": 0.18335166850018336,
+      "grad_norm": 0.3475699722766876,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 250
+    },
+    {
+      "epoch": 0.1906857352401907,
+      "grad_norm": 0.2770358622074127,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 260
+    },
+    {
+      "epoch": 0.19801980198019803,
+      "grad_norm": 0.4310270845890045,
+      "learning_rate": 0.0002,
+      "loss": 1.2238,
+      "step": 270
+    },
+    {
+      "epoch": 0.20535386872020536,
+      "grad_norm": 0.335041880607605,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 280
+    },
+    {
+      "epoch": 0.2126879354602127,
+      "grad_norm": 0.3420602083206177,
+      "learning_rate": 0.0002,
+      "loss": 1.0959,
+      "step": 290
+    },
+    {
+      "epoch": 0.22002200220022003,
+      "grad_norm": 0.325001060962677,
+      "learning_rate": 0.0002,
+      "loss": 1.1232,
+      "step": 300
+    },
+    {
+      "epoch": 0.22735606894022736,
+      "grad_norm": 0.3027827739715576,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 310
+    },
+    {
+      "epoch": 0.2346901356802347,
+      "grad_norm": 0.435550719499588,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 320
+    },
+    {
+      "epoch": 0.24202420242024203,
+      "grad_norm": 0.3884522616863251,
+      "learning_rate": 0.0002,
+      "loss": 1.2045,
+      "step": 330
+    },
+    {
+      "epoch": 0.24935826916024936,
+      "grad_norm": 0.7736002206802368,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 340
+    },
+    {
+      "epoch": 0.2566923359002567,
+      "grad_norm": 0.35052821040153503,
+      "learning_rate": 0.0002,
+      "loss": 1.3606,
+      "step": 350
+    },
+    {
+      "epoch": 0.264026402640264,
+      "grad_norm": 0.3311890959739685,
+      "learning_rate": 0.0002,
+      "loss": 1.2129,
+      "step": 360
+    },
+    {
+      "epoch": 0.27136046938027136,
+      "grad_norm": 0.7473500370979309,
+      "learning_rate": 0.0002,
+      "loss": 1.2219,
+      "step": 370
+    },
+    {
+      "epoch": 0.27869453612027867,
+      "grad_norm": 0.3681875765323639,
+      "learning_rate": 0.0002,
+      "loss": 1.2712,
+      "step": 380
+    },
+    {
+      "epoch": 0.28602860286028603,
+      "grad_norm": 0.3764737844467163,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 390
+    },
+    {
+      "epoch": 0.29336266960029334,
+      "grad_norm": 0.4243989586830139,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 400
+    },
+    {
+      "epoch": 0.3006967363403007,
+      "grad_norm": 0.2658531963825226,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 410
+    },
+    {
+      "epoch": 0.30803080308030806,
+      "grad_norm": 0.3436793386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.1622,
+      "step": 420
+    },
+    {
+      "epoch": 0.31536486982031536,
+      "grad_norm": 0.5101129412651062,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 430
+    },
+    {
+      "epoch": 0.3226989365603227,
+      "grad_norm": 0.3319750726222992,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 440
+    },
+    {
+      "epoch": 0.33003300330033003,
+      "grad_norm": 0.385148286819458,
+      "learning_rate": 0.0002,
+      "loss": 1.1804,
+      "step": 450
+    },
+    {
+      "epoch": 0.3373670700403374,
+      "grad_norm": 0.3477935791015625,
+      "learning_rate": 0.0002,
+      "loss": 1.1808,
+      "step": 460
+    },
+    {
+      "epoch": 0.3447011367803447,
+      "grad_norm": 0.29748716950416565,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 470
+    },
+    {
+      "epoch": 0.35203520352035206,
+      "grad_norm": 0.34083324670791626,
+      "learning_rate": 0.0002,
+      "loss": 1.19,
+      "step": 480
+    },
+    {
+      "epoch": 0.35936927026035936,
+      "grad_norm": 0.36904552578926086,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 490
+    },
+    {
+      "epoch": 0.3667033370003667,
+      "grad_norm": 0.315483033657074,
+      "learning_rate": 0.0002,
+      "loss": 1.2223,
+      "step": 500
+    },
+    {
+      "epoch": 0.37403740374037403,
+      "grad_norm": 0.44897955656051636,
+      "learning_rate": 0.0002,
+      "loss": 1.1461,
+      "step": 510
+    },
+    {
+      "epoch": 0.3813714704803814,
+      "grad_norm": 0.3160701394081116,
+      "learning_rate": 0.0002,
+      "loss": 1.3035,
+      "step": 520
+    },
+    {
+      "epoch": 0.3887055372203887,
+      "grad_norm": 0.29584741592407227,
+      "learning_rate": 0.0002,
+      "loss": 1.3197,
+      "step": 530
+    },
+    {
+      "epoch": 0.39603960396039606,
+      "grad_norm": 0.5430002808570862,
+      "learning_rate": 0.0002,
+      "loss": 1.2983,
+      "step": 540
+    },
+    {
+      "epoch": 0.40337367070040336,
+      "grad_norm": 0.2908070683479309,
+      "learning_rate": 0.0002,
+      "loss": 1.2459,
+      "step": 550
+    },
+    {
+      "epoch": 0.4107077374404107,
+      "grad_norm": 0.35066530108451843,
+      "learning_rate": 0.0002,
+      "loss": 1.2384,
+      "step": 560
+    },
+    {
+      "epoch": 0.41804180418041803,
+      "grad_norm": 0.37588003277778625,
+      "learning_rate": 0.0002,
+      "loss": 1.1784,
+      "step": 570
+    },
+    {
+      "epoch": 0.4253758709204254,
+      "grad_norm": 0.3112126886844635,
+      "learning_rate": 0.0002,
+      "loss": 1.2334,
+      "step": 580
+    },
+    {
+      "epoch": 0.4327099376604327,
+      "grad_norm": 0.35577139258384705,
+      "learning_rate": 0.0002,
+      "loss": 1.1439,
+      "step": 590
+    },
+    {
+      "epoch": 0.44004400440044006,
+      "grad_norm": 0.31706422567367554,
+      "learning_rate": 0.0002,
+      "loss": 1.184,
+      "step": 600
+    },
+    {
+      "epoch": 0.44737807114044736,
+      "grad_norm": 0.3249092102050781,
+      "learning_rate": 0.0002,
+      "loss": 1.2081,
+      "step": 610
+    },
+    {
+      "epoch": 0.4547121378804547,
+      "grad_norm": 0.3842705488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.0824,
+      "step": 620
+    },
+    {
+      "epoch": 0.46204620462046203,
+      "grad_norm": 0.390991747379303,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 630
+    },
+    {
+      "epoch": 0.4693802713604694,
+      "grad_norm": 0.27532413601875305,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 640
+    },
+    {
+      "epoch": 0.4767143381004767,
+      "grad_norm": 0.31412816047668457,
+      "learning_rate": 0.0002,
+      "loss": 1.1058,
+      "step": 650
+    },
+    {
+      "epoch": 0.48404840484048406,
+      "grad_norm": 0.32117101550102234,
+      "learning_rate": 0.0002,
+      "loss": 1.1312,
+      "step": 660
+    },
+    {
+      "epoch": 0.49138247158049136,
+      "grad_norm": 0.3810010254383087,
+      "learning_rate": 0.0002,
+      "loss": 1.2423,
+      "step": 670
+    },
+    {
+      "epoch": 0.4987165383204987,
+      "grad_norm": 0.36289164423942566,
+      "learning_rate": 0.0002,
+      "loss": 1.1978,
+      "step": 680
+    },
+    {
+      "epoch": 0.506050605060506,
+      "grad_norm": 0.34458720684051514,
+      "learning_rate": 0.0002,
+      "loss": 1.2034,
+      "step": 690
+    },
+    {
+      "epoch": 0.5133846718005134,
+      "grad_norm": 0.32844600081443787,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 700
+    },
+    {
+      "epoch": 0.5207187385405208,
+      "grad_norm": 0.3144175708293915,
+      "learning_rate": 0.0002,
+      "loss": 1.0807,
+      "step": 710
+    },
+    {
+      "epoch": 0.528052805280528,
+      "grad_norm": 0.3898887634277344,
+      "learning_rate": 0.0002,
+      "loss": 1.1952,
+      "step": 720
+    },
+    {
+      "epoch": 0.5353868720205354,
+      "grad_norm": 1.3220758438110352,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 730
+    },
+    {
+      "epoch": 0.5427209387605427,
+      "grad_norm": 0.3635874390602112,
+      "learning_rate": 0.0002,
+      "loss": 1.227,
+      "step": 740
+    },
+    {
+      "epoch": 0.5500550055005501,
+      "grad_norm": 0.3138217628002167,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 750
+    },
+    {
+      "epoch": 0.5573890722405573,
+      "grad_norm": 0.4063207805156708,
+      "learning_rate": 0.0002,
+      "loss": 1.1516,
+      "step": 760
+    },
+    {
+      "epoch": 0.5647231389805647,
+      "grad_norm": 0.3926219940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 770
+    },
+    {
+      "epoch": 0.5720572057205721,
+      "grad_norm": 0.31954652070999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1726,
+      "step": 780
+    },
+    {
+      "epoch": 0.5793912724605794,
+      "grad_norm": 0.4248711168766022,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 790
+    },
+    {
+      "epoch": 0.5867253392005867,
+      "grad_norm": 0.643004834651947,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 800
+    },
+    {
+      "epoch": 0.594059405940594,
+      "grad_norm": 0.3479592800140381,
+      "learning_rate": 0.0002,
+      "loss": 1.1793,
+      "step": 810
+    },
+    {
+      "epoch": 0.6013934726806014,
+      "grad_norm": 0.4684754014015198,
+      "learning_rate": 0.0002,
+      "loss": 1.2426,
+      "step": 820
+    },
+    {
+      "epoch": 0.6087275394206088,
+      "grad_norm": 0.3739790916442871,
+      "learning_rate": 0.0002,
+      "loss": 1.2002,
+      "step": 830
+    },
+    {
+      "epoch": 0.6160616061606161,
+      "grad_norm": 0.40884748101234436,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 840
+    },
+    {
+      "epoch": 0.6233956729006234,
+      "grad_norm": 0.9722164273262024,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 850
+    },
+    {
+      "epoch": 0.6307297396406307,
+      "grad_norm": 0.42992347478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.3069,
+      "step": 860
+    },
+    {
+      "epoch": 0.6380638063806381,
+      "grad_norm": 0.36654195189476013,
+      "learning_rate": 0.0002,
+      "loss": 1.1339,
+      "step": 870
+    },
+    {
+      "epoch": 0.6453978731206454,
+      "grad_norm": 0.4113832116127014,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 880
+    },
+    {
+      "epoch": 0.6527319398606527,
+      "grad_norm": 0.2948838770389557,
+      "learning_rate": 0.0002,
+      "loss": 1.2163,
+      "step": 890
+    },
+    {
+      "epoch": 0.6600660066006601,
+      "grad_norm": 0.38330280780792236,
+      "learning_rate": 0.0002,
+      "loss": 1.1081,
+      "step": 900
+    },
+    {
+      "epoch": 0.6674000733406674,
+      "grad_norm": 0.4428867697715759,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 910
+    },
+    {
+      "epoch": 0.6747341400806748,
+      "grad_norm": 0.23659265041351318,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 920
+    },
+    {
+      "epoch": 0.682068206820682,
+      "grad_norm": 0.323685884475708,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 930
+    },
+    {
+      "epoch": 0.6894022735606894,
+      "grad_norm": 0.39157727360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 940
+    },
+    {
+      "epoch": 0.6967363403006968,
+      "grad_norm": 0.27189481258392334,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 950
+    },
+    {
+      "epoch": 0.7040704070407041,
+      "grad_norm": 0.529883861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.1033,
+      "step": 960
+    },
+    {
+      "epoch": 0.7114044737807114,
+      "grad_norm": 0.34758689999580383,
+      "learning_rate": 0.0002,
+      "loss": 1.139,
+      "step": 970
+    },
+    {
+      "epoch": 0.7187385405207187,
+      "grad_norm": 0.831749439239502,
+      "learning_rate": 0.0002,
+      "loss": 1.2197,
+      "step": 980
+    },
+    {
+      "epoch": 0.7260726072607261,
+      "grad_norm": 0.4438304007053375,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 990
+    },
+    {
+      "epoch": 0.7334066740007334,
+      "grad_norm": 0.33840006589889526,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7407407407407407,
+      "grad_norm": 0.3454797863960266,
+      "learning_rate": 0.0002,
+      "loss": 1.254,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7480748074807481,
+      "grad_norm": 0.38999441266059875,
+      "learning_rate": 0.0002,
+      "loss": 1.106,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7554088742207554,
+      "grad_norm": 0.2829911708831787,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7627429409607628,
+      "grad_norm": 0.36918163299560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 1040
+    },
+    {
+      "epoch": 0.77007700770077,
+      "grad_norm": 0.3415680229663849,
+      "learning_rate": 0.0002,
+      "loss": 1.3028,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7774110744407774,
+      "grad_norm": 0.2974182963371277,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7847451411807848,
+      "grad_norm": 0.3880919814109802,
+      "learning_rate": 0.0002,
+      "loss": 1.194,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7920792079207921,
+      "grad_norm": 0.33503302931785583,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7994132746607994,
+      "grad_norm": 0.3728407025337219,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8067473414008067,
+      "grad_norm": 0.3509373664855957,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8140814081408141,
+      "grad_norm": 0.42228564620018005,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8214154748808215,
+      "grad_norm": 0.313467800617218,
+      "learning_rate": 0.0002,
+      "loss": 1.1788,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8287495416208287,
+      "grad_norm": 0.3378850817680359,
+      "learning_rate": 0.0002,
+      "loss": 1.1971,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8360836083608361,
+      "grad_norm": 0.43200382590293884,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8434176751008434,
+      "grad_norm": 0.3309599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.3203,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8507517418408508,
+      "grad_norm": 0.3526846170425415,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1160
+    },
+    {
+      "epoch": 0.858085808580858,
+      "grad_norm": 1.2722247838974,
+      "learning_rate": 0.0002,
+      "loss": 1.0851,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8654198753208654,
+      "grad_norm": 0.34142059087753296,
+      "learning_rate": 0.0002,
+      "loss": 1.0785,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8727539420608728,
+      "grad_norm": 0.3805823028087616,
+      "learning_rate": 0.0002,
+      "loss": 1.2187,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8800880088008801,
+      "grad_norm": 0.3931232690811157,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8874220755408874,
+      "grad_norm": 0.2937372624874115,
+      "learning_rate": 0.0002,
+      "loss": 1.0948,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8947561422808947,
+      "grad_norm": 0.3757196366786957,
+      "learning_rate": 0.0002,
+      "loss": 1.1228,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9020902090209021,
+      "grad_norm": 0.3502705991268158,
+      "learning_rate": 0.0002,
+      "loss": 1.1222,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9094242757609095,
+      "grad_norm": 0.32758915424346924,
+      "learning_rate": 0.0002,
+      "loss": 1.2242,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9167583425009168,
+      "grad_norm": 0.37199416756629944,
+      "learning_rate": 0.0002,
+      "loss": 1.215,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9240924092409241,
+      "grad_norm": 0.3551490604877472,
+      "learning_rate": 0.0002,
+      "loss": 1.1225,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9314264759809314,
+      "grad_norm": 0.2859550714492798,
+      "learning_rate": 0.0002,
+      "loss": 1.1966,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9387605427209388,
+      "grad_norm": 0.427990585565567,
+      "learning_rate": 0.0002,
+      "loss": 1.2186,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9460946094609461,
+      "grad_norm": 0.33717992901802063,
+      "learning_rate": 0.0002,
+      "loss": 1.2848,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9534286762009534,
+      "grad_norm": 0.30225634574890137,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9607627429409608,
+      "grad_norm": 0.385821133852005,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9680968096809681,
+      "grad_norm": 0.35278066992759705,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9754308764209755,
+      "grad_norm": 0.49987098574638367,
+      "learning_rate": 0.0002,
+      "loss": 1.1071,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9827649431609827,
+      "grad_norm": 0.3842747211456299,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9900990099009901,
+      "grad_norm": 0.6274653673171997,
+      "learning_rate": 0.0002,
+      "loss": 1.0862,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9974330766409975,
+      "grad_norm": 0.5239808559417725,
+      "learning_rate": 0.0002,
+      "loss": 1.124,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9996332966629996,
+      "eval_loss": 1.1822267770767212,
+      "eval_runtime": 32.7389,
+      "eval_samples_per_second": 13.165,
+      "eval_steps_per_second": 1.649,
+      "step": 1363
+    },
+    {
+      "epoch": 1.0047671433810048,
+      "grad_norm": 0.45311301946640015,
+      "learning_rate": 0.0002,
+      "loss": 1.096,
+      "step": 1370
+    },
+    {
+      "epoch": 1.012101210121012,
+      "grad_norm": 0.29685574769973755,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0194352768610195,
+      "grad_norm": 0.3290937840938568,
+      "learning_rate": 0.0002,
+      "loss": 1.0302,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0267693436010268,
+      "grad_norm": 0.3801758587360382,
+      "learning_rate": 0.0002,
+      "loss": 1.0295,
+      "step": 1400
+    },
+    {
+      "epoch": 1.034103410341034,
+      "grad_norm": 0.794174313545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0414374770810415,
+      "grad_norm": 0.3854154646396637,
+      "learning_rate": 0.0002,
+      "loss": 1.2232,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0487715438210488,
+      "grad_norm": 0.32702451944351196,
+      "learning_rate": 0.0002,
+      "loss": 1.0652,
+      "step": 1430
+    },
+    {
+      "epoch": 1.056105610561056,
+      "grad_norm": 0.7815203666687012,
+      "learning_rate": 0.0002,
+      "loss": 1.1144,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0634396773010635,
+      "grad_norm": 0.3087436854839325,
+      "learning_rate": 0.0002,
+      "loss": 1.1316,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0707737440410707,
+      "grad_norm": 0.3847602903842926,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0781078107810782,
+      "grad_norm": 0.3693031370639801,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0854418775210855,
+      "grad_norm": 0.4111202359199524,
+      "learning_rate": 0.0002,
+      "loss": 1.0995,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0927759442610927,
+      "grad_norm": 0.41452381014823914,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1001100110011002,
+      "grad_norm": 0.3336445093154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1068,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1074440777411074,
+      "grad_norm": 0.3923407793045044,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1147781444811147,
+      "grad_norm": 0.46215683221817017,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1221122112211221,
+      "grad_norm": 0.3592156767845154,
+      "learning_rate": 0.0002,
+      "loss": 1.1133,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1294462779611294,
+      "grad_norm": 0.361110657453537,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1367803447011369,
+      "grad_norm": 0.5317131280899048,
+      "learning_rate": 0.0002,
+      "loss": 1.1553,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1441144114411441,
+      "grad_norm": 0.3882388174533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1514484781811514,
+      "grad_norm": 0.3259428143501282,
+      "learning_rate": 0.0002,
+      "loss": 1.0805,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1587825449211588,
+      "grad_norm": 0.410935640335083,
+      "learning_rate": 0.0002,
+      "loss": 1.1819,
+      "step": 1580
+    },
+    {
+      "epoch": 1.166116611661166,
+      "grad_norm": 0.44940185546875,
+      "learning_rate": 0.0002,
+      "loss": 1.1143,
+      "step": 1590
+    },
+    {
+      "epoch": 1.1734506784011733,
+      "grad_norm": 0.5106484293937683,
+      "learning_rate": 0.0002,
+      "loss": 1.0334,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1807847451411808,
+      "grad_norm": 0.6603665947914124,
+      "learning_rate": 0.0002,
+      "loss": 1.2376,
+      "step": 1610
+    },
+    {
+      "epoch": 1.188118811881188,
+      "grad_norm": 0.4799964129924774,
+      "learning_rate": 0.0002,
+      "loss": 1.1227,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1954528786211955,
+      "grad_norm": 0.4389883279800415,
+      "learning_rate": 0.0002,
+      "loss": 1.1191,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2027869453612028,
+      "grad_norm": 0.4188813269138336,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 1640
+    },
+    {
+      "epoch": 1.21012101210121,
+      "grad_norm": 0.7132157683372498,
+      "learning_rate": 0.0002,
+      "loss": 1.0605,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2174550788412175,
+      "grad_norm": 0.507480263710022,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2247891455812248,
+      "grad_norm": 0.9452332854270935,
+      "learning_rate": 0.0002,
+      "loss": 0.9948,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2321232123212322,
+      "grad_norm": 0.4121614992618561,
+      "learning_rate": 0.0002,
+      "loss": 1.0228,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2394572790612395,
+      "grad_norm": 0.34230247139930725,
+      "learning_rate": 0.0002,
+      "loss": 1.0366,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2467913458012467,
+      "grad_norm": 0.4026208817958832,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2541254125412542,
+      "grad_norm": 0.46673697233200073,
+      "learning_rate": 0.0002,
+      "loss": 1.0206,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2614594792812615,
+      "grad_norm": 0.38349825143814087,
+      "learning_rate": 0.0002,
+      "loss": 1.0827,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2687935460212687,
+      "grad_norm": 0.4049997627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.0356,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2761276127612762,
+      "grad_norm": 0.3417615294456482,
+      "learning_rate": 0.0002,
+      "loss": 0.9504,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2834616795012834,
+      "grad_norm": 0.4277614951133728,
+      "learning_rate": 0.0002,
+      "loss": 1.094,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2907957462412907,
+      "grad_norm": 0.5864202976226807,
+      "learning_rate": 0.0002,
+      "loss": 0.9938,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2981298129812981,
+      "grad_norm": 0.7097493410110474,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3054638797213054,
+      "grad_norm": 0.3145381212234497,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3127979464613129,
+      "grad_norm": 0.5116165280342102,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3201320132013201,
+      "grad_norm": 0.7469736337661743,
+      "learning_rate": 0.0002,
+      "loss": 1.0765,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3274660799413276,
+      "grad_norm": 0.32272255420684814,
+      "learning_rate": 0.0002,
+      "loss": 1.0663,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3348001466813348,
+      "grad_norm": 0.3534623086452484,
+      "learning_rate": 0.0002,
+      "loss": 0.9887,
+      "step": 1820
+    },
+    {
+      "epoch": 1.342134213421342,
+      "grad_norm": 0.36127907037734985,
+      "learning_rate": 0.0002,
+      "loss": 1.1628,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3494682801613496,
+      "grad_norm": 0.4072401523590088,
+      "learning_rate": 0.0002,
+      "loss": 1.0972,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3568023469013568,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.1267,
+      "step": 1850
+    },
+    {
+      "epoch": 1.364136413641364,
+      "grad_norm": 0.412883460521698,
+      "learning_rate": 0.0002,
+      "loss": 1.0173,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3714704803813715,
+      "grad_norm": 0.3735875189304352,
+      "learning_rate": 0.0002,
+      "loss": 1.0265,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3788045471213788,
+      "grad_norm": 0.39158159494400024,
+      "learning_rate": 0.0002,
+      "loss": 1.1061,
+      "step": 1880
+    },
+    {
+      "epoch": 1.386138613861386,
+      "grad_norm": 0.44431769847869873,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3934726806013935,
+      "grad_norm": 0.37772801518440247,
+      "learning_rate": 0.0002,
+      "loss": 1.0216,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4008067473414008,
+      "grad_norm": 0.4056641757488251,
+      "learning_rate": 0.0002,
+      "loss": 1.0674,
+      "step": 1910
+    },
+    {
+      "epoch": 1.408140814081408,
+      "grad_norm": 0.41612377762794495,
+      "learning_rate": 0.0002,
+      "loss": 1.0256,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4154748808214155,
+      "grad_norm": 0.41153013706207275,
+      "learning_rate": 0.0002,
+      "loss": 1.0467,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4228089475614227,
+      "grad_norm": 0.387845516204834,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4301430143014302,
+      "grad_norm": 0.3809587061405182,
+      "learning_rate": 0.0002,
+      "loss": 1.1094,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4374770810414375,
+      "grad_norm": 0.3625726103782654,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1960
+    },
+    {
+      "epoch": 1.444811147781445,
+      "grad_norm": 0.5294290781021118,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4521452145214522,
+      "grad_norm": 0.39975494146347046,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4594792812614594,
+      "grad_norm": 0.4181167185306549,
+      "learning_rate": 0.0002,
+      "loss": 0.9704,
+      "step": 1990
+    },
+    {
+      "epoch": 1.466813348001467,
+      "grad_norm": 0.42001503705978394,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4741474147414741,
+      "grad_norm": 0.4877578616142273,
+      "learning_rate": 0.0002,
+      "loss": 1.1266,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4814814814814814,
+      "grad_norm": 0.4050969183444977,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4888155482214889,
+      "grad_norm": 0.39068883657455444,
+      "learning_rate": 0.0002,
+      "loss": 1.0562,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4961496149614961,
+      "grad_norm": 0.421282559633255,
+      "learning_rate": 0.0002,
+      "loss": 1.0464,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5034836817015034,
+      "grad_norm": 0.47092297673225403,
+      "learning_rate": 0.0002,
+      "loss": 1.0532,
+      "step": 2050
+    },
+    {
+      "epoch": 1.5108177484415108,
+      "grad_norm": 0.39688974618911743,
+      "learning_rate": 0.0002,
+      "loss": 0.9348,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5181518151815183,
+      "grad_norm": 0.5529879331588745,
+      "learning_rate": 0.0002,
+      "loss": 1.08,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5254858819215253,
+      "grad_norm": 0.4879782199859619,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5328199486615328,
+      "grad_norm": 0.5517361164093018,
+      "learning_rate": 0.0002,
+      "loss": 1.0432,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5401540154015403,
+      "grad_norm": 0.44015637040138245,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5474880821415475,
+      "grad_norm": 0.5435167551040649,
+      "learning_rate": 0.0002,
+      "loss": 1.1873,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5548221488815548,
+      "grad_norm": 0.5714033246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1076,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5621562156215623,
+      "grad_norm": 0.31732529401779175,
+      "learning_rate": 0.0002,
+      "loss": 1.1107,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5694902823615695,
+      "grad_norm": 0.49068278074264526,
+      "learning_rate": 0.0002,
+      "loss": 1.0817,
+      "step": 2140
+    },
+    {
+      "epoch": 1.5768243491015768,
+      "grad_norm": 0.46851542592048645,
+      "learning_rate": 0.0002,
+      "loss": 1.0254,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5841584158415842,
+      "grad_norm": 0.5083092451095581,
+      "learning_rate": 0.0002,
+      "loss": 1.0623,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5914924825815915,
+      "grad_norm": 0.9822936058044434,
+      "learning_rate": 0.0002,
+      "loss": 1.0603,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5988265493215987,
+      "grad_norm": 0.4575989246368408,
+      "learning_rate": 0.0002,
+      "loss": 0.9986,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6061606160616062,
+      "grad_norm": 0.47444286942481995,
+      "learning_rate": 0.0002,
+      "loss": 1.1292,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6134946828016135,
+      "grad_norm": 0.7208226919174194,
+      "learning_rate": 0.0002,
+      "loss": 1.0136,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6208287495416207,
+      "grad_norm": 0.43791481852531433,
+      "learning_rate": 0.0002,
+      "loss": 1.15,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6281628162816282,
+      "grad_norm": 0.5245792865753174,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6354968830216357,
+      "grad_norm": 0.39289429783821106,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6428309497616427,
+      "grad_norm": 0.6106135845184326,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6501650165016502,
+      "grad_norm": 0.3722580671310425,
+      "learning_rate": 0.0002,
+      "loss": 1.0129,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6574990832416576,
+      "grad_norm": 0.3649403750896454,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6648331499816649,
+      "grad_norm": 0.46514248847961426,
+      "learning_rate": 0.0002,
+      "loss": 1.0037,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6721672167216721,
+      "grad_norm": 0.42034927010536194,
+      "learning_rate": 0.0002,
+      "loss": 1.0022,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6795012834616796,
+      "grad_norm": 0.45202910900115967,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6868353502016868,
+      "grad_norm": 0.36257603764533997,
+      "learning_rate": 0.0002,
+      "loss": 1.0866,
+      "step": 2300
+    },
+    {
+      "epoch": 1.694169416941694,
+      "grad_norm": 0.6340323090553284,
+      "learning_rate": 0.0002,
+      "loss": 1.0973,
+      "step": 2310
+    },
+    {
+      "epoch": 1.7015034836817016,
+      "grad_norm": 0.4352878928184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0615,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7088375504217088,
+      "grad_norm": 0.45029792189598083,
+      "learning_rate": 0.0002,
+      "loss": 1.0629,
+      "step": 2330
+    },
+    {
+      "epoch": 1.716171617161716,
+      "grad_norm": 0.3891315758228302,
+      "learning_rate": 0.0002,
+      "loss": 0.9621,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7235056839017235,
+      "grad_norm": 0.35180050134658813,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7308397506417308,
+      "grad_norm": 0.42367449402809143,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 2360
+    },
+    {
+      "epoch": 1.738173817381738,
+      "grad_norm": 0.4553675353527069,
+      "learning_rate": 0.0002,
+      "loss": 1.0376,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7455078841217455,
+      "grad_norm": 0.5944654941558838,
+      "learning_rate": 0.0002,
+      "loss": 1.1467,
+      "step": 2380
+    },
+    {
+      "epoch": 1.752841950861753,
+      "grad_norm": 0.3479664623737335,
+      "learning_rate": 0.0002,
+      "loss": 1.0548,
+      "step": 2390
+    },
+    {
+      "epoch": 1.76017601760176,
+      "grad_norm": 0.3585502505302429,
+      "learning_rate": 0.0002,
+      "loss": 1.0798,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7675100843417675,
+      "grad_norm": 0.4263346493244171,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 2410
+    },
+    {
+      "epoch": 1.774844151081775,
+      "grad_norm": 0.5476409196853638,
+      "learning_rate": 0.0002,
+      "loss": 1.054,
+      "step": 2420
+    },
+    {
+      "epoch": 1.7821782178217822,
+      "grad_norm": 0.3694186508655548,
+      "learning_rate": 0.0002,
+      "loss": 1.1615,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7895122845617895,
+      "grad_norm": 0.9185658693313599,
+      "learning_rate": 0.0002,
+      "loss": 1.1343,
+      "step": 2440
+    },
+    {
+      "epoch": 1.796846351301797,
+      "grad_norm": 0.7171908020973206,
+      "learning_rate": 0.0002,
+      "loss": 1.0764,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8041804180418042,
+      "grad_norm": 0.550658643245697,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 2460
+    },
+    {
+      "epoch": 1.8115144847818114,
+      "grad_norm": 0.4075568914413452,
+      "learning_rate": 0.0002,
+      "loss": 0.9975,
+      "step": 2470
+    },
+    {
+      "epoch": 1.818848551521819,
+      "grad_norm": 0.3790127635002136,
+      "learning_rate": 0.0002,
+      "loss": 1.0935,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8261826182618262,
+      "grad_norm": 0.3576384484767914,
+      "learning_rate": 0.0002,
+      "loss": 0.9839,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8335166850018334,
+      "grad_norm": 0.3919370770454407,
+      "learning_rate": 0.0002,
+      "loss": 1.1369,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8408507517418409,
+      "grad_norm": 0.485083669424057,
+      "learning_rate": 0.0002,
+      "loss": 0.9985,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8481848184818483,
+      "grad_norm": 0.4564347565174103,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8555188852218554,
+      "grad_norm": 0.3613106608390808,
+      "learning_rate": 0.0002,
+      "loss": 1.0944,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8628529519618628,
+      "grad_norm": 0.39600759744644165,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8701870187018703,
+      "grad_norm": 1.123499870300293,
+      "learning_rate": 0.0002,
+      "loss": 0.9453,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8775210854418776,
+      "grad_norm": 0.4612680673599243,
+      "learning_rate": 0.0002,
+      "loss": 1.0635,
+      "step": 2560
+    },
+    {
+      "epoch": 1.8848551521818848,
+      "grad_norm": 0.42745399475097656,
+      "learning_rate": 0.0002,
+      "loss": 1.0087,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8921892189218923,
+      "grad_norm": 0.4055580198764801,
+      "learning_rate": 0.0002,
+      "loss": 1.0102,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8995232856618995,
+      "grad_norm": 0.44174644351005554,
+      "learning_rate": 0.0002,
+      "loss": 1.0177,
+      "step": 2590
+    },
+    {
+      "epoch": 1.9068573524019068,
+      "grad_norm": 1.0228385925292969,
+      "learning_rate": 0.0002,
+      "loss": 0.9886,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9141914191419143,
+      "grad_norm": 0.3496396243572235,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9215254858819215,
+      "grad_norm": 0.4191173017024994,
+      "learning_rate": 0.0002,
+      "loss": 1.0955,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9288595526219288,
+      "grad_norm": 0.6778554916381836,
+      "learning_rate": 0.0002,
+      "loss": 1.0943,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9361936193619362,
+      "grad_norm": 0.41992834210395813,
+      "learning_rate": 0.0002,
+      "loss": 1.0594,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9435276861019435,
+      "grad_norm": 0.8760401010513306,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9508617528419507,
+      "grad_norm": 0.44049209356307983,
+      "learning_rate": 0.0002,
+      "loss": 1.0379,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9581958195819582,
+      "grad_norm": 0.5651928782463074,
+      "learning_rate": 0.0002,
+      "loss": 1.1008,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9655298863219657,
+      "grad_norm": 0.5292727947235107,
+      "learning_rate": 0.0002,
+      "loss": 1.1317,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9728639530619727,
+      "grad_norm": 0.6012240648269653,
+      "learning_rate": 0.0002,
+      "loss": 1.1328,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9801980198019802,
+      "grad_norm": 0.3945149779319763,
+      "learning_rate": 0.0002,
+      "loss": 1.0683,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9875320865419877,
+      "grad_norm": 0.5732627511024475,
+      "learning_rate": 0.0002,
+      "loss": 1.0155,
+      "step": 2710
+    },
+    {
+      "epoch": 1.994866153281995,
+      "grad_norm": 0.3963361084461212,
+      "learning_rate": 0.0002,
+      "loss": 0.9857,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1534006595611572,
+      "eval_runtime": 32.7541,
+      "eval_samples_per_second": 13.159,
+      "eval_steps_per_second": 1.649,
+      "step": 2727
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10904,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "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": 1.2619945399969382e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..2868cff7027115396e695775cacd838522aca295
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b12b6f6817632087b5a5e37d744e25312b96e839de5005320b96bc0c2473c41f
+size 5624
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/README.md b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# 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 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "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": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5682d88ee2c00657eba0aa767d87c4044865cd38
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef302d6be463e56e29f4b08958ebf19ddc0559b0548528c413818f5e2c5a33cc
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2fb0b79ff10e1abd1e4528cbccf466e58c08eafd
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7527c1f97e0d3f127cfe7f87b8021637522d12b2548963523706830f16c9912e
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..b6442894d3deec9f7988bcb5bdf90a11d8f0f19f
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b56c7af373f43eefb93705238cb39dbcfbe71e194b625d87c287dda9ba71d602
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7a1e4ab150dec5b8006a44a3808cfd38f38bee8c
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8935eb574027b699cfe689b5551ebf631eefee73b80f79698d1ad69b75c19ff0
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..9a64707a750ae30fed316e5ed852d68a0db13eeb
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/trainer_state.json
@@ -0,0 +1,2920 @@
+{
+  "best_metric": 1.1534006595611572,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727",
+  "epoch": 2.9996332966629997,
+  "eval_steps": 10,
+  "global_step": 4090,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007334066740007334,
+      "grad_norm": 0.47521963715553284,
+      "learning_rate": 0.0002,
+      "loss": 1.9722,
+      "step": 10
+    },
+    {
+      "epoch": 0.014668133480014669,
+      "grad_norm": 0.5395162105560303,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 20
+    },
+    {
+      "epoch": 0.022002200220022004,
+      "grad_norm": 0.4305780231952667,
+      "learning_rate": 0.0002,
+      "loss": 1.4202,
+      "step": 30
+    },
+    {
+      "epoch": 0.029336266960029337,
+      "grad_norm": 0.6938246488571167,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 40
+    },
+    {
+      "epoch": 0.03667033370003667,
+      "grad_norm": 1.5133819580078125,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 50
+    },
+    {
+      "epoch": 0.04400440044004401,
+      "grad_norm": 0.9173883199691772,
+      "learning_rate": 0.0002,
+      "loss": 1.3132,
+      "step": 60
+    },
+    {
+      "epoch": 0.05133846718005134,
+      "grad_norm": 0.4619861841201782,
+      "learning_rate": 0.0002,
+      "loss": 1.2844,
+      "step": 70
+    },
+    {
+      "epoch": 0.058672533920058674,
+      "grad_norm": 0.46118637919425964,
+      "learning_rate": 0.0002,
+      "loss": 1.2108,
+      "step": 80
+    },
+    {
+      "epoch": 0.066006600660066,
+      "grad_norm": 0.4468648135662079,
+      "learning_rate": 0.0002,
+      "loss": 1.3441,
+      "step": 90
+    },
+    {
+      "epoch": 0.07334066740007333,
+      "grad_norm": 0.46123769879341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1863,
+      "step": 100
+    },
+    {
+      "epoch": 0.08067473414008068,
+      "grad_norm": 0.4859139025211334,
+      "learning_rate": 0.0002,
+      "loss": 1.2772,
+      "step": 110
+    },
+    {
+      "epoch": 0.08800880088008801,
+      "grad_norm": 0.4384922385215759,
+      "learning_rate": 0.0002,
+      "loss": 1.2087,
+      "step": 120
+    },
+    {
+      "epoch": 0.09534286762009535,
+      "grad_norm": 0.39519360661506653,
+      "learning_rate": 0.0002,
+      "loss": 1.2927,
+      "step": 130
+    },
+    {
+      "epoch": 0.10267693436010268,
+      "grad_norm": 0.4049859344959259,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 140
+    },
+    {
+      "epoch": 0.11001100110011001,
+      "grad_norm": 0.4605638086795807,
+      "learning_rate": 0.0002,
+      "loss": 1.293,
+      "step": 150
+    },
+    {
+      "epoch": 0.11734506784011735,
+      "grad_norm": 0.4201928377151489,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 160
+    },
+    {
+      "epoch": 0.12467913458012468,
+      "grad_norm": 0.5367777347564697,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 170
+    },
+    {
+      "epoch": 0.132013201320132,
+      "grad_norm": 0.41752299666404724,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 180
+    },
+    {
+      "epoch": 0.13934726806013933,
+      "grad_norm": 0.31597763299942017,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 190
+    },
+    {
+      "epoch": 0.14668133480014667,
+      "grad_norm": 0.7468788623809814,
+      "learning_rate": 0.0002,
+      "loss": 1.2441,
+      "step": 200
+    },
+    {
+      "epoch": 0.15401540154015403,
+      "grad_norm": 0.3403034508228302,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 210
+    },
+    {
+      "epoch": 0.16134946828016136,
+      "grad_norm": 0.34240293502807617,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 220
+    },
+    {
+      "epoch": 0.1686835350201687,
+      "grad_norm": 0.356158971786499,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 230
+    },
+    {
+      "epoch": 0.17601760176017603,
+      "grad_norm": 0.3448857367038727,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 240
+    },
+    {
+      "epoch": 0.18335166850018336,
+      "grad_norm": 0.3475699722766876,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 250
+    },
+    {
+      "epoch": 0.1906857352401907,
+      "grad_norm": 0.2770358622074127,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 260
+    },
+    {
+      "epoch": 0.19801980198019803,
+      "grad_norm": 0.4310270845890045,
+      "learning_rate": 0.0002,
+      "loss": 1.2238,
+      "step": 270
+    },
+    {
+      "epoch": 0.20535386872020536,
+      "grad_norm": 0.335041880607605,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 280
+    },
+    {
+      "epoch": 0.2126879354602127,
+      "grad_norm": 0.3420602083206177,
+      "learning_rate": 0.0002,
+      "loss": 1.0959,
+      "step": 290
+    },
+    {
+      "epoch": 0.22002200220022003,
+      "grad_norm": 0.325001060962677,
+      "learning_rate": 0.0002,
+      "loss": 1.1232,
+      "step": 300
+    },
+    {
+      "epoch": 0.22735606894022736,
+      "grad_norm": 0.3027827739715576,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 310
+    },
+    {
+      "epoch": 0.2346901356802347,
+      "grad_norm": 0.435550719499588,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 320
+    },
+    {
+      "epoch": 0.24202420242024203,
+      "grad_norm": 0.3884522616863251,
+      "learning_rate": 0.0002,
+      "loss": 1.2045,
+      "step": 330
+    },
+    {
+      "epoch": 0.24935826916024936,
+      "grad_norm": 0.7736002206802368,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 340
+    },
+    {
+      "epoch": 0.2566923359002567,
+      "grad_norm": 0.35052821040153503,
+      "learning_rate": 0.0002,
+      "loss": 1.3606,
+      "step": 350
+    },
+    {
+      "epoch": 0.264026402640264,
+      "grad_norm": 0.3311890959739685,
+      "learning_rate": 0.0002,
+      "loss": 1.2129,
+      "step": 360
+    },
+    {
+      "epoch": 0.27136046938027136,
+      "grad_norm": 0.7473500370979309,
+      "learning_rate": 0.0002,
+      "loss": 1.2219,
+      "step": 370
+    },
+    {
+      "epoch": 0.27869453612027867,
+      "grad_norm": 0.3681875765323639,
+      "learning_rate": 0.0002,
+      "loss": 1.2712,
+      "step": 380
+    },
+    {
+      "epoch": 0.28602860286028603,
+      "grad_norm": 0.3764737844467163,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 390
+    },
+    {
+      "epoch": 0.29336266960029334,
+      "grad_norm": 0.4243989586830139,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 400
+    },
+    {
+      "epoch": 0.3006967363403007,
+      "grad_norm": 0.2658531963825226,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 410
+    },
+    {
+      "epoch": 0.30803080308030806,
+      "grad_norm": 0.3436793386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.1622,
+      "step": 420
+    },
+    {
+      "epoch": 0.31536486982031536,
+      "grad_norm": 0.5101129412651062,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 430
+    },
+    {
+      "epoch": 0.3226989365603227,
+      "grad_norm": 0.3319750726222992,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 440
+    },
+    {
+      "epoch": 0.33003300330033003,
+      "grad_norm": 0.385148286819458,
+      "learning_rate": 0.0002,
+      "loss": 1.1804,
+      "step": 450
+    },
+    {
+      "epoch": 0.3373670700403374,
+      "grad_norm": 0.3477935791015625,
+      "learning_rate": 0.0002,
+      "loss": 1.1808,
+      "step": 460
+    },
+    {
+      "epoch": 0.3447011367803447,
+      "grad_norm": 0.29748716950416565,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 470
+    },
+    {
+      "epoch": 0.35203520352035206,
+      "grad_norm": 0.34083324670791626,
+      "learning_rate": 0.0002,
+      "loss": 1.19,
+      "step": 480
+    },
+    {
+      "epoch": 0.35936927026035936,
+      "grad_norm": 0.36904552578926086,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 490
+    },
+    {
+      "epoch": 0.3667033370003667,
+      "grad_norm": 0.315483033657074,
+      "learning_rate": 0.0002,
+      "loss": 1.2223,
+      "step": 500
+    },
+    {
+      "epoch": 0.37403740374037403,
+      "grad_norm": 0.44897955656051636,
+      "learning_rate": 0.0002,
+      "loss": 1.1461,
+      "step": 510
+    },
+    {
+      "epoch": 0.3813714704803814,
+      "grad_norm": 0.3160701394081116,
+      "learning_rate": 0.0002,
+      "loss": 1.3035,
+      "step": 520
+    },
+    {
+      "epoch": 0.3887055372203887,
+      "grad_norm": 0.29584741592407227,
+      "learning_rate": 0.0002,
+      "loss": 1.3197,
+      "step": 530
+    },
+    {
+      "epoch": 0.39603960396039606,
+      "grad_norm": 0.5430002808570862,
+      "learning_rate": 0.0002,
+      "loss": 1.2983,
+      "step": 540
+    },
+    {
+      "epoch": 0.40337367070040336,
+      "grad_norm": 0.2908070683479309,
+      "learning_rate": 0.0002,
+      "loss": 1.2459,
+      "step": 550
+    },
+    {
+      "epoch": 0.4107077374404107,
+      "grad_norm": 0.35066530108451843,
+      "learning_rate": 0.0002,
+      "loss": 1.2384,
+      "step": 560
+    },
+    {
+      "epoch": 0.41804180418041803,
+      "grad_norm": 0.37588003277778625,
+      "learning_rate": 0.0002,
+      "loss": 1.1784,
+      "step": 570
+    },
+    {
+      "epoch": 0.4253758709204254,
+      "grad_norm": 0.3112126886844635,
+      "learning_rate": 0.0002,
+      "loss": 1.2334,
+      "step": 580
+    },
+    {
+      "epoch": 0.4327099376604327,
+      "grad_norm": 0.35577139258384705,
+      "learning_rate": 0.0002,
+      "loss": 1.1439,
+      "step": 590
+    },
+    {
+      "epoch": 0.44004400440044006,
+      "grad_norm": 0.31706422567367554,
+      "learning_rate": 0.0002,
+      "loss": 1.184,
+      "step": 600
+    },
+    {
+      "epoch": 0.44737807114044736,
+      "grad_norm": 0.3249092102050781,
+      "learning_rate": 0.0002,
+      "loss": 1.2081,
+      "step": 610
+    },
+    {
+      "epoch": 0.4547121378804547,
+      "grad_norm": 0.3842705488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.0824,
+      "step": 620
+    },
+    {
+      "epoch": 0.46204620462046203,
+      "grad_norm": 0.390991747379303,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 630
+    },
+    {
+      "epoch": 0.4693802713604694,
+      "grad_norm": 0.27532413601875305,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 640
+    },
+    {
+      "epoch": 0.4767143381004767,
+      "grad_norm": 0.31412816047668457,
+      "learning_rate": 0.0002,
+      "loss": 1.1058,
+      "step": 650
+    },
+    {
+      "epoch": 0.48404840484048406,
+      "grad_norm": 0.32117101550102234,
+      "learning_rate": 0.0002,
+      "loss": 1.1312,
+      "step": 660
+    },
+    {
+      "epoch": 0.49138247158049136,
+      "grad_norm": 0.3810010254383087,
+      "learning_rate": 0.0002,
+      "loss": 1.2423,
+      "step": 670
+    },
+    {
+      "epoch": 0.4987165383204987,
+      "grad_norm": 0.36289164423942566,
+      "learning_rate": 0.0002,
+      "loss": 1.1978,
+      "step": 680
+    },
+    {
+      "epoch": 0.506050605060506,
+      "grad_norm": 0.34458720684051514,
+      "learning_rate": 0.0002,
+      "loss": 1.2034,
+      "step": 690
+    },
+    {
+      "epoch": 0.5133846718005134,
+      "grad_norm": 0.32844600081443787,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 700
+    },
+    {
+      "epoch": 0.5207187385405208,
+      "grad_norm": 0.3144175708293915,
+      "learning_rate": 0.0002,
+      "loss": 1.0807,
+      "step": 710
+    },
+    {
+      "epoch": 0.528052805280528,
+      "grad_norm": 0.3898887634277344,
+      "learning_rate": 0.0002,
+      "loss": 1.1952,
+      "step": 720
+    },
+    {
+      "epoch": 0.5353868720205354,
+      "grad_norm": 1.3220758438110352,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 730
+    },
+    {
+      "epoch": 0.5427209387605427,
+      "grad_norm": 0.3635874390602112,
+      "learning_rate": 0.0002,
+      "loss": 1.227,
+      "step": 740
+    },
+    {
+      "epoch": 0.5500550055005501,
+      "grad_norm": 0.3138217628002167,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 750
+    },
+    {
+      "epoch": 0.5573890722405573,
+      "grad_norm": 0.4063207805156708,
+      "learning_rate": 0.0002,
+      "loss": 1.1516,
+      "step": 760
+    },
+    {
+      "epoch": 0.5647231389805647,
+      "grad_norm": 0.3926219940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 770
+    },
+    {
+      "epoch": 0.5720572057205721,
+      "grad_norm": 0.31954652070999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1726,
+      "step": 780
+    },
+    {
+      "epoch": 0.5793912724605794,
+      "grad_norm": 0.4248711168766022,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 790
+    },
+    {
+      "epoch": 0.5867253392005867,
+      "grad_norm": 0.643004834651947,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 800
+    },
+    {
+      "epoch": 0.594059405940594,
+      "grad_norm": 0.3479592800140381,
+      "learning_rate": 0.0002,
+      "loss": 1.1793,
+      "step": 810
+    },
+    {
+      "epoch": 0.6013934726806014,
+      "grad_norm": 0.4684754014015198,
+      "learning_rate": 0.0002,
+      "loss": 1.2426,
+      "step": 820
+    },
+    {
+      "epoch": 0.6087275394206088,
+      "grad_norm": 0.3739790916442871,
+      "learning_rate": 0.0002,
+      "loss": 1.2002,
+      "step": 830
+    },
+    {
+      "epoch": 0.6160616061606161,
+      "grad_norm": 0.40884748101234436,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 840
+    },
+    {
+      "epoch": 0.6233956729006234,
+      "grad_norm": 0.9722164273262024,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 850
+    },
+    {
+      "epoch": 0.6307297396406307,
+      "grad_norm": 0.42992347478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.3069,
+      "step": 860
+    },
+    {
+      "epoch": 0.6380638063806381,
+      "grad_norm": 0.36654195189476013,
+      "learning_rate": 0.0002,
+      "loss": 1.1339,
+      "step": 870
+    },
+    {
+      "epoch": 0.6453978731206454,
+      "grad_norm": 0.4113832116127014,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 880
+    },
+    {
+      "epoch": 0.6527319398606527,
+      "grad_norm": 0.2948838770389557,
+      "learning_rate": 0.0002,
+      "loss": 1.2163,
+      "step": 890
+    },
+    {
+      "epoch": 0.6600660066006601,
+      "grad_norm": 0.38330280780792236,
+      "learning_rate": 0.0002,
+      "loss": 1.1081,
+      "step": 900
+    },
+    {
+      "epoch": 0.6674000733406674,
+      "grad_norm": 0.4428867697715759,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 910
+    },
+    {
+      "epoch": 0.6747341400806748,
+      "grad_norm": 0.23659265041351318,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 920
+    },
+    {
+      "epoch": 0.682068206820682,
+      "grad_norm": 0.323685884475708,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 930
+    },
+    {
+      "epoch": 0.6894022735606894,
+      "grad_norm": 0.39157727360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 940
+    },
+    {
+      "epoch": 0.6967363403006968,
+      "grad_norm": 0.27189481258392334,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 950
+    },
+    {
+      "epoch": 0.7040704070407041,
+      "grad_norm": 0.529883861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.1033,
+      "step": 960
+    },
+    {
+      "epoch": 0.7114044737807114,
+      "grad_norm": 0.34758689999580383,
+      "learning_rate": 0.0002,
+      "loss": 1.139,
+      "step": 970
+    },
+    {
+      "epoch": 0.7187385405207187,
+      "grad_norm": 0.831749439239502,
+      "learning_rate": 0.0002,
+      "loss": 1.2197,
+      "step": 980
+    },
+    {
+      "epoch": 0.7260726072607261,
+      "grad_norm": 0.4438304007053375,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 990
+    },
+    {
+      "epoch": 0.7334066740007334,
+      "grad_norm": 0.33840006589889526,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7407407407407407,
+      "grad_norm": 0.3454797863960266,
+      "learning_rate": 0.0002,
+      "loss": 1.254,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7480748074807481,
+      "grad_norm": 0.38999441266059875,
+      "learning_rate": 0.0002,
+      "loss": 1.106,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7554088742207554,
+      "grad_norm": 0.2829911708831787,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7627429409607628,
+      "grad_norm": 0.36918163299560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 1040
+    },
+    {
+      "epoch": 0.77007700770077,
+      "grad_norm": 0.3415680229663849,
+      "learning_rate": 0.0002,
+      "loss": 1.3028,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7774110744407774,
+      "grad_norm": 0.2974182963371277,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7847451411807848,
+      "grad_norm": 0.3880919814109802,
+      "learning_rate": 0.0002,
+      "loss": 1.194,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7920792079207921,
+      "grad_norm": 0.33503302931785583,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7994132746607994,
+      "grad_norm": 0.3728407025337219,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8067473414008067,
+      "grad_norm": 0.3509373664855957,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8140814081408141,
+      "grad_norm": 0.42228564620018005,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8214154748808215,
+      "grad_norm": 0.313467800617218,
+      "learning_rate": 0.0002,
+      "loss": 1.1788,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8287495416208287,
+      "grad_norm": 0.3378850817680359,
+      "learning_rate": 0.0002,
+      "loss": 1.1971,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8360836083608361,
+      "grad_norm": 0.43200382590293884,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8434176751008434,
+      "grad_norm": 0.3309599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.3203,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8507517418408508,
+      "grad_norm": 0.3526846170425415,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1160
+    },
+    {
+      "epoch": 0.858085808580858,
+      "grad_norm": 1.2722247838974,
+      "learning_rate": 0.0002,
+      "loss": 1.0851,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8654198753208654,
+      "grad_norm": 0.34142059087753296,
+      "learning_rate": 0.0002,
+      "loss": 1.0785,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8727539420608728,
+      "grad_norm": 0.3805823028087616,
+      "learning_rate": 0.0002,
+      "loss": 1.2187,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8800880088008801,
+      "grad_norm": 0.3931232690811157,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8874220755408874,
+      "grad_norm": 0.2937372624874115,
+      "learning_rate": 0.0002,
+      "loss": 1.0948,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8947561422808947,
+      "grad_norm": 0.3757196366786957,
+      "learning_rate": 0.0002,
+      "loss": 1.1228,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9020902090209021,
+      "grad_norm": 0.3502705991268158,
+      "learning_rate": 0.0002,
+      "loss": 1.1222,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9094242757609095,
+      "grad_norm": 0.32758915424346924,
+      "learning_rate": 0.0002,
+      "loss": 1.2242,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9167583425009168,
+      "grad_norm": 0.37199416756629944,
+      "learning_rate": 0.0002,
+      "loss": 1.215,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9240924092409241,
+      "grad_norm": 0.3551490604877472,
+      "learning_rate": 0.0002,
+      "loss": 1.1225,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9314264759809314,
+      "grad_norm": 0.2859550714492798,
+      "learning_rate": 0.0002,
+      "loss": 1.1966,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9387605427209388,
+      "grad_norm": 0.427990585565567,
+      "learning_rate": 0.0002,
+      "loss": 1.2186,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9460946094609461,
+      "grad_norm": 0.33717992901802063,
+      "learning_rate": 0.0002,
+      "loss": 1.2848,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9534286762009534,
+      "grad_norm": 0.30225634574890137,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9607627429409608,
+      "grad_norm": 0.385821133852005,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9680968096809681,
+      "grad_norm": 0.35278066992759705,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9754308764209755,
+      "grad_norm": 0.49987098574638367,
+      "learning_rate": 0.0002,
+      "loss": 1.1071,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9827649431609827,
+      "grad_norm": 0.3842747211456299,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9900990099009901,
+      "grad_norm": 0.6274653673171997,
+      "learning_rate": 0.0002,
+      "loss": 1.0862,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9974330766409975,
+      "grad_norm": 0.5239808559417725,
+      "learning_rate": 0.0002,
+      "loss": 1.124,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9996332966629996,
+      "eval_loss": 1.1822267770767212,
+      "eval_runtime": 32.7389,
+      "eval_samples_per_second": 13.165,
+      "eval_steps_per_second": 1.649,
+      "step": 1363
+    },
+    {
+      "epoch": 1.0047671433810048,
+      "grad_norm": 0.45311301946640015,
+      "learning_rate": 0.0002,
+      "loss": 1.096,
+      "step": 1370
+    },
+    {
+      "epoch": 1.012101210121012,
+      "grad_norm": 0.29685574769973755,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0194352768610195,
+      "grad_norm": 0.3290937840938568,
+      "learning_rate": 0.0002,
+      "loss": 1.0302,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0267693436010268,
+      "grad_norm": 0.3801758587360382,
+      "learning_rate": 0.0002,
+      "loss": 1.0295,
+      "step": 1400
+    },
+    {
+      "epoch": 1.034103410341034,
+      "grad_norm": 0.794174313545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0414374770810415,
+      "grad_norm": 0.3854154646396637,
+      "learning_rate": 0.0002,
+      "loss": 1.2232,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0487715438210488,
+      "grad_norm": 0.32702451944351196,
+      "learning_rate": 0.0002,
+      "loss": 1.0652,
+      "step": 1430
+    },
+    {
+      "epoch": 1.056105610561056,
+      "grad_norm": 0.7815203666687012,
+      "learning_rate": 0.0002,
+      "loss": 1.1144,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0634396773010635,
+      "grad_norm": 0.3087436854839325,
+      "learning_rate": 0.0002,
+      "loss": 1.1316,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0707737440410707,
+      "grad_norm": 0.3847602903842926,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0781078107810782,
+      "grad_norm": 0.3693031370639801,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0854418775210855,
+      "grad_norm": 0.4111202359199524,
+      "learning_rate": 0.0002,
+      "loss": 1.0995,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0927759442610927,
+      "grad_norm": 0.41452381014823914,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1001100110011002,
+      "grad_norm": 0.3336445093154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1068,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1074440777411074,
+      "grad_norm": 0.3923407793045044,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1147781444811147,
+      "grad_norm": 0.46215683221817017,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1221122112211221,
+      "grad_norm": 0.3592156767845154,
+      "learning_rate": 0.0002,
+      "loss": 1.1133,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1294462779611294,
+      "grad_norm": 0.361110657453537,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1367803447011369,
+      "grad_norm": 0.5317131280899048,
+      "learning_rate": 0.0002,
+      "loss": 1.1553,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1441144114411441,
+      "grad_norm": 0.3882388174533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1514484781811514,
+      "grad_norm": 0.3259428143501282,
+      "learning_rate": 0.0002,
+      "loss": 1.0805,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1587825449211588,
+      "grad_norm": 0.410935640335083,
+      "learning_rate": 0.0002,
+      "loss": 1.1819,
+      "step": 1580
+    },
+    {
+      "epoch": 1.166116611661166,
+      "grad_norm": 0.44940185546875,
+      "learning_rate": 0.0002,
+      "loss": 1.1143,
+      "step": 1590
+    },
+    {
+      "epoch": 1.1734506784011733,
+      "grad_norm": 0.5106484293937683,
+      "learning_rate": 0.0002,
+      "loss": 1.0334,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1807847451411808,
+      "grad_norm": 0.6603665947914124,
+      "learning_rate": 0.0002,
+      "loss": 1.2376,
+      "step": 1610
+    },
+    {
+      "epoch": 1.188118811881188,
+      "grad_norm": 0.4799964129924774,
+      "learning_rate": 0.0002,
+      "loss": 1.1227,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1954528786211955,
+      "grad_norm": 0.4389883279800415,
+      "learning_rate": 0.0002,
+      "loss": 1.1191,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2027869453612028,
+      "grad_norm": 0.4188813269138336,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 1640
+    },
+    {
+      "epoch": 1.21012101210121,
+      "grad_norm": 0.7132157683372498,
+      "learning_rate": 0.0002,
+      "loss": 1.0605,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2174550788412175,
+      "grad_norm": 0.507480263710022,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2247891455812248,
+      "grad_norm": 0.9452332854270935,
+      "learning_rate": 0.0002,
+      "loss": 0.9948,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2321232123212322,
+      "grad_norm": 0.4121614992618561,
+      "learning_rate": 0.0002,
+      "loss": 1.0228,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2394572790612395,
+      "grad_norm": 0.34230247139930725,
+      "learning_rate": 0.0002,
+      "loss": 1.0366,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2467913458012467,
+      "grad_norm": 0.4026208817958832,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2541254125412542,
+      "grad_norm": 0.46673697233200073,
+      "learning_rate": 0.0002,
+      "loss": 1.0206,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2614594792812615,
+      "grad_norm": 0.38349825143814087,
+      "learning_rate": 0.0002,
+      "loss": 1.0827,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2687935460212687,
+      "grad_norm": 0.4049997627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.0356,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2761276127612762,
+      "grad_norm": 0.3417615294456482,
+      "learning_rate": 0.0002,
+      "loss": 0.9504,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2834616795012834,
+      "grad_norm": 0.4277614951133728,
+      "learning_rate": 0.0002,
+      "loss": 1.094,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2907957462412907,
+      "grad_norm": 0.5864202976226807,
+      "learning_rate": 0.0002,
+      "loss": 0.9938,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2981298129812981,
+      "grad_norm": 0.7097493410110474,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3054638797213054,
+      "grad_norm": 0.3145381212234497,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3127979464613129,
+      "grad_norm": 0.5116165280342102,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3201320132013201,
+      "grad_norm": 0.7469736337661743,
+      "learning_rate": 0.0002,
+      "loss": 1.0765,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3274660799413276,
+      "grad_norm": 0.32272255420684814,
+      "learning_rate": 0.0002,
+      "loss": 1.0663,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3348001466813348,
+      "grad_norm": 0.3534623086452484,
+      "learning_rate": 0.0002,
+      "loss": 0.9887,
+      "step": 1820
+    },
+    {
+      "epoch": 1.342134213421342,
+      "grad_norm": 0.36127907037734985,
+      "learning_rate": 0.0002,
+      "loss": 1.1628,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3494682801613496,
+      "grad_norm": 0.4072401523590088,
+      "learning_rate": 0.0002,
+      "loss": 1.0972,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3568023469013568,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.1267,
+      "step": 1850
+    },
+    {
+      "epoch": 1.364136413641364,
+      "grad_norm": 0.412883460521698,
+      "learning_rate": 0.0002,
+      "loss": 1.0173,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3714704803813715,
+      "grad_norm": 0.3735875189304352,
+      "learning_rate": 0.0002,
+      "loss": 1.0265,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3788045471213788,
+      "grad_norm": 0.39158159494400024,
+      "learning_rate": 0.0002,
+      "loss": 1.1061,
+      "step": 1880
+    },
+    {
+      "epoch": 1.386138613861386,
+      "grad_norm": 0.44431769847869873,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3934726806013935,
+      "grad_norm": 0.37772801518440247,
+      "learning_rate": 0.0002,
+      "loss": 1.0216,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4008067473414008,
+      "grad_norm": 0.4056641757488251,
+      "learning_rate": 0.0002,
+      "loss": 1.0674,
+      "step": 1910
+    },
+    {
+      "epoch": 1.408140814081408,
+      "grad_norm": 0.41612377762794495,
+      "learning_rate": 0.0002,
+      "loss": 1.0256,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4154748808214155,
+      "grad_norm": 0.41153013706207275,
+      "learning_rate": 0.0002,
+      "loss": 1.0467,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4228089475614227,
+      "grad_norm": 0.387845516204834,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4301430143014302,
+      "grad_norm": 0.3809587061405182,
+      "learning_rate": 0.0002,
+      "loss": 1.1094,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4374770810414375,
+      "grad_norm": 0.3625726103782654,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1960
+    },
+    {
+      "epoch": 1.444811147781445,
+      "grad_norm": 0.5294290781021118,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4521452145214522,
+      "grad_norm": 0.39975494146347046,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4594792812614594,
+      "grad_norm": 0.4181167185306549,
+      "learning_rate": 0.0002,
+      "loss": 0.9704,
+      "step": 1990
+    },
+    {
+      "epoch": 1.466813348001467,
+      "grad_norm": 0.42001503705978394,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4741474147414741,
+      "grad_norm": 0.4877578616142273,
+      "learning_rate": 0.0002,
+      "loss": 1.1266,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4814814814814814,
+      "grad_norm": 0.4050969183444977,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4888155482214889,
+      "grad_norm": 0.39068883657455444,
+      "learning_rate": 0.0002,
+      "loss": 1.0562,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4961496149614961,
+      "grad_norm": 0.421282559633255,
+      "learning_rate": 0.0002,
+      "loss": 1.0464,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5034836817015034,
+      "grad_norm": 0.47092297673225403,
+      "learning_rate": 0.0002,
+      "loss": 1.0532,
+      "step": 2050
+    },
+    {
+      "epoch": 1.5108177484415108,
+      "grad_norm": 0.39688974618911743,
+      "learning_rate": 0.0002,
+      "loss": 0.9348,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5181518151815183,
+      "grad_norm": 0.5529879331588745,
+      "learning_rate": 0.0002,
+      "loss": 1.08,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5254858819215253,
+      "grad_norm": 0.4879782199859619,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5328199486615328,
+      "grad_norm": 0.5517361164093018,
+      "learning_rate": 0.0002,
+      "loss": 1.0432,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5401540154015403,
+      "grad_norm": 0.44015637040138245,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5474880821415475,
+      "grad_norm": 0.5435167551040649,
+      "learning_rate": 0.0002,
+      "loss": 1.1873,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5548221488815548,
+      "grad_norm": 0.5714033246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1076,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5621562156215623,
+      "grad_norm": 0.31732529401779175,
+      "learning_rate": 0.0002,
+      "loss": 1.1107,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5694902823615695,
+      "grad_norm": 0.49068278074264526,
+      "learning_rate": 0.0002,
+      "loss": 1.0817,
+      "step": 2140
+    },
+    {
+      "epoch": 1.5768243491015768,
+      "grad_norm": 0.46851542592048645,
+      "learning_rate": 0.0002,
+      "loss": 1.0254,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5841584158415842,
+      "grad_norm": 0.5083092451095581,
+      "learning_rate": 0.0002,
+      "loss": 1.0623,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5914924825815915,
+      "grad_norm": 0.9822936058044434,
+      "learning_rate": 0.0002,
+      "loss": 1.0603,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5988265493215987,
+      "grad_norm": 0.4575989246368408,
+      "learning_rate": 0.0002,
+      "loss": 0.9986,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6061606160616062,
+      "grad_norm": 0.47444286942481995,
+      "learning_rate": 0.0002,
+      "loss": 1.1292,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6134946828016135,
+      "grad_norm": 0.7208226919174194,
+      "learning_rate": 0.0002,
+      "loss": 1.0136,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6208287495416207,
+      "grad_norm": 0.43791481852531433,
+      "learning_rate": 0.0002,
+      "loss": 1.15,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6281628162816282,
+      "grad_norm": 0.5245792865753174,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6354968830216357,
+      "grad_norm": 0.39289429783821106,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6428309497616427,
+      "grad_norm": 0.6106135845184326,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6501650165016502,
+      "grad_norm": 0.3722580671310425,
+      "learning_rate": 0.0002,
+      "loss": 1.0129,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6574990832416576,
+      "grad_norm": 0.3649403750896454,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6648331499816649,
+      "grad_norm": 0.46514248847961426,
+      "learning_rate": 0.0002,
+      "loss": 1.0037,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6721672167216721,
+      "grad_norm": 0.42034927010536194,
+      "learning_rate": 0.0002,
+      "loss": 1.0022,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6795012834616796,
+      "grad_norm": 0.45202910900115967,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6868353502016868,
+      "grad_norm": 0.36257603764533997,
+      "learning_rate": 0.0002,
+      "loss": 1.0866,
+      "step": 2300
+    },
+    {
+      "epoch": 1.694169416941694,
+      "grad_norm": 0.6340323090553284,
+      "learning_rate": 0.0002,
+      "loss": 1.0973,
+      "step": 2310
+    },
+    {
+      "epoch": 1.7015034836817016,
+      "grad_norm": 0.4352878928184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0615,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7088375504217088,
+      "grad_norm": 0.45029792189598083,
+      "learning_rate": 0.0002,
+      "loss": 1.0629,
+      "step": 2330
+    },
+    {
+      "epoch": 1.716171617161716,
+      "grad_norm": 0.3891315758228302,
+      "learning_rate": 0.0002,
+      "loss": 0.9621,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7235056839017235,
+      "grad_norm": 0.35180050134658813,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7308397506417308,
+      "grad_norm": 0.42367449402809143,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 2360
+    },
+    {
+      "epoch": 1.738173817381738,
+      "grad_norm": 0.4553675353527069,
+      "learning_rate": 0.0002,
+      "loss": 1.0376,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7455078841217455,
+      "grad_norm": 0.5944654941558838,
+      "learning_rate": 0.0002,
+      "loss": 1.1467,
+      "step": 2380
+    },
+    {
+      "epoch": 1.752841950861753,
+      "grad_norm": 0.3479664623737335,
+      "learning_rate": 0.0002,
+      "loss": 1.0548,
+      "step": 2390
+    },
+    {
+      "epoch": 1.76017601760176,
+      "grad_norm": 0.3585502505302429,
+      "learning_rate": 0.0002,
+      "loss": 1.0798,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7675100843417675,
+      "grad_norm": 0.4263346493244171,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 2410
+    },
+    {
+      "epoch": 1.774844151081775,
+      "grad_norm": 0.5476409196853638,
+      "learning_rate": 0.0002,
+      "loss": 1.054,
+      "step": 2420
+    },
+    {
+      "epoch": 1.7821782178217822,
+      "grad_norm": 0.3694186508655548,
+      "learning_rate": 0.0002,
+      "loss": 1.1615,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7895122845617895,
+      "grad_norm": 0.9185658693313599,
+      "learning_rate": 0.0002,
+      "loss": 1.1343,
+      "step": 2440
+    },
+    {
+      "epoch": 1.796846351301797,
+      "grad_norm": 0.7171908020973206,
+      "learning_rate": 0.0002,
+      "loss": 1.0764,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8041804180418042,
+      "grad_norm": 0.550658643245697,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 2460
+    },
+    {
+      "epoch": 1.8115144847818114,
+      "grad_norm": 0.4075568914413452,
+      "learning_rate": 0.0002,
+      "loss": 0.9975,
+      "step": 2470
+    },
+    {
+      "epoch": 1.818848551521819,
+      "grad_norm": 0.3790127635002136,
+      "learning_rate": 0.0002,
+      "loss": 1.0935,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8261826182618262,
+      "grad_norm": 0.3576384484767914,
+      "learning_rate": 0.0002,
+      "loss": 0.9839,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8335166850018334,
+      "grad_norm": 0.3919370770454407,
+      "learning_rate": 0.0002,
+      "loss": 1.1369,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8408507517418409,
+      "grad_norm": 0.485083669424057,
+      "learning_rate": 0.0002,
+      "loss": 0.9985,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8481848184818483,
+      "grad_norm": 0.4564347565174103,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8555188852218554,
+      "grad_norm": 0.3613106608390808,
+      "learning_rate": 0.0002,
+      "loss": 1.0944,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8628529519618628,
+      "grad_norm": 0.39600759744644165,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8701870187018703,
+      "grad_norm": 1.123499870300293,
+      "learning_rate": 0.0002,
+      "loss": 0.9453,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8775210854418776,
+      "grad_norm": 0.4612680673599243,
+      "learning_rate": 0.0002,
+      "loss": 1.0635,
+      "step": 2560
+    },
+    {
+      "epoch": 1.8848551521818848,
+      "grad_norm": 0.42745399475097656,
+      "learning_rate": 0.0002,
+      "loss": 1.0087,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8921892189218923,
+      "grad_norm": 0.4055580198764801,
+      "learning_rate": 0.0002,
+      "loss": 1.0102,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8995232856618995,
+      "grad_norm": 0.44174644351005554,
+      "learning_rate": 0.0002,
+      "loss": 1.0177,
+      "step": 2590
+    },
+    {
+      "epoch": 1.9068573524019068,
+      "grad_norm": 1.0228385925292969,
+      "learning_rate": 0.0002,
+      "loss": 0.9886,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9141914191419143,
+      "grad_norm": 0.3496396243572235,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9215254858819215,
+      "grad_norm": 0.4191173017024994,
+      "learning_rate": 0.0002,
+      "loss": 1.0955,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9288595526219288,
+      "grad_norm": 0.6778554916381836,
+      "learning_rate": 0.0002,
+      "loss": 1.0943,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9361936193619362,
+      "grad_norm": 0.41992834210395813,
+      "learning_rate": 0.0002,
+      "loss": 1.0594,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9435276861019435,
+      "grad_norm": 0.8760401010513306,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9508617528419507,
+      "grad_norm": 0.44049209356307983,
+      "learning_rate": 0.0002,
+      "loss": 1.0379,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9581958195819582,
+      "grad_norm": 0.5651928782463074,
+      "learning_rate": 0.0002,
+      "loss": 1.1008,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9655298863219657,
+      "grad_norm": 0.5292727947235107,
+      "learning_rate": 0.0002,
+      "loss": 1.1317,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9728639530619727,
+      "grad_norm": 0.6012240648269653,
+      "learning_rate": 0.0002,
+      "loss": 1.1328,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9801980198019802,
+      "grad_norm": 0.3945149779319763,
+      "learning_rate": 0.0002,
+      "loss": 1.0683,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9875320865419877,
+      "grad_norm": 0.5732627511024475,
+      "learning_rate": 0.0002,
+      "loss": 1.0155,
+      "step": 2710
+    },
+    {
+      "epoch": 1.994866153281995,
+      "grad_norm": 0.3963361084461212,
+      "learning_rate": 0.0002,
+      "loss": 0.9857,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1534006595611572,
+      "eval_runtime": 32.7541,
+      "eval_samples_per_second": 13.159,
+      "eval_steps_per_second": 1.649,
+      "step": 2727
+    },
+    {
+      "epoch": 2.002200220022002,
+      "grad_norm": 0.48628315329551697,
+      "learning_rate": 0.0002,
+      "loss": 0.9624,
+      "step": 2730
+    },
+    {
+      "epoch": 2.0095342867620096,
+      "grad_norm": 0.413875013589859,
+      "learning_rate": 0.0002,
+      "loss": 0.9603,
+      "step": 2740
+    },
+    {
+      "epoch": 2.0168683535020167,
+      "grad_norm": 0.4988735616207123,
+      "learning_rate": 0.0002,
+      "loss": 0.965,
+      "step": 2750
+    },
+    {
+      "epoch": 2.024202420242024,
+      "grad_norm": 0.5634812712669373,
+      "learning_rate": 0.0002,
+      "loss": 0.9677,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0315364869820316,
+      "grad_norm": 0.48302653431892395,
+      "learning_rate": 0.0002,
+      "loss": 0.9547,
+      "step": 2770
+    },
+    {
+      "epoch": 2.038870553722039,
+      "grad_norm": 0.49914175271987915,
+      "learning_rate": 0.0002,
+      "loss": 0.9346,
+      "step": 2780
+    },
+    {
+      "epoch": 2.046204620462046,
+      "grad_norm": 1.14039945602417,
+      "learning_rate": 0.0002,
+      "loss": 0.904,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0535386872020536,
+      "grad_norm": 0.6359720826148987,
+      "learning_rate": 0.0002,
+      "loss": 0.9588,
+      "step": 2800
+    },
+    {
+      "epoch": 2.060872753942061,
+      "grad_norm": 0.4589158296585083,
+      "learning_rate": 0.0002,
+      "loss": 0.9031,
+      "step": 2810
+    },
+    {
+      "epoch": 2.068206820682068,
+      "grad_norm": 0.46255481243133545,
+      "learning_rate": 0.0002,
+      "loss": 0.9438,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0755408874220755,
+      "grad_norm": 0.6232137680053711,
+      "learning_rate": 0.0002,
+      "loss": 0.9464,
+      "step": 2830
+    },
+    {
+      "epoch": 2.082874954162083,
+      "grad_norm": 0.41042178869247437,
+      "learning_rate": 0.0002,
+      "loss": 0.8978,
+      "step": 2840
+    },
+    {
+      "epoch": 2.09020902090209,
+      "grad_norm": 0.5334428548812866,
+      "learning_rate": 0.0002,
+      "loss": 0.8516,
+      "step": 2850
+    },
+    {
+      "epoch": 2.0975430876420975,
+      "grad_norm": 0.8270058631896973,
+      "learning_rate": 0.0002,
+      "loss": 0.9313,
+      "step": 2860
+    },
+    {
+      "epoch": 2.104877154382105,
+      "grad_norm": 0.6624533534049988,
+      "learning_rate": 0.0002,
+      "loss": 1.0064,
+      "step": 2870
+    },
+    {
+      "epoch": 2.112211221122112,
+      "grad_norm": 0.5448863506317139,
+      "learning_rate": 0.0002,
+      "loss": 0.9196,
+      "step": 2880
+    },
+    {
+      "epoch": 2.1195452878621195,
+      "grad_norm": 0.621482789516449,
+      "learning_rate": 0.0002,
+      "loss": 0.887,
+      "step": 2890
+    },
+    {
+      "epoch": 2.126879354602127,
+      "grad_norm": 0.4556255340576172,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1342134213421344,
+      "grad_norm": 0.4620579183101654,
+      "learning_rate": 0.0002,
+      "loss": 0.9323,
+      "step": 2910
+    },
+    {
+      "epoch": 2.1415474880821415,
+      "grad_norm": 0.9602415561676025,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 2920
+    },
+    {
+      "epoch": 2.148881554822149,
+      "grad_norm": 0.587943971157074,
+      "learning_rate": 0.0002,
+      "loss": 0.8826,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1562156215621564,
+      "grad_norm": 0.5121372938156128,
+      "learning_rate": 0.0002,
+      "loss": 0.971,
+      "step": 2940
+    },
+    {
+      "epoch": 2.1635496883021634,
+      "grad_norm": 0.49424484372138977,
+      "learning_rate": 0.0002,
+      "loss": 0.8751,
+      "step": 2950
+    },
+    {
+      "epoch": 2.170883755042171,
+      "grad_norm": 0.6312560439109802,
+      "learning_rate": 0.0002,
+      "loss": 0.8674,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1782178217821784,
+      "grad_norm": 0.5235576629638672,
+      "learning_rate": 0.0002,
+      "loss": 0.9791,
+      "step": 2970
+    },
+    {
+      "epoch": 2.1855518885221854,
+      "grad_norm": 0.5868439674377441,
+      "learning_rate": 0.0002,
+      "loss": 0.9706,
+      "step": 2980
+    },
+    {
+      "epoch": 2.192885955262193,
+      "grad_norm": 0.42302873730659485,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 2990
+    },
+    {
+      "epoch": 2.2002200220022003,
+      "grad_norm": 0.5097725987434387,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 3000
+    },
+    {
+      "epoch": 2.2075540887422074,
+      "grad_norm": 0.5091572403907776,
+      "learning_rate": 0.0002,
+      "loss": 0.9239,
+      "step": 3010
+    },
+    {
+      "epoch": 2.214888155482215,
+      "grad_norm": 0.49433162808418274,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 3020
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.5577368140220642,
+      "learning_rate": 0.0002,
+      "loss": 0.9734,
+      "step": 3030
+    },
+    {
+      "epoch": 2.2295562889622293,
+      "grad_norm": 0.6177583932876587,
+      "learning_rate": 0.0002,
+      "loss": 0.9033,
+      "step": 3040
+    },
+    {
+      "epoch": 2.236890355702237,
+      "grad_norm": 0.5256719589233398,
+      "learning_rate": 0.0002,
+      "loss": 0.9882,
+      "step": 3050
+    },
+    {
+      "epoch": 2.2442244224422443,
+      "grad_norm": 0.5001118183135986,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2515584891822513,
+      "grad_norm": 0.5721249580383301,
+      "learning_rate": 0.0002,
+      "loss": 0.8718,
+      "step": 3070
+    },
+    {
+      "epoch": 2.258892555922259,
+      "grad_norm": 0.5325384140014648,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 3080
+    },
+    {
+      "epoch": 2.2662266226622663,
+      "grad_norm": 0.5719189047813416,
+      "learning_rate": 0.0002,
+      "loss": 0.9843,
+      "step": 3090
+    },
+    {
+      "epoch": 2.2735606894022737,
+      "grad_norm": 0.6337835788726807,
+      "learning_rate": 0.0002,
+      "loss": 0.8633,
+      "step": 3100
+    },
+    {
+      "epoch": 2.2808947561422808,
+      "grad_norm": 0.5381836891174316,
+      "learning_rate": 0.0002,
+      "loss": 0.9962,
+      "step": 3110
+    },
+    {
+      "epoch": 2.2882288228822882,
+      "grad_norm": 0.5408531427383423,
+      "learning_rate": 0.0002,
+      "loss": 0.8265,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2955628896222957,
+      "grad_norm": 0.43705281615257263,
+      "learning_rate": 0.0002,
+      "loss": 1.0325,
+      "step": 3130
+    },
+    {
+      "epoch": 2.3028969563623027,
+      "grad_norm": 0.6454030275344849,
+      "learning_rate": 0.0002,
+      "loss": 0.9388,
+      "step": 3140
+    },
+    {
+      "epoch": 2.31023102310231,
+      "grad_norm": 0.686030387878418,
+      "learning_rate": 0.0002,
+      "loss": 0.954,
+      "step": 3150
+    },
+    {
+      "epoch": 2.3175650898423177,
+      "grad_norm": 0.5123633146286011,
+      "learning_rate": 0.0002,
+      "loss": 0.9403,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3248991565823247,
+      "grad_norm": 0.842506468296051,
+      "learning_rate": 0.0002,
+      "loss": 0.8834,
+      "step": 3170
+    },
+    {
+      "epoch": 2.332233223322332,
+      "grad_norm": 0.5193818807601929,
+      "learning_rate": 0.0002,
+      "loss": 1.0497,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3395672900623397,
+      "grad_norm": 0.5634409189224243,
+      "learning_rate": 0.0002,
+      "loss": 0.9473,
+      "step": 3190
+    },
+    {
+      "epoch": 2.3469013568023467,
+      "grad_norm": 0.6475534439086914,
+      "learning_rate": 0.0002,
+      "loss": 0.8499,
+      "step": 3200
+    },
+    {
+      "epoch": 2.354235423542354,
+      "grad_norm": 1.1503914594650269,
+      "learning_rate": 0.0002,
+      "loss": 0.874,
+      "step": 3210
+    },
+    {
+      "epoch": 2.3615694902823616,
+      "grad_norm": 0.7234905362129211,
+      "learning_rate": 0.0002,
+      "loss": 0.9762,
+      "step": 3220
+    },
+    {
+      "epoch": 2.368903557022369,
+      "grad_norm": 0.664903461933136,
+      "learning_rate": 0.0002,
+      "loss": 0.9007,
+      "step": 3230
+    },
+    {
+      "epoch": 2.376237623762376,
+      "grad_norm": 0.5453006625175476,
+      "learning_rate": 0.0002,
+      "loss": 0.9987,
+      "step": 3240
+    },
+    {
+      "epoch": 2.3835716905023836,
+      "grad_norm": 0.6256654262542725,
+      "learning_rate": 0.0002,
+      "loss": 0.9742,
+      "step": 3250
+    },
+    {
+      "epoch": 2.390905757242391,
+      "grad_norm": 0.5166565179824829,
+      "learning_rate": 0.0002,
+      "loss": 0.9922,
+      "step": 3260
+    },
+    {
+      "epoch": 2.398239823982398,
+      "grad_norm": 0.5699098110198975,
+      "learning_rate": 0.0002,
+      "loss": 0.927,
+      "step": 3270
+    },
+    {
+      "epoch": 2.4055738907224056,
+      "grad_norm": 0.4472540020942688,
+      "learning_rate": 0.0002,
+      "loss": 0.8878,
+      "step": 3280
+    },
+    {
+      "epoch": 2.412907957462413,
+      "grad_norm": 0.6790403127670288,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 3290
+    },
+    {
+      "epoch": 2.42024202420242,
+      "grad_norm": 0.5182185173034668,
+      "learning_rate": 0.0002,
+      "loss": 0.972,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4275760909424275,
+      "grad_norm": 0.564647912979126,
+      "learning_rate": 0.0002,
+      "loss": 0.9775,
+      "step": 3310
+    },
+    {
+      "epoch": 2.434910157682435,
+      "grad_norm": 0.5625313520431519,
+      "learning_rate": 0.0002,
+      "loss": 1.072,
+      "step": 3320
+    },
+    {
+      "epoch": 2.442244224422442,
+      "grad_norm": 0.7496559619903564,
+      "learning_rate": 0.0002,
+      "loss": 0.8798,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4495782911624495,
+      "grad_norm": 0.4779128134250641,
+      "learning_rate": 0.0002,
+      "loss": 0.868,
+      "step": 3340
+    },
+    {
+      "epoch": 2.456912357902457,
+      "grad_norm": 0.578093409538269,
+      "learning_rate": 0.0002,
+      "loss": 1.0316,
+      "step": 3350
+    },
+    {
+      "epoch": 2.4642464246424645,
+      "grad_norm": 0.5456080436706543,
+      "learning_rate": 0.0002,
+      "loss": 0.9282,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4715804913824715,
+      "grad_norm": 0.4769273102283478,
+      "learning_rate": 0.0002,
+      "loss": 0.8409,
+      "step": 3370
+    },
+    {
+      "epoch": 2.478914558122479,
+      "grad_norm": 0.5608189702033997,
+      "learning_rate": 0.0002,
+      "loss": 0.9312,
+      "step": 3380
+    },
+    {
+      "epoch": 2.4862486248624864,
+      "grad_norm": 0.5590165853500366,
+      "learning_rate": 0.0002,
+      "loss": 0.9934,
+      "step": 3390
+    },
+    {
+      "epoch": 2.4935826916024935,
+      "grad_norm": 0.801306962966919,
+      "learning_rate": 0.0002,
+      "loss": 1.025,
+      "step": 3400
+    },
+    {
+      "epoch": 2.500916758342501,
+      "grad_norm": 0.6045624613761902,
+      "learning_rate": 0.0002,
+      "loss": 0.9049,
+      "step": 3410
+    },
+    {
+      "epoch": 2.5082508250825084,
+      "grad_norm": 0.5735858082771301,
+      "learning_rate": 0.0002,
+      "loss": 0.944,
+      "step": 3420
+    },
+    {
+      "epoch": 2.5155848918225154,
+      "grad_norm": 0.6827309131622314,
+      "learning_rate": 0.0002,
+      "loss": 0.9846,
+      "step": 3430
+    },
+    {
+      "epoch": 2.522918958562523,
+      "grad_norm": 0.5702602863311768,
+      "learning_rate": 0.0002,
+      "loss": 0.9789,
+      "step": 3440
+    },
+    {
+      "epoch": 2.5302530253025304,
+      "grad_norm": 0.6674721240997314,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5375870920425374,
+      "grad_norm": 0.5635907649993896,
+      "learning_rate": 0.0002,
+      "loss": 0.914,
+      "step": 3460
+    },
+    {
+      "epoch": 2.544921158782545,
+      "grad_norm": 0.42737770080566406,
+      "learning_rate": 0.0002,
+      "loss": 0.8398,
+      "step": 3470
+    },
+    {
+      "epoch": 2.5522552255225524,
+      "grad_norm": 0.6720691919326782,
+      "learning_rate": 0.0002,
+      "loss": 0.9474,
+      "step": 3480
+    },
+    {
+      "epoch": 2.55958929226256,
+      "grad_norm": 0.8917084336280823,
+      "learning_rate": 0.0002,
+      "loss": 0.8637,
+      "step": 3490
+    },
+    {
+      "epoch": 2.566923359002567,
+      "grad_norm": 0.5134549140930176,
+      "learning_rate": 0.0002,
+      "loss": 0.9257,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5742574257425743,
+      "grad_norm": 0.4951367974281311,
+      "learning_rate": 0.0002,
+      "loss": 0.9362,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5815914924825814,
+      "grad_norm": 0.9438204765319824,
+      "learning_rate": 0.0002,
+      "loss": 0.9184,
+      "step": 3520
+    },
+    {
+      "epoch": 2.588925559222589,
+      "grad_norm": 0.6024714708328247,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5962596259625963,
+      "grad_norm": 0.5248535871505737,
+      "learning_rate": 0.0002,
+      "loss": 0.9298,
+      "step": 3540
+    },
+    {
+      "epoch": 2.6035936927026038,
+      "grad_norm": 0.8677568435668945,
+      "learning_rate": 0.0002,
+      "loss": 0.941,
+      "step": 3550
+    },
+    {
+      "epoch": 2.610927759442611,
+      "grad_norm": 0.82008296251297,
+      "learning_rate": 0.0002,
+      "loss": 0.9253,
+      "step": 3560
+    },
+    {
+      "epoch": 2.6182618261826183,
+      "grad_norm": 0.4724634885787964,
+      "learning_rate": 0.0002,
+      "loss": 0.8429,
+      "step": 3570
+    },
+    {
+      "epoch": 2.6255958929226257,
+      "grad_norm": 0.5434244275093079,
+      "learning_rate": 0.0002,
+      "loss": 0.9058,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6329299596626328,
+      "grad_norm": 0.4948740005493164,
+      "learning_rate": 0.0002,
+      "loss": 0.9379,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6402640264026402,
+      "grad_norm": 0.42109328508377075,
+      "learning_rate": 0.0002,
+      "loss": 0.8718,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6475980931426477,
+      "grad_norm": 0.7979786396026611,
+      "learning_rate": 0.0002,
+      "loss": 0.9809,
+      "step": 3610
+    },
+    {
+      "epoch": 2.654932159882655,
+      "grad_norm": 0.6345919370651245,
+      "learning_rate": 0.0002,
+      "loss": 0.9229,
+      "step": 3620
+    },
+    {
+      "epoch": 2.662266226622662,
+      "grad_norm": 0.4971671402454376,
+      "learning_rate": 0.0002,
+      "loss": 0.8506,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6696002933626697,
+      "grad_norm": 0.6467748284339905,
+      "learning_rate": 0.0002,
+      "loss": 0.8054,
+      "step": 3640
+    },
+    {
+      "epoch": 2.6769343601026767,
+      "grad_norm": 0.4240160286426544,
+      "learning_rate": 0.0002,
+      "loss": 0.9277,
+      "step": 3650
+    },
+    {
+      "epoch": 2.684268426842684,
+      "grad_norm": 0.5179754495620728,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 3660
+    },
+    {
+      "epoch": 2.6916024935826917,
+      "grad_norm": 0.754012405872345,
+      "learning_rate": 0.0002,
+      "loss": 0.9221,
+      "step": 3670
+    },
+    {
+      "epoch": 2.698936560322699,
+      "grad_norm": 0.5141299962997437,
+      "learning_rate": 0.0002,
+      "loss": 0.9194,
+      "step": 3680
+    },
+    {
+      "epoch": 2.706270627062706,
+      "grad_norm": 0.5737819075584412,
+      "learning_rate": 0.0002,
+      "loss": 0.9495,
+      "step": 3690
+    },
+    {
+      "epoch": 2.7136046938027136,
+      "grad_norm": 0.5887577533721924,
+      "learning_rate": 0.0002,
+      "loss": 1.0162,
+      "step": 3700
+    },
+    {
+      "epoch": 2.720938760542721,
+      "grad_norm": 0.6740471720695496,
+      "learning_rate": 0.0002,
+      "loss": 0.9169,
+      "step": 3710
+    },
+    {
+      "epoch": 2.728272827282728,
+      "grad_norm": 0.5879453420639038,
+      "learning_rate": 0.0002,
+      "loss": 0.9297,
+      "step": 3720
+    },
+    {
+      "epoch": 2.7356068940227356,
+      "grad_norm": 0.4858354926109314,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 3730
+    },
+    {
+      "epoch": 2.742940960762743,
+      "grad_norm": 0.5489001870155334,
+      "learning_rate": 0.0002,
+      "loss": 0.9308,
+      "step": 3740
+    },
+    {
+      "epoch": 2.7502750275027505,
+      "grad_norm": 0.8187092542648315,
+      "learning_rate": 0.0002,
+      "loss": 0.894,
+      "step": 3750
+    },
+    {
+      "epoch": 2.7576090942427576,
+      "grad_norm": 0.5666626691818237,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 3760
+    },
+    {
+      "epoch": 2.764943160982765,
+      "grad_norm": 0.5377066135406494,
+      "learning_rate": 0.0002,
+      "loss": 1.0059,
+      "step": 3770
+    },
+    {
+      "epoch": 2.772277227722772,
+      "grad_norm": 0.566330075263977,
+      "learning_rate": 0.0002,
+      "loss": 0.9132,
+      "step": 3780
+    },
+    {
+      "epoch": 2.7796112944627795,
+      "grad_norm": 0.5522832870483398,
+      "learning_rate": 0.0002,
+      "loss": 0.9415,
+      "step": 3790
+    },
+    {
+      "epoch": 2.786945361202787,
+      "grad_norm": 0.5668695569038391,
+      "learning_rate": 0.0002,
+      "loss": 0.8816,
+      "step": 3800
+    },
+    {
+      "epoch": 2.7942794279427945,
+      "grad_norm": 0.7566602826118469,
+      "learning_rate": 0.0002,
+      "loss": 0.8885,
+      "step": 3810
+    },
+    {
+      "epoch": 2.8016134946828015,
+      "grad_norm": 0.5603684782981873,
+      "learning_rate": 0.0002,
+      "loss": 0.8598,
+      "step": 3820
+    },
+    {
+      "epoch": 2.808947561422809,
+      "grad_norm": 0.49122217297554016,
+      "learning_rate": 0.0002,
+      "loss": 0.9602,
+      "step": 3830
+    },
+    {
+      "epoch": 2.816281628162816,
+      "grad_norm": 0.6798251867294312,
+      "learning_rate": 0.0002,
+      "loss": 0.9738,
+      "step": 3840
+    },
+    {
+      "epoch": 2.8236156949028235,
+      "grad_norm": 0.6097991466522217,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 3850
+    },
+    {
+      "epoch": 2.830949761642831,
+      "grad_norm": 0.6675726175308228,
+      "learning_rate": 0.0002,
+      "loss": 0.8672,
+      "step": 3860
+    },
+    {
+      "epoch": 2.8382838283828384,
+      "grad_norm": 0.9223952889442444,
+      "learning_rate": 0.0002,
+      "loss": 0.9324,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8456178951228455,
+      "grad_norm": 0.6020799875259399,
+      "learning_rate": 0.0002,
+      "loss": 0.8767,
+      "step": 3880
+    },
+    {
+      "epoch": 2.852951961862853,
+      "grad_norm": 0.5206381678581238,
+      "learning_rate": 0.0002,
+      "loss": 0.9148,
+      "step": 3890
+    },
+    {
+      "epoch": 2.8602860286028604,
+      "grad_norm": 0.6268777251243591,
+      "learning_rate": 0.0002,
+      "loss": 0.9479,
+      "step": 3900
+    },
+    {
+      "epoch": 2.8676200953428674,
+      "grad_norm": 1.1583497524261475,
+      "learning_rate": 0.0002,
+      "loss": 0.9409,
+      "step": 3910
+    },
+    {
+      "epoch": 2.874954162082875,
+      "grad_norm": 0.7263903021812439,
+      "learning_rate": 0.0002,
+      "loss": 0.895,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8822882288228824,
+      "grad_norm": 0.5369910001754761,
+      "learning_rate": 0.0002,
+      "loss": 0.8786,
+      "step": 3930
+    },
+    {
+      "epoch": 2.88962229556289,
+      "grad_norm": 0.7298350930213928,
+      "learning_rate": 0.0002,
+      "loss": 1.0015,
+      "step": 3940
+    },
+    {
+      "epoch": 2.896956362302897,
+      "grad_norm": 0.577012836933136,
+      "learning_rate": 0.0002,
+      "loss": 0.979,
+      "step": 3950
+    },
+    {
+      "epoch": 2.9042904290429044,
+      "grad_norm": 0.5859594345092773,
+      "learning_rate": 0.0002,
+      "loss": 0.9716,
+      "step": 3960
+    },
+    {
+      "epoch": 2.9116244957829114,
+      "grad_norm": 0.47176122665405273,
+      "learning_rate": 0.0002,
+      "loss": 0.8772,
+      "step": 3970
+    },
+    {
+      "epoch": 2.918958562522919,
+      "grad_norm": 0.9699620604515076,
+      "learning_rate": 0.0002,
+      "loss": 0.8997,
+      "step": 3980
+    },
+    {
+      "epoch": 2.9262926292629263,
+      "grad_norm": 0.7908747792243958,
+      "learning_rate": 0.0002,
+      "loss": 0.9057,
+      "step": 3990
+    },
+    {
+      "epoch": 2.933626696002934,
+      "grad_norm": 0.5777379274368286,
+      "learning_rate": 0.0002,
+      "loss": 0.9462,
+      "step": 4000
+    },
+    {
+      "epoch": 2.940960762742941,
+      "grad_norm": 0.599288284778595,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9482948294829483,
+      "grad_norm": 0.5232274532318115,
+      "learning_rate": 0.0002,
+      "loss": 0.9812,
+      "step": 4020
+    },
+    {
+      "epoch": 2.9556288962229558,
+      "grad_norm": 0.6395137310028076,
+      "learning_rate": 0.0002,
+      "loss": 0.96,
+      "step": 4030
+    },
+    {
+      "epoch": 2.962962962962963,
+      "grad_norm": 0.589260458946228,
+      "learning_rate": 0.0002,
+      "loss": 0.9813,
+      "step": 4040
+    },
+    {
+      "epoch": 2.9702970297029703,
+      "grad_norm": 0.5699581503868103,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 4050
+    },
+    {
+      "epoch": 2.9776310964429777,
+      "grad_norm": 0.528468132019043,
+      "learning_rate": 0.0002,
+      "loss": 0.9585,
+      "step": 4060
+    },
+    {
+      "epoch": 2.984965163182985,
+      "grad_norm": 0.4804670512676239,
+      "learning_rate": 0.0002,
+      "loss": 0.9164,
+      "step": 4070
+    },
+    {
+      "epoch": 2.9922992299229922,
+      "grad_norm": 1.1918889284133911,
+      "learning_rate": 0.0002,
+      "loss": 0.9771,
+      "step": 4080
+    },
+    {
+      "epoch": 2.9996332966629997,
+      "grad_norm": 0.5479103326797485,
+      "learning_rate": 0.0002,
+      "loss": 0.9178,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9996332966629997,
+      "eval_loss": 1.1642853021621704,
+      "eval_runtime": 32.7511,
+      "eval_samples_per_second": 13.16,
+      "eval_steps_per_second": 1.649,
+      "step": 4090
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10904,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "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": 1.8929918099954074e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..2868cff7027115396e695775cacd838522aca295
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-4090/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b12b6f6817632087b5a5e37d744e25312b96e839de5005320b96bc0c2473c41f
+size 5624
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/README.md b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# 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 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "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": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e2f27193d2800afa1ecf38d32ca029cca4c48153
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:783240254d4d99f2734879322ca67b130ada972da1875179a88b93095caa5027
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b7a7b012ba82f504ba58834e0666e367cecb43d3
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aac35d75c36594bce550fceee036251df5ff665be30b898e7461e6cf74d4738b
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6f30acc8ccb3de502804392f9179ca54bf8fbaf6
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3267b86fe975e9097bd132b8cbf8c3b6f2dc70ad372771b130a4b0a82c1b248d
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..85e5a20c021880a6fb5c9efb5621de4c963b391f
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73e6b8312521b90d16ac6a6a64f9ca779b0501086d106cb4f9b043c7499d67c8
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..c251e2b2b002f82fe3e94a1e003a0f7cabe5e71d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/trainer_state.json
@@ -0,0 +1,3880 @@
+{
+  "best_metric": 1.1534006595611572,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 5454,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007334066740007334,
+      "grad_norm": 0.47521963715553284,
+      "learning_rate": 0.0002,
+      "loss": 1.9722,
+      "step": 10
+    },
+    {
+      "epoch": 0.014668133480014669,
+      "grad_norm": 0.5395162105560303,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 20
+    },
+    {
+      "epoch": 0.022002200220022004,
+      "grad_norm": 0.4305780231952667,
+      "learning_rate": 0.0002,
+      "loss": 1.4202,
+      "step": 30
+    },
+    {
+      "epoch": 0.029336266960029337,
+      "grad_norm": 0.6938246488571167,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 40
+    },
+    {
+      "epoch": 0.03667033370003667,
+      "grad_norm": 1.5133819580078125,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 50
+    },
+    {
+      "epoch": 0.04400440044004401,
+      "grad_norm": 0.9173883199691772,
+      "learning_rate": 0.0002,
+      "loss": 1.3132,
+      "step": 60
+    },
+    {
+      "epoch": 0.05133846718005134,
+      "grad_norm": 0.4619861841201782,
+      "learning_rate": 0.0002,
+      "loss": 1.2844,
+      "step": 70
+    },
+    {
+      "epoch": 0.058672533920058674,
+      "grad_norm": 0.46118637919425964,
+      "learning_rate": 0.0002,
+      "loss": 1.2108,
+      "step": 80
+    },
+    {
+      "epoch": 0.066006600660066,
+      "grad_norm": 0.4468648135662079,
+      "learning_rate": 0.0002,
+      "loss": 1.3441,
+      "step": 90
+    },
+    {
+      "epoch": 0.07334066740007333,
+      "grad_norm": 0.46123769879341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1863,
+      "step": 100
+    },
+    {
+      "epoch": 0.08067473414008068,
+      "grad_norm": 0.4859139025211334,
+      "learning_rate": 0.0002,
+      "loss": 1.2772,
+      "step": 110
+    },
+    {
+      "epoch": 0.08800880088008801,
+      "grad_norm": 0.4384922385215759,
+      "learning_rate": 0.0002,
+      "loss": 1.2087,
+      "step": 120
+    },
+    {
+      "epoch": 0.09534286762009535,
+      "grad_norm": 0.39519360661506653,
+      "learning_rate": 0.0002,
+      "loss": 1.2927,
+      "step": 130
+    },
+    {
+      "epoch": 0.10267693436010268,
+      "grad_norm": 0.4049859344959259,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 140
+    },
+    {
+      "epoch": 0.11001100110011001,
+      "grad_norm": 0.4605638086795807,
+      "learning_rate": 0.0002,
+      "loss": 1.293,
+      "step": 150
+    },
+    {
+      "epoch": 0.11734506784011735,
+      "grad_norm": 0.4201928377151489,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 160
+    },
+    {
+      "epoch": 0.12467913458012468,
+      "grad_norm": 0.5367777347564697,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 170
+    },
+    {
+      "epoch": 0.132013201320132,
+      "grad_norm": 0.41752299666404724,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 180
+    },
+    {
+      "epoch": 0.13934726806013933,
+      "grad_norm": 0.31597763299942017,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 190
+    },
+    {
+      "epoch": 0.14668133480014667,
+      "grad_norm": 0.7468788623809814,
+      "learning_rate": 0.0002,
+      "loss": 1.2441,
+      "step": 200
+    },
+    {
+      "epoch": 0.15401540154015403,
+      "grad_norm": 0.3403034508228302,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 210
+    },
+    {
+      "epoch": 0.16134946828016136,
+      "grad_norm": 0.34240293502807617,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 220
+    },
+    {
+      "epoch": 0.1686835350201687,
+      "grad_norm": 0.356158971786499,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 230
+    },
+    {
+      "epoch": 0.17601760176017603,
+      "grad_norm": 0.3448857367038727,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 240
+    },
+    {
+      "epoch": 0.18335166850018336,
+      "grad_norm": 0.3475699722766876,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 250
+    },
+    {
+      "epoch": 0.1906857352401907,
+      "grad_norm": 0.2770358622074127,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 260
+    },
+    {
+      "epoch": 0.19801980198019803,
+      "grad_norm": 0.4310270845890045,
+      "learning_rate": 0.0002,
+      "loss": 1.2238,
+      "step": 270
+    },
+    {
+      "epoch": 0.20535386872020536,
+      "grad_norm": 0.335041880607605,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 280
+    },
+    {
+      "epoch": 0.2126879354602127,
+      "grad_norm": 0.3420602083206177,
+      "learning_rate": 0.0002,
+      "loss": 1.0959,
+      "step": 290
+    },
+    {
+      "epoch": 0.22002200220022003,
+      "grad_norm": 0.325001060962677,
+      "learning_rate": 0.0002,
+      "loss": 1.1232,
+      "step": 300
+    },
+    {
+      "epoch": 0.22735606894022736,
+      "grad_norm": 0.3027827739715576,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 310
+    },
+    {
+      "epoch": 0.2346901356802347,
+      "grad_norm": 0.435550719499588,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 320
+    },
+    {
+      "epoch": 0.24202420242024203,
+      "grad_norm": 0.3884522616863251,
+      "learning_rate": 0.0002,
+      "loss": 1.2045,
+      "step": 330
+    },
+    {
+      "epoch": 0.24935826916024936,
+      "grad_norm": 0.7736002206802368,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 340
+    },
+    {
+      "epoch": 0.2566923359002567,
+      "grad_norm": 0.35052821040153503,
+      "learning_rate": 0.0002,
+      "loss": 1.3606,
+      "step": 350
+    },
+    {
+      "epoch": 0.264026402640264,
+      "grad_norm": 0.3311890959739685,
+      "learning_rate": 0.0002,
+      "loss": 1.2129,
+      "step": 360
+    },
+    {
+      "epoch": 0.27136046938027136,
+      "grad_norm": 0.7473500370979309,
+      "learning_rate": 0.0002,
+      "loss": 1.2219,
+      "step": 370
+    },
+    {
+      "epoch": 0.27869453612027867,
+      "grad_norm": 0.3681875765323639,
+      "learning_rate": 0.0002,
+      "loss": 1.2712,
+      "step": 380
+    },
+    {
+      "epoch": 0.28602860286028603,
+      "grad_norm": 0.3764737844467163,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 390
+    },
+    {
+      "epoch": 0.29336266960029334,
+      "grad_norm": 0.4243989586830139,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 400
+    },
+    {
+      "epoch": 0.3006967363403007,
+      "grad_norm": 0.2658531963825226,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 410
+    },
+    {
+      "epoch": 0.30803080308030806,
+      "grad_norm": 0.3436793386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.1622,
+      "step": 420
+    },
+    {
+      "epoch": 0.31536486982031536,
+      "grad_norm": 0.5101129412651062,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 430
+    },
+    {
+      "epoch": 0.3226989365603227,
+      "grad_norm": 0.3319750726222992,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 440
+    },
+    {
+      "epoch": 0.33003300330033003,
+      "grad_norm": 0.385148286819458,
+      "learning_rate": 0.0002,
+      "loss": 1.1804,
+      "step": 450
+    },
+    {
+      "epoch": 0.3373670700403374,
+      "grad_norm": 0.3477935791015625,
+      "learning_rate": 0.0002,
+      "loss": 1.1808,
+      "step": 460
+    },
+    {
+      "epoch": 0.3447011367803447,
+      "grad_norm": 0.29748716950416565,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 470
+    },
+    {
+      "epoch": 0.35203520352035206,
+      "grad_norm": 0.34083324670791626,
+      "learning_rate": 0.0002,
+      "loss": 1.19,
+      "step": 480
+    },
+    {
+      "epoch": 0.35936927026035936,
+      "grad_norm": 0.36904552578926086,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 490
+    },
+    {
+      "epoch": 0.3667033370003667,
+      "grad_norm": 0.315483033657074,
+      "learning_rate": 0.0002,
+      "loss": 1.2223,
+      "step": 500
+    },
+    {
+      "epoch": 0.37403740374037403,
+      "grad_norm": 0.44897955656051636,
+      "learning_rate": 0.0002,
+      "loss": 1.1461,
+      "step": 510
+    },
+    {
+      "epoch": 0.3813714704803814,
+      "grad_norm": 0.3160701394081116,
+      "learning_rate": 0.0002,
+      "loss": 1.3035,
+      "step": 520
+    },
+    {
+      "epoch": 0.3887055372203887,
+      "grad_norm": 0.29584741592407227,
+      "learning_rate": 0.0002,
+      "loss": 1.3197,
+      "step": 530
+    },
+    {
+      "epoch": 0.39603960396039606,
+      "grad_norm": 0.5430002808570862,
+      "learning_rate": 0.0002,
+      "loss": 1.2983,
+      "step": 540
+    },
+    {
+      "epoch": 0.40337367070040336,
+      "grad_norm": 0.2908070683479309,
+      "learning_rate": 0.0002,
+      "loss": 1.2459,
+      "step": 550
+    },
+    {
+      "epoch": 0.4107077374404107,
+      "grad_norm": 0.35066530108451843,
+      "learning_rate": 0.0002,
+      "loss": 1.2384,
+      "step": 560
+    },
+    {
+      "epoch": 0.41804180418041803,
+      "grad_norm": 0.37588003277778625,
+      "learning_rate": 0.0002,
+      "loss": 1.1784,
+      "step": 570
+    },
+    {
+      "epoch": 0.4253758709204254,
+      "grad_norm": 0.3112126886844635,
+      "learning_rate": 0.0002,
+      "loss": 1.2334,
+      "step": 580
+    },
+    {
+      "epoch": 0.4327099376604327,
+      "grad_norm": 0.35577139258384705,
+      "learning_rate": 0.0002,
+      "loss": 1.1439,
+      "step": 590
+    },
+    {
+      "epoch": 0.44004400440044006,
+      "grad_norm": 0.31706422567367554,
+      "learning_rate": 0.0002,
+      "loss": 1.184,
+      "step": 600
+    },
+    {
+      "epoch": 0.44737807114044736,
+      "grad_norm": 0.3249092102050781,
+      "learning_rate": 0.0002,
+      "loss": 1.2081,
+      "step": 610
+    },
+    {
+      "epoch": 0.4547121378804547,
+      "grad_norm": 0.3842705488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.0824,
+      "step": 620
+    },
+    {
+      "epoch": 0.46204620462046203,
+      "grad_norm": 0.390991747379303,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 630
+    },
+    {
+      "epoch": 0.4693802713604694,
+      "grad_norm": 0.27532413601875305,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 640
+    },
+    {
+      "epoch": 0.4767143381004767,
+      "grad_norm": 0.31412816047668457,
+      "learning_rate": 0.0002,
+      "loss": 1.1058,
+      "step": 650
+    },
+    {
+      "epoch": 0.48404840484048406,
+      "grad_norm": 0.32117101550102234,
+      "learning_rate": 0.0002,
+      "loss": 1.1312,
+      "step": 660
+    },
+    {
+      "epoch": 0.49138247158049136,
+      "grad_norm": 0.3810010254383087,
+      "learning_rate": 0.0002,
+      "loss": 1.2423,
+      "step": 670
+    },
+    {
+      "epoch": 0.4987165383204987,
+      "grad_norm": 0.36289164423942566,
+      "learning_rate": 0.0002,
+      "loss": 1.1978,
+      "step": 680
+    },
+    {
+      "epoch": 0.506050605060506,
+      "grad_norm": 0.34458720684051514,
+      "learning_rate": 0.0002,
+      "loss": 1.2034,
+      "step": 690
+    },
+    {
+      "epoch": 0.5133846718005134,
+      "grad_norm": 0.32844600081443787,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 700
+    },
+    {
+      "epoch": 0.5207187385405208,
+      "grad_norm": 0.3144175708293915,
+      "learning_rate": 0.0002,
+      "loss": 1.0807,
+      "step": 710
+    },
+    {
+      "epoch": 0.528052805280528,
+      "grad_norm": 0.3898887634277344,
+      "learning_rate": 0.0002,
+      "loss": 1.1952,
+      "step": 720
+    },
+    {
+      "epoch": 0.5353868720205354,
+      "grad_norm": 1.3220758438110352,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 730
+    },
+    {
+      "epoch": 0.5427209387605427,
+      "grad_norm": 0.3635874390602112,
+      "learning_rate": 0.0002,
+      "loss": 1.227,
+      "step": 740
+    },
+    {
+      "epoch": 0.5500550055005501,
+      "grad_norm": 0.3138217628002167,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 750
+    },
+    {
+      "epoch": 0.5573890722405573,
+      "grad_norm": 0.4063207805156708,
+      "learning_rate": 0.0002,
+      "loss": 1.1516,
+      "step": 760
+    },
+    {
+      "epoch": 0.5647231389805647,
+      "grad_norm": 0.3926219940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 770
+    },
+    {
+      "epoch": 0.5720572057205721,
+      "grad_norm": 0.31954652070999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1726,
+      "step": 780
+    },
+    {
+      "epoch": 0.5793912724605794,
+      "grad_norm": 0.4248711168766022,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 790
+    },
+    {
+      "epoch": 0.5867253392005867,
+      "grad_norm": 0.643004834651947,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 800
+    },
+    {
+      "epoch": 0.594059405940594,
+      "grad_norm": 0.3479592800140381,
+      "learning_rate": 0.0002,
+      "loss": 1.1793,
+      "step": 810
+    },
+    {
+      "epoch": 0.6013934726806014,
+      "grad_norm": 0.4684754014015198,
+      "learning_rate": 0.0002,
+      "loss": 1.2426,
+      "step": 820
+    },
+    {
+      "epoch": 0.6087275394206088,
+      "grad_norm": 0.3739790916442871,
+      "learning_rate": 0.0002,
+      "loss": 1.2002,
+      "step": 830
+    },
+    {
+      "epoch": 0.6160616061606161,
+      "grad_norm": 0.40884748101234436,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 840
+    },
+    {
+      "epoch": 0.6233956729006234,
+      "grad_norm": 0.9722164273262024,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 850
+    },
+    {
+      "epoch": 0.6307297396406307,
+      "grad_norm": 0.42992347478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.3069,
+      "step": 860
+    },
+    {
+      "epoch": 0.6380638063806381,
+      "grad_norm": 0.36654195189476013,
+      "learning_rate": 0.0002,
+      "loss": 1.1339,
+      "step": 870
+    },
+    {
+      "epoch": 0.6453978731206454,
+      "grad_norm": 0.4113832116127014,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 880
+    },
+    {
+      "epoch": 0.6527319398606527,
+      "grad_norm": 0.2948838770389557,
+      "learning_rate": 0.0002,
+      "loss": 1.2163,
+      "step": 890
+    },
+    {
+      "epoch": 0.6600660066006601,
+      "grad_norm": 0.38330280780792236,
+      "learning_rate": 0.0002,
+      "loss": 1.1081,
+      "step": 900
+    },
+    {
+      "epoch": 0.6674000733406674,
+      "grad_norm": 0.4428867697715759,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 910
+    },
+    {
+      "epoch": 0.6747341400806748,
+      "grad_norm": 0.23659265041351318,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 920
+    },
+    {
+      "epoch": 0.682068206820682,
+      "grad_norm": 0.323685884475708,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 930
+    },
+    {
+      "epoch": 0.6894022735606894,
+      "grad_norm": 0.39157727360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 940
+    },
+    {
+      "epoch": 0.6967363403006968,
+      "grad_norm": 0.27189481258392334,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 950
+    },
+    {
+      "epoch": 0.7040704070407041,
+      "grad_norm": 0.529883861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.1033,
+      "step": 960
+    },
+    {
+      "epoch": 0.7114044737807114,
+      "grad_norm": 0.34758689999580383,
+      "learning_rate": 0.0002,
+      "loss": 1.139,
+      "step": 970
+    },
+    {
+      "epoch": 0.7187385405207187,
+      "grad_norm": 0.831749439239502,
+      "learning_rate": 0.0002,
+      "loss": 1.2197,
+      "step": 980
+    },
+    {
+      "epoch": 0.7260726072607261,
+      "grad_norm": 0.4438304007053375,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 990
+    },
+    {
+      "epoch": 0.7334066740007334,
+      "grad_norm": 0.33840006589889526,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7407407407407407,
+      "grad_norm": 0.3454797863960266,
+      "learning_rate": 0.0002,
+      "loss": 1.254,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7480748074807481,
+      "grad_norm": 0.38999441266059875,
+      "learning_rate": 0.0002,
+      "loss": 1.106,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7554088742207554,
+      "grad_norm": 0.2829911708831787,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7627429409607628,
+      "grad_norm": 0.36918163299560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 1040
+    },
+    {
+      "epoch": 0.77007700770077,
+      "grad_norm": 0.3415680229663849,
+      "learning_rate": 0.0002,
+      "loss": 1.3028,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7774110744407774,
+      "grad_norm": 0.2974182963371277,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7847451411807848,
+      "grad_norm": 0.3880919814109802,
+      "learning_rate": 0.0002,
+      "loss": 1.194,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7920792079207921,
+      "grad_norm": 0.33503302931785583,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7994132746607994,
+      "grad_norm": 0.3728407025337219,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8067473414008067,
+      "grad_norm": 0.3509373664855957,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8140814081408141,
+      "grad_norm": 0.42228564620018005,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8214154748808215,
+      "grad_norm": 0.313467800617218,
+      "learning_rate": 0.0002,
+      "loss": 1.1788,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8287495416208287,
+      "grad_norm": 0.3378850817680359,
+      "learning_rate": 0.0002,
+      "loss": 1.1971,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8360836083608361,
+      "grad_norm": 0.43200382590293884,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8434176751008434,
+      "grad_norm": 0.3309599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.3203,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8507517418408508,
+      "grad_norm": 0.3526846170425415,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1160
+    },
+    {
+      "epoch": 0.858085808580858,
+      "grad_norm": 1.2722247838974,
+      "learning_rate": 0.0002,
+      "loss": 1.0851,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8654198753208654,
+      "grad_norm": 0.34142059087753296,
+      "learning_rate": 0.0002,
+      "loss": 1.0785,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8727539420608728,
+      "grad_norm": 0.3805823028087616,
+      "learning_rate": 0.0002,
+      "loss": 1.2187,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8800880088008801,
+      "grad_norm": 0.3931232690811157,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8874220755408874,
+      "grad_norm": 0.2937372624874115,
+      "learning_rate": 0.0002,
+      "loss": 1.0948,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8947561422808947,
+      "grad_norm": 0.3757196366786957,
+      "learning_rate": 0.0002,
+      "loss": 1.1228,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9020902090209021,
+      "grad_norm": 0.3502705991268158,
+      "learning_rate": 0.0002,
+      "loss": 1.1222,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9094242757609095,
+      "grad_norm": 0.32758915424346924,
+      "learning_rate": 0.0002,
+      "loss": 1.2242,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9167583425009168,
+      "grad_norm": 0.37199416756629944,
+      "learning_rate": 0.0002,
+      "loss": 1.215,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9240924092409241,
+      "grad_norm": 0.3551490604877472,
+      "learning_rate": 0.0002,
+      "loss": 1.1225,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9314264759809314,
+      "grad_norm": 0.2859550714492798,
+      "learning_rate": 0.0002,
+      "loss": 1.1966,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9387605427209388,
+      "grad_norm": 0.427990585565567,
+      "learning_rate": 0.0002,
+      "loss": 1.2186,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9460946094609461,
+      "grad_norm": 0.33717992901802063,
+      "learning_rate": 0.0002,
+      "loss": 1.2848,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9534286762009534,
+      "grad_norm": 0.30225634574890137,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9607627429409608,
+      "grad_norm": 0.385821133852005,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9680968096809681,
+      "grad_norm": 0.35278066992759705,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9754308764209755,
+      "grad_norm": 0.49987098574638367,
+      "learning_rate": 0.0002,
+      "loss": 1.1071,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9827649431609827,
+      "grad_norm": 0.3842747211456299,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9900990099009901,
+      "grad_norm": 0.6274653673171997,
+      "learning_rate": 0.0002,
+      "loss": 1.0862,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9974330766409975,
+      "grad_norm": 0.5239808559417725,
+      "learning_rate": 0.0002,
+      "loss": 1.124,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9996332966629996,
+      "eval_loss": 1.1822267770767212,
+      "eval_runtime": 32.7389,
+      "eval_samples_per_second": 13.165,
+      "eval_steps_per_second": 1.649,
+      "step": 1363
+    },
+    {
+      "epoch": 1.0047671433810048,
+      "grad_norm": 0.45311301946640015,
+      "learning_rate": 0.0002,
+      "loss": 1.096,
+      "step": 1370
+    },
+    {
+      "epoch": 1.012101210121012,
+      "grad_norm": 0.29685574769973755,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0194352768610195,
+      "grad_norm": 0.3290937840938568,
+      "learning_rate": 0.0002,
+      "loss": 1.0302,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0267693436010268,
+      "grad_norm": 0.3801758587360382,
+      "learning_rate": 0.0002,
+      "loss": 1.0295,
+      "step": 1400
+    },
+    {
+      "epoch": 1.034103410341034,
+      "grad_norm": 0.794174313545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0414374770810415,
+      "grad_norm": 0.3854154646396637,
+      "learning_rate": 0.0002,
+      "loss": 1.2232,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0487715438210488,
+      "grad_norm": 0.32702451944351196,
+      "learning_rate": 0.0002,
+      "loss": 1.0652,
+      "step": 1430
+    },
+    {
+      "epoch": 1.056105610561056,
+      "grad_norm": 0.7815203666687012,
+      "learning_rate": 0.0002,
+      "loss": 1.1144,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0634396773010635,
+      "grad_norm": 0.3087436854839325,
+      "learning_rate": 0.0002,
+      "loss": 1.1316,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0707737440410707,
+      "grad_norm": 0.3847602903842926,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0781078107810782,
+      "grad_norm": 0.3693031370639801,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0854418775210855,
+      "grad_norm": 0.4111202359199524,
+      "learning_rate": 0.0002,
+      "loss": 1.0995,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0927759442610927,
+      "grad_norm": 0.41452381014823914,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1001100110011002,
+      "grad_norm": 0.3336445093154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1068,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1074440777411074,
+      "grad_norm": 0.3923407793045044,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1147781444811147,
+      "grad_norm": 0.46215683221817017,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1221122112211221,
+      "grad_norm": 0.3592156767845154,
+      "learning_rate": 0.0002,
+      "loss": 1.1133,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1294462779611294,
+      "grad_norm": 0.361110657453537,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1367803447011369,
+      "grad_norm": 0.5317131280899048,
+      "learning_rate": 0.0002,
+      "loss": 1.1553,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1441144114411441,
+      "grad_norm": 0.3882388174533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1514484781811514,
+      "grad_norm": 0.3259428143501282,
+      "learning_rate": 0.0002,
+      "loss": 1.0805,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1587825449211588,
+      "grad_norm": 0.410935640335083,
+      "learning_rate": 0.0002,
+      "loss": 1.1819,
+      "step": 1580
+    },
+    {
+      "epoch": 1.166116611661166,
+      "grad_norm": 0.44940185546875,
+      "learning_rate": 0.0002,
+      "loss": 1.1143,
+      "step": 1590
+    },
+    {
+      "epoch": 1.1734506784011733,
+      "grad_norm": 0.5106484293937683,
+      "learning_rate": 0.0002,
+      "loss": 1.0334,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1807847451411808,
+      "grad_norm": 0.6603665947914124,
+      "learning_rate": 0.0002,
+      "loss": 1.2376,
+      "step": 1610
+    },
+    {
+      "epoch": 1.188118811881188,
+      "grad_norm": 0.4799964129924774,
+      "learning_rate": 0.0002,
+      "loss": 1.1227,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1954528786211955,
+      "grad_norm": 0.4389883279800415,
+      "learning_rate": 0.0002,
+      "loss": 1.1191,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2027869453612028,
+      "grad_norm": 0.4188813269138336,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 1640
+    },
+    {
+      "epoch": 1.21012101210121,
+      "grad_norm": 0.7132157683372498,
+      "learning_rate": 0.0002,
+      "loss": 1.0605,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2174550788412175,
+      "grad_norm": 0.507480263710022,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2247891455812248,
+      "grad_norm": 0.9452332854270935,
+      "learning_rate": 0.0002,
+      "loss": 0.9948,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2321232123212322,
+      "grad_norm": 0.4121614992618561,
+      "learning_rate": 0.0002,
+      "loss": 1.0228,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2394572790612395,
+      "grad_norm": 0.34230247139930725,
+      "learning_rate": 0.0002,
+      "loss": 1.0366,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2467913458012467,
+      "grad_norm": 0.4026208817958832,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2541254125412542,
+      "grad_norm": 0.46673697233200073,
+      "learning_rate": 0.0002,
+      "loss": 1.0206,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2614594792812615,
+      "grad_norm": 0.38349825143814087,
+      "learning_rate": 0.0002,
+      "loss": 1.0827,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2687935460212687,
+      "grad_norm": 0.4049997627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.0356,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2761276127612762,
+      "grad_norm": 0.3417615294456482,
+      "learning_rate": 0.0002,
+      "loss": 0.9504,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2834616795012834,
+      "grad_norm": 0.4277614951133728,
+      "learning_rate": 0.0002,
+      "loss": 1.094,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2907957462412907,
+      "grad_norm": 0.5864202976226807,
+      "learning_rate": 0.0002,
+      "loss": 0.9938,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2981298129812981,
+      "grad_norm": 0.7097493410110474,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3054638797213054,
+      "grad_norm": 0.3145381212234497,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3127979464613129,
+      "grad_norm": 0.5116165280342102,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3201320132013201,
+      "grad_norm": 0.7469736337661743,
+      "learning_rate": 0.0002,
+      "loss": 1.0765,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3274660799413276,
+      "grad_norm": 0.32272255420684814,
+      "learning_rate": 0.0002,
+      "loss": 1.0663,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3348001466813348,
+      "grad_norm": 0.3534623086452484,
+      "learning_rate": 0.0002,
+      "loss": 0.9887,
+      "step": 1820
+    },
+    {
+      "epoch": 1.342134213421342,
+      "grad_norm": 0.36127907037734985,
+      "learning_rate": 0.0002,
+      "loss": 1.1628,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3494682801613496,
+      "grad_norm": 0.4072401523590088,
+      "learning_rate": 0.0002,
+      "loss": 1.0972,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3568023469013568,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.1267,
+      "step": 1850
+    },
+    {
+      "epoch": 1.364136413641364,
+      "grad_norm": 0.412883460521698,
+      "learning_rate": 0.0002,
+      "loss": 1.0173,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3714704803813715,
+      "grad_norm": 0.3735875189304352,
+      "learning_rate": 0.0002,
+      "loss": 1.0265,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3788045471213788,
+      "grad_norm": 0.39158159494400024,
+      "learning_rate": 0.0002,
+      "loss": 1.1061,
+      "step": 1880
+    },
+    {
+      "epoch": 1.386138613861386,
+      "grad_norm": 0.44431769847869873,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3934726806013935,
+      "grad_norm": 0.37772801518440247,
+      "learning_rate": 0.0002,
+      "loss": 1.0216,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4008067473414008,
+      "grad_norm": 0.4056641757488251,
+      "learning_rate": 0.0002,
+      "loss": 1.0674,
+      "step": 1910
+    },
+    {
+      "epoch": 1.408140814081408,
+      "grad_norm": 0.41612377762794495,
+      "learning_rate": 0.0002,
+      "loss": 1.0256,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4154748808214155,
+      "grad_norm": 0.41153013706207275,
+      "learning_rate": 0.0002,
+      "loss": 1.0467,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4228089475614227,
+      "grad_norm": 0.387845516204834,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4301430143014302,
+      "grad_norm": 0.3809587061405182,
+      "learning_rate": 0.0002,
+      "loss": 1.1094,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4374770810414375,
+      "grad_norm": 0.3625726103782654,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1960
+    },
+    {
+      "epoch": 1.444811147781445,
+      "grad_norm": 0.5294290781021118,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4521452145214522,
+      "grad_norm": 0.39975494146347046,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4594792812614594,
+      "grad_norm": 0.4181167185306549,
+      "learning_rate": 0.0002,
+      "loss": 0.9704,
+      "step": 1990
+    },
+    {
+      "epoch": 1.466813348001467,
+      "grad_norm": 0.42001503705978394,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4741474147414741,
+      "grad_norm": 0.4877578616142273,
+      "learning_rate": 0.0002,
+      "loss": 1.1266,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4814814814814814,
+      "grad_norm": 0.4050969183444977,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4888155482214889,
+      "grad_norm": 0.39068883657455444,
+      "learning_rate": 0.0002,
+      "loss": 1.0562,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4961496149614961,
+      "grad_norm": 0.421282559633255,
+      "learning_rate": 0.0002,
+      "loss": 1.0464,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5034836817015034,
+      "grad_norm": 0.47092297673225403,
+      "learning_rate": 0.0002,
+      "loss": 1.0532,
+      "step": 2050
+    },
+    {
+      "epoch": 1.5108177484415108,
+      "grad_norm": 0.39688974618911743,
+      "learning_rate": 0.0002,
+      "loss": 0.9348,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5181518151815183,
+      "grad_norm": 0.5529879331588745,
+      "learning_rate": 0.0002,
+      "loss": 1.08,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5254858819215253,
+      "grad_norm": 0.4879782199859619,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5328199486615328,
+      "grad_norm": 0.5517361164093018,
+      "learning_rate": 0.0002,
+      "loss": 1.0432,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5401540154015403,
+      "grad_norm": 0.44015637040138245,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5474880821415475,
+      "grad_norm": 0.5435167551040649,
+      "learning_rate": 0.0002,
+      "loss": 1.1873,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5548221488815548,
+      "grad_norm": 0.5714033246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1076,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5621562156215623,
+      "grad_norm": 0.31732529401779175,
+      "learning_rate": 0.0002,
+      "loss": 1.1107,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5694902823615695,
+      "grad_norm": 0.49068278074264526,
+      "learning_rate": 0.0002,
+      "loss": 1.0817,
+      "step": 2140
+    },
+    {
+      "epoch": 1.5768243491015768,
+      "grad_norm": 0.46851542592048645,
+      "learning_rate": 0.0002,
+      "loss": 1.0254,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5841584158415842,
+      "grad_norm": 0.5083092451095581,
+      "learning_rate": 0.0002,
+      "loss": 1.0623,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5914924825815915,
+      "grad_norm": 0.9822936058044434,
+      "learning_rate": 0.0002,
+      "loss": 1.0603,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5988265493215987,
+      "grad_norm": 0.4575989246368408,
+      "learning_rate": 0.0002,
+      "loss": 0.9986,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6061606160616062,
+      "grad_norm": 0.47444286942481995,
+      "learning_rate": 0.0002,
+      "loss": 1.1292,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6134946828016135,
+      "grad_norm": 0.7208226919174194,
+      "learning_rate": 0.0002,
+      "loss": 1.0136,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6208287495416207,
+      "grad_norm": 0.43791481852531433,
+      "learning_rate": 0.0002,
+      "loss": 1.15,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6281628162816282,
+      "grad_norm": 0.5245792865753174,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6354968830216357,
+      "grad_norm": 0.39289429783821106,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6428309497616427,
+      "grad_norm": 0.6106135845184326,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6501650165016502,
+      "grad_norm": 0.3722580671310425,
+      "learning_rate": 0.0002,
+      "loss": 1.0129,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6574990832416576,
+      "grad_norm": 0.3649403750896454,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6648331499816649,
+      "grad_norm": 0.46514248847961426,
+      "learning_rate": 0.0002,
+      "loss": 1.0037,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6721672167216721,
+      "grad_norm": 0.42034927010536194,
+      "learning_rate": 0.0002,
+      "loss": 1.0022,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6795012834616796,
+      "grad_norm": 0.45202910900115967,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6868353502016868,
+      "grad_norm": 0.36257603764533997,
+      "learning_rate": 0.0002,
+      "loss": 1.0866,
+      "step": 2300
+    },
+    {
+      "epoch": 1.694169416941694,
+      "grad_norm": 0.6340323090553284,
+      "learning_rate": 0.0002,
+      "loss": 1.0973,
+      "step": 2310
+    },
+    {
+      "epoch": 1.7015034836817016,
+      "grad_norm": 0.4352878928184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0615,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7088375504217088,
+      "grad_norm": 0.45029792189598083,
+      "learning_rate": 0.0002,
+      "loss": 1.0629,
+      "step": 2330
+    },
+    {
+      "epoch": 1.716171617161716,
+      "grad_norm": 0.3891315758228302,
+      "learning_rate": 0.0002,
+      "loss": 0.9621,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7235056839017235,
+      "grad_norm": 0.35180050134658813,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7308397506417308,
+      "grad_norm": 0.42367449402809143,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 2360
+    },
+    {
+      "epoch": 1.738173817381738,
+      "grad_norm": 0.4553675353527069,
+      "learning_rate": 0.0002,
+      "loss": 1.0376,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7455078841217455,
+      "grad_norm": 0.5944654941558838,
+      "learning_rate": 0.0002,
+      "loss": 1.1467,
+      "step": 2380
+    },
+    {
+      "epoch": 1.752841950861753,
+      "grad_norm": 0.3479664623737335,
+      "learning_rate": 0.0002,
+      "loss": 1.0548,
+      "step": 2390
+    },
+    {
+      "epoch": 1.76017601760176,
+      "grad_norm": 0.3585502505302429,
+      "learning_rate": 0.0002,
+      "loss": 1.0798,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7675100843417675,
+      "grad_norm": 0.4263346493244171,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 2410
+    },
+    {
+      "epoch": 1.774844151081775,
+      "grad_norm": 0.5476409196853638,
+      "learning_rate": 0.0002,
+      "loss": 1.054,
+      "step": 2420
+    },
+    {
+      "epoch": 1.7821782178217822,
+      "grad_norm": 0.3694186508655548,
+      "learning_rate": 0.0002,
+      "loss": 1.1615,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7895122845617895,
+      "grad_norm": 0.9185658693313599,
+      "learning_rate": 0.0002,
+      "loss": 1.1343,
+      "step": 2440
+    },
+    {
+      "epoch": 1.796846351301797,
+      "grad_norm": 0.7171908020973206,
+      "learning_rate": 0.0002,
+      "loss": 1.0764,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8041804180418042,
+      "grad_norm": 0.550658643245697,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 2460
+    },
+    {
+      "epoch": 1.8115144847818114,
+      "grad_norm": 0.4075568914413452,
+      "learning_rate": 0.0002,
+      "loss": 0.9975,
+      "step": 2470
+    },
+    {
+      "epoch": 1.818848551521819,
+      "grad_norm": 0.3790127635002136,
+      "learning_rate": 0.0002,
+      "loss": 1.0935,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8261826182618262,
+      "grad_norm": 0.3576384484767914,
+      "learning_rate": 0.0002,
+      "loss": 0.9839,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8335166850018334,
+      "grad_norm": 0.3919370770454407,
+      "learning_rate": 0.0002,
+      "loss": 1.1369,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8408507517418409,
+      "grad_norm": 0.485083669424057,
+      "learning_rate": 0.0002,
+      "loss": 0.9985,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8481848184818483,
+      "grad_norm": 0.4564347565174103,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8555188852218554,
+      "grad_norm": 0.3613106608390808,
+      "learning_rate": 0.0002,
+      "loss": 1.0944,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8628529519618628,
+      "grad_norm": 0.39600759744644165,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8701870187018703,
+      "grad_norm": 1.123499870300293,
+      "learning_rate": 0.0002,
+      "loss": 0.9453,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8775210854418776,
+      "grad_norm": 0.4612680673599243,
+      "learning_rate": 0.0002,
+      "loss": 1.0635,
+      "step": 2560
+    },
+    {
+      "epoch": 1.8848551521818848,
+      "grad_norm": 0.42745399475097656,
+      "learning_rate": 0.0002,
+      "loss": 1.0087,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8921892189218923,
+      "grad_norm": 0.4055580198764801,
+      "learning_rate": 0.0002,
+      "loss": 1.0102,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8995232856618995,
+      "grad_norm": 0.44174644351005554,
+      "learning_rate": 0.0002,
+      "loss": 1.0177,
+      "step": 2590
+    },
+    {
+      "epoch": 1.9068573524019068,
+      "grad_norm": 1.0228385925292969,
+      "learning_rate": 0.0002,
+      "loss": 0.9886,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9141914191419143,
+      "grad_norm": 0.3496396243572235,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9215254858819215,
+      "grad_norm": 0.4191173017024994,
+      "learning_rate": 0.0002,
+      "loss": 1.0955,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9288595526219288,
+      "grad_norm": 0.6778554916381836,
+      "learning_rate": 0.0002,
+      "loss": 1.0943,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9361936193619362,
+      "grad_norm": 0.41992834210395813,
+      "learning_rate": 0.0002,
+      "loss": 1.0594,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9435276861019435,
+      "grad_norm": 0.8760401010513306,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9508617528419507,
+      "grad_norm": 0.44049209356307983,
+      "learning_rate": 0.0002,
+      "loss": 1.0379,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9581958195819582,
+      "grad_norm": 0.5651928782463074,
+      "learning_rate": 0.0002,
+      "loss": 1.1008,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9655298863219657,
+      "grad_norm": 0.5292727947235107,
+      "learning_rate": 0.0002,
+      "loss": 1.1317,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9728639530619727,
+      "grad_norm": 0.6012240648269653,
+      "learning_rate": 0.0002,
+      "loss": 1.1328,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9801980198019802,
+      "grad_norm": 0.3945149779319763,
+      "learning_rate": 0.0002,
+      "loss": 1.0683,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9875320865419877,
+      "grad_norm": 0.5732627511024475,
+      "learning_rate": 0.0002,
+      "loss": 1.0155,
+      "step": 2710
+    },
+    {
+      "epoch": 1.994866153281995,
+      "grad_norm": 0.3963361084461212,
+      "learning_rate": 0.0002,
+      "loss": 0.9857,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1534006595611572,
+      "eval_runtime": 32.7541,
+      "eval_samples_per_second": 13.159,
+      "eval_steps_per_second": 1.649,
+      "step": 2727
+    },
+    {
+      "epoch": 2.002200220022002,
+      "grad_norm": 0.48628315329551697,
+      "learning_rate": 0.0002,
+      "loss": 0.9624,
+      "step": 2730
+    },
+    {
+      "epoch": 2.0095342867620096,
+      "grad_norm": 0.413875013589859,
+      "learning_rate": 0.0002,
+      "loss": 0.9603,
+      "step": 2740
+    },
+    {
+      "epoch": 2.0168683535020167,
+      "grad_norm": 0.4988735616207123,
+      "learning_rate": 0.0002,
+      "loss": 0.965,
+      "step": 2750
+    },
+    {
+      "epoch": 2.024202420242024,
+      "grad_norm": 0.5634812712669373,
+      "learning_rate": 0.0002,
+      "loss": 0.9677,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0315364869820316,
+      "grad_norm": 0.48302653431892395,
+      "learning_rate": 0.0002,
+      "loss": 0.9547,
+      "step": 2770
+    },
+    {
+      "epoch": 2.038870553722039,
+      "grad_norm": 0.49914175271987915,
+      "learning_rate": 0.0002,
+      "loss": 0.9346,
+      "step": 2780
+    },
+    {
+      "epoch": 2.046204620462046,
+      "grad_norm": 1.14039945602417,
+      "learning_rate": 0.0002,
+      "loss": 0.904,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0535386872020536,
+      "grad_norm": 0.6359720826148987,
+      "learning_rate": 0.0002,
+      "loss": 0.9588,
+      "step": 2800
+    },
+    {
+      "epoch": 2.060872753942061,
+      "grad_norm": 0.4589158296585083,
+      "learning_rate": 0.0002,
+      "loss": 0.9031,
+      "step": 2810
+    },
+    {
+      "epoch": 2.068206820682068,
+      "grad_norm": 0.46255481243133545,
+      "learning_rate": 0.0002,
+      "loss": 0.9438,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0755408874220755,
+      "grad_norm": 0.6232137680053711,
+      "learning_rate": 0.0002,
+      "loss": 0.9464,
+      "step": 2830
+    },
+    {
+      "epoch": 2.082874954162083,
+      "grad_norm": 0.41042178869247437,
+      "learning_rate": 0.0002,
+      "loss": 0.8978,
+      "step": 2840
+    },
+    {
+      "epoch": 2.09020902090209,
+      "grad_norm": 0.5334428548812866,
+      "learning_rate": 0.0002,
+      "loss": 0.8516,
+      "step": 2850
+    },
+    {
+      "epoch": 2.0975430876420975,
+      "grad_norm": 0.8270058631896973,
+      "learning_rate": 0.0002,
+      "loss": 0.9313,
+      "step": 2860
+    },
+    {
+      "epoch": 2.104877154382105,
+      "grad_norm": 0.6624533534049988,
+      "learning_rate": 0.0002,
+      "loss": 1.0064,
+      "step": 2870
+    },
+    {
+      "epoch": 2.112211221122112,
+      "grad_norm": 0.5448863506317139,
+      "learning_rate": 0.0002,
+      "loss": 0.9196,
+      "step": 2880
+    },
+    {
+      "epoch": 2.1195452878621195,
+      "grad_norm": 0.621482789516449,
+      "learning_rate": 0.0002,
+      "loss": 0.887,
+      "step": 2890
+    },
+    {
+      "epoch": 2.126879354602127,
+      "grad_norm": 0.4556255340576172,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1342134213421344,
+      "grad_norm": 0.4620579183101654,
+      "learning_rate": 0.0002,
+      "loss": 0.9323,
+      "step": 2910
+    },
+    {
+      "epoch": 2.1415474880821415,
+      "grad_norm": 0.9602415561676025,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 2920
+    },
+    {
+      "epoch": 2.148881554822149,
+      "grad_norm": 0.587943971157074,
+      "learning_rate": 0.0002,
+      "loss": 0.8826,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1562156215621564,
+      "grad_norm": 0.5121372938156128,
+      "learning_rate": 0.0002,
+      "loss": 0.971,
+      "step": 2940
+    },
+    {
+      "epoch": 2.1635496883021634,
+      "grad_norm": 0.49424484372138977,
+      "learning_rate": 0.0002,
+      "loss": 0.8751,
+      "step": 2950
+    },
+    {
+      "epoch": 2.170883755042171,
+      "grad_norm": 0.6312560439109802,
+      "learning_rate": 0.0002,
+      "loss": 0.8674,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1782178217821784,
+      "grad_norm": 0.5235576629638672,
+      "learning_rate": 0.0002,
+      "loss": 0.9791,
+      "step": 2970
+    },
+    {
+      "epoch": 2.1855518885221854,
+      "grad_norm": 0.5868439674377441,
+      "learning_rate": 0.0002,
+      "loss": 0.9706,
+      "step": 2980
+    },
+    {
+      "epoch": 2.192885955262193,
+      "grad_norm": 0.42302873730659485,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 2990
+    },
+    {
+      "epoch": 2.2002200220022003,
+      "grad_norm": 0.5097725987434387,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 3000
+    },
+    {
+      "epoch": 2.2075540887422074,
+      "grad_norm": 0.5091572403907776,
+      "learning_rate": 0.0002,
+      "loss": 0.9239,
+      "step": 3010
+    },
+    {
+      "epoch": 2.214888155482215,
+      "grad_norm": 0.49433162808418274,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 3020
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.5577368140220642,
+      "learning_rate": 0.0002,
+      "loss": 0.9734,
+      "step": 3030
+    },
+    {
+      "epoch": 2.2295562889622293,
+      "grad_norm": 0.6177583932876587,
+      "learning_rate": 0.0002,
+      "loss": 0.9033,
+      "step": 3040
+    },
+    {
+      "epoch": 2.236890355702237,
+      "grad_norm": 0.5256719589233398,
+      "learning_rate": 0.0002,
+      "loss": 0.9882,
+      "step": 3050
+    },
+    {
+      "epoch": 2.2442244224422443,
+      "grad_norm": 0.5001118183135986,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2515584891822513,
+      "grad_norm": 0.5721249580383301,
+      "learning_rate": 0.0002,
+      "loss": 0.8718,
+      "step": 3070
+    },
+    {
+      "epoch": 2.258892555922259,
+      "grad_norm": 0.5325384140014648,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 3080
+    },
+    {
+      "epoch": 2.2662266226622663,
+      "grad_norm": 0.5719189047813416,
+      "learning_rate": 0.0002,
+      "loss": 0.9843,
+      "step": 3090
+    },
+    {
+      "epoch": 2.2735606894022737,
+      "grad_norm": 0.6337835788726807,
+      "learning_rate": 0.0002,
+      "loss": 0.8633,
+      "step": 3100
+    },
+    {
+      "epoch": 2.2808947561422808,
+      "grad_norm": 0.5381836891174316,
+      "learning_rate": 0.0002,
+      "loss": 0.9962,
+      "step": 3110
+    },
+    {
+      "epoch": 2.2882288228822882,
+      "grad_norm": 0.5408531427383423,
+      "learning_rate": 0.0002,
+      "loss": 0.8265,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2955628896222957,
+      "grad_norm": 0.43705281615257263,
+      "learning_rate": 0.0002,
+      "loss": 1.0325,
+      "step": 3130
+    },
+    {
+      "epoch": 2.3028969563623027,
+      "grad_norm": 0.6454030275344849,
+      "learning_rate": 0.0002,
+      "loss": 0.9388,
+      "step": 3140
+    },
+    {
+      "epoch": 2.31023102310231,
+      "grad_norm": 0.686030387878418,
+      "learning_rate": 0.0002,
+      "loss": 0.954,
+      "step": 3150
+    },
+    {
+      "epoch": 2.3175650898423177,
+      "grad_norm": 0.5123633146286011,
+      "learning_rate": 0.0002,
+      "loss": 0.9403,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3248991565823247,
+      "grad_norm": 0.842506468296051,
+      "learning_rate": 0.0002,
+      "loss": 0.8834,
+      "step": 3170
+    },
+    {
+      "epoch": 2.332233223322332,
+      "grad_norm": 0.5193818807601929,
+      "learning_rate": 0.0002,
+      "loss": 1.0497,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3395672900623397,
+      "grad_norm": 0.5634409189224243,
+      "learning_rate": 0.0002,
+      "loss": 0.9473,
+      "step": 3190
+    },
+    {
+      "epoch": 2.3469013568023467,
+      "grad_norm": 0.6475534439086914,
+      "learning_rate": 0.0002,
+      "loss": 0.8499,
+      "step": 3200
+    },
+    {
+      "epoch": 2.354235423542354,
+      "grad_norm": 1.1503914594650269,
+      "learning_rate": 0.0002,
+      "loss": 0.874,
+      "step": 3210
+    },
+    {
+      "epoch": 2.3615694902823616,
+      "grad_norm": 0.7234905362129211,
+      "learning_rate": 0.0002,
+      "loss": 0.9762,
+      "step": 3220
+    },
+    {
+      "epoch": 2.368903557022369,
+      "grad_norm": 0.664903461933136,
+      "learning_rate": 0.0002,
+      "loss": 0.9007,
+      "step": 3230
+    },
+    {
+      "epoch": 2.376237623762376,
+      "grad_norm": 0.5453006625175476,
+      "learning_rate": 0.0002,
+      "loss": 0.9987,
+      "step": 3240
+    },
+    {
+      "epoch": 2.3835716905023836,
+      "grad_norm": 0.6256654262542725,
+      "learning_rate": 0.0002,
+      "loss": 0.9742,
+      "step": 3250
+    },
+    {
+      "epoch": 2.390905757242391,
+      "grad_norm": 0.5166565179824829,
+      "learning_rate": 0.0002,
+      "loss": 0.9922,
+      "step": 3260
+    },
+    {
+      "epoch": 2.398239823982398,
+      "grad_norm": 0.5699098110198975,
+      "learning_rate": 0.0002,
+      "loss": 0.927,
+      "step": 3270
+    },
+    {
+      "epoch": 2.4055738907224056,
+      "grad_norm": 0.4472540020942688,
+      "learning_rate": 0.0002,
+      "loss": 0.8878,
+      "step": 3280
+    },
+    {
+      "epoch": 2.412907957462413,
+      "grad_norm": 0.6790403127670288,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 3290
+    },
+    {
+      "epoch": 2.42024202420242,
+      "grad_norm": 0.5182185173034668,
+      "learning_rate": 0.0002,
+      "loss": 0.972,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4275760909424275,
+      "grad_norm": 0.564647912979126,
+      "learning_rate": 0.0002,
+      "loss": 0.9775,
+      "step": 3310
+    },
+    {
+      "epoch": 2.434910157682435,
+      "grad_norm": 0.5625313520431519,
+      "learning_rate": 0.0002,
+      "loss": 1.072,
+      "step": 3320
+    },
+    {
+      "epoch": 2.442244224422442,
+      "grad_norm": 0.7496559619903564,
+      "learning_rate": 0.0002,
+      "loss": 0.8798,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4495782911624495,
+      "grad_norm": 0.4779128134250641,
+      "learning_rate": 0.0002,
+      "loss": 0.868,
+      "step": 3340
+    },
+    {
+      "epoch": 2.456912357902457,
+      "grad_norm": 0.578093409538269,
+      "learning_rate": 0.0002,
+      "loss": 1.0316,
+      "step": 3350
+    },
+    {
+      "epoch": 2.4642464246424645,
+      "grad_norm": 0.5456080436706543,
+      "learning_rate": 0.0002,
+      "loss": 0.9282,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4715804913824715,
+      "grad_norm": 0.4769273102283478,
+      "learning_rate": 0.0002,
+      "loss": 0.8409,
+      "step": 3370
+    },
+    {
+      "epoch": 2.478914558122479,
+      "grad_norm": 0.5608189702033997,
+      "learning_rate": 0.0002,
+      "loss": 0.9312,
+      "step": 3380
+    },
+    {
+      "epoch": 2.4862486248624864,
+      "grad_norm": 0.5590165853500366,
+      "learning_rate": 0.0002,
+      "loss": 0.9934,
+      "step": 3390
+    },
+    {
+      "epoch": 2.4935826916024935,
+      "grad_norm": 0.801306962966919,
+      "learning_rate": 0.0002,
+      "loss": 1.025,
+      "step": 3400
+    },
+    {
+      "epoch": 2.500916758342501,
+      "grad_norm": 0.6045624613761902,
+      "learning_rate": 0.0002,
+      "loss": 0.9049,
+      "step": 3410
+    },
+    {
+      "epoch": 2.5082508250825084,
+      "grad_norm": 0.5735858082771301,
+      "learning_rate": 0.0002,
+      "loss": 0.944,
+      "step": 3420
+    },
+    {
+      "epoch": 2.5155848918225154,
+      "grad_norm": 0.6827309131622314,
+      "learning_rate": 0.0002,
+      "loss": 0.9846,
+      "step": 3430
+    },
+    {
+      "epoch": 2.522918958562523,
+      "grad_norm": 0.5702602863311768,
+      "learning_rate": 0.0002,
+      "loss": 0.9789,
+      "step": 3440
+    },
+    {
+      "epoch": 2.5302530253025304,
+      "grad_norm": 0.6674721240997314,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5375870920425374,
+      "grad_norm": 0.5635907649993896,
+      "learning_rate": 0.0002,
+      "loss": 0.914,
+      "step": 3460
+    },
+    {
+      "epoch": 2.544921158782545,
+      "grad_norm": 0.42737770080566406,
+      "learning_rate": 0.0002,
+      "loss": 0.8398,
+      "step": 3470
+    },
+    {
+      "epoch": 2.5522552255225524,
+      "grad_norm": 0.6720691919326782,
+      "learning_rate": 0.0002,
+      "loss": 0.9474,
+      "step": 3480
+    },
+    {
+      "epoch": 2.55958929226256,
+      "grad_norm": 0.8917084336280823,
+      "learning_rate": 0.0002,
+      "loss": 0.8637,
+      "step": 3490
+    },
+    {
+      "epoch": 2.566923359002567,
+      "grad_norm": 0.5134549140930176,
+      "learning_rate": 0.0002,
+      "loss": 0.9257,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5742574257425743,
+      "grad_norm": 0.4951367974281311,
+      "learning_rate": 0.0002,
+      "loss": 0.9362,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5815914924825814,
+      "grad_norm": 0.9438204765319824,
+      "learning_rate": 0.0002,
+      "loss": 0.9184,
+      "step": 3520
+    },
+    {
+      "epoch": 2.588925559222589,
+      "grad_norm": 0.6024714708328247,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5962596259625963,
+      "grad_norm": 0.5248535871505737,
+      "learning_rate": 0.0002,
+      "loss": 0.9298,
+      "step": 3540
+    },
+    {
+      "epoch": 2.6035936927026038,
+      "grad_norm": 0.8677568435668945,
+      "learning_rate": 0.0002,
+      "loss": 0.941,
+      "step": 3550
+    },
+    {
+      "epoch": 2.610927759442611,
+      "grad_norm": 0.82008296251297,
+      "learning_rate": 0.0002,
+      "loss": 0.9253,
+      "step": 3560
+    },
+    {
+      "epoch": 2.6182618261826183,
+      "grad_norm": 0.4724634885787964,
+      "learning_rate": 0.0002,
+      "loss": 0.8429,
+      "step": 3570
+    },
+    {
+      "epoch": 2.6255958929226257,
+      "grad_norm": 0.5434244275093079,
+      "learning_rate": 0.0002,
+      "loss": 0.9058,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6329299596626328,
+      "grad_norm": 0.4948740005493164,
+      "learning_rate": 0.0002,
+      "loss": 0.9379,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6402640264026402,
+      "grad_norm": 0.42109328508377075,
+      "learning_rate": 0.0002,
+      "loss": 0.8718,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6475980931426477,
+      "grad_norm": 0.7979786396026611,
+      "learning_rate": 0.0002,
+      "loss": 0.9809,
+      "step": 3610
+    },
+    {
+      "epoch": 2.654932159882655,
+      "grad_norm": 0.6345919370651245,
+      "learning_rate": 0.0002,
+      "loss": 0.9229,
+      "step": 3620
+    },
+    {
+      "epoch": 2.662266226622662,
+      "grad_norm": 0.4971671402454376,
+      "learning_rate": 0.0002,
+      "loss": 0.8506,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6696002933626697,
+      "grad_norm": 0.6467748284339905,
+      "learning_rate": 0.0002,
+      "loss": 0.8054,
+      "step": 3640
+    },
+    {
+      "epoch": 2.6769343601026767,
+      "grad_norm": 0.4240160286426544,
+      "learning_rate": 0.0002,
+      "loss": 0.9277,
+      "step": 3650
+    },
+    {
+      "epoch": 2.684268426842684,
+      "grad_norm": 0.5179754495620728,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 3660
+    },
+    {
+      "epoch": 2.6916024935826917,
+      "grad_norm": 0.754012405872345,
+      "learning_rate": 0.0002,
+      "loss": 0.9221,
+      "step": 3670
+    },
+    {
+      "epoch": 2.698936560322699,
+      "grad_norm": 0.5141299962997437,
+      "learning_rate": 0.0002,
+      "loss": 0.9194,
+      "step": 3680
+    },
+    {
+      "epoch": 2.706270627062706,
+      "grad_norm": 0.5737819075584412,
+      "learning_rate": 0.0002,
+      "loss": 0.9495,
+      "step": 3690
+    },
+    {
+      "epoch": 2.7136046938027136,
+      "grad_norm": 0.5887577533721924,
+      "learning_rate": 0.0002,
+      "loss": 1.0162,
+      "step": 3700
+    },
+    {
+      "epoch": 2.720938760542721,
+      "grad_norm": 0.6740471720695496,
+      "learning_rate": 0.0002,
+      "loss": 0.9169,
+      "step": 3710
+    },
+    {
+      "epoch": 2.728272827282728,
+      "grad_norm": 0.5879453420639038,
+      "learning_rate": 0.0002,
+      "loss": 0.9297,
+      "step": 3720
+    },
+    {
+      "epoch": 2.7356068940227356,
+      "grad_norm": 0.4858354926109314,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 3730
+    },
+    {
+      "epoch": 2.742940960762743,
+      "grad_norm": 0.5489001870155334,
+      "learning_rate": 0.0002,
+      "loss": 0.9308,
+      "step": 3740
+    },
+    {
+      "epoch": 2.7502750275027505,
+      "grad_norm": 0.8187092542648315,
+      "learning_rate": 0.0002,
+      "loss": 0.894,
+      "step": 3750
+    },
+    {
+      "epoch": 2.7576090942427576,
+      "grad_norm": 0.5666626691818237,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 3760
+    },
+    {
+      "epoch": 2.764943160982765,
+      "grad_norm": 0.5377066135406494,
+      "learning_rate": 0.0002,
+      "loss": 1.0059,
+      "step": 3770
+    },
+    {
+      "epoch": 2.772277227722772,
+      "grad_norm": 0.566330075263977,
+      "learning_rate": 0.0002,
+      "loss": 0.9132,
+      "step": 3780
+    },
+    {
+      "epoch": 2.7796112944627795,
+      "grad_norm": 0.5522832870483398,
+      "learning_rate": 0.0002,
+      "loss": 0.9415,
+      "step": 3790
+    },
+    {
+      "epoch": 2.786945361202787,
+      "grad_norm": 0.5668695569038391,
+      "learning_rate": 0.0002,
+      "loss": 0.8816,
+      "step": 3800
+    },
+    {
+      "epoch": 2.7942794279427945,
+      "grad_norm": 0.7566602826118469,
+      "learning_rate": 0.0002,
+      "loss": 0.8885,
+      "step": 3810
+    },
+    {
+      "epoch": 2.8016134946828015,
+      "grad_norm": 0.5603684782981873,
+      "learning_rate": 0.0002,
+      "loss": 0.8598,
+      "step": 3820
+    },
+    {
+      "epoch": 2.808947561422809,
+      "grad_norm": 0.49122217297554016,
+      "learning_rate": 0.0002,
+      "loss": 0.9602,
+      "step": 3830
+    },
+    {
+      "epoch": 2.816281628162816,
+      "grad_norm": 0.6798251867294312,
+      "learning_rate": 0.0002,
+      "loss": 0.9738,
+      "step": 3840
+    },
+    {
+      "epoch": 2.8236156949028235,
+      "grad_norm": 0.6097991466522217,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 3850
+    },
+    {
+      "epoch": 2.830949761642831,
+      "grad_norm": 0.6675726175308228,
+      "learning_rate": 0.0002,
+      "loss": 0.8672,
+      "step": 3860
+    },
+    {
+      "epoch": 2.8382838283828384,
+      "grad_norm": 0.9223952889442444,
+      "learning_rate": 0.0002,
+      "loss": 0.9324,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8456178951228455,
+      "grad_norm": 0.6020799875259399,
+      "learning_rate": 0.0002,
+      "loss": 0.8767,
+      "step": 3880
+    },
+    {
+      "epoch": 2.852951961862853,
+      "grad_norm": 0.5206381678581238,
+      "learning_rate": 0.0002,
+      "loss": 0.9148,
+      "step": 3890
+    },
+    {
+      "epoch": 2.8602860286028604,
+      "grad_norm": 0.6268777251243591,
+      "learning_rate": 0.0002,
+      "loss": 0.9479,
+      "step": 3900
+    },
+    {
+      "epoch": 2.8676200953428674,
+      "grad_norm": 1.1583497524261475,
+      "learning_rate": 0.0002,
+      "loss": 0.9409,
+      "step": 3910
+    },
+    {
+      "epoch": 2.874954162082875,
+      "grad_norm": 0.7263903021812439,
+      "learning_rate": 0.0002,
+      "loss": 0.895,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8822882288228824,
+      "grad_norm": 0.5369910001754761,
+      "learning_rate": 0.0002,
+      "loss": 0.8786,
+      "step": 3930
+    },
+    {
+      "epoch": 2.88962229556289,
+      "grad_norm": 0.7298350930213928,
+      "learning_rate": 0.0002,
+      "loss": 1.0015,
+      "step": 3940
+    },
+    {
+      "epoch": 2.896956362302897,
+      "grad_norm": 0.577012836933136,
+      "learning_rate": 0.0002,
+      "loss": 0.979,
+      "step": 3950
+    },
+    {
+      "epoch": 2.9042904290429044,
+      "grad_norm": 0.5859594345092773,
+      "learning_rate": 0.0002,
+      "loss": 0.9716,
+      "step": 3960
+    },
+    {
+      "epoch": 2.9116244957829114,
+      "grad_norm": 0.47176122665405273,
+      "learning_rate": 0.0002,
+      "loss": 0.8772,
+      "step": 3970
+    },
+    {
+      "epoch": 2.918958562522919,
+      "grad_norm": 0.9699620604515076,
+      "learning_rate": 0.0002,
+      "loss": 0.8997,
+      "step": 3980
+    },
+    {
+      "epoch": 2.9262926292629263,
+      "grad_norm": 0.7908747792243958,
+      "learning_rate": 0.0002,
+      "loss": 0.9057,
+      "step": 3990
+    },
+    {
+      "epoch": 2.933626696002934,
+      "grad_norm": 0.5777379274368286,
+      "learning_rate": 0.0002,
+      "loss": 0.9462,
+      "step": 4000
+    },
+    {
+      "epoch": 2.940960762742941,
+      "grad_norm": 0.599288284778595,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9482948294829483,
+      "grad_norm": 0.5232274532318115,
+      "learning_rate": 0.0002,
+      "loss": 0.9812,
+      "step": 4020
+    },
+    {
+      "epoch": 2.9556288962229558,
+      "grad_norm": 0.6395137310028076,
+      "learning_rate": 0.0002,
+      "loss": 0.96,
+      "step": 4030
+    },
+    {
+      "epoch": 2.962962962962963,
+      "grad_norm": 0.589260458946228,
+      "learning_rate": 0.0002,
+      "loss": 0.9813,
+      "step": 4040
+    },
+    {
+      "epoch": 2.9702970297029703,
+      "grad_norm": 0.5699581503868103,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 4050
+    },
+    {
+      "epoch": 2.9776310964429777,
+      "grad_norm": 0.528468132019043,
+      "learning_rate": 0.0002,
+      "loss": 0.9585,
+      "step": 4060
+    },
+    {
+      "epoch": 2.984965163182985,
+      "grad_norm": 0.4804670512676239,
+      "learning_rate": 0.0002,
+      "loss": 0.9164,
+      "step": 4070
+    },
+    {
+      "epoch": 2.9922992299229922,
+      "grad_norm": 1.1918889284133911,
+      "learning_rate": 0.0002,
+      "loss": 0.9771,
+      "step": 4080
+    },
+    {
+      "epoch": 2.9996332966629997,
+      "grad_norm": 0.5479103326797485,
+      "learning_rate": 0.0002,
+      "loss": 0.9178,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9996332966629997,
+      "eval_loss": 1.1642853021621704,
+      "eval_runtime": 32.7511,
+      "eval_samples_per_second": 13.16,
+      "eval_steps_per_second": 1.649,
+      "step": 4090
+    },
+    {
+      "epoch": 3.006967363403007,
+      "grad_norm": 0.7430027723312378,
+      "learning_rate": 0.0002,
+      "loss": 0.7981,
+      "step": 4100
+    },
+    {
+      "epoch": 3.014301430143014,
+      "grad_norm": 0.6293647289276123,
+      "learning_rate": 0.0002,
+      "loss": 0.7871,
+      "step": 4110
+    },
+    {
+      "epoch": 3.0216354968830217,
+      "grad_norm": 0.6191329956054688,
+      "learning_rate": 0.0002,
+      "loss": 0.78,
+      "step": 4120
+    },
+    {
+      "epoch": 3.028969563623029,
+      "grad_norm": 0.7959313988685608,
+      "learning_rate": 0.0002,
+      "loss": 0.7618,
+      "step": 4130
+    },
+    {
+      "epoch": 3.036303630363036,
+      "grad_norm": 0.5956351161003113,
+      "learning_rate": 0.0002,
+      "loss": 0.8039,
+      "step": 4140
+    },
+    {
+      "epoch": 3.0436376971030437,
+      "grad_norm": 0.670383632183075,
+      "learning_rate": 0.0002,
+      "loss": 0.7477,
+      "step": 4150
+    },
+    {
+      "epoch": 3.050971763843051,
+      "grad_norm": 0.6414518356323242,
+      "learning_rate": 0.0002,
+      "loss": 0.7984,
+      "step": 4160
+    },
+    {
+      "epoch": 3.058305830583058,
+      "grad_norm": 0.7928852438926697,
+      "learning_rate": 0.0002,
+      "loss": 0.7369,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0656398973230656,
+      "grad_norm": 0.6211121082305908,
+      "learning_rate": 0.0002,
+      "loss": 0.7914,
+      "step": 4180
+    },
+    {
+      "epoch": 3.072973964063073,
+      "grad_norm": 0.6237057447433472,
+      "learning_rate": 0.0002,
+      "loss": 0.7365,
+      "step": 4190
+    },
+    {
+      "epoch": 3.08030803080308,
+      "grad_norm": 0.6522233486175537,
+      "learning_rate": 0.0002,
+      "loss": 0.702,
+      "step": 4200
+    },
+    {
+      "epoch": 3.0876420975430876,
+      "grad_norm": 0.9396848678588867,
+      "learning_rate": 0.0002,
+      "loss": 0.7646,
+      "step": 4210
+    },
+    {
+      "epoch": 3.094976164283095,
+      "grad_norm": 0.8003010749816895,
+      "learning_rate": 0.0002,
+      "loss": 0.7559,
+      "step": 4220
+    },
+    {
+      "epoch": 3.102310231023102,
+      "grad_norm": 0.6733810305595398,
+      "learning_rate": 0.0002,
+      "loss": 0.711,
+      "step": 4230
+    },
+    {
+      "epoch": 3.1096442977631096,
+      "grad_norm": 0.6365828514099121,
+      "learning_rate": 0.0002,
+      "loss": 0.696,
+      "step": 4240
+    },
+    {
+      "epoch": 3.116978364503117,
+      "grad_norm": 1.0805548429489136,
+      "learning_rate": 0.0002,
+      "loss": 0.8362,
+      "step": 4250
+    },
+    {
+      "epoch": 3.1243124312431245,
+      "grad_norm": 0.7262141108512878,
+      "learning_rate": 0.0002,
+      "loss": 0.7651,
+      "step": 4260
+    },
+    {
+      "epoch": 3.1316464979831315,
+      "grad_norm": 0.5500539541244507,
+      "learning_rate": 0.0002,
+      "loss": 0.7304,
+      "step": 4270
+    },
+    {
+      "epoch": 3.138980564723139,
+      "grad_norm": 0.793912947177887,
+      "learning_rate": 0.0002,
+      "loss": 0.7721,
+      "step": 4280
+    },
+    {
+      "epoch": 3.1463146314631465,
+      "grad_norm": 1.2540518045425415,
+      "learning_rate": 0.0002,
+      "loss": 0.7708,
+      "step": 4290
+    },
+    {
+      "epoch": 3.1536486982031535,
+      "grad_norm": 0.7020077705383301,
+      "learning_rate": 0.0002,
+      "loss": 0.782,
+      "step": 4300
+    },
+    {
+      "epoch": 3.160982764943161,
+      "grad_norm": 0.5111123323440552,
+      "learning_rate": 0.0002,
+      "loss": 0.7253,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1683168316831685,
+      "grad_norm": 0.7172090411186218,
+      "learning_rate": 0.0002,
+      "loss": 0.8159,
+      "step": 4320
+    },
+    {
+      "epoch": 3.1756508984231755,
+      "grad_norm": 0.6343168616294861,
+      "learning_rate": 0.0002,
+      "loss": 0.6962,
+      "step": 4330
+    },
+    {
+      "epoch": 3.182984965163183,
+      "grad_norm": 0.9563672542572021,
+      "learning_rate": 0.0002,
+      "loss": 0.7938,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1903190319031904,
+      "grad_norm": 1.0225574970245361,
+      "learning_rate": 0.0002,
+      "loss": 0.7385,
+      "step": 4350
+    },
+    {
+      "epoch": 3.1976530986431975,
+      "grad_norm": 1.1633386611938477,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 4360
+    },
+    {
+      "epoch": 3.204987165383205,
+      "grad_norm": 0.8915148973464966,
+      "learning_rate": 0.0002,
+      "loss": 0.7259,
+      "step": 4370
+    },
+    {
+      "epoch": 3.2123212321232124,
+      "grad_norm": 0.9156812429428101,
+      "learning_rate": 0.0002,
+      "loss": 0.8061,
+      "step": 4380
+    },
+    {
+      "epoch": 3.21965529886322,
+      "grad_norm": 0.6363258957862854,
+      "learning_rate": 0.0002,
+      "loss": 0.8189,
+      "step": 4390
+    },
+    {
+      "epoch": 3.226989365603227,
+      "grad_norm": 0.579099178314209,
+      "learning_rate": 0.0002,
+      "loss": 0.7996,
+      "step": 4400
+    },
+    {
+      "epoch": 3.2343234323432344,
+      "grad_norm": 0.8778146505355835,
+      "learning_rate": 0.0002,
+      "loss": 0.8592,
+      "step": 4410
+    },
+    {
+      "epoch": 3.241657499083242,
+      "grad_norm": 0.8356770873069763,
+      "learning_rate": 0.0002,
+      "loss": 0.8281,
+      "step": 4420
+    },
+    {
+      "epoch": 3.248991565823249,
+      "grad_norm": 0.702032208442688,
+      "learning_rate": 0.0002,
+      "loss": 0.8484,
+      "step": 4430
+    },
+    {
+      "epoch": 3.2563256325632564,
+      "grad_norm": 0.6386539340019226,
+      "learning_rate": 0.0002,
+      "loss": 0.7227,
+      "step": 4440
+    },
+    {
+      "epoch": 3.263659699303264,
+      "grad_norm": 0.7008408904075623,
+      "learning_rate": 0.0002,
+      "loss": 0.8374,
+      "step": 4450
+    },
+    {
+      "epoch": 3.270993766043271,
+      "grad_norm": 0.9556332230567932,
+      "learning_rate": 0.0002,
+      "loss": 0.7572,
+      "step": 4460
+    },
+    {
+      "epoch": 3.2783278327832783,
+      "grad_norm": 0.5667835474014282,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 4470
+    },
+    {
+      "epoch": 3.285661899523286,
+      "grad_norm": 0.8239172697067261,
+      "learning_rate": 0.0002,
+      "loss": 0.8152,
+      "step": 4480
+    },
+    {
+      "epoch": 3.292995966263293,
+      "grad_norm": 0.7045050859451294,
+      "learning_rate": 0.0002,
+      "loss": 0.756,
+      "step": 4490
+    },
+    {
+      "epoch": 3.3003300330033003,
+      "grad_norm": 0.7131434082984924,
+      "learning_rate": 0.0002,
+      "loss": 0.7655,
+      "step": 4500
+    },
+    {
+      "epoch": 3.3076640997433078,
+      "grad_norm": 0.6924910545349121,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 4510
+    },
+    {
+      "epoch": 3.3149981664833152,
+      "grad_norm": 0.8945356607437134,
+      "learning_rate": 0.0002,
+      "loss": 0.736,
+      "step": 4520
+    },
+    {
+      "epoch": 3.3223322332233223,
+      "grad_norm": 0.6546903252601624,
+      "learning_rate": 0.0002,
+      "loss": 0.7575,
+      "step": 4530
+    },
+    {
+      "epoch": 3.3296662999633297,
+      "grad_norm": 0.8206679224967957,
+      "learning_rate": 0.0002,
+      "loss": 0.7893,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3370003667033368,
+      "grad_norm": 0.6482203602790833,
+      "learning_rate": 0.0002,
+      "loss": 0.7502,
+      "step": 4550
+    },
+    {
+      "epoch": 3.3443344334433442,
+      "grad_norm": 0.7558760046958923,
+      "learning_rate": 0.0002,
+      "loss": 0.8172,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3516685001833517,
+      "grad_norm": 0.7794756889343262,
+      "learning_rate": 0.0002,
+      "loss": 0.744,
+      "step": 4570
+    },
+    {
+      "epoch": 3.359002566923359,
+      "grad_norm": 0.7382805943489075,
+      "learning_rate": 0.0002,
+      "loss": 0.7385,
+      "step": 4580
+    },
+    {
+      "epoch": 3.366336633663366,
+      "grad_norm": 0.5912511944770813,
+      "learning_rate": 0.0002,
+      "loss": 0.8511,
+      "step": 4590
+    },
+    {
+      "epoch": 3.3736707004033737,
+      "grad_norm": 0.7444885969161987,
+      "learning_rate": 0.0002,
+      "loss": 0.8272,
+      "step": 4600
+    },
+    {
+      "epoch": 3.381004767143381,
+      "grad_norm": 0.7354922890663147,
+      "learning_rate": 0.0002,
+      "loss": 0.7927,
+      "step": 4610
+    },
+    {
+      "epoch": 3.388338833883388,
+      "grad_norm": 0.7685934901237488,
+      "learning_rate": 0.0002,
+      "loss": 0.7183,
+      "step": 4620
+    },
+    {
+      "epoch": 3.3956729006233957,
+      "grad_norm": 0.61041259765625,
+      "learning_rate": 0.0002,
+      "loss": 0.7436,
+      "step": 4630
+    },
+    {
+      "epoch": 3.403006967363403,
+      "grad_norm": 0.6820451021194458,
+      "learning_rate": 0.0002,
+      "loss": 0.7661,
+      "step": 4640
+    },
+    {
+      "epoch": 3.41034103410341,
+      "grad_norm": 0.5819534063339233,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 4650
+    },
+    {
+      "epoch": 3.4176751008434176,
+      "grad_norm": 0.705410897731781,
+      "learning_rate": 0.0002,
+      "loss": 0.7314,
+      "step": 4660
+    },
+    {
+      "epoch": 3.425009167583425,
+      "grad_norm": 0.8052892088890076,
+      "learning_rate": 0.0002,
+      "loss": 0.7901,
+      "step": 4670
+    },
+    {
+      "epoch": 3.432343234323432,
+      "grad_norm": 0.7746483087539673,
+      "learning_rate": 0.0002,
+      "loss": 0.7298,
+      "step": 4680
+    },
+    {
+      "epoch": 3.4396773010634396,
+      "grad_norm": 0.7713689804077148,
+      "learning_rate": 0.0002,
+      "loss": 0.7976,
+      "step": 4690
+    },
+    {
+      "epoch": 3.447011367803447,
+      "grad_norm": 0.810371994972229,
+      "learning_rate": 0.0002,
+      "loss": 0.7427,
+      "step": 4700
+    },
+    {
+      "epoch": 3.4543454345434546,
+      "grad_norm": 0.7702969312667847,
+      "learning_rate": 0.0002,
+      "loss": 0.7594,
+      "step": 4710
+    },
+    {
+      "epoch": 3.4616795012834616,
+      "grad_norm": 0.7069268822669983,
+      "learning_rate": 0.0002,
+      "loss": 0.7957,
+      "step": 4720
+    },
+    {
+      "epoch": 3.469013568023469,
+      "grad_norm": 0.7640359401702881,
+      "learning_rate": 0.0002,
+      "loss": 0.8199,
+      "step": 4730
+    },
+    {
+      "epoch": 3.4763476347634765,
+      "grad_norm": 0.8661707639694214,
+      "learning_rate": 0.0002,
+      "loss": 0.6875,
+      "step": 4740
+    },
+    {
+      "epoch": 3.4836817015034836,
+      "grad_norm": 0.9970282912254333,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 4750
+    },
+    {
+      "epoch": 3.491015768243491,
+      "grad_norm": 0.5824355483055115,
+      "learning_rate": 0.0002,
+      "loss": 0.8462,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4983498349834985,
+      "grad_norm": 1.3072649240493774,
+      "learning_rate": 0.0002,
+      "loss": 0.851,
+      "step": 4770
+    },
+    {
+      "epoch": 3.5056839017235055,
+      "grad_norm": 0.873978316783905,
+      "learning_rate": 0.0002,
+      "loss": 0.9101,
+      "step": 4780
+    },
+    {
+      "epoch": 3.513017968463513,
+      "grad_norm": 0.5526657104492188,
+      "learning_rate": 0.0002,
+      "loss": 0.7403,
+      "step": 4790
+    },
+    {
+      "epoch": 3.5203520352035205,
+      "grad_norm": 0.790894627571106,
+      "learning_rate": 0.0002,
+      "loss": 0.7921,
+      "step": 4800
+    },
+    {
+      "epoch": 3.5276861019435275,
+      "grad_norm": 0.8119630217552185,
+      "learning_rate": 0.0002,
+      "loss": 0.831,
+      "step": 4810
+    },
+    {
+      "epoch": 3.535020168683535,
+      "grad_norm": 0.633212149143219,
+      "learning_rate": 0.0002,
+      "loss": 0.7351,
+      "step": 4820
+    },
+    {
+      "epoch": 3.5423542354235424,
+      "grad_norm": 0.703029990196228,
+      "learning_rate": 0.0002,
+      "loss": 0.8505,
+      "step": 4830
+    },
+    {
+      "epoch": 3.54968830216355,
+      "grad_norm": 0.7603771686553955,
+      "learning_rate": 0.0002,
+      "loss": 0.7204,
+      "step": 4840
+    },
+    {
+      "epoch": 3.557022368903557,
+      "grad_norm": 0.6260480880737305,
+      "learning_rate": 0.0002,
+      "loss": 0.8868,
+      "step": 4850
+    },
+    {
+      "epoch": 3.5643564356435644,
+      "grad_norm": 0.8203664422035217,
+      "learning_rate": 0.0002,
+      "loss": 0.8137,
+      "step": 4860
+    },
+    {
+      "epoch": 3.5716905023835714,
+      "grad_norm": 0.7793813347816467,
+      "learning_rate": 0.0002,
+      "loss": 0.8821,
+      "step": 4870
+    },
+    {
+      "epoch": 3.579024569123579,
+      "grad_norm": 0.7667397260665894,
+      "learning_rate": 0.0002,
+      "loss": 0.8164,
+      "step": 4880
+    },
+    {
+      "epoch": 3.5863586358635864,
+      "grad_norm": 0.8198829889297485,
+      "learning_rate": 0.0002,
+      "loss": 0.7597,
+      "step": 4890
+    },
+    {
+      "epoch": 3.593692702603594,
+      "grad_norm": 0.7689233422279358,
+      "learning_rate": 0.0002,
+      "loss": 0.7027,
+      "step": 4900
+    },
+    {
+      "epoch": 3.601026769343601,
+      "grad_norm": 0.7870983481407166,
+      "learning_rate": 0.0002,
+      "loss": 0.804,
+      "step": 4910
+    },
+    {
+      "epoch": 3.6083608360836084,
+      "grad_norm": 0.8133853077888489,
+      "learning_rate": 0.0002,
+      "loss": 0.8269,
+      "step": 4920
+    },
+    {
+      "epoch": 3.615694902823616,
+      "grad_norm": 1.308401346206665,
+      "learning_rate": 0.0002,
+      "loss": 0.8515,
+      "step": 4930
+    },
+    {
+      "epoch": 3.623028969563623,
+      "grad_norm": 0.7131121754646301,
+      "learning_rate": 0.0002,
+      "loss": 0.8494,
+      "step": 4940
+    },
+    {
+      "epoch": 3.6303630363036303,
+      "grad_norm": 0.6825910210609436,
+      "learning_rate": 0.0002,
+      "loss": 0.7235,
+      "step": 4950
+    },
+    {
+      "epoch": 3.637697103043638,
+      "grad_norm": 0.7254678606987,
+      "learning_rate": 0.0002,
+      "loss": 0.7824,
+      "step": 4960
+    },
+    {
+      "epoch": 3.6450311697836453,
+      "grad_norm": 0.8045085072517395,
+      "learning_rate": 0.0002,
+      "loss": 0.7983,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6523652365236523,
+      "grad_norm": 0.6991777420043945,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6596993032636598,
+      "grad_norm": 0.7804713249206543,
+      "learning_rate": 0.0002,
+      "loss": 0.7806,
+      "step": 4990
+    },
+    {
+      "epoch": 3.667033370003667,
+      "grad_norm": 0.8525708317756653,
+      "learning_rate": 0.0002,
+      "loss": 0.8402,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6743674367436743,
+      "grad_norm": 0.7959994673728943,
+      "learning_rate": 0.0002,
+      "loss": 0.8496,
+      "step": 5010
+    },
+    {
+      "epoch": 3.6817015034836817,
+      "grad_norm": 0.8103628158569336,
+      "learning_rate": 0.0002,
+      "loss": 0.8022,
+      "step": 5020
+    },
+    {
+      "epoch": 3.689035570223689,
+      "grad_norm": 0.7517836093902588,
+      "learning_rate": 0.0002,
+      "loss": 0.7376,
+      "step": 5030
+    },
+    {
+      "epoch": 3.6963696369636962,
+      "grad_norm": 0.6878514289855957,
+      "learning_rate": 0.0002,
+      "loss": 0.8375,
+      "step": 5040
+    },
+    {
+      "epoch": 3.7037037037037037,
+      "grad_norm": 1.2371820211410522,
+      "learning_rate": 0.0002,
+      "loss": 0.7998,
+      "step": 5050
+    },
+    {
+      "epoch": 3.711037770443711,
+      "grad_norm": 0.6567103862762451,
+      "learning_rate": 0.0002,
+      "loss": 0.6941,
+      "step": 5060
+    },
+    {
+      "epoch": 3.718371837183718,
+      "grad_norm": 1.1254922151565552,
+      "learning_rate": 0.0002,
+      "loss": 0.8465,
+      "step": 5070
+    },
+    {
+      "epoch": 3.7257059039237257,
+      "grad_norm": 0.6796132326126099,
+      "learning_rate": 0.0002,
+      "loss": 0.8365,
+      "step": 5080
+    },
+    {
+      "epoch": 3.733039970663733,
+      "grad_norm": 0.7285300493240356,
+      "learning_rate": 0.0002,
+      "loss": 0.7818,
+      "step": 5090
+    },
+    {
+      "epoch": 3.7403740374037406,
+      "grad_norm": 0.8931500911712646,
+      "learning_rate": 0.0002,
+      "loss": 0.8581,
+      "step": 5100
+    },
+    {
+      "epoch": 3.7477081041437477,
+      "grad_norm": 0.6256856918334961,
+      "learning_rate": 0.0002,
+      "loss": 0.8181,
+      "step": 5110
+    },
+    {
+      "epoch": 3.755042170883755,
+      "grad_norm": 0.79310142993927,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 5120
+    },
+    {
+      "epoch": 3.762376237623762,
+      "grad_norm": 0.6594041585922241,
+      "learning_rate": 0.0002,
+      "loss": 0.8235,
+      "step": 5130
+    },
+    {
+      "epoch": 3.7697103043637696,
+      "grad_norm": 0.7029327750205994,
+      "learning_rate": 0.0002,
+      "loss": 0.6925,
+      "step": 5140
+    },
+    {
+      "epoch": 3.777044371103777,
+      "grad_norm": 0.5880070328712463,
+      "learning_rate": 0.0002,
+      "loss": 0.7457,
+      "step": 5150
+    },
+    {
+      "epoch": 3.7843784378437846,
+      "grad_norm": 0.7578945159912109,
+      "learning_rate": 0.0002,
+      "loss": 0.8716,
+      "step": 5160
+    },
+    {
+      "epoch": 3.7917125045837916,
+      "grad_norm": 0.8276378512382507,
+      "learning_rate": 0.0002,
+      "loss": 0.8819,
+      "step": 5170
+    },
+    {
+      "epoch": 3.799046571323799,
+      "grad_norm": 0.7627953886985779,
+      "learning_rate": 0.0002,
+      "loss": 0.7559,
+      "step": 5180
+    },
+    {
+      "epoch": 3.806380638063806,
+      "grad_norm": 0.8169086575508118,
+      "learning_rate": 0.0002,
+      "loss": 0.7665,
+      "step": 5190
+    },
+    {
+      "epoch": 3.8137147048038136,
+      "grad_norm": 0.6605030298233032,
+      "learning_rate": 0.0002,
+      "loss": 0.761,
+      "step": 5200
+    },
+    {
+      "epoch": 3.821048771543821,
+      "grad_norm": 0.5837286114692688,
+      "learning_rate": 0.0002,
+      "loss": 0.8804,
+      "step": 5210
+    },
+    {
+      "epoch": 3.8283828382838285,
+      "grad_norm": 1.2422157526016235,
+      "learning_rate": 0.0002,
+      "loss": 0.8369,
+      "step": 5220
+    },
+    {
+      "epoch": 3.8357169050238356,
+      "grad_norm": 0.6589220762252808,
+      "learning_rate": 0.0002,
+      "loss": 0.8431,
+      "step": 5230
+    },
+    {
+      "epoch": 3.843050971763843,
+      "grad_norm": 0.8567556142807007,
+      "learning_rate": 0.0002,
+      "loss": 0.7686,
+      "step": 5240
+    },
+    {
+      "epoch": 3.8503850385038505,
+      "grad_norm": 0.6490627527236938,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8577191052438575,
+      "grad_norm": 0.620232880115509,
+      "learning_rate": 0.0002,
+      "loss": 0.7386,
+      "step": 5260
+    },
+    {
+      "epoch": 3.865053171983865,
+      "grad_norm": 0.7685128450393677,
+      "learning_rate": 0.0002,
+      "loss": 0.9192,
+      "step": 5270
+    },
+    {
+      "epoch": 3.8723872387238725,
+      "grad_norm": 0.8113296627998352,
+      "learning_rate": 0.0002,
+      "loss": 0.872,
+      "step": 5280
+    },
+    {
+      "epoch": 3.87972130546388,
+      "grad_norm": 0.8092675805091858,
+      "learning_rate": 0.0002,
+      "loss": 0.7156,
+      "step": 5290
+    },
+    {
+      "epoch": 3.887055372203887,
+      "grad_norm": 0.583570122718811,
+      "learning_rate": 0.0002,
+      "loss": 0.7325,
+      "step": 5300
+    },
+    {
+      "epoch": 3.8943894389438944,
+      "grad_norm": 1.712363600730896,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 5310
+    },
+    {
+      "epoch": 3.9017235056839015,
+      "grad_norm": 0.6673534512519836,
+      "learning_rate": 0.0002,
+      "loss": 0.7537,
+      "step": 5320
+    },
+    {
+      "epoch": 3.909057572423909,
+      "grad_norm": 1.9770312309265137,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 5330
+    },
+    {
+      "epoch": 3.9163916391639164,
+      "grad_norm": 0.6430999636650085,
+      "learning_rate": 0.0002,
+      "loss": 0.8793,
+      "step": 5340
+    },
+    {
+      "epoch": 3.923725705903924,
+      "grad_norm": 1.0159571170806885,
+      "learning_rate": 0.0002,
+      "loss": 0.839,
+      "step": 5350
+    },
+    {
+      "epoch": 3.931059772643931,
+      "grad_norm": 0.8607584834098816,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 5360
+    },
+    {
+      "epoch": 3.9383938393839384,
+      "grad_norm": 0.6967900991439819,
+      "learning_rate": 0.0002,
+      "loss": 0.7261,
+      "step": 5370
+    },
+    {
+      "epoch": 3.945727906123946,
+      "grad_norm": 0.7683077454566956,
+      "learning_rate": 0.0002,
+      "loss": 0.8456,
+      "step": 5380
+    },
+    {
+      "epoch": 3.953061972863953,
+      "grad_norm": 0.6805762648582458,
+      "learning_rate": 0.0002,
+      "loss": 0.7682,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9603960396039604,
+      "grad_norm": 0.7033619284629822,
+      "learning_rate": 0.0002,
+      "loss": 0.7746,
+      "step": 5400
+    },
+    {
+      "epoch": 3.967730106343968,
+      "grad_norm": 0.966112494468689,
+      "learning_rate": 0.0002,
+      "loss": 0.8393,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9750641730839753,
+      "grad_norm": 0.8467881083488464,
+      "learning_rate": 0.0002,
+      "loss": 0.8316,
+      "step": 5420
+    },
+    {
+      "epoch": 3.9823982398239823,
+      "grad_norm": 0.8005317449569702,
+      "learning_rate": 0.0002,
+      "loss": 0.8084,
+      "step": 5430
+    },
+    {
+      "epoch": 3.98973230656399,
+      "grad_norm": 1.1615241765975952,
+      "learning_rate": 0.0002,
+      "loss": 0.7168,
+      "step": 5440
+    },
+    {
+      "epoch": 3.997066373303997,
+      "grad_norm": 0.6121614575386047,
+      "learning_rate": 0.0002,
+      "loss": 0.8263,
+      "step": 5450
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.1834222078323364,
+      "eval_runtime": 32.7569,
+      "eval_samples_per_second": 13.158,
+      "eval_steps_per_second": 1.649,
+      "step": 5454
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10904,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "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": 2.5239890799938765e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..2868cff7027115396e695775cacd838522aca295
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-5454/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b12b6f6817632087b5a5e37d744e25312b96e839de5005320b96bc0c2473c41f
+size 5624
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/README.md b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# 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 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "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": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5bb16ab95ca096920c5d5b7350054e10507d7989
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e4f634851d34aa044ae96f13caf049e805ff844286cc74d8fab46081770f1af
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f4dcb5cad10181ec10fa32fd6b3c7466201c6eda
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44ef702ec102cc98397673136cc4e894d33991fd0940b046e486268bfa17deae
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..8475fa8f8b545c12e801def539ab1fd87bbece8c
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6c5c3e36480b97449f54a91b6bb67c1ce97ff2182fabcb54951279599efded4
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b73922fca4c14f7bb64a94872248c0d53c64160f
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a2fccf9ed41af18b1ff88d3cd93d214635edb273f565215977f96017eaa5a6f
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..52f9aa15b7a2d8c6eee09b65ea501dc7b26b6e98
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/trainer_state.json
@@ -0,0 +1,4840 @@
+{
+  "best_metric": 1.1534006595611572,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727",
+  "epoch": 4.999633296662999,
+  "eval_steps": 10,
+  "global_step": 6817,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007334066740007334,
+      "grad_norm": 0.47521963715553284,
+      "learning_rate": 0.0002,
+      "loss": 1.9722,
+      "step": 10
+    },
+    {
+      "epoch": 0.014668133480014669,
+      "grad_norm": 0.5395162105560303,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 20
+    },
+    {
+      "epoch": 0.022002200220022004,
+      "grad_norm": 0.4305780231952667,
+      "learning_rate": 0.0002,
+      "loss": 1.4202,
+      "step": 30
+    },
+    {
+      "epoch": 0.029336266960029337,
+      "grad_norm": 0.6938246488571167,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 40
+    },
+    {
+      "epoch": 0.03667033370003667,
+      "grad_norm": 1.5133819580078125,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 50
+    },
+    {
+      "epoch": 0.04400440044004401,
+      "grad_norm": 0.9173883199691772,
+      "learning_rate": 0.0002,
+      "loss": 1.3132,
+      "step": 60
+    },
+    {
+      "epoch": 0.05133846718005134,
+      "grad_norm": 0.4619861841201782,
+      "learning_rate": 0.0002,
+      "loss": 1.2844,
+      "step": 70
+    },
+    {
+      "epoch": 0.058672533920058674,
+      "grad_norm": 0.46118637919425964,
+      "learning_rate": 0.0002,
+      "loss": 1.2108,
+      "step": 80
+    },
+    {
+      "epoch": 0.066006600660066,
+      "grad_norm": 0.4468648135662079,
+      "learning_rate": 0.0002,
+      "loss": 1.3441,
+      "step": 90
+    },
+    {
+      "epoch": 0.07334066740007333,
+      "grad_norm": 0.46123769879341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1863,
+      "step": 100
+    },
+    {
+      "epoch": 0.08067473414008068,
+      "grad_norm": 0.4859139025211334,
+      "learning_rate": 0.0002,
+      "loss": 1.2772,
+      "step": 110
+    },
+    {
+      "epoch": 0.08800880088008801,
+      "grad_norm": 0.4384922385215759,
+      "learning_rate": 0.0002,
+      "loss": 1.2087,
+      "step": 120
+    },
+    {
+      "epoch": 0.09534286762009535,
+      "grad_norm": 0.39519360661506653,
+      "learning_rate": 0.0002,
+      "loss": 1.2927,
+      "step": 130
+    },
+    {
+      "epoch": 0.10267693436010268,
+      "grad_norm": 0.4049859344959259,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 140
+    },
+    {
+      "epoch": 0.11001100110011001,
+      "grad_norm": 0.4605638086795807,
+      "learning_rate": 0.0002,
+      "loss": 1.293,
+      "step": 150
+    },
+    {
+      "epoch": 0.11734506784011735,
+      "grad_norm": 0.4201928377151489,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 160
+    },
+    {
+      "epoch": 0.12467913458012468,
+      "grad_norm": 0.5367777347564697,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 170
+    },
+    {
+      "epoch": 0.132013201320132,
+      "grad_norm": 0.41752299666404724,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 180
+    },
+    {
+      "epoch": 0.13934726806013933,
+      "grad_norm": 0.31597763299942017,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 190
+    },
+    {
+      "epoch": 0.14668133480014667,
+      "grad_norm": 0.7468788623809814,
+      "learning_rate": 0.0002,
+      "loss": 1.2441,
+      "step": 200
+    },
+    {
+      "epoch": 0.15401540154015403,
+      "grad_norm": 0.3403034508228302,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 210
+    },
+    {
+      "epoch": 0.16134946828016136,
+      "grad_norm": 0.34240293502807617,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 220
+    },
+    {
+      "epoch": 0.1686835350201687,
+      "grad_norm": 0.356158971786499,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 230
+    },
+    {
+      "epoch": 0.17601760176017603,
+      "grad_norm": 0.3448857367038727,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 240
+    },
+    {
+      "epoch": 0.18335166850018336,
+      "grad_norm": 0.3475699722766876,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 250
+    },
+    {
+      "epoch": 0.1906857352401907,
+      "grad_norm": 0.2770358622074127,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 260
+    },
+    {
+      "epoch": 0.19801980198019803,
+      "grad_norm": 0.4310270845890045,
+      "learning_rate": 0.0002,
+      "loss": 1.2238,
+      "step": 270
+    },
+    {
+      "epoch": 0.20535386872020536,
+      "grad_norm": 0.335041880607605,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 280
+    },
+    {
+      "epoch": 0.2126879354602127,
+      "grad_norm": 0.3420602083206177,
+      "learning_rate": 0.0002,
+      "loss": 1.0959,
+      "step": 290
+    },
+    {
+      "epoch": 0.22002200220022003,
+      "grad_norm": 0.325001060962677,
+      "learning_rate": 0.0002,
+      "loss": 1.1232,
+      "step": 300
+    },
+    {
+      "epoch": 0.22735606894022736,
+      "grad_norm": 0.3027827739715576,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 310
+    },
+    {
+      "epoch": 0.2346901356802347,
+      "grad_norm": 0.435550719499588,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 320
+    },
+    {
+      "epoch": 0.24202420242024203,
+      "grad_norm": 0.3884522616863251,
+      "learning_rate": 0.0002,
+      "loss": 1.2045,
+      "step": 330
+    },
+    {
+      "epoch": 0.24935826916024936,
+      "grad_norm": 0.7736002206802368,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 340
+    },
+    {
+      "epoch": 0.2566923359002567,
+      "grad_norm": 0.35052821040153503,
+      "learning_rate": 0.0002,
+      "loss": 1.3606,
+      "step": 350
+    },
+    {
+      "epoch": 0.264026402640264,
+      "grad_norm": 0.3311890959739685,
+      "learning_rate": 0.0002,
+      "loss": 1.2129,
+      "step": 360
+    },
+    {
+      "epoch": 0.27136046938027136,
+      "grad_norm": 0.7473500370979309,
+      "learning_rate": 0.0002,
+      "loss": 1.2219,
+      "step": 370
+    },
+    {
+      "epoch": 0.27869453612027867,
+      "grad_norm": 0.3681875765323639,
+      "learning_rate": 0.0002,
+      "loss": 1.2712,
+      "step": 380
+    },
+    {
+      "epoch": 0.28602860286028603,
+      "grad_norm": 0.3764737844467163,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 390
+    },
+    {
+      "epoch": 0.29336266960029334,
+      "grad_norm": 0.4243989586830139,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 400
+    },
+    {
+      "epoch": 0.3006967363403007,
+      "grad_norm": 0.2658531963825226,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 410
+    },
+    {
+      "epoch": 0.30803080308030806,
+      "grad_norm": 0.3436793386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.1622,
+      "step": 420
+    },
+    {
+      "epoch": 0.31536486982031536,
+      "grad_norm": 0.5101129412651062,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 430
+    },
+    {
+      "epoch": 0.3226989365603227,
+      "grad_norm": 0.3319750726222992,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 440
+    },
+    {
+      "epoch": 0.33003300330033003,
+      "grad_norm": 0.385148286819458,
+      "learning_rate": 0.0002,
+      "loss": 1.1804,
+      "step": 450
+    },
+    {
+      "epoch": 0.3373670700403374,
+      "grad_norm": 0.3477935791015625,
+      "learning_rate": 0.0002,
+      "loss": 1.1808,
+      "step": 460
+    },
+    {
+      "epoch": 0.3447011367803447,
+      "grad_norm": 0.29748716950416565,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 470
+    },
+    {
+      "epoch": 0.35203520352035206,
+      "grad_norm": 0.34083324670791626,
+      "learning_rate": 0.0002,
+      "loss": 1.19,
+      "step": 480
+    },
+    {
+      "epoch": 0.35936927026035936,
+      "grad_norm": 0.36904552578926086,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 490
+    },
+    {
+      "epoch": 0.3667033370003667,
+      "grad_norm": 0.315483033657074,
+      "learning_rate": 0.0002,
+      "loss": 1.2223,
+      "step": 500
+    },
+    {
+      "epoch": 0.37403740374037403,
+      "grad_norm": 0.44897955656051636,
+      "learning_rate": 0.0002,
+      "loss": 1.1461,
+      "step": 510
+    },
+    {
+      "epoch": 0.3813714704803814,
+      "grad_norm": 0.3160701394081116,
+      "learning_rate": 0.0002,
+      "loss": 1.3035,
+      "step": 520
+    },
+    {
+      "epoch": 0.3887055372203887,
+      "grad_norm": 0.29584741592407227,
+      "learning_rate": 0.0002,
+      "loss": 1.3197,
+      "step": 530
+    },
+    {
+      "epoch": 0.39603960396039606,
+      "grad_norm": 0.5430002808570862,
+      "learning_rate": 0.0002,
+      "loss": 1.2983,
+      "step": 540
+    },
+    {
+      "epoch": 0.40337367070040336,
+      "grad_norm": 0.2908070683479309,
+      "learning_rate": 0.0002,
+      "loss": 1.2459,
+      "step": 550
+    },
+    {
+      "epoch": 0.4107077374404107,
+      "grad_norm": 0.35066530108451843,
+      "learning_rate": 0.0002,
+      "loss": 1.2384,
+      "step": 560
+    },
+    {
+      "epoch": 0.41804180418041803,
+      "grad_norm": 0.37588003277778625,
+      "learning_rate": 0.0002,
+      "loss": 1.1784,
+      "step": 570
+    },
+    {
+      "epoch": 0.4253758709204254,
+      "grad_norm": 0.3112126886844635,
+      "learning_rate": 0.0002,
+      "loss": 1.2334,
+      "step": 580
+    },
+    {
+      "epoch": 0.4327099376604327,
+      "grad_norm": 0.35577139258384705,
+      "learning_rate": 0.0002,
+      "loss": 1.1439,
+      "step": 590
+    },
+    {
+      "epoch": 0.44004400440044006,
+      "grad_norm": 0.31706422567367554,
+      "learning_rate": 0.0002,
+      "loss": 1.184,
+      "step": 600
+    },
+    {
+      "epoch": 0.44737807114044736,
+      "grad_norm": 0.3249092102050781,
+      "learning_rate": 0.0002,
+      "loss": 1.2081,
+      "step": 610
+    },
+    {
+      "epoch": 0.4547121378804547,
+      "grad_norm": 0.3842705488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.0824,
+      "step": 620
+    },
+    {
+      "epoch": 0.46204620462046203,
+      "grad_norm": 0.390991747379303,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 630
+    },
+    {
+      "epoch": 0.4693802713604694,
+      "grad_norm": 0.27532413601875305,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 640
+    },
+    {
+      "epoch": 0.4767143381004767,
+      "grad_norm": 0.31412816047668457,
+      "learning_rate": 0.0002,
+      "loss": 1.1058,
+      "step": 650
+    },
+    {
+      "epoch": 0.48404840484048406,
+      "grad_norm": 0.32117101550102234,
+      "learning_rate": 0.0002,
+      "loss": 1.1312,
+      "step": 660
+    },
+    {
+      "epoch": 0.49138247158049136,
+      "grad_norm": 0.3810010254383087,
+      "learning_rate": 0.0002,
+      "loss": 1.2423,
+      "step": 670
+    },
+    {
+      "epoch": 0.4987165383204987,
+      "grad_norm": 0.36289164423942566,
+      "learning_rate": 0.0002,
+      "loss": 1.1978,
+      "step": 680
+    },
+    {
+      "epoch": 0.506050605060506,
+      "grad_norm": 0.34458720684051514,
+      "learning_rate": 0.0002,
+      "loss": 1.2034,
+      "step": 690
+    },
+    {
+      "epoch": 0.5133846718005134,
+      "grad_norm": 0.32844600081443787,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 700
+    },
+    {
+      "epoch": 0.5207187385405208,
+      "grad_norm": 0.3144175708293915,
+      "learning_rate": 0.0002,
+      "loss": 1.0807,
+      "step": 710
+    },
+    {
+      "epoch": 0.528052805280528,
+      "grad_norm": 0.3898887634277344,
+      "learning_rate": 0.0002,
+      "loss": 1.1952,
+      "step": 720
+    },
+    {
+      "epoch": 0.5353868720205354,
+      "grad_norm": 1.3220758438110352,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 730
+    },
+    {
+      "epoch": 0.5427209387605427,
+      "grad_norm": 0.3635874390602112,
+      "learning_rate": 0.0002,
+      "loss": 1.227,
+      "step": 740
+    },
+    {
+      "epoch": 0.5500550055005501,
+      "grad_norm": 0.3138217628002167,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 750
+    },
+    {
+      "epoch": 0.5573890722405573,
+      "grad_norm": 0.4063207805156708,
+      "learning_rate": 0.0002,
+      "loss": 1.1516,
+      "step": 760
+    },
+    {
+      "epoch": 0.5647231389805647,
+      "grad_norm": 0.3926219940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 770
+    },
+    {
+      "epoch": 0.5720572057205721,
+      "grad_norm": 0.31954652070999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1726,
+      "step": 780
+    },
+    {
+      "epoch": 0.5793912724605794,
+      "grad_norm": 0.4248711168766022,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 790
+    },
+    {
+      "epoch": 0.5867253392005867,
+      "grad_norm": 0.643004834651947,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 800
+    },
+    {
+      "epoch": 0.594059405940594,
+      "grad_norm": 0.3479592800140381,
+      "learning_rate": 0.0002,
+      "loss": 1.1793,
+      "step": 810
+    },
+    {
+      "epoch": 0.6013934726806014,
+      "grad_norm": 0.4684754014015198,
+      "learning_rate": 0.0002,
+      "loss": 1.2426,
+      "step": 820
+    },
+    {
+      "epoch": 0.6087275394206088,
+      "grad_norm": 0.3739790916442871,
+      "learning_rate": 0.0002,
+      "loss": 1.2002,
+      "step": 830
+    },
+    {
+      "epoch": 0.6160616061606161,
+      "grad_norm": 0.40884748101234436,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 840
+    },
+    {
+      "epoch": 0.6233956729006234,
+      "grad_norm": 0.9722164273262024,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 850
+    },
+    {
+      "epoch": 0.6307297396406307,
+      "grad_norm": 0.42992347478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.3069,
+      "step": 860
+    },
+    {
+      "epoch": 0.6380638063806381,
+      "grad_norm": 0.36654195189476013,
+      "learning_rate": 0.0002,
+      "loss": 1.1339,
+      "step": 870
+    },
+    {
+      "epoch": 0.6453978731206454,
+      "grad_norm": 0.4113832116127014,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 880
+    },
+    {
+      "epoch": 0.6527319398606527,
+      "grad_norm": 0.2948838770389557,
+      "learning_rate": 0.0002,
+      "loss": 1.2163,
+      "step": 890
+    },
+    {
+      "epoch": 0.6600660066006601,
+      "grad_norm": 0.38330280780792236,
+      "learning_rate": 0.0002,
+      "loss": 1.1081,
+      "step": 900
+    },
+    {
+      "epoch": 0.6674000733406674,
+      "grad_norm": 0.4428867697715759,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 910
+    },
+    {
+      "epoch": 0.6747341400806748,
+      "grad_norm": 0.23659265041351318,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 920
+    },
+    {
+      "epoch": 0.682068206820682,
+      "grad_norm": 0.323685884475708,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 930
+    },
+    {
+      "epoch": 0.6894022735606894,
+      "grad_norm": 0.39157727360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 940
+    },
+    {
+      "epoch": 0.6967363403006968,
+      "grad_norm": 0.27189481258392334,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 950
+    },
+    {
+      "epoch": 0.7040704070407041,
+      "grad_norm": 0.529883861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.1033,
+      "step": 960
+    },
+    {
+      "epoch": 0.7114044737807114,
+      "grad_norm": 0.34758689999580383,
+      "learning_rate": 0.0002,
+      "loss": 1.139,
+      "step": 970
+    },
+    {
+      "epoch": 0.7187385405207187,
+      "grad_norm": 0.831749439239502,
+      "learning_rate": 0.0002,
+      "loss": 1.2197,
+      "step": 980
+    },
+    {
+      "epoch": 0.7260726072607261,
+      "grad_norm": 0.4438304007053375,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 990
+    },
+    {
+      "epoch": 0.7334066740007334,
+      "grad_norm": 0.33840006589889526,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7407407407407407,
+      "grad_norm": 0.3454797863960266,
+      "learning_rate": 0.0002,
+      "loss": 1.254,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7480748074807481,
+      "grad_norm": 0.38999441266059875,
+      "learning_rate": 0.0002,
+      "loss": 1.106,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7554088742207554,
+      "grad_norm": 0.2829911708831787,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7627429409607628,
+      "grad_norm": 0.36918163299560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 1040
+    },
+    {
+      "epoch": 0.77007700770077,
+      "grad_norm": 0.3415680229663849,
+      "learning_rate": 0.0002,
+      "loss": 1.3028,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7774110744407774,
+      "grad_norm": 0.2974182963371277,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7847451411807848,
+      "grad_norm": 0.3880919814109802,
+      "learning_rate": 0.0002,
+      "loss": 1.194,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7920792079207921,
+      "grad_norm": 0.33503302931785583,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7994132746607994,
+      "grad_norm": 0.3728407025337219,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8067473414008067,
+      "grad_norm": 0.3509373664855957,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8140814081408141,
+      "grad_norm": 0.42228564620018005,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8214154748808215,
+      "grad_norm": 0.313467800617218,
+      "learning_rate": 0.0002,
+      "loss": 1.1788,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8287495416208287,
+      "grad_norm": 0.3378850817680359,
+      "learning_rate": 0.0002,
+      "loss": 1.1971,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8360836083608361,
+      "grad_norm": 0.43200382590293884,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8434176751008434,
+      "grad_norm": 0.3309599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.3203,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8507517418408508,
+      "grad_norm": 0.3526846170425415,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1160
+    },
+    {
+      "epoch": 0.858085808580858,
+      "grad_norm": 1.2722247838974,
+      "learning_rate": 0.0002,
+      "loss": 1.0851,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8654198753208654,
+      "grad_norm": 0.34142059087753296,
+      "learning_rate": 0.0002,
+      "loss": 1.0785,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8727539420608728,
+      "grad_norm": 0.3805823028087616,
+      "learning_rate": 0.0002,
+      "loss": 1.2187,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8800880088008801,
+      "grad_norm": 0.3931232690811157,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8874220755408874,
+      "grad_norm": 0.2937372624874115,
+      "learning_rate": 0.0002,
+      "loss": 1.0948,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8947561422808947,
+      "grad_norm": 0.3757196366786957,
+      "learning_rate": 0.0002,
+      "loss": 1.1228,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9020902090209021,
+      "grad_norm": 0.3502705991268158,
+      "learning_rate": 0.0002,
+      "loss": 1.1222,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9094242757609095,
+      "grad_norm": 0.32758915424346924,
+      "learning_rate": 0.0002,
+      "loss": 1.2242,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9167583425009168,
+      "grad_norm": 0.37199416756629944,
+      "learning_rate": 0.0002,
+      "loss": 1.215,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9240924092409241,
+      "grad_norm": 0.3551490604877472,
+      "learning_rate": 0.0002,
+      "loss": 1.1225,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9314264759809314,
+      "grad_norm": 0.2859550714492798,
+      "learning_rate": 0.0002,
+      "loss": 1.1966,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9387605427209388,
+      "grad_norm": 0.427990585565567,
+      "learning_rate": 0.0002,
+      "loss": 1.2186,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9460946094609461,
+      "grad_norm": 0.33717992901802063,
+      "learning_rate": 0.0002,
+      "loss": 1.2848,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9534286762009534,
+      "grad_norm": 0.30225634574890137,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9607627429409608,
+      "grad_norm": 0.385821133852005,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9680968096809681,
+      "grad_norm": 0.35278066992759705,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9754308764209755,
+      "grad_norm": 0.49987098574638367,
+      "learning_rate": 0.0002,
+      "loss": 1.1071,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9827649431609827,
+      "grad_norm": 0.3842747211456299,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9900990099009901,
+      "grad_norm": 0.6274653673171997,
+      "learning_rate": 0.0002,
+      "loss": 1.0862,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9974330766409975,
+      "grad_norm": 0.5239808559417725,
+      "learning_rate": 0.0002,
+      "loss": 1.124,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9996332966629996,
+      "eval_loss": 1.1822267770767212,
+      "eval_runtime": 32.7389,
+      "eval_samples_per_second": 13.165,
+      "eval_steps_per_second": 1.649,
+      "step": 1363
+    },
+    {
+      "epoch": 1.0047671433810048,
+      "grad_norm": 0.45311301946640015,
+      "learning_rate": 0.0002,
+      "loss": 1.096,
+      "step": 1370
+    },
+    {
+      "epoch": 1.012101210121012,
+      "grad_norm": 0.29685574769973755,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0194352768610195,
+      "grad_norm": 0.3290937840938568,
+      "learning_rate": 0.0002,
+      "loss": 1.0302,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0267693436010268,
+      "grad_norm": 0.3801758587360382,
+      "learning_rate": 0.0002,
+      "loss": 1.0295,
+      "step": 1400
+    },
+    {
+      "epoch": 1.034103410341034,
+      "grad_norm": 0.794174313545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0414374770810415,
+      "grad_norm": 0.3854154646396637,
+      "learning_rate": 0.0002,
+      "loss": 1.2232,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0487715438210488,
+      "grad_norm": 0.32702451944351196,
+      "learning_rate": 0.0002,
+      "loss": 1.0652,
+      "step": 1430
+    },
+    {
+      "epoch": 1.056105610561056,
+      "grad_norm": 0.7815203666687012,
+      "learning_rate": 0.0002,
+      "loss": 1.1144,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0634396773010635,
+      "grad_norm": 0.3087436854839325,
+      "learning_rate": 0.0002,
+      "loss": 1.1316,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0707737440410707,
+      "grad_norm": 0.3847602903842926,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0781078107810782,
+      "grad_norm": 0.3693031370639801,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0854418775210855,
+      "grad_norm": 0.4111202359199524,
+      "learning_rate": 0.0002,
+      "loss": 1.0995,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0927759442610927,
+      "grad_norm": 0.41452381014823914,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1001100110011002,
+      "grad_norm": 0.3336445093154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1068,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1074440777411074,
+      "grad_norm": 0.3923407793045044,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1147781444811147,
+      "grad_norm": 0.46215683221817017,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1221122112211221,
+      "grad_norm": 0.3592156767845154,
+      "learning_rate": 0.0002,
+      "loss": 1.1133,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1294462779611294,
+      "grad_norm": 0.361110657453537,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1367803447011369,
+      "grad_norm": 0.5317131280899048,
+      "learning_rate": 0.0002,
+      "loss": 1.1553,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1441144114411441,
+      "grad_norm": 0.3882388174533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1514484781811514,
+      "grad_norm": 0.3259428143501282,
+      "learning_rate": 0.0002,
+      "loss": 1.0805,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1587825449211588,
+      "grad_norm": 0.410935640335083,
+      "learning_rate": 0.0002,
+      "loss": 1.1819,
+      "step": 1580
+    },
+    {
+      "epoch": 1.166116611661166,
+      "grad_norm": 0.44940185546875,
+      "learning_rate": 0.0002,
+      "loss": 1.1143,
+      "step": 1590
+    },
+    {
+      "epoch": 1.1734506784011733,
+      "grad_norm": 0.5106484293937683,
+      "learning_rate": 0.0002,
+      "loss": 1.0334,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1807847451411808,
+      "grad_norm": 0.6603665947914124,
+      "learning_rate": 0.0002,
+      "loss": 1.2376,
+      "step": 1610
+    },
+    {
+      "epoch": 1.188118811881188,
+      "grad_norm": 0.4799964129924774,
+      "learning_rate": 0.0002,
+      "loss": 1.1227,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1954528786211955,
+      "grad_norm": 0.4389883279800415,
+      "learning_rate": 0.0002,
+      "loss": 1.1191,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2027869453612028,
+      "grad_norm": 0.4188813269138336,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 1640
+    },
+    {
+      "epoch": 1.21012101210121,
+      "grad_norm": 0.7132157683372498,
+      "learning_rate": 0.0002,
+      "loss": 1.0605,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2174550788412175,
+      "grad_norm": 0.507480263710022,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2247891455812248,
+      "grad_norm": 0.9452332854270935,
+      "learning_rate": 0.0002,
+      "loss": 0.9948,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2321232123212322,
+      "grad_norm": 0.4121614992618561,
+      "learning_rate": 0.0002,
+      "loss": 1.0228,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2394572790612395,
+      "grad_norm": 0.34230247139930725,
+      "learning_rate": 0.0002,
+      "loss": 1.0366,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2467913458012467,
+      "grad_norm": 0.4026208817958832,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2541254125412542,
+      "grad_norm": 0.46673697233200073,
+      "learning_rate": 0.0002,
+      "loss": 1.0206,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2614594792812615,
+      "grad_norm": 0.38349825143814087,
+      "learning_rate": 0.0002,
+      "loss": 1.0827,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2687935460212687,
+      "grad_norm": 0.4049997627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.0356,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2761276127612762,
+      "grad_norm": 0.3417615294456482,
+      "learning_rate": 0.0002,
+      "loss": 0.9504,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2834616795012834,
+      "grad_norm": 0.4277614951133728,
+      "learning_rate": 0.0002,
+      "loss": 1.094,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2907957462412907,
+      "grad_norm": 0.5864202976226807,
+      "learning_rate": 0.0002,
+      "loss": 0.9938,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2981298129812981,
+      "grad_norm": 0.7097493410110474,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3054638797213054,
+      "grad_norm": 0.3145381212234497,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3127979464613129,
+      "grad_norm": 0.5116165280342102,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3201320132013201,
+      "grad_norm": 0.7469736337661743,
+      "learning_rate": 0.0002,
+      "loss": 1.0765,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3274660799413276,
+      "grad_norm": 0.32272255420684814,
+      "learning_rate": 0.0002,
+      "loss": 1.0663,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3348001466813348,
+      "grad_norm": 0.3534623086452484,
+      "learning_rate": 0.0002,
+      "loss": 0.9887,
+      "step": 1820
+    },
+    {
+      "epoch": 1.342134213421342,
+      "grad_norm": 0.36127907037734985,
+      "learning_rate": 0.0002,
+      "loss": 1.1628,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3494682801613496,
+      "grad_norm": 0.4072401523590088,
+      "learning_rate": 0.0002,
+      "loss": 1.0972,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3568023469013568,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.1267,
+      "step": 1850
+    },
+    {
+      "epoch": 1.364136413641364,
+      "grad_norm": 0.412883460521698,
+      "learning_rate": 0.0002,
+      "loss": 1.0173,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3714704803813715,
+      "grad_norm": 0.3735875189304352,
+      "learning_rate": 0.0002,
+      "loss": 1.0265,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3788045471213788,
+      "grad_norm": 0.39158159494400024,
+      "learning_rate": 0.0002,
+      "loss": 1.1061,
+      "step": 1880
+    },
+    {
+      "epoch": 1.386138613861386,
+      "grad_norm": 0.44431769847869873,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3934726806013935,
+      "grad_norm": 0.37772801518440247,
+      "learning_rate": 0.0002,
+      "loss": 1.0216,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4008067473414008,
+      "grad_norm": 0.4056641757488251,
+      "learning_rate": 0.0002,
+      "loss": 1.0674,
+      "step": 1910
+    },
+    {
+      "epoch": 1.408140814081408,
+      "grad_norm": 0.41612377762794495,
+      "learning_rate": 0.0002,
+      "loss": 1.0256,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4154748808214155,
+      "grad_norm": 0.41153013706207275,
+      "learning_rate": 0.0002,
+      "loss": 1.0467,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4228089475614227,
+      "grad_norm": 0.387845516204834,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4301430143014302,
+      "grad_norm": 0.3809587061405182,
+      "learning_rate": 0.0002,
+      "loss": 1.1094,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4374770810414375,
+      "grad_norm": 0.3625726103782654,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1960
+    },
+    {
+      "epoch": 1.444811147781445,
+      "grad_norm": 0.5294290781021118,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4521452145214522,
+      "grad_norm": 0.39975494146347046,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4594792812614594,
+      "grad_norm": 0.4181167185306549,
+      "learning_rate": 0.0002,
+      "loss": 0.9704,
+      "step": 1990
+    },
+    {
+      "epoch": 1.466813348001467,
+      "grad_norm": 0.42001503705978394,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4741474147414741,
+      "grad_norm": 0.4877578616142273,
+      "learning_rate": 0.0002,
+      "loss": 1.1266,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4814814814814814,
+      "grad_norm": 0.4050969183444977,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4888155482214889,
+      "grad_norm": 0.39068883657455444,
+      "learning_rate": 0.0002,
+      "loss": 1.0562,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4961496149614961,
+      "grad_norm": 0.421282559633255,
+      "learning_rate": 0.0002,
+      "loss": 1.0464,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5034836817015034,
+      "grad_norm": 0.47092297673225403,
+      "learning_rate": 0.0002,
+      "loss": 1.0532,
+      "step": 2050
+    },
+    {
+      "epoch": 1.5108177484415108,
+      "grad_norm": 0.39688974618911743,
+      "learning_rate": 0.0002,
+      "loss": 0.9348,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5181518151815183,
+      "grad_norm": 0.5529879331588745,
+      "learning_rate": 0.0002,
+      "loss": 1.08,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5254858819215253,
+      "grad_norm": 0.4879782199859619,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5328199486615328,
+      "grad_norm": 0.5517361164093018,
+      "learning_rate": 0.0002,
+      "loss": 1.0432,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5401540154015403,
+      "grad_norm": 0.44015637040138245,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5474880821415475,
+      "grad_norm": 0.5435167551040649,
+      "learning_rate": 0.0002,
+      "loss": 1.1873,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5548221488815548,
+      "grad_norm": 0.5714033246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1076,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5621562156215623,
+      "grad_norm": 0.31732529401779175,
+      "learning_rate": 0.0002,
+      "loss": 1.1107,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5694902823615695,
+      "grad_norm": 0.49068278074264526,
+      "learning_rate": 0.0002,
+      "loss": 1.0817,
+      "step": 2140
+    },
+    {
+      "epoch": 1.5768243491015768,
+      "grad_norm": 0.46851542592048645,
+      "learning_rate": 0.0002,
+      "loss": 1.0254,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5841584158415842,
+      "grad_norm": 0.5083092451095581,
+      "learning_rate": 0.0002,
+      "loss": 1.0623,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5914924825815915,
+      "grad_norm": 0.9822936058044434,
+      "learning_rate": 0.0002,
+      "loss": 1.0603,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5988265493215987,
+      "grad_norm": 0.4575989246368408,
+      "learning_rate": 0.0002,
+      "loss": 0.9986,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6061606160616062,
+      "grad_norm": 0.47444286942481995,
+      "learning_rate": 0.0002,
+      "loss": 1.1292,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6134946828016135,
+      "grad_norm": 0.7208226919174194,
+      "learning_rate": 0.0002,
+      "loss": 1.0136,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6208287495416207,
+      "grad_norm": 0.43791481852531433,
+      "learning_rate": 0.0002,
+      "loss": 1.15,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6281628162816282,
+      "grad_norm": 0.5245792865753174,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6354968830216357,
+      "grad_norm": 0.39289429783821106,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6428309497616427,
+      "grad_norm": 0.6106135845184326,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6501650165016502,
+      "grad_norm": 0.3722580671310425,
+      "learning_rate": 0.0002,
+      "loss": 1.0129,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6574990832416576,
+      "grad_norm": 0.3649403750896454,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6648331499816649,
+      "grad_norm": 0.46514248847961426,
+      "learning_rate": 0.0002,
+      "loss": 1.0037,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6721672167216721,
+      "grad_norm": 0.42034927010536194,
+      "learning_rate": 0.0002,
+      "loss": 1.0022,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6795012834616796,
+      "grad_norm": 0.45202910900115967,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6868353502016868,
+      "grad_norm": 0.36257603764533997,
+      "learning_rate": 0.0002,
+      "loss": 1.0866,
+      "step": 2300
+    },
+    {
+      "epoch": 1.694169416941694,
+      "grad_norm": 0.6340323090553284,
+      "learning_rate": 0.0002,
+      "loss": 1.0973,
+      "step": 2310
+    },
+    {
+      "epoch": 1.7015034836817016,
+      "grad_norm": 0.4352878928184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0615,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7088375504217088,
+      "grad_norm": 0.45029792189598083,
+      "learning_rate": 0.0002,
+      "loss": 1.0629,
+      "step": 2330
+    },
+    {
+      "epoch": 1.716171617161716,
+      "grad_norm": 0.3891315758228302,
+      "learning_rate": 0.0002,
+      "loss": 0.9621,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7235056839017235,
+      "grad_norm": 0.35180050134658813,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7308397506417308,
+      "grad_norm": 0.42367449402809143,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 2360
+    },
+    {
+      "epoch": 1.738173817381738,
+      "grad_norm": 0.4553675353527069,
+      "learning_rate": 0.0002,
+      "loss": 1.0376,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7455078841217455,
+      "grad_norm": 0.5944654941558838,
+      "learning_rate": 0.0002,
+      "loss": 1.1467,
+      "step": 2380
+    },
+    {
+      "epoch": 1.752841950861753,
+      "grad_norm": 0.3479664623737335,
+      "learning_rate": 0.0002,
+      "loss": 1.0548,
+      "step": 2390
+    },
+    {
+      "epoch": 1.76017601760176,
+      "grad_norm": 0.3585502505302429,
+      "learning_rate": 0.0002,
+      "loss": 1.0798,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7675100843417675,
+      "grad_norm": 0.4263346493244171,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 2410
+    },
+    {
+      "epoch": 1.774844151081775,
+      "grad_norm": 0.5476409196853638,
+      "learning_rate": 0.0002,
+      "loss": 1.054,
+      "step": 2420
+    },
+    {
+      "epoch": 1.7821782178217822,
+      "grad_norm": 0.3694186508655548,
+      "learning_rate": 0.0002,
+      "loss": 1.1615,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7895122845617895,
+      "grad_norm": 0.9185658693313599,
+      "learning_rate": 0.0002,
+      "loss": 1.1343,
+      "step": 2440
+    },
+    {
+      "epoch": 1.796846351301797,
+      "grad_norm": 0.7171908020973206,
+      "learning_rate": 0.0002,
+      "loss": 1.0764,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8041804180418042,
+      "grad_norm": 0.550658643245697,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 2460
+    },
+    {
+      "epoch": 1.8115144847818114,
+      "grad_norm": 0.4075568914413452,
+      "learning_rate": 0.0002,
+      "loss": 0.9975,
+      "step": 2470
+    },
+    {
+      "epoch": 1.818848551521819,
+      "grad_norm": 0.3790127635002136,
+      "learning_rate": 0.0002,
+      "loss": 1.0935,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8261826182618262,
+      "grad_norm": 0.3576384484767914,
+      "learning_rate": 0.0002,
+      "loss": 0.9839,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8335166850018334,
+      "grad_norm": 0.3919370770454407,
+      "learning_rate": 0.0002,
+      "loss": 1.1369,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8408507517418409,
+      "grad_norm": 0.485083669424057,
+      "learning_rate": 0.0002,
+      "loss": 0.9985,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8481848184818483,
+      "grad_norm": 0.4564347565174103,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8555188852218554,
+      "grad_norm": 0.3613106608390808,
+      "learning_rate": 0.0002,
+      "loss": 1.0944,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8628529519618628,
+      "grad_norm": 0.39600759744644165,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8701870187018703,
+      "grad_norm": 1.123499870300293,
+      "learning_rate": 0.0002,
+      "loss": 0.9453,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8775210854418776,
+      "grad_norm": 0.4612680673599243,
+      "learning_rate": 0.0002,
+      "loss": 1.0635,
+      "step": 2560
+    },
+    {
+      "epoch": 1.8848551521818848,
+      "grad_norm": 0.42745399475097656,
+      "learning_rate": 0.0002,
+      "loss": 1.0087,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8921892189218923,
+      "grad_norm": 0.4055580198764801,
+      "learning_rate": 0.0002,
+      "loss": 1.0102,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8995232856618995,
+      "grad_norm": 0.44174644351005554,
+      "learning_rate": 0.0002,
+      "loss": 1.0177,
+      "step": 2590
+    },
+    {
+      "epoch": 1.9068573524019068,
+      "grad_norm": 1.0228385925292969,
+      "learning_rate": 0.0002,
+      "loss": 0.9886,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9141914191419143,
+      "grad_norm": 0.3496396243572235,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9215254858819215,
+      "grad_norm": 0.4191173017024994,
+      "learning_rate": 0.0002,
+      "loss": 1.0955,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9288595526219288,
+      "grad_norm": 0.6778554916381836,
+      "learning_rate": 0.0002,
+      "loss": 1.0943,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9361936193619362,
+      "grad_norm": 0.41992834210395813,
+      "learning_rate": 0.0002,
+      "loss": 1.0594,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9435276861019435,
+      "grad_norm": 0.8760401010513306,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9508617528419507,
+      "grad_norm": 0.44049209356307983,
+      "learning_rate": 0.0002,
+      "loss": 1.0379,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9581958195819582,
+      "grad_norm": 0.5651928782463074,
+      "learning_rate": 0.0002,
+      "loss": 1.1008,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9655298863219657,
+      "grad_norm": 0.5292727947235107,
+      "learning_rate": 0.0002,
+      "loss": 1.1317,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9728639530619727,
+      "grad_norm": 0.6012240648269653,
+      "learning_rate": 0.0002,
+      "loss": 1.1328,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9801980198019802,
+      "grad_norm": 0.3945149779319763,
+      "learning_rate": 0.0002,
+      "loss": 1.0683,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9875320865419877,
+      "grad_norm": 0.5732627511024475,
+      "learning_rate": 0.0002,
+      "loss": 1.0155,
+      "step": 2710
+    },
+    {
+      "epoch": 1.994866153281995,
+      "grad_norm": 0.3963361084461212,
+      "learning_rate": 0.0002,
+      "loss": 0.9857,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1534006595611572,
+      "eval_runtime": 32.7541,
+      "eval_samples_per_second": 13.159,
+      "eval_steps_per_second": 1.649,
+      "step": 2727
+    },
+    {
+      "epoch": 2.002200220022002,
+      "grad_norm": 0.48628315329551697,
+      "learning_rate": 0.0002,
+      "loss": 0.9624,
+      "step": 2730
+    },
+    {
+      "epoch": 2.0095342867620096,
+      "grad_norm": 0.413875013589859,
+      "learning_rate": 0.0002,
+      "loss": 0.9603,
+      "step": 2740
+    },
+    {
+      "epoch": 2.0168683535020167,
+      "grad_norm": 0.4988735616207123,
+      "learning_rate": 0.0002,
+      "loss": 0.965,
+      "step": 2750
+    },
+    {
+      "epoch": 2.024202420242024,
+      "grad_norm": 0.5634812712669373,
+      "learning_rate": 0.0002,
+      "loss": 0.9677,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0315364869820316,
+      "grad_norm": 0.48302653431892395,
+      "learning_rate": 0.0002,
+      "loss": 0.9547,
+      "step": 2770
+    },
+    {
+      "epoch": 2.038870553722039,
+      "grad_norm": 0.49914175271987915,
+      "learning_rate": 0.0002,
+      "loss": 0.9346,
+      "step": 2780
+    },
+    {
+      "epoch": 2.046204620462046,
+      "grad_norm": 1.14039945602417,
+      "learning_rate": 0.0002,
+      "loss": 0.904,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0535386872020536,
+      "grad_norm": 0.6359720826148987,
+      "learning_rate": 0.0002,
+      "loss": 0.9588,
+      "step": 2800
+    },
+    {
+      "epoch": 2.060872753942061,
+      "grad_norm": 0.4589158296585083,
+      "learning_rate": 0.0002,
+      "loss": 0.9031,
+      "step": 2810
+    },
+    {
+      "epoch": 2.068206820682068,
+      "grad_norm": 0.46255481243133545,
+      "learning_rate": 0.0002,
+      "loss": 0.9438,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0755408874220755,
+      "grad_norm": 0.6232137680053711,
+      "learning_rate": 0.0002,
+      "loss": 0.9464,
+      "step": 2830
+    },
+    {
+      "epoch": 2.082874954162083,
+      "grad_norm": 0.41042178869247437,
+      "learning_rate": 0.0002,
+      "loss": 0.8978,
+      "step": 2840
+    },
+    {
+      "epoch": 2.09020902090209,
+      "grad_norm": 0.5334428548812866,
+      "learning_rate": 0.0002,
+      "loss": 0.8516,
+      "step": 2850
+    },
+    {
+      "epoch": 2.0975430876420975,
+      "grad_norm": 0.8270058631896973,
+      "learning_rate": 0.0002,
+      "loss": 0.9313,
+      "step": 2860
+    },
+    {
+      "epoch": 2.104877154382105,
+      "grad_norm": 0.6624533534049988,
+      "learning_rate": 0.0002,
+      "loss": 1.0064,
+      "step": 2870
+    },
+    {
+      "epoch": 2.112211221122112,
+      "grad_norm": 0.5448863506317139,
+      "learning_rate": 0.0002,
+      "loss": 0.9196,
+      "step": 2880
+    },
+    {
+      "epoch": 2.1195452878621195,
+      "grad_norm": 0.621482789516449,
+      "learning_rate": 0.0002,
+      "loss": 0.887,
+      "step": 2890
+    },
+    {
+      "epoch": 2.126879354602127,
+      "grad_norm": 0.4556255340576172,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1342134213421344,
+      "grad_norm": 0.4620579183101654,
+      "learning_rate": 0.0002,
+      "loss": 0.9323,
+      "step": 2910
+    },
+    {
+      "epoch": 2.1415474880821415,
+      "grad_norm": 0.9602415561676025,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 2920
+    },
+    {
+      "epoch": 2.148881554822149,
+      "grad_norm": 0.587943971157074,
+      "learning_rate": 0.0002,
+      "loss": 0.8826,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1562156215621564,
+      "grad_norm": 0.5121372938156128,
+      "learning_rate": 0.0002,
+      "loss": 0.971,
+      "step": 2940
+    },
+    {
+      "epoch": 2.1635496883021634,
+      "grad_norm": 0.49424484372138977,
+      "learning_rate": 0.0002,
+      "loss": 0.8751,
+      "step": 2950
+    },
+    {
+      "epoch": 2.170883755042171,
+      "grad_norm": 0.6312560439109802,
+      "learning_rate": 0.0002,
+      "loss": 0.8674,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1782178217821784,
+      "grad_norm": 0.5235576629638672,
+      "learning_rate": 0.0002,
+      "loss": 0.9791,
+      "step": 2970
+    },
+    {
+      "epoch": 2.1855518885221854,
+      "grad_norm": 0.5868439674377441,
+      "learning_rate": 0.0002,
+      "loss": 0.9706,
+      "step": 2980
+    },
+    {
+      "epoch": 2.192885955262193,
+      "grad_norm": 0.42302873730659485,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 2990
+    },
+    {
+      "epoch": 2.2002200220022003,
+      "grad_norm": 0.5097725987434387,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 3000
+    },
+    {
+      "epoch": 2.2075540887422074,
+      "grad_norm": 0.5091572403907776,
+      "learning_rate": 0.0002,
+      "loss": 0.9239,
+      "step": 3010
+    },
+    {
+      "epoch": 2.214888155482215,
+      "grad_norm": 0.49433162808418274,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 3020
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.5577368140220642,
+      "learning_rate": 0.0002,
+      "loss": 0.9734,
+      "step": 3030
+    },
+    {
+      "epoch": 2.2295562889622293,
+      "grad_norm": 0.6177583932876587,
+      "learning_rate": 0.0002,
+      "loss": 0.9033,
+      "step": 3040
+    },
+    {
+      "epoch": 2.236890355702237,
+      "grad_norm": 0.5256719589233398,
+      "learning_rate": 0.0002,
+      "loss": 0.9882,
+      "step": 3050
+    },
+    {
+      "epoch": 2.2442244224422443,
+      "grad_norm": 0.5001118183135986,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2515584891822513,
+      "grad_norm": 0.5721249580383301,
+      "learning_rate": 0.0002,
+      "loss": 0.8718,
+      "step": 3070
+    },
+    {
+      "epoch": 2.258892555922259,
+      "grad_norm": 0.5325384140014648,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 3080
+    },
+    {
+      "epoch": 2.2662266226622663,
+      "grad_norm": 0.5719189047813416,
+      "learning_rate": 0.0002,
+      "loss": 0.9843,
+      "step": 3090
+    },
+    {
+      "epoch": 2.2735606894022737,
+      "grad_norm": 0.6337835788726807,
+      "learning_rate": 0.0002,
+      "loss": 0.8633,
+      "step": 3100
+    },
+    {
+      "epoch": 2.2808947561422808,
+      "grad_norm": 0.5381836891174316,
+      "learning_rate": 0.0002,
+      "loss": 0.9962,
+      "step": 3110
+    },
+    {
+      "epoch": 2.2882288228822882,
+      "grad_norm": 0.5408531427383423,
+      "learning_rate": 0.0002,
+      "loss": 0.8265,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2955628896222957,
+      "grad_norm": 0.43705281615257263,
+      "learning_rate": 0.0002,
+      "loss": 1.0325,
+      "step": 3130
+    },
+    {
+      "epoch": 2.3028969563623027,
+      "grad_norm": 0.6454030275344849,
+      "learning_rate": 0.0002,
+      "loss": 0.9388,
+      "step": 3140
+    },
+    {
+      "epoch": 2.31023102310231,
+      "grad_norm": 0.686030387878418,
+      "learning_rate": 0.0002,
+      "loss": 0.954,
+      "step": 3150
+    },
+    {
+      "epoch": 2.3175650898423177,
+      "grad_norm": 0.5123633146286011,
+      "learning_rate": 0.0002,
+      "loss": 0.9403,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3248991565823247,
+      "grad_norm": 0.842506468296051,
+      "learning_rate": 0.0002,
+      "loss": 0.8834,
+      "step": 3170
+    },
+    {
+      "epoch": 2.332233223322332,
+      "grad_norm": 0.5193818807601929,
+      "learning_rate": 0.0002,
+      "loss": 1.0497,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3395672900623397,
+      "grad_norm": 0.5634409189224243,
+      "learning_rate": 0.0002,
+      "loss": 0.9473,
+      "step": 3190
+    },
+    {
+      "epoch": 2.3469013568023467,
+      "grad_norm": 0.6475534439086914,
+      "learning_rate": 0.0002,
+      "loss": 0.8499,
+      "step": 3200
+    },
+    {
+      "epoch": 2.354235423542354,
+      "grad_norm": 1.1503914594650269,
+      "learning_rate": 0.0002,
+      "loss": 0.874,
+      "step": 3210
+    },
+    {
+      "epoch": 2.3615694902823616,
+      "grad_norm": 0.7234905362129211,
+      "learning_rate": 0.0002,
+      "loss": 0.9762,
+      "step": 3220
+    },
+    {
+      "epoch": 2.368903557022369,
+      "grad_norm": 0.664903461933136,
+      "learning_rate": 0.0002,
+      "loss": 0.9007,
+      "step": 3230
+    },
+    {
+      "epoch": 2.376237623762376,
+      "grad_norm": 0.5453006625175476,
+      "learning_rate": 0.0002,
+      "loss": 0.9987,
+      "step": 3240
+    },
+    {
+      "epoch": 2.3835716905023836,
+      "grad_norm": 0.6256654262542725,
+      "learning_rate": 0.0002,
+      "loss": 0.9742,
+      "step": 3250
+    },
+    {
+      "epoch": 2.390905757242391,
+      "grad_norm": 0.5166565179824829,
+      "learning_rate": 0.0002,
+      "loss": 0.9922,
+      "step": 3260
+    },
+    {
+      "epoch": 2.398239823982398,
+      "grad_norm": 0.5699098110198975,
+      "learning_rate": 0.0002,
+      "loss": 0.927,
+      "step": 3270
+    },
+    {
+      "epoch": 2.4055738907224056,
+      "grad_norm": 0.4472540020942688,
+      "learning_rate": 0.0002,
+      "loss": 0.8878,
+      "step": 3280
+    },
+    {
+      "epoch": 2.412907957462413,
+      "grad_norm": 0.6790403127670288,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 3290
+    },
+    {
+      "epoch": 2.42024202420242,
+      "grad_norm": 0.5182185173034668,
+      "learning_rate": 0.0002,
+      "loss": 0.972,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4275760909424275,
+      "grad_norm": 0.564647912979126,
+      "learning_rate": 0.0002,
+      "loss": 0.9775,
+      "step": 3310
+    },
+    {
+      "epoch": 2.434910157682435,
+      "grad_norm": 0.5625313520431519,
+      "learning_rate": 0.0002,
+      "loss": 1.072,
+      "step": 3320
+    },
+    {
+      "epoch": 2.442244224422442,
+      "grad_norm": 0.7496559619903564,
+      "learning_rate": 0.0002,
+      "loss": 0.8798,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4495782911624495,
+      "grad_norm": 0.4779128134250641,
+      "learning_rate": 0.0002,
+      "loss": 0.868,
+      "step": 3340
+    },
+    {
+      "epoch": 2.456912357902457,
+      "grad_norm": 0.578093409538269,
+      "learning_rate": 0.0002,
+      "loss": 1.0316,
+      "step": 3350
+    },
+    {
+      "epoch": 2.4642464246424645,
+      "grad_norm": 0.5456080436706543,
+      "learning_rate": 0.0002,
+      "loss": 0.9282,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4715804913824715,
+      "grad_norm": 0.4769273102283478,
+      "learning_rate": 0.0002,
+      "loss": 0.8409,
+      "step": 3370
+    },
+    {
+      "epoch": 2.478914558122479,
+      "grad_norm": 0.5608189702033997,
+      "learning_rate": 0.0002,
+      "loss": 0.9312,
+      "step": 3380
+    },
+    {
+      "epoch": 2.4862486248624864,
+      "grad_norm": 0.5590165853500366,
+      "learning_rate": 0.0002,
+      "loss": 0.9934,
+      "step": 3390
+    },
+    {
+      "epoch": 2.4935826916024935,
+      "grad_norm": 0.801306962966919,
+      "learning_rate": 0.0002,
+      "loss": 1.025,
+      "step": 3400
+    },
+    {
+      "epoch": 2.500916758342501,
+      "grad_norm": 0.6045624613761902,
+      "learning_rate": 0.0002,
+      "loss": 0.9049,
+      "step": 3410
+    },
+    {
+      "epoch": 2.5082508250825084,
+      "grad_norm": 0.5735858082771301,
+      "learning_rate": 0.0002,
+      "loss": 0.944,
+      "step": 3420
+    },
+    {
+      "epoch": 2.5155848918225154,
+      "grad_norm": 0.6827309131622314,
+      "learning_rate": 0.0002,
+      "loss": 0.9846,
+      "step": 3430
+    },
+    {
+      "epoch": 2.522918958562523,
+      "grad_norm": 0.5702602863311768,
+      "learning_rate": 0.0002,
+      "loss": 0.9789,
+      "step": 3440
+    },
+    {
+      "epoch": 2.5302530253025304,
+      "grad_norm": 0.6674721240997314,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5375870920425374,
+      "grad_norm": 0.5635907649993896,
+      "learning_rate": 0.0002,
+      "loss": 0.914,
+      "step": 3460
+    },
+    {
+      "epoch": 2.544921158782545,
+      "grad_norm": 0.42737770080566406,
+      "learning_rate": 0.0002,
+      "loss": 0.8398,
+      "step": 3470
+    },
+    {
+      "epoch": 2.5522552255225524,
+      "grad_norm": 0.6720691919326782,
+      "learning_rate": 0.0002,
+      "loss": 0.9474,
+      "step": 3480
+    },
+    {
+      "epoch": 2.55958929226256,
+      "grad_norm": 0.8917084336280823,
+      "learning_rate": 0.0002,
+      "loss": 0.8637,
+      "step": 3490
+    },
+    {
+      "epoch": 2.566923359002567,
+      "grad_norm": 0.5134549140930176,
+      "learning_rate": 0.0002,
+      "loss": 0.9257,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5742574257425743,
+      "grad_norm": 0.4951367974281311,
+      "learning_rate": 0.0002,
+      "loss": 0.9362,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5815914924825814,
+      "grad_norm": 0.9438204765319824,
+      "learning_rate": 0.0002,
+      "loss": 0.9184,
+      "step": 3520
+    },
+    {
+      "epoch": 2.588925559222589,
+      "grad_norm": 0.6024714708328247,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5962596259625963,
+      "grad_norm": 0.5248535871505737,
+      "learning_rate": 0.0002,
+      "loss": 0.9298,
+      "step": 3540
+    },
+    {
+      "epoch": 2.6035936927026038,
+      "grad_norm": 0.8677568435668945,
+      "learning_rate": 0.0002,
+      "loss": 0.941,
+      "step": 3550
+    },
+    {
+      "epoch": 2.610927759442611,
+      "grad_norm": 0.82008296251297,
+      "learning_rate": 0.0002,
+      "loss": 0.9253,
+      "step": 3560
+    },
+    {
+      "epoch": 2.6182618261826183,
+      "grad_norm": 0.4724634885787964,
+      "learning_rate": 0.0002,
+      "loss": 0.8429,
+      "step": 3570
+    },
+    {
+      "epoch": 2.6255958929226257,
+      "grad_norm": 0.5434244275093079,
+      "learning_rate": 0.0002,
+      "loss": 0.9058,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6329299596626328,
+      "grad_norm": 0.4948740005493164,
+      "learning_rate": 0.0002,
+      "loss": 0.9379,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6402640264026402,
+      "grad_norm": 0.42109328508377075,
+      "learning_rate": 0.0002,
+      "loss": 0.8718,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6475980931426477,
+      "grad_norm": 0.7979786396026611,
+      "learning_rate": 0.0002,
+      "loss": 0.9809,
+      "step": 3610
+    },
+    {
+      "epoch": 2.654932159882655,
+      "grad_norm": 0.6345919370651245,
+      "learning_rate": 0.0002,
+      "loss": 0.9229,
+      "step": 3620
+    },
+    {
+      "epoch": 2.662266226622662,
+      "grad_norm": 0.4971671402454376,
+      "learning_rate": 0.0002,
+      "loss": 0.8506,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6696002933626697,
+      "grad_norm": 0.6467748284339905,
+      "learning_rate": 0.0002,
+      "loss": 0.8054,
+      "step": 3640
+    },
+    {
+      "epoch": 2.6769343601026767,
+      "grad_norm": 0.4240160286426544,
+      "learning_rate": 0.0002,
+      "loss": 0.9277,
+      "step": 3650
+    },
+    {
+      "epoch": 2.684268426842684,
+      "grad_norm": 0.5179754495620728,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 3660
+    },
+    {
+      "epoch": 2.6916024935826917,
+      "grad_norm": 0.754012405872345,
+      "learning_rate": 0.0002,
+      "loss": 0.9221,
+      "step": 3670
+    },
+    {
+      "epoch": 2.698936560322699,
+      "grad_norm": 0.5141299962997437,
+      "learning_rate": 0.0002,
+      "loss": 0.9194,
+      "step": 3680
+    },
+    {
+      "epoch": 2.706270627062706,
+      "grad_norm": 0.5737819075584412,
+      "learning_rate": 0.0002,
+      "loss": 0.9495,
+      "step": 3690
+    },
+    {
+      "epoch": 2.7136046938027136,
+      "grad_norm": 0.5887577533721924,
+      "learning_rate": 0.0002,
+      "loss": 1.0162,
+      "step": 3700
+    },
+    {
+      "epoch": 2.720938760542721,
+      "grad_norm": 0.6740471720695496,
+      "learning_rate": 0.0002,
+      "loss": 0.9169,
+      "step": 3710
+    },
+    {
+      "epoch": 2.728272827282728,
+      "grad_norm": 0.5879453420639038,
+      "learning_rate": 0.0002,
+      "loss": 0.9297,
+      "step": 3720
+    },
+    {
+      "epoch": 2.7356068940227356,
+      "grad_norm": 0.4858354926109314,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 3730
+    },
+    {
+      "epoch": 2.742940960762743,
+      "grad_norm": 0.5489001870155334,
+      "learning_rate": 0.0002,
+      "loss": 0.9308,
+      "step": 3740
+    },
+    {
+      "epoch": 2.7502750275027505,
+      "grad_norm": 0.8187092542648315,
+      "learning_rate": 0.0002,
+      "loss": 0.894,
+      "step": 3750
+    },
+    {
+      "epoch": 2.7576090942427576,
+      "grad_norm": 0.5666626691818237,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 3760
+    },
+    {
+      "epoch": 2.764943160982765,
+      "grad_norm": 0.5377066135406494,
+      "learning_rate": 0.0002,
+      "loss": 1.0059,
+      "step": 3770
+    },
+    {
+      "epoch": 2.772277227722772,
+      "grad_norm": 0.566330075263977,
+      "learning_rate": 0.0002,
+      "loss": 0.9132,
+      "step": 3780
+    },
+    {
+      "epoch": 2.7796112944627795,
+      "grad_norm": 0.5522832870483398,
+      "learning_rate": 0.0002,
+      "loss": 0.9415,
+      "step": 3790
+    },
+    {
+      "epoch": 2.786945361202787,
+      "grad_norm": 0.5668695569038391,
+      "learning_rate": 0.0002,
+      "loss": 0.8816,
+      "step": 3800
+    },
+    {
+      "epoch": 2.7942794279427945,
+      "grad_norm": 0.7566602826118469,
+      "learning_rate": 0.0002,
+      "loss": 0.8885,
+      "step": 3810
+    },
+    {
+      "epoch": 2.8016134946828015,
+      "grad_norm": 0.5603684782981873,
+      "learning_rate": 0.0002,
+      "loss": 0.8598,
+      "step": 3820
+    },
+    {
+      "epoch": 2.808947561422809,
+      "grad_norm": 0.49122217297554016,
+      "learning_rate": 0.0002,
+      "loss": 0.9602,
+      "step": 3830
+    },
+    {
+      "epoch": 2.816281628162816,
+      "grad_norm": 0.6798251867294312,
+      "learning_rate": 0.0002,
+      "loss": 0.9738,
+      "step": 3840
+    },
+    {
+      "epoch": 2.8236156949028235,
+      "grad_norm": 0.6097991466522217,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 3850
+    },
+    {
+      "epoch": 2.830949761642831,
+      "grad_norm": 0.6675726175308228,
+      "learning_rate": 0.0002,
+      "loss": 0.8672,
+      "step": 3860
+    },
+    {
+      "epoch": 2.8382838283828384,
+      "grad_norm": 0.9223952889442444,
+      "learning_rate": 0.0002,
+      "loss": 0.9324,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8456178951228455,
+      "grad_norm": 0.6020799875259399,
+      "learning_rate": 0.0002,
+      "loss": 0.8767,
+      "step": 3880
+    },
+    {
+      "epoch": 2.852951961862853,
+      "grad_norm": 0.5206381678581238,
+      "learning_rate": 0.0002,
+      "loss": 0.9148,
+      "step": 3890
+    },
+    {
+      "epoch": 2.8602860286028604,
+      "grad_norm": 0.6268777251243591,
+      "learning_rate": 0.0002,
+      "loss": 0.9479,
+      "step": 3900
+    },
+    {
+      "epoch": 2.8676200953428674,
+      "grad_norm": 1.1583497524261475,
+      "learning_rate": 0.0002,
+      "loss": 0.9409,
+      "step": 3910
+    },
+    {
+      "epoch": 2.874954162082875,
+      "grad_norm": 0.7263903021812439,
+      "learning_rate": 0.0002,
+      "loss": 0.895,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8822882288228824,
+      "grad_norm": 0.5369910001754761,
+      "learning_rate": 0.0002,
+      "loss": 0.8786,
+      "step": 3930
+    },
+    {
+      "epoch": 2.88962229556289,
+      "grad_norm": 0.7298350930213928,
+      "learning_rate": 0.0002,
+      "loss": 1.0015,
+      "step": 3940
+    },
+    {
+      "epoch": 2.896956362302897,
+      "grad_norm": 0.577012836933136,
+      "learning_rate": 0.0002,
+      "loss": 0.979,
+      "step": 3950
+    },
+    {
+      "epoch": 2.9042904290429044,
+      "grad_norm": 0.5859594345092773,
+      "learning_rate": 0.0002,
+      "loss": 0.9716,
+      "step": 3960
+    },
+    {
+      "epoch": 2.9116244957829114,
+      "grad_norm": 0.47176122665405273,
+      "learning_rate": 0.0002,
+      "loss": 0.8772,
+      "step": 3970
+    },
+    {
+      "epoch": 2.918958562522919,
+      "grad_norm": 0.9699620604515076,
+      "learning_rate": 0.0002,
+      "loss": 0.8997,
+      "step": 3980
+    },
+    {
+      "epoch": 2.9262926292629263,
+      "grad_norm": 0.7908747792243958,
+      "learning_rate": 0.0002,
+      "loss": 0.9057,
+      "step": 3990
+    },
+    {
+      "epoch": 2.933626696002934,
+      "grad_norm": 0.5777379274368286,
+      "learning_rate": 0.0002,
+      "loss": 0.9462,
+      "step": 4000
+    },
+    {
+      "epoch": 2.940960762742941,
+      "grad_norm": 0.599288284778595,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9482948294829483,
+      "grad_norm": 0.5232274532318115,
+      "learning_rate": 0.0002,
+      "loss": 0.9812,
+      "step": 4020
+    },
+    {
+      "epoch": 2.9556288962229558,
+      "grad_norm": 0.6395137310028076,
+      "learning_rate": 0.0002,
+      "loss": 0.96,
+      "step": 4030
+    },
+    {
+      "epoch": 2.962962962962963,
+      "grad_norm": 0.589260458946228,
+      "learning_rate": 0.0002,
+      "loss": 0.9813,
+      "step": 4040
+    },
+    {
+      "epoch": 2.9702970297029703,
+      "grad_norm": 0.5699581503868103,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 4050
+    },
+    {
+      "epoch": 2.9776310964429777,
+      "grad_norm": 0.528468132019043,
+      "learning_rate": 0.0002,
+      "loss": 0.9585,
+      "step": 4060
+    },
+    {
+      "epoch": 2.984965163182985,
+      "grad_norm": 0.4804670512676239,
+      "learning_rate": 0.0002,
+      "loss": 0.9164,
+      "step": 4070
+    },
+    {
+      "epoch": 2.9922992299229922,
+      "grad_norm": 1.1918889284133911,
+      "learning_rate": 0.0002,
+      "loss": 0.9771,
+      "step": 4080
+    },
+    {
+      "epoch": 2.9996332966629997,
+      "grad_norm": 0.5479103326797485,
+      "learning_rate": 0.0002,
+      "loss": 0.9178,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9996332966629997,
+      "eval_loss": 1.1642853021621704,
+      "eval_runtime": 32.7511,
+      "eval_samples_per_second": 13.16,
+      "eval_steps_per_second": 1.649,
+      "step": 4090
+    },
+    {
+      "epoch": 3.006967363403007,
+      "grad_norm": 0.7430027723312378,
+      "learning_rate": 0.0002,
+      "loss": 0.7981,
+      "step": 4100
+    },
+    {
+      "epoch": 3.014301430143014,
+      "grad_norm": 0.6293647289276123,
+      "learning_rate": 0.0002,
+      "loss": 0.7871,
+      "step": 4110
+    },
+    {
+      "epoch": 3.0216354968830217,
+      "grad_norm": 0.6191329956054688,
+      "learning_rate": 0.0002,
+      "loss": 0.78,
+      "step": 4120
+    },
+    {
+      "epoch": 3.028969563623029,
+      "grad_norm": 0.7959313988685608,
+      "learning_rate": 0.0002,
+      "loss": 0.7618,
+      "step": 4130
+    },
+    {
+      "epoch": 3.036303630363036,
+      "grad_norm": 0.5956351161003113,
+      "learning_rate": 0.0002,
+      "loss": 0.8039,
+      "step": 4140
+    },
+    {
+      "epoch": 3.0436376971030437,
+      "grad_norm": 0.670383632183075,
+      "learning_rate": 0.0002,
+      "loss": 0.7477,
+      "step": 4150
+    },
+    {
+      "epoch": 3.050971763843051,
+      "grad_norm": 0.6414518356323242,
+      "learning_rate": 0.0002,
+      "loss": 0.7984,
+      "step": 4160
+    },
+    {
+      "epoch": 3.058305830583058,
+      "grad_norm": 0.7928852438926697,
+      "learning_rate": 0.0002,
+      "loss": 0.7369,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0656398973230656,
+      "grad_norm": 0.6211121082305908,
+      "learning_rate": 0.0002,
+      "loss": 0.7914,
+      "step": 4180
+    },
+    {
+      "epoch": 3.072973964063073,
+      "grad_norm": 0.6237057447433472,
+      "learning_rate": 0.0002,
+      "loss": 0.7365,
+      "step": 4190
+    },
+    {
+      "epoch": 3.08030803080308,
+      "grad_norm": 0.6522233486175537,
+      "learning_rate": 0.0002,
+      "loss": 0.702,
+      "step": 4200
+    },
+    {
+      "epoch": 3.0876420975430876,
+      "grad_norm": 0.9396848678588867,
+      "learning_rate": 0.0002,
+      "loss": 0.7646,
+      "step": 4210
+    },
+    {
+      "epoch": 3.094976164283095,
+      "grad_norm": 0.8003010749816895,
+      "learning_rate": 0.0002,
+      "loss": 0.7559,
+      "step": 4220
+    },
+    {
+      "epoch": 3.102310231023102,
+      "grad_norm": 0.6733810305595398,
+      "learning_rate": 0.0002,
+      "loss": 0.711,
+      "step": 4230
+    },
+    {
+      "epoch": 3.1096442977631096,
+      "grad_norm": 0.6365828514099121,
+      "learning_rate": 0.0002,
+      "loss": 0.696,
+      "step": 4240
+    },
+    {
+      "epoch": 3.116978364503117,
+      "grad_norm": 1.0805548429489136,
+      "learning_rate": 0.0002,
+      "loss": 0.8362,
+      "step": 4250
+    },
+    {
+      "epoch": 3.1243124312431245,
+      "grad_norm": 0.7262141108512878,
+      "learning_rate": 0.0002,
+      "loss": 0.7651,
+      "step": 4260
+    },
+    {
+      "epoch": 3.1316464979831315,
+      "grad_norm": 0.5500539541244507,
+      "learning_rate": 0.0002,
+      "loss": 0.7304,
+      "step": 4270
+    },
+    {
+      "epoch": 3.138980564723139,
+      "grad_norm": 0.793912947177887,
+      "learning_rate": 0.0002,
+      "loss": 0.7721,
+      "step": 4280
+    },
+    {
+      "epoch": 3.1463146314631465,
+      "grad_norm": 1.2540518045425415,
+      "learning_rate": 0.0002,
+      "loss": 0.7708,
+      "step": 4290
+    },
+    {
+      "epoch": 3.1536486982031535,
+      "grad_norm": 0.7020077705383301,
+      "learning_rate": 0.0002,
+      "loss": 0.782,
+      "step": 4300
+    },
+    {
+      "epoch": 3.160982764943161,
+      "grad_norm": 0.5111123323440552,
+      "learning_rate": 0.0002,
+      "loss": 0.7253,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1683168316831685,
+      "grad_norm": 0.7172090411186218,
+      "learning_rate": 0.0002,
+      "loss": 0.8159,
+      "step": 4320
+    },
+    {
+      "epoch": 3.1756508984231755,
+      "grad_norm": 0.6343168616294861,
+      "learning_rate": 0.0002,
+      "loss": 0.6962,
+      "step": 4330
+    },
+    {
+      "epoch": 3.182984965163183,
+      "grad_norm": 0.9563672542572021,
+      "learning_rate": 0.0002,
+      "loss": 0.7938,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1903190319031904,
+      "grad_norm": 1.0225574970245361,
+      "learning_rate": 0.0002,
+      "loss": 0.7385,
+      "step": 4350
+    },
+    {
+      "epoch": 3.1976530986431975,
+      "grad_norm": 1.1633386611938477,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 4360
+    },
+    {
+      "epoch": 3.204987165383205,
+      "grad_norm": 0.8915148973464966,
+      "learning_rate": 0.0002,
+      "loss": 0.7259,
+      "step": 4370
+    },
+    {
+      "epoch": 3.2123212321232124,
+      "grad_norm": 0.9156812429428101,
+      "learning_rate": 0.0002,
+      "loss": 0.8061,
+      "step": 4380
+    },
+    {
+      "epoch": 3.21965529886322,
+      "grad_norm": 0.6363258957862854,
+      "learning_rate": 0.0002,
+      "loss": 0.8189,
+      "step": 4390
+    },
+    {
+      "epoch": 3.226989365603227,
+      "grad_norm": 0.579099178314209,
+      "learning_rate": 0.0002,
+      "loss": 0.7996,
+      "step": 4400
+    },
+    {
+      "epoch": 3.2343234323432344,
+      "grad_norm": 0.8778146505355835,
+      "learning_rate": 0.0002,
+      "loss": 0.8592,
+      "step": 4410
+    },
+    {
+      "epoch": 3.241657499083242,
+      "grad_norm": 0.8356770873069763,
+      "learning_rate": 0.0002,
+      "loss": 0.8281,
+      "step": 4420
+    },
+    {
+      "epoch": 3.248991565823249,
+      "grad_norm": 0.702032208442688,
+      "learning_rate": 0.0002,
+      "loss": 0.8484,
+      "step": 4430
+    },
+    {
+      "epoch": 3.2563256325632564,
+      "grad_norm": 0.6386539340019226,
+      "learning_rate": 0.0002,
+      "loss": 0.7227,
+      "step": 4440
+    },
+    {
+      "epoch": 3.263659699303264,
+      "grad_norm": 0.7008408904075623,
+      "learning_rate": 0.0002,
+      "loss": 0.8374,
+      "step": 4450
+    },
+    {
+      "epoch": 3.270993766043271,
+      "grad_norm": 0.9556332230567932,
+      "learning_rate": 0.0002,
+      "loss": 0.7572,
+      "step": 4460
+    },
+    {
+      "epoch": 3.2783278327832783,
+      "grad_norm": 0.5667835474014282,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 4470
+    },
+    {
+      "epoch": 3.285661899523286,
+      "grad_norm": 0.8239172697067261,
+      "learning_rate": 0.0002,
+      "loss": 0.8152,
+      "step": 4480
+    },
+    {
+      "epoch": 3.292995966263293,
+      "grad_norm": 0.7045050859451294,
+      "learning_rate": 0.0002,
+      "loss": 0.756,
+      "step": 4490
+    },
+    {
+      "epoch": 3.3003300330033003,
+      "grad_norm": 0.7131434082984924,
+      "learning_rate": 0.0002,
+      "loss": 0.7655,
+      "step": 4500
+    },
+    {
+      "epoch": 3.3076640997433078,
+      "grad_norm": 0.6924910545349121,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 4510
+    },
+    {
+      "epoch": 3.3149981664833152,
+      "grad_norm": 0.8945356607437134,
+      "learning_rate": 0.0002,
+      "loss": 0.736,
+      "step": 4520
+    },
+    {
+      "epoch": 3.3223322332233223,
+      "grad_norm": 0.6546903252601624,
+      "learning_rate": 0.0002,
+      "loss": 0.7575,
+      "step": 4530
+    },
+    {
+      "epoch": 3.3296662999633297,
+      "grad_norm": 0.8206679224967957,
+      "learning_rate": 0.0002,
+      "loss": 0.7893,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3370003667033368,
+      "grad_norm": 0.6482203602790833,
+      "learning_rate": 0.0002,
+      "loss": 0.7502,
+      "step": 4550
+    },
+    {
+      "epoch": 3.3443344334433442,
+      "grad_norm": 0.7558760046958923,
+      "learning_rate": 0.0002,
+      "loss": 0.8172,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3516685001833517,
+      "grad_norm": 0.7794756889343262,
+      "learning_rate": 0.0002,
+      "loss": 0.744,
+      "step": 4570
+    },
+    {
+      "epoch": 3.359002566923359,
+      "grad_norm": 0.7382805943489075,
+      "learning_rate": 0.0002,
+      "loss": 0.7385,
+      "step": 4580
+    },
+    {
+      "epoch": 3.366336633663366,
+      "grad_norm": 0.5912511944770813,
+      "learning_rate": 0.0002,
+      "loss": 0.8511,
+      "step": 4590
+    },
+    {
+      "epoch": 3.3736707004033737,
+      "grad_norm": 0.7444885969161987,
+      "learning_rate": 0.0002,
+      "loss": 0.8272,
+      "step": 4600
+    },
+    {
+      "epoch": 3.381004767143381,
+      "grad_norm": 0.7354922890663147,
+      "learning_rate": 0.0002,
+      "loss": 0.7927,
+      "step": 4610
+    },
+    {
+      "epoch": 3.388338833883388,
+      "grad_norm": 0.7685934901237488,
+      "learning_rate": 0.0002,
+      "loss": 0.7183,
+      "step": 4620
+    },
+    {
+      "epoch": 3.3956729006233957,
+      "grad_norm": 0.61041259765625,
+      "learning_rate": 0.0002,
+      "loss": 0.7436,
+      "step": 4630
+    },
+    {
+      "epoch": 3.403006967363403,
+      "grad_norm": 0.6820451021194458,
+      "learning_rate": 0.0002,
+      "loss": 0.7661,
+      "step": 4640
+    },
+    {
+      "epoch": 3.41034103410341,
+      "grad_norm": 0.5819534063339233,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 4650
+    },
+    {
+      "epoch": 3.4176751008434176,
+      "grad_norm": 0.705410897731781,
+      "learning_rate": 0.0002,
+      "loss": 0.7314,
+      "step": 4660
+    },
+    {
+      "epoch": 3.425009167583425,
+      "grad_norm": 0.8052892088890076,
+      "learning_rate": 0.0002,
+      "loss": 0.7901,
+      "step": 4670
+    },
+    {
+      "epoch": 3.432343234323432,
+      "grad_norm": 0.7746483087539673,
+      "learning_rate": 0.0002,
+      "loss": 0.7298,
+      "step": 4680
+    },
+    {
+      "epoch": 3.4396773010634396,
+      "grad_norm": 0.7713689804077148,
+      "learning_rate": 0.0002,
+      "loss": 0.7976,
+      "step": 4690
+    },
+    {
+      "epoch": 3.447011367803447,
+      "grad_norm": 0.810371994972229,
+      "learning_rate": 0.0002,
+      "loss": 0.7427,
+      "step": 4700
+    },
+    {
+      "epoch": 3.4543454345434546,
+      "grad_norm": 0.7702969312667847,
+      "learning_rate": 0.0002,
+      "loss": 0.7594,
+      "step": 4710
+    },
+    {
+      "epoch": 3.4616795012834616,
+      "grad_norm": 0.7069268822669983,
+      "learning_rate": 0.0002,
+      "loss": 0.7957,
+      "step": 4720
+    },
+    {
+      "epoch": 3.469013568023469,
+      "grad_norm": 0.7640359401702881,
+      "learning_rate": 0.0002,
+      "loss": 0.8199,
+      "step": 4730
+    },
+    {
+      "epoch": 3.4763476347634765,
+      "grad_norm": 0.8661707639694214,
+      "learning_rate": 0.0002,
+      "loss": 0.6875,
+      "step": 4740
+    },
+    {
+      "epoch": 3.4836817015034836,
+      "grad_norm": 0.9970282912254333,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 4750
+    },
+    {
+      "epoch": 3.491015768243491,
+      "grad_norm": 0.5824355483055115,
+      "learning_rate": 0.0002,
+      "loss": 0.8462,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4983498349834985,
+      "grad_norm": 1.3072649240493774,
+      "learning_rate": 0.0002,
+      "loss": 0.851,
+      "step": 4770
+    },
+    {
+      "epoch": 3.5056839017235055,
+      "grad_norm": 0.873978316783905,
+      "learning_rate": 0.0002,
+      "loss": 0.9101,
+      "step": 4780
+    },
+    {
+      "epoch": 3.513017968463513,
+      "grad_norm": 0.5526657104492188,
+      "learning_rate": 0.0002,
+      "loss": 0.7403,
+      "step": 4790
+    },
+    {
+      "epoch": 3.5203520352035205,
+      "grad_norm": 0.790894627571106,
+      "learning_rate": 0.0002,
+      "loss": 0.7921,
+      "step": 4800
+    },
+    {
+      "epoch": 3.5276861019435275,
+      "grad_norm": 0.8119630217552185,
+      "learning_rate": 0.0002,
+      "loss": 0.831,
+      "step": 4810
+    },
+    {
+      "epoch": 3.535020168683535,
+      "grad_norm": 0.633212149143219,
+      "learning_rate": 0.0002,
+      "loss": 0.7351,
+      "step": 4820
+    },
+    {
+      "epoch": 3.5423542354235424,
+      "grad_norm": 0.703029990196228,
+      "learning_rate": 0.0002,
+      "loss": 0.8505,
+      "step": 4830
+    },
+    {
+      "epoch": 3.54968830216355,
+      "grad_norm": 0.7603771686553955,
+      "learning_rate": 0.0002,
+      "loss": 0.7204,
+      "step": 4840
+    },
+    {
+      "epoch": 3.557022368903557,
+      "grad_norm": 0.6260480880737305,
+      "learning_rate": 0.0002,
+      "loss": 0.8868,
+      "step": 4850
+    },
+    {
+      "epoch": 3.5643564356435644,
+      "grad_norm": 0.8203664422035217,
+      "learning_rate": 0.0002,
+      "loss": 0.8137,
+      "step": 4860
+    },
+    {
+      "epoch": 3.5716905023835714,
+      "grad_norm": 0.7793813347816467,
+      "learning_rate": 0.0002,
+      "loss": 0.8821,
+      "step": 4870
+    },
+    {
+      "epoch": 3.579024569123579,
+      "grad_norm": 0.7667397260665894,
+      "learning_rate": 0.0002,
+      "loss": 0.8164,
+      "step": 4880
+    },
+    {
+      "epoch": 3.5863586358635864,
+      "grad_norm": 0.8198829889297485,
+      "learning_rate": 0.0002,
+      "loss": 0.7597,
+      "step": 4890
+    },
+    {
+      "epoch": 3.593692702603594,
+      "grad_norm": 0.7689233422279358,
+      "learning_rate": 0.0002,
+      "loss": 0.7027,
+      "step": 4900
+    },
+    {
+      "epoch": 3.601026769343601,
+      "grad_norm": 0.7870983481407166,
+      "learning_rate": 0.0002,
+      "loss": 0.804,
+      "step": 4910
+    },
+    {
+      "epoch": 3.6083608360836084,
+      "grad_norm": 0.8133853077888489,
+      "learning_rate": 0.0002,
+      "loss": 0.8269,
+      "step": 4920
+    },
+    {
+      "epoch": 3.615694902823616,
+      "grad_norm": 1.308401346206665,
+      "learning_rate": 0.0002,
+      "loss": 0.8515,
+      "step": 4930
+    },
+    {
+      "epoch": 3.623028969563623,
+      "grad_norm": 0.7131121754646301,
+      "learning_rate": 0.0002,
+      "loss": 0.8494,
+      "step": 4940
+    },
+    {
+      "epoch": 3.6303630363036303,
+      "grad_norm": 0.6825910210609436,
+      "learning_rate": 0.0002,
+      "loss": 0.7235,
+      "step": 4950
+    },
+    {
+      "epoch": 3.637697103043638,
+      "grad_norm": 0.7254678606987,
+      "learning_rate": 0.0002,
+      "loss": 0.7824,
+      "step": 4960
+    },
+    {
+      "epoch": 3.6450311697836453,
+      "grad_norm": 0.8045085072517395,
+      "learning_rate": 0.0002,
+      "loss": 0.7983,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6523652365236523,
+      "grad_norm": 0.6991777420043945,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6596993032636598,
+      "grad_norm": 0.7804713249206543,
+      "learning_rate": 0.0002,
+      "loss": 0.7806,
+      "step": 4990
+    },
+    {
+      "epoch": 3.667033370003667,
+      "grad_norm": 0.8525708317756653,
+      "learning_rate": 0.0002,
+      "loss": 0.8402,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6743674367436743,
+      "grad_norm": 0.7959994673728943,
+      "learning_rate": 0.0002,
+      "loss": 0.8496,
+      "step": 5010
+    },
+    {
+      "epoch": 3.6817015034836817,
+      "grad_norm": 0.8103628158569336,
+      "learning_rate": 0.0002,
+      "loss": 0.8022,
+      "step": 5020
+    },
+    {
+      "epoch": 3.689035570223689,
+      "grad_norm": 0.7517836093902588,
+      "learning_rate": 0.0002,
+      "loss": 0.7376,
+      "step": 5030
+    },
+    {
+      "epoch": 3.6963696369636962,
+      "grad_norm": 0.6878514289855957,
+      "learning_rate": 0.0002,
+      "loss": 0.8375,
+      "step": 5040
+    },
+    {
+      "epoch": 3.7037037037037037,
+      "grad_norm": 1.2371820211410522,
+      "learning_rate": 0.0002,
+      "loss": 0.7998,
+      "step": 5050
+    },
+    {
+      "epoch": 3.711037770443711,
+      "grad_norm": 0.6567103862762451,
+      "learning_rate": 0.0002,
+      "loss": 0.6941,
+      "step": 5060
+    },
+    {
+      "epoch": 3.718371837183718,
+      "grad_norm": 1.1254922151565552,
+      "learning_rate": 0.0002,
+      "loss": 0.8465,
+      "step": 5070
+    },
+    {
+      "epoch": 3.7257059039237257,
+      "grad_norm": 0.6796132326126099,
+      "learning_rate": 0.0002,
+      "loss": 0.8365,
+      "step": 5080
+    },
+    {
+      "epoch": 3.733039970663733,
+      "grad_norm": 0.7285300493240356,
+      "learning_rate": 0.0002,
+      "loss": 0.7818,
+      "step": 5090
+    },
+    {
+      "epoch": 3.7403740374037406,
+      "grad_norm": 0.8931500911712646,
+      "learning_rate": 0.0002,
+      "loss": 0.8581,
+      "step": 5100
+    },
+    {
+      "epoch": 3.7477081041437477,
+      "grad_norm": 0.6256856918334961,
+      "learning_rate": 0.0002,
+      "loss": 0.8181,
+      "step": 5110
+    },
+    {
+      "epoch": 3.755042170883755,
+      "grad_norm": 0.79310142993927,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 5120
+    },
+    {
+      "epoch": 3.762376237623762,
+      "grad_norm": 0.6594041585922241,
+      "learning_rate": 0.0002,
+      "loss": 0.8235,
+      "step": 5130
+    },
+    {
+      "epoch": 3.7697103043637696,
+      "grad_norm": 0.7029327750205994,
+      "learning_rate": 0.0002,
+      "loss": 0.6925,
+      "step": 5140
+    },
+    {
+      "epoch": 3.777044371103777,
+      "grad_norm": 0.5880070328712463,
+      "learning_rate": 0.0002,
+      "loss": 0.7457,
+      "step": 5150
+    },
+    {
+      "epoch": 3.7843784378437846,
+      "grad_norm": 0.7578945159912109,
+      "learning_rate": 0.0002,
+      "loss": 0.8716,
+      "step": 5160
+    },
+    {
+      "epoch": 3.7917125045837916,
+      "grad_norm": 0.8276378512382507,
+      "learning_rate": 0.0002,
+      "loss": 0.8819,
+      "step": 5170
+    },
+    {
+      "epoch": 3.799046571323799,
+      "grad_norm": 0.7627953886985779,
+      "learning_rate": 0.0002,
+      "loss": 0.7559,
+      "step": 5180
+    },
+    {
+      "epoch": 3.806380638063806,
+      "grad_norm": 0.8169086575508118,
+      "learning_rate": 0.0002,
+      "loss": 0.7665,
+      "step": 5190
+    },
+    {
+      "epoch": 3.8137147048038136,
+      "grad_norm": 0.6605030298233032,
+      "learning_rate": 0.0002,
+      "loss": 0.761,
+      "step": 5200
+    },
+    {
+      "epoch": 3.821048771543821,
+      "grad_norm": 0.5837286114692688,
+      "learning_rate": 0.0002,
+      "loss": 0.8804,
+      "step": 5210
+    },
+    {
+      "epoch": 3.8283828382838285,
+      "grad_norm": 1.2422157526016235,
+      "learning_rate": 0.0002,
+      "loss": 0.8369,
+      "step": 5220
+    },
+    {
+      "epoch": 3.8357169050238356,
+      "grad_norm": 0.6589220762252808,
+      "learning_rate": 0.0002,
+      "loss": 0.8431,
+      "step": 5230
+    },
+    {
+      "epoch": 3.843050971763843,
+      "grad_norm": 0.8567556142807007,
+      "learning_rate": 0.0002,
+      "loss": 0.7686,
+      "step": 5240
+    },
+    {
+      "epoch": 3.8503850385038505,
+      "grad_norm": 0.6490627527236938,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8577191052438575,
+      "grad_norm": 0.620232880115509,
+      "learning_rate": 0.0002,
+      "loss": 0.7386,
+      "step": 5260
+    },
+    {
+      "epoch": 3.865053171983865,
+      "grad_norm": 0.7685128450393677,
+      "learning_rate": 0.0002,
+      "loss": 0.9192,
+      "step": 5270
+    },
+    {
+      "epoch": 3.8723872387238725,
+      "grad_norm": 0.8113296627998352,
+      "learning_rate": 0.0002,
+      "loss": 0.872,
+      "step": 5280
+    },
+    {
+      "epoch": 3.87972130546388,
+      "grad_norm": 0.8092675805091858,
+      "learning_rate": 0.0002,
+      "loss": 0.7156,
+      "step": 5290
+    },
+    {
+      "epoch": 3.887055372203887,
+      "grad_norm": 0.583570122718811,
+      "learning_rate": 0.0002,
+      "loss": 0.7325,
+      "step": 5300
+    },
+    {
+      "epoch": 3.8943894389438944,
+      "grad_norm": 1.712363600730896,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 5310
+    },
+    {
+      "epoch": 3.9017235056839015,
+      "grad_norm": 0.6673534512519836,
+      "learning_rate": 0.0002,
+      "loss": 0.7537,
+      "step": 5320
+    },
+    {
+      "epoch": 3.909057572423909,
+      "grad_norm": 1.9770312309265137,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 5330
+    },
+    {
+      "epoch": 3.9163916391639164,
+      "grad_norm": 0.6430999636650085,
+      "learning_rate": 0.0002,
+      "loss": 0.8793,
+      "step": 5340
+    },
+    {
+      "epoch": 3.923725705903924,
+      "grad_norm": 1.0159571170806885,
+      "learning_rate": 0.0002,
+      "loss": 0.839,
+      "step": 5350
+    },
+    {
+      "epoch": 3.931059772643931,
+      "grad_norm": 0.8607584834098816,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 5360
+    },
+    {
+      "epoch": 3.9383938393839384,
+      "grad_norm": 0.6967900991439819,
+      "learning_rate": 0.0002,
+      "loss": 0.7261,
+      "step": 5370
+    },
+    {
+      "epoch": 3.945727906123946,
+      "grad_norm": 0.7683077454566956,
+      "learning_rate": 0.0002,
+      "loss": 0.8456,
+      "step": 5380
+    },
+    {
+      "epoch": 3.953061972863953,
+      "grad_norm": 0.6805762648582458,
+      "learning_rate": 0.0002,
+      "loss": 0.7682,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9603960396039604,
+      "grad_norm": 0.7033619284629822,
+      "learning_rate": 0.0002,
+      "loss": 0.7746,
+      "step": 5400
+    },
+    {
+      "epoch": 3.967730106343968,
+      "grad_norm": 0.966112494468689,
+      "learning_rate": 0.0002,
+      "loss": 0.8393,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9750641730839753,
+      "grad_norm": 0.8467881083488464,
+      "learning_rate": 0.0002,
+      "loss": 0.8316,
+      "step": 5420
+    },
+    {
+      "epoch": 3.9823982398239823,
+      "grad_norm": 0.8005317449569702,
+      "learning_rate": 0.0002,
+      "loss": 0.8084,
+      "step": 5430
+    },
+    {
+      "epoch": 3.98973230656399,
+      "grad_norm": 1.1615241765975952,
+      "learning_rate": 0.0002,
+      "loss": 0.7168,
+      "step": 5440
+    },
+    {
+      "epoch": 3.997066373303997,
+      "grad_norm": 0.6121614575386047,
+      "learning_rate": 0.0002,
+      "loss": 0.8263,
+      "step": 5450
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.1834222078323364,
+      "eval_runtime": 32.7569,
+      "eval_samples_per_second": 13.158,
+      "eval_steps_per_second": 1.649,
+      "step": 5454
+    },
+    {
+      "epoch": 4.004400440044004,
+      "grad_norm": 0.6055727005004883,
+      "learning_rate": 0.0002,
+      "loss": 0.7267,
+      "step": 5460
+    },
+    {
+      "epoch": 4.011734506784012,
+      "grad_norm": 0.8232647180557251,
+      "learning_rate": 0.0002,
+      "loss": 0.5766,
+      "step": 5470
+    },
+    {
+      "epoch": 4.019068573524019,
+      "grad_norm": 0.7739192247390747,
+      "learning_rate": 0.0002,
+      "loss": 0.6489,
+      "step": 5480
+    },
+    {
+      "epoch": 4.026402640264027,
+      "grad_norm": 0.6264950633049011,
+      "learning_rate": 0.0002,
+      "loss": 0.5978,
+      "step": 5490
+    },
+    {
+      "epoch": 4.033736707004033,
+      "grad_norm": 1.4798702001571655,
+      "learning_rate": 0.0002,
+      "loss": 0.6392,
+      "step": 5500
+    },
+    {
+      "epoch": 4.041070773744041,
+      "grad_norm": 0.9538470506668091,
+      "learning_rate": 0.0002,
+      "loss": 0.6143,
+      "step": 5510
+    },
+    {
+      "epoch": 4.048404840484048,
+      "grad_norm": 0.834561288356781,
+      "learning_rate": 0.0002,
+      "loss": 0.6056,
+      "step": 5520
+    },
+    {
+      "epoch": 4.055738907224056,
+      "grad_norm": 0.6407850384712219,
+      "learning_rate": 0.0002,
+      "loss": 0.6077,
+      "step": 5530
+    },
+    {
+      "epoch": 4.063072973964063,
+      "grad_norm": 0.9035961627960205,
+      "learning_rate": 0.0002,
+      "loss": 0.6733,
+      "step": 5540
+    },
+    {
+      "epoch": 4.070407040704071,
+      "grad_norm": 0.842812716960907,
+      "learning_rate": 0.0002,
+      "loss": 0.5854,
+      "step": 5550
+    },
+    {
+      "epoch": 4.077741107444078,
+      "grad_norm": 0.8197882175445557,
+      "learning_rate": 0.0002,
+      "loss": 0.654,
+      "step": 5560
+    },
+    {
+      "epoch": 4.085075174184085,
+      "grad_norm": 0.8652673959732056,
+      "learning_rate": 0.0002,
+      "loss": 0.5919,
+      "step": 5570
+    },
+    {
+      "epoch": 4.092409240924092,
+      "grad_norm": 0.8048318028450012,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 5580
+    },
+    {
+      "epoch": 4.0997433076641,
+      "grad_norm": 0.9604969024658203,
+      "learning_rate": 0.0002,
+      "loss": 0.6487,
+      "step": 5590
+    },
+    {
+      "epoch": 4.107077374404107,
+      "grad_norm": 1.244756817817688,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 5600
+    },
+    {
+      "epoch": 4.114411441144115,
+      "grad_norm": 0.7975269556045532,
+      "learning_rate": 0.0002,
+      "loss": 0.6489,
+      "step": 5610
+    },
+    {
+      "epoch": 4.121745507884122,
+      "grad_norm": 0.6130099296569824,
+      "learning_rate": 0.0002,
+      "loss": 0.6445,
+      "step": 5620
+    },
+    {
+      "epoch": 4.129079574624129,
+      "grad_norm": 0.7793202996253967,
+      "learning_rate": 0.0002,
+      "loss": 0.6024,
+      "step": 5630
+    },
+    {
+      "epoch": 4.136413641364136,
+      "grad_norm": 1.187238335609436,
+      "learning_rate": 0.0002,
+      "loss": 0.5723,
+      "step": 5640
+    },
+    {
+      "epoch": 4.143747708104144,
+      "grad_norm": 0.8450375199317932,
+      "learning_rate": 0.0002,
+      "loss": 0.6385,
+      "step": 5650
+    },
+    {
+      "epoch": 4.151081774844151,
+      "grad_norm": 0.9006940126419067,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 5660
+    },
+    {
+      "epoch": 4.158415841584159,
+      "grad_norm": 0.9447154998779297,
+      "learning_rate": 0.0002,
+      "loss": 0.6179,
+      "step": 5670
+    },
+    {
+      "epoch": 4.165749908324166,
+      "grad_norm": 0.798032283782959,
+      "learning_rate": 0.0002,
+      "loss": 0.6476,
+      "step": 5680
+    },
+    {
+      "epoch": 4.1730839750641735,
+      "grad_norm": 0.65578693151474,
+      "learning_rate": 0.0002,
+      "loss": 0.6666,
+      "step": 5690
+    },
+    {
+      "epoch": 4.18041804180418,
+      "grad_norm": 1.0864700078964233,
+      "learning_rate": 0.0002,
+      "loss": 0.701,
+      "step": 5700
+    },
+    {
+      "epoch": 4.187752108544188,
+      "grad_norm": 0.7344121932983398,
+      "learning_rate": 0.0002,
+      "loss": 0.6895,
+      "step": 5710
+    },
+    {
+      "epoch": 4.195086175284195,
+      "grad_norm": 0.9722456932067871,
+      "learning_rate": 0.0002,
+      "loss": 0.6659,
+      "step": 5720
+    },
+    {
+      "epoch": 4.2024202420242025,
+      "grad_norm": 1.263814926147461,
+      "learning_rate": 0.0002,
+      "loss": 0.6887,
+      "step": 5730
+    },
+    {
+      "epoch": 4.20975430876421,
+      "grad_norm": 0.9622581005096436,
+      "learning_rate": 0.0002,
+      "loss": 0.608,
+      "step": 5740
+    },
+    {
+      "epoch": 4.2170883755042174,
+      "grad_norm": 0.8497143387794495,
+      "learning_rate": 0.0002,
+      "loss": 0.6221,
+      "step": 5750
+    },
+    {
+      "epoch": 4.224422442244224,
+      "grad_norm": 0.8248446583747864,
+      "learning_rate": 0.0002,
+      "loss": 0.6322,
+      "step": 5760
+    },
+    {
+      "epoch": 4.2317565089842315,
+      "grad_norm": 1.2544798851013184,
+      "learning_rate": 0.0002,
+      "loss": 0.6045,
+      "step": 5770
+    },
+    {
+      "epoch": 4.239090575724239,
+      "grad_norm": 0.8224676251411438,
+      "learning_rate": 0.0002,
+      "loss": 0.641,
+      "step": 5780
+    },
+    {
+      "epoch": 4.2464246424642464,
+      "grad_norm": 0.8924877047538757,
+      "learning_rate": 0.0002,
+      "loss": 0.6399,
+      "step": 5790
+    },
+    {
+      "epoch": 4.253758709204254,
+      "grad_norm": 0.8545848727226257,
+      "learning_rate": 0.0002,
+      "loss": 0.6845,
+      "step": 5800
+    },
+    {
+      "epoch": 4.261092775944261,
+      "grad_norm": 0.8081067800521851,
+      "learning_rate": 0.0002,
+      "loss": 0.6669,
+      "step": 5810
+    },
+    {
+      "epoch": 4.268426842684269,
+      "grad_norm": 0.7111002802848816,
+      "learning_rate": 0.0002,
+      "loss": 0.6149,
+      "step": 5820
+    },
+    {
+      "epoch": 4.2757609094242754,
+      "grad_norm": 0.8696979880332947,
+      "learning_rate": 0.0002,
+      "loss": 0.6343,
+      "step": 5830
+    },
+    {
+      "epoch": 4.283094976164283,
+      "grad_norm": 0.821401834487915,
+      "learning_rate": 0.0002,
+      "loss": 0.6384,
+      "step": 5840
+    },
+    {
+      "epoch": 4.29042904290429,
+      "grad_norm": 0.888908326625824,
+      "learning_rate": 0.0002,
+      "loss": 0.6912,
+      "step": 5850
+    },
+    {
+      "epoch": 4.297763109644298,
+      "grad_norm": 1.9380123615264893,
+      "learning_rate": 0.0002,
+      "loss": 0.6061,
+      "step": 5860
+    },
+    {
+      "epoch": 4.305097176384305,
+      "grad_norm": 1.121774435043335,
+      "learning_rate": 0.0002,
+      "loss": 0.6766,
+      "step": 5870
+    },
+    {
+      "epoch": 4.312431243124313,
+      "grad_norm": 0.9238282442092896,
+      "learning_rate": 0.0002,
+      "loss": 0.7205,
+      "step": 5880
+    },
+    {
+      "epoch": 4.319765309864319,
+      "grad_norm": 0.7321620583534241,
+      "learning_rate": 0.0002,
+      "loss": 0.6351,
+      "step": 5890
+    },
+    {
+      "epoch": 4.327099376604327,
+      "grad_norm": 0.8739548325538635,
+      "learning_rate": 0.0002,
+      "loss": 0.6404,
+      "step": 5900
+    },
+    {
+      "epoch": 4.334433443344334,
+      "grad_norm": 0.9686012268066406,
+      "learning_rate": 0.0002,
+      "loss": 0.5892,
+      "step": 5910
+    },
+    {
+      "epoch": 4.341767510084342,
+      "grad_norm": 0.9033839106559753,
+      "learning_rate": 0.0002,
+      "loss": 0.641,
+      "step": 5920
+    },
+    {
+      "epoch": 4.349101576824349,
+      "grad_norm": 0.8131115436553955,
+      "learning_rate": 0.0002,
+      "loss": 0.6456,
+      "step": 5930
+    },
+    {
+      "epoch": 4.356435643564357,
+      "grad_norm": 0.8942412734031677,
+      "learning_rate": 0.0002,
+      "loss": 0.5826,
+      "step": 5940
+    },
+    {
+      "epoch": 4.363769710304364,
+      "grad_norm": 0.8439112901687622,
+      "learning_rate": 0.0002,
+      "loss": 0.7336,
+      "step": 5950
+    },
+    {
+      "epoch": 4.371103777044371,
+      "grad_norm": 0.9176713228225708,
+      "learning_rate": 0.0002,
+      "loss": 0.6537,
+      "step": 5960
+    },
+    {
+      "epoch": 4.378437843784378,
+      "grad_norm": 0.6799634695053101,
+      "learning_rate": 0.0002,
+      "loss": 0.6792,
+      "step": 5970
+    },
+    {
+      "epoch": 4.385771910524386,
+      "grad_norm": 1.0435824394226074,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 5980
+    },
+    {
+      "epoch": 4.393105977264393,
+      "grad_norm": 0.997937798500061,
+      "learning_rate": 0.0002,
+      "loss": 0.68,
+      "step": 5990
+    },
+    {
+      "epoch": 4.400440044004401,
+      "grad_norm": 1.0308842658996582,
+      "learning_rate": 0.0002,
+      "loss": 0.6604,
+      "step": 6000
+    },
+    {
+      "epoch": 4.407774110744408,
+      "grad_norm": 1.3683775663375854,
+      "learning_rate": 0.0002,
+      "loss": 0.6402,
+      "step": 6010
+    },
+    {
+      "epoch": 4.415108177484415,
+      "grad_norm": 0.7569534182548523,
+      "learning_rate": 0.0002,
+      "loss": 0.7027,
+      "step": 6020
+    },
+    {
+      "epoch": 4.422442244224422,
+      "grad_norm": 1.089978575706482,
+      "learning_rate": 0.0002,
+      "loss": 0.5949,
+      "step": 6030
+    },
+    {
+      "epoch": 4.42977631096443,
+      "grad_norm": 0.7522459626197815,
+      "learning_rate": 0.0002,
+      "loss": 0.6353,
+      "step": 6040
+    },
+    {
+      "epoch": 4.437110377704437,
+      "grad_norm": 0.6709823608398438,
+      "learning_rate": 0.0002,
+      "loss": 0.5852,
+      "step": 6050
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.6992089748382568,
+      "learning_rate": 0.0002,
+      "loss": 0.6718,
+      "step": 6060
+    },
+    {
+      "epoch": 4.451778511184452,
+      "grad_norm": 1.0182931423187256,
+      "learning_rate": 0.0002,
+      "loss": 0.6933,
+      "step": 6070
+    },
+    {
+      "epoch": 4.459112577924459,
+      "grad_norm": 1.0685160160064697,
+      "learning_rate": 0.0002,
+      "loss": 0.6255,
+      "step": 6080
+    },
+    {
+      "epoch": 4.466446644664466,
+      "grad_norm": 0.8295124769210815,
+      "learning_rate": 0.0002,
+      "loss": 0.6086,
+      "step": 6090
+    },
+    {
+      "epoch": 4.473780711404474,
+      "grad_norm": 1.1862998008728027,
+      "learning_rate": 0.0002,
+      "loss": 0.6359,
+      "step": 6100
+    },
+    {
+      "epoch": 4.481114778144481,
+      "grad_norm": 0.7400273084640503,
+      "learning_rate": 0.0002,
+      "loss": 0.638,
+      "step": 6110
+    },
+    {
+      "epoch": 4.488448844884489,
+      "grad_norm": 0.7098417282104492,
+      "learning_rate": 0.0002,
+      "loss": 0.6854,
+      "step": 6120
+    },
+    {
+      "epoch": 4.495782911624496,
+      "grad_norm": 0.9745053648948669,
+      "learning_rate": 0.0002,
+      "loss": 0.6976,
+      "step": 6130
+    },
+    {
+      "epoch": 4.503116978364503,
+      "grad_norm": 0.8638797998428345,
+      "learning_rate": 0.0002,
+      "loss": 0.605,
+      "step": 6140
+    },
+    {
+      "epoch": 4.51045104510451,
+      "grad_norm": 0.8291046619415283,
+      "learning_rate": 0.0002,
+      "loss": 0.6491,
+      "step": 6150
+    },
+    {
+      "epoch": 4.517785111844518,
+      "grad_norm": 1.0301737785339355,
+      "learning_rate": 0.0002,
+      "loss": 0.6457,
+      "step": 6160
+    },
+    {
+      "epoch": 4.525119178584525,
+      "grad_norm": 1.1996512413024902,
+      "learning_rate": 0.0002,
+      "loss": 0.6742,
+      "step": 6170
+    },
+    {
+      "epoch": 4.5324532453245325,
+      "grad_norm": 1.151038408279419,
+      "learning_rate": 0.0002,
+      "loss": 0.6484,
+      "step": 6180
+    },
+    {
+      "epoch": 4.53978731206454,
+      "grad_norm": 0.8385201096534729,
+      "learning_rate": 0.0002,
+      "loss": 0.668,
+      "step": 6190
+    },
+    {
+      "epoch": 4.5471213788045475,
+      "grad_norm": 0.8969188332557678,
+      "learning_rate": 0.0002,
+      "loss": 0.6381,
+      "step": 6200
+    },
+    {
+      "epoch": 4.554455445544555,
+      "grad_norm": 1.60659658908844,
+      "learning_rate": 0.0002,
+      "loss": 0.7141,
+      "step": 6210
+    },
+    {
+      "epoch": 4.5617895122845615,
+      "grad_norm": 0.9356731176376343,
+      "learning_rate": 0.0002,
+      "loss": 0.6388,
+      "step": 6220
+    },
+    {
+      "epoch": 4.569123579024569,
+      "grad_norm": 0.95856773853302,
+      "learning_rate": 0.0002,
+      "loss": 0.7393,
+      "step": 6230
+    },
+    {
+      "epoch": 4.5764576457645765,
+      "grad_norm": 1.1162524223327637,
+      "learning_rate": 0.0002,
+      "loss": 0.6554,
+      "step": 6240
+    },
+    {
+      "epoch": 4.583791712504584,
+      "grad_norm": 0.8809238076210022,
+      "learning_rate": 0.0002,
+      "loss": 0.6012,
+      "step": 6250
+    },
+    {
+      "epoch": 4.591125779244591,
+      "grad_norm": 0.890738844871521,
+      "learning_rate": 0.0002,
+      "loss": 0.648,
+      "step": 6260
+    },
+    {
+      "epoch": 4.598459845984598,
+      "grad_norm": 0.918684720993042,
+      "learning_rate": 0.0002,
+      "loss": 0.6663,
+      "step": 6270
+    },
+    {
+      "epoch": 4.6057939127246055,
+      "grad_norm": 0.8156296610832214,
+      "learning_rate": 0.0002,
+      "loss": 0.5992,
+      "step": 6280
+    },
+    {
+      "epoch": 4.613127979464613,
+      "grad_norm": 1.046634316444397,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 6290
+    },
+    {
+      "epoch": 4.62046204620462,
+      "grad_norm": 0.7725525498390198,
+      "learning_rate": 0.0002,
+      "loss": 0.7023,
+      "step": 6300
+    },
+    {
+      "epoch": 4.627796112944628,
+      "grad_norm": 0.9992046356201172,
+      "learning_rate": 0.0002,
+      "loss": 0.6414,
+      "step": 6310
+    },
+    {
+      "epoch": 4.635130179684635,
+      "grad_norm": 0.8480095267295837,
+      "learning_rate": 0.0002,
+      "loss": 0.6201,
+      "step": 6320
+    },
+    {
+      "epoch": 4.642464246424643,
+      "grad_norm": 0.7061955332756042,
+      "learning_rate": 0.0002,
+      "loss": 0.6869,
+      "step": 6330
+    },
+    {
+      "epoch": 4.649798313164649,
+      "grad_norm": 1.0354212522506714,
+      "learning_rate": 0.0002,
+      "loss": 0.6828,
+      "step": 6340
+    },
+    {
+      "epoch": 4.657132379904657,
+      "grad_norm": 1.0081377029418945,
+      "learning_rate": 0.0002,
+      "loss": 0.6651,
+      "step": 6350
+    },
+    {
+      "epoch": 4.664466446644664,
+      "grad_norm": 1.2904249429702759,
+      "learning_rate": 0.0002,
+      "loss": 0.726,
+      "step": 6360
+    },
+    {
+      "epoch": 4.671800513384672,
+      "grad_norm": 0.9248910546302795,
+      "learning_rate": 0.0002,
+      "loss": 0.7148,
+      "step": 6370
+    },
+    {
+      "epoch": 4.679134580124679,
+      "grad_norm": 0.9907804131507874,
+      "learning_rate": 0.0002,
+      "loss": 0.6961,
+      "step": 6380
+    },
+    {
+      "epoch": 4.686468646864687,
+      "grad_norm": 1.201143741607666,
+      "learning_rate": 0.0002,
+      "loss": 0.6163,
+      "step": 6390
+    },
+    {
+      "epoch": 4.693802713604693,
+      "grad_norm": 0.8709394335746765,
+      "learning_rate": 0.0002,
+      "loss": 0.6762,
+      "step": 6400
+    },
+    {
+      "epoch": 4.701136780344701,
+      "grad_norm": 0.7468608021736145,
+      "learning_rate": 0.0002,
+      "loss": 0.7217,
+      "step": 6410
+    },
+    {
+      "epoch": 4.708470847084708,
+      "grad_norm": 0.8607903718948364,
+      "learning_rate": 0.0002,
+      "loss": 0.6548,
+      "step": 6420
+    },
+    {
+      "epoch": 4.715804913824716,
+      "grad_norm": 0.9840512871742249,
+      "learning_rate": 0.0002,
+      "loss": 0.6449,
+      "step": 6430
+    },
+    {
+      "epoch": 4.723138980564723,
+      "grad_norm": 0.8328204154968262,
+      "learning_rate": 0.0002,
+      "loss": 0.685,
+      "step": 6440
+    },
+    {
+      "epoch": 4.730473047304731,
+      "grad_norm": 0.924505352973938,
+      "learning_rate": 0.0002,
+      "loss": 0.697,
+      "step": 6450
+    },
+    {
+      "epoch": 4.737807114044738,
+      "grad_norm": 0.8897685408592224,
+      "learning_rate": 0.0002,
+      "loss": 0.7422,
+      "step": 6460
+    },
+    {
+      "epoch": 4.745141180784745,
+      "grad_norm": 0.9605024456977844,
+      "learning_rate": 0.0002,
+      "loss": 0.6842,
+      "step": 6470
+    },
+    {
+      "epoch": 4.752475247524752,
+      "grad_norm": 0.8150759935379028,
+      "learning_rate": 0.0002,
+      "loss": 0.6488,
+      "step": 6480
+    },
+    {
+      "epoch": 4.75980931426476,
+      "grad_norm": 0.8128412961959839,
+      "learning_rate": 0.0002,
+      "loss": 0.6606,
+      "step": 6490
+    },
+    {
+      "epoch": 4.767143381004767,
+      "grad_norm": 0.7381404638290405,
+      "learning_rate": 0.0002,
+      "loss": 0.6729,
+      "step": 6500
+    },
+    {
+      "epoch": 4.774477447744775,
+      "grad_norm": 1.0565853118896484,
+      "learning_rate": 0.0002,
+      "loss": 0.6713,
+      "step": 6510
+    },
+    {
+      "epoch": 4.781811514484782,
+      "grad_norm": 0.9298134446144104,
+      "learning_rate": 0.0002,
+      "loss": 0.6496,
+      "step": 6520
+    },
+    {
+      "epoch": 4.789145581224789,
+      "grad_norm": 1.0145525932312012,
+      "learning_rate": 0.0002,
+      "loss": 0.7279,
+      "step": 6530
+    },
+    {
+      "epoch": 4.796479647964796,
+      "grad_norm": 0.92259681224823,
+      "learning_rate": 0.0002,
+      "loss": 0.5986,
+      "step": 6540
+    },
+    {
+      "epoch": 4.803813714704804,
+      "grad_norm": 0.7881024479866028,
+      "learning_rate": 0.0002,
+      "loss": 0.63,
+      "step": 6550
+    },
+    {
+      "epoch": 4.811147781444811,
+      "grad_norm": 1.4935206174850464,
+      "learning_rate": 0.0002,
+      "loss": 0.7134,
+      "step": 6560
+    },
+    {
+      "epoch": 4.818481848184819,
+      "grad_norm": 0.8612369298934937,
+      "learning_rate": 0.0002,
+      "loss": 0.6695,
+      "step": 6570
+    },
+    {
+      "epoch": 4.825815914924826,
+      "grad_norm": 1.0118653774261475,
+      "learning_rate": 0.0002,
+      "loss": 0.779,
+      "step": 6580
+    },
+    {
+      "epoch": 4.833149981664834,
+      "grad_norm": 1.1303809881210327,
+      "learning_rate": 0.0002,
+      "loss": 0.6991,
+      "step": 6590
+    },
+    {
+      "epoch": 4.84048404840484,
+      "grad_norm": 0.9112492203712463,
+      "learning_rate": 0.0002,
+      "loss": 0.7887,
+      "step": 6600
+    },
+    {
+      "epoch": 4.847818115144848,
+      "grad_norm": 0.864762544631958,
+      "learning_rate": 0.0002,
+      "loss": 0.7699,
+      "step": 6610
+    },
+    {
+      "epoch": 4.855152181884855,
+      "grad_norm": 0.9090572595596313,
+      "learning_rate": 0.0002,
+      "loss": 0.7347,
+      "step": 6620
+    },
+    {
+      "epoch": 4.862486248624863,
+      "grad_norm": 1.014953374862671,
+      "learning_rate": 0.0002,
+      "loss": 0.6608,
+      "step": 6630
+    },
+    {
+      "epoch": 4.86982031536487,
+      "grad_norm": 1.0702149868011475,
+      "learning_rate": 0.0002,
+      "loss": 0.6429,
+      "step": 6640
+    },
+    {
+      "epoch": 4.8771543821048775,
+      "grad_norm": 1.002135157585144,
+      "learning_rate": 0.0002,
+      "loss": 0.6943,
+      "step": 6650
+    },
+    {
+      "epoch": 4.884488448844884,
+      "grad_norm": 0.862545907497406,
+      "learning_rate": 0.0002,
+      "loss": 0.7225,
+      "step": 6660
+    },
+    {
+      "epoch": 4.891822515584892,
+      "grad_norm": 0.7302131056785583,
+      "learning_rate": 0.0002,
+      "loss": 0.6206,
+      "step": 6670
+    },
+    {
+      "epoch": 4.899156582324899,
+      "grad_norm": 0.8380730152130127,
+      "learning_rate": 0.0002,
+      "loss": 0.7175,
+      "step": 6680
+    },
+    {
+      "epoch": 4.9064906490649065,
+      "grad_norm": 0.7956018447875977,
+      "learning_rate": 0.0002,
+      "loss": 0.645,
+      "step": 6690
+    },
+    {
+      "epoch": 4.913824715804914,
+      "grad_norm": 0.6717583537101746,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 6700
+    },
+    {
+      "epoch": 4.9211587825449215,
+      "grad_norm": 1.09099280834198,
+      "learning_rate": 0.0002,
+      "loss": 0.6942,
+      "step": 6710
+    },
+    {
+      "epoch": 4.928492849284929,
+      "grad_norm": 0.8589889407157898,
+      "learning_rate": 0.0002,
+      "loss": 0.7533,
+      "step": 6720
+    },
+    {
+      "epoch": 4.9358269160249355,
+      "grad_norm": 1.0046314001083374,
+      "learning_rate": 0.0002,
+      "loss": 0.66,
+      "step": 6730
+    },
+    {
+      "epoch": 4.943160982764943,
+      "grad_norm": 0.8559659123420715,
+      "learning_rate": 0.0002,
+      "loss": 0.6864,
+      "step": 6740
+    },
+    {
+      "epoch": 4.9504950495049505,
+      "grad_norm": 0.8588525652885437,
+      "learning_rate": 0.0002,
+      "loss": 0.6847,
+      "step": 6750
+    },
+    {
+      "epoch": 4.957829116244958,
+      "grad_norm": 0.9192708134651184,
+      "learning_rate": 0.0002,
+      "loss": 0.6428,
+      "step": 6760
+    },
+    {
+      "epoch": 4.965163182984965,
+      "grad_norm": 1.051398754119873,
+      "learning_rate": 0.0002,
+      "loss": 0.6873,
+      "step": 6770
+    },
+    {
+      "epoch": 4.972497249724973,
+      "grad_norm": 0.9111362099647522,
+      "learning_rate": 0.0002,
+      "loss": 0.7249,
+      "step": 6780
+    },
+    {
+      "epoch": 4.9798313164649795,
+      "grad_norm": 0.7305638194084167,
+      "learning_rate": 0.0002,
+      "loss": 0.7613,
+      "step": 6790
+    },
+    {
+      "epoch": 4.987165383204987,
+      "grad_norm": 1.118837594985962,
+      "learning_rate": 0.0002,
+      "loss": 0.6747,
+      "step": 6800
+    },
+    {
+      "epoch": 4.994499449944994,
+      "grad_norm": 0.9075239300727844,
+      "learning_rate": 0.0002,
+      "loss": 0.6412,
+      "step": 6810
+    },
+    {
+      "epoch": 4.999633296662999,
+      "eval_loss": 1.2361247539520264,
+      "eval_runtime": 32.7325,
+      "eval_samples_per_second": 13.167,
+      "eval_steps_per_second": 1.65,
+      "step": 6817
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10904,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "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": 3.1549863499923456e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..2868cff7027115396e695775cacd838522aca295
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-6817/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b12b6f6817632087b5a5e37d744e25312b96e839de5005320b96bc0c2473c41f
+size 5624
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/README.md b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# 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 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "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": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2f84d0c20b043a302eae9c8985fa4b100bc27236
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc82fdb16bc9320508ca18eb240c5dc89509a062e490c9a6ec2f233ff61a96a6
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..704898495d039c14404a6c9ddccd0d6fb21c9cfa
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13e882cee82cf7301b2f56c95a7c530935d8b933b312beb0d6854b0d7d72c702
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..063e77d95a8152895f0ef43259d8b615ce355576
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aca658d9e71a3ab50f947b897f388acf6699ad2bd6b92adae6468a2433e3aa3d
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..668af33cb4c86c5dba1e08264d59f56026f1d3ae
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fe03bf7018c0aa325897c70d0d73733264f489b69d7fd26965086e90396ce58
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..233d2cad79f5f691aa15964b0f5532581caea619
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/trainer_state.json
@@ -0,0 +1,5807 @@
+{
+  "best_metric": 1.1534006595611572,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 8181,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007334066740007334,
+      "grad_norm": 0.47521963715553284,
+      "learning_rate": 0.0002,
+      "loss": 1.9722,
+      "step": 10
+    },
+    {
+      "epoch": 0.014668133480014669,
+      "grad_norm": 0.5395162105560303,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 20
+    },
+    {
+      "epoch": 0.022002200220022004,
+      "grad_norm": 0.4305780231952667,
+      "learning_rate": 0.0002,
+      "loss": 1.4202,
+      "step": 30
+    },
+    {
+      "epoch": 0.029336266960029337,
+      "grad_norm": 0.6938246488571167,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 40
+    },
+    {
+      "epoch": 0.03667033370003667,
+      "grad_norm": 1.5133819580078125,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 50
+    },
+    {
+      "epoch": 0.04400440044004401,
+      "grad_norm": 0.9173883199691772,
+      "learning_rate": 0.0002,
+      "loss": 1.3132,
+      "step": 60
+    },
+    {
+      "epoch": 0.05133846718005134,
+      "grad_norm": 0.4619861841201782,
+      "learning_rate": 0.0002,
+      "loss": 1.2844,
+      "step": 70
+    },
+    {
+      "epoch": 0.058672533920058674,
+      "grad_norm": 0.46118637919425964,
+      "learning_rate": 0.0002,
+      "loss": 1.2108,
+      "step": 80
+    },
+    {
+      "epoch": 0.066006600660066,
+      "grad_norm": 0.4468648135662079,
+      "learning_rate": 0.0002,
+      "loss": 1.3441,
+      "step": 90
+    },
+    {
+      "epoch": 0.07334066740007333,
+      "grad_norm": 0.46123769879341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1863,
+      "step": 100
+    },
+    {
+      "epoch": 0.08067473414008068,
+      "grad_norm": 0.4859139025211334,
+      "learning_rate": 0.0002,
+      "loss": 1.2772,
+      "step": 110
+    },
+    {
+      "epoch": 0.08800880088008801,
+      "grad_norm": 0.4384922385215759,
+      "learning_rate": 0.0002,
+      "loss": 1.2087,
+      "step": 120
+    },
+    {
+      "epoch": 0.09534286762009535,
+      "grad_norm": 0.39519360661506653,
+      "learning_rate": 0.0002,
+      "loss": 1.2927,
+      "step": 130
+    },
+    {
+      "epoch": 0.10267693436010268,
+      "grad_norm": 0.4049859344959259,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 140
+    },
+    {
+      "epoch": 0.11001100110011001,
+      "grad_norm": 0.4605638086795807,
+      "learning_rate": 0.0002,
+      "loss": 1.293,
+      "step": 150
+    },
+    {
+      "epoch": 0.11734506784011735,
+      "grad_norm": 0.4201928377151489,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 160
+    },
+    {
+      "epoch": 0.12467913458012468,
+      "grad_norm": 0.5367777347564697,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 170
+    },
+    {
+      "epoch": 0.132013201320132,
+      "grad_norm": 0.41752299666404724,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 180
+    },
+    {
+      "epoch": 0.13934726806013933,
+      "grad_norm": 0.31597763299942017,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 190
+    },
+    {
+      "epoch": 0.14668133480014667,
+      "grad_norm": 0.7468788623809814,
+      "learning_rate": 0.0002,
+      "loss": 1.2441,
+      "step": 200
+    },
+    {
+      "epoch": 0.15401540154015403,
+      "grad_norm": 0.3403034508228302,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 210
+    },
+    {
+      "epoch": 0.16134946828016136,
+      "grad_norm": 0.34240293502807617,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 220
+    },
+    {
+      "epoch": 0.1686835350201687,
+      "grad_norm": 0.356158971786499,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 230
+    },
+    {
+      "epoch": 0.17601760176017603,
+      "grad_norm": 0.3448857367038727,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 240
+    },
+    {
+      "epoch": 0.18335166850018336,
+      "grad_norm": 0.3475699722766876,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 250
+    },
+    {
+      "epoch": 0.1906857352401907,
+      "grad_norm": 0.2770358622074127,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 260
+    },
+    {
+      "epoch": 0.19801980198019803,
+      "grad_norm": 0.4310270845890045,
+      "learning_rate": 0.0002,
+      "loss": 1.2238,
+      "step": 270
+    },
+    {
+      "epoch": 0.20535386872020536,
+      "grad_norm": 0.335041880607605,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 280
+    },
+    {
+      "epoch": 0.2126879354602127,
+      "grad_norm": 0.3420602083206177,
+      "learning_rate": 0.0002,
+      "loss": 1.0959,
+      "step": 290
+    },
+    {
+      "epoch": 0.22002200220022003,
+      "grad_norm": 0.325001060962677,
+      "learning_rate": 0.0002,
+      "loss": 1.1232,
+      "step": 300
+    },
+    {
+      "epoch": 0.22735606894022736,
+      "grad_norm": 0.3027827739715576,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 310
+    },
+    {
+      "epoch": 0.2346901356802347,
+      "grad_norm": 0.435550719499588,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 320
+    },
+    {
+      "epoch": 0.24202420242024203,
+      "grad_norm": 0.3884522616863251,
+      "learning_rate": 0.0002,
+      "loss": 1.2045,
+      "step": 330
+    },
+    {
+      "epoch": 0.24935826916024936,
+      "grad_norm": 0.7736002206802368,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 340
+    },
+    {
+      "epoch": 0.2566923359002567,
+      "grad_norm": 0.35052821040153503,
+      "learning_rate": 0.0002,
+      "loss": 1.3606,
+      "step": 350
+    },
+    {
+      "epoch": 0.264026402640264,
+      "grad_norm": 0.3311890959739685,
+      "learning_rate": 0.0002,
+      "loss": 1.2129,
+      "step": 360
+    },
+    {
+      "epoch": 0.27136046938027136,
+      "grad_norm": 0.7473500370979309,
+      "learning_rate": 0.0002,
+      "loss": 1.2219,
+      "step": 370
+    },
+    {
+      "epoch": 0.27869453612027867,
+      "grad_norm": 0.3681875765323639,
+      "learning_rate": 0.0002,
+      "loss": 1.2712,
+      "step": 380
+    },
+    {
+      "epoch": 0.28602860286028603,
+      "grad_norm": 0.3764737844467163,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 390
+    },
+    {
+      "epoch": 0.29336266960029334,
+      "grad_norm": 0.4243989586830139,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 400
+    },
+    {
+      "epoch": 0.3006967363403007,
+      "grad_norm": 0.2658531963825226,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 410
+    },
+    {
+      "epoch": 0.30803080308030806,
+      "grad_norm": 0.3436793386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.1622,
+      "step": 420
+    },
+    {
+      "epoch": 0.31536486982031536,
+      "grad_norm": 0.5101129412651062,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 430
+    },
+    {
+      "epoch": 0.3226989365603227,
+      "grad_norm": 0.3319750726222992,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 440
+    },
+    {
+      "epoch": 0.33003300330033003,
+      "grad_norm": 0.385148286819458,
+      "learning_rate": 0.0002,
+      "loss": 1.1804,
+      "step": 450
+    },
+    {
+      "epoch": 0.3373670700403374,
+      "grad_norm": 0.3477935791015625,
+      "learning_rate": 0.0002,
+      "loss": 1.1808,
+      "step": 460
+    },
+    {
+      "epoch": 0.3447011367803447,
+      "grad_norm": 0.29748716950416565,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 470
+    },
+    {
+      "epoch": 0.35203520352035206,
+      "grad_norm": 0.34083324670791626,
+      "learning_rate": 0.0002,
+      "loss": 1.19,
+      "step": 480
+    },
+    {
+      "epoch": 0.35936927026035936,
+      "grad_norm": 0.36904552578926086,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 490
+    },
+    {
+      "epoch": 0.3667033370003667,
+      "grad_norm": 0.315483033657074,
+      "learning_rate": 0.0002,
+      "loss": 1.2223,
+      "step": 500
+    },
+    {
+      "epoch": 0.37403740374037403,
+      "grad_norm": 0.44897955656051636,
+      "learning_rate": 0.0002,
+      "loss": 1.1461,
+      "step": 510
+    },
+    {
+      "epoch": 0.3813714704803814,
+      "grad_norm": 0.3160701394081116,
+      "learning_rate": 0.0002,
+      "loss": 1.3035,
+      "step": 520
+    },
+    {
+      "epoch": 0.3887055372203887,
+      "grad_norm": 0.29584741592407227,
+      "learning_rate": 0.0002,
+      "loss": 1.3197,
+      "step": 530
+    },
+    {
+      "epoch": 0.39603960396039606,
+      "grad_norm": 0.5430002808570862,
+      "learning_rate": 0.0002,
+      "loss": 1.2983,
+      "step": 540
+    },
+    {
+      "epoch": 0.40337367070040336,
+      "grad_norm": 0.2908070683479309,
+      "learning_rate": 0.0002,
+      "loss": 1.2459,
+      "step": 550
+    },
+    {
+      "epoch": 0.4107077374404107,
+      "grad_norm": 0.35066530108451843,
+      "learning_rate": 0.0002,
+      "loss": 1.2384,
+      "step": 560
+    },
+    {
+      "epoch": 0.41804180418041803,
+      "grad_norm": 0.37588003277778625,
+      "learning_rate": 0.0002,
+      "loss": 1.1784,
+      "step": 570
+    },
+    {
+      "epoch": 0.4253758709204254,
+      "grad_norm": 0.3112126886844635,
+      "learning_rate": 0.0002,
+      "loss": 1.2334,
+      "step": 580
+    },
+    {
+      "epoch": 0.4327099376604327,
+      "grad_norm": 0.35577139258384705,
+      "learning_rate": 0.0002,
+      "loss": 1.1439,
+      "step": 590
+    },
+    {
+      "epoch": 0.44004400440044006,
+      "grad_norm": 0.31706422567367554,
+      "learning_rate": 0.0002,
+      "loss": 1.184,
+      "step": 600
+    },
+    {
+      "epoch": 0.44737807114044736,
+      "grad_norm": 0.3249092102050781,
+      "learning_rate": 0.0002,
+      "loss": 1.2081,
+      "step": 610
+    },
+    {
+      "epoch": 0.4547121378804547,
+      "grad_norm": 0.3842705488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.0824,
+      "step": 620
+    },
+    {
+      "epoch": 0.46204620462046203,
+      "grad_norm": 0.390991747379303,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 630
+    },
+    {
+      "epoch": 0.4693802713604694,
+      "grad_norm": 0.27532413601875305,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 640
+    },
+    {
+      "epoch": 0.4767143381004767,
+      "grad_norm": 0.31412816047668457,
+      "learning_rate": 0.0002,
+      "loss": 1.1058,
+      "step": 650
+    },
+    {
+      "epoch": 0.48404840484048406,
+      "grad_norm": 0.32117101550102234,
+      "learning_rate": 0.0002,
+      "loss": 1.1312,
+      "step": 660
+    },
+    {
+      "epoch": 0.49138247158049136,
+      "grad_norm": 0.3810010254383087,
+      "learning_rate": 0.0002,
+      "loss": 1.2423,
+      "step": 670
+    },
+    {
+      "epoch": 0.4987165383204987,
+      "grad_norm": 0.36289164423942566,
+      "learning_rate": 0.0002,
+      "loss": 1.1978,
+      "step": 680
+    },
+    {
+      "epoch": 0.506050605060506,
+      "grad_norm": 0.34458720684051514,
+      "learning_rate": 0.0002,
+      "loss": 1.2034,
+      "step": 690
+    },
+    {
+      "epoch": 0.5133846718005134,
+      "grad_norm": 0.32844600081443787,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 700
+    },
+    {
+      "epoch": 0.5207187385405208,
+      "grad_norm": 0.3144175708293915,
+      "learning_rate": 0.0002,
+      "loss": 1.0807,
+      "step": 710
+    },
+    {
+      "epoch": 0.528052805280528,
+      "grad_norm": 0.3898887634277344,
+      "learning_rate": 0.0002,
+      "loss": 1.1952,
+      "step": 720
+    },
+    {
+      "epoch": 0.5353868720205354,
+      "grad_norm": 1.3220758438110352,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 730
+    },
+    {
+      "epoch": 0.5427209387605427,
+      "grad_norm": 0.3635874390602112,
+      "learning_rate": 0.0002,
+      "loss": 1.227,
+      "step": 740
+    },
+    {
+      "epoch": 0.5500550055005501,
+      "grad_norm": 0.3138217628002167,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 750
+    },
+    {
+      "epoch": 0.5573890722405573,
+      "grad_norm": 0.4063207805156708,
+      "learning_rate": 0.0002,
+      "loss": 1.1516,
+      "step": 760
+    },
+    {
+      "epoch": 0.5647231389805647,
+      "grad_norm": 0.3926219940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 770
+    },
+    {
+      "epoch": 0.5720572057205721,
+      "grad_norm": 0.31954652070999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1726,
+      "step": 780
+    },
+    {
+      "epoch": 0.5793912724605794,
+      "grad_norm": 0.4248711168766022,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 790
+    },
+    {
+      "epoch": 0.5867253392005867,
+      "grad_norm": 0.643004834651947,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 800
+    },
+    {
+      "epoch": 0.594059405940594,
+      "grad_norm": 0.3479592800140381,
+      "learning_rate": 0.0002,
+      "loss": 1.1793,
+      "step": 810
+    },
+    {
+      "epoch": 0.6013934726806014,
+      "grad_norm": 0.4684754014015198,
+      "learning_rate": 0.0002,
+      "loss": 1.2426,
+      "step": 820
+    },
+    {
+      "epoch": 0.6087275394206088,
+      "grad_norm": 0.3739790916442871,
+      "learning_rate": 0.0002,
+      "loss": 1.2002,
+      "step": 830
+    },
+    {
+      "epoch": 0.6160616061606161,
+      "grad_norm": 0.40884748101234436,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 840
+    },
+    {
+      "epoch": 0.6233956729006234,
+      "grad_norm": 0.9722164273262024,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 850
+    },
+    {
+      "epoch": 0.6307297396406307,
+      "grad_norm": 0.42992347478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.3069,
+      "step": 860
+    },
+    {
+      "epoch": 0.6380638063806381,
+      "grad_norm": 0.36654195189476013,
+      "learning_rate": 0.0002,
+      "loss": 1.1339,
+      "step": 870
+    },
+    {
+      "epoch": 0.6453978731206454,
+      "grad_norm": 0.4113832116127014,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 880
+    },
+    {
+      "epoch": 0.6527319398606527,
+      "grad_norm": 0.2948838770389557,
+      "learning_rate": 0.0002,
+      "loss": 1.2163,
+      "step": 890
+    },
+    {
+      "epoch": 0.6600660066006601,
+      "grad_norm": 0.38330280780792236,
+      "learning_rate": 0.0002,
+      "loss": 1.1081,
+      "step": 900
+    },
+    {
+      "epoch": 0.6674000733406674,
+      "grad_norm": 0.4428867697715759,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 910
+    },
+    {
+      "epoch": 0.6747341400806748,
+      "grad_norm": 0.23659265041351318,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 920
+    },
+    {
+      "epoch": 0.682068206820682,
+      "grad_norm": 0.323685884475708,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 930
+    },
+    {
+      "epoch": 0.6894022735606894,
+      "grad_norm": 0.39157727360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 940
+    },
+    {
+      "epoch": 0.6967363403006968,
+      "grad_norm": 0.27189481258392334,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 950
+    },
+    {
+      "epoch": 0.7040704070407041,
+      "grad_norm": 0.529883861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.1033,
+      "step": 960
+    },
+    {
+      "epoch": 0.7114044737807114,
+      "grad_norm": 0.34758689999580383,
+      "learning_rate": 0.0002,
+      "loss": 1.139,
+      "step": 970
+    },
+    {
+      "epoch": 0.7187385405207187,
+      "grad_norm": 0.831749439239502,
+      "learning_rate": 0.0002,
+      "loss": 1.2197,
+      "step": 980
+    },
+    {
+      "epoch": 0.7260726072607261,
+      "grad_norm": 0.4438304007053375,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 990
+    },
+    {
+      "epoch": 0.7334066740007334,
+      "grad_norm": 0.33840006589889526,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7407407407407407,
+      "grad_norm": 0.3454797863960266,
+      "learning_rate": 0.0002,
+      "loss": 1.254,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7480748074807481,
+      "grad_norm": 0.38999441266059875,
+      "learning_rate": 0.0002,
+      "loss": 1.106,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7554088742207554,
+      "grad_norm": 0.2829911708831787,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7627429409607628,
+      "grad_norm": 0.36918163299560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 1040
+    },
+    {
+      "epoch": 0.77007700770077,
+      "grad_norm": 0.3415680229663849,
+      "learning_rate": 0.0002,
+      "loss": 1.3028,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7774110744407774,
+      "grad_norm": 0.2974182963371277,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7847451411807848,
+      "grad_norm": 0.3880919814109802,
+      "learning_rate": 0.0002,
+      "loss": 1.194,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7920792079207921,
+      "grad_norm": 0.33503302931785583,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7994132746607994,
+      "grad_norm": 0.3728407025337219,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8067473414008067,
+      "grad_norm": 0.3509373664855957,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8140814081408141,
+      "grad_norm": 0.42228564620018005,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8214154748808215,
+      "grad_norm": 0.313467800617218,
+      "learning_rate": 0.0002,
+      "loss": 1.1788,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8287495416208287,
+      "grad_norm": 0.3378850817680359,
+      "learning_rate": 0.0002,
+      "loss": 1.1971,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8360836083608361,
+      "grad_norm": 0.43200382590293884,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8434176751008434,
+      "grad_norm": 0.3309599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.3203,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8507517418408508,
+      "grad_norm": 0.3526846170425415,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1160
+    },
+    {
+      "epoch": 0.858085808580858,
+      "grad_norm": 1.2722247838974,
+      "learning_rate": 0.0002,
+      "loss": 1.0851,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8654198753208654,
+      "grad_norm": 0.34142059087753296,
+      "learning_rate": 0.0002,
+      "loss": 1.0785,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8727539420608728,
+      "grad_norm": 0.3805823028087616,
+      "learning_rate": 0.0002,
+      "loss": 1.2187,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8800880088008801,
+      "grad_norm": 0.3931232690811157,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8874220755408874,
+      "grad_norm": 0.2937372624874115,
+      "learning_rate": 0.0002,
+      "loss": 1.0948,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8947561422808947,
+      "grad_norm": 0.3757196366786957,
+      "learning_rate": 0.0002,
+      "loss": 1.1228,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9020902090209021,
+      "grad_norm": 0.3502705991268158,
+      "learning_rate": 0.0002,
+      "loss": 1.1222,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9094242757609095,
+      "grad_norm": 0.32758915424346924,
+      "learning_rate": 0.0002,
+      "loss": 1.2242,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9167583425009168,
+      "grad_norm": 0.37199416756629944,
+      "learning_rate": 0.0002,
+      "loss": 1.215,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9240924092409241,
+      "grad_norm": 0.3551490604877472,
+      "learning_rate": 0.0002,
+      "loss": 1.1225,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9314264759809314,
+      "grad_norm": 0.2859550714492798,
+      "learning_rate": 0.0002,
+      "loss": 1.1966,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9387605427209388,
+      "grad_norm": 0.427990585565567,
+      "learning_rate": 0.0002,
+      "loss": 1.2186,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9460946094609461,
+      "grad_norm": 0.33717992901802063,
+      "learning_rate": 0.0002,
+      "loss": 1.2848,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9534286762009534,
+      "grad_norm": 0.30225634574890137,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9607627429409608,
+      "grad_norm": 0.385821133852005,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9680968096809681,
+      "grad_norm": 0.35278066992759705,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9754308764209755,
+      "grad_norm": 0.49987098574638367,
+      "learning_rate": 0.0002,
+      "loss": 1.1071,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9827649431609827,
+      "grad_norm": 0.3842747211456299,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9900990099009901,
+      "grad_norm": 0.6274653673171997,
+      "learning_rate": 0.0002,
+      "loss": 1.0862,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9974330766409975,
+      "grad_norm": 0.5239808559417725,
+      "learning_rate": 0.0002,
+      "loss": 1.124,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9996332966629996,
+      "eval_loss": 1.1822267770767212,
+      "eval_runtime": 32.7389,
+      "eval_samples_per_second": 13.165,
+      "eval_steps_per_second": 1.649,
+      "step": 1363
+    },
+    {
+      "epoch": 1.0047671433810048,
+      "grad_norm": 0.45311301946640015,
+      "learning_rate": 0.0002,
+      "loss": 1.096,
+      "step": 1370
+    },
+    {
+      "epoch": 1.012101210121012,
+      "grad_norm": 0.29685574769973755,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0194352768610195,
+      "grad_norm": 0.3290937840938568,
+      "learning_rate": 0.0002,
+      "loss": 1.0302,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0267693436010268,
+      "grad_norm": 0.3801758587360382,
+      "learning_rate": 0.0002,
+      "loss": 1.0295,
+      "step": 1400
+    },
+    {
+      "epoch": 1.034103410341034,
+      "grad_norm": 0.794174313545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0414374770810415,
+      "grad_norm": 0.3854154646396637,
+      "learning_rate": 0.0002,
+      "loss": 1.2232,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0487715438210488,
+      "grad_norm": 0.32702451944351196,
+      "learning_rate": 0.0002,
+      "loss": 1.0652,
+      "step": 1430
+    },
+    {
+      "epoch": 1.056105610561056,
+      "grad_norm": 0.7815203666687012,
+      "learning_rate": 0.0002,
+      "loss": 1.1144,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0634396773010635,
+      "grad_norm": 0.3087436854839325,
+      "learning_rate": 0.0002,
+      "loss": 1.1316,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0707737440410707,
+      "grad_norm": 0.3847602903842926,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0781078107810782,
+      "grad_norm": 0.3693031370639801,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0854418775210855,
+      "grad_norm": 0.4111202359199524,
+      "learning_rate": 0.0002,
+      "loss": 1.0995,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0927759442610927,
+      "grad_norm": 0.41452381014823914,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1001100110011002,
+      "grad_norm": 0.3336445093154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1068,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1074440777411074,
+      "grad_norm": 0.3923407793045044,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1147781444811147,
+      "grad_norm": 0.46215683221817017,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1221122112211221,
+      "grad_norm": 0.3592156767845154,
+      "learning_rate": 0.0002,
+      "loss": 1.1133,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1294462779611294,
+      "grad_norm": 0.361110657453537,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1367803447011369,
+      "grad_norm": 0.5317131280899048,
+      "learning_rate": 0.0002,
+      "loss": 1.1553,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1441144114411441,
+      "grad_norm": 0.3882388174533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1514484781811514,
+      "grad_norm": 0.3259428143501282,
+      "learning_rate": 0.0002,
+      "loss": 1.0805,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1587825449211588,
+      "grad_norm": 0.410935640335083,
+      "learning_rate": 0.0002,
+      "loss": 1.1819,
+      "step": 1580
+    },
+    {
+      "epoch": 1.166116611661166,
+      "grad_norm": 0.44940185546875,
+      "learning_rate": 0.0002,
+      "loss": 1.1143,
+      "step": 1590
+    },
+    {
+      "epoch": 1.1734506784011733,
+      "grad_norm": 0.5106484293937683,
+      "learning_rate": 0.0002,
+      "loss": 1.0334,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1807847451411808,
+      "grad_norm": 0.6603665947914124,
+      "learning_rate": 0.0002,
+      "loss": 1.2376,
+      "step": 1610
+    },
+    {
+      "epoch": 1.188118811881188,
+      "grad_norm": 0.4799964129924774,
+      "learning_rate": 0.0002,
+      "loss": 1.1227,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1954528786211955,
+      "grad_norm": 0.4389883279800415,
+      "learning_rate": 0.0002,
+      "loss": 1.1191,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2027869453612028,
+      "grad_norm": 0.4188813269138336,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 1640
+    },
+    {
+      "epoch": 1.21012101210121,
+      "grad_norm": 0.7132157683372498,
+      "learning_rate": 0.0002,
+      "loss": 1.0605,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2174550788412175,
+      "grad_norm": 0.507480263710022,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2247891455812248,
+      "grad_norm": 0.9452332854270935,
+      "learning_rate": 0.0002,
+      "loss": 0.9948,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2321232123212322,
+      "grad_norm": 0.4121614992618561,
+      "learning_rate": 0.0002,
+      "loss": 1.0228,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2394572790612395,
+      "grad_norm": 0.34230247139930725,
+      "learning_rate": 0.0002,
+      "loss": 1.0366,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2467913458012467,
+      "grad_norm": 0.4026208817958832,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2541254125412542,
+      "grad_norm": 0.46673697233200073,
+      "learning_rate": 0.0002,
+      "loss": 1.0206,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2614594792812615,
+      "grad_norm": 0.38349825143814087,
+      "learning_rate": 0.0002,
+      "loss": 1.0827,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2687935460212687,
+      "grad_norm": 0.4049997627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.0356,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2761276127612762,
+      "grad_norm": 0.3417615294456482,
+      "learning_rate": 0.0002,
+      "loss": 0.9504,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2834616795012834,
+      "grad_norm": 0.4277614951133728,
+      "learning_rate": 0.0002,
+      "loss": 1.094,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2907957462412907,
+      "grad_norm": 0.5864202976226807,
+      "learning_rate": 0.0002,
+      "loss": 0.9938,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2981298129812981,
+      "grad_norm": 0.7097493410110474,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3054638797213054,
+      "grad_norm": 0.3145381212234497,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3127979464613129,
+      "grad_norm": 0.5116165280342102,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3201320132013201,
+      "grad_norm": 0.7469736337661743,
+      "learning_rate": 0.0002,
+      "loss": 1.0765,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3274660799413276,
+      "grad_norm": 0.32272255420684814,
+      "learning_rate": 0.0002,
+      "loss": 1.0663,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3348001466813348,
+      "grad_norm": 0.3534623086452484,
+      "learning_rate": 0.0002,
+      "loss": 0.9887,
+      "step": 1820
+    },
+    {
+      "epoch": 1.342134213421342,
+      "grad_norm": 0.36127907037734985,
+      "learning_rate": 0.0002,
+      "loss": 1.1628,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3494682801613496,
+      "grad_norm": 0.4072401523590088,
+      "learning_rate": 0.0002,
+      "loss": 1.0972,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3568023469013568,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.1267,
+      "step": 1850
+    },
+    {
+      "epoch": 1.364136413641364,
+      "grad_norm": 0.412883460521698,
+      "learning_rate": 0.0002,
+      "loss": 1.0173,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3714704803813715,
+      "grad_norm": 0.3735875189304352,
+      "learning_rate": 0.0002,
+      "loss": 1.0265,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3788045471213788,
+      "grad_norm": 0.39158159494400024,
+      "learning_rate": 0.0002,
+      "loss": 1.1061,
+      "step": 1880
+    },
+    {
+      "epoch": 1.386138613861386,
+      "grad_norm": 0.44431769847869873,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3934726806013935,
+      "grad_norm": 0.37772801518440247,
+      "learning_rate": 0.0002,
+      "loss": 1.0216,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4008067473414008,
+      "grad_norm": 0.4056641757488251,
+      "learning_rate": 0.0002,
+      "loss": 1.0674,
+      "step": 1910
+    },
+    {
+      "epoch": 1.408140814081408,
+      "grad_norm": 0.41612377762794495,
+      "learning_rate": 0.0002,
+      "loss": 1.0256,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4154748808214155,
+      "grad_norm": 0.41153013706207275,
+      "learning_rate": 0.0002,
+      "loss": 1.0467,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4228089475614227,
+      "grad_norm": 0.387845516204834,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4301430143014302,
+      "grad_norm": 0.3809587061405182,
+      "learning_rate": 0.0002,
+      "loss": 1.1094,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4374770810414375,
+      "grad_norm": 0.3625726103782654,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1960
+    },
+    {
+      "epoch": 1.444811147781445,
+      "grad_norm": 0.5294290781021118,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4521452145214522,
+      "grad_norm": 0.39975494146347046,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4594792812614594,
+      "grad_norm": 0.4181167185306549,
+      "learning_rate": 0.0002,
+      "loss": 0.9704,
+      "step": 1990
+    },
+    {
+      "epoch": 1.466813348001467,
+      "grad_norm": 0.42001503705978394,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4741474147414741,
+      "grad_norm": 0.4877578616142273,
+      "learning_rate": 0.0002,
+      "loss": 1.1266,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4814814814814814,
+      "grad_norm": 0.4050969183444977,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4888155482214889,
+      "grad_norm": 0.39068883657455444,
+      "learning_rate": 0.0002,
+      "loss": 1.0562,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4961496149614961,
+      "grad_norm": 0.421282559633255,
+      "learning_rate": 0.0002,
+      "loss": 1.0464,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5034836817015034,
+      "grad_norm": 0.47092297673225403,
+      "learning_rate": 0.0002,
+      "loss": 1.0532,
+      "step": 2050
+    },
+    {
+      "epoch": 1.5108177484415108,
+      "grad_norm": 0.39688974618911743,
+      "learning_rate": 0.0002,
+      "loss": 0.9348,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5181518151815183,
+      "grad_norm": 0.5529879331588745,
+      "learning_rate": 0.0002,
+      "loss": 1.08,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5254858819215253,
+      "grad_norm": 0.4879782199859619,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5328199486615328,
+      "grad_norm": 0.5517361164093018,
+      "learning_rate": 0.0002,
+      "loss": 1.0432,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5401540154015403,
+      "grad_norm": 0.44015637040138245,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5474880821415475,
+      "grad_norm": 0.5435167551040649,
+      "learning_rate": 0.0002,
+      "loss": 1.1873,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5548221488815548,
+      "grad_norm": 0.5714033246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1076,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5621562156215623,
+      "grad_norm": 0.31732529401779175,
+      "learning_rate": 0.0002,
+      "loss": 1.1107,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5694902823615695,
+      "grad_norm": 0.49068278074264526,
+      "learning_rate": 0.0002,
+      "loss": 1.0817,
+      "step": 2140
+    },
+    {
+      "epoch": 1.5768243491015768,
+      "grad_norm": 0.46851542592048645,
+      "learning_rate": 0.0002,
+      "loss": 1.0254,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5841584158415842,
+      "grad_norm": 0.5083092451095581,
+      "learning_rate": 0.0002,
+      "loss": 1.0623,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5914924825815915,
+      "grad_norm": 0.9822936058044434,
+      "learning_rate": 0.0002,
+      "loss": 1.0603,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5988265493215987,
+      "grad_norm": 0.4575989246368408,
+      "learning_rate": 0.0002,
+      "loss": 0.9986,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6061606160616062,
+      "grad_norm": 0.47444286942481995,
+      "learning_rate": 0.0002,
+      "loss": 1.1292,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6134946828016135,
+      "grad_norm": 0.7208226919174194,
+      "learning_rate": 0.0002,
+      "loss": 1.0136,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6208287495416207,
+      "grad_norm": 0.43791481852531433,
+      "learning_rate": 0.0002,
+      "loss": 1.15,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6281628162816282,
+      "grad_norm": 0.5245792865753174,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6354968830216357,
+      "grad_norm": 0.39289429783821106,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6428309497616427,
+      "grad_norm": 0.6106135845184326,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6501650165016502,
+      "grad_norm": 0.3722580671310425,
+      "learning_rate": 0.0002,
+      "loss": 1.0129,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6574990832416576,
+      "grad_norm": 0.3649403750896454,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6648331499816649,
+      "grad_norm": 0.46514248847961426,
+      "learning_rate": 0.0002,
+      "loss": 1.0037,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6721672167216721,
+      "grad_norm": 0.42034927010536194,
+      "learning_rate": 0.0002,
+      "loss": 1.0022,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6795012834616796,
+      "grad_norm": 0.45202910900115967,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6868353502016868,
+      "grad_norm": 0.36257603764533997,
+      "learning_rate": 0.0002,
+      "loss": 1.0866,
+      "step": 2300
+    },
+    {
+      "epoch": 1.694169416941694,
+      "grad_norm": 0.6340323090553284,
+      "learning_rate": 0.0002,
+      "loss": 1.0973,
+      "step": 2310
+    },
+    {
+      "epoch": 1.7015034836817016,
+      "grad_norm": 0.4352878928184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0615,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7088375504217088,
+      "grad_norm": 0.45029792189598083,
+      "learning_rate": 0.0002,
+      "loss": 1.0629,
+      "step": 2330
+    },
+    {
+      "epoch": 1.716171617161716,
+      "grad_norm": 0.3891315758228302,
+      "learning_rate": 0.0002,
+      "loss": 0.9621,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7235056839017235,
+      "grad_norm": 0.35180050134658813,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7308397506417308,
+      "grad_norm": 0.42367449402809143,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 2360
+    },
+    {
+      "epoch": 1.738173817381738,
+      "grad_norm": 0.4553675353527069,
+      "learning_rate": 0.0002,
+      "loss": 1.0376,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7455078841217455,
+      "grad_norm": 0.5944654941558838,
+      "learning_rate": 0.0002,
+      "loss": 1.1467,
+      "step": 2380
+    },
+    {
+      "epoch": 1.752841950861753,
+      "grad_norm": 0.3479664623737335,
+      "learning_rate": 0.0002,
+      "loss": 1.0548,
+      "step": 2390
+    },
+    {
+      "epoch": 1.76017601760176,
+      "grad_norm": 0.3585502505302429,
+      "learning_rate": 0.0002,
+      "loss": 1.0798,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7675100843417675,
+      "grad_norm": 0.4263346493244171,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 2410
+    },
+    {
+      "epoch": 1.774844151081775,
+      "grad_norm": 0.5476409196853638,
+      "learning_rate": 0.0002,
+      "loss": 1.054,
+      "step": 2420
+    },
+    {
+      "epoch": 1.7821782178217822,
+      "grad_norm": 0.3694186508655548,
+      "learning_rate": 0.0002,
+      "loss": 1.1615,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7895122845617895,
+      "grad_norm": 0.9185658693313599,
+      "learning_rate": 0.0002,
+      "loss": 1.1343,
+      "step": 2440
+    },
+    {
+      "epoch": 1.796846351301797,
+      "grad_norm": 0.7171908020973206,
+      "learning_rate": 0.0002,
+      "loss": 1.0764,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8041804180418042,
+      "grad_norm": 0.550658643245697,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 2460
+    },
+    {
+      "epoch": 1.8115144847818114,
+      "grad_norm": 0.4075568914413452,
+      "learning_rate": 0.0002,
+      "loss": 0.9975,
+      "step": 2470
+    },
+    {
+      "epoch": 1.818848551521819,
+      "grad_norm": 0.3790127635002136,
+      "learning_rate": 0.0002,
+      "loss": 1.0935,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8261826182618262,
+      "grad_norm": 0.3576384484767914,
+      "learning_rate": 0.0002,
+      "loss": 0.9839,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8335166850018334,
+      "grad_norm": 0.3919370770454407,
+      "learning_rate": 0.0002,
+      "loss": 1.1369,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8408507517418409,
+      "grad_norm": 0.485083669424057,
+      "learning_rate": 0.0002,
+      "loss": 0.9985,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8481848184818483,
+      "grad_norm": 0.4564347565174103,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8555188852218554,
+      "grad_norm": 0.3613106608390808,
+      "learning_rate": 0.0002,
+      "loss": 1.0944,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8628529519618628,
+      "grad_norm": 0.39600759744644165,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8701870187018703,
+      "grad_norm": 1.123499870300293,
+      "learning_rate": 0.0002,
+      "loss": 0.9453,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8775210854418776,
+      "grad_norm": 0.4612680673599243,
+      "learning_rate": 0.0002,
+      "loss": 1.0635,
+      "step": 2560
+    },
+    {
+      "epoch": 1.8848551521818848,
+      "grad_norm": 0.42745399475097656,
+      "learning_rate": 0.0002,
+      "loss": 1.0087,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8921892189218923,
+      "grad_norm": 0.4055580198764801,
+      "learning_rate": 0.0002,
+      "loss": 1.0102,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8995232856618995,
+      "grad_norm": 0.44174644351005554,
+      "learning_rate": 0.0002,
+      "loss": 1.0177,
+      "step": 2590
+    },
+    {
+      "epoch": 1.9068573524019068,
+      "grad_norm": 1.0228385925292969,
+      "learning_rate": 0.0002,
+      "loss": 0.9886,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9141914191419143,
+      "grad_norm": 0.3496396243572235,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9215254858819215,
+      "grad_norm": 0.4191173017024994,
+      "learning_rate": 0.0002,
+      "loss": 1.0955,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9288595526219288,
+      "grad_norm": 0.6778554916381836,
+      "learning_rate": 0.0002,
+      "loss": 1.0943,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9361936193619362,
+      "grad_norm": 0.41992834210395813,
+      "learning_rate": 0.0002,
+      "loss": 1.0594,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9435276861019435,
+      "grad_norm": 0.8760401010513306,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9508617528419507,
+      "grad_norm": 0.44049209356307983,
+      "learning_rate": 0.0002,
+      "loss": 1.0379,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9581958195819582,
+      "grad_norm": 0.5651928782463074,
+      "learning_rate": 0.0002,
+      "loss": 1.1008,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9655298863219657,
+      "grad_norm": 0.5292727947235107,
+      "learning_rate": 0.0002,
+      "loss": 1.1317,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9728639530619727,
+      "grad_norm": 0.6012240648269653,
+      "learning_rate": 0.0002,
+      "loss": 1.1328,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9801980198019802,
+      "grad_norm": 0.3945149779319763,
+      "learning_rate": 0.0002,
+      "loss": 1.0683,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9875320865419877,
+      "grad_norm": 0.5732627511024475,
+      "learning_rate": 0.0002,
+      "loss": 1.0155,
+      "step": 2710
+    },
+    {
+      "epoch": 1.994866153281995,
+      "grad_norm": 0.3963361084461212,
+      "learning_rate": 0.0002,
+      "loss": 0.9857,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1534006595611572,
+      "eval_runtime": 32.7541,
+      "eval_samples_per_second": 13.159,
+      "eval_steps_per_second": 1.649,
+      "step": 2727
+    },
+    {
+      "epoch": 2.002200220022002,
+      "grad_norm": 0.48628315329551697,
+      "learning_rate": 0.0002,
+      "loss": 0.9624,
+      "step": 2730
+    },
+    {
+      "epoch": 2.0095342867620096,
+      "grad_norm": 0.413875013589859,
+      "learning_rate": 0.0002,
+      "loss": 0.9603,
+      "step": 2740
+    },
+    {
+      "epoch": 2.0168683535020167,
+      "grad_norm": 0.4988735616207123,
+      "learning_rate": 0.0002,
+      "loss": 0.965,
+      "step": 2750
+    },
+    {
+      "epoch": 2.024202420242024,
+      "grad_norm": 0.5634812712669373,
+      "learning_rate": 0.0002,
+      "loss": 0.9677,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0315364869820316,
+      "grad_norm": 0.48302653431892395,
+      "learning_rate": 0.0002,
+      "loss": 0.9547,
+      "step": 2770
+    },
+    {
+      "epoch": 2.038870553722039,
+      "grad_norm": 0.49914175271987915,
+      "learning_rate": 0.0002,
+      "loss": 0.9346,
+      "step": 2780
+    },
+    {
+      "epoch": 2.046204620462046,
+      "grad_norm": 1.14039945602417,
+      "learning_rate": 0.0002,
+      "loss": 0.904,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0535386872020536,
+      "grad_norm": 0.6359720826148987,
+      "learning_rate": 0.0002,
+      "loss": 0.9588,
+      "step": 2800
+    },
+    {
+      "epoch": 2.060872753942061,
+      "grad_norm": 0.4589158296585083,
+      "learning_rate": 0.0002,
+      "loss": 0.9031,
+      "step": 2810
+    },
+    {
+      "epoch": 2.068206820682068,
+      "grad_norm": 0.46255481243133545,
+      "learning_rate": 0.0002,
+      "loss": 0.9438,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0755408874220755,
+      "grad_norm": 0.6232137680053711,
+      "learning_rate": 0.0002,
+      "loss": 0.9464,
+      "step": 2830
+    },
+    {
+      "epoch": 2.082874954162083,
+      "grad_norm": 0.41042178869247437,
+      "learning_rate": 0.0002,
+      "loss": 0.8978,
+      "step": 2840
+    },
+    {
+      "epoch": 2.09020902090209,
+      "grad_norm": 0.5334428548812866,
+      "learning_rate": 0.0002,
+      "loss": 0.8516,
+      "step": 2850
+    },
+    {
+      "epoch": 2.0975430876420975,
+      "grad_norm": 0.8270058631896973,
+      "learning_rate": 0.0002,
+      "loss": 0.9313,
+      "step": 2860
+    },
+    {
+      "epoch": 2.104877154382105,
+      "grad_norm": 0.6624533534049988,
+      "learning_rate": 0.0002,
+      "loss": 1.0064,
+      "step": 2870
+    },
+    {
+      "epoch": 2.112211221122112,
+      "grad_norm": 0.5448863506317139,
+      "learning_rate": 0.0002,
+      "loss": 0.9196,
+      "step": 2880
+    },
+    {
+      "epoch": 2.1195452878621195,
+      "grad_norm": 0.621482789516449,
+      "learning_rate": 0.0002,
+      "loss": 0.887,
+      "step": 2890
+    },
+    {
+      "epoch": 2.126879354602127,
+      "grad_norm": 0.4556255340576172,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1342134213421344,
+      "grad_norm": 0.4620579183101654,
+      "learning_rate": 0.0002,
+      "loss": 0.9323,
+      "step": 2910
+    },
+    {
+      "epoch": 2.1415474880821415,
+      "grad_norm": 0.9602415561676025,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 2920
+    },
+    {
+      "epoch": 2.148881554822149,
+      "grad_norm": 0.587943971157074,
+      "learning_rate": 0.0002,
+      "loss": 0.8826,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1562156215621564,
+      "grad_norm": 0.5121372938156128,
+      "learning_rate": 0.0002,
+      "loss": 0.971,
+      "step": 2940
+    },
+    {
+      "epoch": 2.1635496883021634,
+      "grad_norm": 0.49424484372138977,
+      "learning_rate": 0.0002,
+      "loss": 0.8751,
+      "step": 2950
+    },
+    {
+      "epoch": 2.170883755042171,
+      "grad_norm": 0.6312560439109802,
+      "learning_rate": 0.0002,
+      "loss": 0.8674,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1782178217821784,
+      "grad_norm": 0.5235576629638672,
+      "learning_rate": 0.0002,
+      "loss": 0.9791,
+      "step": 2970
+    },
+    {
+      "epoch": 2.1855518885221854,
+      "grad_norm": 0.5868439674377441,
+      "learning_rate": 0.0002,
+      "loss": 0.9706,
+      "step": 2980
+    },
+    {
+      "epoch": 2.192885955262193,
+      "grad_norm": 0.42302873730659485,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 2990
+    },
+    {
+      "epoch": 2.2002200220022003,
+      "grad_norm": 0.5097725987434387,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 3000
+    },
+    {
+      "epoch": 2.2075540887422074,
+      "grad_norm": 0.5091572403907776,
+      "learning_rate": 0.0002,
+      "loss": 0.9239,
+      "step": 3010
+    },
+    {
+      "epoch": 2.214888155482215,
+      "grad_norm": 0.49433162808418274,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 3020
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.5577368140220642,
+      "learning_rate": 0.0002,
+      "loss": 0.9734,
+      "step": 3030
+    },
+    {
+      "epoch": 2.2295562889622293,
+      "grad_norm": 0.6177583932876587,
+      "learning_rate": 0.0002,
+      "loss": 0.9033,
+      "step": 3040
+    },
+    {
+      "epoch": 2.236890355702237,
+      "grad_norm": 0.5256719589233398,
+      "learning_rate": 0.0002,
+      "loss": 0.9882,
+      "step": 3050
+    },
+    {
+      "epoch": 2.2442244224422443,
+      "grad_norm": 0.5001118183135986,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2515584891822513,
+      "grad_norm": 0.5721249580383301,
+      "learning_rate": 0.0002,
+      "loss": 0.8718,
+      "step": 3070
+    },
+    {
+      "epoch": 2.258892555922259,
+      "grad_norm": 0.5325384140014648,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 3080
+    },
+    {
+      "epoch": 2.2662266226622663,
+      "grad_norm": 0.5719189047813416,
+      "learning_rate": 0.0002,
+      "loss": 0.9843,
+      "step": 3090
+    },
+    {
+      "epoch": 2.2735606894022737,
+      "grad_norm": 0.6337835788726807,
+      "learning_rate": 0.0002,
+      "loss": 0.8633,
+      "step": 3100
+    },
+    {
+      "epoch": 2.2808947561422808,
+      "grad_norm": 0.5381836891174316,
+      "learning_rate": 0.0002,
+      "loss": 0.9962,
+      "step": 3110
+    },
+    {
+      "epoch": 2.2882288228822882,
+      "grad_norm": 0.5408531427383423,
+      "learning_rate": 0.0002,
+      "loss": 0.8265,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2955628896222957,
+      "grad_norm": 0.43705281615257263,
+      "learning_rate": 0.0002,
+      "loss": 1.0325,
+      "step": 3130
+    },
+    {
+      "epoch": 2.3028969563623027,
+      "grad_norm": 0.6454030275344849,
+      "learning_rate": 0.0002,
+      "loss": 0.9388,
+      "step": 3140
+    },
+    {
+      "epoch": 2.31023102310231,
+      "grad_norm": 0.686030387878418,
+      "learning_rate": 0.0002,
+      "loss": 0.954,
+      "step": 3150
+    },
+    {
+      "epoch": 2.3175650898423177,
+      "grad_norm": 0.5123633146286011,
+      "learning_rate": 0.0002,
+      "loss": 0.9403,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3248991565823247,
+      "grad_norm": 0.842506468296051,
+      "learning_rate": 0.0002,
+      "loss": 0.8834,
+      "step": 3170
+    },
+    {
+      "epoch": 2.332233223322332,
+      "grad_norm": 0.5193818807601929,
+      "learning_rate": 0.0002,
+      "loss": 1.0497,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3395672900623397,
+      "grad_norm": 0.5634409189224243,
+      "learning_rate": 0.0002,
+      "loss": 0.9473,
+      "step": 3190
+    },
+    {
+      "epoch": 2.3469013568023467,
+      "grad_norm": 0.6475534439086914,
+      "learning_rate": 0.0002,
+      "loss": 0.8499,
+      "step": 3200
+    },
+    {
+      "epoch": 2.354235423542354,
+      "grad_norm": 1.1503914594650269,
+      "learning_rate": 0.0002,
+      "loss": 0.874,
+      "step": 3210
+    },
+    {
+      "epoch": 2.3615694902823616,
+      "grad_norm": 0.7234905362129211,
+      "learning_rate": 0.0002,
+      "loss": 0.9762,
+      "step": 3220
+    },
+    {
+      "epoch": 2.368903557022369,
+      "grad_norm": 0.664903461933136,
+      "learning_rate": 0.0002,
+      "loss": 0.9007,
+      "step": 3230
+    },
+    {
+      "epoch": 2.376237623762376,
+      "grad_norm": 0.5453006625175476,
+      "learning_rate": 0.0002,
+      "loss": 0.9987,
+      "step": 3240
+    },
+    {
+      "epoch": 2.3835716905023836,
+      "grad_norm": 0.6256654262542725,
+      "learning_rate": 0.0002,
+      "loss": 0.9742,
+      "step": 3250
+    },
+    {
+      "epoch": 2.390905757242391,
+      "grad_norm": 0.5166565179824829,
+      "learning_rate": 0.0002,
+      "loss": 0.9922,
+      "step": 3260
+    },
+    {
+      "epoch": 2.398239823982398,
+      "grad_norm": 0.5699098110198975,
+      "learning_rate": 0.0002,
+      "loss": 0.927,
+      "step": 3270
+    },
+    {
+      "epoch": 2.4055738907224056,
+      "grad_norm": 0.4472540020942688,
+      "learning_rate": 0.0002,
+      "loss": 0.8878,
+      "step": 3280
+    },
+    {
+      "epoch": 2.412907957462413,
+      "grad_norm": 0.6790403127670288,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 3290
+    },
+    {
+      "epoch": 2.42024202420242,
+      "grad_norm": 0.5182185173034668,
+      "learning_rate": 0.0002,
+      "loss": 0.972,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4275760909424275,
+      "grad_norm": 0.564647912979126,
+      "learning_rate": 0.0002,
+      "loss": 0.9775,
+      "step": 3310
+    },
+    {
+      "epoch": 2.434910157682435,
+      "grad_norm": 0.5625313520431519,
+      "learning_rate": 0.0002,
+      "loss": 1.072,
+      "step": 3320
+    },
+    {
+      "epoch": 2.442244224422442,
+      "grad_norm": 0.7496559619903564,
+      "learning_rate": 0.0002,
+      "loss": 0.8798,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4495782911624495,
+      "grad_norm": 0.4779128134250641,
+      "learning_rate": 0.0002,
+      "loss": 0.868,
+      "step": 3340
+    },
+    {
+      "epoch": 2.456912357902457,
+      "grad_norm": 0.578093409538269,
+      "learning_rate": 0.0002,
+      "loss": 1.0316,
+      "step": 3350
+    },
+    {
+      "epoch": 2.4642464246424645,
+      "grad_norm": 0.5456080436706543,
+      "learning_rate": 0.0002,
+      "loss": 0.9282,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4715804913824715,
+      "grad_norm": 0.4769273102283478,
+      "learning_rate": 0.0002,
+      "loss": 0.8409,
+      "step": 3370
+    },
+    {
+      "epoch": 2.478914558122479,
+      "grad_norm": 0.5608189702033997,
+      "learning_rate": 0.0002,
+      "loss": 0.9312,
+      "step": 3380
+    },
+    {
+      "epoch": 2.4862486248624864,
+      "grad_norm": 0.5590165853500366,
+      "learning_rate": 0.0002,
+      "loss": 0.9934,
+      "step": 3390
+    },
+    {
+      "epoch": 2.4935826916024935,
+      "grad_norm": 0.801306962966919,
+      "learning_rate": 0.0002,
+      "loss": 1.025,
+      "step": 3400
+    },
+    {
+      "epoch": 2.500916758342501,
+      "grad_norm": 0.6045624613761902,
+      "learning_rate": 0.0002,
+      "loss": 0.9049,
+      "step": 3410
+    },
+    {
+      "epoch": 2.5082508250825084,
+      "grad_norm": 0.5735858082771301,
+      "learning_rate": 0.0002,
+      "loss": 0.944,
+      "step": 3420
+    },
+    {
+      "epoch": 2.5155848918225154,
+      "grad_norm": 0.6827309131622314,
+      "learning_rate": 0.0002,
+      "loss": 0.9846,
+      "step": 3430
+    },
+    {
+      "epoch": 2.522918958562523,
+      "grad_norm": 0.5702602863311768,
+      "learning_rate": 0.0002,
+      "loss": 0.9789,
+      "step": 3440
+    },
+    {
+      "epoch": 2.5302530253025304,
+      "grad_norm": 0.6674721240997314,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5375870920425374,
+      "grad_norm": 0.5635907649993896,
+      "learning_rate": 0.0002,
+      "loss": 0.914,
+      "step": 3460
+    },
+    {
+      "epoch": 2.544921158782545,
+      "grad_norm": 0.42737770080566406,
+      "learning_rate": 0.0002,
+      "loss": 0.8398,
+      "step": 3470
+    },
+    {
+      "epoch": 2.5522552255225524,
+      "grad_norm": 0.6720691919326782,
+      "learning_rate": 0.0002,
+      "loss": 0.9474,
+      "step": 3480
+    },
+    {
+      "epoch": 2.55958929226256,
+      "grad_norm": 0.8917084336280823,
+      "learning_rate": 0.0002,
+      "loss": 0.8637,
+      "step": 3490
+    },
+    {
+      "epoch": 2.566923359002567,
+      "grad_norm": 0.5134549140930176,
+      "learning_rate": 0.0002,
+      "loss": 0.9257,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5742574257425743,
+      "grad_norm": 0.4951367974281311,
+      "learning_rate": 0.0002,
+      "loss": 0.9362,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5815914924825814,
+      "grad_norm": 0.9438204765319824,
+      "learning_rate": 0.0002,
+      "loss": 0.9184,
+      "step": 3520
+    },
+    {
+      "epoch": 2.588925559222589,
+      "grad_norm": 0.6024714708328247,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5962596259625963,
+      "grad_norm": 0.5248535871505737,
+      "learning_rate": 0.0002,
+      "loss": 0.9298,
+      "step": 3540
+    },
+    {
+      "epoch": 2.6035936927026038,
+      "grad_norm": 0.8677568435668945,
+      "learning_rate": 0.0002,
+      "loss": 0.941,
+      "step": 3550
+    },
+    {
+      "epoch": 2.610927759442611,
+      "grad_norm": 0.82008296251297,
+      "learning_rate": 0.0002,
+      "loss": 0.9253,
+      "step": 3560
+    },
+    {
+      "epoch": 2.6182618261826183,
+      "grad_norm": 0.4724634885787964,
+      "learning_rate": 0.0002,
+      "loss": 0.8429,
+      "step": 3570
+    },
+    {
+      "epoch": 2.6255958929226257,
+      "grad_norm": 0.5434244275093079,
+      "learning_rate": 0.0002,
+      "loss": 0.9058,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6329299596626328,
+      "grad_norm": 0.4948740005493164,
+      "learning_rate": 0.0002,
+      "loss": 0.9379,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6402640264026402,
+      "grad_norm": 0.42109328508377075,
+      "learning_rate": 0.0002,
+      "loss": 0.8718,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6475980931426477,
+      "grad_norm": 0.7979786396026611,
+      "learning_rate": 0.0002,
+      "loss": 0.9809,
+      "step": 3610
+    },
+    {
+      "epoch": 2.654932159882655,
+      "grad_norm": 0.6345919370651245,
+      "learning_rate": 0.0002,
+      "loss": 0.9229,
+      "step": 3620
+    },
+    {
+      "epoch": 2.662266226622662,
+      "grad_norm": 0.4971671402454376,
+      "learning_rate": 0.0002,
+      "loss": 0.8506,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6696002933626697,
+      "grad_norm": 0.6467748284339905,
+      "learning_rate": 0.0002,
+      "loss": 0.8054,
+      "step": 3640
+    },
+    {
+      "epoch": 2.6769343601026767,
+      "grad_norm": 0.4240160286426544,
+      "learning_rate": 0.0002,
+      "loss": 0.9277,
+      "step": 3650
+    },
+    {
+      "epoch": 2.684268426842684,
+      "grad_norm": 0.5179754495620728,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 3660
+    },
+    {
+      "epoch": 2.6916024935826917,
+      "grad_norm": 0.754012405872345,
+      "learning_rate": 0.0002,
+      "loss": 0.9221,
+      "step": 3670
+    },
+    {
+      "epoch": 2.698936560322699,
+      "grad_norm": 0.5141299962997437,
+      "learning_rate": 0.0002,
+      "loss": 0.9194,
+      "step": 3680
+    },
+    {
+      "epoch": 2.706270627062706,
+      "grad_norm": 0.5737819075584412,
+      "learning_rate": 0.0002,
+      "loss": 0.9495,
+      "step": 3690
+    },
+    {
+      "epoch": 2.7136046938027136,
+      "grad_norm": 0.5887577533721924,
+      "learning_rate": 0.0002,
+      "loss": 1.0162,
+      "step": 3700
+    },
+    {
+      "epoch": 2.720938760542721,
+      "grad_norm": 0.6740471720695496,
+      "learning_rate": 0.0002,
+      "loss": 0.9169,
+      "step": 3710
+    },
+    {
+      "epoch": 2.728272827282728,
+      "grad_norm": 0.5879453420639038,
+      "learning_rate": 0.0002,
+      "loss": 0.9297,
+      "step": 3720
+    },
+    {
+      "epoch": 2.7356068940227356,
+      "grad_norm": 0.4858354926109314,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 3730
+    },
+    {
+      "epoch": 2.742940960762743,
+      "grad_norm": 0.5489001870155334,
+      "learning_rate": 0.0002,
+      "loss": 0.9308,
+      "step": 3740
+    },
+    {
+      "epoch": 2.7502750275027505,
+      "grad_norm": 0.8187092542648315,
+      "learning_rate": 0.0002,
+      "loss": 0.894,
+      "step": 3750
+    },
+    {
+      "epoch": 2.7576090942427576,
+      "grad_norm": 0.5666626691818237,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 3760
+    },
+    {
+      "epoch": 2.764943160982765,
+      "grad_norm": 0.5377066135406494,
+      "learning_rate": 0.0002,
+      "loss": 1.0059,
+      "step": 3770
+    },
+    {
+      "epoch": 2.772277227722772,
+      "grad_norm": 0.566330075263977,
+      "learning_rate": 0.0002,
+      "loss": 0.9132,
+      "step": 3780
+    },
+    {
+      "epoch": 2.7796112944627795,
+      "grad_norm": 0.5522832870483398,
+      "learning_rate": 0.0002,
+      "loss": 0.9415,
+      "step": 3790
+    },
+    {
+      "epoch": 2.786945361202787,
+      "grad_norm": 0.5668695569038391,
+      "learning_rate": 0.0002,
+      "loss": 0.8816,
+      "step": 3800
+    },
+    {
+      "epoch": 2.7942794279427945,
+      "grad_norm": 0.7566602826118469,
+      "learning_rate": 0.0002,
+      "loss": 0.8885,
+      "step": 3810
+    },
+    {
+      "epoch": 2.8016134946828015,
+      "grad_norm": 0.5603684782981873,
+      "learning_rate": 0.0002,
+      "loss": 0.8598,
+      "step": 3820
+    },
+    {
+      "epoch": 2.808947561422809,
+      "grad_norm": 0.49122217297554016,
+      "learning_rate": 0.0002,
+      "loss": 0.9602,
+      "step": 3830
+    },
+    {
+      "epoch": 2.816281628162816,
+      "grad_norm": 0.6798251867294312,
+      "learning_rate": 0.0002,
+      "loss": 0.9738,
+      "step": 3840
+    },
+    {
+      "epoch": 2.8236156949028235,
+      "grad_norm": 0.6097991466522217,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 3850
+    },
+    {
+      "epoch": 2.830949761642831,
+      "grad_norm": 0.6675726175308228,
+      "learning_rate": 0.0002,
+      "loss": 0.8672,
+      "step": 3860
+    },
+    {
+      "epoch": 2.8382838283828384,
+      "grad_norm": 0.9223952889442444,
+      "learning_rate": 0.0002,
+      "loss": 0.9324,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8456178951228455,
+      "grad_norm": 0.6020799875259399,
+      "learning_rate": 0.0002,
+      "loss": 0.8767,
+      "step": 3880
+    },
+    {
+      "epoch": 2.852951961862853,
+      "grad_norm": 0.5206381678581238,
+      "learning_rate": 0.0002,
+      "loss": 0.9148,
+      "step": 3890
+    },
+    {
+      "epoch": 2.8602860286028604,
+      "grad_norm": 0.6268777251243591,
+      "learning_rate": 0.0002,
+      "loss": 0.9479,
+      "step": 3900
+    },
+    {
+      "epoch": 2.8676200953428674,
+      "grad_norm": 1.1583497524261475,
+      "learning_rate": 0.0002,
+      "loss": 0.9409,
+      "step": 3910
+    },
+    {
+      "epoch": 2.874954162082875,
+      "grad_norm": 0.7263903021812439,
+      "learning_rate": 0.0002,
+      "loss": 0.895,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8822882288228824,
+      "grad_norm": 0.5369910001754761,
+      "learning_rate": 0.0002,
+      "loss": 0.8786,
+      "step": 3930
+    },
+    {
+      "epoch": 2.88962229556289,
+      "grad_norm": 0.7298350930213928,
+      "learning_rate": 0.0002,
+      "loss": 1.0015,
+      "step": 3940
+    },
+    {
+      "epoch": 2.896956362302897,
+      "grad_norm": 0.577012836933136,
+      "learning_rate": 0.0002,
+      "loss": 0.979,
+      "step": 3950
+    },
+    {
+      "epoch": 2.9042904290429044,
+      "grad_norm": 0.5859594345092773,
+      "learning_rate": 0.0002,
+      "loss": 0.9716,
+      "step": 3960
+    },
+    {
+      "epoch": 2.9116244957829114,
+      "grad_norm": 0.47176122665405273,
+      "learning_rate": 0.0002,
+      "loss": 0.8772,
+      "step": 3970
+    },
+    {
+      "epoch": 2.918958562522919,
+      "grad_norm": 0.9699620604515076,
+      "learning_rate": 0.0002,
+      "loss": 0.8997,
+      "step": 3980
+    },
+    {
+      "epoch": 2.9262926292629263,
+      "grad_norm": 0.7908747792243958,
+      "learning_rate": 0.0002,
+      "loss": 0.9057,
+      "step": 3990
+    },
+    {
+      "epoch": 2.933626696002934,
+      "grad_norm": 0.5777379274368286,
+      "learning_rate": 0.0002,
+      "loss": 0.9462,
+      "step": 4000
+    },
+    {
+      "epoch": 2.940960762742941,
+      "grad_norm": 0.599288284778595,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9482948294829483,
+      "grad_norm": 0.5232274532318115,
+      "learning_rate": 0.0002,
+      "loss": 0.9812,
+      "step": 4020
+    },
+    {
+      "epoch": 2.9556288962229558,
+      "grad_norm": 0.6395137310028076,
+      "learning_rate": 0.0002,
+      "loss": 0.96,
+      "step": 4030
+    },
+    {
+      "epoch": 2.962962962962963,
+      "grad_norm": 0.589260458946228,
+      "learning_rate": 0.0002,
+      "loss": 0.9813,
+      "step": 4040
+    },
+    {
+      "epoch": 2.9702970297029703,
+      "grad_norm": 0.5699581503868103,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 4050
+    },
+    {
+      "epoch": 2.9776310964429777,
+      "grad_norm": 0.528468132019043,
+      "learning_rate": 0.0002,
+      "loss": 0.9585,
+      "step": 4060
+    },
+    {
+      "epoch": 2.984965163182985,
+      "grad_norm": 0.4804670512676239,
+      "learning_rate": 0.0002,
+      "loss": 0.9164,
+      "step": 4070
+    },
+    {
+      "epoch": 2.9922992299229922,
+      "grad_norm": 1.1918889284133911,
+      "learning_rate": 0.0002,
+      "loss": 0.9771,
+      "step": 4080
+    },
+    {
+      "epoch": 2.9996332966629997,
+      "grad_norm": 0.5479103326797485,
+      "learning_rate": 0.0002,
+      "loss": 0.9178,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9996332966629997,
+      "eval_loss": 1.1642853021621704,
+      "eval_runtime": 32.7511,
+      "eval_samples_per_second": 13.16,
+      "eval_steps_per_second": 1.649,
+      "step": 4090
+    },
+    {
+      "epoch": 3.006967363403007,
+      "grad_norm": 0.7430027723312378,
+      "learning_rate": 0.0002,
+      "loss": 0.7981,
+      "step": 4100
+    },
+    {
+      "epoch": 3.014301430143014,
+      "grad_norm": 0.6293647289276123,
+      "learning_rate": 0.0002,
+      "loss": 0.7871,
+      "step": 4110
+    },
+    {
+      "epoch": 3.0216354968830217,
+      "grad_norm": 0.6191329956054688,
+      "learning_rate": 0.0002,
+      "loss": 0.78,
+      "step": 4120
+    },
+    {
+      "epoch": 3.028969563623029,
+      "grad_norm": 0.7959313988685608,
+      "learning_rate": 0.0002,
+      "loss": 0.7618,
+      "step": 4130
+    },
+    {
+      "epoch": 3.036303630363036,
+      "grad_norm": 0.5956351161003113,
+      "learning_rate": 0.0002,
+      "loss": 0.8039,
+      "step": 4140
+    },
+    {
+      "epoch": 3.0436376971030437,
+      "grad_norm": 0.670383632183075,
+      "learning_rate": 0.0002,
+      "loss": 0.7477,
+      "step": 4150
+    },
+    {
+      "epoch": 3.050971763843051,
+      "grad_norm": 0.6414518356323242,
+      "learning_rate": 0.0002,
+      "loss": 0.7984,
+      "step": 4160
+    },
+    {
+      "epoch": 3.058305830583058,
+      "grad_norm": 0.7928852438926697,
+      "learning_rate": 0.0002,
+      "loss": 0.7369,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0656398973230656,
+      "grad_norm": 0.6211121082305908,
+      "learning_rate": 0.0002,
+      "loss": 0.7914,
+      "step": 4180
+    },
+    {
+      "epoch": 3.072973964063073,
+      "grad_norm": 0.6237057447433472,
+      "learning_rate": 0.0002,
+      "loss": 0.7365,
+      "step": 4190
+    },
+    {
+      "epoch": 3.08030803080308,
+      "grad_norm": 0.6522233486175537,
+      "learning_rate": 0.0002,
+      "loss": 0.702,
+      "step": 4200
+    },
+    {
+      "epoch": 3.0876420975430876,
+      "grad_norm": 0.9396848678588867,
+      "learning_rate": 0.0002,
+      "loss": 0.7646,
+      "step": 4210
+    },
+    {
+      "epoch": 3.094976164283095,
+      "grad_norm": 0.8003010749816895,
+      "learning_rate": 0.0002,
+      "loss": 0.7559,
+      "step": 4220
+    },
+    {
+      "epoch": 3.102310231023102,
+      "grad_norm": 0.6733810305595398,
+      "learning_rate": 0.0002,
+      "loss": 0.711,
+      "step": 4230
+    },
+    {
+      "epoch": 3.1096442977631096,
+      "grad_norm": 0.6365828514099121,
+      "learning_rate": 0.0002,
+      "loss": 0.696,
+      "step": 4240
+    },
+    {
+      "epoch": 3.116978364503117,
+      "grad_norm": 1.0805548429489136,
+      "learning_rate": 0.0002,
+      "loss": 0.8362,
+      "step": 4250
+    },
+    {
+      "epoch": 3.1243124312431245,
+      "grad_norm": 0.7262141108512878,
+      "learning_rate": 0.0002,
+      "loss": 0.7651,
+      "step": 4260
+    },
+    {
+      "epoch": 3.1316464979831315,
+      "grad_norm": 0.5500539541244507,
+      "learning_rate": 0.0002,
+      "loss": 0.7304,
+      "step": 4270
+    },
+    {
+      "epoch": 3.138980564723139,
+      "grad_norm": 0.793912947177887,
+      "learning_rate": 0.0002,
+      "loss": 0.7721,
+      "step": 4280
+    },
+    {
+      "epoch": 3.1463146314631465,
+      "grad_norm": 1.2540518045425415,
+      "learning_rate": 0.0002,
+      "loss": 0.7708,
+      "step": 4290
+    },
+    {
+      "epoch": 3.1536486982031535,
+      "grad_norm": 0.7020077705383301,
+      "learning_rate": 0.0002,
+      "loss": 0.782,
+      "step": 4300
+    },
+    {
+      "epoch": 3.160982764943161,
+      "grad_norm": 0.5111123323440552,
+      "learning_rate": 0.0002,
+      "loss": 0.7253,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1683168316831685,
+      "grad_norm": 0.7172090411186218,
+      "learning_rate": 0.0002,
+      "loss": 0.8159,
+      "step": 4320
+    },
+    {
+      "epoch": 3.1756508984231755,
+      "grad_norm": 0.6343168616294861,
+      "learning_rate": 0.0002,
+      "loss": 0.6962,
+      "step": 4330
+    },
+    {
+      "epoch": 3.182984965163183,
+      "grad_norm": 0.9563672542572021,
+      "learning_rate": 0.0002,
+      "loss": 0.7938,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1903190319031904,
+      "grad_norm": 1.0225574970245361,
+      "learning_rate": 0.0002,
+      "loss": 0.7385,
+      "step": 4350
+    },
+    {
+      "epoch": 3.1976530986431975,
+      "grad_norm": 1.1633386611938477,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 4360
+    },
+    {
+      "epoch": 3.204987165383205,
+      "grad_norm": 0.8915148973464966,
+      "learning_rate": 0.0002,
+      "loss": 0.7259,
+      "step": 4370
+    },
+    {
+      "epoch": 3.2123212321232124,
+      "grad_norm": 0.9156812429428101,
+      "learning_rate": 0.0002,
+      "loss": 0.8061,
+      "step": 4380
+    },
+    {
+      "epoch": 3.21965529886322,
+      "grad_norm": 0.6363258957862854,
+      "learning_rate": 0.0002,
+      "loss": 0.8189,
+      "step": 4390
+    },
+    {
+      "epoch": 3.226989365603227,
+      "grad_norm": 0.579099178314209,
+      "learning_rate": 0.0002,
+      "loss": 0.7996,
+      "step": 4400
+    },
+    {
+      "epoch": 3.2343234323432344,
+      "grad_norm": 0.8778146505355835,
+      "learning_rate": 0.0002,
+      "loss": 0.8592,
+      "step": 4410
+    },
+    {
+      "epoch": 3.241657499083242,
+      "grad_norm": 0.8356770873069763,
+      "learning_rate": 0.0002,
+      "loss": 0.8281,
+      "step": 4420
+    },
+    {
+      "epoch": 3.248991565823249,
+      "grad_norm": 0.702032208442688,
+      "learning_rate": 0.0002,
+      "loss": 0.8484,
+      "step": 4430
+    },
+    {
+      "epoch": 3.2563256325632564,
+      "grad_norm": 0.6386539340019226,
+      "learning_rate": 0.0002,
+      "loss": 0.7227,
+      "step": 4440
+    },
+    {
+      "epoch": 3.263659699303264,
+      "grad_norm": 0.7008408904075623,
+      "learning_rate": 0.0002,
+      "loss": 0.8374,
+      "step": 4450
+    },
+    {
+      "epoch": 3.270993766043271,
+      "grad_norm": 0.9556332230567932,
+      "learning_rate": 0.0002,
+      "loss": 0.7572,
+      "step": 4460
+    },
+    {
+      "epoch": 3.2783278327832783,
+      "grad_norm": 0.5667835474014282,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 4470
+    },
+    {
+      "epoch": 3.285661899523286,
+      "grad_norm": 0.8239172697067261,
+      "learning_rate": 0.0002,
+      "loss": 0.8152,
+      "step": 4480
+    },
+    {
+      "epoch": 3.292995966263293,
+      "grad_norm": 0.7045050859451294,
+      "learning_rate": 0.0002,
+      "loss": 0.756,
+      "step": 4490
+    },
+    {
+      "epoch": 3.3003300330033003,
+      "grad_norm": 0.7131434082984924,
+      "learning_rate": 0.0002,
+      "loss": 0.7655,
+      "step": 4500
+    },
+    {
+      "epoch": 3.3076640997433078,
+      "grad_norm": 0.6924910545349121,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 4510
+    },
+    {
+      "epoch": 3.3149981664833152,
+      "grad_norm": 0.8945356607437134,
+      "learning_rate": 0.0002,
+      "loss": 0.736,
+      "step": 4520
+    },
+    {
+      "epoch": 3.3223322332233223,
+      "grad_norm": 0.6546903252601624,
+      "learning_rate": 0.0002,
+      "loss": 0.7575,
+      "step": 4530
+    },
+    {
+      "epoch": 3.3296662999633297,
+      "grad_norm": 0.8206679224967957,
+      "learning_rate": 0.0002,
+      "loss": 0.7893,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3370003667033368,
+      "grad_norm": 0.6482203602790833,
+      "learning_rate": 0.0002,
+      "loss": 0.7502,
+      "step": 4550
+    },
+    {
+      "epoch": 3.3443344334433442,
+      "grad_norm": 0.7558760046958923,
+      "learning_rate": 0.0002,
+      "loss": 0.8172,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3516685001833517,
+      "grad_norm": 0.7794756889343262,
+      "learning_rate": 0.0002,
+      "loss": 0.744,
+      "step": 4570
+    },
+    {
+      "epoch": 3.359002566923359,
+      "grad_norm": 0.7382805943489075,
+      "learning_rate": 0.0002,
+      "loss": 0.7385,
+      "step": 4580
+    },
+    {
+      "epoch": 3.366336633663366,
+      "grad_norm": 0.5912511944770813,
+      "learning_rate": 0.0002,
+      "loss": 0.8511,
+      "step": 4590
+    },
+    {
+      "epoch": 3.3736707004033737,
+      "grad_norm": 0.7444885969161987,
+      "learning_rate": 0.0002,
+      "loss": 0.8272,
+      "step": 4600
+    },
+    {
+      "epoch": 3.381004767143381,
+      "grad_norm": 0.7354922890663147,
+      "learning_rate": 0.0002,
+      "loss": 0.7927,
+      "step": 4610
+    },
+    {
+      "epoch": 3.388338833883388,
+      "grad_norm": 0.7685934901237488,
+      "learning_rate": 0.0002,
+      "loss": 0.7183,
+      "step": 4620
+    },
+    {
+      "epoch": 3.3956729006233957,
+      "grad_norm": 0.61041259765625,
+      "learning_rate": 0.0002,
+      "loss": 0.7436,
+      "step": 4630
+    },
+    {
+      "epoch": 3.403006967363403,
+      "grad_norm": 0.6820451021194458,
+      "learning_rate": 0.0002,
+      "loss": 0.7661,
+      "step": 4640
+    },
+    {
+      "epoch": 3.41034103410341,
+      "grad_norm": 0.5819534063339233,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 4650
+    },
+    {
+      "epoch": 3.4176751008434176,
+      "grad_norm": 0.705410897731781,
+      "learning_rate": 0.0002,
+      "loss": 0.7314,
+      "step": 4660
+    },
+    {
+      "epoch": 3.425009167583425,
+      "grad_norm": 0.8052892088890076,
+      "learning_rate": 0.0002,
+      "loss": 0.7901,
+      "step": 4670
+    },
+    {
+      "epoch": 3.432343234323432,
+      "grad_norm": 0.7746483087539673,
+      "learning_rate": 0.0002,
+      "loss": 0.7298,
+      "step": 4680
+    },
+    {
+      "epoch": 3.4396773010634396,
+      "grad_norm": 0.7713689804077148,
+      "learning_rate": 0.0002,
+      "loss": 0.7976,
+      "step": 4690
+    },
+    {
+      "epoch": 3.447011367803447,
+      "grad_norm": 0.810371994972229,
+      "learning_rate": 0.0002,
+      "loss": 0.7427,
+      "step": 4700
+    },
+    {
+      "epoch": 3.4543454345434546,
+      "grad_norm": 0.7702969312667847,
+      "learning_rate": 0.0002,
+      "loss": 0.7594,
+      "step": 4710
+    },
+    {
+      "epoch": 3.4616795012834616,
+      "grad_norm": 0.7069268822669983,
+      "learning_rate": 0.0002,
+      "loss": 0.7957,
+      "step": 4720
+    },
+    {
+      "epoch": 3.469013568023469,
+      "grad_norm": 0.7640359401702881,
+      "learning_rate": 0.0002,
+      "loss": 0.8199,
+      "step": 4730
+    },
+    {
+      "epoch": 3.4763476347634765,
+      "grad_norm": 0.8661707639694214,
+      "learning_rate": 0.0002,
+      "loss": 0.6875,
+      "step": 4740
+    },
+    {
+      "epoch": 3.4836817015034836,
+      "grad_norm": 0.9970282912254333,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 4750
+    },
+    {
+      "epoch": 3.491015768243491,
+      "grad_norm": 0.5824355483055115,
+      "learning_rate": 0.0002,
+      "loss": 0.8462,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4983498349834985,
+      "grad_norm": 1.3072649240493774,
+      "learning_rate": 0.0002,
+      "loss": 0.851,
+      "step": 4770
+    },
+    {
+      "epoch": 3.5056839017235055,
+      "grad_norm": 0.873978316783905,
+      "learning_rate": 0.0002,
+      "loss": 0.9101,
+      "step": 4780
+    },
+    {
+      "epoch": 3.513017968463513,
+      "grad_norm": 0.5526657104492188,
+      "learning_rate": 0.0002,
+      "loss": 0.7403,
+      "step": 4790
+    },
+    {
+      "epoch": 3.5203520352035205,
+      "grad_norm": 0.790894627571106,
+      "learning_rate": 0.0002,
+      "loss": 0.7921,
+      "step": 4800
+    },
+    {
+      "epoch": 3.5276861019435275,
+      "grad_norm": 0.8119630217552185,
+      "learning_rate": 0.0002,
+      "loss": 0.831,
+      "step": 4810
+    },
+    {
+      "epoch": 3.535020168683535,
+      "grad_norm": 0.633212149143219,
+      "learning_rate": 0.0002,
+      "loss": 0.7351,
+      "step": 4820
+    },
+    {
+      "epoch": 3.5423542354235424,
+      "grad_norm": 0.703029990196228,
+      "learning_rate": 0.0002,
+      "loss": 0.8505,
+      "step": 4830
+    },
+    {
+      "epoch": 3.54968830216355,
+      "grad_norm": 0.7603771686553955,
+      "learning_rate": 0.0002,
+      "loss": 0.7204,
+      "step": 4840
+    },
+    {
+      "epoch": 3.557022368903557,
+      "grad_norm": 0.6260480880737305,
+      "learning_rate": 0.0002,
+      "loss": 0.8868,
+      "step": 4850
+    },
+    {
+      "epoch": 3.5643564356435644,
+      "grad_norm": 0.8203664422035217,
+      "learning_rate": 0.0002,
+      "loss": 0.8137,
+      "step": 4860
+    },
+    {
+      "epoch": 3.5716905023835714,
+      "grad_norm": 0.7793813347816467,
+      "learning_rate": 0.0002,
+      "loss": 0.8821,
+      "step": 4870
+    },
+    {
+      "epoch": 3.579024569123579,
+      "grad_norm": 0.7667397260665894,
+      "learning_rate": 0.0002,
+      "loss": 0.8164,
+      "step": 4880
+    },
+    {
+      "epoch": 3.5863586358635864,
+      "grad_norm": 0.8198829889297485,
+      "learning_rate": 0.0002,
+      "loss": 0.7597,
+      "step": 4890
+    },
+    {
+      "epoch": 3.593692702603594,
+      "grad_norm": 0.7689233422279358,
+      "learning_rate": 0.0002,
+      "loss": 0.7027,
+      "step": 4900
+    },
+    {
+      "epoch": 3.601026769343601,
+      "grad_norm": 0.7870983481407166,
+      "learning_rate": 0.0002,
+      "loss": 0.804,
+      "step": 4910
+    },
+    {
+      "epoch": 3.6083608360836084,
+      "grad_norm": 0.8133853077888489,
+      "learning_rate": 0.0002,
+      "loss": 0.8269,
+      "step": 4920
+    },
+    {
+      "epoch": 3.615694902823616,
+      "grad_norm": 1.308401346206665,
+      "learning_rate": 0.0002,
+      "loss": 0.8515,
+      "step": 4930
+    },
+    {
+      "epoch": 3.623028969563623,
+      "grad_norm": 0.7131121754646301,
+      "learning_rate": 0.0002,
+      "loss": 0.8494,
+      "step": 4940
+    },
+    {
+      "epoch": 3.6303630363036303,
+      "grad_norm": 0.6825910210609436,
+      "learning_rate": 0.0002,
+      "loss": 0.7235,
+      "step": 4950
+    },
+    {
+      "epoch": 3.637697103043638,
+      "grad_norm": 0.7254678606987,
+      "learning_rate": 0.0002,
+      "loss": 0.7824,
+      "step": 4960
+    },
+    {
+      "epoch": 3.6450311697836453,
+      "grad_norm": 0.8045085072517395,
+      "learning_rate": 0.0002,
+      "loss": 0.7983,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6523652365236523,
+      "grad_norm": 0.6991777420043945,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6596993032636598,
+      "grad_norm": 0.7804713249206543,
+      "learning_rate": 0.0002,
+      "loss": 0.7806,
+      "step": 4990
+    },
+    {
+      "epoch": 3.667033370003667,
+      "grad_norm": 0.8525708317756653,
+      "learning_rate": 0.0002,
+      "loss": 0.8402,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6743674367436743,
+      "grad_norm": 0.7959994673728943,
+      "learning_rate": 0.0002,
+      "loss": 0.8496,
+      "step": 5010
+    },
+    {
+      "epoch": 3.6817015034836817,
+      "grad_norm": 0.8103628158569336,
+      "learning_rate": 0.0002,
+      "loss": 0.8022,
+      "step": 5020
+    },
+    {
+      "epoch": 3.689035570223689,
+      "grad_norm": 0.7517836093902588,
+      "learning_rate": 0.0002,
+      "loss": 0.7376,
+      "step": 5030
+    },
+    {
+      "epoch": 3.6963696369636962,
+      "grad_norm": 0.6878514289855957,
+      "learning_rate": 0.0002,
+      "loss": 0.8375,
+      "step": 5040
+    },
+    {
+      "epoch": 3.7037037037037037,
+      "grad_norm": 1.2371820211410522,
+      "learning_rate": 0.0002,
+      "loss": 0.7998,
+      "step": 5050
+    },
+    {
+      "epoch": 3.711037770443711,
+      "grad_norm": 0.6567103862762451,
+      "learning_rate": 0.0002,
+      "loss": 0.6941,
+      "step": 5060
+    },
+    {
+      "epoch": 3.718371837183718,
+      "grad_norm": 1.1254922151565552,
+      "learning_rate": 0.0002,
+      "loss": 0.8465,
+      "step": 5070
+    },
+    {
+      "epoch": 3.7257059039237257,
+      "grad_norm": 0.6796132326126099,
+      "learning_rate": 0.0002,
+      "loss": 0.8365,
+      "step": 5080
+    },
+    {
+      "epoch": 3.733039970663733,
+      "grad_norm": 0.7285300493240356,
+      "learning_rate": 0.0002,
+      "loss": 0.7818,
+      "step": 5090
+    },
+    {
+      "epoch": 3.7403740374037406,
+      "grad_norm": 0.8931500911712646,
+      "learning_rate": 0.0002,
+      "loss": 0.8581,
+      "step": 5100
+    },
+    {
+      "epoch": 3.7477081041437477,
+      "grad_norm": 0.6256856918334961,
+      "learning_rate": 0.0002,
+      "loss": 0.8181,
+      "step": 5110
+    },
+    {
+      "epoch": 3.755042170883755,
+      "grad_norm": 0.79310142993927,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 5120
+    },
+    {
+      "epoch": 3.762376237623762,
+      "grad_norm": 0.6594041585922241,
+      "learning_rate": 0.0002,
+      "loss": 0.8235,
+      "step": 5130
+    },
+    {
+      "epoch": 3.7697103043637696,
+      "grad_norm": 0.7029327750205994,
+      "learning_rate": 0.0002,
+      "loss": 0.6925,
+      "step": 5140
+    },
+    {
+      "epoch": 3.777044371103777,
+      "grad_norm": 0.5880070328712463,
+      "learning_rate": 0.0002,
+      "loss": 0.7457,
+      "step": 5150
+    },
+    {
+      "epoch": 3.7843784378437846,
+      "grad_norm": 0.7578945159912109,
+      "learning_rate": 0.0002,
+      "loss": 0.8716,
+      "step": 5160
+    },
+    {
+      "epoch": 3.7917125045837916,
+      "grad_norm": 0.8276378512382507,
+      "learning_rate": 0.0002,
+      "loss": 0.8819,
+      "step": 5170
+    },
+    {
+      "epoch": 3.799046571323799,
+      "grad_norm": 0.7627953886985779,
+      "learning_rate": 0.0002,
+      "loss": 0.7559,
+      "step": 5180
+    },
+    {
+      "epoch": 3.806380638063806,
+      "grad_norm": 0.8169086575508118,
+      "learning_rate": 0.0002,
+      "loss": 0.7665,
+      "step": 5190
+    },
+    {
+      "epoch": 3.8137147048038136,
+      "grad_norm": 0.6605030298233032,
+      "learning_rate": 0.0002,
+      "loss": 0.761,
+      "step": 5200
+    },
+    {
+      "epoch": 3.821048771543821,
+      "grad_norm": 0.5837286114692688,
+      "learning_rate": 0.0002,
+      "loss": 0.8804,
+      "step": 5210
+    },
+    {
+      "epoch": 3.8283828382838285,
+      "grad_norm": 1.2422157526016235,
+      "learning_rate": 0.0002,
+      "loss": 0.8369,
+      "step": 5220
+    },
+    {
+      "epoch": 3.8357169050238356,
+      "grad_norm": 0.6589220762252808,
+      "learning_rate": 0.0002,
+      "loss": 0.8431,
+      "step": 5230
+    },
+    {
+      "epoch": 3.843050971763843,
+      "grad_norm": 0.8567556142807007,
+      "learning_rate": 0.0002,
+      "loss": 0.7686,
+      "step": 5240
+    },
+    {
+      "epoch": 3.8503850385038505,
+      "grad_norm": 0.6490627527236938,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8577191052438575,
+      "grad_norm": 0.620232880115509,
+      "learning_rate": 0.0002,
+      "loss": 0.7386,
+      "step": 5260
+    },
+    {
+      "epoch": 3.865053171983865,
+      "grad_norm": 0.7685128450393677,
+      "learning_rate": 0.0002,
+      "loss": 0.9192,
+      "step": 5270
+    },
+    {
+      "epoch": 3.8723872387238725,
+      "grad_norm": 0.8113296627998352,
+      "learning_rate": 0.0002,
+      "loss": 0.872,
+      "step": 5280
+    },
+    {
+      "epoch": 3.87972130546388,
+      "grad_norm": 0.8092675805091858,
+      "learning_rate": 0.0002,
+      "loss": 0.7156,
+      "step": 5290
+    },
+    {
+      "epoch": 3.887055372203887,
+      "grad_norm": 0.583570122718811,
+      "learning_rate": 0.0002,
+      "loss": 0.7325,
+      "step": 5300
+    },
+    {
+      "epoch": 3.8943894389438944,
+      "grad_norm": 1.712363600730896,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 5310
+    },
+    {
+      "epoch": 3.9017235056839015,
+      "grad_norm": 0.6673534512519836,
+      "learning_rate": 0.0002,
+      "loss": 0.7537,
+      "step": 5320
+    },
+    {
+      "epoch": 3.909057572423909,
+      "grad_norm": 1.9770312309265137,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 5330
+    },
+    {
+      "epoch": 3.9163916391639164,
+      "grad_norm": 0.6430999636650085,
+      "learning_rate": 0.0002,
+      "loss": 0.8793,
+      "step": 5340
+    },
+    {
+      "epoch": 3.923725705903924,
+      "grad_norm": 1.0159571170806885,
+      "learning_rate": 0.0002,
+      "loss": 0.839,
+      "step": 5350
+    },
+    {
+      "epoch": 3.931059772643931,
+      "grad_norm": 0.8607584834098816,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 5360
+    },
+    {
+      "epoch": 3.9383938393839384,
+      "grad_norm": 0.6967900991439819,
+      "learning_rate": 0.0002,
+      "loss": 0.7261,
+      "step": 5370
+    },
+    {
+      "epoch": 3.945727906123946,
+      "grad_norm": 0.7683077454566956,
+      "learning_rate": 0.0002,
+      "loss": 0.8456,
+      "step": 5380
+    },
+    {
+      "epoch": 3.953061972863953,
+      "grad_norm": 0.6805762648582458,
+      "learning_rate": 0.0002,
+      "loss": 0.7682,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9603960396039604,
+      "grad_norm": 0.7033619284629822,
+      "learning_rate": 0.0002,
+      "loss": 0.7746,
+      "step": 5400
+    },
+    {
+      "epoch": 3.967730106343968,
+      "grad_norm": 0.966112494468689,
+      "learning_rate": 0.0002,
+      "loss": 0.8393,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9750641730839753,
+      "grad_norm": 0.8467881083488464,
+      "learning_rate": 0.0002,
+      "loss": 0.8316,
+      "step": 5420
+    },
+    {
+      "epoch": 3.9823982398239823,
+      "grad_norm": 0.8005317449569702,
+      "learning_rate": 0.0002,
+      "loss": 0.8084,
+      "step": 5430
+    },
+    {
+      "epoch": 3.98973230656399,
+      "grad_norm": 1.1615241765975952,
+      "learning_rate": 0.0002,
+      "loss": 0.7168,
+      "step": 5440
+    },
+    {
+      "epoch": 3.997066373303997,
+      "grad_norm": 0.6121614575386047,
+      "learning_rate": 0.0002,
+      "loss": 0.8263,
+      "step": 5450
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.1834222078323364,
+      "eval_runtime": 32.7569,
+      "eval_samples_per_second": 13.158,
+      "eval_steps_per_second": 1.649,
+      "step": 5454
+    },
+    {
+      "epoch": 4.004400440044004,
+      "grad_norm": 0.6055727005004883,
+      "learning_rate": 0.0002,
+      "loss": 0.7267,
+      "step": 5460
+    },
+    {
+      "epoch": 4.011734506784012,
+      "grad_norm": 0.8232647180557251,
+      "learning_rate": 0.0002,
+      "loss": 0.5766,
+      "step": 5470
+    },
+    {
+      "epoch": 4.019068573524019,
+      "grad_norm": 0.7739192247390747,
+      "learning_rate": 0.0002,
+      "loss": 0.6489,
+      "step": 5480
+    },
+    {
+      "epoch": 4.026402640264027,
+      "grad_norm": 0.6264950633049011,
+      "learning_rate": 0.0002,
+      "loss": 0.5978,
+      "step": 5490
+    },
+    {
+      "epoch": 4.033736707004033,
+      "grad_norm": 1.4798702001571655,
+      "learning_rate": 0.0002,
+      "loss": 0.6392,
+      "step": 5500
+    },
+    {
+      "epoch": 4.041070773744041,
+      "grad_norm": 0.9538470506668091,
+      "learning_rate": 0.0002,
+      "loss": 0.6143,
+      "step": 5510
+    },
+    {
+      "epoch": 4.048404840484048,
+      "grad_norm": 0.834561288356781,
+      "learning_rate": 0.0002,
+      "loss": 0.6056,
+      "step": 5520
+    },
+    {
+      "epoch": 4.055738907224056,
+      "grad_norm": 0.6407850384712219,
+      "learning_rate": 0.0002,
+      "loss": 0.6077,
+      "step": 5530
+    },
+    {
+      "epoch": 4.063072973964063,
+      "grad_norm": 0.9035961627960205,
+      "learning_rate": 0.0002,
+      "loss": 0.6733,
+      "step": 5540
+    },
+    {
+      "epoch": 4.070407040704071,
+      "grad_norm": 0.842812716960907,
+      "learning_rate": 0.0002,
+      "loss": 0.5854,
+      "step": 5550
+    },
+    {
+      "epoch": 4.077741107444078,
+      "grad_norm": 0.8197882175445557,
+      "learning_rate": 0.0002,
+      "loss": 0.654,
+      "step": 5560
+    },
+    {
+      "epoch": 4.085075174184085,
+      "grad_norm": 0.8652673959732056,
+      "learning_rate": 0.0002,
+      "loss": 0.5919,
+      "step": 5570
+    },
+    {
+      "epoch": 4.092409240924092,
+      "grad_norm": 0.8048318028450012,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 5580
+    },
+    {
+      "epoch": 4.0997433076641,
+      "grad_norm": 0.9604969024658203,
+      "learning_rate": 0.0002,
+      "loss": 0.6487,
+      "step": 5590
+    },
+    {
+      "epoch": 4.107077374404107,
+      "grad_norm": 1.244756817817688,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 5600
+    },
+    {
+      "epoch": 4.114411441144115,
+      "grad_norm": 0.7975269556045532,
+      "learning_rate": 0.0002,
+      "loss": 0.6489,
+      "step": 5610
+    },
+    {
+      "epoch": 4.121745507884122,
+      "grad_norm": 0.6130099296569824,
+      "learning_rate": 0.0002,
+      "loss": 0.6445,
+      "step": 5620
+    },
+    {
+      "epoch": 4.129079574624129,
+      "grad_norm": 0.7793202996253967,
+      "learning_rate": 0.0002,
+      "loss": 0.6024,
+      "step": 5630
+    },
+    {
+      "epoch": 4.136413641364136,
+      "grad_norm": 1.187238335609436,
+      "learning_rate": 0.0002,
+      "loss": 0.5723,
+      "step": 5640
+    },
+    {
+      "epoch": 4.143747708104144,
+      "grad_norm": 0.8450375199317932,
+      "learning_rate": 0.0002,
+      "loss": 0.6385,
+      "step": 5650
+    },
+    {
+      "epoch": 4.151081774844151,
+      "grad_norm": 0.9006940126419067,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 5660
+    },
+    {
+      "epoch": 4.158415841584159,
+      "grad_norm": 0.9447154998779297,
+      "learning_rate": 0.0002,
+      "loss": 0.6179,
+      "step": 5670
+    },
+    {
+      "epoch": 4.165749908324166,
+      "grad_norm": 0.798032283782959,
+      "learning_rate": 0.0002,
+      "loss": 0.6476,
+      "step": 5680
+    },
+    {
+      "epoch": 4.1730839750641735,
+      "grad_norm": 0.65578693151474,
+      "learning_rate": 0.0002,
+      "loss": 0.6666,
+      "step": 5690
+    },
+    {
+      "epoch": 4.18041804180418,
+      "grad_norm": 1.0864700078964233,
+      "learning_rate": 0.0002,
+      "loss": 0.701,
+      "step": 5700
+    },
+    {
+      "epoch": 4.187752108544188,
+      "grad_norm": 0.7344121932983398,
+      "learning_rate": 0.0002,
+      "loss": 0.6895,
+      "step": 5710
+    },
+    {
+      "epoch": 4.195086175284195,
+      "grad_norm": 0.9722456932067871,
+      "learning_rate": 0.0002,
+      "loss": 0.6659,
+      "step": 5720
+    },
+    {
+      "epoch": 4.2024202420242025,
+      "grad_norm": 1.263814926147461,
+      "learning_rate": 0.0002,
+      "loss": 0.6887,
+      "step": 5730
+    },
+    {
+      "epoch": 4.20975430876421,
+      "grad_norm": 0.9622581005096436,
+      "learning_rate": 0.0002,
+      "loss": 0.608,
+      "step": 5740
+    },
+    {
+      "epoch": 4.2170883755042174,
+      "grad_norm": 0.8497143387794495,
+      "learning_rate": 0.0002,
+      "loss": 0.6221,
+      "step": 5750
+    },
+    {
+      "epoch": 4.224422442244224,
+      "grad_norm": 0.8248446583747864,
+      "learning_rate": 0.0002,
+      "loss": 0.6322,
+      "step": 5760
+    },
+    {
+      "epoch": 4.2317565089842315,
+      "grad_norm": 1.2544798851013184,
+      "learning_rate": 0.0002,
+      "loss": 0.6045,
+      "step": 5770
+    },
+    {
+      "epoch": 4.239090575724239,
+      "grad_norm": 0.8224676251411438,
+      "learning_rate": 0.0002,
+      "loss": 0.641,
+      "step": 5780
+    },
+    {
+      "epoch": 4.2464246424642464,
+      "grad_norm": 0.8924877047538757,
+      "learning_rate": 0.0002,
+      "loss": 0.6399,
+      "step": 5790
+    },
+    {
+      "epoch": 4.253758709204254,
+      "grad_norm": 0.8545848727226257,
+      "learning_rate": 0.0002,
+      "loss": 0.6845,
+      "step": 5800
+    },
+    {
+      "epoch": 4.261092775944261,
+      "grad_norm": 0.8081067800521851,
+      "learning_rate": 0.0002,
+      "loss": 0.6669,
+      "step": 5810
+    },
+    {
+      "epoch": 4.268426842684269,
+      "grad_norm": 0.7111002802848816,
+      "learning_rate": 0.0002,
+      "loss": 0.6149,
+      "step": 5820
+    },
+    {
+      "epoch": 4.2757609094242754,
+      "grad_norm": 0.8696979880332947,
+      "learning_rate": 0.0002,
+      "loss": 0.6343,
+      "step": 5830
+    },
+    {
+      "epoch": 4.283094976164283,
+      "grad_norm": 0.821401834487915,
+      "learning_rate": 0.0002,
+      "loss": 0.6384,
+      "step": 5840
+    },
+    {
+      "epoch": 4.29042904290429,
+      "grad_norm": 0.888908326625824,
+      "learning_rate": 0.0002,
+      "loss": 0.6912,
+      "step": 5850
+    },
+    {
+      "epoch": 4.297763109644298,
+      "grad_norm": 1.9380123615264893,
+      "learning_rate": 0.0002,
+      "loss": 0.6061,
+      "step": 5860
+    },
+    {
+      "epoch": 4.305097176384305,
+      "grad_norm": 1.121774435043335,
+      "learning_rate": 0.0002,
+      "loss": 0.6766,
+      "step": 5870
+    },
+    {
+      "epoch": 4.312431243124313,
+      "grad_norm": 0.9238282442092896,
+      "learning_rate": 0.0002,
+      "loss": 0.7205,
+      "step": 5880
+    },
+    {
+      "epoch": 4.319765309864319,
+      "grad_norm": 0.7321620583534241,
+      "learning_rate": 0.0002,
+      "loss": 0.6351,
+      "step": 5890
+    },
+    {
+      "epoch": 4.327099376604327,
+      "grad_norm": 0.8739548325538635,
+      "learning_rate": 0.0002,
+      "loss": 0.6404,
+      "step": 5900
+    },
+    {
+      "epoch": 4.334433443344334,
+      "grad_norm": 0.9686012268066406,
+      "learning_rate": 0.0002,
+      "loss": 0.5892,
+      "step": 5910
+    },
+    {
+      "epoch": 4.341767510084342,
+      "grad_norm": 0.9033839106559753,
+      "learning_rate": 0.0002,
+      "loss": 0.641,
+      "step": 5920
+    },
+    {
+      "epoch": 4.349101576824349,
+      "grad_norm": 0.8131115436553955,
+      "learning_rate": 0.0002,
+      "loss": 0.6456,
+      "step": 5930
+    },
+    {
+      "epoch": 4.356435643564357,
+      "grad_norm": 0.8942412734031677,
+      "learning_rate": 0.0002,
+      "loss": 0.5826,
+      "step": 5940
+    },
+    {
+      "epoch": 4.363769710304364,
+      "grad_norm": 0.8439112901687622,
+      "learning_rate": 0.0002,
+      "loss": 0.7336,
+      "step": 5950
+    },
+    {
+      "epoch": 4.371103777044371,
+      "grad_norm": 0.9176713228225708,
+      "learning_rate": 0.0002,
+      "loss": 0.6537,
+      "step": 5960
+    },
+    {
+      "epoch": 4.378437843784378,
+      "grad_norm": 0.6799634695053101,
+      "learning_rate": 0.0002,
+      "loss": 0.6792,
+      "step": 5970
+    },
+    {
+      "epoch": 4.385771910524386,
+      "grad_norm": 1.0435824394226074,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 5980
+    },
+    {
+      "epoch": 4.393105977264393,
+      "grad_norm": 0.997937798500061,
+      "learning_rate": 0.0002,
+      "loss": 0.68,
+      "step": 5990
+    },
+    {
+      "epoch": 4.400440044004401,
+      "grad_norm": 1.0308842658996582,
+      "learning_rate": 0.0002,
+      "loss": 0.6604,
+      "step": 6000
+    },
+    {
+      "epoch": 4.407774110744408,
+      "grad_norm": 1.3683775663375854,
+      "learning_rate": 0.0002,
+      "loss": 0.6402,
+      "step": 6010
+    },
+    {
+      "epoch": 4.415108177484415,
+      "grad_norm": 0.7569534182548523,
+      "learning_rate": 0.0002,
+      "loss": 0.7027,
+      "step": 6020
+    },
+    {
+      "epoch": 4.422442244224422,
+      "grad_norm": 1.089978575706482,
+      "learning_rate": 0.0002,
+      "loss": 0.5949,
+      "step": 6030
+    },
+    {
+      "epoch": 4.42977631096443,
+      "grad_norm": 0.7522459626197815,
+      "learning_rate": 0.0002,
+      "loss": 0.6353,
+      "step": 6040
+    },
+    {
+      "epoch": 4.437110377704437,
+      "grad_norm": 0.6709823608398438,
+      "learning_rate": 0.0002,
+      "loss": 0.5852,
+      "step": 6050
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.6992089748382568,
+      "learning_rate": 0.0002,
+      "loss": 0.6718,
+      "step": 6060
+    },
+    {
+      "epoch": 4.451778511184452,
+      "grad_norm": 1.0182931423187256,
+      "learning_rate": 0.0002,
+      "loss": 0.6933,
+      "step": 6070
+    },
+    {
+      "epoch": 4.459112577924459,
+      "grad_norm": 1.0685160160064697,
+      "learning_rate": 0.0002,
+      "loss": 0.6255,
+      "step": 6080
+    },
+    {
+      "epoch": 4.466446644664466,
+      "grad_norm": 0.8295124769210815,
+      "learning_rate": 0.0002,
+      "loss": 0.6086,
+      "step": 6090
+    },
+    {
+      "epoch": 4.473780711404474,
+      "grad_norm": 1.1862998008728027,
+      "learning_rate": 0.0002,
+      "loss": 0.6359,
+      "step": 6100
+    },
+    {
+      "epoch": 4.481114778144481,
+      "grad_norm": 0.7400273084640503,
+      "learning_rate": 0.0002,
+      "loss": 0.638,
+      "step": 6110
+    },
+    {
+      "epoch": 4.488448844884489,
+      "grad_norm": 0.7098417282104492,
+      "learning_rate": 0.0002,
+      "loss": 0.6854,
+      "step": 6120
+    },
+    {
+      "epoch": 4.495782911624496,
+      "grad_norm": 0.9745053648948669,
+      "learning_rate": 0.0002,
+      "loss": 0.6976,
+      "step": 6130
+    },
+    {
+      "epoch": 4.503116978364503,
+      "grad_norm": 0.8638797998428345,
+      "learning_rate": 0.0002,
+      "loss": 0.605,
+      "step": 6140
+    },
+    {
+      "epoch": 4.51045104510451,
+      "grad_norm": 0.8291046619415283,
+      "learning_rate": 0.0002,
+      "loss": 0.6491,
+      "step": 6150
+    },
+    {
+      "epoch": 4.517785111844518,
+      "grad_norm": 1.0301737785339355,
+      "learning_rate": 0.0002,
+      "loss": 0.6457,
+      "step": 6160
+    },
+    {
+      "epoch": 4.525119178584525,
+      "grad_norm": 1.1996512413024902,
+      "learning_rate": 0.0002,
+      "loss": 0.6742,
+      "step": 6170
+    },
+    {
+      "epoch": 4.5324532453245325,
+      "grad_norm": 1.151038408279419,
+      "learning_rate": 0.0002,
+      "loss": 0.6484,
+      "step": 6180
+    },
+    {
+      "epoch": 4.53978731206454,
+      "grad_norm": 0.8385201096534729,
+      "learning_rate": 0.0002,
+      "loss": 0.668,
+      "step": 6190
+    },
+    {
+      "epoch": 4.5471213788045475,
+      "grad_norm": 0.8969188332557678,
+      "learning_rate": 0.0002,
+      "loss": 0.6381,
+      "step": 6200
+    },
+    {
+      "epoch": 4.554455445544555,
+      "grad_norm": 1.60659658908844,
+      "learning_rate": 0.0002,
+      "loss": 0.7141,
+      "step": 6210
+    },
+    {
+      "epoch": 4.5617895122845615,
+      "grad_norm": 0.9356731176376343,
+      "learning_rate": 0.0002,
+      "loss": 0.6388,
+      "step": 6220
+    },
+    {
+      "epoch": 4.569123579024569,
+      "grad_norm": 0.95856773853302,
+      "learning_rate": 0.0002,
+      "loss": 0.7393,
+      "step": 6230
+    },
+    {
+      "epoch": 4.5764576457645765,
+      "grad_norm": 1.1162524223327637,
+      "learning_rate": 0.0002,
+      "loss": 0.6554,
+      "step": 6240
+    },
+    {
+      "epoch": 4.583791712504584,
+      "grad_norm": 0.8809238076210022,
+      "learning_rate": 0.0002,
+      "loss": 0.6012,
+      "step": 6250
+    },
+    {
+      "epoch": 4.591125779244591,
+      "grad_norm": 0.890738844871521,
+      "learning_rate": 0.0002,
+      "loss": 0.648,
+      "step": 6260
+    },
+    {
+      "epoch": 4.598459845984598,
+      "grad_norm": 0.918684720993042,
+      "learning_rate": 0.0002,
+      "loss": 0.6663,
+      "step": 6270
+    },
+    {
+      "epoch": 4.6057939127246055,
+      "grad_norm": 0.8156296610832214,
+      "learning_rate": 0.0002,
+      "loss": 0.5992,
+      "step": 6280
+    },
+    {
+      "epoch": 4.613127979464613,
+      "grad_norm": 1.046634316444397,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 6290
+    },
+    {
+      "epoch": 4.62046204620462,
+      "grad_norm": 0.7725525498390198,
+      "learning_rate": 0.0002,
+      "loss": 0.7023,
+      "step": 6300
+    },
+    {
+      "epoch": 4.627796112944628,
+      "grad_norm": 0.9992046356201172,
+      "learning_rate": 0.0002,
+      "loss": 0.6414,
+      "step": 6310
+    },
+    {
+      "epoch": 4.635130179684635,
+      "grad_norm": 0.8480095267295837,
+      "learning_rate": 0.0002,
+      "loss": 0.6201,
+      "step": 6320
+    },
+    {
+      "epoch": 4.642464246424643,
+      "grad_norm": 0.7061955332756042,
+      "learning_rate": 0.0002,
+      "loss": 0.6869,
+      "step": 6330
+    },
+    {
+      "epoch": 4.649798313164649,
+      "grad_norm": 1.0354212522506714,
+      "learning_rate": 0.0002,
+      "loss": 0.6828,
+      "step": 6340
+    },
+    {
+      "epoch": 4.657132379904657,
+      "grad_norm": 1.0081377029418945,
+      "learning_rate": 0.0002,
+      "loss": 0.6651,
+      "step": 6350
+    },
+    {
+      "epoch": 4.664466446644664,
+      "grad_norm": 1.2904249429702759,
+      "learning_rate": 0.0002,
+      "loss": 0.726,
+      "step": 6360
+    },
+    {
+      "epoch": 4.671800513384672,
+      "grad_norm": 0.9248910546302795,
+      "learning_rate": 0.0002,
+      "loss": 0.7148,
+      "step": 6370
+    },
+    {
+      "epoch": 4.679134580124679,
+      "grad_norm": 0.9907804131507874,
+      "learning_rate": 0.0002,
+      "loss": 0.6961,
+      "step": 6380
+    },
+    {
+      "epoch": 4.686468646864687,
+      "grad_norm": 1.201143741607666,
+      "learning_rate": 0.0002,
+      "loss": 0.6163,
+      "step": 6390
+    },
+    {
+      "epoch": 4.693802713604693,
+      "grad_norm": 0.8709394335746765,
+      "learning_rate": 0.0002,
+      "loss": 0.6762,
+      "step": 6400
+    },
+    {
+      "epoch": 4.701136780344701,
+      "grad_norm": 0.7468608021736145,
+      "learning_rate": 0.0002,
+      "loss": 0.7217,
+      "step": 6410
+    },
+    {
+      "epoch": 4.708470847084708,
+      "grad_norm": 0.8607903718948364,
+      "learning_rate": 0.0002,
+      "loss": 0.6548,
+      "step": 6420
+    },
+    {
+      "epoch": 4.715804913824716,
+      "grad_norm": 0.9840512871742249,
+      "learning_rate": 0.0002,
+      "loss": 0.6449,
+      "step": 6430
+    },
+    {
+      "epoch": 4.723138980564723,
+      "grad_norm": 0.8328204154968262,
+      "learning_rate": 0.0002,
+      "loss": 0.685,
+      "step": 6440
+    },
+    {
+      "epoch": 4.730473047304731,
+      "grad_norm": 0.924505352973938,
+      "learning_rate": 0.0002,
+      "loss": 0.697,
+      "step": 6450
+    },
+    {
+      "epoch": 4.737807114044738,
+      "grad_norm": 0.8897685408592224,
+      "learning_rate": 0.0002,
+      "loss": 0.7422,
+      "step": 6460
+    },
+    {
+      "epoch": 4.745141180784745,
+      "grad_norm": 0.9605024456977844,
+      "learning_rate": 0.0002,
+      "loss": 0.6842,
+      "step": 6470
+    },
+    {
+      "epoch": 4.752475247524752,
+      "grad_norm": 0.8150759935379028,
+      "learning_rate": 0.0002,
+      "loss": 0.6488,
+      "step": 6480
+    },
+    {
+      "epoch": 4.75980931426476,
+      "grad_norm": 0.8128412961959839,
+      "learning_rate": 0.0002,
+      "loss": 0.6606,
+      "step": 6490
+    },
+    {
+      "epoch": 4.767143381004767,
+      "grad_norm": 0.7381404638290405,
+      "learning_rate": 0.0002,
+      "loss": 0.6729,
+      "step": 6500
+    },
+    {
+      "epoch": 4.774477447744775,
+      "grad_norm": 1.0565853118896484,
+      "learning_rate": 0.0002,
+      "loss": 0.6713,
+      "step": 6510
+    },
+    {
+      "epoch": 4.781811514484782,
+      "grad_norm": 0.9298134446144104,
+      "learning_rate": 0.0002,
+      "loss": 0.6496,
+      "step": 6520
+    },
+    {
+      "epoch": 4.789145581224789,
+      "grad_norm": 1.0145525932312012,
+      "learning_rate": 0.0002,
+      "loss": 0.7279,
+      "step": 6530
+    },
+    {
+      "epoch": 4.796479647964796,
+      "grad_norm": 0.92259681224823,
+      "learning_rate": 0.0002,
+      "loss": 0.5986,
+      "step": 6540
+    },
+    {
+      "epoch": 4.803813714704804,
+      "grad_norm": 0.7881024479866028,
+      "learning_rate": 0.0002,
+      "loss": 0.63,
+      "step": 6550
+    },
+    {
+      "epoch": 4.811147781444811,
+      "grad_norm": 1.4935206174850464,
+      "learning_rate": 0.0002,
+      "loss": 0.7134,
+      "step": 6560
+    },
+    {
+      "epoch": 4.818481848184819,
+      "grad_norm": 0.8612369298934937,
+      "learning_rate": 0.0002,
+      "loss": 0.6695,
+      "step": 6570
+    },
+    {
+      "epoch": 4.825815914924826,
+      "grad_norm": 1.0118653774261475,
+      "learning_rate": 0.0002,
+      "loss": 0.779,
+      "step": 6580
+    },
+    {
+      "epoch": 4.833149981664834,
+      "grad_norm": 1.1303809881210327,
+      "learning_rate": 0.0002,
+      "loss": 0.6991,
+      "step": 6590
+    },
+    {
+      "epoch": 4.84048404840484,
+      "grad_norm": 0.9112492203712463,
+      "learning_rate": 0.0002,
+      "loss": 0.7887,
+      "step": 6600
+    },
+    {
+      "epoch": 4.847818115144848,
+      "grad_norm": 0.864762544631958,
+      "learning_rate": 0.0002,
+      "loss": 0.7699,
+      "step": 6610
+    },
+    {
+      "epoch": 4.855152181884855,
+      "grad_norm": 0.9090572595596313,
+      "learning_rate": 0.0002,
+      "loss": 0.7347,
+      "step": 6620
+    },
+    {
+      "epoch": 4.862486248624863,
+      "grad_norm": 1.014953374862671,
+      "learning_rate": 0.0002,
+      "loss": 0.6608,
+      "step": 6630
+    },
+    {
+      "epoch": 4.86982031536487,
+      "grad_norm": 1.0702149868011475,
+      "learning_rate": 0.0002,
+      "loss": 0.6429,
+      "step": 6640
+    },
+    {
+      "epoch": 4.8771543821048775,
+      "grad_norm": 1.002135157585144,
+      "learning_rate": 0.0002,
+      "loss": 0.6943,
+      "step": 6650
+    },
+    {
+      "epoch": 4.884488448844884,
+      "grad_norm": 0.862545907497406,
+      "learning_rate": 0.0002,
+      "loss": 0.7225,
+      "step": 6660
+    },
+    {
+      "epoch": 4.891822515584892,
+      "grad_norm": 0.7302131056785583,
+      "learning_rate": 0.0002,
+      "loss": 0.6206,
+      "step": 6670
+    },
+    {
+      "epoch": 4.899156582324899,
+      "grad_norm": 0.8380730152130127,
+      "learning_rate": 0.0002,
+      "loss": 0.7175,
+      "step": 6680
+    },
+    {
+      "epoch": 4.9064906490649065,
+      "grad_norm": 0.7956018447875977,
+      "learning_rate": 0.0002,
+      "loss": 0.645,
+      "step": 6690
+    },
+    {
+      "epoch": 4.913824715804914,
+      "grad_norm": 0.6717583537101746,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 6700
+    },
+    {
+      "epoch": 4.9211587825449215,
+      "grad_norm": 1.09099280834198,
+      "learning_rate": 0.0002,
+      "loss": 0.6942,
+      "step": 6710
+    },
+    {
+      "epoch": 4.928492849284929,
+      "grad_norm": 0.8589889407157898,
+      "learning_rate": 0.0002,
+      "loss": 0.7533,
+      "step": 6720
+    },
+    {
+      "epoch": 4.9358269160249355,
+      "grad_norm": 1.0046314001083374,
+      "learning_rate": 0.0002,
+      "loss": 0.66,
+      "step": 6730
+    },
+    {
+      "epoch": 4.943160982764943,
+      "grad_norm": 0.8559659123420715,
+      "learning_rate": 0.0002,
+      "loss": 0.6864,
+      "step": 6740
+    },
+    {
+      "epoch": 4.9504950495049505,
+      "grad_norm": 0.8588525652885437,
+      "learning_rate": 0.0002,
+      "loss": 0.6847,
+      "step": 6750
+    },
+    {
+      "epoch": 4.957829116244958,
+      "grad_norm": 0.9192708134651184,
+      "learning_rate": 0.0002,
+      "loss": 0.6428,
+      "step": 6760
+    },
+    {
+      "epoch": 4.965163182984965,
+      "grad_norm": 1.051398754119873,
+      "learning_rate": 0.0002,
+      "loss": 0.6873,
+      "step": 6770
+    },
+    {
+      "epoch": 4.972497249724973,
+      "grad_norm": 0.9111362099647522,
+      "learning_rate": 0.0002,
+      "loss": 0.7249,
+      "step": 6780
+    },
+    {
+      "epoch": 4.9798313164649795,
+      "grad_norm": 0.7305638194084167,
+      "learning_rate": 0.0002,
+      "loss": 0.7613,
+      "step": 6790
+    },
+    {
+      "epoch": 4.987165383204987,
+      "grad_norm": 1.118837594985962,
+      "learning_rate": 0.0002,
+      "loss": 0.6747,
+      "step": 6800
+    },
+    {
+      "epoch": 4.994499449944994,
+      "grad_norm": 0.9075239300727844,
+      "learning_rate": 0.0002,
+      "loss": 0.6412,
+      "step": 6810
+    },
+    {
+      "epoch": 4.999633296662999,
+      "eval_loss": 1.2361247539520264,
+      "eval_runtime": 32.7325,
+      "eval_samples_per_second": 13.167,
+      "eval_steps_per_second": 1.65,
+      "step": 6817
+    },
+    {
+      "epoch": 5.001833516685002,
+      "grad_norm": 1.0541315078735352,
+      "learning_rate": 0.0002,
+      "loss": 0.7091,
+      "step": 6820
+    },
+    {
+      "epoch": 5.009167583425009,
+      "grad_norm": 0.9750140905380249,
+      "learning_rate": 0.0002,
+      "loss": 0.4882,
+      "step": 6830
+    },
+    {
+      "epoch": 5.016501650165017,
+      "grad_norm": 0.931838870048523,
+      "learning_rate": 0.0002,
+      "loss": 0.6022,
+      "step": 6840
+    },
+    {
+      "epoch": 5.023835716905023,
+      "grad_norm": 1.110278844833374,
+      "learning_rate": 0.0002,
+      "loss": 0.5194,
+      "step": 6850
+    },
+    {
+      "epoch": 5.031169783645031,
+      "grad_norm": 1.0670180320739746,
+      "learning_rate": 0.0002,
+      "loss": 0.4676,
+      "step": 6860
+    },
+    {
+      "epoch": 5.038503850385038,
+      "grad_norm": 0.8762092590332031,
+      "learning_rate": 0.0002,
+      "loss": 0.4374,
+      "step": 6870
+    },
+    {
+      "epoch": 5.045837917125046,
+      "grad_norm": 1.1169432401657104,
+      "learning_rate": 0.0002,
+      "loss": 0.505,
+      "step": 6880
+    },
+    {
+      "epoch": 5.053171983865053,
+      "grad_norm": 1.005491018295288,
+      "learning_rate": 0.0002,
+      "loss": 0.5114,
+      "step": 6890
+    },
+    {
+      "epoch": 5.060506050605061,
+      "grad_norm": 1.1751841306686401,
+      "learning_rate": 0.0002,
+      "loss": 0.5221,
+      "step": 6900
+    },
+    {
+      "epoch": 5.067840117345068,
+      "grad_norm": 0.8501367568969727,
+      "learning_rate": 0.0002,
+      "loss": 0.451,
+      "step": 6910
+    },
+    {
+      "epoch": 5.075174184085075,
+      "grad_norm": 0.9795131683349609,
+      "learning_rate": 0.0002,
+      "loss": 0.5292,
+      "step": 6920
+    },
+    {
+      "epoch": 5.082508250825082,
+      "grad_norm": 0.8929879665374756,
+      "learning_rate": 0.0002,
+      "loss": 0.5234,
+      "step": 6930
+    },
+    {
+      "epoch": 5.08984231756509,
+      "grad_norm": 1.0156651735305786,
+      "learning_rate": 0.0002,
+      "loss": 0.5378,
+      "step": 6940
+    },
+    {
+      "epoch": 5.097176384305097,
+      "grad_norm": 1.0974335670471191,
+      "learning_rate": 0.0002,
+      "loss": 0.5241,
+      "step": 6950
+    },
+    {
+      "epoch": 5.104510451045105,
+      "grad_norm": 1.7015666961669922,
+      "learning_rate": 0.0002,
+      "loss": 0.5705,
+      "step": 6960
+    },
+    {
+      "epoch": 5.111844517785112,
+      "grad_norm": 1.0343226194381714,
+      "learning_rate": 0.0002,
+      "loss": 0.523,
+      "step": 6970
+    },
+    {
+      "epoch": 5.119178584525119,
+      "grad_norm": 1.3072983026504517,
+      "learning_rate": 0.0002,
+      "loss": 0.4616,
+      "step": 6980
+    },
+    {
+      "epoch": 5.126512651265126,
+      "grad_norm": 1.038986086845398,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 6990
+    },
+    {
+      "epoch": 5.133846718005134,
+      "grad_norm": 0.8638386130332947,
+      "learning_rate": 0.0002,
+      "loss": 0.4616,
+      "step": 7000
+    },
+    {
+      "epoch": 5.141180784745141,
+      "grad_norm": 0.8326523900032043,
+      "learning_rate": 0.0002,
+      "loss": 0.5294,
+      "step": 7010
+    },
+    {
+      "epoch": 5.148514851485149,
+      "grad_norm": 1.0976895093917847,
+      "learning_rate": 0.0002,
+      "loss": 0.5021,
+      "step": 7020
+    },
+    {
+      "epoch": 5.155848918225156,
+      "grad_norm": 1.0077873468399048,
+      "learning_rate": 0.0002,
+      "loss": 0.4677,
+      "step": 7030
+    },
+    {
+      "epoch": 5.163182984965164,
+      "grad_norm": 1.0662257671356201,
+      "learning_rate": 0.0002,
+      "loss": 0.5262,
+      "step": 7040
+    },
+    {
+      "epoch": 5.17051705170517,
+      "grad_norm": 1.206271767616272,
+      "learning_rate": 0.0002,
+      "loss": 0.5484,
+      "step": 7050
+    },
+    {
+      "epoch": 5.177851118445178,
+      "grad_norm": 1.1990262269973755,
+      "learning_rate": 0.0002,
+      "loss": 0.4817,
+      "step": 7060
+    },
+    {
+      "epoch": 5.185185185185185,
+      "grad_norm": 1.0207163095474243,
+      "learning_rate": 0.0002,
+      "loss": 0.6048,
+      "step": 7070
+    },
+    {
+      "epoch": 5.192519251925193,
+      "grad_norm": 1.2783987522125244,
+      "learning_rate": 0.0002,
+      "loss": 0.4816,
+      "step": 7080
+    },
+    {
+      "epoch": 5.1998533186652,
+      "grad_norm": 1.1592512130737305,
+      "learning_rate": 0.0002,
+      "loss": 0.5322,
+      "step": 7090
+    },
+    {
+      "epoch": 5.2071873854052075,
+      "grad_norm": 1.1053160429000854,
+      "learning_rate": 0.0002,
+      "loss": 0.5472,
+      "step": 7100
+    },
+    {
+      "epoch": 5.214521452145214,
+      "grad_norm": 1.1925510168075562,
+      "learning_rate": 0.0002,
+      "loss": 0.4986,
+      "step": 7110
+    },
+    {
+      "epoch": 5.221855518885222,
+      "grad_norm": 1.0714877843856812,
+      "learning_rate": 0.0002,
+      "loss": 0.5065,
+      "step": 7120
+    },
+    {
+      "epoch": 5.229189585625229,
+      "grad_norm": 0.9451011419296265,
+      "learning_rate": 0.0002,
+      "loss": 0.5209,
+      "step": 7130
+    },
+    {
+      "epoch": 5.2365236523652365,
+      "grad_norm": 1.03838050365448,
+      "learning_rate": 0.0002,
+      "loss": 0.5298,
+      "step": 7140
+    },
+    {
+      "epoch": 5.243857719105244,
+      "grad_norm": 0.9204146265983582,
+      "learning_rate": 0.0002,
+      "loss": 0.4848,
+      "step": 7150
+    },
+    {
+      "epoch": 5.2511917858452515,
+      "grad_norm": 1.0142229795455933,
+      "learning_rate": 0.0002,
+      "loss": 0.5164,
+      "step": 7160
+    },
+    {
+      "epoch": 5.258525852585258,
+      "grad_norm": 1.4432005882263184,
+      "learning_rate": 0.0002,
+      "loss": 0.5092,
+      "step": 7170
+    },
+    {
+      "epoch": 5.2658599193252655,
+      "grad_norm": 1.1239633560180664,
+      "learning_rate": 0.0002,
+      "loss": 0.5133,
+      "step": 7180
+    },
+    {
+      "epoch": 5.273193986065273,
+      "grad_norm": 0.7012821435928345,
+      "learning_rate": 0.0002,
+      "loss": 0.4969,
+      "step": 7190
+    },
+    {
+      "epoch": 5.2805280528052805,
+      "grad_norm": 1.3499128818511963,
+      "learning_rate": 0.0002,
+      "loss": 0.5466,
+      "step": 7200
+    },
+    {
+      "epoch": 5.287862119545288,
+      "grad_norm": 0.9498730897903442,
+      "learning_rate": 0.0002,
+      "loss": 0.5282,
+      "step": 7210
+    },
+    {
+      "epoch": 5.295196186285295,
+      "grad_norm": 0.9552369117736816,
+      "learning_rate": 0.0002,
+      "loss": 0.5051,
+      "step": 7220
+    },
+    {
+      "epoch": 5.302530253025303,
+      "grad_norm": 0.7610348463058472,
+      "learning_rate": 0.0002,
+      "loss": 0.5329,
+      "step": 7230
+    },
+    {
+      "epoch": 5.3098643197653095,
+      "grad_norm": 1.0314512252807617,
+      "learning_rate": 0.0002,
+      "loss": 0.468,
+      "step": 7240
+    },
+    {
+      "epoch": 5.317198386505317,
+      "grad_norm": 1.0534334182739258,
+      "learning_rate": 0.0002,
+      "loss": 0.5367,
+      "step": 7250
+    },
+    {
+      "epoch": 5.324532453245324,
+      "grad_norm": 1.2553406953811646,
+      "learning_rate": 0.0002,
+      "loss": 0.5491,
+      "step": 7260
+    },
+    {
+      "epoch": 5.331866519985332,
+      "grad_norm": 0.7061691880226135,
+      "learning_rate": 0.0002,
+      "loss": 0.5218,
+      "step": 7270
+    },
+    {
+      "epoch": 5.339200586725339,
+      "grad_norm": 0.9652578830718994,
+      "learning_rate": 0.0002,
+      "loss": 0.5625,
+      "step": 7280
+    },
+    {
+      "epoch": 5.346534653465347,
+      "grad_norm": 1.114788293838501,
+      "learning_rate": 0.0002,
+      "loss": 0.5608,
+      "step": 7290
+    },
+    {
+      "epoch": 5.353868720205353,
+      "grad_norm": 1.0940049886703491,
+      "learning_rate": 0.0002,
+      "loss": 0.578,
+      "step": 7300
+    },
+    {
+      "epoch": 5.361202786945361,
+      "grad_norm": 1.0151008367538452,
+      "learning_rate": 0.0002,
+      "loss": 0.5256,
+      "step": 7310
+    },
+    {
+      "epoch": 5.368536853685368,
+      "grad_norm": 1.0369552373886108,
+      "learning_rate": 0.0002,
+      "loss": 0.5377,
+      "step": 7320
+    },
+    {
+      "epoch": 5.375870920425376,
+      "grad_norm": 0.8489866256713867,
+      "learning_rate": 0.0002,
+      "loss": 0.5028,
+      "step": 7330
+    },
+    {
+      "epoch": 5.383204987165383,
+      "grad_norm": 1.1031713485717773,
+      "learning_rate": 0.0002,
+      "loss": 0.5937,
+      "step": 7340
+    },
+    {
+      "epoch": 5.390539053905391,
+      "grad_norm": 0.9094716310501099,
+      "learning_rate": 0.0002,
+      "loss": 0.5355,
+      "step": 7350
+    },
+    {
+      "epoch": 5.397873120645398,
+      "grad_norm": 0.9530431032180786,
+      "learning_rate": 0.0002,
+      "loss": 0.5406,
+      "step": 7360
+    },
+    {
+      "epoch": 5.405207187385405,
+      "grad_norm": 0.9633604884147644,
+      "learning_rate": 0.0002,
+      "loss": 0.529,
+      "step": 7370
+    },
+    {
+      "epoch": 5.412541254125412,
+      "grad_norm": 0.9541662335395813,
+      "learning_rate": 0.0002,
+      "loss": 0.5315,
+      "step": 7380
+    },
+    {
+      "epoch": 5.41987532086542,
+      "grad_norm": 1.0459771156311035,
+      "learning_rate": 0.0002,
+      "loss": 0.6774,
+      "step": 7390
+    },
+    {
+      "epoch": 5.427209387605427,
+      "grad_norm": 1.027388334274292,
+      "learning_rate": 0.0002,
+      "loss": 0.5737,
+      "step": 7400
+    },
+    {
+      "epoch": 5.434543454345435,
+      "grad_norm": 0.7267653346061707,
+      "learning_rate": 0.0002,
+      "loss": 0.556,
+      "step": 7410
+    },
+    {
+      "epoch": 5.441877521085442,
+      "grad_norm": 1.020142674446106,
+      "learning_rate": 0.0002,
+      "loss": 0.4581,
+      "step": 7420
+    },
+    {
+      "epoch": 5.449211587825449,
+      "grad_norm": 1.044754147529602,
+      "learning_rate": 0.0002,
+      "loss": 0.4853,
+      "step": 7430
+    },
+    {
+      "epoch": 5.456545654565456,
+      "grad_norm": 1.5476195812225342,
+      "learning_rate": 0.0002,
+      "loss": 0.5666,
+      "step": 7440
+    },
+    {
+      "epoch": 5.463879721305464,
+      "grad_norm": 0.9879506826400757,
+      "learning_rate": 0.0002,
+      "loss": 0.5302,
+      "step": 7450
+    },
+    {
+      "epoch": 5.471213788045471,
+      "grad_norm": 1.2562980651855469,
+      "learning_rate": 0.0002,
+      "loss": 0.591,
+      "step": 7460
+    },
+    {
+      "epoch": 5.478547854785479,
+      "grad_norm": 1.3051384687423706,
+      "learning_rate": 0.0002,
+      "loss": 0.5188,
+      "step": 7470
+    },
+    {
+      "epoch": 5.485881921525486,
+      "grad_norm": 1.0511597394943237,
+      "learning_rate": 0.0002,
+      "loss": 0.5658,
+      "step": 7480
+    },
+    {
+      "epoch": 5.493215988265494,
+      "grad_norm": 1.0380817651748657,
+      "learning_rate": 0.0002,
+      "loss": 0.6327,
+      "step": 7490
+    },
+    {
+      "epoch": 5.5005500550055,
+      "grad_norm": 1.170274257659912,
+      "learning_rate": 0.0002,
+      "loss": 0.5356,
+      "step": 7500
+    },
+    {
+      "epoch": 5.507884121745508,
+      "grad_norm": 1.3356517553329468,
+      "learning_rate": 0.0002,
+      "loss": 0.5405,
+      "step": 7510
+    },
+    {
+      "epoch": 5.515218188485515,
+      "grad_norm": 1.0727124214172363,
+      "learning_rate": 0.0002,
+      "loss": 0.5305,
+      "step": 7520
+    },
+    {
+      "epoch": 5.522552255225523,
+      "grad_norm": 1.0110199451446533,
+      "learning_rate": 0.0002,
+      "loss": 0.5543,
+      "step": 7530
+    },
+    {
+      "epoch": 5.52988632196553,
+      "grad_norm": 1.3086743354797363,
+      "learning_rate": 0.0002,
+      "loss": 0.5962,
+      "step": 7540
+    },
+    {
+      "epoch": 5.537220388705538,
+      "grad_norm": 1.1904916763305664,
+      "learning_rate": 0.0002,
+      "loss": 0.5512,
+      "step": 7550
+    },
+    {
+      "epoch": 5.544554455445544,
+      "grad_norm": 0.9466280937194824,
+      "learning_rate": 0.0002,
+      "loss": 0.5915,
+      "step": 7560
+    },
+    {
+      "epoch": 5.551888522185552,
+      "grad_norm": 1.1237901449203491,
+      "learning_rate": 0.0002,
+      "loss": 0.5573,
+      "step": 7570
+    },
+    {
+      "epoch": 5.559222588925559,
+      "grad_norm": 0.9590660333633423,
+      "learning_rate": 0.0002,
+      "loss": 0.5383,
+      "step": 7580
+    },
+    {
+      "epoch": 5.566556655665567,
+      "grad_norm": 1.0890778303146362,
+      "learning_rate": 0.0002,
+      "loss": 0.5594,
+      "step": 7590
+    },
+    {
+      "epoch": 5.573890722405574,
+      "grad_norm": 0.7206931114196777,
+      "learning_rate": 0.0002,
+      "loss": 0.5698,
+      "step": 7600
+    },
+    {
+      "epoch": 5.5812247891455815,
+      "grad_norm": 1.2884514331817627,
+      "learning_rate": 0.0002,
+      "loss": 0.5511,
+      "step": 7610
+    },
+    {
+      "epoch": 5.588558855885589,
+      "grad_norm": 0.7798039317131042,
+      "learning_rate": 0.0002,
+      "loss": 0.5279,
+      "step": 7620
+    },
+    {
+      "epoch": 5.595892922625596,
+      "grad_norm": 1.166046142578125,
+      "learning_rate": 0.0002,
+      "loss": 0.4847,
+      "step": 7630
+    },
+    {
+      "epoch": 5.603226989365603,
+      "grad_norm": 1.0150201320648193,
+      "learning_rate": 0.0002,
+      "loss": 0.5821,
+      "step": 7640
+    },
+    {
+      "epoch": 5.6105610561056105,
+      "grad_norm": 1.0449682474136353,
+      "learning_rate": 0.0002,
+      "loss": 0.5296,
+      "step": 7650
+    },
+    {
+      "epoch": 5.617895122845618,
+      "grad_norm": 0.9310530424118042,
+      "learning_rate": 0.0002,
+      "loss": 0.5431,
+      "step": 7660
+    },
+    {
+      "epoch": 5.6252291895856255,
+      "grad_norm": 0.9117933511734009,
+      "learning_rate": 0.0002,
+      "loss": 0.5234,
+      "step": 7670
+    },
+    {
+      "epoch": 5.632563256325633,
+      "grad_norm": 1.1475164890289307,
+      "learning_rate": 0.0002,
+      "loss": 0.5807,
+      "step": 7680
+    },
+    {
+      "epoch": 5.6398973230656395,
+      "grad_norm": 1.066809058189392,
+      "learning_rate": 0.0002,
+      "loss": 0.5816,
+      "step": 7690
+    },
+    {
+      "epoch": 5.647231389805647,
+      "grad_norm": 1.2834991216659546,
+      "learning_rate": 0.0002,
+      "loss": 0.551,
+      "step": 7700
+    },
+    {
+      "epoch": 5.6545654565456545,
+      "grad_norm": 1.2245112657546997,
+      "learning_rate": 0.0002,
+      "loss": 0.5914,
+      "step": 7710
+    },
+    {
+      "epoch": 5.661899523285662,
+      "grad_norm": 1.1424106359481812,
+      "learning_rate": 0.0002,
+      "loss": 0.5552,
+      "step": 7720
+    },
+    {
+      "epoch": 5.669233590025669,
+      "grad_norm": 1.0673892498016357,
+      "learning_rate": 0.0002,
+      "loss": 0.559,
+      "step": 7730
+    },
+    {
+      "epoch": 5.676567656765677,
+      "grad_norm": 1.4312121868133545,
+      "learning_rate": 0.0002,
+      "loss": 0.544,
+      "step": 7740
+    },
+    {
+      "epoch": 5.683901723505684,
+      "grad_norm": 0.9976982474327087,
+      "learning_rate": 0.0002,
+      "loss": 0.5576,
+      "step": 7750
+    },
+    {
+      "epoch": 5.691235790245691,
+      "grad_norm": 0.9464678168296814,
+      "learning_rate": 0.0002,
+      "loss": 0.4855,
+      "step": 7760
+    },
+    {
+      "epoch": 5.698569856985698,
+      "grad_norm": 1.010995626449585,
+      "learning_rate": 0.0002,
+      "loss": 0.5363,
+      "step": 7770
+    },
+    {
+      "epoch": 5.705903923725706,
+      "grad_norm": 1.3787750005722046,
+      "learning_rate": 0.0002,
+      "loss": 0.5873,
+      "step": 7780
+    },
+    {
+      "epoch": 5.713237990465713,
+      "grad_norm": 1.020922303199768,
+      "learning_rate": 0.0002,
+      "loss": 0.6234,
+      "step": 7790
+    },
+    {
+      "epoch": 5.720572057205721,
+      "grad_norm": 0.9748636484146118,
+      "learning_rate": 0.0002,
+      "loss": 0.5337,
+      "step": 7800
+    },
+    {
+      "epoch": 5.727906123945728,
+      "grad_norm": 1.3077744245529175,
+      "learning_rate": 0.0002,
+      "loss": 0.5507,
+      "step": 7810
+    },
+    {
+      "epoch": 5.735240190685735,
+      "grad_norm": 1.4770057201385498,
+      "learning_rate": 0.0002,
+      "loss": 0.558,
+      "step": 7820
+    },
+    {
+      "epoch": 5.742574257425742,
+      "grad_norm": 1.6349090337753296,
+      "learning_rate": 0.0002,
+      "loss": 0.5571,
+      "step": 7830
+    },
+    {
+      "epoch": 5.74990832416575,
+      "grad_norm": 0.9818630814552307,
+      "learning_rate": 0.0002,
+      "loss": 0.5056,
+      "step": 7840
+    },
+    {
+      "epoch": 5.757242390905757,
+      "grad_norm": 0.9659715890884399,
+      "learning_rate": 0.0002,
+      "loss": 0.5495,
+      "step": 7850
+    },
+    {
+      "epoch": 5.764576457645765,
+      "grad_norm": 0.9269950985908508,
+      "learning_rate": 0.0002,
+      "loss": 0.5628,
+      "step": 7860
+    },
+    {
+      "epoch": 5.771910524385772,
+      "grad_norm": 1.0099073648452759,
+      "learning_rate": 0.0002,
+      "loss": 0.5594,
+      "step": 7870
+    },
+    {
+      "epoch": 5.77924459112578,
+      "grad_norm": 0.9123615026473999,
+      "learning_rate": 0.0002,
+      "loss": 0.5912,
+      "step": 7880
+    },
+    {
+      "epoch": 5.786578657865786,
+      "grad_norm": 1.1542246341705322,
+      "learning_rate": 0.0002,
+      "loss": 0.6054,
+      "step": 7890
+    },
+    {
+      "epoch": 5.793912724605794,
+      "grad_norm": 1.0792022943496704,
+      "learning_rate": 0.0002,
+      "loss": 0.5829,
+      "step": 7900
+    },
+    {
+      "epoch": 5.801246791345801,
+      "grad_norm": 0.95615553855896,
+      "learning_rate": 0.0002,
+      "loss": 0.504,
+      "step": 7910
+    },
+    {
+      "epoch": 5.808580858085809,
+      "grad_norm": 1.2471332550048828,
+      "learning_rate": 0.0002,
+      "loss": 0.5918,
+      "step": 7920
+    },
+    {
+      "epoch": 5.815914924825816,
+      "grad_norm": 1.0189851522445679,
+      "learning_rate": 0.0002,
+      "loss": 0.5719,
+      "step": 7930
+    },
+    {
+      "epoch": 5.823248991565823,
+      "grad_norm": 1.3309742212295532,
+      "learning_rate": 0.0002,
+      "loss": 0.5958,
+      "step": 7940
+    },
+    {
+      "epoch": 5.83058305830583,
+      "grad_norm": 1.2930549383163452,
+      "learning_rate": 0.0002,
+      "loss": 0.6255,
+      "step": 7950
+    },
+    {
+      "epoch": 5.837917125045838,
+      "grad_norm": 0.8216308951377869,
+      "learning_rate": 0.0002,
+      "loss": 0.5301,
+      "step": 7960
+    },
+    {
+      "epoch": 5.845251191785845,
+      "grad_norm": 1.1205775737762451,
+      "learning_rate": 0.0002,
+      "loss": 0.5397,
+      "step": 7970
+    },
+    {
+      "epoch": 5.852585258525853,
+      "grad_norm": 0.851298451423645,
+      "learning_rate": 0.0002,
+      "loss": 0.5903,
+      "step": 7980
+    },
+    {
+      "epoch": 5.85991932526586,
+      "grad_norm": 0.8797095417976379,
+      "learning_rate": 0.0002,
+      "loss": 0.5981,
+      "step": 7990
+    },
+    {
+      "epoch": 5.867253392005868,
+      "grad_norm": 1.5784614086151123,
+      "learning_rate": 0.0002,
+      "loss": 0.6106,
+      "step": 8000
+    },
+    {
+      "epoch": 5.874587458745875,
+      "grad_norm": 1.1531187295913696,
+      "learning_rate": 0.0002,
+      "loss": 0.5956,
+      "step": 8010
+    },
+    {
+      "epoch": 5.881921525485882,
+      "grad_norm": 1.2469146251678467,
+      "learning_rate": 0.0002,
+      "loss": 0.6289,
+      "step": 8020
+    },
+    {
+      "epoch": 5.889255592225889,
+      "grad_norm": 1.0784350633621216,
+      "learning_rate": 0.0002,
+      "loss": 0.5827,
+      "step": 8030
+    },
+    {
+      "epoch": 5.896589658965897,
+      "grad_norm": 1.1311599016189575,
+      "learning_rate": 0.0002,
+      "loss": 0.6339,
+      "step": 8040
+    },
+    {
+      "epoch": 5.903923725705904,
+      "grad_norm": 0.9654512405395508,
+      "learning_rate": 0.0002,
+      "loss": 0.5815,
+      "step": 8050
+    },
+    {
+      "epoch": 5.9112577924459115,
+      "grad_norm": 1.3288270235061646,
+      "learning_rate": 0.0002,
+      "loss": 0.6198,
+      "step": 8060
+    },
+    {
+      "epoch": 5.918591859185918,
+      "grad_norm": 1.12800931930542,
+      "learning_rate": 0.0002,
+      "loss": 0.6515,
+      "step": 8070
+    },
+    {
+      "epoch": 5.925925925925926,
+      "grad_norm": 0.9449917674064636,
+      "learning_rate": 0.0002,
+      "loss": 0.5684,
+      "step": 8080
+    },
+    {
+      "epoch": 5.933259992665933,
+      "grad_norm": 1.1532357931137085,
+      "learning_rate": 0.0002,
+      "loss": 0.6063,
+      "step": 8090
+    },
+    {
+      "epoch": 5.9405940594059405,
+      "grad_norm": 1.2211151123046875,
+      "learning_rate": 0.0002,
+      "loss": 0.5318,
+      "step": 8100
+    },
+    {
+      "epoch": 5.947928126145948,
+      "grad_norm": 1.3459105491638184,
+      "learning_rate": 0.0002,
+      "loss": 0.6512,
+      "step": 8110
+    },
+    {
+      "epoch": 5.9552621928859555,
+      "grad_norm": 1.251999855041504,
+      "learning_rate": 0.0002,
+      "loss": 0.5952,
+      "step": 8120
+    },
+    {
+      "epoch": 5.962596259625963,
+      "grad_norm": 1.5682506561279297,
+      "learning_rate": 0.0002,
+      "loss": 0.6203,
+      "step": 8130
+    },
+    {
+      "epoch": 5.9699303263659695,
+      "grad_norm": 0.926075279712677,
+      "learning_rate": 0.0002,
+      "loss": 0.6253,
+      "step": 8140
+    },
+    {
+      "epoch": 5.977264393105977,
+      "grad_norm": 0.9622511863708496,
+      "learning_rate": 0.0002,
+      "loss": 0.5545,
+      "step": 8150
+    },
+    {
+      "epoch": 5.9845984598459845,
+      "grad_norm": 0.9633373618125916,
+      "learning_rate": 0.0002,
+      "loss": 0.5518,
+      "step": 8160
+    },
+    {
+      "epoch": 5.991932526585992,
+      "grad_norm": 0.8960476517677307,
+      "learning_rate": 0.0002,
+      "loss": 0.5831,
+      "step": 8170
+    },
+    {
+      "epoch": 5.999266593325999,
+      "grad_norm": 0.9372805953025818,
+      "learning_rate": 0.0002,
+      "loss": 0.5442,
+      "step": 8180
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.3233846426010132,
+      "eval_runtime": 32.7419,
+      "eval_samples_per_second": 13.164,
+      "eval_steps_per_second": 1.649,
+      "step": 8181
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10904,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "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": 3.785983619990815e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..2868cff7027115396e695775cacd838522aca295
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-8181/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b12b6f6817632087b5a5e37d744e25312b96e839de5005320b96bc0c2473c41f
+size 5624
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/README.md b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# 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 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "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": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7cf9e959e32046d22d776998db9f62da2f301a35
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9bf2e18a4874146b707384d0e34b6a5ffdda8564101e38b2f64f81beb762de
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7de281b3259db1c7a5f003c28543f8476618c91d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c69583b3ea1be11ba1035d9f27d924ff5ebff08b05184c101a63cee86e38850
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..2c0afe783aec92385b20bf53b88de93cf5800656
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdfd0044ec7097c0370428a56f4983cc513ae1f01d46173d06bfb7c1a890bdb6
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8c18e8ea45f03d57945d8cedf736c7cb140071a1
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14322ec31c8207dffb4e78821ba511f674f7d5cc97a6efc3327c3ce666090d22
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..178d72b936e4bf0bc804a7530ac341e5d22400f0
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/trainer_state.json
@@ -0,0 +1,6767 @@
+{
+  "best_metric": 1.1534006595611572,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727",
+  "epoch": 6.999633296662999,
+  "eval_steps": 10,
+  "global_step": 9544,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007334066740007334,
+      "grad_norm": 0.47521963715553284,
+      "learning_rate": 0.0002,
+      "loss": 1.9722,
+      "step": 10
+    },
+    {
+      "epoch": 0.014668133480014669,
+      "grad_norm": 0.5395162105560303,
+      "learning_rate": 0.0002,
+      "loss": 1.4821,
+      "step": 20
+    },
+    {
+      "epoch": 0.022002200220022004,
+      "grad_norm": 0.4305780231952667,
+      "learning_rate": 0.0002,
+      "loss": 1.4202,
+      "step": 30
+    },
+    {
+      "epoch": 0.029336266960029337,
+      "grad_norm": 0.6938246488571167,
+      "learning_rate": 0.0002,
+      "loss": 1.4271,
+      "step": 40
+    },
+    {
+      "epoch": 0.03667033370003667,
+      "grad_norm": 1.5133819580078125,
+      "learning_rate": 0.0002,
+      "loss": 1.3112,
+      "step": 50
+    },
+    {
+      "epoch": 0.04400440044004401,
+      "grad_norm": 0.9173883199691772,
+      "learning_rate": 0.0002,
+      "loss": 1.3132,
+      "step": 60
+    },
+    {
+      "epoch": 0.05133846718005134,
+      "grad_norm": 0.4619861841201782,
+      "learning_rate": 0.0002,
+      "loss": 1.2844,
+      "step": 70
+    },
+    {
+      "epoch": 0.058672533920058674,
+      "grad_norm": 0.46118637919425964,
+      "learning_rate": 0.0002,
+      "loss": 1.2108,
+      "step": 80
+    },
+    {
+      "epoch": 0.066006600660066,
+      "grad_norm": 0.4468648135662079,
+      "learning_rate": 0.0002,
+      "loss": 1.3441,
+      "step": 90
+    },
+    {
+      "epoch": 0.07334066740007333,
+      "grad_norm": 0.46123769879341125,
+      "learning_rate": 0.0002,
+      "loss": 1.1863,
+      "step": 100
+    },
+    {
+      "epoch": 0.08067473414008068,
+      "grad_norm": 0.4859139025211334,
+      "learning_rate": 0.0002,
+      "loss": 1.2772,
+      "step": 110
+    },
+    {
+      "epoch": 0.08800880088008801,
+      "grad_norm": 0.4384922385215759,
+      "learning_rate": 0.0002,
+      "loss": 1.2087,
+      "step": 120
+    },
+    {
+      "epoch": 0.09534286762009535,
+      "grad_norm": 0.39519360661506653,
+      "learning_rate": 0.0002,
+      "loss": 1.2927,
+      "step": 130
+    },
+    {
+      "epoch": 0.10267693436010268,
+      "grad_norm": 0.4049859344959259,
+      "learning_rate": 0.0002,
+      "loss": 1.2349,
+      "step": 140
+    },
+    {
+      "epoch": 0.11001100110011001,
+      "grad_norm": 0.4605638086795807,
+      "learning_rate": 0.0002,
+      "loss": 1.293,
+      "step": 150
+    },
+    {
+      "epoch": 0.11734506784011735,
+      "grad_norm": 0.4201928377151489,
+      "learning_rate": 0.0002,
+      "loss": 1.2659,
+      "step": 160
+    },
+    {
+      "epoch": 0.12467913458012468,
+      "grad_norm": 0.5367777347564697,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 170
+    },
+    {
+      "epoch": 0.132013201320132,
+      "grad_norm": 0.41752299666404724,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 180
+    },
+    {
+      "epoch": 0.13934726806013933,
+      "grad_norm": 0.31597763299942017,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 190
+    },
+    {
+      "epoch": 0.14668133480014667,
+      "grad_norm": 0.7468788623809814,
+      "learning_rate": 0.0002,
+      "loss": 1.2441,
+      "step": 200
+    },
+    {
+      "epoch": 0.15401540154015403,
+      "grad_norm": 0.3403034508228302,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 210
+    },
+    {
+      "epoch": 0.16134946828016136,
+      "grad_norm": 0.34240293502807617,
+      "learning_rate": 0.0002,
+      "loss": 1.2439,
+      "step": 220
+    },
+    {
+      "epoch": 0.1686835350201687,
+      "grad_norm": 0.356158971786499,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 230
+    },
+    {
+      "epoch": 0.17601760176017603,
+      "grad_norm": 0.3448857367038727,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 240
+    },
+    {
+      "epoch": 0.18335166850018336,
+      "grad_norm": 0.3475699722766876,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 250
+    },
+    {
+      "epoch": 0.1906857352401907,
+      "grad_norm": 0.2770358622074127,
+      "learning_rate": 0.0002,
+      "loss": 1.1551,
+      "step": 260
+    },
+    {
+      "epoch": 0.19801980198019803,
+      "grad_norm": 0.4310270845890045,
+      "learning_rate": 0.0002,
+      "loss": 1.2238,
+      "step": 270
+    },
+    {
+      "epoch": 0.20535386872020536,
+      "grad_norm": 0.335041880607605,
+      "learning_rate": 0.0002,
+      "loss": 1.2917,
+      "step": 280
+    },
+    {
+      "epoch": 0.2126879354602127,
+      "grad_norm": 0.3420602083206177,
+      "learning_rate": 0.0002,
+      "loss": 1.0959,
+      "step": 290
+    },
+    {
+      "epoch": 0.22002200220022003,
+      "grad_norm": 0.325001060962677,
+      "learning_rate": 0.0002,
+      "loss": 1.1232,
+      "step": 300
+    },
+    {
+      "epoch": 0.22735606894022736,
+      "grad_norm": 0.3027827739715576,
+      "learning_rate": 0.0002,
+      "loss": 1.2007,
+      "step": 310
+    },
+    {
+      "epoch": 0.2346901356802347,
+      "grad_norm": 0.435550719499588,
+      "learning_rate": 0.0002,
+      "loss": 1.1803,
+      "step": 320
+    },
+    {
+      "epoch": 0.24202420242024203,
+      "grad_norm": 0.3884522616863251,
+      "learning_rate": 0.0002,
+      "loss": 1.2045,
+      "step": 330
+    },
+    {
+      "epoch": 0.24935826916024936,
+      "grad_norm": 0.7736002206802368,
+      "learning_rate": 0.0002,
+      "loss": 1.2481,
+      "step": 340
+    },
+    {
+      "epoch": 0.2566923359002567,
+      "grad_norm": 0.35052821040153503,
+      "learning_rate": 0.0002,
+      "loss": 1.3606,
+      "step": 350
+    },
+    {
+      "epoch": 0.264026402640264,
+      "grad_norm": 0.3311890959739685,
+      "learning_rate": 0.0002,
+      "loss": 1.2129,
+      "step": 360
+    },
+    {
+      "epoch": 0.27136046938027136,
+      "grad_norm": 0.7473500370979309,
+      "learning_rate": 0.0002,
+      "loss": 1.2219,
+      "step": 370
+    },
+    {
+      "epoch": 0.27869453612027867,
+      "grad_norm": 0.3681875765323639,
+      "learning_rate": 0.0002,
+      "loss": 1.2712,
+      "step": 380
+    },
+    {
+      "epoch": 0.28602860286028603,
+      "grad_norm": 0.3764737844467163,
+      "learning_rate": 0.0002,
+      "loss": 1.2258,
+      "step": 390
+    },
+    {
+      "epoch": 0.29336266960029334,
+      "grad_norm": 0.4243989586830139,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 400
+    },
+    {
+      "epoch": 0.3006967363403007,
+      "grad_norm": 0.2658531963825226,
+      "learning_rate": 0.0002,
+      "loss": 1.199,
+      "step": 410
+    },
+    {
+      "epoch": 0.30803080308030806,
+      "grad_norm": 0.3436793386936188,
+      "learning_rate": 0.0002,
+      "loss": 1.1622,
+      "step": 420
+    },
+    {
+      "epoch": 0.31536486982031536,
+      "grad_norm": 0.5101129412651062,
+      "learning_rate": 0.0002,
+      "loss": 1.2953,
+      "step": 430
+    },
+    {
+      "epoch": 0.3226989365603227,
+      "grad_norm": 0.3319750726222992,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 440
+    },
+    {
+      "epoch": 0.33003300330033003,
+      "grad_norm": 0.385148286819458,
+      "learning_rate": 0.0002,
+      "loss": 1.1804,
+      "step": 450
+    },
+    {
+      "epoch": 0.3373670700403374,
+      "grad_norm": 0.3477935791015625,
+      "learning_rate": 0.0002,
+      "loss": 1.1808,
+      "step": 460
+    },
+    {
+      "epoch": 0.3447011367803447,
+      "grad_norm": 0.29748716950416565,
+      "learning_rate": 0.0002,
+      "loss": 1.1877,
+      "step": 470
+    },
+    {
+      "epoch": 0.35203520352035206,
+      "grad_norm": 0.34083324670791626,
+      "learning_rate": 0.0002,
+      "loss": 1.19,
+      "step": 480
+    },
+    {
+      "epoch": 0.35936927026035936,
+      "grad_norm": 0.36904552578926086,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 490
+    },
+    {
+      "epoch": 0.3667033370003667,
+      "grad_norm": 0.315483033657074,
+      "learning_rate": 0.0002,
+      "loss": 1.2223,
+      "step": 500
+    },
+    {
+      "epoch": 0.37403740374037403,
+      "grad_norm": 0.44897955656051636,
+      "learning_rate": 0.0002,
+      "loss": 1.1461,
+      "step": 510
+    },
+    {
+      "epoch": 0.3813714704803814,
+      "grad_norm": 0.3160701394081116,
+      "learning_rate": 0.0002,
+      "loss": 1.3035,
+      "step": 520
+    },
+    {
+      "epoch": 0.3887055372203887,
+      "grad_norm": 0.29584741592407227,
+      "learning_rate": 0.0002,
+      "loss": 1.3197,
+      "step": 530
+    },
+    {
+      "epoch": 0.39603960396039606,
+      "grad_norm": 0.5430002808570862,
+      "learning_rate": 0.0002,
+      "loss": 1.2983,
+      "step": 540
+    },
+    {
+      "epoch": 0.40337367070040336,
+      "grad_norm": 0.2908070683479309,
+      "learning_rate": 0.0002,
+      "loss": 1.2459,
+      "step": 550
+    },
+    {
+      "epoch": 0.4107077374404107,
+      "grad_norm": 0.35066530108451843,
+      "learning_rate": 0.0002,
+      "loss": 1.2384,
+      "step": 560
+    },
+    {
+      "epoch": 0.41804180418041803,
+      "grad_norm": 0.37588003277778625,
+      "learning_rate": 0.0002,
+      "loss": 1.1784,
+      "step": 570
+    },
+    {
+      "epoch": 0.4253758709204254,
+      "grad_norm": 0.3112126886844635,
+      "learning_rate": 0.0002,
+      "loss": 1.2334,
+      "step": 580
+    },
+    {
+      "epoch": 0.4327099376604327,
+      "grad_norm": 0.35577139258384705,
+      "learning_rate": 0.0002,
+      "loss": 1.1439,
+      "step": 590
+    },
+    {
+      "epoch": 0.44004400440044006,
+      "grad_norm": 0.31706422567367554,
+      "learning_rate": 0.0002,
+      "loss": 1.184,
+      "step": 600
+    },
+    {
+      "epoch": 0.44737807114044736,
+      "grad_norm": 0.3249092102050781,
+      "learning_rate": 0.0002,
+      "loss": 1.2081,
+      "step": 610
+    },
+    {
+      "epoch": 0.4547121378804547,
+      "grad_norm": 0.3842705488204956,
+      "learning_rate": 0.0002,
+      "loss": 1.0824,
+      "step": 620
+    },
+    {
+      "epoch": 0.46204620462046203,
+      "grad_norm": 0.390991747379303,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 630
+    },
+    {
+      "epoch": 0.4693802713604694,
+      "grad_norm": 0.27532413601875305,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 640
+    },
+    {
+      "epoch": 0.4767143381004767,
+      "grad_norm": 0.31412816047668457,
+      "learning_rate": 0.0002,
+      "loss": 1.1058,
+      "step": 650
+    },
+    {
+      "epoch": 0.48404840484048406,
+      "grad_norm": 0.32117101550102234,
+      "learning_rate": 0.0002,
+      "loss": 1.1312,
+      "step": 660
+    },
+    {
+      "epoch": 0.49138247158049136,
+      "grad_norm": 0.3810010254383087,
+      "learning_rate": 0.0002,
+      "loss": 1.2423,
+      "step": 670
+    },
+    {
+      "epoch": 0.4987165383204987,
+      "grad_norm": 0.36289164423942566,
+      "learning_rate": 0.0002,
+      "loss": 1.1978,
+      "step": 680
+    },
+    {
+      "epoch": 0.506050605060506,
+      "grad_norm": 0.34458720684051514,
+      "learning_rate": 0.0002,
+      "loss": 1.2034,
+      "step": 690
+    },
+    {
+      "epoch": 0.5133846718005134,
+      "grad_norm": 0.32844600081443787,
+      "learning_rate": 0.0002,
+      "loss": 1.1756,
+      "step": 700
+    },
+    {
+      "epoch": 0.5207187385405208,
+      "grad_norm": 0.3144175708293915,
+      "learning_rate": 0.0002,
+      "loss": 1.0807,
+      "step": 710
+    },
+    {
+      "epoch": 0.528052805280528,
+      "grad_norm": 0.3898887634277344,
+      "learning_rate": 0.0002,
+      "loss": 1.1952,
+      "step": 720
+    },
+    {
+      "epoch": 0.5353868720205354,
+      "grad_norm": 1.3220758438110352,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 730
+    },
+    {
+      "epoch": 0.5427209387605427,
+      "grad_norm": 0.3635874390602112,
+      "learning_rate": 0.0002,
+      "loss": 1.227,
+      "step": 740
+    },
+    {
+      "epoch": 0.5500550055005501,
+      "grad_norm": 0.3138217628002167,
+      "learning_rate": 0.0002,
+      "loss": 1.2169,
+      "step": 750
+    },
+    {
+      "epoch": 0.5573890722405573,
+      "grad_norm": 0.4063207805156708,
+      "learning_rate": 0.0002,
+      "loss": 1.1516,
+      "step": 760
+    },
+    {
+      "epoch": 0.5647231389805647,
+      "grad_norm": 0.3926219940185547,
+      "learning_rate": 0.0002,
+      "loss": 1.1954,
+      "step": 770
+    },
+    {
+      "epoch": 0.5720572057205721,
+      "grad_norm": 0.31954652070999146,
+      "learning_rate": 0.0002,
+      "loss": 1.1726,
+      "step": 780
+    },
+    {
+      "epoch": 0.5793912724605794,
+      "grad_norm": 0.4248711168766022,
+      "learning_rate": 0.0002,
+      "loss": 1.2977,
+      "step": 790
+    },
+    {
+      "epoch": 0.5867253392005867,
+      "grad_norm": 0.643004834651947,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 800
+    },
+    {
+      "epoch": 0.594059405940594,
+      "grad_norm": 0.3479592800140381,
+      "learning_rate": 0.0002,
+      "loss": 1.1793,
+      "step": 810
+    },
+    {
+      "epoch": 0.6013934726806014,
+      "grad_norm": 0.4684754014015198,
+      "learning_rate": 0.0002,
+      "loss": 1.2426,
+      "step": 820
+    },
+    {
+      "epoch": 0.6087275394206088,
+      "grad_norm": 0.3739790916442871,
+      "learning_rate": 0.0002,
+      "loss": 1.2002,
+      "step": 830
+    },
+    {
+      "epoch": 0.6160616061606161,
+      "grad_norm": 0.40884748101234436,
+      "learning_rate": 0.0002,
+      "loss": 1.2139,
+      "step": 840
+    },
+    {
+      "epoch": 0.6233956729006234,
+      "grad_norm": 0.9722164273262024,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 850
+    },
+    {
+      "epoch": 0.6307297396406307,
+      "grad_norm": 0.42992347478866577,
+      "learning_rate": 0.0002,
+      "loss": 1.3069,
+      "step": 860
+    },
+    {
+      "epoch": 0.6380638063806381,
+      "grad_norm": 0.36654195189476013,
+      "learning_rate": 0.0002,
+      "loss": 1.1339,
+      "step": 870
+    },
+    {
+      "epoch": 0.6453978731206454,
+      "grad_norm": 0.4113832116127014,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 880
+    },
+    {
+      "epoch": 0.6527319398606527,
+      "grad_norm": 0.2948838770389557,
+      "learning_rate": 0.0002,
+      "loss": 1.2163,
+      "step": 890
+    },
+    {
+      "epoch": 0.6600660066006601,
+      "grad_norm": 0.38330280780792236,
+      "learning_rate": 0.0002,
+      "loss": 1.1081,
+      "step": 900
+    },
+    {
+      "epoch": 0.6674000733406674,
+      "grad_norm": 0.4428867697715759,
+      "learning_rate": 0.0002,
+      "loss": 1.1342,
+      "step": 910
+    },
+    {
+      "epoch": 0.6747341400806748,
+      "grad_norm": 0.23659265041351318,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 920
+    },
+    {
+      "epoch": 0.682068206820682,
+      "grad_norm": 0.323685884475708,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 930
+    },
+    {
+      "epoch": 0.6894022735606894,
+      "grad_norm": 0.39157727360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.0853,
+      "step": 940
+    },
+    {
+      "epoch": 0.6967363403006968,
+      "grad_norm": 0.27189481258392334,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 950
+    },
+    {
+      "epoch": 0.7040704070407041,
+      "grad_norm": 0.529883861541748,
+      "learning_rate": 0.0002,
+      "loss": 1.1033,
+      "step": 960
+    },
+    {
+      "epoch": 0.7114044737807114,
+      "grad_norm": 0.34758689999580383,
+      "learning_rate": 0.0002,
+      "loss": 1.139,
+      "step": 970
+    },
+    {
+      "epoch": 0.7187385405207187,
+      "grad_norm": 0.831749439239502,
+      "learning_rate": 0.0002,
+      "loss": 1.2197,
+      "step": 980
+    },
+    {
+      "epoch": 0.7260726072607261,
+      "grad_norm": 0.4438304007053375,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 990
+    },
+    {
+      "epoch": 0.7334066740007334,
+      "grad_norm": 0.33840006589889526,
+      "learning_rate": 0.0002,
+      "loss": 1.1021,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7407407407407407,
+      "grad_norm": 0.3454797863960266,
+      "learning_rate": 0.0002,
+      "loss": 1.254,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7480748074807481,
+      "grad_norm": 0.38999441266059875,
+      "learning_rate": 0.0002,
+      "loss": 1.106,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7554088742207554,
+      "grad_norm": 0.2829911708831787,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7627429409607628,
+      "grad_norm": 0.36918163299560547,
+      "learning_rate": 0.0002,
+      "loss": 1.2123,
+      "step": 1040
+    },
+    {
+      "epoch": 0.77007700770077,
+      "grad_norm": 0.3415680229663849,
+      "learning_rate": 0.0002,
+      "loss": 1.3028,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7774110744407774,
+      "grad_norm": 0.2974182963371277,
+      "learning_rate": 0.0002,
+      "loss": 1.1939,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7847451411807848,
+      "grad_norm": 0.3880919814109802,
+      "learning_rate": 0.0002,
+      "loss": 1.194,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7920792079207921,
+      "grad_norm": 0.33503302931785583,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7994132746607994,
+      "grad_norm": 0.3728407025337219,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8067473414008067,
+      "grad_norm": 0.3509373664855957,
+      "learning_rate": 0.0002,
+      "loss": 1.0835,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8140814081408141,
+      "grad_norm": 0.42228564620018005,
+      "learning_rate": 0.0002,
+      "loss": 1.2661,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8214154748808215,
+      "grad_norm": 0.313467800617218,
+      "learning_rate": 0.0002,
+      "loss": 1.1788,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8287495416208287,
+      "grad_norm": 0.3378850817680359,
+      "learning_rate": 0.0002,
+      "loss": 1.1971,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8360836083608361,
+      "grad_norm": 0.43200382590293884,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8434176751008434,
+      "grad_norm": 0.3309599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.3203,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8507517418408508,
+      "grad_norm": 0.3526846170425415,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1160
+    },
+    {
+      "epoch": 0.858085808580858,
+      "grad_norm": 1.2722247838974,
+      "learning_rate": 0.0002,
+      "loss": 1.0851,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8654198753208654,
+      "grad_norm": 0.34142059087753296,
+      "learning_rate": 0.0002,
+      "loss": 1.0785,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8727539420608728,
+      "grad_norm": 0.3805823028087616,
+      "learning_rate": 0.0002,
+      "loss": 1.2187,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8800880088008801,
+      "grad_norm": 0.3931232690811157,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8874220755408874,
+      "grad_norm": 0.2937372624874115,
+      "learning_rate": 0.0002,
+      "loss": 1.0948,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8947561422808947,
+      "grad_norm": 0.3757196366786957,
+      "learning_rate": 0.0002,
+      "loss": 1.1228,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9020902090209021,
+      "grad_norm": 0.3502705991268158,
+      "learning_rate": 0.0002,
+      "loss": 1.1222,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9094242757609095,
+      "grad_norm": 0.32758915424346924,
+      "learning_rate": 0.0002,
+      "loss": 1.2242,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9167583425009168,
+      "grad_norm": 0.37199416756629944,
+      "learning_rate": 0.0002,
+      "loss": 1.215,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9240924092409241,
+      "grad_norm": 0.3551490604877472,
+      "learning_rate": 0.0002,
+      "loss": 1.1225,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9314264759809314,
+      "grad_norm": 0.2859550714492798,
+      "learning_rate": 0.0002,
+      "loss": 1.1966,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9387605427209388,
+      "grad_norm": 0.427990585565567,
+      "learning_rate": 0.0002,
+      "loss": 1.2186,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9460946094609461,
+      "grad_norm": 0.33717992901802063,
+      "learning_rate": 0.0002,
+      "loss": 1.2848,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9534286762009534,
+      "grad_norm": 0.30225634574890137,
+      "learning_rate": 0.0002,
+      "loss": 1.1656,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9607627429409608,
+      "grad_norm": 0.385821133852005,
+      "learning_rate": 0.0002,
+      "loss": 1.2404,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9680968096809681,
+      "grad_norm": 0.35278066992759705,
+      "learning_rate": 0.0002,
+      "loss": 1.1932,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9754308764209755,
+      "grad_norm": 0.49987098574638367,
+      "learning_rate": 0.0002,
+      "loss": 1.1071,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9827649431609827,
+      "grad_norm": 0.3842747211456299,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9900990099009901,
+      "grad_norm": 0.6274653673171997,
+      "learning_rate": 0.0002,
+      "loss": 1.0862,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9974330766409975,
+      "grad_norm": 0.5239808559417725,
+      "learning_rate": 0.0002,
+      "loss": 1.124,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9996332966629996,
+      "eval_loss": 1.1822267770767212,
+      "eval_runtime": 32.7389,
+      "eval_samples_per_second": 13.165,
+      "eval_steps_per_second": 1.649,
+      "step": 1363
+    },
+    {
+      "epoch": 1.0047671433810048,
+      "grad_norm": 0.45311301946640015,
+      "learning_rate": 0.0002,
+      "loss": 1.096,
+      "step": 1370
+    },
+    {
+      "epoch": 1.012101210121012,
+      "grad_norm": 0.29685574769973755,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0194352768610195,
+      "grad_norm": 0.3290937840938568,
+      "learning_rate": 0.0002,
+      "loss": 1.0302,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0267693436010268,
+      "grad_norm": 0.3801758587360382,
+      "learning_rate": 0.0002,
+      "loss": 1.0295,
+      "step": 1400
+    },
+    {
+      "epoch": 1.034103410341034,
+      "grad_norm": 0.794174313545227,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0414374770810415,
+      "grad_norm": 0.3854154646396637,
+      "learning_rate": 0.0002,
+      "loss": 1.2232,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0487715438210488,
+      "grad_norm": 0.32702451944351196,
+      "learning_rate": 0.0002,
+      "loss": 1.0652,
+      "step": 1430
+    },
+    {
+      "epoch": 1.056105610561056,
+      "grad_norm": 0.7815203666687012,
+      "learning_rate": 0.0002,
+      "loss": 1.1144,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0634396773010635,
+      "grad_norm": 0.3087436854839325,
+      "learning_rate": 0.0002,
+      "loss": 1.1316,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0707737440410707,
+      "grad_norm": 0.3847602903842926,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0781078107810782,
+      "grad_norm": 0.3693031370639801,
+      "learning_rate": 0.0002,
+      "loss": 1.1428,
+      "step": 1470
+    },
+    {
+      "epoch": 1.0854418775210855,
+      "grad_norm": 0.4111202359199524,
+      "learning_rate": 0.0002,
+      "loss": 1.0995,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0927759442610927,
+      "grad_norm": 0.41452381014823914,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1001100110011002,
+      "grad_norm": 0.3336445093154907,
+      "learning_rate": 0.0002,
+      "loss": 1.1068,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1074440777411074,
+      "grad_norm": 0.3923407793045044,
+      "learning_rate": 0.0002,
+      "loss": 1.0556,
+      "step": 1510
+    },
+    {
+      "epoch": 1.1147781444811147,
+      "grad_norm": 0.46215683221817017,
+      "learning_rate": 0.0002,
+      "loss": 1.1644,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1221122112211221,
+      "grad_norm": 0.3592156767845154,
+      "learning_rate": 0.0002,
+      "loss": 1.1133,
+      "step": 1530
+    },
+    {
+      "epoch": 1.1294462779611294,
+      "grad_norm": 0.361110657453537,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1367803447011369,
+      "grad_norm": 0.5317131280899048,
+      "learning_rate": 0.0002,
+      "loss": 1.1553,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1441144114411441,
+      "grad_norm": 0.3882388174533844,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1514484781811514,
+      "grad_norm": 0.3259428143501282,
+      "learning_rate": 0.0002,
+      "loss": 1.0805,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1587825449211588,
+      "grad_norm": 0.410935640335083,
+      "learning_rate": 0.0002,
+      "loss": 1.1819,
+      "step": 1580
+    },
+    {
+      "epoch": 1.166116611661166,
+      "grad_norm": 0.44940185546875,
+      "learning_rate": 0.0002,
+      "loss": 1.1143,
+      "step": 1590
+    },
+    {
+      "epoch": 1.1734506784011733,
+      "grad_norm": 0.5106484293937683,
+      "learning_rate": 0.0002,
+      "loss": 1.0334,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1807847451411808,
+      "grad_norm": 0.6603665947914124,
+      "learning_rate": 0.0002,
+      "loss": 1.2376,
+      "step": 1610
+    },
+    {
+      "epoch": 1.188118811881188,
+      "grad_norm": 0.4799964129924774,
+      "learning_rate": 0.0002,
+      "loss": 1.1227,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1954528786211955,
+      "grad_norm": 0.4389883279800415,
+      "learning_rate": 0.0002,
+      "loss": 1.1191,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2027869453612028,
+      "grad_norm": 0.4188813269138336,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 1640
+    },
+    {
+      "epoch": 1.21012101210121,
+      "grad_norm": 0.7132157683372498,
+      "learning_rate": 0.0002,
+      "loss": 1.0605,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2174550788412175,
+      "grad_norm": 0.507480263710022,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2247891455812248,
+      "grad_norm": 0.9452332854270935,
+      "learning_rate": 0.0002,
+      "loss": 0.9948,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2321232123212322,
+      "grad_norm": 0.4121614992618561,
+      "learning_rate": 0.0002,
+      "loss": 1.0228,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2394572790612395,
+      "grad_norm": 0.34230247139930725,
+      "learning_rate": 0.0002,
+      "loss": 1.0366,
+      "step": 1690
+    },
+    {
+      "epoch": 1.2467913458012467,
+      "grad_norm": 0.4026208817958832,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2541254125412542,
+      "grad_norm": 0.46673697233200073,
+      "learning_rate": 0.0002,
+      "loss": 1.0206,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2614594792812615,
+      "grad_norm": 0.38349825143814087,
+      "learning_rate": 0.0002,
+      "loss": 1.0827,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2687935460212687,
+      "grad_norm": 0.4049997627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.0356,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2761276127612762,
+      "grad_norm": 0.3417615294456482,
+      "learning_rate": 0.0002,
+      "loss": 0.9504,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2834616795012834,
+      "grad_norm": 0.4277614951133728,
+      "learning_rate": 0.0002,
+      "loss": 1.094,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2907957462412907,
+      "grad_norm": 0.5864202976226807,
+      "learning_rate": 0.0002,
+      "loss": 0.9938,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2981298129812981,
+      "grad_norm": 0.7097493410110474,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3054638797213054,
+      "grad_norm": 0.3145381212234497,
+      "learning_rate": 0.0002,
+      "loss": 1.1132,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3127979464613129,
+      "grad_norm": 0.5116165280342102,
+      "learning_rate": 0.0002,
+      "loss": 1.1099,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3201320132013201,
+      "grad_norm": 0.7469736337661743,
+      "learning_rate": 0.0002,
+      "loss": 1.0765,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3274660799413276,
+      "grad_norm": 0.32272255420684814,
+      "learning_rate": 0.0002,
+      "loss": 1.0663,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3348001466813348,
+      "grad_norm": 0.3534623086452484,
+      "learning_rate": 0.0002,
+      "loss": 0.9887,
+      "step": 1820
+    },
+    {
+      "epoch": 1.342134213421342,
+      "grad_norm": 0.36127907037734985,
+      "learning_rate": 0.0002,
+      "loss": 1.1628,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3494682801613496,
+      "grad_norm": 0.4072401523590088,
+      "learning_rate": 0.0002,
+      "loss": 1.0972,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3568023469013568,
+      "grad_norm": 0.3769161105155945,
+      "learning_rate": 0.0002,
+      "loss": 1.1267,
+      "step": 1850
+    },
+    {
+      "epoch": 1.364136413641364,
+      "grad_norm": 0.412883460521698,
+      "learning_rate": 0.0002,
+      "loss": 1.0173,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3714704803813715,
+      "grad_norm": 0.3735875189304352,
+      "learning_rate": 0.0002,
+      "loss": 1.0265,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3788045471213788,
+      "grad_norm": 0.39158159494400024,
+      "learning_rate": 0.0002,
+      "loss": 1.1061,
+      "step": 1880
+    },
+    {
+      "epoch": 1.386138613861386,
+      "grad_norm": 0.44431769847869873,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3934726806013935,
+      "grad_norm": 0.37772801518440247,
+      "learning_rate": 0.0002,
+      "loss": 1.0216,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4008067473414008,
+      "grad_norm": 0.4056641757488251,
+      "learning_rate": 0.0002,
+      "loss": 1.0674,
+      "step": 1910
+    },
+    {
+      "epoch": 1.408140814081408,
+      "grad_norm": 0.41612377762794495,
+      "learning_rate": 0.0002,
+      "loss": 1.0256,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4154748808214155,
+      "grad_norm": 0.41153013706207275,
+      "learning_rate": 0.0002,
+      "loss": 1.0467,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4228089475614227,
+      "grad_norm": 0.387845516204834,
+      "learning_rate": 0.0002,
+      "loss": 1.1062,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4301430143014302,
+      "grad_norm": 0.3809587061405182,
+      "learning_rate": 0.0002,
+      "loss": 1.1094,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4374770810414375,
+      "grad_norm": 0.3625726103782654,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1960
+    },
+    {
+      "epoch": 1.444811147781445,
+      "grad_norm": 0.5294290781021118,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4521452145214522,
+      "grad_norm": 0.39975494146347046,
+      "learning_rate": 0.0002,
+      "loss": 1.1114,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4594792812614594,
+      "grad_norm": 0.4181167185306549,
+      "learning_rate": 0.0002,
+      "loss": 0.9704,
+      "step": 1990
+    },
+    {
+      "epoch": 1.466813348001467,
+      "grad_norm": 0.42001503705978394,
+      "learning_rate": 0.0002,
+      "loss": 1.1146,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4741474147414741,
+      "grad_norm": 0.4877578616142273,
+      "learning_rate": 0.0002,
+      "loss": 1.1266,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4814814814814814,
+      "grad_norm": 0.4050969183444977,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4888155482214889,
+      "grad_norm": 0.39068883657455444,
+      "learning_rate": 0.0002,
+      "loss": 1.0562,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4961496149614961,
+      "grad_norm": 0.421282559633255,
+      "learning_rate": 0.0002,
+      "loss": 1.0464,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5034836817015034,
+      "grad_norm": 0.47092297673225403,
+      "learning_rate": 0.0002,
+      "loss": 1.0532,
+      "step": 2050
+    },
+    {
+      "epoch": 1.5108177484415108,
+      "grad_norm": 0.39688974618911743,
+      "learning_rate": 0.0002,
+      "loss": 0.9348,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5181518151815183,
+      "grad_norm": 0.5529879331588745,
+      "learning_rate": 0.0002,
+      "loss": 1.08,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5254858819215253,
+      "grad_norm": 0.4879782199859619,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5328199486615328,
+      "grad_norm": 0.5517361164093018,
+      "learning_rate": 0.0002,
+      "loss": 1.0432,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5401540154015403,
+      "grad_norm": 0.44015637040138245,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5474880821415475,
+      "grad_norm": 0.5435167551040649,
+      "learning_rate": 0.0002,
+      "loss": 1.1873,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5548221488815548,
+      "grad_norm": 0.5714033246040344,
+      "learning_rate": 0.0002,
+      "loss": 1.1076,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5621562156215623,
+      "grad_norm": 0.31732529401779175,
+      "learning_rate": 0.0002,
+      "loss": 1.1107,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5694902823615695,
+      "grad_norm": 0.49068278074264526,
+      "learning_rate": 0.0002,
+      "loss": 1.0817,
+      "step": 2140
+    },
+    {
+      "epoch": 1.5768243491015768,
+      "grad_norm": 0.46851542592048645,
+      "learning_rate": 0.0002,
+      "loss": 1.0254,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5841584158415842,
+      "grad_norm": 0.5083092451095581,
+      "learning_rate": 0.0002,
+      "loss": 1.0623,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5914924825815915,
+      "grad_norm": 0.9822936058044434,
+      "learning_rate": 0.0002,
+      "loss": 1.0603,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5988265493215987,
+      "grad_norm": 0.4575989246368408,
+      "learning_rate": 0.0002,
+      "loss": 0.9986,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6061606160616062,
+      "grad_norm": 0.47444286942481995,
+      "learning_rate": 0.0002,
+      "loss": 1.1292,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6134946828016135,
+      "grad_norm": 0.7208226919174194,
+      "learning_rate": 0.0002,
+      "loss": 1.0136,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6208287495416207,
+      "grad_norm": 0.43791481852531433,
+      "learning_rate": 0.0002,
+      "loss": 1.15,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6281628162816282,
+      "grad_norm": 0.5245792865753174,
+      "learning_rate": 0.0002,
+      "loss": 1.0961,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6354968830216357,
+      "grad_norm": 0.39289429783821106,
+      "learning_rate": 0.0002,
+      "loss": 0.9957,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6428309497616427,
+      "grad_norm": 0.6106135845184326,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6501650165016502,
+      "grad_norm": 0.3722580671310425,
+      "learning_rate": 0.0002,
+      "loss": 1.0129,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6574990832416576,
+      "grad_norm": 0.3649403750896454,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6648331499816649,
+      "grad_norm": 0.46514248847961426,
+      "learning_rate": 0.0002,
+      "loss": 1.0037,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6721672167216721,
+      "grad_norm": 0.42034927010536194,
+      "learning_rate": 0.0002,
+      "loss": 1.0022,
+      "step": 2280
+    },
+    {
+      "epoch": 1.6795012834616796,
+      "grad_norm": 0.45202910900115967,
+      "learning_rate": 0.0002,
+      "loss": 1.1362,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6868353502016868,
+      "grad_norm": 0.36257603764533997,
+      "learning_rate": 0.0002,
+      "loss": 1.0866,
+      "step": 2300
+    },
+    {
+      "epoch": 1.694169416941694,
+      "grad_norm": 0.6340323090553284,
+      "learning_rate": 0.0002,
+      "loss": 1.0973,
+      "step": 2310
+    },
+    {
+      "epoch": 1.7015034836817016,
+      "grad_norm": 0.4352878928184509,
+      "learning_rate": 0.0002,
+      "loss": 1.0615,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7088375504217088,
+      "grad_norm": 0.45029792189598083,
+      "learning_rate": 0.0002,
+      "loss": 1.0629,
+      "step": 2330
+    },
+    {
+      "epoch": 1.716171617161716,
+      "grad_norm": 0.3891315758228302,
+      "learning_rate": 0.0002,
+      "loss": 0.9621,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7235056839017235,
+      "grad_norm": 0.35180050134658813,
+      "learning_rate": 0.0002,
+      "loss": 0.9779,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7308397506417308,
+      "grad_norm": 0.42367449402809143,
+      "learning_rate": 0.0002,
+      "loss": 1.0368,
+      "step": 2360
+    },
+    {
+      "epoch": 1.738173817381738,
+      "grad_norm": 0.4553675353527069,
+      "learning_rate": 0.0002,
+      "loss": 1.0376,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7455078841217455,
+      "grad_norm": 0.5944654941558838,
+      "learning_rate": 0.0002,
+      "loss": 1.1467,
+      "step": 2380
+    },
+    {
+      "epoch": 1.752841950861753,
+      "grad_norm": 0.3479664623737335,
+      "learning_rate": 0.0002,
+      "loss": 1.0548,
+      "step": 2390
+    },
+    {
+      "epoch": 1.76017601760176,
+      "grad_norm": 0.3585502505302429,
+      "learning_rate": 0.0002,
+      "loss": 1.0798,
+      "step": 2400
+    },
+    {
+      "epoch": 1.7675100843417675,
+      "grad_norm": 0.4263346493244171,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 2410
+    },
+    {
+      "epoch": 1.774844151081775,
+      "grad_norm": 0.5476409196853638,
+      "learning_rate": 0.0002,
+      "loss": 1.054,
+      "step": 2420
+    },
+    {
+      "epoch": 1.7821782178217822,
+      "grad_norm": 0.3694186508655548,
+      "learning_rate": 0.0002,
+      "loss": 1.1615,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7895122845617895,
+      "grad_norm": 0.9185658693313599,
+      "learning_rate": 0.0002,
+      "loss": 1.1343,
+      "step": 2440
+    },
+    {
+      "epoch": 1.796846351301797,
+      "grad_norm": 0.7171908020973206,
+      "learning_rate": 0.0002,
+      "loss": 1.0764,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8041804180418042,
+      "grad_norm": 0.550658643245697,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 2460
+    },
+    {
+      "epoch": 1.8115144847818114,
+      "grad_norm": 0.4075568914413452,
+      "learning_rate": 0.0002,
+      "loss": 0.9975,
+      "step": 2470
+    },
+    {
+      "epoch": 1.818848551521819,
+      "grad_norm": 0.3790127635002136,
+      "learning_rate": 0.0002,
+      "loss": 1.0935,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8261826182618262,
+      "grad_norm": 0.3576384484767914,
+      "learning_rate": 0.0002,
+      "loss": 0.9839,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8335166850018334,
+      "grad_norm": 0.3919370770454407,
+      "learning_rate": 0.0002,
+      "loss": 1.1369,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8408507517418409,
+      "grad_norm": 0.485083669424057,
+      "learning_rate": 0.0002,
+      "loss": 0.9985,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8481848184818483,
+      "grad_norm": 0.4564347565174103,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8555188852218554,
+      "grad_norm": 0.3613106608390808,
+      "learning_rate": 0.0002,
+      "loss": 1.0944,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8628529519618628,
+      "grad_norm": 0.39600759744644165,
+      "learning_rate": 0.0002,
+      "loss": 1.0819,
+      "step": 2540
+    },
+    {
+      "epoch": 1.8701870187018703,
+      "grad_norm": 1.123499870300293,
+      "learning_rate": 0.0002,
+      "loss": 0.9453,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8775210854418776,
+      "grad_norm": 0.4612680673599243,
+      "learning_rate": 0.0002,
+      "loss": 1.0635,
+      "step": 2560
+    },
+    {
+      "epoch": 1.8848551521818848,
+      "grad_norm": 0.42745399475097656,
+      "learning_rate": 0.0002,
+      "loss": 1.0087,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8921892189218923,
+      "grad_norm": 0.4055580198764801,
+      "learning_rate": 0.0002,
+      "loss": 1.0102,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8995232856618995,
+      "grad_norm": 0.44174644351005554,
+      "learning_rate": 0.0002,
+      "loss": 1.0177,
+      "step": 2590
+    },
+    {
+      "epoch": 1.9068573524019068,
+      "grad_norm": 1.0228385925292969,
+      "learning_rate": 0.0002,
+      "loss": 0.9886,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9141914191419143,
+      "grad_norm": 0.3496396243572235,
+      "learning_rate": 0.0002,
+      "loss": 1.0857,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9215254858819215,
+      "grad_norm": 0.4191173017024994,
+      "learning_rate": 0.0002,
+      "loss": 1.0955,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9288595526219288,
+      "grad_norm": 0.6778554916381836,
+      "learning_rate": 0.0002,
+      "loss": 1.0943,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9361936193619362,
+      "grad_norm": 0.41992834210395813,
+      "learning_rate": 0.0002,
+      "loss": 1.0594,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9435276861019435,
+      "grad_norm": 0.8760401010513306,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9508617528419507,
+      "grad_norm": 0.44049209356307983,
+      "learning_rate": 0.0002,
+      "loss": 1.0379,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9581958195819582,
+      "grad_norm": 0.5651928782463074,
+      "learning_rate": 0.0002,
+      "loss": 1.1008,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9655298863219657,
+      "grad_norm": 0.5292727947235107,
+      "learning_rate": 0.0002,
+      "loss": 1.1317,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9728639530619727,
+      "grad_norm": 0.6012240648269653,
+      "learning_rate": 0.0002,
+      "loss": 1.1328,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9801980198019802,
+      "grad_norm": 0.3945149779319763,
+      "learning_rate": 0.0002,
+      "loss": 1.0683,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9875320865419877,
+      "grad_norm": 0.5732627511024475,
+      "learning_rate": 0.0002,
+      "loss": 1.0155,
+      "step": 2710
+    },
+    {
+      "epoch": 1.994866153281995,
+      "grad_norm": 0.3963361084461212,
+      "learning_rate": 0.0002,
+      "loss": 0.9857,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1534006595611572,
+      "eval_runtime": 32.7541,
+      "eval_samples_per_second": 13.159,
+      "eval_steps_per_second": 1.649,
+      "step": 2727
+    },
+    {
+      "epoch": 2.002200220022002,
+      "grad_norm": 0.48628315329551697,
+      "learning_rate": 0.0002,
+      "loss": 0.9624,
+      "step": 2730
+    },
+    {
+      "epoch": 2.0095342867620096,
+      "grad_norm": 0.413875013589859,
+      "learning_rate": 0.0002,
+      "loss": 0.9603,
+      "step": 2740
+    },
+    {
+      "epoch": 2.0168683535020167,
+      "grad_norm": 0.4988735616207123,
+      "learning_rate": 0.0002,
+      "loss": 0.965,
+      "step": 2750
+    },
+    {
+      "epoch": 2.024202420242024,
+      "grad_norm": 0.5634812712669373,
+      "learning_rate": 0.0002,
+      "loss": 0.9677,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0315364869820316,
+      "grad_norm": 0.48302653431892395,
+      "learning_rate": 0.0002,
+      "loss": 0.9547,
+      "step": 2770
+    },
+    {
+      "epoch": 2.038870553722039,
+      "grad_norm": 0.49914175271987915,
+      "learning_rate": 0.0002,
+      "loss": 0.9346,
+      "step": 2780
+    },
+    {
+      "epoch": 2.046204620462046,
+      "grad_norm": 1.14039945602417,
+      "learning_rate": 0.0002,
+      "loss": 0.904,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0535386872020536,
+      "grad_norm": 0.6359720826148987,
+      "learning_rate": 0.0002,
+      "loss": 0.9588,
+      "step": 2800
+    },
+    {
+      "epoch": 2.060872753942061,
+      "grad_norm": 0.4589158296585083,
+      "learning_rate": 0.0002,
+      "loss": 0.9031,
+      "step": 2810
+    },
+    {
+      "epoch": 2.068206820682068,
+      "grad_norm": 0.46255481243133545,
+      "learning_rate": 0.0002,
+      "loss": 0.9438,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0755408874220755,
+      "grad_norm": 0.6232137680053711,
+      "learning_rate": 0.0002,
+      "loss": 0.9464,
+      "step": 2830
+    },
+    {
+      "epoch": 2.082874954162083,
+      "grad_norm": 0.41042178869247437,
+      "learning_rate": 0.0002,
+      "loss": 0.8978,
+      "step": 2840
+    },
+    {
+      "epoch": 2.09020902090209,
+      "grad_norm": 0.5334428548812866,
+      "learning_rate": 0.0002,
+      "loss": 0.8516,
+      "step": 2850
+    },
+    {
+      "epoch": 2.0975430876420975,
+      "grad_norm": 0.8270058631896973,
+      "learning_rate": 0.0002,
+      "loss": 0.9313,
+      "step": 2860
+    },
+    {
+      "epoch": 2.104877154382105,
+      "grad_norm": 0.6624533534049988,
+      "learning_rate": 0.0002,
+      "loss": 1.0064,
+      "step": 2870
+    },
+    {
+      "epoch": 2.112211221122112,
+      "grad_norm": 0.5448863506317139,
+      "learning_rate": 0.0002,
+      "loss": 0.9196,
+      "step": 2880
+    },
+    {
+      "epoch": 2.1195452878621195,
+      "grad_norm": 0.621482789516449,
+      "learning_rate": 0.0002,
+      "loss": 0.887,
+      "step": 2890
+    },
+    {
+      "epoch": 2.126879354602127,
+      "grad_norm": 0.4556255340576172,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1342134213421344,
+      "grad_norm": 0.4620579183101654,
+      "learning_rate": 0.0002,
+      "loss": 0.9323,
+      "step": 2910
+    },
+    {
+      "epoch": 2.1415474880821415,
+      "grad_norm": 0.9602415561676025,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 2920
+    },
+    {
+      "epoch": 2.148881554822149,
+      "grad_norm": 0.587943971157074,
+      "learning_rate": 0.0002,
+      "loss": 0.8826,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1562156215621564,
+      "grad_norm": 0.5121372938156128,
+      "learning_rate": 0.0002,
+      "loss": 0.971,
+      "step": 2940
+    },
+    {
+      "epoch": 2.1635496883021634,
+      "grad_norm": 0.49424484372138977,
+      "learning_rate": 0.0002,
+      "loss": 0.8751,
+      "step": 2950
+    },
+    {
+      "epoch": 2.170883755042171,
+      "grad_norm": 0.6312560439109802,
+      "learning_rate": 0.0002,
+      "loss": 0.8674,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1782178217821784,
+      "grad_norm": 0.5235576629638672,
+      "learning_rate": 0.0002,
+      "loss": 0.9791,
+      "step": 2970
+    },
+    {
+      "epoch": 2.1855518885221854,
+      "grad_norm": 0.5868439674377441,
+      "learning_rate": 0.0002,
+      "loss": 0.9706,
+      "step": 2980
+    },
+    {
+      "epoch": 2.192885955262193,
+      "grad_norm": 0.42302873730659485,
+      "learning_rate": 0.0002,
+      "loss": 0.9338,
+      "step": 2990
+    },
+    {
+      "epoch": 2.2002200220022003,
+      "grad_norm": 0.5097725987434387,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 3000
+    },
+    {
+      "epoch": 2.2075540887422074,
+      "grad_norm": 0.5091572403907776,
+      "learning_rate": 0.0002,
+      "loss": 0.9239,
+      "step": 3010
+    },
+    {
+      "epoch": 2.214888155482215,
+      "grad_norm": 0.49433162808418274,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 3020
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.5577368140220642,
+      "learning_rate": 0.0002,
+      "loss": 0.9734,
+      "step": 3030
+    },
+    {
+      "epoch": 2.2295562889622293,
+      "grad_norm": 0.6177583932876587,
+      "learning_rate": 0.0002,
+      "loss": 0.9033,
+      "step": 3040
+    },
+    {
+      "epoch": 2.236890355702237,
+      "grad_norm": 0.5256719589233398,
+      "learning_rate": 0.0002,
+      "loss": 0.9882,
+      "step": 3050
+    },
+    {
+      "epoch": 2.2442244224422443,
+      "grad_norm": 0.5001118183135986,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2515584891822513,
+      "grad_norm": 0.5721249580383301,
+      "learning_rate": 0.0002,
+      "loss": 0.8718,
+      "step": 3070
+    },
+    {
+      "epoch": 2.258892555922259,
+      "grad_norm": 0.5325384140014648,
+      "learning_rate": 0.0002,
+      "loss": 1.0648,
+      "step": 3080
+    },
+    {
+      "epoch": 2.2662266226622663,
+      "grad_norm": 0.5719189047813416,
+      "learning_rate": 0.0002,
+      "loss": 0.9843,
+      "step": 3090
+    },
+    {
+      "epoch": 2.2735606894022737,
+      "grad_norm": 0.6337835788726807,
+      "learning_rate": 0.0002,
+      "loss": 0.8633,
+      "step": 3100
+    },
+    {
+      "epoch": 2.2808947561422808,
+      "grad_norm": 0.5381836891174316,
+      "learning_rate": 0.0002,
+      "loss": 0.9962,
+      "step": 3110
+    },
+    {
+      "epoch": 2.2882288228822882,
+      "grad_norm": 0.5408531427383423,
+      "learning_rate": 0.0002,
+      "loss": 0.8265,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2955628896222957,
+      "grad_norm": 0.43705281615257263,
+      "learning_rate": 0.0002,
+      "loss": 1.0325,
+      "step": 3130
+    },
+    {
+      "epoch": 2.3028969563623027,
+      "grad_norm": 0.6454030275344849,
+      "learning_rate": 0.0002,
+      "loss": 0.9388,
+      "step": 3140
+    },
+    {
+      "epoch": 2.31023102310231,
+      "grad_norm": 0.686030387878418,
+      "learning_rate": 0.0002,
+      "loss": 0.954,
+      "step": 3150
+    },
+    {
+      "epoch": 2.3175650898423177,
+      "grad_norm": 0.5123633146286011,
+      "learning_rate": 0.0002,
+      "loss": 0.9403,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3248991565823247,
+      "grad_norm": 0.842506468296051,
+      "learning_rate": 0.0002,
+      "loss": 0.8834,
+      "step": 3170
+    },
+    {
+      "epoch": 2.332233223322332,
+      "grad_norm": 0.5193818807601929,
+      "learning_rate": 0.0002,
+      "loss": 1.0497,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3395672900623397,
+      "grad_norm": 0.5634409189224243,
+      "learning_rate": 0.0002,
+      "loss": 0.9473,
+      "step": 3190
+    },
+    {
+      "epoch": 2.3469013568023467,
+      "grad_norm": 0.6475534439086914,
+      "learning_rate": 0.0002,
+      "loss": 0.8499,
+      "step": 3200
+    },
+    {
+      "epoch": 2.354235423542354,
+      "grad_norm": 1.1503914594650269,
+      "learning_rate": 0.0002,
+      "loss": 0.874,
+      "step": 3210
+    },
+    {
+      "epoch": 2.3615694902823616,
+      "grad_norm": 0.7234905362129211,
+      "learning_rate": 0.0002,
+      "loss": 0.9762,
+      "step": 3220
+    },
+    {
+      "epoch": 2.368903557022369,
+      "grad_norm": 0.664903461933136,
+      "learning_rate": 0.0002,
+      "loss": 0.9007,
+      "step": 3230
+    },
+    {
+      "epoch": 2.376237623762376,
+      "grad_norm": 0.5453006625175476,
+      "learning_rate": 0.0002,
+      "loss": 0.9987,
+      "step": 3240
+    },
+    {
+      "epoch": 2.3835716905023836,
+      "grad_norm": 0.6256654262542725,
+      "learning_rate": 0.0002,
+      "loss": 0.9742,
+      "step": 3250
+    },
+    {
+      "epoch": 2.390905757242391,
+      "grad_norm": 0.5166565179824829,
+      "learning_rate": 0.0002,
+      "loss": 0.9922,
+      "step": 3260
+    },
+    {
+      "epoch": 2.398239823982398,
+      "grad_norm": 0.5699098110198975,
+      "learning_rate": 0.0002,
+      "loss": 0.927,
+      "step": 3270
+    },
+    {
+      "epoch": 2.4055738907224056,
+      "grad_norm": 0.4472540020942688,
+      "learning_rate": 0.0002,
+      "loss": 0.8878,
+      "step": 3280
+    },
+    {
+      "epoch": 2.412907957462413,
+      "grad_norm": 0.6790403127670288,
+      "learning_rate": 0.0002,
+      "loss": 0.9439,
+      "step": 3290
+    },
+    {
+      "epoch": 2.42024202420242,
+      "grad_norm": 0.5182185173034668,
+      "learning_rate": 0.0002,
+      "loss": 0.972,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4275760909424275,
+      "grad_norm": 0.564647912979126,
+      "learning_rate": 0.0002,
+      "loss": 0.9775,
+      "step": 3310
+    },
+    {
+      "epoch": 2.434910157682435,
+      "grad_norm": 0.5625313520431519,
+      "learning_rate": 0.0002,
+      "loss": 1.072,
+      "step": 3320
+    },
+    {
+      "epoch": 2.442244224422442,
+      "grad_norm": 0.7496559619903564,
+      "learning_rate": 0.0002,
+      "loss": 0.8798,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4495782911624495,
+      "grad_norm": 0.4779128134250641,
+      "learning_rate": 0.0002,
+      "loss": 0.868,
+      "step": 3340
+    },
+    {
+      "epoch": 2.456912357902457,
+      "grad_norm": 0.578093409538269,
+      "learning_rate": 0.0002,
+      "loss": 1.0316,
+      "step": 3350
+    },
+    {
+      "epoch": 2.4642464246424645,
+      "grad_norm": 0.5456080436706543,
+      "learning_rate": 0.0002,
+      "loss": 0.9282,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4715804913824715,
+      "grad_norm": 0.4769273102283478,
+      "learning_rate": 0.0002,
+      "loss": 0.8409,
+      "step": 3370
+    },
+    {
+      "epoch": 2.478914558122479,
+      "grad_norm": 0.5608189702033997,
+      "learning_rate": 0.0002,
+      "loss": 0.9312,
+      "step": 3380
+    },
+    {
+      "epoch": 2.4862486248624864,
+      "grad_norm": 0.5590165853500366,
+      "learning_rate": 0.0002,
+      "loss": 0.9934,
+      "step": 3390
+    },
+    {
+      "epoch": 2.4935826916024935,
+      "grad_norm": 0.801306962966919,
+      "learning_rate": 0.0002,
+      "loss": 1.025,
+      "step": 3400
+    },
+    {
+      "epoch": 2.500916758342501,
+      "grad_norm": 0.6045624613761902,
+      "learning_rate": 0.0002,
+      "loss": 0.9049,
+      "step": 3410
+    },
+    {
+      "epoch": 2.5082508250825084,
+      "grad_norm": 0.5735858082771301,
+      "learning_rate": 0.0002,
+      "loss": 0.944,
+      "step": 3420
+    },
+    {
+      "epoch": 2.5155848918225154,
+      "grad_norm": 0.6827309131622314,
+      "learning_rate": 0.0002,
+      "loss": 0.9846,
+      "step": 3430
+    },
+    {
+      "epoch": 2.522918958562523,
+      "grad_norm": 0.5702602863311768,
+      "learning_rate": 0.0002,
+      "loss": 0.9789,
+      "step": 3440
+    },
+    {
+      "epoch": 2.5302530253025304,
+      "grad_norm": 0.6674721240997314,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5375870920425374,
+      "grad_norm": 0.5635907649993896,
+      "learning_rate": 0.0002,
+      "loss": 0.914,
+      "step": 3460
+    },
+    {
+      "epoch": 2.544921158782545,
+      "grad_norm": 0.42737770080566406,
+      "learning_rate": 0.0002,
+      "loss": 0.8398,
+      "step": 3470
+    },
+    {
+      "epoch": 2.5522552255225524,
+      "grad_norm": 0.6720691919326782,
+      "learning_rate": 0.0002,
+      "loss": 0.9474,
+      "step": 3480
+    },
+    {
+      "epoch": 2.55958929226256,
+      "grad_norm": 0.8917084336280823,
+      "learning_rate": 0.0002,
+      "loss": 0.8637,
+      "step": 3490
+    },
+    {
+      "epoch": 2.566923359002567,
+      "grad_norm": 0.5134549140930176,
+      "learning_rate": 0.0002,
+      "loss": 0.9257,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5742574257425743,
+      "grad_norm": 0.4951367974281311,
+      "learning_rate": 0.0002,
+      "loss": 0.9362,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5815914924825814,
+      "grad_norm": 0.9438204765319824,
+      "learning_rate": 0.0002,
+      "loss": 0.9184,
+      "step": 3520
+    },
+    {
+      "epoch": 2.588925559222589,
+      "grad_norm": 0.6024714708328247,
+      "learning_rate": 0.0002,
+      "loss": 0.8939,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5962596259625963,
+      "grad_norm": 0.5248535871505737,
+      "learning_rate": 0.0002,
+      "loss": 0.9298,
+      "step": 3540
+    },
+    {
+      "epoch": 2.6035936927026038,
+      "grad_norm": 0.8677568435668945,
+      "learning_rate": 0.0002,
+      "loss": 0.941,
+      "step": 3550
+    },
+    {
+      "epoch": 2.610927759442611,
+      "grad_norm": 0.82008296251297,
+      "learning_rate": 0.0002,
+      "loss": 0.9253,
+      "step": 3560
+    },
+    {
+      "epoch": 2.6182618261826183,
+      "grad_norm": 0.4724634885787964,
+      "learning_rate": 0.0002,
+      "loss": 0.8429,
+      "step": 3570
+    },
+    {
+      "epoch": 2.6255958929226257,
+      "grad_norm": 0.5434244275093079,
+      "learning_rate": 0.0002,
+      "loss": 0.9058,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6329299596626328,
+      "grad_norm": 0.4948740005493164,
+      "learning_rate": 0.0002,
+      "loss": 0.9379,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6402640264026402,
+      "grad_norm": 0.42109328508377075,
+      "learning_rate": 0.0002,
+      "loss": 0.8718,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6475980931426477,
+      "grad_norm": 0.7979786396026611,
+      "learning_rate": 0.0002,
+      "loss": 0.9809,
+      "step": 3610
+    },
+    {
+      "epoch": 2.654932159882655,
+      "grad_norm": 0.6345919370651245,
+      "learning_rate": 0.0002,
+      "loss": 0.9229,
+      "step": 3620
+    },
+    {
+      "epoch": 2.662266226622662,
+      "grad_norm": 0.4971671402454376,
+      "learning_rate": 0.0002,
+      "loss": 0.8506,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6696002933626697,
+      "grad_norm": 0.6467748284339905,
+      "learning_rate": 0.0002,
+      "loss": 0.8054,
+      "step": 3640
+    },
+    {
+      "epoch": 2.6769343601026767,
+      "grad_norm": 0.4240160286426544,
+      "learning_rate": 0.0002,
+      "loss": 0.9277,
+      "step": 3650
+    },
+    {
+      "epoch": 2.684268426842684,
+      "grad_norm": 0.5179754495620728,
+      "learning_rate": 0.0002,
+      "loss": 0.8213,
+      "step": 3660
+    },
+    {
+      "epoch": 2.6916024935826917,
+      "grad_norm": 0.754012405872345,
+      "learning_rate": 0.0002,
+      "loss": 0.9221,
+      "step": 3670
+    },
+    {
+      "epoch": 2.698936560322699,
+      "grad_norm": 0.5141299962997437,
+      "learning_rate": 0.0002,
+      "loss": 0.9194,
+      "step": 3680
+    },
+    {
+      "epoch": 2.706270627062706,
+      "grad_norm": 0.5737819075584412,
+      "learning_rate": 0.0002,
+      "loss": 0.9495,
+      "step": 3690
+    },
+    {
+      "epoch": 2.7136046938027136,
+      "grad_norm": 0.5887577533721924,
+      "learning_rate": 0.0002,
+      "loss": 1.0162,
+      "step": 3700
+    },
+    {
+      "epoch": 2.720938760542721,
+      "grad_norm": 0.6740471720695496,
+      "learning_rate": 0.0002,
+      "loss": 0.9169,
+      "step": 3710
+    },
+    {
+      "epoch": 2.728272827282728,
+      "grad_norm": 0.5879453420639038,
+      "learning_rate": 0.0002,
+      "loss": 0.9297,
+      "step": 3720
+    },
+    {
+      "epoch": 2.7356068940227356,
+      "grad_norm": 0.4858354926109314,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 3730
+    },
+    {
+      "epoch": 2.742940960762743,
+      "grad_norm": 0.5489001870155334,
+      "learning_rate": 0.0002,
+      "loss": 0.9308,
+      "step": 3740
+    },
+    {
+      "epoch": 2.7502750275027505,
+      "grad_norm": 0.8187092542648315,
+      "learning_rate": 0.0002,
+      "loss": 0.894,
+      "step": 3750
+    },
+    {
+      "epoch": 2.7576090942427576,
+      "grad_norm": 0.5666626691818237,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 3760
+    },
+    {
+      "epoch": 2.764943160982765,
+      "grad_norm": 0.5377066135406494,
+      "learning_rate": 0.0002,
+      "loss": 1.0059,
+      "step": 3770
+    },
+    {
+      "epoch": 2.772277227722772,
+      "grad_norm": 0.566330075263977,
+      "learning_rate": 0.0002,
+      "loss": 0.9132,
+      "step": 3780
+    },
+    {
+      "epoch": 2.7796112944627795,
+      "grad_norm": 0.5522832870483398,
+      "learning_rate": 0.0002,
+      "loss": 0.9415,
+      "step": 3790
+    },
+    {
+      "epoch": 2.786945361202787,
+      "grad_norm": 0.5668695569038391,
+      "learning_rate": 0.0002,
+      "loss": 0.8816,
+      "step": 3800
+    },
+    {
+      "epoch": 2.7942794279427945,
+      "grad_norm": 0.7566602826118469,
+      "learning_rate": 0.0002,
+      "loss": 0.8885,
+      "step": 3810
+    },
+    {
+      "epoch": 2.8016134946828015,
+      "grad_norm": 0.5603684782981873,
+      "learning_rate": 0.0002,
+      "loss": 0.8598,
+      "step": 3820
+    },
+    {
+      "epoch": 2.808947561422809,
+      "grad_norm": 0.49122217297554016,
+      "learning_rate": 0.0002,
+      "loss": 0.9602,
+      "step": 3830
+    },
+    {
+      "epoch": 2.816281628162816,
+      "grad_norm": 0.6798251867294312,
+      "learning_rate": 0.0002,
+      "loss": 0.9738,
+      "step": 3840
+    },
+    {
+      "epoch": 2.8236156949028235,
+      "grad_norm": 0.6097991466522217,
+      "learning_rate": 0.0002,
+      "loss": 0.9533,
+      "step": 3850
+    },
+    {
+      "epoch": 2.830949761642831,
+      "grad_norm": 0.6675726175308228,
+      "learning_rate": 0.0002,
+      "loss": 0.8672,
+      "step": 3860
+    },
+    {
+      "epoch": 2.8382838283828384,
+      "grad_norm": 0.9223952889442444,
+      "learning_rate": 0.0002,
+      "loss": 0.9324,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8456178951228455,
+      "grad_norm": 0.6020799875259399,
+      "learning_rate": 0.0002,
+      "loss": 0.8767,
+      "step": 3880
+    },
+    {
+      "epoch": 2.852951961862853,
+      "grad_norm": 0.5206381678581238,
+      "learning_rate": 0.0002,
+      "loss": 0.9148,
+      "step": 3890
+    },
+    {
+      "epoch": 2.8602860286028604,
+      "grad_norm": 0.6268777251243591,
+      "learning_rate": 0.0002,
+      "loss": 0.9479,
+      "step": 3900
+    },
+    {
+      "epoch": 2.8676200953428674,
+      "grad_norm": 1.1583497524261475,
+      "learning_rate": 0.0002,
+      "loss": 0.9409,
+      "step": 3910
+    },
+    {
+      "epoch": 2.874954162082875,
+      "grad_norm": 0.7263903021812439,
+      "learning_rate": 0.0002,
+      "loss": 0.895,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8822882288228824,
+      "grad_norm": 0.5369910001754761,
+      "learning_rate": 0.0002,
+      "loss": 0.8786,
+      "step": 3930
+    },
+    {
+      "epoch": 2.88962229556289,
+      "grad_norm": 0.7298350930213928,
+      "learning_rate": 0.0002,
+      "loss": 1.0015,
+      "step": 3940
+    },
+    {
+      "epoch": 2.896956362302897,
+      "grad_norm": 0.577012836933136,
+      "learning_rate": 0.0002,
+      "loss": 0.979,
+      "step": 3950
+    },
+    {
+      "epoch": 2.9042904290429044,
+      "grad_norm": 0.5859594345092773,
+      "learning_rate": 0.0002,
+      "loss": 0.9716,
+      "step": 3960
+    },
+    {
+      "epoch": 2.9116244957829114,
+      "grad_norm": 0.47176122665405273,
+      "learning_rate": 0.0002,
+      "loss": 0.8772,
+      "step": 3970
+    },
+    {
+      "epoch": 2.918958562522919,
+      "grad_norm": 0.9699620604515076,
+      "learning_rate": 0.0002,
+      "loss": 0.8997,
+      "step": 3980
+    },
+    {
+      "epoch": 2.9262926292629263,
+      "grad_norm": 0.7908747792243958,
+      "learning_rate": 0.0002,
+      "loss": 0.9057,
+      "step": 3990
+    },
+    {
+      "epoch": 2.933626696002934,
+      "grad_norm": 0.5777379274368286,
+      "learning_rate": 0.0002,
+      "loss": 0.9462,
+      "step": 4000
+    },
+    {
+      "epoch": 2.940960762742941,
+      "grad_norm": 0.599288284778595,
+      "learning_rate": 0.0002,
+      "loss": 0.9358,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9482948294829483,
+      "grad_norm": 0.5232274532318115,
+      "learning_rate": 0.0002,
+      "loss": 0.9812,
+      "step": 4020
+    },
+    {
+      "epoch": 2.9556288962229558,
+      "grad_norm": 0.6395137310028076,
+      "learning_rate": 0.0002,
+      "loss": 0.96,
+      "step": 4030
+    },
+    {
+      "epoch": 2.962962962962963,
+      "grad_norm": 0.589260458946228,
+      "learning_rate": 0.0002,
+      "loss": 0.9813,
+      "step": 4040
+    },
+    {
+      "epoch": 2.9702970297029703,
+      "grad_norm": 0.5699581503868103,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 4050
+    },
+    {
+      "epoch": 2.9776310964429777,
+      "grad_norm": 0.528468132019043,
+      "learning_rate": 0.0002,
+      "loss": 0.9585,
+      "step": 4060
+    },
+    {
+      "epoch": 2.984965163182985,
+      "grad_norm": 0.4804670512676239,
+      "learning_rate": 0.0002,
+      "loss": 0.9164,
+      "step": 4070
+    },
+    {
+      "epoch": 2.9922992299229922,
+      "grad_norm": 1.1918889284133911,
+      "learning_rate": 0.0002,
+      "loss": 0.9771,
+      "step": 4080
+    },
+    {
+      "epoch": 2.9996332966629997,
+      "grad_norm": 0.5479103326797485,
+      "learning_rate": 0.0002,
+      "loss": 0.9178,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9996332966629997,
+      "eval_loss": 1.1642853021621704,
+      "eval_runtime": 32.7511,
+      "eval_samples_per_second": 13.16,
+      "eval_steps_per_second": 1.649,
+      "step": 4090
+    },
+    {
+      "epoch": 3.006967363403007,
+      "grad_norm": 0.7430027723312378,
+      "learning_rate": 0.0002,
+      "loss": 0.7981,
+      "step": 4100
+    },
+    {
+      "epoch": 3.014301430143014,
+      "grad_norm": 0.6293647289276123,
+      "learning_rate": 0.0002,
+      "loss": 0.7871,
+      "step": 4110
+    },
+    {
+      "epoch": 3.0216354968830217,
+      "grad_norm": 0.6191329956054688,
+      "learning_rate": 0.0002,
+      "loss": 0.78,
+      "step": 4120
+    },
+    {
+      "epoch": 3.028969563623029,
+      "grad_norm": 0.7959313988685608,
+      "learning_rate": 0.0002,
+      "loss": 0.7618,
+      "step": 4130
+    },
+    {
+      "epoch": 3.036303630363036,
+      "grad_norm": 0.5956351161003113,
+      "learning_rate": 0.0002,
+      "loss": 0.8039,
+      "step": 4140
+    },
+    {
+      "epoch": 3.0436376971030437,
+      "grad_norm": 0.670383632183075,
+      "learning_rate": 0.0002,
+      "loss": 0.7477,
+      "step": 4150
+    },
+    {
+      "epoch": 3.050971763843051,
+      "grad_norm": 0.6414518356323242,
+      "learning_rate": 0.0002,
+      "loss": 0.7984,
+      "step": 4160
+    },
+    {
+      "epoch": 3.058305830583058,
+      "grad_norm": 0.7928852438926697,
+      "learning_rate": 0.0002,
+      "loss": 0.7369,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0656398973230656,
+      "grad_norm": 0.6211121082305908,
+      "learning_rate": 0.0002,
+      "loss": 0.7914,
+      "step": 4180
+    },
+    {
+      "epoch": 3.072973964063073,
+      "grad_norm": 0.6237057447433472,
+      "learning_rate": 0.0002,
+      "loss": 0.7365,
+      "step": 4190
+    },
+    {
+      "epoch": 3.08030803080308,
+      "grad_norm": 0.6522233486175537,
+      "learning_rate": 0.0002,
+      "loss": 0.702,
+      "step": 4200
+    },
+    {
+      "epoch": 3.0876420975430876,
+      "grad_norm": 0.9396848678588867,
+      "learning_rate": 0.0002,
+      "loss": 0.7646,
+      "step": 4210
+    },
+    {
+      "epoch": 3.094976164283095,
+      "grad_norm": 0.8003010749816895,
+      "learning_rate": 0.0002,
+      "loss": 0.7559,
+      "step": 4220
+    },
+    {
+      "epoch": 3.102310231023102,
+      "grad_norm": 0.6733810305595398,
+      "learning_rate": 0.0002,
+      "loss": 0.711,
+      "step": 4230
+    },
+    {
+      "epoch": 3.1096442977631096,
+      "grad_norm": 0.6365828514099121,
+      "learning_rate": 0.0002,
+      "loss": 0.696,
+      "step": 4240
+    },
+    {
+      "epoch": 3.116978364503117,
+      "grad_norm": 1.0805548429489136,
+      "learning_rate": 0.0002,
+      "loss": 0.8362,
+      "step": 4250
+    },
+    {
+      "epoch": 3.1243124312431245,
+      "grad_norm": 0.7262141108512878,
+      "learning_rate": 0.0002,
+      "loss": 0.7651,
+      "step": 4260
+    },
+    {
+      "epoch": 3.1316464979831315,
+      "grad_norm": 0.5500539541244507,
+      "learning_rate": 0.0002,
+      "loss": 0.7304,
+      "step": 4270
+    },
+    {
+      "epoch": 3.138980564723139,
+      "grad_norm": 0.793912947177887,
+      "learning_rate": 0.0002,
+      "loss": 0.7721,
+      "step": 4280
+    },
+    {
+      "epoch": 3.1463146314631465,
+      "grad_norm": 1.2540518045425415,
+      "learning_rate": 0.0002,
+      "loss": 0.7708,
+      "step": 4290
+    },
+    {
+      "epoch": 3.1536486982031535,
+      "grad_norm": 0.7020077705383301,
+      "learning_rate": 0.0002,
+      "loss": 0.782,
+      "step": 4300
+    },
+    {
+      "epoch": 3.160982764943161,
+      "grad_norm": 0.5111123323440552,
+      "learning_rate": 0.0002,
+      "loss": 0.7253,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1683168316831685,
+      "grad_norm": 0.7172090411186218,
+      "learning_rate": 0.0002,
+      "loss": 0.8159,
+      "step": 4320
+    },
+    {
+      "epoch": 3.1756508984231755,
+      "grad_norm": 0.6343168616294861,
+      "learning_rate": 0.0002,
+      "loss": 0.6962,
+      "step": 4330
+    },
+    {
+      "epoch": 3.182984965163183,
+      "grad_norm": 0.9563672542572021,
+      "learning_rate": 0.0002,
+      "loss": 0.7938,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1903190319031904,
+      "grad_norm": 1.0225574970245361,
+      "learning_rate": 0.0002,
+      "loss": 0.7385,
+      "step": 4350
+    },
+    {
+      "epoch": 3.1976530986431975,
+      "grad_norm": 1.1633386611938477,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 4360
+    },
+    {
+      "epoch": 3.204987165383205,
+      "grad_norm": 0.8915148973464966,
+      "learning_rate": 0.0002,
+      "loss": 0.7259,
+      "step": 4370
+    },
+    {
+      "epoch": 3.2123212321232124,
+      "grad_norm": 0.9156812429428101,
+      "learning_rate": 0.0002,
+      "loss": 0.8061,
+      "step": 4380
+    },
+    {
+      "epoch": 3.21965529886322,
+      "grad_norm": 0.6363258957862854,
+      "learning_rate": 0.0002,
+      "loss": 0.8189,
+      "step": 4390
+    },
+    {
+      "epoch": 3.226989365603227,
+      "grad_norm": 0.579099178314209,
+      "learning_rate": 0.0002,
+      "loss": 0.7996,
+      "step": 4400
+    },
+    {
+      "epoch": 3.2343234323432344,
+      "grad_norm": 0.8778146505355835,
+      "learning_rate": 0.0002,
+      "loss": 0.8592,
+      "step": 4410
+    },
+    {
+      "epoch": 3.241657499083242,
+      "grad_norm": 0.8356770873069763,
+      "learning_rate": 0.0002,
+      "loss": 0.8281,
+      "step": 4420
+    },
+    {
+      "epoch": 3.248991565823249,
+      "grad_norm": 0.702032208442688,
+      "learning_rate": 0.0002,
+      "loss": 0.8484,
+      "step": 4430
+    },
+    {
+      "epoch": 3.2563256325632564,
+      "grad_norm": 0.6386539340019226,
+      "learning_rate": 0.0002,
+      "loss": 0.7227,
+      "step": 4440
+    },
+    {
+      "epoch": 3.263659699303264,
+      "grad_norm": 0.7008408904075623,
+      "learning_rate": 0.0002,
+      "loss": 0.8374,
+      "step": 4450
+    },
+    {
+      "epoch": 3.270993766043271,
+      "grad_norm": 0.9556332230567932,
+      "learning_rate": 0.0002,
+      "loss": 0.7572,
+      "step": 4460
+    },
+    {
+      "epoch": 3.2783278327832783,
+      "grad_norm": 0.5667835474014282,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 4470
+    },
+    {
+      "epoch": 3.285661899523286,
+      "grad_norm": 0.8239172697067261,
+      "learning_rate": 0.0002,
+      "loss": 0.8152,
+      "step": 4480
+    },
+    {
+      "epoch": 3.292995966263293,
+      "grad_norm": 0.7045050859451294,
+      "learning_rate": 0.0002,
+      "loss": 0.756,
+      "step": 4490
+    },
+    {
+      "epoch": 3.3003300330033003,
+      "grad_norm": 0.7131434082984924,
+      "learning_rate": 0.0002,
+      "loss": 0.7655,
+      "step": 4500
+    },
+    {
+      "epoch": 3.3076640997433078,
+      "grad_norm": 0.6924910545349121,
+      "learning_rate": 0.0002,
+      "loss": 0.836,
+      "step": 4510
+    },
+    {
+      "epoch": 3.3149981664833152,
+      "grad_norm": 0.8945356607437134,
+      "learning_rate": 0.0002,
+      "loss": 0.736,
+      "step": 4520
+    },
+    {
+      "epoch": 3.3223322332233223,
+      "grad_norm": 0.6546903252601624,
+      "learning_rate": 0.0002,
+      "loss": 0.7575,
+      "step": 4530
+    },
+    {
+      "epoch": 3.3296662999633297,
+      "grad_norm": 0.8206679224967957,
+      "learning_rate": 0.0002,
+      "loss": 0.7893,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3370003667033368,
+      "grad_norm": 0.6482203602790833,
+      "learning_rate": 0.0002,
+      "loss": 0.7502,
+      "step": 4550
+    },
+    {
+      "epoch": 3.3443344334433442,
+      "grad_norm": 0.7558760046958923,
+      "learning_rate": 0.0002,
+      "loss": 0.8172,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3516685001833517,
+      "grad_norm": 0.7794756889343262,
+      "learning_rate": 0.0002,
+      "loss": 0.744,
+      "step": 4570
+    },
+    {
+      "epoch": 3.359002566923359,
+      "grad_norm": 0.7382805943489075,
+      "learning_rate": 0.0002,
+      "loss": 0.7385,
+      "step": 4580
+    },
+    {
+      "epoch": 3.366336633663366,
+      "grad_norm": 0.5912511944770813,
+      "learning_rate": 0.0002,
+      "loss": 0.8511,
+      "step": 4590
+    },
+    {
+      "epoch": 3.3736707004033737,
+      "grad_norm": 0.7444885969161987,
+      "learning_rate": 0.0002,
+      "loss": 0.8272,
+      "step": 4600
+    },
+    {
+      "epoch": 3.381004767143381,
+      "grad_norm": 0.7354922890663147,
+      "learning_rate": 0.0002,
+      "loss": 0.7927,
+      "step": 4610
+    },
+    {
+      "epoch": 3.388338833883388,
+      "grad_norm": 0.7685934901237488,
+      "learning_rate": 0.0002,
+      "loss": 0.7183,
+      "step": 4620
+    },
+    {
+      "epoch": 3.3956729006233957,
+      "grad_norm": 0.61041259765625,
+      "learning_rate": 0.0002,
+      "loss": 0.7436,
+      "step": 4630
+    },
+    {
+      "epoch": 3.403006967363403,
+      "grad_norm": 0.6820451021194458,
+      "learning_rate": 0.0002,
+      "loss": 0.7661,
+      "step": 4640
+    },
+    {
+      "epoch": 3.41034103410341,
+      "grad_norm": 0.5819534063339233,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 4650
+    },
+    {
+      "epoch": 3.4176751008434176,
+      "grad_norm": 0.705410897731781,
+      "learning_rate": 0.0002,
+      "loss": 0.7314,
+      "step": 4660
+    },
+    {
+      "epoch": 3.425009167583425,
+      "grad_norm": 0.8052892088890076,
+      "learning_rate": 0.0002,
+      "loss": 0.7901,
+      "step": 4670
+    },
+    {
+      "epoch": 3.432343234323432,
+      "grad_norm": 0.7746483087539673,
+      "learning_rate": 0.0002,
+      "loss": 0.7298,
+      "step": 4680
+    },
+    {
+      "epoch": 3.4396773010634396,
+      "grad_norm": 0.7713689804077148,
+      "learning_rate": 0.0002,
+      "loss": 0.7976,
+      "step": 4690
+    },
+    {
+      "epoch": 3.447011367803447,
+      "grad_norm": 0.810371994972229,
+      "learning_rate": 0.0002,
+      "loss": 0.7427,
+      "step": 4700
+    },
+    {
+      "epoch": 3.4543454345434546,
+      "grad_norm": 0.7702969312667847,
+      "learning_rate": 0.0002,
+      "loss": 0.7594,
+      "step": 4710
+    },
+    {
+      "epoch": 3.4616795012834616,
+      "grad_norm": 0.7069268822669983,
+      "learning_rate": 0.0002,
+      "loss": 0.7957,
+      "step": 4720
+    },
+    {
+      "epoch": 3.469013568023469,
+      "grad_norm": 0.7640359401702881,
+      "learning_rate": 0.0002,
+      "loss": 0.8199,
+      "step": 4730
+    },
+    {
+      "epoch": 3.4763476347634765,
+      "grad_norm": 0.8661707639694214,
+      "learning_rate": 0.0002,
+      "loss": 0.6875,
+      "step": 4740
+    },
+    {
+      "epoch": 3.4836817015034836,
+      "grad_norm": 0.9970282912254333,
+      "learning_rate": 0.0002,
+      "loss": 0.8528,
+      "step": 4750
+    },
+    {
+      "epoch": 3.491015768243491,
+      "grad_norm": 0.5824355483055115,
+      "learning_rate": 0.0002,
+      "loss": 0.8462,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4983498349834985,
+      "grad_norm": 1.3072649240493774,
+      "learning_rate": 0.0002,
+      "loss": 0.851,
+      "step": 4770
+    },
+    {
+      "epoch": 3.5056839017235055,
+      "grad_norm": 0.873978316783905,
+      "learning_rate": 0.0002,
+      "loss": 0.9101,
+      "step": 4780
+    },
+    {
+      "epoch": 3.513017968463513,
+      "grad_norm": 0.5526657104492188,
+      "learning_rate": 0.0002,
+      "loss": 0.7403,
+      "step": 4790
+    },
+    {
+      "epoch": 3.5203520352035205,
+      "grad_norm": 0.790894627571106,
+      "learning_rate": 0.0002,
+      "loss": 0.7921,
+      "step": 4800
+    },
+    {
+      "epoch": 3.5276861019435275,
+      "grad_norm": 0.8119630217552185,
+      "learning_rate": 0.0002,
+      "loss": 0.831,
+      "step": 4810
+    },
+    {
+      "epoch": 3.535020168683535,
+      "grad_norm": 0.633212149143219,
+      "learning_rate": 0.0002,
+      "loss": 0.7351,
+      "step": 4820
+    },
+    {
+      "epoch": 3.5423542354235424,
+      "grad_norm": 0.703029990196228,
+      "learning_rate": 0.0002,
+      "loss": 0.8505,
+      "step": 4830
+    },
+    {
+      "epoch": 3.54968830216355,
+      "grad_norm": 0.7603771686553955,
+      "learning_rate": 0.0002,
+      "loss": 0.7204,
+      "step": 4840
+    },
+    {
+      "epoch": 3.557022368903557,
+      "grad_norm": 0.6260480880737305,
+      "learning_rate": 0.0002,
+      "loss": 0.8868,
+      "step": 4850
+    },
+    {
+      "epoch": 3.5643564356435644,
+      "grad_norm": 0.8203664422035217,
+      "learning_rate": 0.0002,
+      "loss": 0.8137,
+      "step": 4860
+    },
+    {
+      "epoch": 3.5716905023835714,
+      "grad_norm": 0.7793813347816467,
+      "learning_rate": 0.0002,
+      "loss": 0.8821,
+      "step": 4870
+    },
+    {
+      "epoch": 3.579024569123579,
+      "grad_norm": 0.7667397260665894,
+      "learning_rate": 0.0002,
+      "loss": 0.8164,
+      "step": 4880
+    },
+    {
+      "epoch": 3.5863586358635864,
+      "grad_norm": 0.8198829889297485,
+      "learning_rate": 0.0002,
+      "loss": 0.7597,
+      "step": 4890
+    },
+    {
+      "epoch": 3.593692702603594,
+      "grad_norm": 0.7689233422279358,
+      "learning_rate": 0.0002,
+      "loss": 0.7027,
+      "step": 4900
+    },
+    {
+      "epoch": 3.601026769343601,
+      "grad_norm": 0.7870983481407166,
+      "learning_rate": 0.0002,
+      "loss": 0.804,
+      "step": 4910
+    },
+    {
+      "epoch": 3.6083608360836084,
+      "grad_norm": 0.8133853077888489,
+      "learning_rate": 0.0002,
+      "loss": 0.8269,
+      "step": 4920
+    },
+    {
+      "epoch": 3.615694902823616,
+      "grad_norm": 1.308401346206665,
+      "learning_rate": 0.0002,
+      "loss": 0.8515,
+      "step": 4930
+    },
+    {
+      "epoch": 3.623028969563623,
+      "grad_norm": 0.7131121754646301,
+      "learning_rate": 0.0002,
+      "loss": 0.8494,
+      "step": 4940
+    },
+    {
+      "epoch": 3.6303630363036303,
+      "grad_norm": 0.6825910210609436,
+      "learning_rate": 0.0002,
+      "loss": 0.7235,
+      "step": 4950
+    },
+    {
+      "epoch": 3.637697103043638,
+      "grad_norm": 0.7254678606987,
+      "learning_rate": 0.0002,
+      "loss": 0.7824,
+      "step": 4960
+    },
+    {
+      "epoch": 3.6450311697836453,
+      "grad_norm": 0.8045085072517395,
+      "learning_rate": 0.0002,
+      "loss": 0.7983,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6523652365236523,
+      "grad_norm": 0.6991777420043945,
+      "learning_rate": 0.0002,
+      "loss": 0.8223,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6596993032636598,
+      "grad_norm": 0.7804713249206543,
+      "learning_rate": 0.0002,
+      "loss": 0.7806,
+      "step": 4990
+    },
+    {
+      "epoch": 3.667033370003667,
+      "grad_norm": 0.8525708317756653,
+      "learning_rate": 0.0002,
+      "loss": 0.8402,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6743674367436743,
+      "grad_norm": 0.7959994673728943,
+      "learning_rate": 0.0002,
+      "loss": 0.8496,
+      "step": 5010
+    },
+    {
+      "epoch": 3.6817015034836817,
+      "grad_norm": 0.8103628158569336,
+      "learning_rate": 0.0002,
+      "loss": 0.8022,
+      "step": 5020
+    },
+    {
+      "epoch": 3.689035570223689,
+      "grad_norm": 0.7517836093902588,
+      "learning_rate": 0.0002,
+      "loss": 0.7376,
+      "step": 5030
+    },
+    {
+      "epoch": 3.6963696369636962,
+      "grad_norm": 0.6878514289855957,
+      "learning_rate": 0.0002,
+      "loss": 0.8375,
+      "step": 5040
+    },
+    {
+      "epoch": 3.7037037037037037,
+      "grad_norm": 1.2371820211410522,
+      "learning_rate": 0.0002,
+      "loss": 0.7998,
+      "step": 5050
+    },
+    {
+      "epoch": 3.711037770443711,
+      "grad_norm": 0.6567103862762451,
+      "learning_rate": 0.0002,
+      "loss": 0.6941,
+      "step": 5060
+    },
+    {
+      "epoch": 3.718371837183718,
+      "grad_norm": 1.1254922151565552,
+      "learning_rate": 0.0002,
+      "loss": 0.8465,
+      "step": 5070
+    },
+    {
+      "epoch": 3.7257059039237257,
+      "grad_norm": 0.6796132326126099,
+      "learning_rate": 0.0002,
+      "loss": 0.8365,
+      "step": 5080
+    },
+    {
+      "epoch": 3.733039970663733,
+      "grad_norm": 0.7285300493240356,
+      "learning_rate": 0.0002,
+      "loss": 0.7818,
+      "step": 5090
+    },
+    {
+      "epoch": 3.7403740374037406,
+      "grad_norm": 0.8931500911712646,
+      "learning_rate": 0.0002,
+      "loss": 0.8581,
+      "step": 5100
+    },
+    {
+      "epoch": 3.7477081041437477,
+      "grad_norm": 0.6256856918334961,
+      "learning_rate": 0.0002,
+      "loss": 0.8181,
+      "step": 5110
+    },
+    {
+      "epoch": 3.755042170883755,
+      "grad_norm": 0.79310142993927,
+      "learning_rate": 0.0002,
+      "loss": 0.743,
+      "step": 5120
+    },
+    {
+      "epoch": 3.762376237623762,
+      "grad_norm": 0.6594041585922241,
+      "learning_rate": 0.0002,
+      "loss": 0.8235,
+      "step": 5130
+    },
+    {
+      "epoch": 3.7697103043637696,
+      "grad_norm": 0.7029327750205994,
+      "learning_rate": 0.0002,
+      "loss": 0.6925,
+      "step": 5140
+    },
+    {
+      "epoch": 3.777044371103777,
+      "grad_norm": 0.5880070328712463,
+      "learning_rate": 0.0002,
+      "loss": 0.7457,
+      "step": 5150
+    },
+    {
+      "epoch": 3.7843784378437846,
+      "grad_norm": 0.7578945159912109,
+      "learning_rate": 0.0002,
+      "loss": 0.8716,
+      "step": 5160
+    },
+    {
+      "epoch": 3.7917125045837916,
+      "grad_norm": 0.8276378512382507,
+      "learning_rate": 0.0002,
+      "loss": 0.8819,
+      "step": 5170
+    },
+    {
+      "epoch": 3.799046571323799,
+      "grad_norm": 0.7627953886985779,
+      "learning_rate": 0.0002,
+      "loss": 0.7559,
+      "step": 5180
+    },
+    {
+      "epoch": 3.806380638063806,
+      "grad_norm": 0.8169086575508118,
+      "learning_rate": 0.0002,
+      "loss": 0.7665,
+      "step": 5190
+    },
+    {
+      "epoch": 3.8137147048038136,
+      "grad_norm": 0.6605030298233032,
+      "learning_rate": 0.0002,
+      "loss": 0.761,
+      "step": 5200
+    },
+    {
+      "epoch": 3.821048771543821,
+      "grad_norm": 0.5837286114692688,
+      "learning_rate": 0.0002,
+      "loss": 0.8804,
+      "step": 5210
+    },
+    {
+      "epoch": 3.8283828382838285,
+      "grad_norm": 1.2422157526016235,
+      "learning_rate": 0.0002,
+      "loss": 0.8369,
+      "step": 5220
+    },
+    {
+      "epoch": 3.8357169050238356,
+      "grad_norm": 0.6589220762252808,
+      "learning_rate": 0.0002,
+      "loss": 0.8431,
+      "step": 5230
+    },
+    {
+      "epoch": 3.843050971763843,
+      "grad_norm": 0.8567556142807007,
+      "learning_rate": 0.0002,
+      "loss": 0.7686,
+      "step": 5240
+    },
+    {
+      "epoch": 3.8503850385038505,
+      "grad_norm": 0.6490627527236938,
+      "learning_rate": 0.0002,
+      "loss": 0.8652,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8577191052438575,
+      "grad_norm": 0.620232880115509,
+      "learning_rate": 0.0002,
+      "loss": 0.7386,
+      "step": 5260
+    },
+    {
+      "epoch": 3.865053171983865,
+      "grad_norm": 0.7685128450393677,
+      "learning_rate": 0.0002,
+      "loss": 0.9192,
+      "step": 5270
+    },
+    {
+      "epoch": 3.8723872387238725,
+      "grad_norm": 0.8113296627998352,
+      "learning_rate": 0.0002,
+      "loss": 0.872,
+      "step": 5280
+    },
+    {
+      "epoch": 3.87972130546388,
+      "grad_norm": 0.8092675805091858,
+      "learning_rate": 0.0002,
+      "loss": 0.7156,
+      "step": 5290
+    },
+    {
+      "epoch": 3.887055372203887,
+      "grad_norm": 0.583570122718811,
+      "learning_rate": 0.0002,
+      "loss": 0.7325,
+      "step": 5300
+    },
+    {
+      "epoch": 3.8943894389438944,
+      "grad_norm": 1.712363600730896,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 5310
+    },
+    {
+      "epoch": 3.9017235056839015,
+      "grad_norm": 0.6673534512519836,
+      "learning_rate": 0.0002,
+      "loss": 0.7537,
+      "step": 5320
+    },
+    {
+      "epoch": 3.909057572423909,
+      "grad_norm": 1.9770312309265137,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 5330
+    },
+    {
+      "epoch": 3.9163916391639164,
+      "grad_norm": 0.6430999636650085,
+      "learning_rate": 0.0002,
+      "loss": 0.8793,
+      "step": 5340
+    },
+    {
+      "epoch": 3.923725705903924,
+      "grad_norm": 1.0159571170806885,
+      "learning_rate": 0.0002,
+      "loss": 0.839,
+      "step": 5350
+    },
+    {
+      "epoch": 3.931059772643931,
+      "grad_norm": 0.8607584834098816,
+      "learning_rate": 0.0002,
+      "loss": 0.9332,
+      "step": 5360
+    },
+    {
+      "epoch": 3.9383938393839384,
+      "grad_norm": 0.6967900991439819,
+      "learning_rate": 0.0002,
+      "loss": 0.7261,
+      "step": 5370
+    },
+    {
+      "epoch": 3.945727906123946,
+      "grad_norm": 0.7683077454566956,
+      "learning_rate": 0.0002,
+      "loss": 0.8456,
+      "step": 5380
+    },
+    {
+      "epoch": 3.953061972863953,
+      "grad_norm": 0.6805762648582458,
+      "learning_rate": 0.0002,
+      "loss": 0.7682,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9603960396039604,
+      "grad_norm": 0.7033619284629822,
+      "learning_rate": 0.0002,
+      "loss": 0.7746,
+      "step": 5400
+    },
+    {
+      "epoch": 3.967730106343968,
+      "grad_norm": 0.966112494468689,
+      "learning_rate": 0.0002,
+      "loss": 0.8393,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9750641730839753,
+      "grad_norm": 0.8467881083488464,
+      "learning_rate": 0.0002,
+      "loss": 0.8316,
+      "step": 5420
+    },
+    {
+      "epoch": 3.9823982398239823,
+      "grad_norm": 0.8005317449569702,
+      "learning_rate": 0.0002,
+      "loss": 0.8084,
+      "step": 5430
+    },
+    {
+      "epoch": 3.98973230656399,
+      "grad_norm": 1.1615241765975952,
+      "learning_rate": 0.0002,
+      "loss": 0.7168,
+      "step": 5440
+    },
+    {
+      "epoch": 3.997066373303997,
+      "grad_norm": 0.6121614575386047,
+      "learning_rate": 0.0002,
+      "loss": 0.8263,
+      "step": 5450
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.1834222078323364,
+      "eval_runtime": 32.7569,
+      "eval_samples_per_second": 13.158,
+      "eval_steps_per_second": 1.649,
+      "step": 5454
+    },
+    {
+      "epoch": 4.004400440044004,
+      "grad_norm": 0.6055727005004883,
+      "learning_rate": 0.0002,
+      "loss": 0.7267,
+      "step": 5460
+    },
+    {
+      "epoch": 4.011734506784012,
+      "grad_norm": 0.8232647180557251,
+      "learning_rate": 0.0002,
+      "loss": 0.5766,
+      "step": 5470
+    },
+    {
+      "epoch": 4.019068573524019,
+      "grad_norm": 0.7739192247390747,
+      "learning_rate": 0.0002,
+      "loss": 0.6489,
+      "step": 5480
+    },
+    {
+      "epoch": 4.026402640264027,
+      "grad_norm": 0.6264950633049011,
+      "learning_rate": 0.0002,
+      "loss": 0.5978,
+      "step": 5490
+    },
+    {
+      "epoch": 4.033736707004033,
+      "grad_norm": 1.4798702001571655,
+      "learning_rate": 0.0002,
+      "loss": 0.6392,
+      "step": 5500
+    },
+    {
+      "epoch": 4.041070773744041,
+      "grad_norm": 0.9538470506668091,
+      "learning_rate": 0.0002,
+      "loss": 0.6143,
+      "step": 5510
+    },
+    {
+      "epoch": 4.048404840484048,
+      "grad_norm": 0.834561288356781,
+      "learning_rate": 0.0002,
+      "loss": 0.6056,
+      "step": 5520
+    },
+    {
+      "epoch": 4.055738907224056,
+      "grad_norm": 0.6407850384712219,
+      "learning_rate": 0.0002,
+      "loss": 0.6077,
+      "step": 5530
+    },
+    {
+      "epoch": 4.063072973964063,
+      "grad_norm": 0.9035961627960205,
+      "learning_rate": 0.0002,
+      "loss": 0.6733,
+      "step": 5540
+    },
+    {
+      "epoch": 4.070407040704071,
+      "grad_norm": 0.842812716960907,
+      "learning_rate": 0.0002,
+      "loss": 0.5854,
+      "step": 5550
+    },
+    {
+      "epoch": 4.077741107444078,
+      "grad_norm": 0.8197882175445557,
+      "learning_rate": 0.0002,
+      "loss": 0.654,
+      "step": 5560
+    },
+    {
+      "epoch": 4.085075174184085,
+      "grad_norm": 0.8652673959732056,
+      "learning_rate": 0.0002,
+      "loss": 0.5919,
+      "step": 5570
+    },
+    {
+      "epoch": 4.092409240924092,
+      "grad_norm": 0.8048318028450012,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 5580
+    },
+    {
+      "epoch": 4.0997433076641,
+      "grad_norm": 0.9604969024658203,
+      "learning_rate": 0.0002,
+      "loss": 0.6487,
+      "step": 5590
+    },
+    {
+      "epoch": 4.107077374404107,
+      "grad_norm": 1.244756817817688,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 5600
+    },
+    {
+      "epoch": 4.114411441144115,
+      "grad_norm": 0.7975269556045532,
+      "learning_rate": 0.0002,
+      "loss": 0.6489,
+      "step": 5610
+    },
+    {
+      "epoch": 4.121745507884122,
+      "grad_norm": 0.6130099296569824,
+      "learning_rate": 0.0002,
+      "loss": 0.6445,
+      "step": 5620
+    },
+    {
+      "epoch": 4.129079574624129,
+      "grad_norm": 0.7793202996253967,
+      "learning_rate": 0.0002,
+      "loss": 0.6024,
+      "step": 5630
+    },
+    {
+      "epoch": 4.136413641364136,
+      "grad_norm": 1.187238335609436,
+      "learning_rate": 0.0002,
+      "loss": 0.5723,
+      "step": 5640
+    },
+    {
+      "epoch": 4.143747708104144,
+      "grad_norm": 0.8450375199317932,
+      "learning_rate": 0.0002,
+      "loss": 0.6385,
+      "step": 5650
+    },
+    {
+      "epoch": 4.151081774844151,
+      "grad_norm": 0.9006940126419067,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 5660
+    },
+    {
+      "epoch": 4.158415841584159,
+      "grad_norm": 0.9447154998779297,
+      "learning_rate": 0.0002,
+      "loss": 0.6179,
+      "step": 5670
+    },
+    {
+      "epoch": 4.165749908324166,
+      "grad_norm": 0.798032283782959,
+      "learning_rate": 0.0002,
+      "loss": 0.6476,
+      "step": 5680
+    },
+    {
+      "epoch": 4.1730839750641735,
+      "grad_norm": 0.65578693151474,
+      "learning_rate": 0.0002,
+      "loss": 0.6666,
+      "step": 5690
+    },
+    {
+      "epoch": 4.18041804180418,
+      "grad_norm": 1.0864700078964233,
+      "learning_rate": 0.0002,
+      "loss": 0.701,
+      "step": 5700
+    },
+    {
+      "epoch": 4.187752108544188,
+      "grad_norm": 0.7344121932983398,
+      "learning_rate": 0.0002,
+      "loss": 0.6895,
+      "step": 5710
+    },
+    {
+      "epoch": 4.195086175284195,
+      "grad_norm": 0.9722456932067871,
+      "learning_rate": 0.0002,
+      "loss": 0.6659,
+      "step": 5720
+    },
+    {
+      "epoch": 4.2024202420242025,
+      "grad_norm": 1.263814926147461,
+      "learning_rate": 0.0002,
+      "loss": 0.6887,
+      "step": 5730
+    },
+    {
+      "epoch": 4.20975430876421,
+      "grad_norm": 0.9622581005096436,
+      "learning_rate": 0.0002,
+      "loss": 0.608,
+      "step": 5740
+    },
+    {
+      "epoch": 4.2170883755042174,
+      "grad_norm": 0.8497143387794495,
+      "learning_rate": 0.0002,
+      "loss": 0.6221,
+      "step": 5750
+    },
+    {
+      "epoch": 4.224422442244224,
+      "grad_norm": 0.8248446583747864,
+      "learning_rate": 0.0002,
+      "loss": 0.6322,
+      "step": 5760
+    },
+    {
+      "epoch": 4.2317565089842315,
+      "grad_norm": 1.2544798851013184,
+      "learning_rate": 0.0002,
+      "loss": 0.6045,
+      "step": 5770
+    },
+    {
+      "epoch": 4.239090575724239,
+      "grad_norm": 0.8224676251411438,
+      "learning_rate": 0.0002,
+      "loss": 0.641,
+      "step": 5780
+    },
+    {
+      "epoch": 4.2464246424642464,
+      "grad_norm": 0.8924877047538757,
+      "learning_rate": 0.0002,
+      "loss": 0.6399,
+      "step": 5790
+    },
+    {
+      "epoch": 4.253758709204254,
+      "grad_norm": 0.8545848727226257,
+      "learning_rate": 0.0002,
+      "loss": 0.6845,
+      "step": 5800
+    },
+    {
+      "epoch": 4.261092775944261,
+      "grad_norm": 0.8081067800521851,
+      "learning_rate": 0.0002,
+      "loss": 0.6669,
+      "step": 5810
+    },
+    {
+      "epoch": 4.268426842684269,
+      "grad_norm": 0.7111002802848816,
+      "learning_rate": 0.0002,
+      "loss": 0.6149,
+      "step": 5820
+    },
+    {
+      "epoch": 4.2757609094242754,
+      "grad_norm": 0.8696979880332947,
+      "learning_rate": 0.0002,
+      "loss": 0.6343,
+      "step": 5830
+    },
+    {
+      "epoch": 4.283094976164283,
+      "grad_norm": 0.821401834487915,
+      "learning_rate": 0.0002,
+      "loss": 0.6384,
+      "step": 5840
+    },
+    {
+      "epoch": 4.29042904290429,
+      "grad_norm": 0.888908326625824,
+      "learning_rate": 0.0002,
+      "loss": 0.6912,
+      "step": 5850
+    },
+    {
+      "epoch": 4.297763109644298,
+      "grad_norm": 1.9380123615264893,
+      "learning_rate": 0.0002,
+      "loss": 0.6061,
+      "step": 5860
+    },
+    {
+      "epoch": 4.305097176384305,
+      "grad_norm": 1.121774435043335,
+      "learning_rate": 0.0002,
+      "loss": 0.6766,
+      "step": 5870
+    },
+    {
+      "epoch": 4.312431243124313,
+      "grad_norm": 0.9238282442092896,
+      "learning_rate": 0.0002,
+      "loss": 0.7205,
+      "step": 5880
+    },
+    {
+      "epoch": 4.319765309864319,
+      "grad_norm": 0.7321620583534241,
+      "learning_rate": 0.0002,
+      "loss": 0.6351,
+      "step": 5890
+    },
+    {
+      "epoch": 4.327099376604327,
+      "grad_norm": 0.8739548325538635,
+      "learning_rate": 0.0002,
+      "loss": 0.6404,
+      "step": 5900
+    },
+    {
+      "epoch": 4.334433443344334,
+      "grad_norm": 0.9686012268066406,
+      "learning_rate": 0.0002,
+      "loss": 0.5892,
+      "step": 5910
+    },
+    {
+      "epoch": 4.341767510084342,
+      "grad_norm": 0.9033839106559753,
+      "learning_rate": 0.0002,
+      "loss": 0.641,
+      "step": 5920
+    },
+    {
+      "epoch": 4.349101576824349,
+      "grad_norm": 0.8131115436553955,
+      "learning_rate": 0.0002,
+      "loss": 0.6456,
+      "step": 5930
+    },
+    {
+      "epoch": 4.356435643564357,
+      "grad_norm": 0.8942412734031677,
+      "learning_rate": 0.0002,
+      "loss": 0.5826,
+      "step": 5940
+    },
+    {
+      "epoch": 4.363769710304364,
+      "grad_norm": 0.8439112901687622,
+      "learning_rate": 0.0002,
+      "loss": 0.7336,
+      "step": 5950
+    },
+    {
+      "epoch": 4.371103777044371,
+      "grad_norm": 0.9176713228225708,
+      "learning_rate": 0.0002,
+      "loss": 0.6537,
+      "step": 5960
+    },
+    {
+      "epoch": 4.378437843784378,
+      "grad_norm": 0.6799634695053101,
+      "learning_rate": 0.0002,
+      "loss": 0.6792,
+      "step": 5970
+    },
+    {
+      "epoch": 4.385771910524386,
+      "grad_norm": 1.0435824394226074,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 5980
+    },
+    {
+      "epoch": 4.393105977264393,
+      "grad_norm": 0.997937798500061,
+      "learning_rate": 0.0002,
+      "loss": 0.68,
+      "step": 5990
+    },
+    {
+      "epoch": 4.400440044004401,
+      "grad_norm": 1.0308842658996582,
+      "learning_rate": 0.0002,
+      "loss": 0.6604,
+      "step": 6000
+    },
+    {
+      "epoch": 4.407774110744408,
+      "grad_norm": 1.3683775663375854,
+      "learning_rate": 0.0002,
+      "loss": 0.6402,
+      "step": 6010
+    },
+    {
+      "epoch": 4.415108177484415,
+      "grad_norm": 0.7569534182548523,
+      "learning_rate": 0.0002,
+      "loss": 0.7027,
+      "step": 6020
+    },
+    {
+      "epoch": 4.422442244224422,
+      "grad_norm": 1.089978575706482,
+      "learning_rate": 0.0002,
+      "loss": 0.5949,
+      "step": 6030
+    },
+    {
+      "epoch": 4.42977631096443,
+      "grad_norm": 0.7522459626197815,
+      "learning_rate": 0.0002,
+      "loss": 0.6353,
+      "step": 6040
+    },
+    {
+      "epoch": 4.437110377704437,
+      "grad_norm": 0.6709823608398438,
+      "learning_rate": 0.0002,
+      "loss": 0.5852,
+      "step": 6050
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.6992089748382568,
+      "learning_rate": 0.0002,
+      "loss": 0.6718,
+      "step": 6060
+    },
+    {
+      "epoch": 4.451778511184452,
+      "grad_norm": 1.0182931423187256,
+      "learning_rate": 0.0002,
+      "loss": 0.6933,
+      "step": 6070
+    },
+    {
+      "epoch": 4.459112577924459,
+      "grad_norm": 1.0685160160064697,
+      "learning_rate": 0.0002,
+      "loss": 0.6255,
+      "step": 6080
+    },
+    {
+      "epoch": 4.466446644664466,
+      "grad_norm": 0.8295124769210815,
+      "learning_rate": 0.0002,
+      "loss": 0.6086,
+      "step": 6090
+    },
+    {
+      "epoch": 4.473780711404474,
+      "grad_norm": 1.1862998008728027,
+      "learning_rate": 0.0002,
+      "loss": 0.6359,
+      "step": 6100
+    },
+    {
+      "epoch": 4.481114778144481,
+      "grad_norm": 0.7400273084640503,
+      "learning_rate": 0.0002,
+      "loss": 0.638,
+      "step": 6110
+    },
+    {
+      "epoch": 4.488448844884489,
+      "grad_norm": 0.7098417282104492,
+      "learning_rate": 0.0002,
+      "loss": 0.6854,
+      "step": 6120
+    },
+    {
+      "epoch": 4.495782911624496,
+      "grad_norm": 0.9745053648948669,
+      "learning_rate": 0.0002,
+      "loss": 0.6976,
+      "step": 6130
+    },
+    {
+      "epoch": 4.503116978364503,
+      "grad_norm": 0.8638797998428345,
+      "learning_rate": 0.0002,
+      "loss": 0.605,
+      "step": 6140
+    },
+    {
+      "epoch": 4.51045104510451,
+      "grad_norm": 0.8291046619415283,
+      "learning_rate": 0.0002,
+      "loss": 0.6491,
+      "step": 6150
+    },
+    {
+      "epoch": 4.517785111844518,
+      "grad_norm": 1.0301737785339355,
+      "learning_rate": 0.0002,
+      "loss": 0.6457,
+      "step": 6160
+    },
+    {
+      "epoch": 4.525119178584525,
+      "grad_norm": 1.1996512413024902,
+      "learning_rate": 0.0002,
+      "loss": 0.6742,
+      "step": 6170
+    },
+    {
+      "epoch": 4.5324532453245325,
+      "grad_norm": 1.151038408279419,
+      "learning_rate": 0.0002,
+      "loss": 0.6484,
+      "step": 6180
+    },
+    {
+      "epoch": 4.53978731206454,
+      "grad_norm": 0.8385201096534729,
+      "learning_rate": 0.0002,
+      "loss": 0.668,
+      "step": 6190
+    },
+    {
+      "epoch": 4.5471213788045475,
+      "grad_norm": 0.8969188332557678,
+      "learning_rate": 0.0002,
+      "loss": 0.6381,
+      "step": 6200
+    },
+    {
+      "epoch": 4.554455445544555,
+      "grad_norm": 1.60659658908844,
+      "learning_rate": 0.0002,
+      "loss": 0.7141,
+      "step": 6210
+    },
+    {
+      "epoch": 4.5617895122845615,
+      "grad_norm": 0.9356731176376343,
+      "learning_rate": 0.0002,
+      "loss": 0.6388,
+      "step": 6220
+    },
+    {
+      "epoch": 4.569123579024569,
+      "grad_norm": 0.95856773853302,
+      "learning_rate": 0.0002,
+      "loss": 0.7393,
+      "step": 6230
+    },
+    {
+      "epoch": 4.5764576457645765,
+      "grad_norm": 1.1162524223327637,
+      "learning_rate": 0.0002,
+      "loss": 0.6554,
+      "step": 6240
+    },
+    {
+      "epoch": 4.583791712504584,
+      "grad_norm": 0.8809238076210022,
+      "learning_rate": 0.0002,
+      "loss": 0.6012,
+      "step": 6250
+    },
+    {
+      "epoch": 4.591125779244591,
+      "grad_norm": 0.890738844871521,
+      "learning_rate": 0.0002,
+      "loss": 0.648,
+      "step": 6260
+    },
+    {
+      "epoch": 4.598459845984598,
+      "grad_norm": 0.918684720993042,
+      "learning_rate": 0.0002,
+      "loss": 0.6663,
+      "step": 6270
+    },
+    {
+      "epoch": 4.6057939127246055,
+      "grad_norm": 0.8156296610832214,
+      "learning_rate": 0.0002,
+      "loss": 0.5992,
+      "step": 6280
+    },
+    {
+      "epoch": 4.613127979464613,
+      "grad_norm": 1.046634316444397,
+      "learning_rate": 0.0002,
+      "loss": 0.723,
+      "step": 6290
+    },
+    {
+      "epoch": 4.62046204620462,
+      "grad_norm": 0.7725525498390198,
+      "learning_rate": 0.0002,
+      "loss": 0.7023,
+      "step": 6300
+    },
+    {
+      "epoch": 4.627796112944628,
+      "grad_norm": 0.9992046356201172,
+      "learning_rate": 0.0002,
+      "loss": 0.6414,
+      "step": 6310
+    },
+    {
+      "epoch": 4.635130179684635,
+      "grad_norm": 0.8480095267295837,
+      "learning_rate": 0.0002,
+      "loss": 0.6201,
+      "step": 6320
+    },
+    {
+      "epoch": 4.642464246424643,
+      "grad_norm": 0.7061955332756042,
+      "learning_rate": 0.0002,
+      "loss": 0.6869,
+      "step": 6330
+    },
+    {
+      "epoch": 4.649798313164649,
+      "grad_norm": 1.0354212522506714,
+      "learning_rate": 0.0002,
+      "loss": 0.6828,
+      "step": 6340
+    },
+    {
+      "epoch": 4.657132379904657,
+      "grad_norm": 1.0081377029418945,
+      "learning_rate": 0.0002,
+      "loss": 0.6651,
+      "step": 6350
+    },
+    {
+      "epoch": 4.664466446644664,
+      "grad_norm": 1.2904249429702759,
+      "learning_rate": 0.0002,
+      "loss": 0.726,
+      "step": 6360
+    },
+    {
+      "epoch": 4.671800513384672,
+      "grad_norm": 0.9248910546302795,
+      "learning_rate": 0.0002,
+      "loss": 0.7148,
+      "step": 6370
+    },
+    {
+      "epoch": 4.679134580124679,
+      "grad_norm": 0.9907804131507874,
+      "learning_rate": 0.0002,
+      "loss": 0.6961,
+      "step": 6380
+    },
+    {
+      "epoch": 4.686468646864687,
+      "grad_norm": 1.201143741607666,
+      "learning_rate": 0.0002,
+      "loss": 0.6163,
+      "step": 6390
+    },
+    {
+      "epoch": 4.693802713604693,
+      "grad_norm": 0.8709394335746765,
+      "learning_rate": 0.0002,
+      "loss": 0.6762,
+      "step": 6400
+    },
+    {
+      "epoch": 4.701136780344701,
+      "grad_norm": 0.7468608021736145,
+      "learning_rate": 0.0002,
+      "loss": 0.7217,
+      "step": 6410
+    },
+    {
+      "epoch": 4.708470847084708,
+      "grad_norm": 0.8607903718948364,
+      "learning_rate": 0.0002,
+      "loss": 0.6548,
+      "step": 6420
+    },
+    {
+      "epoch": 4.715804913824716,
+      "grad_norm": 0.9840512871742249,
+      "learning_rate": 0.0002,
+      "loss": 0.6449,
+      "step": 6430
+    },
+    {
+      "epoch": 4.723138980564723,
+      "grad_norm": 0.8328204154968262,
+      "learning_rate": 0.0002,
+      "loss": 0.685,
+      "step": 6440
+    },
+    {
+      "epoch": 4.730473047304731,
+      "grad_norm": 0.924505352973938,
+      "learning_rate": 0.0002,
+      "loss": 0.697,
+      "step": 6450
+    },
+    {
+      "epoch": 4.737807114044738,
+      "grad_norm": 0.8897685408592224,
+      "learning_rate": 0.0002,
+      "loss": 0.7422,
+      "step": 6460
+    },
+    {
+      "epoch": 4.745141180784745,
+      "grad_norm": 0.9605024456977844,
+      "learning_rate": 0.0002,
+      "loss": 0.6842,
+      "step": 6470
+    },
+    {
+      "epoch": 4.752475247524752,
+      "grad_norm": 0.8150759935379028,
+      "learning_rate": 0.0002,
+      "loss": 0.6488,
+      "step": 6480
+    },
+    {
+      "epoch": 4.75980931426476,
+      "grad_norm": 0.8128412961959839,
+      "learning_rate": 0.0002,
+      "loss": 0.6606,
+      "step": 6490
+    },
+    {
+      "epoch": 4.767143381004767,
+      "grad_norm": 0.7381404638290405,
+      "learning_rate": 0.0002,
+      "loss": 0.6729,
+      "step": 6500
+    },
+    {
+      "epoch": 4.774477447744775,
+      "grad_norm": 1.0565853118896484,
+      "learning_rate": 0.0002,
+      "loss": 0.6713,
+      "step": 6510
+    },
+    {
+      "epoch": 4.781811514484782,
+      "grad_norm": 0.9298134446144104,
+      "learning_rate": 0.0002,
+      "loss": 0.6496,
+      "step": 6520
+    },
+    {
+      "epoch": 4.789145581224789,
+      "grad_norm": 1.0145525932312012,
+      "learning_rate": 0.0002,
+      "loss": 0.7279,
+      "step": 6530
+    },
+    {
+      "epoch": 4.796479647964796,
+      "grad_norm": 0.92259681224823,
+      "learning_rate": 0.0002,
+      "loss": 0.5986,
+      "step": 6540
+    },
+    {
+      "epoch": 4.803813714704804,
+      "grad_norm": 0.7881024479866028,
+      "learning_rate": 0.0002,
+      "loss": 0.63,
+      "step": 6550
+    },
+    {
+      "epoch": 4.811147781444811,
+      "grad_norm": 1.4935206174850464,
+      "learning_rate": 0.0002,
+      "loss": 0.7134,
+      "step": 6560
+    },
+    {
+      "epoch": 4.818481848184819,
+      "grad_norm": 0.8612369298934937,
+      "learning_rate": 0.0002,
+      "loss": 0.6695,
+      "step": 6570
+    },
+    {
+      "epoch": 4.825815914924826,
+      "grad_norm": 1.0118653774261475,
+      "learning_rate": 0.0002,
+      "loss": 0.779,
+      "step": 6580
+    },
+    {
+      "epoch": 4.833149981664834,
+      "grad_norm": 1.1303809881210327,
+      "learning_rate": 0.0002,
+      "loss": 0.6991,
+      "step": 6590
+    },
+    {
+      "epoch": 4.84048404840484,
+      "grad_norm": 0.9112492203712463,
+      "learning_rate": 0.0002,
+      "loss": 0.7887,
+      "step": 6600
+    },
+    {
+      "epoch": 4.847818115144848,
+      "grad_norm": 0.864762544631958,
+      "learning_rate": 0.0002,
+      "loss": 0.7699,
+      "step": 6610
+    },
+    {
+      "epoch": 4.855152181884855,
+      "grad_norm": 0.9090572595596313,
+      "learning_rate": 0.0002,
+      "loss": 0.7347,
+      "step": 6620
+    },
+    {
+      "epoch": 4.862486248624863,
+      "grad_norm": 1.014953374862671,
+      "learning_rate": 0.0002,
+      "loss": 0.6608,
+      "step": 6630
+    },
+    {
+      "epoch": 4.86982031536487,
+      "grad_norm": 1.0702149868011475,
+      "learning_rate": 0.0002,
+      "loss": 0.6429,
+      "step": 6640
+    },
+    {
+      "epoch": 4.8771543821048775,
+      "grad_norm": 1.002135157585144,
+      "learning_rate": 0.0002,
+      "loss": 0.6943,
+      "step": 6650
+    },
+    {
+      "epoch": 4.884488448844884,
+      "grad_norm": 0.862545907497406,
+      "learning_rate": 0.0002,
+      "loss": 0.7225,
+      "step": 6660
+    },
+    {
+      "epoch": 4.891822515584892,
+      "grad_norm": 0.7302131056785583,
+      "learning_rate": 0.0002,
+      "loss": 0.6206,
+      "step": 6670
+    },
+    {
+      "epoch": 4.899156582324899,
+      "grad_norm": 0.8380730152130127,
+      "learning_rate": 0.0002,
+      "loss": 0.7175,
+      "step": 6680
+    },
+    {
+      "epoch": 4.9064906490649065,
+      "grad_norm": 0.7956018447875977,
+      "learning_rate": 0.0002,
+      "loss": 0.645,
+      "step": 6690
+    },
+    {
+      "epoch": 4.913824715804914,
+      "grad_norm": 0.6717583537101746,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 6700
+    },
+    {
+      "epoch": 4.9211587825449215,
+      "grad_norm": 1.09099280834198,
+      "learning_rate": 0.0002,
+      "loss": 0.6942,
+      "step": 6710
+    },
+    {
+      "epoch": 4.928492849284929,
+      "grad_norm": 0.8589889407157898,
+      "learning_rate": 0.0002,
+      "loss": 0.7533,
+      "step": 6720
+    },
+    {
+      "epoch": 4.9358269160249355,
+      "grad_norm": 1.0046314001083374,
+      "learning_rate": 0.0002,
+      "loss": 0.66,
+      "step": 6730
+    },
+    {
+      "epoch": 4.943160982764943,
+      "grad_norm": 0.8559659123420715,
+      "learning_rate": 0.0002,
+      "loss": 0.6864,
+      "step": 6740
+    },
+    {
+      "epoch": 4.9504950495049505,
+      "grad_norm": 0.8588525652885437,
+      "learning_rate": 0.0002,
+      "loss": 0.6847,
+      "step": 6750
+    },
+    {
+      "epoch": 4.957829116244958,
+      "grad_norm": 0.9192708134651184,
+      "learning_rate": 0.0002,
+      "loss": 0.6428,
+      "step": 6760
+    },
+    {
+      "epoch": 4.965163182984965,
+      "grad_norm": 1.051398754119873,
+      "learning_rate": 0.0002,
+      "loss": 0.6873,
+      "step": 6770
+    },
+    {
+      "epoch": 4.972497249724973,
+      "grad_norm": 0.9111362099647522,
+      "learning_rate": 0.0002,
+      "loss": 0.7249,
+      "step": 6780
+    },
+    {
+      "epoch": 4.9798313164649795,
+      "grad_norm": 0.7305638194084167,
+      "learning_rate": 0.0002,
+      "loss": 0.7613,
+      "step": 6790
+    },
+    {
+      "epoch": 4.987165383204987,
+      "grad_norm": 1.118837594985962,
+      "learning_rate": 0.0002,
+      "loss": 0.6747,
+      "step": 6800
+    },
+    {
+      "epoch": 4.994499449944994,
+      "grad_norm": 0.9075239300727844,
+      "learning_rate": 0.0002,
+      "loss": 0.6412,
+      "step": 6810
+    },
+    {
+      "epoch": 4.999633296662999,
+      "eval_loss": 1.2361247539520264,
+      "eval_runtime": 32.7325,
+      "eval_samples_per_second": 13.167,
+      "eval_steps_per_second": 1.65,
+      "step": 6817
+    },
+    {
+      "epoch": 5.001833516685002,
+      "grad_norm": 1.0541315078735352,
+      "learning_rate": 0.0002,
+      "loss": 0.7091,
+      "step": 6820
+    },
+    {
+      "epoch": 5.009167583425009,
+      "grad_norm": 0.9750140905380249,
+      "learning_rate": 0.0002,
+      "loss": 0.4882,
+      "step": 6830
+    },
+    {
+      "epoch": 5.016501650165017,
+      "grad_norm": 0.931838870048523,
+      "learning_rate": 0.0002,
+      "loss": 0.6022,
+      "step": 6840
+    },
+    {
+      "epoch": 5.023835716905023,
+      "grad_norm": 1.110278844833374,
+      "learning_rate": 0.0002,
+      "loss": 0.5194,
+      "step": 6850
+    },
+    {
+      "epoch": 5.031169783645031,
+      "grad_norm": 1.0670180320739746,
+      "learning_rate": 0.0002,
+      "loss": 0.4676,
+      "step": 6860
+    },
+    {
+      "epoch": 5.038503850385038,
+      "grad_norm": 0.8762092590332031,
+      "learning_rate": 0.0002,
+      "loss": 0.4374,
+      "step": 6870
+    },
+    {
+      "epoch": 5.045837917125046,
+      "grad_norm": 1.1169432401657104,
+      "learning_rate": 0.0002,
+      "loss": 0.505,
+      "step": 6880
+    },
+    {
+      "epoch": 5.053171983865053,
+      "grad_norm": 1.005491018295288,
+      "learning_rate": 0.0002,
+      "loss": 0.5114,
+      "step": 6890
+    },
+    {
+      "epoch": 5.060506050605061,
+      "grad_norm": 1.1751841306686401,
+      "learning_rate": 0.0002,
+      "loss": 0.5221,
+      "step": 6900
+    },
+    {
+      "epoch": 5.067840117345068,
+      "grad_norm": 0.8501367568969727,
+      "learning_rate": 0.0002,
+      "loss": 0.451,
+      "step": 6910
+    },
+    {
+      "epoch": 5.075174184085075,
+      "grad_norm": 0.9795131683349609,
+      "learning_rate": 0.0002,
+      "loss": 0.5292,
+      "step": 6920
+    },
+    {
+      "epoch": 5.082508250825082,
+      "grad_norm": 0.8929879665374756,
+      "learning_rate": 0.0002,
+      "loss": 0.5234,
+      "step": 6930
+    },
+    {
+      "epoch": 5.08984231756509,
+      "grad_norm": 1.0156651735305786,
+      "learning_rate": 0.0002,
+      "loss": 0.5378,
+      "step": 6940
+    },
+    {
+      "epoch": 5.097176384305097,
+      "grad_norm": 1.0974335670471191,
+      "learning_rate": 0.0002,
+      "loss": 0.5241,
+      "step": 6950
+    },
+    {
+      "epoch": 5.104510451045105,
+      "grad_norm": 1.7015666961669922,
+      "learning_rate": 0.0002,
+      "loss": 0.5705,
+      "step": 6960
+    },
+    {
+      "epoch": 5.111844517785112,
+      "grad_norm": 1.0343226194381714,
+      "learning_rate": 0.0002,
+      "loss": 0.523,
+      "step": 6970
+    },
+    {
+      "epoch": 5.119178584525119,
+      "grad_norm": 1.3072983026504517,
+      "learning_rate": 0.0002,
+      "loss": 0.4616,
+      "step": 6980
+    },
+    {
+      "epoch": 5.126512651265126,
+      "grad_norm": 1.038986086845398,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 6990
+    },
+    {
+      "epoch": 5.133846718005134,
+      "grad_norm": 0.8638386130332947,
+      "learning_rate": 0.0002,
+      "loss": 0.4616,
+      "step": 7000
+    },
+    {
+      "epoch": 5.141180784745141,
+      "grad_norm": 0.8326523900032043,
+      "learning_rate": 0.0002,
+      "loss": 0.5294,
+      "step": 7010
+    },
+    {
+      "epoch": 5.148514851485149,
+      "grad_norm": 1.0976895093917847,
+      "learning_rate": 0.0002,
+      "loss": 0.5021,
+      "step": 7020
+    },
+    {
+      "epoch": 5.155848918225156,
+      "grad_norm": 1.0077873468399048,
+      "learning_rate": 0.0002,
+      "loss": 0.4677,
+      "step": 7030
+    },
+    {
+      "epoch": 5.163182984965164,
+      "grad_norm": 1.0662257671356201,
+      "learning_rate": 0.0002,
+      "loss": 0.5262,
+      "step": 7040
+    },
+    {
+      "epoch": 5.17051705170517,
+      "grad_norm": 1.206271767616272,
+      "learning_rate": 0.0002,
+      "loss": 0.5484,
+      "step": 7050
+    },
+    {
+      "epoch": 5.177851118445178,
+      "grad_norm": 1.1990262269973755,
+      "learning_rate": 0.0002,
+      "loss": 0.4817,
+      "step": 7060
+    },
+    {
+      "epoch": 5.185185185185185,
+      "grad_norm": 1.0207163095474243,
+      "learning_rate": 0.0002,
+      "loss": 0.6048,
+      "step": 7070
+    },
+    {
+      "epoch": 5.192519251925193,
+      "grad_norm": 1.2783987522125244,
+      "learning_rate": 0.0002,
+      "loss": 0.4816,
+      "step": 7080
+    },
+    {
+      "epoch": 5.1998533186652,
+      "grad_norm": 1.1592512130737305,
+      "learning_rate": 0.0002,
+      "loss": 0.5322,
+      "step": 7090
+    },
+    {
+      "epoch": 5.2071873854052075,
+      "grad_norm": 1.1053160429000854,
+      "learning_rate": 0.0002,
+      "loss": 0.5472,
+      "step": 7100
+    },
+    {
+      "epoch": 5.214521452145214,
+      "grad_norm": 1.1925510168075562,
+      "learning_rate": 0.0002,
+      "loss": 0.4986,
+      "step": 7110
+    },
+    {
+      "epoch": 5.221855518885222,
+      "grad_norm": 1.0714877843856812,
+      "learning_rate": 0.0002,
+      "loss": 0.5065,
+      "step": 7120
+    },
+    {
+      "epoch": 5.229189585625229,
+      "grad_norm": 0.9451011419296265,
+      "learning_rate": 0.0002,
+      "loss": 0.5209,
+      "step": 7130
+    },
+    {
+      "epoch": 5.2365236523652365,
+      "grad_norm": 1.03838050365448,
+      "learning_rate": 0.0002,
+      "loss": 0.5298,
+      "step": 7140
+    },
+    {
+      "epoch": 5.243857719105244,
+      "grad_norm": 0.9204146265983582,
+      "learning_rate": 0.0002,
+      "loss": 0.4848,
+      "step": 7150
+    },
+    {
+      "epoch": 5.2511917858452515,
+      "grad_norm": 1.0142229795455933,
+      "learning_rate": 0.0002,
+      "loss": 0.5164,
+      "step": 7160
+    },
+    {
+      "epoch": 5.258525852585258,
+      "grad_norm": 1.4432005882263184,
+      "learning_rate": 0.0002,
+      "loss": 0.5092,
+      "step": 7170
+    },
+    {
+      "epoch": 5.2658599193252655,
+      "grad_norm": 1.1239633560180664,
+      "learning_rate": 0.0002,
+      "loss": 0.5133,
+      "step": 7180
+    },
+    {
+      "epoch": 5.273193986065273,
+      "grad_norm": 0.7012821435928345,
+      "learning_rate": 0.0002,
+      "loss": 0.4969,
+      "step": 7190
+    },
+    {
+      "epoch": 5.2805280528052805,
+      "grad_norm": 1.3499128818511963,
+      "learning_rate": 0.0002,
+      "loss": 0.5466,
+      "step": 7200
+    },
+    {
+      "epoch": 5.287862119545288,
+      "grad_norm": 0.9498730897903442,
+      "learning_rate": 0.0002,
+      "loss": 0.5282,
+      "step": 7210
+    },
+    {
+      "epoch": 5.295196186285295,
+      "grad_norm": 0.9552369117736816,
+      "learning_rate": 0.0002,
+      "loss": 0.5051,
+      "step": 7220
+    },
+    {
+      "epoch": 5.302530253025303,
+      "grad_norm": 0.7610348463058472,
+      "learning_rate": 0.0002,
+      "loss": 0.5329,
+      "step": 7230
+    },
+    {
+      "epoch": 5.3098643197653095,
+      "grad_norm": 1.0314512252807617,
+      "learning_rate": 0.0002,
+      "loss": 0.468,
+      "step": 7240
+    },
+    {
+      "epoch": 5.317198386505317,
+      "grad_norm": 1.0534334182739258,
+      "learning_rate": 0.0002,
+      "loss": 0.5367,
+      "step": 7250
+    },
+    {
+      "epoch": 5.324532453245324,
+      "grad_norm": 1.2553406953811646,
+      "learning_rate": 0.0002,
+      "loss": 0.5491,
+      "step": 7260
+    },
+    {
+      "epoch": 5.331866519985332,
+      "grad_norm": 0.7061691880226135,
+      "learning_rate": 0.0002,
+      "loss": 0.5218,
+      "step": 7270
+    },
+    {
+      "epoch": 5.339200586725339,
+      "grad_norm": 0.9652578830718994,
+      "learning_rate": 0.0002,
+      "loss": 0.5625,
+      "step": 7280
+    },
+    {
+      "epoch": 5.346534653465347,
+      "grad_norm": 1.114788293838501,
+      "learning_rate": 0.0002,
+      "loss": 0.5608,
+      "step": 7290
+    },
+    {
+      "epoch": 5.353868720205353,
+      "grad_norm": 1.0940049886703491,
+      "learning_rate": 0.0002,
+      "loss": 0.578,
+      "step": 7300
+    },
+    {
+      "epoch": 5.361202786945361,
+      "grad_norm": 1.0151008367538452,
+      "learning_rate": 0.0002,
+      "loss": 0.5256,
+      "step": 7310
+    },
+    {
+      "epoch": 5.368536853685368,
+      "grad_norm": 1.0369552373886108,
+      "learning_rate": 0.0002,
+      "loss": 0.5377,
+      "step": 7320
+    },
+    {
+      "epoch": 5.375870920425376,
+      "grad_norm": 0.8489866256713867,
+      "learning_rate": 0.0002,
+      "loss": 0.5028,
+      "step": 7330
+    },
+    {
+      "epoch": 5.383204987165383,
+      "grad_norm": 1.1031713485717773,
+      "learning_rate": 0.0002,
+      "loss": 0.5937,
+      "step": 7340
+    },
+    {
+      "epoch": 5.390539053905391,
+      "grad_norm": 0.9094716310501099,
+      "learning_rate": 0.0002,
+      "loss": 0.5355,
+      "step": 7350
+    },
+    {
+      "epoch": 5.397873120645398,
+      "grad_norm": 0.9530431032180786,
+      "learning_rate": 0.0002,
+      "loss": 0.5406,
+      "step": 7360
+    },
+    {
+      "epoch": 5.405207187385405,
+      "grad_norm": 0.9633604884147644,
+      "learning_rate": 0.0002,
+      "loss": 0.529,
+      "step": 7370
+    },
+    {
+      "epoch": 5.412541254125412,
+      "grad_norm": 0.9541662335395813,
+      "learning_rate": 0.0002,
+      "loss": 0.5315,
+      "step": 7380
+    },
+    {
+      "epoch": 5.41987532086542,
+      "grad_norm": 1.0459771156311035,
+      "learning_rate": 0.0002,
+      "loss": 0.6774,
+      "step": 7390
+    },
+    {
+      "epoch": 5.427209387605427,
+      "grad_norm": 1.027388334274292,
+      "learning_rate": 0.0002,
+      "loss": 0.5737,
+      "step": 7400
+    },
+    {
+      "epoch": 5.434543454345435,
+      "grad_norm": 0.7267653346061707,
+      "learning_rate": 0.0002,
+      "loss": 0.556,
+      "step": 7410
+    },
+    {
+      "epoch": 5.441877521085442,
+      "grad_norm": 1.020142674446106,
+      "learning_rate": 0.0002,
+      "loss": 0.4581,
+      "step": 7420
+    },
+    {
+      "epoch": 5.449211587825449,
+      "grad_norm": 1.044754147529602,
+      "learning_rate": 0.0002,
+      "loss": 0.4853,
+      "step": 7430
+    },
+    {
+      "epoch": 5.456545654565456,
+      "grad_norm": 1.5476195812225342,
+      "learning_rate": 0.0002,
+      "loss": 0.5666,
+      "step": 7440
+    },
+    {
+      "epoch": 5.463879721305464,
+      "grad_norm": 0.9879506826400757,
+      "learning_rate": 0.0002,
+      "loss": 0.5302,
+      "step": 7450
+    },
+    {
+      "epoch": 5.471213788045471,
+      "grad_norm": 1.2562980651855469,
+      "learning_rate": 0.0002,
+      "loss": 0.591,
+      "step": 7460
+    },
+    {
+      "epoch": 5.478547854785479,
+      "grad_norm": 1.3051384687423706,
+      "learning_rate": 0.0002,
+      "loss": 0.5188,
+      "step": 7470
+    },
+    {
+      "epoch": 5.485881921525486,
+      "grad_norm": 1.0511597394943237,
+      "learning_rate": 0.0002,
+      "loss": 0.5658,
+      "step": 7480
+    },
+    {
+      "epoch": 5.493215988265494,
+      "grad_norm": 1.0380817651748657,
+      "learning_rate": 0.0002,
+      "loss": 0.6327,
+      "step": 7490
+    },
+    {
+      "epoch": 5.5005500550055,
+      "grad_norm": 1.170274257659912,
+      "learning_rate": 0.0002,
+      "loss": 0.5356,
+      "step": 7500
+    },
+    {
+      "epoch": 5.507884121745508,
+      "grad_norm": 1.3356517553329468,
+      "learning_rate": 0.0002,
+      "loss": 0.5405,
+      "step": 7510
+    },
+    {
+      "epoch": 5.515218188485515,
+      "grad_norm": 1.0727124214172363,
+      "learning_rate": 0.0002,
+      "loss": 0.5305,
+      "step": 7520
+    },
+    {
+      "epoch": 5.522552255225523,
+      "grad_norm": 1.0110199451446533,
+      "learning_rate": 0.0002,
+      "loss": 0.5543,
+      "step": 7530
+    },
+    {
+      "epoch": 5.52988632196553,
+      "grad_norm": 1.3086743354797363,
+      "learning_rate": 0.0002,
+      "loss": 0.5962,
+      "step": 7540
+    },
+    {
+      "epoch": 5.537220388705538,
+      "grad_norm": 1.1904916763305664,
+      "learning_rate": 0.0002,
+      "loss": 0.5512,
+      "step": 7550
+    },
+    {
+      "epoch": 5.544554455445544,
+      "grad_norm": 0.9466280937194824,
+      "learning_rate": 0.0002,
+      "loss": 0.5915,
+      "step": 7560
+    },
+    {
+      "epoch": 5.551888522185552,
+      "grad_norm": 1.1237901449203491,
+      "learning_rate": 0.0002,
+      "loss": 0.5573,
+      "step": 7570
+    },
+    {
+      "epoch": 5.559222588925559,
+      "grad_norm": 0.9590660333633423,
+      "learning_rate": 0.0002,
+      "loss": 0.5383,
+      "step": 7580
+    },
+    {
+      "epoch": 5.566556655665567,
+      "grad_norm": 1.0890778303146362,
+      "learning_rate": 0.0002,
+      "loss": 0.5594,
+      "step": 7590
+    },
+    {
+      "epoch": 5.573890722405574,
+      "grad_norm": 0.7206931114196777,
+      "learning_rate": 0.0002,
+      "loss": 0.5698,
+      "step": 7600
+    },
+    {
+      "epoch": 5.5812247891455815,
+      "grad_norm": 1.2884514331817627,
+      "learning_rate": 0.0002,
+      "loss": 0.5511,
+      "step": 7610
+    },
+    {
+      "epoch": 5.588558855885589,
+      "grad_norm": 0.7798039317131042,
+      "learning_rate": 0.0002,
+      "loss": 0.5279,
+      "step": 7620
+    },
+    {
+      "epoch": 5.595892922625596,
+      "grad_norm": 1.166046142578125,
+      "learning_rate": 0.0002,
+      "loss": 0.4847,
+      "step": 7630
+    },
+    {
+      "epoch": 5.603226989365603,
+      "grad_norm": 1.0150201320648193,
+      "learning_rate": 0.0002,
+      "loss": 0.5821,
+      "step": 7640
+    },
+    {
+      "epoch": 5.6105610561056105,
+      "grad_norm": 1.0449682474136353,
+      "learning_rate": 0.0002,
+      "loss": 0.5296,
+      "step": 7650
+    },
+    {
+      "epoch": 5.617895122845618,
+      "grad_norm": 0.9310530424118042,
+      "learning_rate": 0.0002,
+      "loss": 0.5431,
+      "step": 7660
+    },
+    {
+      "epoch": 5.6252291895856255,
+      "grad_norm": 0.9117933511734009,
+      "learning_rate": 0.0002,
+      "loss": 0.5234,
+      "step": 7670
+    },
+    {
+      "epoch": 5.632563256325633,
+      "grad_norm": 1.1475164890289307,
+      "learning_rate": 0.0002,
+      "loss": 0.5807,
+      "step": 7680
+    },
+    {
+      "epoch": 5.6398973230656395,
+      "grad_norm": 1.066809058189392,
+      "learning_rate": 0.0002,
+      "loss": 0.5816,
+      "step": 7690
+    },
+    {
+      "epoch": 5.647231389805647,
+      "grad_norm": 1.2834991216659546,
+      "learning_rate": 0.0002,
+      "loss": 0.551,
+      "step": 7700
+    },
+    {
+      "epoch": 5.6545654565456545,
+      "grad_norm": 1.2245112657546997,
+      "learning_rate": 0.0002,
+      "loss": 0.5914,
+      "step": 7710
+    },
+    {
+      "epoch": 5.661899523285662,
+      "grad_norm": 1.1424106359481812,
+      "learning_rate": 0.0002,
+      "loss": 0.5552,
+      "step": 7720
+    },
+    {
+      "epoch": 5.669233590025669,
+      "grad_norm": 1.0673892498016357,
+      "learning_rate": 0.0002,
+      "loss": 0.559,
+      "step": 7730
+    },
+    {
+      "epoch": 5.676567656765677,
+      "grad_norm": 1.4312121868133545,
+      "learning_rate": 0.0002,
+      "loss": 0.544,
+      "step": 7740
+    },
+    {
+      "epoch": 5.683901723505684,
+      "grad_norm": 0.9976982474327087,
+      "learning_rate": 0.0002,
+      "loss": 0.5576,
+      "step": 7750
+    },
+    {
+      "epoch": 5.691235790245691,
+      "grad_norm": 0.9464678168296814,
+      "learning_rate": 0.0002,
+      "loss": 0.4855,
+      "step": 7760
+    },
+    {
+      "epoch": 5.698569856985698,
+      "grad_norm": 1.010995626449585,
+      "learning_rate": 0.0002,
+      "loss": 0.5363,
+      "step": 7770
+    },
+    {
+      "epoch": 5.705903923725706,
+      "grad_norm": 1.3787750005722046,
+      "learning_rate": 0.0002,
+      "loss": 0.5873,
+      "step": 7780
+    },
+    {
+      "epoch": 5.713237990465713,
+      "grad_norm": 1.020922303199768,
+      "learning_rate": 0.0002,
+      "loss": 0.6234,
+      "step": 7790
+    },
+    {
+      "epoch": 5.720572057205721,
+      "grad_norm": 0.9748636484146118,
+      "learning_rate": 0.0002,
+      "loss": 0.5337,
+      "step": 7800
+    },
+    {
+      "epoch": 5.727906123945728,
+      "grad_norm": 1.3077744245529175,
+      "learning_rate": 0.0002,
+      "loss": 0.5507,
+      "step": 7810
+    },
+    {
+      "epoch": 5.735240190685735,
+      "grad_norm": 1.4770057201385498,
+      "learning_rate": 0.0002,
+      "loss": 0.558,
+      "step": 7820
+    },
+    {
+      "epoch": 5.742574257425742,
+      "grad_norm": 1.6349090337753296,
+      "learning_rate": 0.0002,
+      "loss": 0.5571,
+      "step": 7830
+    },
+    {
+      "epoch": 5.74990832416575,
+      "grad_norm": 0.9818630814552307,
+      "learning_rate": 0.0002,
+      "loss": 0.5056,
+      "step": 7840
+    },
+    {
+      "epoch": 5.757242390905757,
+      "grad_norm": 0.9659715890884399,
+      "learning_rate": 0.0002,
+      "loss": 0.5495,
+      "step": 7850
+    },
+    {
+      "epoch": 5.764576457645765,
+      "grad_norm": 0.9269950985908508,
+      "learning_rate": 0.0002,
+      "loss": 0.5628,
+      "step": 7860
+    },
+    {
+      "epoch": 5.771910524385772,
+      "grad_norm": 1.0099073648452759,
+      "learning_rate": 0.0002,
+      "loss": 0.5594,
+      "step": 7870
+    },
+    {
+      "epoch": 5.77924459112578,
+      "grad_norm": 0.9123615026473999,
+      "learning_rate": 0.0002,
+      "loss": 0.5912,
+      "step": 7880
+    },
+    {
+      "epoch": 5.786578657865786,
+      "grad_norm": 1.1542246341705322,
+      "learning_rate": 0.0002,
+      "loss": 0.6054,
+      "step": 7890
+    },
+    {
+      "epoch": 5.793912724605794,
+      "grad_norm": 1.0792022943496704,
+      "learning_rate": 0.0002,
+      "loss": 0.5829,
+      "step": 7900
+    },
+    {
+      "epoch": 5.801246791345801,
+      "grad_norm": 0.95615553855896,
+      "learning_rate": 0.0002,
+      "loss": 0.504,
+      "step": 7910
+    },
+    {
+      "epoch": 5.808580858085809,
+      "grad_norm": 1.2471332550048828,
+      "learning_rate": 0.0002,
+      "loss": 0.5918,
+      "step": 7920
+    },
+    {
+      "epoch": 5.815914924825816,
+      "grad_norm": 1.0189851522445679,
+      "learning_rate": 0.0002,
+      "loss": 0.5719,
+      "step": 7930
+    },
+    {
+      "epoch": 5.823248991565823,
+      "grad_norm": 1.3309742212295532,
+      "learning_rate": 0.0002,
+      "loss": 0.5958,
+      "step": 7940
+    },
+    {
+      "epoch": 5.83058305830583,
+      "grad_norm": 1.2930549383163452,
+      "learning_rate": 0.0002,
+      "loss": 0.6255,
+      "step": 7950
+    },
+    {
+      "epoch": 5.837917125045838,
+      "grad_norm": 0.8216308951377869,
+      "learning_rate": 0.0002,
+      "loss": 0.5301,
+      "step": 7960
+    },
+    {
+      "epoch": 5.845251191785845,
+      "grad_norm": 1.1205775737762451,
+      "learning_rate": 0.0002,
+      "loss": 0.5397,
+      "step": 7970
+    },
+    {
+      "epoch": 5.852585258525853,
+      "grad_norm": 0.851298451423645,
+      "learning_rate": 0.0002,
+      "loss": 0.5903,
+      "step": 7980
+    },
+    {
+      "epoch": 5.85991932526586,
+      "grad_norm": 0.8797095417976379,
+      "learning_rate": 0.0002,
+      "loss": 0.5981,
+      "step": 7990
+    },
+    {
+      "epoch": 5.867253392005868,
+      "grad_norm": 1.5784614086151123,
+      "learning_rate": 0.0002,
+      "loss": 0.6106,
+      "step": 8000
+    },
+    {
+      "epoch": 5.874587458745875,
+      "grad_norm": 1.1531187295913696,
+      "learning_rate": 0.0002,
+      "loss": 0.5956,
+      "step": 8010
+    },
+    {
+      "epoch": 5.881921525485882,
+      "grad_norm": 1.2469146251678467,
+      "learning_rate": 0.0002,
+      "loss": 0.6289,
+      "step": 8020
+    },
+    {
+      "epoch": 5.889255592225889,
+      "grad_norm": 1.0784350633621216,
+      "learning_rate": 0.0002,
+      "loss": 0.5827,
+      "step": 8030
+    },
+    {
+      "epoch": 5.896589658965897,
+      "grad_norm": 1.1311599016189575,
+      "learning_rate": 0.0002,
+      "loss": 0.6339,
+      "step": 8040
+    },
+    {
+      "epoch": 5.903923725705904,
+      "grad_norm": 0.9654512405395508,
+      "learning_rate": 0.0002,
+      "loss": 0.5815,
+      "step": 8050
+    },
+    {
+      "epoch": 5.9112577924459115,
+      "grad_norm": 1.3288270235061646,
+      "learning_rate": 0.0002,
+      "loss": 0.6198,
+      "step": 8060
+    },
+    {
+      "epoch": 5.918591859185918,
+      "grad_norm": 1.12800931930542,
+      "learning_rate": 0.0002,
+      "loss": 0.6515,
+      "step": 8070
+    },
+    {
+      "epoch": 5.925925925925926,
+      "grad_norm": 0.9449917674064636,
+      "learning_rate": 0.0002,
+      "loss": 0.5684,
+      "step": 8080
+    },
+    {
+      "epoch": 5.933259992665933,
+      "grad_norm": 1.1532357931137085,
+      "learning_rate": 0.0002,
+      "loss": 0.6063,
+      "step": 8090
+    },
+    {
+      "epoch": 5.9405940594059405,
+      "grad_norm": 1.2211151123046875,
+      "learning_rate": 0.0002,
+      "loss": 0.5318,
+      "step": 8100
+    },
+    {
+      "epoch": 5.947928126145948,
+      "grad_norm": 1.3459105491638184,
+      "learning_rate": 0.0002,
+      "loss": 0.6512,
+      "step": 8110
+    },
+    {
+      "epoch": 5.9552621928859555,
+      "grad_norm": 1.251999855041504,
+      "learning_rate": 0.0002,
+      "loss": 0.5952,
+      "step": 8120
+    },
+    {
+      "epoch": 5.962596259625963,
+      "grad_norm": 1.5682506561279297,
+      "learning_rate": 0.0002,
+      "loss": 0.6203,
+      "step": 8130
+    },
+    {
+      "epoch": 5.9699303263659695,
+      "grad_norm": 0.926075279712677,
+      "learning_rate": 0.0002,
+      "loss": 0.6253,
+      "step": 8140
+    },
+    {
+      "epoch": 5.977264393105977,
+      "grad_norm": 0.9622511863708496,
+      "learning_rate": 0.0002,
+      "loss": 0.5545,
+      "step": 8150
+    },
+    {
+      "epoch": 5.9845984598459845,
+      "grad_norm": 0.9633373618125916,
+      "learning_rate": 0.0002,
+      "loss": 0.5518,
+      "step": 8160
+    },
+    {
+      "epoch": 5.991932526585992,
+      "grad_norm": 0.8960476517677307,
+      "learning_rate": 0.0002,
+      "loss": 0.5831,
+      "step": 8170
+    },
+    {
+      "epoch": 5.999266593325999,
+      "grad_norm": 0.9372805953025818,
+      "learning_rate": 0.0002,
+      "loss": 0.5442,
+      "step": 8180
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.3233846426010132,
+      "eval_runtime": 32.7419,
+      "eval_samples_per_second": 13.164,
+      "eval_steps_per_second": 1.649,
+      "step": 8181
+    },
+    {
+      "epoch": 6.006600660066007,
+      "grad_norm": 1.1900787353515625,
+      "learning_rate": 0.0002,
+      "loss": 0.4644,
+      "step": 8190
+    },
+    {
+      "epoch": 6.013934726806014,
+      "grad_norm": 1.1448326110839844,
+      "learning_rate": 0.0002,
+      "loss": 0.4509,
+      "step": 8200
+    },
+    {
+      "epoch": 6.021268793546021,
+      "grad_norm": 1.1848368644714355,
+      "learning_rate": 0.0002,
+      "loss": 0.3667,
+      "step": 8210
+    },
+    {
+      "epoch": 6.028602860286028,
+      "grad_norm": 1.2315572500228882,
+      "learning_rate": 0.0002,
+      "loss": 0.4315,
+      "step": 8220
+    },
+    {
+      "epoch": 6.035936927026036,
+      "grad_norm": 1.2214244604110718,
+      "learning_rate": 0.0002,
+      "loss": 0.3541,
+      "step": 8230
+    },
+    {
+      "epoch": 6.043270993766043,
+      "grad_norm": 0.9455513954162598,
+      "learning_rate": 0.0002,
+      "loss": 0.4025,
+      "step": 8240
+    },
+    {
+      "epoch": 6.050605060506051,
+      "grad_norm": 0.9574248790740967,
+      "learning_rate": 0.0002,
+      "loss": 0.4448,
+      "step": 8250
+    },
+    {
+      "epoch": 6.057939127246058,
+      "grad_norm": 1.1022400856018066,
+      "learning_rate": 0.0002,
+      "loss": 0.4271,
+      "step": 8260
+    },
+    {
+      "epoch": 6.065273193986065,
+      "grad_norm": 0.9555122256278992,
+      "learning_rate": 0.0002,
+      "loss": 0.3603,
+      "step": 8270
+    },
+    {
+      "epoch": 6.072607260726072,
+      "grad_norm": 1.1956106424331665,
+      "learning_rate": 0.0002,
+      "loss": 0.4324,
+      "step": 8280
+    },
+    {
+      "epoch": 6.07994132746608,
+      "grad_norm": 1.3110876083374023,
+      "learning_rate": 0.0002,
+      "loss": 0.3924,
+      "step": 8290
+    },
+    {
+      "epoch": 6.087275394206087,
+      "grad_norm": 1.1293374300003052,
+      "learning_rate": 0.0002,
+      "loss": 0.3664,
+      "step": 8300
+    },
+    {
+      "epoch": 6.094609460946095,
+      "grad_norm": 0.9176164269447327,
+      "learning_rate": 0.0002,
+      "loss": 0.385,
+      "step": 8310
+    },
+    {
+      "epoch": 6.101943527686102,
+      "grad_norm": 0.9751231670379639,
+      "learning_rate": 0.0002,
+      "loss": 0.4142,
+      "step": 8320
+    },
+    {
+      "epoch": 6.109277594426109,
+      "grad_norm": 1.0536044836044312,
+      "learning_rate": 0.0002,
+      "loss": 0.4356,
+      "step": 8330
+    },
+    {
+      "epoch": 6.116611661166116,
+      "grad_norm": 1.289342999458313,
+      "learning_rate": 0.0002,
+      "loss": 0.409,
+      "step": 8340
+    },
+    {
+      "epoch": 6.123945727906124,
+      "grad_norm": 1.1773661375045776,
+      "learning_rate": 0.0002,
+      "loss": 0.4121,
+      "step": 8350
+    },
+    {
+      "epoch": 6.131279794646131,
+      "grad_norm": 1.2450661659240723,
+      "learning_rate": 0.0002,
+      "loss": 0.4499,
+      "step": 8360
+    },
+    {
+      "epoch": 6.138613861386139,
+      "grad_norm": 1.3965914249420166,
+      "learning_rate": 0.0002,
+      "loss": 0.4467,
+      "step": 8370
+    },
+    {
+      "epoch": 6.145947928126146,
+      "grad_norm": 1.3530808687210083,
+      "learning_rate": 0.0002,
+      "loss": 0.4024,
+      "step": 8380
+    },
+    {
+      "epoch": 6.153281994866154,
+      "grad_norm": 1.296276330947876,
+      "learning_rate": 0.0002,
+      "loss": 0.4658,
+      "step": 8390
+    },
+    {
+      "epoch": 6.16061606160616,
+      "grad_norm": 0.9759053587913513,
+      "learning_rate": 0.0002,
+      "loss": 0.5073,
+      "step": 8400
+    },
+    {
+      "epoch": 6.167950128346168,
+      "grad_norm": 1.2110707759857178,
+      "learning_rate": 0.0002,
+      "loss": 0.4718,
+      "step": 8410
+    },
+    {
+      "epoch": 6.175284195086175,
+      "grad_norm": 1.312226414680481,
+      "learning_rate": 0.0002,
+      "loss": 0.4453,
+      "step": 8420
+    },
+    {
+      "epoch": 6.182618261826183,
+      "grad_norm": 1.1696736812591553,
+      "learning_rate": 0.0002,
+      "loss": 0.4183,
+      "step": 8430
+    },
+    {
+      "epoch": 6.18995232856619,
+      "grad_norm": 1.260304570198059,
+      "learning_rate": 0.0002,
+      "loss": 0.4546,
+      "step": 8440
+    },
+    {
+      "epoch": 6.197286395306198,
+      "grad_norm": 1.472961187362671,
+      "learning_rate": 0.0002,
+      "loss": 0.4137,
+      "step": 8450
+    },
+    {
+      "epoch": 6.204620462046204,
+      "grad_norm": 1.3618475198745728,
+      "learning_rate": 0.0002,
+      "loss": 0.42,
+      "step": 8460
+    },
+    {
+      "epoch": 6.211954528786212,
+      "grad_norm": 1.2544318437576294,
+      "learning_rate": 0.0002,
+      "loss": 0.415,
+      "step": 8470
+    },
+    {
+      "epoch": 6.219288595526219,
+      "grad_norm": 1.205898642539978,
+      "learning_rate": 0.0002,
+      "loss": 0.3907,
+      "step": 8480
+    },
+    {
+      "epoch": 6.226622662266227,
+      "grad_norm": 0.9984724521636963,
+      "learning_rate": 0.0002,
+      "loss": 0.4431,
+      "step": 8490
+    },
+    {
+      "epoch": 6.233956729006234,
+      "grad_norm": 1.3184109926223755,
+      "learning_rate": 0.0002,
+      "loss": 0.4768,
+      "step": 8500
+    },
+    {
+      "epoch": 6.241290795746242,
+      "grad_norm": 1.135520100593567,
+      "learning_rate": 0.0002,
+      "loss": 0.3859,
+      "step": 8510
+    },
+    {
+      "epoch": 6.248624862486249,
+      "grad_norm": 1.4528400897979736,
+      "learning_rate": 0.0002,
+      "loss": 0.4159,
+      "step": 8520
+    },
+    {
+      "epoch": 6.255958929226256,
+      "grad_norm": 1.1222716569900513,
+      "learning_rate": 0.0002,
+      "loss": 0.4347,
+      "step": 8530
+    },
+    {
+      "epoch": 6.263292995966263,
+      "grad_norm": 1.7878046035766602,
+      "learning_rate": 0.0002,
+      "loss": 0.4581,
+      "step": 8540
+    },
+    {
+      "epoch": 6.270627062706271,
+      "grad_norm": 0.9789481163024902,
+      "learning_rate": 0.0002,
+      "loss": 0.4298,
+      "step": 8550
+    },
+    {
+      "epoch": 6.277961129446278,
+      "grad_norm": 1.151977300643921,
+      "learning_rate": 0.0002,
+      "loss": 0.4316,
+      "step": 8560
+    },
+    {
+      "epoch": 6.2852951961862855,
+      "grad_norm": 1.389968752861023,
+      "learning_rate": 0.0002,
+      "loss": 0.428,
+      "step": 8570
+    },
+    {
+      "epoch": 6.292629262926293,
+      "grad_norm": 0.884211003780365,
+      "learning_rate": 0.0002,
+      "loss": 0.3903,
+      "step": 8580
+    },
+    {
+      "epoch": 6.2999633296663,
+      "grad_norm": 1.3604296445846558,
+      "learning_rate": 0.0002,
+      "loss": 0.4611,
+      "step": 8590
+    },
+    {
+      "epoch": 6.307297396406307,
+      "grad_norm": 1.1845694780349731,
+      "learning_rate": 0.0002,
+      "loss": 0.4183,
+      "step": 8600
+    },
+    {
+      "epoch": 6.3146314631463145,
+      "grad_norm": 1.3231550455093384,
+      "learning_rate": 0.0002,
+      "loss": 0.472,
+      "step": 8610
+    },
+    {
+      "epoch": 6.321965529886322,
+      "grad_norm": 0.9546721577644348,
+      "learning_rate": 0.0002,
+      "loss": 0.3922,
+      "step": 8620
+    },
+    {
+      "epoch": 6.3292995966263295,
+      "grad_norm": 1.2329787015914917,
+      "learning_rate": 0.0002,
+      "loss": 0.4395,
+      "step": 8630
+    },
+    {
+      "epoch": 6.336633663366337,
+      "grad_norm": 1.0240199565887451,
+      "learning_rate": 0.0002,
+      "loss": 0.4344,
+      "step": 8640
+    },
+    {
+      "epoch": 6.343967730106344,
+      "grad_norm": 1.1866962909698486,
+      "learning_rate": 0.0002,
+      "loss": 0.4529,
+      "step": 8650
+    },
+    {
+      "epoch": 6.351301796846351,
+      "grad_norm": 1.2819687128067017,
+      "learning_rate": 0.0002,
+      "loss": 0.4575,
+      "step": 8660
+    },
+    {
+      "epoch": 6.3586358635863585,
+      "grad_norm": 0.9654944539070129,
+      "learning_rate": 0.0002,
+      "loss": 0.455,
+      "step": 8670
+    },
+    {
+      "epoch": 6.365969930326366,
+      "grad_norm": 0.9443874955177307,
+      "learning_rate": 0.0002,
+      "loss": 0.4739,
+      "step": 8680
+    },
+    {
+      "epoch": 6.373303997066373,
+      "grad_norm": 1.2914115190505981,
+      "learning_rate": 0.0002,
+      "loss": 0.435,
+      "step": 8690
+    },
+    {
+      "epoch": 6.380638063806381,
+      "grad_norm": 1.4558709859848022,
+      "learning_rate": 0.0002,
+      "loss": 0.4392,
+      "step": 8700
+    },
+    {
+      "epoch": 6.387972130546388,
+      "grad_norm": 1.3255952596664429,
+      "learning_rate": 0.0002,
+      "loss": 0.4398,
+      "step": 8710
+    },
+    {
+      "epoch": 6.395306197286395,
+      "grad_norm": 1.348742961883545,
+      "learning_rate": 0.0002,
+      "loss": 0.4451,
+      "step": 8720
+    },
+    {
+      "epoch": 6.402640264026402,
+      "grad_norm": 1.0096025466918945,
+      "learning_rate": 0.0002,
+      "loss": 0.41,
+      "step": 8730
+    },
+    {
+      "epoch": 6.40997433076641,
+      "grad_norm": 1.1720590591430664,
+      "learning_rate": 0.0002,
+      "loss": 0.4459,
+      "step": 8740
+    },
+    {
+      "epoch": 6.417308397506417,
+      "grad_norm": 1.1803077459335327,
+      "learning_rate": 0.0002,
+      "loss": 0.5059,
+      "step": 8750
+    },
+    {
+      "epoch": 6.424642464246425,
+      "grad_norm": 1.3649998903274536,
+      "learning_rate": 0.0002,
+      "loss": 0.4539,
+      "step": 8760
+    },
+    {
+      "epoch": 6.431976530986432,
+      "grad_norm": 1.1503992080688477,
+      "learning_rate": 0.0002,
+      "loss": 0.4171,
+      "step": 8770
+    },
+    {
+      "epoch": 6.43931059772644,
+      "grad_norm": 1.1537176370620728,
+      "learning_rate": 0.0002,
+      "loss": 0.488,
+      "step": 8780
+    },
+    {
+      "epoch": 6.446644664466446,
+      "grad_norm": 0.9743003845214844,
+      "learning_rate": 0.0002,
+      "loss": 0.4167,
+      "step": 8790
+    },
+    {
+      "epoch": 6.453978731206454,
+      "grad_norm": 0.9097744822502136,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 8800
+    },
+    {
+      "epoch": 6.461312797946461,
+      "grad_norm": 2.0174002647399902,
+      "learning_rate": 0.0002,
+      "loss": 0.4809,
+      "step": 8810
+    },
+    {
+      "epoch": 6.468646864686469,
+      "grad_norm": 1.0809309482574463,
+      "learning_rate": 0.0002,
+      "loss": 0.4879,
+      "step": 8820
+    },
+    {
+      "epoch": 6.475980931426476,
+      "grad_norm": 1.100294828414917,
+      "learning_rate": 0.0002,
+      "loss": 0.4235,
+      "step": 8830
+    },
+    {
+      "epoch": 6.483314998166484,
+      "grad_norm": 1.3707489967346191,
+      "learning_rate": 0.0002,
+      "loss": 0.4251,
+      "step": 8840
+    },
+    {
+      "epoch": 6.49064906490649,
+      "grad_norm": 1.1304761171340942,
+      "learning_rate": 0.0002,
+      "loss": 0.4533,
+      "step": 8850
+    },
+    {
+      "epoch": 6.497983131646498,
+      "grad_norm": 1.2171573638916016,
+      "learning_rate": 0.0002,
+      "loss": 0.4596,
+      "step": 8860
+    },
+    {
+      "epoch": 6.505317198386505,
+      "grad_norm": 1.0452901124954224,
+      "learning_rate": 0.0002,
+      "loss": 0.4694,
+      "step": 8870
+    },
+    {
+      "epoch": 6.512651265126513,
+      "grad_norm": 1.197298526763916,
+      "learning_rate": 0.0002,
+      "loss": 0.4855,
+      "step": 8880
+    },
+    {
+      "epoch": 6.51998533186652,
+      "grad_norm": 0.9179880619049072,
+      "learning_rate": 0.0002,
+      "loss": 0.4167,
+      "step": 8890
+    },
+    {
+      "epoch": 6.527319398606528,
+      "grad_norm": 1.415079951286316,
+      "learning_rate": 0.0002,
+      "loss": 0.445,
+      "step": 8900
+    },
+    {
+      "epoch": 6.534653465346535,
+      "grad_norm": 1.1032487154006958,
+      "learning_rate": 0.0002,
+      "loss": 0.424,
+      "step": 8910
+    },
+    {
+      "epoch": 6.541987532086542,
+      "grad_norm": 1.2295007705688477,
+      "learning_rate": 0.0002,
+      "loss": 0.4496,
+      "step": 8920
+    },
+    {
+      "epoch": 6.549321598826549,
+      "grad_norm": 1.4223219156265259,
+      "learning_rate": 0.0002,
+      "loss": 0.4755,
+      "step": 8930
+    },
+    {
+      "epoch": 6.556655665566557,
+      "grad_norm": 1.2785786390304565,
+      "learning_rate": 0.0002,
+      "loss": 0.4597,
+      "step": 8940
+    },
+    {
+      "epoch": 6.563989732306564,
+      "grad_norm": 1.3514775037765503,
+      "learning_rate": 0.0002,
+      "loss": 0.4651,
+      "step": 8950
+    },
+    {
+      "epoch": 6.571323799046572,
+      "grad_norm": 1.107937216758728,
+      "learning_rate": 0.0002,
+      "loss": 0.4961,
+      "step": 8960
+    },
+    {
+      "epoch": 6.578657865786578,
+      "grad_norm": 1.2839902639389038,
+      "learning_rate": 0.0002,
+      "loss": 0.4954,
+      "step": 8970
+    },
+    {
+      "epoch": 6.585991932526586,
+      "grad_norm": 0.9793244004249573,
+      "learning_rate": 0.0002,
+      "loss": 0.4207,
+      "step": 8980
+    },
+    {
+      "epoch": 6.593325999266593,
+      "grad_norm": 1.3403126001358032,
+      "learning_rate": 0.0002,
+      "loss": 0.4989,
+      "step": 8990
+    },
+    {
+      "epoch": 6.600660066006601,
+      "grad_norm": 1.2612813711166382,
+      "learning_rate": 0.0002,
+      "loss": 0.465,
+      "step": 9000
+    },
+    {
+      "epoch": 6.607994132746608,
+      "grad_norm": 1.4347625970840454,
+      "learning_rate": 0.0002,
+      "loss": 0.4589,
+      "step": 9010
+    },
+    {
+      "epoch": 6.6153281994866155,
+      "grad_norm": 1.225921869277954,
+      "learning_rate": 0.0002,
+      "loss": 0.4864,
+      "step": 9020
+    },
+    {
+      "epoch": 6.622662266226623,
+      "grad_norm": 1.033644676208496,
+      "learning_rate": 0.0002,
+      "loss": 0.4364,
+      "step": 9030
+    },
+    {
+      "epoch": 6.6299963329666305,
+      "grad_norm": 1.1791894435882568,
+      "learning_rate": 0.0002,
+      "loss": 0.4698,
+      "step": 9040
+    },
+    {
+      "epoch": 6.637330399706637,
+      "grad_norm": 1.0968137979507446,
+      "learning_rate": 0.0002,
+      "loss": 0.4908,
+      "step": 9050
+    },
+    {
+      "epoch": 6.6446644664466445,
+      "grad_norm": 1.5639140605926514,
+      "learning_rate": 0.0002,
+      "loss": 0.4346,
+      "step": 9060
+    },
+    {
+      "epoch": 6.651998533186652,
+      "grad_norm": 1.4158905744552612,
+      "learning_rate": 0.0002,
+      "loss": 0.4627,
+      "step": 9070
+    },
+    {
+      "epoch": 6.6593325999266595,
+      "grad_norm": 1.2120254039764404,
+      "learning_rate": 0.0002,
+      "loss": 0.4619,
+      "step": 9080
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.1866531372070312,
+      "learning_rate": 0.0002,
+      "loss": 0.4564,
+      "step": 9090
+    },
+    {
+      "epoch": 6.6740007334066735,
+      "grad_norm": 1.2704026699066162,
+      "learning_rate": 0.0002,
+      "loss": 0.5175,
+      "step": 9100
+    },
+    {
+      "epoch": 6.681334800146681,
+      "grad_norm": 1.1878353357315063,
+      "learning_rate": 0.0002,
+      "loss": 0.4859,
+      "step": 9110
+    },
+    {
+      "epoch": 6.6886688668866885,
+      "grad_norm": 1.193995475769043,
+      "learning_rate": 0.0002,
+      "loss": 0.4657,
+      "step": 9120
+    },
+    {
+      "epoch": 6.696002933626696,
+      "grad_norm": 1.2927545309066772,
+      "learning_rate": 0.0002,
+      "loss": 0.4939,
+      "step": 9130
+    },
+    {
+      "epoch": 6.703337000366703,
+      "grad_norm": 1.0770703554153442,
+      "learning_rate": 0.0002,
+      "loss": 0.4157,
+      "step": 9140
+    },
+    {
+      "epoch": 6.710671067106711,
+      "grad_norm": 1.2200851440429688,
+      "learning_rate": 0.0002,
+      "loss": 0.4571,
+      "step": 9150
+    },
+    {
+      "epoch": 6.718005133846718,
+      "grad_norm": 1.293891191482544,
+      "learning_rate": 0.0002,
+      "loss": 0.4605,
+      "step": 9160
+    },
+    {
+      "epoch": 6.725339200586725,
+      "grad_norm": 1.9376052618026733,
+      "learning_rate": 0.0002,
+      "loss": 0.5328,
+      "step": 9170
+    },
+    {
+      "epoch": 6.732673267326732,
+      "grad_norm": 1.0353254079818726,
+      "learning_rate": 0.0002,
+      "loss": 0.4861,
+      "step": 9180
+    },
+    {
+      "epoch": 6.74000733406674,
+      "grad_norm": 1.1274057626724243,
+      "learning_rate": 0.0002,
+      "loss": 0.5668,
+      "step": 9190
+    },
+    {
+      "epoch": 6.747341400806747,
+      "grad_norm": 1.3344064950942993,
+      "learning_rate": 0.0002,
+      "loss": 0.4486,
+      "step": 9200
+    },
+    {
+      "epoch": 6.754675467546755,
+      "grad_norm": 1.303621768951416,
+      "learning_rate": 0.0002,
+      "loss": 0.49,
+      "step": 9210
+    },
+    {
+      "epoch": 6.762009534286762,
+      "grad_norm": 1.2327780723571777,
+      "learning_rate": 0.0002,
+      "loss": 0.5059,
+      "step": 9220
+    },
+    {
+      "epoch": 6.769343601026769,
+      "grad_norm": 1.3513109683990479,
+      "learning_rate": 0.0002,
+      "loss": 0.486,
+      "step": 9230
+    },
+    {
+      "epoch": 6.776677667766776,
+      "grad_norm": 1.4762850999832153,
+      "learning_rate": 0.0002,
+      "loss": 0.5254,
+      "step": 9240
+    },
+    {
+      "epoch": 6.784011734506784,
+      "grad_norm": 1.0967189073562622,
+      "learning_rate": 0.0002,
+      "loss": 0.4181,
+      "step": 9250
+    },
+    {
+      "epoch": 6.791345801246791,
+      "grad_norm": 0.933936357498169,
+      "learning_rate": 0.0002,
+      "loss": 0.4862,
+      "step": 9260
+    },
+    {
+      "epoch": 6.798679867986799,
+      "grad_norm": 1.065553903579712,
+      "learning_rate": 0.0002,
+      "loss": 0.4667,
+      "step": 9270
+    },
+    {
+      "epoch": 6.806013934726806,
+      "grad_norm": 1.2044163942337036,
+      "learning_rate": 0.0002,
+      "loss": 0.5164,
+      "step": 9280
+    },
+    {
+      "epoch": 6.813348001466814,
+      "grad_norm": 1.404137134552002,
+      "learning_rate": 0.0002,
+      "loss": 0.4648,
+      "step": 9290
+    },
+    {
+      "epoch": 6.82068206820682,
+      "grad_norm": 1.4005582332611084,
+      "learning_rate": 0.0002,
+      "loss": 0.4442,
+      "step": 9300
+    },
+    {
+      "epoch": 6.828016134946828,
+      "grad_norm": 1.1771104335784912,
+      "learning_rate": 0.0002,
+      "loss": 0.459,
+      "step": 9310
+    },
+    {
+      "epoch": 6.835350201686835,
+      "grad_norm": 1.191933035850525,
+      "learning_rate": 0.0002,
+      "loss": 0.5059,
+      "step": 9320
+    },
+    {
+      "epoch": 6.842684268426843,
+      "grad_norm": 1.3395432233810425,
+      "learning_rate": 0.0002,
+      "loss": 0.4733,
+      "step": 9330
+    },
+    {
+      "epoch": 6.85001833516685,
+      "grad_norm": 1.4145503044128418,
+      "learning_rate": 0.0002,
+      "loss": 0.4882,
+      "step": 9340
+    },
+    {
+      "epoch": 6.857352401906858,
+      "grad_norm": 1.1128839254379272,
+      "learning_rate": 0.0002,
+      "loss": 0.4872,
+      "step": 9350
+    },
+    {
+      "epoch": 6.864686468646864,
+      "grad_norm": 1.0771174430847168,
+      "learning_rate": 0.0002,
+      "loss": 0.4909,
+      "step": 9360
+    },
+    {
+      "epoch": 6.872020535386872,
+      "grad_norm": 1.1089814901351929,
+      "learning_rate": 0.0002,
+      "loss": 0.4739,
+      "step": 9370
+    },
+    {
+      "epoch": 6.879354602126879,
+      "grad_norm": 1.078444004058838,
+      "learning_rate": 0.0002,
+      "loss": 0.4854,
+      "step": 9380
+    },
+    {
+      "epoch": 6.886688668866887,
+      "grad_norm": 1.3676636219024658,
+      "learning_rate": 0.0002,
+      "loss": 0.4904,
+      "step": 9390
+    },
+    {
+      "epoch": 6.894022735606894,
+      "grad_norm": 0.8973749876022339,
+      "learning_rate": 0.0002,
+      "loss": 0.4854,
+      "step": 9400
+    },
+    {
+      "epoch": 6.901356802346902,
+      "grad_norm": 1.141552448272705,
+      "learning_rate": 0.0002,
+      "loss": 0.4274,
+      "step": 9410
+    },
+    {
+      "epoch": 6.908690869086909,
+      "grad_norm": 0.8345359563827515,
+      "learning_rate": 0.0002,
+      "loss": 0.4972,
+      "step": 9420
+    },
+    {
+      "epoch": 6.916024935826916,
+      "grad_norm": 1.1602197885513306,
+      "learning_rate": 0.0002,
+      "loss": 0.5218,
+      "step": 9430
+    },
+    {
+      "epoch": 6.923359002566923,
+      "grad_norm": 1.275466799736023,
+      "learning_rate": 0.0002,
+      "loss": 0.4911,
+      "step": 9440
+    },
+    {
+      "epoch": 6.930693069306931,
+      "grad_norm": 0.9186071157455444,
+      "learning_rate": 0.0002,
+      "loss": 0.4904,
+      "step": 9450
+    },
+    {
+      "epoch": 6.938027136046938,
+      "grad_norm": 0.9069198966026306,
+      "learning_rate": 0.0002,
+      "loss": 0.4604,
+      "step": 9460
+    },
+    {
+      "epoch": 6.945361202786946,
+      "grad_norm": 1.2331899404525757,
+      "learning_rate": 0.0002,
+      "loss": 0.4363,
+      "step": 9470
+    },
+    {
+      "epoch": 6.952695269526953,
+      "grad_norm": 0.8685150742530823,
+      "learning_rate": 0.0002,
+      "loss": 0.4815,
+      "step": 9480
+    },
+    {
+      "epoch": 6.96002933626696,
+      "grad_norm": 1.4067939519882202,
+      "learning_rate": 0.0002,
+      "loss": 0.4424,
+      "step": 9490
+    },
+    {
+      "epoch": 6.967363403006967,
+      "grad_norm": 1.1864029169082642,
+      "learning_rate": 0.0002,
+      "loss": 0.5089,
+      "step": 9500
+    },
+    {
+      "epoch": 6.974697469746975,
+      "grad_norm": 1.3697725534439087,
+      "learning_rate": 0.0002,
+      "loss": 0.4906,
+      "step": 9510
+    },
+    {
+      "epoch": 6.982031536486982,
+      "grad_norm": 1.1632893085479736,
+      "learning_rate": 0.0002,
+      "loss": 0.4797,
+      "step": 9520
+    },
+    {
+      "epoch": 6.9893656032269895,
+      "grad_norm": 1.1447268724441528,
+      "learning_rate": 0.0002,
+      "loss": 0.4526,
+      "step": 9530
+    },
+    {
+      "epoch": 6.996699669966997,
+      "grad_norm": 1.5017213821411133,
+      "learning_rate": 0.0002,
+      "loss": 0.4627,
+      "step": 9540
+    },
+    {
+      "epoch": 6.999633296662999,
+      "eval_loss": 1.4178194999694824,
+      "eval_runtime": 32.7488,
+      "eval_samples_per_second": 13.161,
+      "eval_steps_per_second": 1.649,
+      "step": 9544
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10904,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "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": 4.416980889989284e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..2868cff7027115396e695775cacd838522aca295
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-9544/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b12b6f6817632087b5a5e37d744e25312b96e839de5005320b96bc0c2473c41f
+size 5624
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..2868cff7027115396e695775cacd838522aca295
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b12b6f6817632087b5a5e37d744e25312b96e839de5005320b96bc0c2473c41f
+size 5624
diff --git a/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/training_log.jsonl b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..bcd616a1f7d0877989def9df545f8a8fe95fdfb0
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/training_log.jsonl
@@ -0,0 +1,8 @@
+{"epoch": 0.9996332966629996, "step": 1363, "epoch_duration": 1472.4565567970276, "total_accumulated_duration": 1472.4565567970276, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7672.4541015625}, "peak_memory_usage": {"GPU_0": 9688.99365234375}, "avg_memory_reserved": {"GPU_0": 10406.0}, "peak_memory_reserved": {"GPU_0": 10406.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 1.9722, "grad_norm": 0.47521963715553284, "learning_rate": 0.0002, "epoch": 0.007334066740007334, "step": 10}, {"loss": 1.4821, "grad_norm": 0.5395162105560303, "learning_rate": 0.0002, "epoch": 0.014668133480014669, "step": 20}, {"loss": 1.4202, "grad_norm": 0.4305780231952667, "learning_rate": 0.0002, "epoch": 0.022002200220022004, "step": 30}, {"loss": 1.4271, "grad_norm": 0.6938246488571167, "learning_rate": 0.0002, "epoch": 0.029336266960029337, "step": 40}, {"loss": 1.3112, "grad_norm": 1.5133819580078125, "learning_rate": 0.0002, "epoch": 0.03667033370003667, "step": 50}, {"loss": 1.3132, "grad_norm": 0.9173883199691772, "learning_rate": 0.0002, "epoch": 0.04400440044004401, "step": 60}, {"loss": 1.2844, "grad_norm": 0.4619861841201782, "learning_rate": 0.0002, "epoch": 0.05133846718005134, "step": 70}, {"loss": 1.2108, "grad_norm": 0.46118637919425964, "learning_rate": 0.0002, "epoch": 0.058672533920058674, "step": 80}, {"loss": 1.3441, "grad_norm": 0.4468648135662079, "learning_rate": 0.0002, "epoch": 0.066006600660066, "step": 90}, {"loss": 1.1863, "grad_norm": 0.46123769879341125, "learning_rate": 0.0002, "epoch": 0.07334066740007333, "step": 100}, {"loss": 1.2772, "grad_norm": 0.4859139025211334, "learning_rate": 0.0002, "epoch": 0.08067473414008068, "step": 110}, {"loss": 1.2087, "grad_norm": 0.4384922385215759, "learning_rate": 0.0002, "epoch": 0.08800880088008801, "step": 120}, {"loss": 1.2927, "grad_norm": 0.39519360661506653, "learning_rate": 0.0002, "epoch": 0.09534286762009535, "step": 130}, {"loss": 1.2349, "grad_norm": 0.4049859344959259, "learning_rate": 0.0002, "epoch": 0.10267693436010268, "step": 140}, {"loss": 1.293, "grad_norm": 0.4605638086795807, "learning_rate": 0.0002, "epoch": 0.11001100110011001, "step": 150}, {"loss": 1.2659, "grad_norm": 0.4201928377151489, "learning_rate": 0.0002, "epoch": 0.11734506784011735, "step": 160}, {"loss": 1.3961, "grad_norm": 0.5367777347564697, "learning_rate": 0.0002, "epoch": 0.12467913458012468, "step": 170}, {"loss": 1.2481, "grad_norm": 0.41752299666404724, "learning_rate": 0.0002, "epoch": 0.132013201320132, "step": 180}, {"loss": 1.207, "grad_norm": 0.31597763299942017, "learning_rate": 0.0002, "epoch": 0.13934726806013933, "step": 190}, {"loss": 1.2441, "grad_norm": 0.7468788623809814, "learning_rate": 0.0002, "epoch": 0.14668133480014667, "step": 200}, {"loss": 1.199, "grad_norm": 0.3403034508228302, "learning_rate": 0.0002, "epoch": 0.15401540154015403, "step": 210}, {"loss": 1.2439, "grad_norm": 0.34240293502807617, "learning_rate": 0.0002, "epoch": 0.16134946828016136, "step": 220}, {"loss": 1.2022, "grad_norm": 0.356158971786499, "learning_rate": 0.0002, "epoch": 0.1686835350201687, "step": 230}, {"loss": 1.207, "grad_norm": 0.3448857367038727, "learning_rate": 0.0002, "epoch": 0.17601760176017603, "step": 240}, {"loss": 1.2156, "grad_norm": 0.3475699722766876, "learning_rate": 0.0002, "epoch": 0.18335166850018336, "step": 250}, {"loss": 1.1551, "grad_norm": 0.2770358622074127, "learning_rate": 0.0002, "epoch": 0.1906857352401907, "step": 260}, {"loss": 1.2238, "grad_norm": 0.4310270845890045, "learning_rate": 0.0002, "epoch": 0.19801980198019803, "step": 270}, {"loss": 1.2917, "grad_norm": 0.335041880607605, "learning_rate": 0.0002, "epoch": 0.20535386872020536, "step": 280}, {"loss": 1.0959, "grad_norm": 0.3420602083206177, "learning_rate": 0.0002, "epoch": 0.2126879354602127, "step": 290}, {"loss": 1.1232, "grad_norm": 0.325001060962677, "learning_rate": 0.0002, "epoch": 0.22002200220022003, "step": 300}, {"loss": 1.2007, "grad_norm": 0.3027827739715576, "learning_rate": 0.0002, "epoch": 0.22735606894022736, "step": 310}, {"loss": 1.1803, "grad_norm": 0.435550719499588, "learning_rate": 0.0002, "epoch": 0.2346901356802347, "step": 320}, {"loss": 1.2045, "grad_norm": 0.3884522616863251, "learning_rate": 0.0002, "epoch": 0.24202420242024203, "step": 330}, {"loss": 1.2481, "grad_norm": 0.7736002206802368, "learning_rate": 0.0002, "epoch": 0.24935826916024936, "step": 340}, {"loss": 1.3606, "grad_norm": 0.35052821040153503, "learning_rate": 0.0002, "epoch": 0.2566923359002567, "step": 350}, {"loss": 1.2129, "grad_norm": 0.3311890959739685, "learning_rate": 0.0002, "epoch": 0.264026402640264, "step": 360}, {"loss": 1.2219, "grad_norm": 0.7473500370979309, "learning_rate": 0.0002, "epoch": 0.27136046938027136, "step": 370}, {"loss": 1.2712, "grad_norm": 0.3681875765323639, "learning_rate": 0.0002, "epoch": 0.27869453612027867, "step": 380}, {"loss": 1.2258, "grad_norm": 0.3764737844467163, "learning_rate": 0.0002, "epoch": 0.28602860286028603, "step": 390}, {"loss": 1.1917, "grad_norm": 0.4243989586830139, "learning_rate": 0.0002, "epoch": 0.29336266960029334, "step": 400}, {"loss": 1.199, "grad_norm": 0.2658531963825226, "learning_rate": 0.0002, "epoch": 0.3006967363403007, "step": 410}, {"loss": 1.1622, "grad_norm": 0.3436793386936188, "learning_rate": 0.0002, "epoch": 0.30803080308030806, "step": 420}, {"loss": 1.2953, "grad_norm": 0.5101129412651062, "learning_rate": 0.0002, "epoch": 0.31536486982031536, "step": 430}, {"loss": 1.1557, "grad_norm": 0.3319750726222992, "learning_rate": 0.0002, "epoch": 0.3226989365603227, "step": 440}, {"loss": 1.1804, "grad_norm": 0.385148286819458, "learning_rate": 0.0002, "epoch": 0.33003300330033003, "step": 450}, {"loss": 1.1808, "grad_norm": 0.3477935791015625, "learning_rate": 0.0002, "epoch": 0.3373670700403374, "step": 460}, {"loss": 1.1877, "grad_norm": 0.29748716950416565, "learning_rate": 0.0002, "epoch": 0.3447011367803447, "step": 470}, {"loss": 1.19, "grad_norm": 0.34083324670791626, "learning_rate": 0.0002, "epoch": 0.35203520352035206, "step": 480}, {"loss": 1.2, "grad_norm": 0.36904552578926086, "learning_rate": 0.0002, "epoch": 0.35936927026035936, "step": 490}, {"loss": 1.2223, "grad_norm": 0.315483033657074, "learning_rate": 0.0002, "epoch": 0.3667033370003667, "step": 500}, {"loss": 1.1461, "grad_norm": 0.44897955656051636, "learning_rate": 0.0002, "epoch": 0.37403740374037403, "step": 510}, {"loss": 1.3035, "grad_norm": 0.3160701394081116, "learning_rate": 0.0002, "epoch": 0.3813714704803814, "step": 520}, {"loss": 1.3197, "grad_norm": 0.29584741592407227, "learning_rate": 0.0002, "epoch": 0.3887055372203887, "step": 530}, {"loss": 1.2983, "grad_norm": 0.5430002808570862, "learning_rate": 0.0002, "epoch": 0.39603960396039606, "step": 540}, {"loss": 1.2459, "grad_norm": 0.2908070683479309, "learning_rate": 0.0002, "epoch": 0.40337367070040336, "step": 550}, {"loss": 1.2384, "grad_norm": 0.35066530108451843, "learning_rate": 0.0002, "epoch": 0.4107077374404107, "step": 560}, {"loss": 1.1784, "grad_norm": 0.37588003277778625, "learning_rate": 0.0002, "epoch": 0.41804180418041803, "step": 570}, {"loss": 1.2334, "grad_norm": 0.3112126886844635, "learning_rate": 0.0002, "epoch": 0.4253758709204254, "step": 580}, {"loss": 1.1439, "grad_norm": 0.35577139258384705, "learning_rate": 0.0002, "epoch": 0.4327099376604327, "step": 590}, {"loss": 1.184, "grad_norm": 0.31706422567367554, "learning_rate": 0.0002, "epoch": 0.44004400440044006, "step": 600}, {"loss": 1.2081, "grad_norm": 0.3249092102050781, "learning_rate": 0.0002, "epoch": 0.44737807114044736, "step": 610}, {"loss": 1.0824, "grad_norm": 0.3842705488204956, "learning_rate": 0.0002, "epoch": 0.4547121378804547, "step": 620}, {"loss": 1.2257, "grad_norm": 0.390991747379303, "learning_rate": 0.0002, "epoch": 0.46204620462046203, "step": 630}, {"loss": 1.1954, "grad_norm": 0.27532413601875305, "learning_rate": 0.0002, "epoch": 0.4693802713604694, "step": 640}, {"loss": 1.1058, "grad_norm": 0.31412816047668457, "learning_rate": 0.0002, "epoch": 0.4767143381004767, "step": 650}, {"loss": 1.1312, "grad_norm": 0.32117101550102234, "learning_rate": 0.0002, "epoch": 0.48404840484048406, "step": 660}, {"loss": 1.2423, "grad_norm": 0.3810010254383087, "learning_rate": 0.0002, "epoch": 0.49138247158049136, "step": 670}, {"loss": 1.1978, "grad_norm": 0.36289164423942566, "learning_rate": 0.0002, "epoch": 0.4987165383204987, "step": 680}, {"loss": 1.2034, "grad_norm": 0.34458720684051514, "learning_rate": 0.0002, "epoch": 0.506050605060506, "step": 690}, {"loss": 1.1756, "grad_norm": 0.32844600081443787, "learning_rate": 0.0002, "epoch": 0.5133846718005134, "step": 700}, {"loss": 1.0807, "grad_norm": 0.3144175708293915, "learning_rate": 0.0002, "epoch": 0.5207187385405208, "step": 710}, {"loss": 1.1952, "grad_norm": 0.3898887634277344, "learning_rate": 0.0002, "epoch": 0.528052805280528, "step": 720}, {"loss": 1.1244, "grad_norm": 1.3220758438110352, "learning_rate": 0.0002, "epoch": 0.5353868720205354, "step": 730}, {"loss": 1.227, "grad_norm": 0.3635874390602112, "learning_rate": 0.0002, "epoch": 0.5427209387605427, "step": 740}, {"loss": 1.2169, "grad_norm": 0.3138217628002167, "learning_rate": 0.0002, "epoch": 0.5500550055005501, "step": 750}, {"loss": 1.1516, "grad_norm": 0.4063207805156708, "learning_rate": 0.0002, "epoch": 0.5573890722405573, "step": 760}, {"loss": 1.1954, "grad_norm": 0.3926219940185547, "learning_rate": 0.0002, "epoch": 0.5647231389805647, "step": 770}, {"loss": 1.1726, "grad_norm": 0.31954652070999146, "learning_rate": 0.0002, "epoch": 0.5720572057205721, "step": 780}, {"loss": 1.2977, "grad_norm": 0.4248711168766022, "learning_rate": 0.0002, "epoch": 0.5793912724605794, "step": 790}, {"loss": 1.1728, "grad_norm": 0.643004834651947, "learning_rate": 0.0002, "epoch": 0.5867253392005867, "step": 800}, {"loss": 1.1793, "grad_norm": 0.3479592800140381, "learning_rate": 0.0002, "epoch": 0.594059405940594, "step": 810}, {"loss": 1.2426, "grad_norm": 0.4684754014015198, "learning_rate": 0.0002, "epoch": 0.6013934726806014, "step": 820}, {"loss": 1.2002, "grad_norm": 0.3739790916442871, "learning_rate": 0.0002, "epoch": 0.6087275394206088, "step": 830}, {"loss": 1.2139, "grad_norm": 0.40884748101234436, "learning_rate": 0.0002, "epoch": 0.6160616061606161, "step": 840}, {"loss": 1.1557, "grad_norm": 0.9722164273262024, "learning_rate": 0.0002, "epoch": 0.6233956729006234, "step": 850}, {"loss": 1.3069, "grad_norm": 0.42992347478866577, "learning_rate": 0.0002, "epoch": 0.6307297396406307, "step": 860}, {"loss": 1.1339, "grad_norm": 0.36654195189476013, "learning_rate": 0.0002, "epoch": 0.6380638063806381, "step": 870}, {"loss": 1.1932, "grad_norm": 0.4113832116127014, "learning_rate": 0.0002, "epoch": 0.6453978731206454, "step": 880}, {"loss": 1.2163, "grad_norm": 0.2948838770389557, "learning_rate": 0.0002, "epoch": 0.6527319398606527, "step": 890}, {"loss": 1.1081, "grad_norm": 0.38330280780792236, "learning_rate": 0.0002, "epoch": 0.6600660066006601, "step": 900}, {"loss": 1.1342, "grad_norm": 0.4428867697715759, "learning_rate": 0.0002, "epoch": 0.6674000733406674, "step": 910}, {"loss": 1.1021, "grad_norm": 0.23659265041351318, "learning_rate": 0.0002, "epoch": 0.6747341400806748, "step": 920}, {"loss": 1.1226, "grad_norm": 0.323685884475708, "learning_rate": 0.0002, "epoch": 0.682068206820682, "step": 930}, {"loss": 1.0853, "grad_norm": 0.39157727360725403, "learning_rate": 0.0002, "epoch": 0.6894022735606894, "step": 940}, {"loss": 1.1435, "grad_norm": 0.27189481258392334, "learning_rate": 0.0002, "epoch": 0.6967363403006968, "step": 950}, {"loss": 1.1033, "grad_norm": 0.529883861541748, "learning_rate": 0.0002, "epoch": 0.7040704070407041, "step": 960}, {"loss": 1.139, "grad_norm": 0.34758689999580383, "learning_rate": 0.0002, "epoch": 0.7114044737807114, "step": 970}, {"loss": 1.2197, "grad_norm": 0.831749439239502, "learning_rate": 0.0002, "epoch": 0.7187385405207187, "step": 980}, {"loss": 1.158, "grad_norm": 0.4438304007053375, "learning_rate": 0.0002, "epoch": 0.7260726072607261, "step": 990}, {"loss": 1.1021, "grad_norm": 0.33840006589889526, "learning_rate": 0.0002, "epoch": 0.7334066740007334, "step": 1000}, {"loss": 1.254, "grad_norm": 0.3454797863960266, "learning_rate": 0.0002, "epoch": 0.7407407407407407, "step": 1010}, {"loss": 1.106, "grad_norm": 0.38999441266059875, "learning_rate": 0.0002, "epoch": 0.7480748074807481, "step": 1020}, {"loss": 1.1428, "grad_norm": 0.2829911708831787, "learning_rate": 0.0002, "epoch": 0.7554088742207554, "step": 1030}, {"loss": 1.2123, "grad_norm": 0.36918163299560547, "learning_rate": 0.0002, "epoch": 0.7627429409607628, "step": 1040}, {"loss": 1.3028, "grad_norm": 0.3415680229663849, "learning_rate": 0.0002, "epoch": 0.77007700770077, "step": 1050}, {"loss": 1.1939, "grad_norm": 0.2974182963371277, "learning_rate": 0.0002, "epoch": 0.7774110744407774, "step": 1060}, {"loss": 1.194, "grad_norm": 0.3880919814109802, "learning_rate": 0.0002, "epoch": 0.7847451411807848, "step": 1070}, {"loss": 1.1095, "grad_norm": 0.33503302931785583, "learning_rate": 0.0002, "epoch": 0.7920792079207921, "step": 1080}, {"loss": 1.2111, "grad_norm": 0.3728407025337219, "learning_rate": 0.0002, "epoch": 0.7994132746607994, "step": 1090}, {"loss": 1.0835, "grad_norm": 0.3509373664855957, "learning_rate": 0.0002, "epoch": 0.8067473414008067, "step": 1100}, {"loss": 1.2661, "grad_norm": 0.42228564620018005, "learning_rate": 0.0002, "epoch": 0.8140814081408141, "step": 1110}, {"loss": 1.1788, "grad_norm": 0.313467800617218, "learning_rate": 0.0002, "epoch": 0.8214154748808215, "step": 1120}, {"loss": 1.1971, "grad_norm": 0.3378850817680359, "learning_rate": 0.0002, "epoch": 0.8287495416208287, "step": 1130}, {"loss": 1.1238, "grad_norm": 0.43200382590293884, "learning_rate": 0.0002, "epoch": 0.8360836083608361, "step": 1140}, {"loss": 1.3203, "grad_norm": 0.3309599459171295, "learning_rate": 0.0002, "epoch": 0.8434176751008434, "step": 1150}, {"loss": 1.1062, "grad_norm": 0.3526846170425415, "learning_rate": 0.0002, "epoch": 0.8507517418408508, "step": 1160}, {"loss": 1.0851, "grad_norm": 1.2722247838974, "learning_rate": 0.0002, "epoch": 0.858085808580858, "step": 1170}, {"loss": 1.0785, "grad_norm": 0.34142059087753296, "learning_rate": 0.0002, "epoch": 0.8654198753208654, "step": 1180}, {"loss": 1.2187, "grad_norm": 0.3805823028087616, "learning_rate": 0.0002, "epoch": 0.8727539420608728, "step": 1190}, {"loss": 1.1215, "grad_norm": 0.3931232690811157, "learning_rate": 0.0002, "epoch": 0.8800880088008801, "step": 1200}, {"loss": 1.0948, "grad_norm": 0.2937372624874115, "learning_rate": 0.0002, "epoch": 0.8874220755408874, "step": 1210}, {"loss": 1.1228, "grad_norm": 0.3757196366786957, "learning_rate": 0.0002, "epoch": 0.8947561422808947, "step": 1220}, {"loss": 1.1222, "grad_norm": 0.3502705991268158, "learning_rate": 0.0002, "epoch": 0.9020902090209021, "step": 1230}, {"loss": 1.2242, "grad_norm": 0.32758915424346924, "learning_rate": 0.0002, "epoch": 0.9094242757609095, "step": 1240}, {"loss": 1.215, "grad_norm": 0.37199416756629944, "learning_rate": 0.0002, "epoch": 0.9167583425009168, "step": 1250}, {"loss": 1.1225, "grad_norm": 0.3551490604877472, "learning_rate": 0.0002, "epoch": 0.9240924092409241, "step": 1260}, {"loss": 1.1966, "grad_norm": 0.2859550714492798, "learning_rate": 0.0002, "epoch": 0.9314264759809314, "step": 1270}, {"loss": 1.2186, "grad_norm": 0.427990585565567, "learning_rate": 0.0002, "epoch": 0.9387605427209388, "step": 1280}, {"loss": 1.2848, "grad_norm": 0.33717992901802063, "learning_rate": 0.0002, "epoch": 0.9460946094609461, "step": 1290}, {"loss": 1.1656, "grad_norm": 0.30225634574890137, "learning_rate": 0.0002, "epoch": 0.9534286762009534, "step": 1300}, {"loss": 1.2404, "grad_norm": 0.385821133852005, "learning_rate": 0.0002, "epoch": 0.9607627429409608, "step": 1310}, {"loss": 1.1932, "grad_norm": 0.35278066992759705, "learning_rate": 0.0002, "epoch": 0.9680968096809681, "step": 1320}, {"loss": 1.1071, "grad_norm": 0.49987098574638367, "learning_rate": 0.0002, "epoch": 0.9754308764209755, "step": 1330}, {"loss": 1.2259, "grad_norm": 0.3842747211456299, "learning_rate": 0.0002, "epoch": 0.9827649431609827, "step": 1340}, {"loss": 1.0862, "grad_norm": 0.6274653673171997, "learning_rate": 0.0002, "epoch": 0.9900990099009901, "step": 1350}, {"loss": 1.124, "grad_norm": 0.5239808559417725, "learning_rate": 0.0002, "epoch": 0.9974330766409975, "step": 1360}]}
+{"epoch": 2.0, "step": 2727, "epoch_duration": 1469.6491115093231, "total_accumulated_duration": 2942.1056683063507, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7568.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-1363", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 1.9722, "grad_norm": 0.47521963715553284, "learning_rate": 0.0002, "epoch": 0.007334066740007334, "step": 10}, {"loss": 1.4821, "grad_norm": 0.5395162105560303, "learning_rate": 0.0002, "epoch": 0.014668133480014669, "step": 20}, {"loss": 1.4202, "grad_norm": 0.4305780231952667, "learning_rate": 0.0002, "epoch": 0.022002200220022004, "step": 30}, {"loss": 1.4271, "grad_norm": 0.6938246488571167, "learning_rate": 0.0002, "epoch": 0.029336266960029337, "step": 40}, {"loss": 1.3112, "grad_norm": 1.5133819580078125, "learning_rate": 0.0002, "epoch": 0.03667033370003667, "step": 50}, {"loss": 1.3132, "grad_norm": 0.9173883199691772, "learning_rate": 0.0002, "epoch": 0.04400440044004401, "step": 60}, {"loss": 1.2844, "grad_norm": 0.4619861841201782, "learning_rate": 0.0002, "epoch": 0.05133846718005134, "step": 70}, {"loss": 1.2108, "grad_norm": 0.46118637919425964, "learning_rate": 0.0002, "epoch": 0.058672533920058674, "step": 80}, {"loss": 1.3441, "grad_norm": 0.4468648135662079, "learning_rate": 0.0002, "epoch": 0.066006600660066, "step": 90}, {"loss": 1.1863, "grad_norm": 0.46123769879341125, "learning_rate": 0.0002, "epoch": 0.07334066740007333, "step": 100}, {"loss": 1.2772, "grad_norm": 0.4859139025211334, "learning_rate": 0.0002, "epoch": 0.08067473414008068, "step": 110}, {"loss": 1.2087, "grad_norm": 0.4384922385215759, "learning_rate": 0.0002, "epoch": 0.08800880088008801, "step": 120}, {"loss": 1.2927, "grad_norm": 0.39519360661506653, "learning_rate": 0.0002, "epoch": 0.09534286762009535, "step": 130}, {"loss": 1.2349, "grad_norm": 0.4049859344959259, "learning_rate": 0.0002, "epoch": 0.10267693436010268, "step": 140}, {"loss": 1.293, "grad_norm": 0.4605638086795807, "learning_rate": 0.0002, "epoch": 0.11001100110011001, "step": 150}, {"loss": 1.2659, "grad_norm": 0.4201928377151489, "learning_rate": 0.0002, "epoch": 0.11734506784011735, "step": 160}, {"loss": 1.3961, "grad_norm": 0.5367777347564697, "learning_rate": 0.0002, "epoch": 0.12467913458012468, "step": 170}, {"loss": 1.2481, "grad_norm": 0.41752299666404724, "learning_rate": 0.0002, "epoch": 0.132013201320132, "step": 180}, {"loss": 1.207, "grad_norm": 0.31597763299942017, "learning_rate": 0.0002, "epoch": 0.13934726806013933, "step": 190}, {"loss": 1.2441, "grad_norm": 0.7468788623809814, "learning_rate": 0.0002, "epoch": 0.14668133480014667, "step": 200}, {"loss": 1.199, "grad_norm": 0.3403034508228302, "learning_rate": 0.0002, "epoch": 0.15401540154015403, "step": 210}, {"loss": 1.2439, "grad_norm": 0.34240293502807617, "learning_rate": 0.0002, "epoch": 0.16134946828016136, "step": 220}, {"loss": 1.2022, "grad_norm": 0.356158971786499, "learning_rate": 0.0002, "epoch": 0.1686835350201687, "step": 230}, {"loss": 1.207, "grad_norm": 0.3448857367038727, "learning_rate": 0.0002, "epoch": 0.17601760176017603, "step": 240}, {"loss": 1.2156, "grad_norm": 0.3475699722766876, "learning_rate": 0.0002, "epoch": 0.18335166850018336, "step": 250}, {"loss": 1.1551, "grad_norm": 0.2770358622074127, "learning_rate": 0.0002, "epoch": 0.1906857352401907, "step": 260}, {"loss": 1.2238, "grad_norm": 0.4310270845890045, "learning_rate": 0.0002, "epoch": 0.19801980198019803, "step": 270}, {"loss": 1.2917, "grad_norm": 0.335041880607605, "learning_rate": 0.0002, "epoch": 0.20535386872020536, "step": 280}, {"loss": 1.0959, "grad_norm": 0.3420602083206177, "learning_rate": 0.0002, "epoch": 0.2126879354602127, "step": 290}, {"loss": 1.1232, "grad_norm": 0.325001060962677, "learning_rate": 0.0002, "epoch": 0.22002200220022003, "step": 300}, {"loss": 1.2007, "grad_norm": 0.3027827739715576, "learning_rate": 0.0002, "epoch": 0.22735606894022736, "step": 310}, {"loss": 1.1803, "grad_norm": 0.435550719499588, "learning_rate": 0.0002, "epoch": 0.2346901356802347, "step": 320}, {"loss": 1.2045, "grad_norm": 0.3884522616863251, "learning_rate": 0.0002, "epoch": 0.24202420242024203, "step": 330}, {"loss": 1.2481, "grad_norm": 0.7736002206802368, "learning_rate": 0.0002, "epoch": 0.24935826916024936, "step": 340}, {"loss": 1.3606, "grad_norm": 0.35052821040153503, "learning_rate": 0.0002, "epoch": 0.2566923359002567, "step": 350}, {"loss": 1.2129, "grad_norm": 0.3311890959739685, "learning_rate": 0.0002, "epoch": 0.264026402640264, "step": 360}, {"loss": 1.2219, "grad_norm": 0.7473500370979309, "learning_rate": 0.0002, "epoch": 0.27136046938027136, "step": 370}, {"loss": 1.2712, "grad_norm": 0.3681875765323639, "learning_rate": 0.0002, "epoch": 0.27869453612027867, "step": 380}, {"loss": 1.2258, "grad_norm": 0.3764737844467163, "learning_rate": 0.0002, "epoch": 0.28602860286028603, "step": 390}, {"loss": 1.1917, "grad_norm": 0.4243989586830139, "learning_rate": 0.0002, "epoch": 0.29336266960029334, "step": 400}, {"loss": 1.199, "grad_norm": 0.2658531963825226, "learning_rate": 0.0002, "epoch": 0.3006967363403007, "step": 410}, {"loss": 1.1622, "grad_norm": 0.3436793386936188, "learning_rate": 0.0002, "epoch": 0.30803080308030806, "step": 420}, {"loss": 1.2953, "grad_norm": 0.5101129412651062, "learning_rate": 0.0002, "epoch": 0.31536486982031536, "step": 430}, {"loss": 1.1557, "grad_norm": 0.3319750726222992, "learning_rate": 0.0002, "epoch": 0.3226989365603227, "step": 440}, {"loss": 1.1804, "grad_norm": 0.385148286819458, "learning_rate": 0.0002, "epoch": 0.33003300330033003, "step": 450}, {"loss": 1.1808, "grad_norm": 0.3477935791015625, "learning_rate": 0.0002, "epoch": 0.3373670700403374, "step": 460}, {"loss": 1.1877, "grad_norm": 0.29748716950416565, "learning_rate": 0.0002, "epoch": 0.3447011367803447, "step": 470}, {"loss": 1.19, "grad_norm": 0.34083324670791626, "learning_rate": 0.0002, "epoch": 0.35203520352035206, "step": 480}, {"loss": 1.2, "grad_norm": 0.36904552578926086, "learning_rate": 0.0002, "epoch": 0.35936927026035936, "step": 490}, {"loss": 1.2223, "grad_norm": 0.315483033657074, "learning_rate": 0.0002, "epoch": 0.3667033370003667, "step": 500}, {"loss": 1.1461, "grad_norm": 0.44897955656051636, "learning_rate": 0.0002, "epoch": 0.37403740374037403, "step": 510}, {"loss": 1.3035, "grad_norm": 0.3160701394081116, "learning_rate": 0.0002, "epoch": 0.3813714704803814, "step": 520}, {"loss": 1.3197, "grad_norm": 0.29584741592407227, "learning_rate": 0.0002, "epoch": 0.3887055372203887, "step": 530}, {"loss": 1.2983, "grad_norm": 0.5430002808570862, "learning_rate": 0.0002, "epoch": 0.39603960396039606, "step": 540}, {"loss": 1.2459, "grad_norm": 0.2908070683479309, "learning_rate": 0.0002, "epoch": 0.40337367070040336, "step": 550}, {"loss": 1.2384, "grad_norm": 0.35066530108451843, "learning_rate": 0.0002, "epoch": 0.4107077374404107, "step": 560}, {"loss": 1.1784, "grad_norm": 0.37588003277778625, "learning_rate": 0.0002, "epoch": 0.41804180418041803, "step": 570}, {"loss": 1.2334, "grad_norm": 0.3112126886844635, "learning_rate": 0.0002, "epoch": 0.4253758709204254, "step": 580}, {"loss": 1.1439, "grad_norm": 0.35577139258384705, "learning_rate": 0.0002, "epoch": 0.4327099376604327, "step": 590}, {"loss": 1.184, "grad_norm": 0.31706422567367554, "learning_rate": 0.0002, "epoch": 0.44004400440044006, "step": 600}, {"loss": 1.2081, "grad_norm": 0.3249092102050781, "learning_rate": 0.0002, "epoch": 0.44737807114044736, "step": 610}, {"loss": 1.0824, "grad_norm": 0.3842705488204956, "learning_rate": 0.0002, "epoch": 0.4547121378804547, "step": 620}, {"loss": 1.2257, "grad_norm": 0.390991747379303, "learning_rate": 0.0002, "epoch": 0.46204620462046203, "step": 630}, {"loss": 1.1954, "grad_norm": 0.27532413601875305, "learning_rate": 0.0002, "epoch": 0.4693802713604694, "step": 640}, {"loss": 1.1058, "grad_norm": 0.31412816047668457, "learning_rate": 0.0002, "epoch": 0.4767143381004767, "step": 650}, {"loss": 1.1312, "grad_norm": 0.32117101550102234, "learning_rate": 0.0002, "epoch": 0.48404840484048406, "step": 660}, {"loss": 1.2423, "grad_norm": 0.3810010254383087, "learning_rate": 0.0002, "epoch": 0.49138247158049136, "step": 670}, {"loss": 1.1978, "grad_norm": 0.36289164423942566, "learning_rate": 0.0002, "epoch": 0.4987165383204987, "step": 680}, {"loss": 1.2034, "grad_norm": 0.34458720684051514, "learning_rate": 0.0002, "epoch": 0.506050605060506, "step": 690}, {"loss": 1.1756, "grad_norm": 0.32844600081443787, "learning_rate": 0.0002, "epoch": 0.5133846718005134, "step": 700}, {"loss": 1.0807, "grad_norm": 0.3144175708293915, "learning_rate": 0.0002, "epoch": 0.5207187385405208, "step": 710}, {"loss": 1.1952, "grad_norm": 0.3898887634277344, "learning_rate": 0.0002, "epoch": 0.528052805280528, "step": 720}, {"loss": 1.1244, "grad_norm": 1.3220758438110352, "learning_rate": 0.0002, "epoch": 0.5353868720205354, "step": 730}, {"loss": 1.227, "grad_norm": 0.3635874390602112, "learning_rate": 0.0002, "epoch": 0.5427209387605427, "step": 740}, {"loss": 1.2169, "grad_norm": 0.3138217628002167, "learning_rate": 0.0002, "epoch": 0.5500550055005501, "step": 750}, {"loss": 1.1516, "grad_norm": 0.4063207805156708, "learning_rate": 0.0002, "epoch": 0.5573890722405573, "step": 760}, {"loss": 1.1954, "grad_norm": 0.3926219940185547, "learning_rate": 0.0002, "epoch": 0.5647231389805647, "step": 770}, {"loss": 1.1726, "grad_norm": 0.31954652070999146, "learning_rate": 0.0002, "epoch": 0.5720572057205721, "step": 780}, {"loss": 1.2977, "grad_norm": 0.4248711168766022, "learning_rate": 0.0002, "epoch": 0.5793912724605794, "step": 790}, {"loss": 1.1728, "grad_norm": 0.643004834651947, "learning_rate": 0.0002, "epoch": 0.5867253392005867, "step": 800}, {"loss": 1.1793, "grad_norm": 0.3479592800140381, "learning_rate": 0.0002, "epoch": 0.594059405940594, "step": 810}, {"loss": 1.2426, "grad_norm": 0.4684754014015198, "learning_rate": 0.0002, "epoch": 0.6013934726806014, "step": 820}, {"loss": 1.2002, "grad_norm": 0.3739790916442871, "learning_rate": 0.0002, "epoch": 0.6087275394206088, "step": 830}, {"loss": 1.2139, "grad_norm": 0.40884748101234436, "learning_rate": 0.0002, "epoch": 0.6160616061606161, "step": 840}, {"loss": 1.1557, "grad_norm": 0.9722164273262024, "learning_rate": 0.0002, "epoch": 0.6233956729006234, "step": 850}, {"loss": 1.3069, "grad_norm": 0.42992347478866577, "learning_rate": 0.0002, "epoch": 0.6307297396406307, "step": 860}, {"loss": 1.1339, "grad_norm": 0.36654195189476013, "learning_rate": 0.0002, "epoch": 0.6380638063806381, "step": 870}, {"loss": 1.1932, "grad_norm": 0.4113832116127014, "learning_rate": 0.0002, "epoch": 0.6453978731206454, "step": 880}, {"loss": 1.2163, "grad_norm": 0.2948838770389557, "learning_rate": 0.0002, "epoch": 0.6527319398606527, "step": 890}, {"loss": 1.1081, "grad_norm": 0.38330280780792236, "learning_rate": 0.0002, "epoch": 0.6600660066006601, "step": 900}, {"loss": 1.1342, "grad_norm": 0.4428867697715759, "learning_rate": 0.0002, "epoch": 0.6674000733406674, "step": 910}, {"loss": 1.1021, "grad_norm": 0.23659265041351318, "learning_rate": 0.0002, "epoch": 0.6747341400806748, "step": 920}, {"loss": 1.1226, "grad_norm": 0.323685884475708, "learning_rate": 0.0002, "epoch": 0.682068206820682, "step": 930}, {"loss": 1.0853, "grad_norm": 0.39157727360725403, "learning_rate": 0.0002, "epoch": 0.6894022735606894, "step": 940}, {"loss": 1.1435, "grad_norm": 0.27189481258392334, "learning_rate": 0.0002, "epoch": 0.6967363403006968, "step": 950}, {"loss": 1.1033, "grad_norm": 0.529883861541748, "learning_rate": 0.0002, "epoch": 0.7040704070407041, "step": 960}, {"loss": 1.139, "grad_norm": 0.34758689999580383, "learning_rate": 0.0002, "epoch": 0.7114044737807114, "step": 970}, {"loss": 1.2197, "grad_norm": 0.831749439239502, "learning_rate": 0.0002, "epoch": 0.7187385405207187, "step": 980}, {"loss": 1.158, "grad_norm": 0.4438304007053375, "learning_rate": 0.0002, "epoch": 0.7260726072607261, "step": 990}, {"loss": 1.1021, "grad_norm": 0.33840006589889526, "learning_rate": 0.0002, "epoch": 0.7334066740007334, "step": 1000}, {"loss": 1.254, "grad_norm": 0.3454797863960266, "learning_rate": 0.0002, "epoch": 0.7407407407407407, "step": 1010}, {"loss": 1.106, "grad_norm": 0.38999441266059875, "learning_rate": 0.0002, "epoch": 0.7480748074807481, "step": 1020}, {"loss": 1.1428, "grad_norm": 0.2829911708831787, "learning_rate": 0.0002, "epoch": 0.7554088742207554, "step": 1030}, {"loss": 1.2123, "grad_norm": 0.36918163299560547, "learning_rate": 0.0002, "epoch": 0.7627429409607628, "step": 1040}, {"loss": 1.3028, "grad_norm": 0.3415680229663849, "learning_rate": 0.0002, "epoch": 0.77007700770077, "step": 1050}, {"loss": 1.1939, "grad_norm": 0.2974182963371277, "learning_rate": 0.0002, "epoch": 0.7774110744407774, "step": 1060}, {"loss": 1.194, "grad_norm": 0.3880919814109802, "learning_rate": 0.0002, "epoch": 0.7847451411807848, "step": 1070}, {"loss": 1.1095, "grad_norm": 0.33503302931785583, "learning_rate": 0.0002, "epoch": 0.7920792079207921, "step": 1080}, {"loss": 1.2111, "grad_norm": 0.3728407025337219, "learning_rate": 0.0002, "epoch": 0.7994132746607994, "step": 1090}, {"loss": 1.0835, "grad_norm": 0.3509373664855957, "learning_rate": 0.0002, "epoch": 0.8067473414008067, "step": 1100}, {"loss": 1.2661, "grad_norm": 0.42228564620018005, "learning_rate": 0.0002, "epoch": 0.8140814081408141, "step": 1110}, {"loss": 1.1788, "grad_norm": 0.313467800617218, "learning_rate": 0.0002, "epoch": 0.8214154748808215, "step": 1120}, {"loss": 1.1971, "grad_norm": 0.3378850817680359, "learning_rate": 0.0002, "epoch": 0.8287495416208287, "step": 1130}, {"loss": 1.1238, "grad_norm": 0.43200382590293884, "learning_rate": 0.0002, "epoch": 0.8360836083608361, "step": 1140}, {"loss": 1.3203, "grad_norm": 0.3309599459171295, "learning_rate": 0.0002, "epoch": 0.8434176751008434, "step": 1150}, {"loss": 1.1062, "grad_norm": 0.3526846170425415, "learning_rate": 0.0002, "epoch": 0.8507517418408508, "step": 1160}, {"loss": 1.0851, "grad_norm": 1.2722247838974, "learning_rate": 0.0002, "epoch": 0.858085808580858, "step": 1170}, {"loss": 1.0785, "grad_norm": 0.34142059087753296, "learning_rate": 0.0002, "epoch": 0.8654198753208654, "step": 1180}, {"loss": 1.2187, "grad_norm": 0.3805823028087616, "learning_rate": 0.0002, "epoch": 0.8727539420608728, "step": 1190}, {"loss": 1.1215, "grad_norm": 0.3931232690811157, "learning_rate": 0.0002, "epoch": 0.8800880088008801, "step": 1200}, {"loss": 1.0948, "grad_norm": 0.2937372624874115, "learning_rate": 0.0002, "epoch": 0.8874220755408874, "step": 1210}, {"loss": 1.1228, "grad_norm": 0.3757196366786957, "learning_rate": 0.0002, "epoch": 0.8947561422808947, "step": 1220}, {"loss": 1.1222, "grad_norm": 0.3502705991268158, "learning_rate": 0.0002, "epoch": 0.9020902090209021, "step": 1230}, {"loss": 1.2242, "grad_norm": 0.32758915424346924, "learning_rate": 0.0002, "epoch": 0.9094242757609095, "step": 1240}, {"loss": 1.215, "grad_norm": 0.37199416756629944, "learning_rate": 0.0002, "epoch": 0.9167583425009168, "step": 1250}, {"loss": 1.1225, "grad_norm": 0.3551490604877472, "learning_rate": 0.0002, "epoch": 0.9240924092409241, "step": 1260}, {"loss": 1.1966, "grad_norm": 0.2859550714492798, "learning_rate": 0.0002, "epoch": 0.9314264759809314, "step": 1270}, {"loss": 1.2186, "grad_norm": 0.427990585565567, "learning_rate": 0.0002, "epoch": 0.9387605427209388, "step": 1280}, {"loss": 1.2848, "grad_norm": 0.33717992901802063, "learning_rate": 0.0002, "epoch": 0.9460946094609461, "step": 1290}, {"loss": 1.1656, "grad_norm": 0.30225634574890137, "learning_rate": 0.0002, "epoch": 0.9534286762009534, "step": 1300}, {"loss": 1.2404, "grad_norm": 0.385821133852005, "learning_rate": 0.0002, "epoch": 0.9607627429409608, "step": 1310}, {"loss": 1.1932, "grad_norm": 0.35278066992759705, "learning_rate": 0.0002, "epoch": 0.9680968096809681, "step": 1320}, {"loss": 1.1071, "grad_norm": 0.49987098574638367, "learning_rate": 0.0002, "epoch": 0.9754308764209755, "step": 1330}, {"loss": 1.2259, "grad_norm": 0.3842747211456299, "learning_rate": 0.0002, "epoch": 0.9827649431609827, "step": 1340}, {"loss": 1.0862, "grad_norm": 0.6274653673171997, "learning_rate": 0.0002, "epoch": 0.9900990099009901, "step": 1350}, {"loss": 1.124, "grad_norm": 0.5239808559417725, "learning_rate": 0.0002, "epoch": 0.9974330766409975, "step": 1360}, {"eval_loss": 1.1822267770767212, "eval_runtime": 32.7389, "eval_samples_per_second": 13.165, "eval_steps_per_second": 1.649, "epoch": 0.9996332966629996, "step": 1363}, {"loss": 1.096, "grad_norm": 0.45311301946640015, "learning_rate": 0.0002, "epoch": 1.0047671433810048, "step": 1370}, {"loss": 1.0143, "grad_norm": 0.29685574769973755, "learning_rate": 0.0002, "epoch": 1.012101210121012, "step": 1380}, {"loss": 1.0302, "grad_norm": 0.3290937840938568, "learning_rate": 0.0002, "epoch": 1.0194352768610195, "step": 1390}, {"loss": 1.0295, "grad_norm": 0.3801758587360382, "learning_rate": 0.0002, "epoch": 1.0267693436010268, "step": 1400}, {"loss": 1.1226, "grad_norm": 0.794174313545227, "learning_rate": 0.0002, "epoch": 1.034103410341034, "step": 1410}, {"loss": 1.2232, "grad_norm": 0.3854154646396637, "learning_rate": 0.0002, "epoch": 1.0414374770810415, "step": 1420}, {"loss": 1.0652, "grad_norm": 0.32702451944351196, "learning_rate": 0.0002, "epoch": 1.0487715438210488, "step": 1430}, {"loss": 1.1144, "grad_norm": 0.7815203666687012, "learning_rate": 0.0002, "epoch": 1.056105610561056, "step": 1440}, {"loss": 1.1316, "grad_norm": 0.3087436854839325, "learning_rate": 0.0002, "epoch": 1.0634396773010635, "step": 1450}, {"loss": 1.1124, "grad_norm": 0.3847602903842926, "learning_rate": 0.0002, "epoch": 1.0707737440410707, "step": 1460}, {"loss": 1.1428, "grad_norm": 0.3693031370639801, "learning_rate": 0.0002, "epoch": 1.0781078107810782, "step": 1470}, {"loss": 1.0995, "grad_norm": 0.4111202359199524, "learning_rate": 0.0002, "epoch": 1.0854418775210855, "step": 1480}, {"loss": 1.0961, "grad_norm": 0.41452381014823914, "learning_rate": 0.0002, "epoch": 1.0927759442610927, "step": 1490}, {"loss": 1.1068, "grad_norm": 0.3336445093154907, "learning_rate": 0.0002, "epoch": 1.1001100110011002, "step": 1500}, {"loss": 1.0556, "grad_norm": 0.3923407793045044, "learning_rate": 0.0002, "epoch": 1.1074440777411074, "step": 1510}, {"loss": 1.1644, "grad_norm": 0.46215683221817017, "learning_rate": 0.0002, "epoch": 1.1147781444811147, "step": 1520}, {"loss": 1.1133, "grad_norm": 0.3592156767845154, "learning_rate": 0.0002, "epoch": 1.1221122112211221, "step": 1530}, {"loss": 1.0957, "grad_norm": 0.361110657453537, "learning_rate": 0.0002, "epoch": 1.1294462779611294, "step": 1540}, {"loss": 1.1553, "grad_norm": 0.5317131280899048, "learning_rate": 0.0002, "epoch": 1.1367803447011369, "step": 1550}, {"loss": 1.0368, "grad_norm": 0.3882388174533844, "learning_rate": 0.0002, "epoch": 1.1441144114411441, "step": 1560}, {"loss": 1.0805, "grad_norm": 0.3259428143501282, "learning_rate": 0.0002, "epoch": 1.1514484781811514, "step": 1570}, {"loss": 1.1819, "grad_norm": 0.410935640335083, "learning_rate": 0.0002, "epoch": 1.1587825449211588, "step": 1580}, {"loss": 1.1143, "grad_norm": 0.44940185546875, "learning_rate": 0.0002, "epoch": 1.166116611661166, "step": 1590}, {"loss": 1.0334, "grad_norm": 0.5106484293937683, "learning_rate": 0.0002, "epoch": 1.1734506784011733, "step": 1600}, {"loss": 1.2376, "grad_norm": 0.6603665947914124, "learning_rate": 0.0002, "epoch": 1.1807847451411808, "step": 1610}, {"loss": 1.1227, "grad_norm": 0.4799964129924774, "learning_rate": 0.0002, "epoch": 1.188118811881188, "step": 1620}, {"loss": 1.1191, "grad_norm": 0.4389883279800415, "learning_rate": 0.0002, "epoch": 1.1954528786211955, "step": 1630}, {"loss": 1.0667, "grad_norm": 0.4188813269138336, "learning_rate": 0.0002, "epoch": 1.2027869453612028, "step": 1640}, {"loss": 1.0605, "grad_norm": 0.7132157683372498, "learning_rate": 0.0002, "epoch": 1.21012101210121, "step": 1650}, {"loss": 1.0204, "grad_norm": 0.507480263710022, "learning_rate": 0.0002, "epoch": 1.2174550788412175, "step": 1660}, {"loss": 0.9948, "grad_norm": 0.9452332854270935, "learning_rate": 0.0002, "epoch": 1.2247891455812248, "step": 1670}, {"loss": 1.0228, "grad_norm": 0.4121614992618561, "learning_rate": 0.0002, "epoch": 1.2321232123212322, "step": 1680}, {"loss": 1.0366, "grad_norm": 0.34230247139930725, "learning_rate": 0.0002, "epoch": 1.2394572790612395, "step": 1690}, {"loss": 1.1289, "grad_norm": 0.4026208817958832, "learning_rate": 0.0002, "epoch": 1.2467913458012467, "step": 1700}, {"loss": 1.0206, "grad_norm": 0.46673697233200073, "learning_rate": 0.0002, "epoch": 1.2541254125412542, "step": 1710}, {"loss": 1.0827, "grad_norm": 0.38349825143814087, "learning_rate": 0.0002, "epoch": 1.2614594792812615, "step": 1720}, {"loss": 1.0356, "grad_norm": 0.4049997627735138, "learning_rate": 0.0002, "epoch": 1.2687935460212687, "step": 1730}, {"loss": 0.9504, "grad_norm": 0.3417615294456482, "learning_rate": 0.0002, "epoch": 1.2761276127612762, "step": 1740}, {"loss": 1.094, "grad_norm": 0.4277614951133728, "learning_rate": 0.0002, "epoch": 1.2834616795012834, "step": 1750}, {"loss": 0.9938, "grad_norm": 0.5864202976226807, "learning_rate": 0.0002, "epoch": 1.2907957462412907, "step": 1760}, {"loss": 1.1167, "grad_norm": 0.7097493410110474, "learning_rate": 0.0002, "epoch": 1.2981298129812981, "step": 1770}, {"loss": 1.1132, "grad_norm": 0.3145381212234497, "learning_rate": 0.0002, "epoch": 1.3054638797213054, "step": 1780}, {"loss": 1.1099, "grad_norm": 0.5116165280342102, "learning_rate": 0.0002, "epoch": 1.3127979464613129, "step": 1790}, {"loss": 1.0765, "grad_norm": 0.7469736337661743, "learning_rate": 0.0002, "epoch": 1.3201320132013201, "step": 1800}, {"loss": 1.0663, "grad_norm": 0.32272255420684814, "learning_rate": 0.0002, "epoch": 1.3274660799413276, "step": 1810}, {"loss": 0.9887, "grad_norm": 0.3534623086452484, "learning_rate": 0.0002, "epoch": 1.3348001466813348, "step": 1820}, {"loss": 1.1628, "grad_norm": 0.36127907037734985, "learning_rate": 0.0002, "epoch": 1.342134213421342, "step": 1830}, {"loss": 1.0972, "grad_norm": 0.4072401523590088, "learning_rate": 0.0002, "epoch": 1.3494682801613496, "step": 1840}, {"loss": 1.1267, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.3568023469013568, "step": 1850}, {"loss": 1.0173, "grad_norm": 0.412883460521698, "learning_rate": 0.0002, "epoch": 1.364136413641364, "step": 1860}, {"loss": 1.0265, "grad_norm": 0.3735875189304352, "learning_rate": 0.0002, "epoch": 1.3714704803813715, "step": 1870}, {"loss": 1.1061, "grad_norm": 0.39158159494400024, "learning_rate": 0.0002, "epoch": 1.3788045471213788, "step": 1880}, {"loss": 1.0433, "grad_norm": 0.44431769847869873, "learning_rate": 0.0002, "epoch": 1.386138613861386, "step": 1890}, {"loss": 1.0216, "grad_norm": 0.37772801518440247, "learning_rate": 0.0002, "epoch": 1.3934726806013935, "step": 1900}, {"loss": 1.0674, "grad_norm": 0.4056641757488251, "learning_rate": 0.0002, "epoch": 1.4008067473414008, "step": 1910}, {"loss": 1.0256, "grad_norm": 0.41612377762794495, "learning_rate": 0.0002, "epoch": 1.408140814081408, "step": 1920}, {"loss": 1.0467, "grad_norm": 0.41153013706207275, "learning_rate": 0.0002, "epoch": 1.4154748808214155, "step": 1930}, {"loss": 1.1062, "grad_norm": 0.387845516204834, "learning_rate": 0.0002, "epoch": 1.4228089475614227, "step": 1940}, {"loss": 1.1094, "grad_norm": 0.3809587061405182, "learning_rate": 0.0002, "epoch": 1.4301430143014302, "step": 1950}, {"loss": 1.0461, "grad_norm": 0.3625726103782654, "learning_rate": 0.0002, "epoch": 1.4374770810414375, "step": 1960}, {"loss": 0.9983, "grad_norm": 0.5294290781021118, "learning_rate": 0.0002, "epoch": 1.444811147781445, "step": 1970}, {"loss": 1.1114, "grad_norm": 0.39975494146347046, "learning_rate": 0.0002, "epoch": 1.4521452145214522, "step": 1980}, {"loss": 0.9704, "grad_norm": 0.4181167185306549, "learning_rate": 0.0002, "epoch": 1.4594792812614594, "step": 1990}, {"loss": 1.1146, "grad_norm": 0.42001503705978394, "learning_rate": 0.0002, "epoch": 1.466813348001467, "step": 2000}, {"loss": 1.1266, "grad_norm": 0.4877578616142273, "learning_rate": 0.0002, "epoch": 1.4741474147414741, "step": 2010}, {"loss": 1.1012, "grad_norm": 0.4050969183444977, "learning_rate": 0.0002, "epoch": 1.4814814814814814, "step": 2020}, {"loss": 1.0562, "grad_norm": 0.39068883657455444, "learning_rate": 0.0002, "epoch": 1.4888155482214889, "step": 2030}, {"loss": 1.0464, "grad_norm": 0.421282559633255, "learning_rate": 0.0002, "epoch": 1.4961496149614961, "step": 2040}, {"loss": 1.0532, "grad_norm": 0.47092297673225403, "learning_rate": 0.0002, "epoch": 1.5034836817015034, "step": 2050}, {"loss": 0.9348, "grad_norm": 0.39688974618911743, "learning_rate": 0.0002, "epoch": 1.5108177484415108, "step": 2060}, {"loss": 1.08, "grad_norm": 0.5529879331588745, "learning_rate": 0.0002, "epoch": 1.5181518151815183, "step": 2070}, {"loss": 1.1836, "grad_norm": 0.4879782199859619, "learning_rate": 0.0002, "epoch": 1.5254858819215253, "step": 2080}, {"loss": 1.0432, "grad_norm": 0.5517361164093018, "learning_rate": 0.0002, "epoch": 1.5328199486615328, "step": 2090}, {"loss": 1.0433, "grad_norm": 0.44015637040138245, "learning_rate": 0.0002, "epoch": 1.5401540154015403, "step": 2100}, {"loss": 1.1873, "grad_norm": 0.5435167551040649, "learning_rate": 0.0002, "epoch": 1.5474880821415475, "step": 2110}, {"loss": 1.1076, "grad_norm": 0.5714033246040344, "learning_rate": 0.0002, "epoch": 1.5548221488815548, "step": 2120}, {"loss": 1.1107, "grad_norm": 0.31732529401779175, "learning_rate": 0.0002, "epoch": 1.5621562156215623, "step": 2130}, {"loss": 1.0817, "grad_norm": 0.49068278074264526, "learning_rate": 0.0002, "epoch": 1.5694902823615695, "step": 2140}, {"loss": 1.0254, "grad_norm": 0.46851542592048645, "learning_rate": 0.0002, "epoch": 1.5768243491015768, "step": 2150}, {"loss": 1.0623, "grad_norm": 0.5083092451095581, "learning_rate": 0.0002, "epoch": 1.5841584158415842, "step": 2160}, {"loss": 1.0603, "grad_norm": 0.9822936058044434, "learning_rate": 0.0002, "epoch": 1.5914924825815915, "step": 2170}, {"loss": 0.9986, "grad_norm": 0.4575989246368408, "learning_rate": 0.0002, "epoch": 1.5988265493215987, "step": 2180}, {"loss": 1.1292, "grad_norm": 0.47444286942481995, "learning_rate": 0.0002, "epoch": 1.6061606160616062, "step": 2190}, {"loss": 1.0136, "grad_norm": 0.7208226919174194, "learning_rate": 0.0002, "epoch": 1.6134946828016135, "step": 2200}, {"loss": 1.15, "grad_norm": 0.43791481852531433, "learning_rate": 0.0002, "epoch": 1.6208287495416207, "step": 2210}, {"loss": 1.0961, "grad_norm": 0.5245792865753174, "learning_rate": 0.0002, "epoch": 1.6281628162816282, "step": 2220}, {"loss": 0.9957, "grad_norm": 0.39289429783821106, "learning_rate": 0.0002, "epoch": 1.6354968830216357, "step": 2230}, {"loss": 1.133, "grad_norm": 0.6106135845184326, "learning_rate": 0.0002, "epoch": 1.6428309497616427, "step": 2240}, {"loss": 1.0129, "grad_norm": 0.3722580671310425, "learning_rate": 0.0002, "epoch": 1.6501650165016502, "step": 2250}, {"loss": 1.0446, "grad_norm": 0.3649403750896454, "learning_rate": 0.0002, "epoch": 1.6574990832416576, "step": 2260}, {"loss": 1.0037, "grad_norm": 0.46514248847961426, "learning_rate": 0.0002, "epoch": 1.6648331499816649, "step": 2270}, {"loss": 1.0022, "grad_norm": 0.42034927010536194, "learning_rate": 0.0002, "epoch": 1.6721672167216721, "step": 2280}, {"loss": 1.1362, "grad_norm": 0.45202910900115967, "learning_rate": 0.0002, "epoch": 1.6795012834616796, "step": 2290}, {"loss": 1.0866, "grad_norm": 0.36257603764533997, "learning_rate": 0.0002, "epoch": 1.6868353502016868, "step": 2300}, {"loss": 1.0973, "grad_norm": 0.6340323090553284, "learning_rate": 0.0002, "epoch": 1.694169416941694, "step": 2310}, {"loss": 1.0615, "grad_norm": 0.4352878928184509, "learning_rate": 0.0002, "epoch": 1.7015034836817016, "step": 2320}, {"loss": 1.0629, "grad_norm": 0.45029792189598083, "learning_rate": 0.0002, "epoch": 1.7088375504217088, "step": 2330}, {"loss": 0.9621, "grad_norm": 0.3891315758228302, "learning_rate": 0.0002, "epoch": 1.716171617161716, "step": 2340}, {"loss": 0.9779, "grad_norm": 0.35180050134658813, "learning_rate": 0.0002, "epoch": 1.7235056839017235, "step": 2350}, {"loss": 1.0368, "grad_norm": 0.42367449402809143, "learning_rate": 0.0002, "epoch": 1.7308397506417308, "step": 2360}, {"loss": 1.0376, "grad_norm": 0.4553675353527069, "learning_rate": 0.0002, "epoch": 1.738173817381738, "step": 2370}, {"loss": 1.1467, "grad_norm": 0.5944654941558838, "learning_rate": 0.0002, "epoch": 1.7455078841217455, "step": 2380}, {"loss": 1.0548, "grad_norm": 0.3479664623737335, "learning_rate": 0.0002, "epoch": 1.752841950861753, "step": 2390}, {"loss": 1.0798, "grad_norm": 0.3585502505302429, "learning_rate": 0.0002, "epoch": 1.76017601760176, "step": 2400}, {"loss": 1.0983, "grad_norm": 0.4263346493244171, "learning_rate": 0.0002, "epoch": 1.7675100843417675, "step": 2410}, {"loss": 1.054, "grad_norm": 0.5476409196853638, "learning_rate": 0.0002, "epoch": 1.774844151081775, "step": 2420}, {"loss": 1.1615, "grad_norm": 0.3694186508655548, "learning_rate": 0.0002, "epoch": 1.7821782178217822, "step": 2430}, {"loss": 1.1343, "grad_norm": 0.9185658693313599, "learning_rate": 0.0002, "epoch": 1.7895122845617895, "step": 2440}, {"loss": 1.0764, "grad_norm": 0.7171908020973206, "learning_rate": 0.0002, "epoch": 1.796846351301797, "step": 2450}, {"loss": 1.1154, "grad_norm": 0.550658643245697, "learning_rate": 0.0002, "epoch": 1.8041804180418042, "step": 2460}, {"loss": 0.9975, "grad_norm": 0.4075568914413452, "learning_rate": 0.0002, "epoch": 1.8115144847818114, "step": 2470}, {"loss": 1.0935, "grad_norm": 0.3790127635002136, "learning_rate": 0.0002, "epoch": 1.818848551521819, "step": 2480}, {"loss": 0.9839, "grad_norm": 0.3576384484767914, "learning_rate": 0.0002, "epoch": 1.8261826182618262, "step": 2490}, {"loss": 1.1369, "grad_norm": 0.3919370770454407, "learning_rate": 0.0002, "epoch": 1.8335166850018334, "step": 2500}, {"loss": 0.9985, "grad_norm": 0.485083669424057, "learning_rate": 0.0002, "epoch": 1.8408507517418409, "step": 2510}, {"loss": 1.1585, "grad_norm": 0.4564347565174103, "learning_rate": 0.0002, "epoch": 1.8481848184818483, "step": 2520}, {"loss": 1.0944, "grad_norm": 0.3613106608390808, "learning_rate": 0.0002, "epoch": 1.8555188852218554, "step": 2530}, {"loss": 1.0819, "grad_norm": 0.39600759744644165, "learning_rate": 0.0002, "epoch": 1.8628529519618628, "step": 2540}, {"loss": 0.9453, "grad_norm": 1.123499870300293, "learning_rate": 0.0002, "epoch": 1.8701870187018703, "step": 2550}, {"loss": 1.0635, "grad_norm": 0.4612680673599243, "learning_rate": 0.0002, "epoch": 1.8775210854418776, "step": 2560}, {"loss": 1.0087, "grad_norm": 0.42745399475097656, "learning_rate": 0.0002, "epoch": 1.8848551521818848, "step": 2570}, {"loss": 1.0102, "grad_norm": 0.4055580198764801, "learning_rate": 0.0002, "epoch": 1.8921892189218923, "step": 2580}, {"loss": 1.0177, "grad_norm": 0.44174644351005554, "learning_rate": 0.0002, "epoch": 1.8995232856618995, "step": 2590}, {"loss": 0.9886, "grad_norm": 1.0228385925292969, "learning_rate": 0.0002, "epoch": 1.9068573524019068, "step": 2600}, {"loss": 1.0857, "grad_norm": 0.3496396243572235, "learning_rate": 0.0002, "epoch": 1.9141914191419143, "step": 2610}, {"loss": 1.0955, "grad_norm": 0.4191173017024994, "learning_rate": 0.0002, "epoch": 1.9215254858819215, "step": 2620}, {"loss": 1.0943, "grad_norm": 0.6778554916381836, "learning_rate": 0.0002, "epoch": 1.9288595526219288, "step": 2630}, {"loss": 1.0594, "grad_norm": 0.41992834210395813, "learning_rate": 0.0002, "epoch": 1.9361936193619362, "step": 2640}, {"loss": 1.1159, "grad_norm": 0.8760401010513306, "learning_rate": 0.0002, "epoch": 1.9435276861019435, "step": 2650}, {"loss": 1.0379, "grad_norm": 0.44049209356307983, "learning_rate": 0.0002, "epoch": 1.9508617528419507, "step": 2660}, {"loss": 1.1008, "grad_norm": 0.5651928782463074, "learning_rate": 0.0002, "epoch": 1.9581958195819582, "step": 2670}, {"loss": 1.1317, "grad_norm": 0.5292727947235107, "learning_rate": 0.0002, "epoch": 1.9655298863219657, "step": 2680}, {"loss": 1.1328, "grad_norm": 0.6012240648269653, "learning_rate": 0.0002, "epoch": 1.9728639530619727, "step": 2690}, {"loss": 1.0683, "grad_norm": 0.3945149779319763, "learning_rate": 0.0002, "epoch": 1.9801980198019802, "step": 2700}, {"loss": 1.0155, "grad_norm": 0.5732627511024475, "learning_rate": 0.0002, "epoch": 1.9875320865419877, "step": 2710}, {"loss": 0.9857, "grad_norm": 0.3963361084461212, "learning_rate": 0.0002, "epoch": 1.994866153281995, "step": 2720}]}
+{"epoch": 2.9996332966629997, "step": 4090, "epoch_duration": 1492.1944043636322, "total_accumulated_duration": 4434.300072669983, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7672.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 1.9722, "grad_norm": 0.47521963715553284, "learning_rate": 0.0002, "epoch": 0.007334066740007334, "step": 10}, {"loss": 1.4821, "grad_norm": 0.5395162105560303, "learning_rate": 0.0002, "epoch": 0.014668133480014669, "step": 20}, {"loss": 1.4202, "grad_norm": 0.4305780231952667, "learning_rate": 0.0002, "epoch": 0.022002200220022004, "step": 30}, {"loss": 1.4271, "grad_norm": 0.6938246488571167, "learning_rate": 0.0002, "epoch": 0.029336266960029337, "step": 40}, {"loss": 1.3112, "grad_norm": 1.5133819580078125, "learning_rate": 0.0002, "epoch": 0.03667033370003667, "step": 50}, {"loss": 1.3132, "grad_norm": 0.9173883199691772, "learning_rate": 0.0002, "epoch": 0.04400440044004401, "step": 60}, {"loss": 1.2844, "grad_norm": 0.4619861841201782, "learning_rate": 0.0002, "epoch": 0.05133846718005134, "step": 70}, {"loss": 1.2108, "grad_norm": 0.46118637919425964, "learning_rate": 0.0002, "epoch": 0.058672533920058674, "step": 80}, {"loss": 1.3441, "grad_norm": 0.4468648135662079, "learning_rate": 0.0002, "epoch": 0.066006600660066, "step": 90}, {"loss": 1.1863, "grad_norm": 0.46123769879341125, "learning_rate": 0.0002, "epoch": 0.07334066740007333, "step": 100}, {"loss": 1.2772, "grad_norm": 0.4859139025211334, "learning_rate": 0.0002, "epoch": 0.08067473414008068, "step": 110}, {"loss": 1.2087, "grad_norm": 0.4384922385215759, "learning_rate": 0.0002, "epoch": 0.08800880088008801, "step": 120}, {"loss": 1.2927, "grad_norm": 0.39519360661506653, "learning_rate": 0.0002, "epoch": 0.09534286762009535, "step": 130}, {"loss": 1.2349, "grad_norm": 0.4049859344959259, "learning_rate": 0.0002, "epoch": 0.10267693436010268, "step": 140}, {"loss": 1.293, "grad_norm": 0.4605638086795807, "learning_rate": 0.0002, "epoch": 0.11001100110011001, "step": 150}, {"loss": 1.2659, "grad_norm": 0.4201928377151489, "learning_rate": 0.0002, "epoch": 0.11734506784011735, "step": 160}, {"loss": 1.3961, "grad_norm": 0.5367777347564697, "learning_rate": 0.0002, "epoch": 0.12467913458012468, "step": 170}, {"loss": 1.2481, "grad_norm": 0.41752299666404724, "learning_rate": 0.0002, "epoch": 0.132013201320132, "step": 180}, {"loss": 1.207, "grad_norm": 0.31597763299942017, "learning_rate": 0.0002, "epoch": 0.13934726806013933, "step": 190}, {"loss": 1.2441, "grad_norm": 0.7468788623809814, "learning_rate": 0.0002, "epoch": 0.14668133480014667, "step": 200}, {"loss": 1.199, "grad_norm": 0.3403034508228302, "learning_rate": 0.0002, "epoch": 0.15401540154015403, "step": 210}, {"loss": 1.2439, "grad_norm": 0.34240293502807617, "learning_rate": 0.0002, "epoch": 0.16134946828016136, "step": 220}, {"loss": 1.2022, "grad_norm": 0.356158971786499, "learning_rate": 0.0002, "epoch": 0.1686835350201687, "step": 230}, {"loss": 1.207, "grad_norm": 0.3448857367038727, "learning_rate": 0.0002, "epoch": 0.17601760176017603, "step": 240}, {"loss": 1.2156, "grad_norm": 0.3475699722766876, "learning_rate": 0.0002, "epoch": 0.18335166850018336, "step": 250}, {"loss": 1.1551, "grad_norm": 0.2770358622074127, "learning_rate": 0.0002, "epoch": 0.1906857352401907, "step": 260}, {"loss": 1.2238, "grad_norm": 0.4310270845890045, "learning_rate": 0.0002, "epoch": 0.19801980198019803, "step": 270}, {"loss": 1.2917, "grad_norm": 0.335041880607605, "learning_rate": 0.0002, "epoch": 0.20535386872020536, "step": 280}, {"loss": 1.0959, "grad_norm": 0.3420602083206177, "learning_rate": 0.0002, "epoch": 0.2126879354602127, "step": 290}, {"loss": 1.1232, "grad_norm": 0.325001060962677, "learning_rate": 0.0002, "epoch": 0.22002200220022003, "step": 300}, {"loss": 1.2007, "grad_norm": 0.3027827739715576, "learning_rate": 0.0002, "epoch": 0.22735606894022736, "step": 310}, {"loss": 1.1803, "grad_norm": 0.435550719499588, "learning_rate": 0.0002, "epoch": 0.2346901356802347, "step": 320}, {"loss": 1.2045, "grad_norm": 0.3884522616863251, "learning_rate": 0.0002, "epoch": 0.24202420242024203, "step": 330}, {"loss": 1.2481, "grad_norm": 0.7736002206802368, "learning_rate": 0.0002, "epoch": 0.24935826916024936, "step": 340}, {"loss": 1.3606, "grad_norm": 0.35052821040153503, "learning_rate": 0.0002, "epoch": 0.2566923359002567, "step": 350}, {"loss": 1.2129, "grad_norm": 0.3311890959739685, "learning_rate": 0.0002, "epoch": 0.264026402640264, "step": 360}, {"loss": 1.2219, "grad_norm": 0.7473500370979309, "learning_rate": 0.0002, "epoch": 0.27136046938027136, "step": 370}, {"loss": 1.2712, "grad_norm": 0.3681875765323639, "learning_rate": 0.0002, "epoch": 0.27869453612027867, "step": 380}, {"loss": 1.2258, "grad_norm": 0.3764737844467163, "learning_rate": 0.0002, "epoch": 0.28602860286028603, "step": 390}, {"loss": 1.1917, "grad_norm": 0.4243989586830139, "learning_rate": 0.0002, "epoch": 0.29336266960029334, "step": 400}, {"loss": 1.199, "grad_norm": 0.2658531963825226, "learning_rate": 0.0002, "epoch": 0.3006967363403007, "step": 410}, {"loss": 1.1622, "grad_norm": 0.3436793386936188, "learning_rate": 0.0002, "epoch": 0.30803080308030806, "step": 420}, {"loss": 1.2953, "grad_norm": 0.5101129412651062, "learning_rate": 0.0002, "epoch": 0.31536486982031536, "step": 430}, {"loss": 1.1557, "grad_norm": 0.3319750726222992, "learning_rate": 0.0002, "epoch": 0.3226989365603227, "step": 440}, {"loss": 1.1804, "grad_norm": 0.385148286819458, "learning_rate": 0.0002, "epoch": 0.33003300330033003, "step": 450}, {"loss": 1.1808, "grad_norm": 0.3477935791015625, "learning_rate": 0.0002, "epoch": 0.3373670700403374, "step": 460}, {"loss": 1.1877, "grad_norm": 0.29748716950416565, "learning_rate": 0.0002, "epoch": 0.3447011367803447, "step": 470}, {"loss": 1.19, "grad_norm": 0.34083324670791626, "learning_rate": 0.0002, "epoch": 0.35203520352035206, "step": 480}, {"loss": 1.2, "grad_norm": 0.36904552578926086, "learning_rate": 0.0002, "epoch": 0.35936927026035936, "step": 490}, {"loss": 1.2223, "grad_norm": 0.315483033657074, "learning_rate": 0.0002, "epoch": 0.3667033370003667, "step": 500}, {"loss": 1.1461, "grad_norm": 0.44897955656051636, "learning_rate": 0.0002, "epoch": 0.37403740374037403, "step": 510}, {"loss": 1.3035, "grad_norm": 0.3160701394081116, "learning_rate": 0.0002, "epoch": 0.3813714704803814, "step": 520}, {"loss": 1.3197, "grad_norm": 0.29584741592407227, "learning_rate": 0.0002, "epoch": 0.3887055372203887, "step": 530}, {"loss": 1.2983, "grad_norm": 0.5430002808570862, "learning_rate": 0.0002, "epoch": 0.39603960396039606, "step": 540}, {"loss": 1.2459, "grad_norm": 0.2908070683479309, "learning_rate": 0.0002, "epoch": 0.40337367070040336, "step": 550}, {"loss": 1.2384, "grad_norm": 0.35066530108451843, "learning_rate": 0.0002, "epoch": 0.4107077374404107, "step": 560}, {"loss": 1.1784, "grad_norm": 0.37588003277778625, "learning_rate": 0.0002, "epoch": 0.41804180418041803, "step": 570}, {"loss": 1.2334, "grad_norm": 0.3112126886844635, "learning_rate": 0.0002, "epoch": 0.4253758709204254, "step": 580}, {"loss": 1.1439, "grad_norm": 0.35577139258384705, "learning_rate": 0.0002, "epoch": 0.4327099376604327, "step": 590}, {"loss": 1.184, "grad_norm": 0.31706422567367554, "learning_rate": 0.0002, "epoch": 0.44004400440044006, "step": 600}, {"loss": 1.2081, "grad_norm": 0.3249092102050781, "learning_rate": 0.0002, "epoch": 0.44737807114044736, "step": 610}, {"loss": 1.0824, "grad_norm": 0.3842705488204956, "learning_rate": 0.0002, "epoch": 0.4547121378804547, "step": 620}, {"loss": 1.2257, "grad_norm": 0.390991747379303, "learning_rate": 0.0002, "epoch": 0.46204620462046203, "step": 630}, {"loss": 1.1954, "grad_norm": 0.27532413601875305, "learning_rate": 0.0002, "epoch": 0.4693802713604694, "step": 640}, {"loss": 1.1058, "grad_norm": 0.31412816047668457, "learning_rate": 0.0002, "epoch": 0.4767143381004767, "step": 650}, {"loss": 1.1312, "grad_norm": 0.32117101550102234, "learning_rate": 0.0002, "epoch": 0.48404840484048406, "step": 660}, {"loss": 1.2423, "grad_norm": 0.3810010254383087, "learning_rate": 0.0002, "epoch": 0.49138247158049136, "step": 670}, {"loss": 1.1978, "grad_norm": 0.36289164423942566, "learning_rate": 0.0002, "epoch": 0.4987165383204987, "step": 680}, {"loss": 1.2034, "grad_norm": 0.34458720684051514, "learning_rate": 0.0002, "epoch": 0.506050605060506, "step": 690}, {"loss": 1.1756, "grad_norm": 0.32844600081443787, "learning_rate": 0.0002, "epoch": 0.5133846718005134, "step": 700}, {"loss": 1.0807, "grad_norm": 0.3144175708293915, "learning_rate": 0.0002, "epoch": 0.5207187385405208, "step": 710}, {"loss": 1.1952, "grad_norm": 0.3898887634277344, "learning_rate": 0.0002, "epoch": 0.528052805280528, "step": 720}, {"loss": 1.1244, "grad_norm": 1.3220758438110352, "learning_rate": 0.0002, "epoch": 0.5353868720205354, "step": 730}, {"loss": 1.227, "grad_norm": 0.3635874390602112, "learning_rate": 0.0002, "epoch": 0.5427209387605427, "step": 740}, {"loss": 1.2169, "grad_norm": 0.3138217628002167, "learning_rate": 0.0002, "epoch": 0.5500550055005501, "step": 750}, {"loss": 1.1516, "grad_norm": 0.4063207805156708, "learning_rate": 0.0002, "epoch": 0.5573890722405573, "step": 760}, {"loss": 1.1954, "grad_norm": 0.3926219940185547, "learning_rate": 0.0002, "epoch": 0.5647231389805647, "step": 770}, {"loss": 1.1726, "grad_norm": 0.31954652070999146, "learning_rate": 0.0002, "epoch": 0.5720572057205721, "step": 780}, {"loss": 1.2977, "grad_norm": 0.4248711168766022, "learning_rate": 0.0002, "epoch": 0.5793912724605794, "step": 790}, {"loss": 1.1728, "grad_norm": 0.643004834651947, "learning_rate": 0.0002, "epoch": 0.5867253392005867, "step": 800}, {"loss": 1.1793, "grad_norm": 0.3479592800140381, "learning_rate": 0.0002, "epoch": 0.594059405940594, "step": 810}, {"loss": 1.2426, "grad_norm": 0.4684754014015198, "learning_rate": 0.0002, "epoch": 0.6013934726806014, "step": 820}, {"loss": 1.2002, "grad_norm": 0.3739790916442871, "learning_rate": 0.0002, "epoch": 0.6087275394206088, "step": 830}, {"loss": 1.2139, "grad_norm": 0.40884748101234436, "learning_rate": 0.0002, "epoch": 0.6160616061606161, "step": 840}, {"loss": 1.1557, "grad_norm": 0.9722164273262024, "learning_rate": 0.0002, "epoch": 0.6233956729006234, "step": 850}, {"loss": 1.3069, "grad_norm": 0.42992347478866577, "learning_rate": 0.0002, "epoch": 0.6307297396406307, "step": 860}, {"loss": 1.1339, "grad_norm": 0.36654195189476013, "learning_rate": 0.0002, "epoch": 0.6380638063806381, "step": 870}, {"loss": 1.1932, "grad_norm": 0.4113832116127014, "learning_rate": 0.0002, "epoch": 0.6453978731206454, "step": 880}, {"loss": 1.2163, "grad_norm": 0.2948838770389557, "learning_rate": 0.0002, "epoch": 0.6527319398606527, "step": 890}, {"loss": 1.1081, "grad_norm": 0.38330280780792236, "learning_rate": 0.0002, "epoch": 0.6600660066006601, "step": 900}, {"loss": 1.1342, "grad_norm": 0.4428867697715759, "learning_rate": 0.0002, "epoch": 0.6674000733406674, "step": 910}, {"loss": 1.1021, "grad_norm": 0.23659265041351318, "learning_rate": 0.0002, "epoch": 0.6747341400806748, "step": 920}, {"loss": 1.1226, "grad_norm": 0.323685884475708, "learning_rate": 0.0002, "epoch": 0.682068206820682, "step": 930}, {"loss": 1.0853, "grad_norm": 0.39157727360725403, "learning_rate": 0.0002, "epoch": 0.6894022735606894, "step": 940}, {"loss": 1.1435, "grad_norm": 0.27189481258392334, "learning_rate": 0.0002, "epoch": 0.6967363403006968, "step": 950}, {"loss": 1.1033, "grad_norm": 0.529883861541748, "learning_rate": 0.0002, "epoch": 0.7040704070407041, "step": 960}, {"loss": 1.139, "grad_norm": 0.34758689999580383, "learning_rate": 0.0002, "epoch": 0.7114044737807114, "step": 970}, {"loss": 1.2197, "grad_norm": 0.831749439239502, "learning_rate": 0.0002, "epoch": 0.7187385405207187, "step": 980}, {"loss": 1.158, "grad_norm": 0.4438304007053375, "learning_rate": 0.0002, "epoch": 0.7260726072607261, "step": 990}, {"loss": 1.1021, "grad_norm": 0.33840006589889526, "learning_rate": 0.0002, "epoch": 0.7334066740007334, "step": 1000}, {"loss": 1.254, "grad_norm": 0.3454797863960266, "learning_rate": 0.0002, "epoch": 0.7407407407407407, "step": 1010}, {"loss": 1.106, "grad_norm": 0.38999441266059875, "learning_rate": 0.0002, "epoch": 0.7480748074807481, "step": 1020}, {"loss": 1.1428, "grad_norm": 0.2829911708831787, "learning_rate": 0.0002, "epoch": 0.7554088742207554, "step": 1030}, {"loss": 1.2123, "grad_norm": 0.36918163299560547, "learning_rate": 0.0002, "epoch": 0.7627429409607628, "step": 1040}, {"loss": 1.3028, "grad_norm": 0.3415680229663849, "learning_rate": 0.0002, "epoch": 0.77007700770077, "step": 1050}, {"loss": 1.1939, "grad_norm": 0.2974182963371277, "learning_rate": 0.0002, "epoch": 0.7774110744407774, "step": 1060}, {"loss": 1.194, "grad_norm": 0.3880919814109802, "learning_rate": 0.0002, "epoch": 0.7847451411807848, "step": 1070}, {"loss": 1.1095, "grad_norm": 0.33503302931785583, "learning_rate": 0.0002, "epoch": 0.7920792079207921, "step": 1080}, {"loss": 1.2111, "grad_norm": 0.3728407025337219, "learning_rate": 0.0002, "epoch": 0.7994132746607994, "step": 1090}, {"loss": 1.0835, "grad_norm": 0.3509373664855957, "learning_rate": 0.0002, "epoch": 0.8067473414008067, "step": 1100}, {"loss": 1.2661, "grad_norm": 0.42228564620018005, "learning_rate": 0.0002, "epoch": 0.8140814081408141, "step": 1110}, {"loss": 1.1788, "grad_norm": 0.313467800617218, "learning_rate": 0.0002, "epoch": 0.8214154748808215, "step": 1120}, {"loss": 1.1971, "grad_norm": 0.3378850817680359, "learning_rate": 0.0002, "epoch": 0.8287495416208287, "step": 1130}, {"loss": 1.1238, "grad_norm": 0.43200382590293884, "learning_rate": 0.0002, "epoch": 0.8360836083608361, "step": 1140}, {"loss": 1.3203, "grad_norm": 0.3309599459171295, "learning_rate": 0.0002, "epoch": 0.8434176751008434, "step": 1150}, {"loss": 1.1062, "grad_norm": 0.3526846170425415, "learning_rate": 0.0002, "epoch": 0.8507517418408508, "step": 1160}, {"loss": 1.0851, "grad_norm": 1.2722247838974, "learning_rate": 0.0002, "epoch": 0.858085808580858, "step": 1170}, {"loss": 1.0785, "grad_norm": 0.34142059087753296, "learning_rate": 0.0002, "epoch": 0.8654198753208654, "step": 1180}, {"loss": 1.2187, "grad_norm": 0.3805823028087616, "learning_rate": 0.0002, "epoch": 0.8727539420608728, "step": 1190}, {"loss": 1.1215, "grad_norm": 0.3931232690811157, "learning_rate": 0.0002, "epoch": 0.8800880088008801, "step": 1200}, {"loss": 1.0948, "grad_norm": 0.2937372624874115, "learning_rate": 0.0002, "epoch": 0.8874220755408874, "step": 1210}, {"loss": 1.1228, "grad_norm": 0.3757196366786957, "learning_rate": 0.0002, "epoch": 0.8947561422808947, "step": 1220}, {"loss": 1.1222, "grad_norm": 0.3502705991268158, "learning_rate": 0.0002, "epoch": 0.9020902090209021, "step": 1230}, {"loss": 1.2242, "grad_norm": 0.32758915424346924, "learning_rate": 0.0002, "epoch": 0.9094242757609095, "step": 1240}, {"loss": 1.215, "grad_norm": 0.37199416756629944, "learning_rate": 0.0002, "epoch": 0.9167583425009168, "step": 1250}, {"loss": 1.1225, "grad_norm": 0.3551490604877472, "learning_rate": 0.0002, "epoch": 0.9240924092409241, "step": 1260}, {"loss": 1.1966, "grad_norm": 0.2859550714492798, "learning_rate": 0.0002, "epoch": 0.9314264759809314, "step": 1270}, {"loss": 1.2186, "grad_norm": 0.427990585565567, "learning_rate": 0.0002, "epoch": 0.9387605427209388, "step": 1280}, {"loss": 1.2848, "grad_norm": 0.33717992901802063, "learning_rate": 0.0002, "epoch": 0.9460946094609461, "step": 1290}, {"loss": 1.1656, "grad_norm": 0.30225634574890137, "learning_rate": 0.0002, "epoch": 0.9534286762009534, "step": 1300}, {"loss": 1.2404, "grad_norm": 0.385821133852005, "learning_rate": 0.0002, "epoch": 0.9607627429409608, "step": 1310}, {"loss": 1.1932, "grad_norm": 0.35278066992759705, "learning_rate": 0.0002, "epoch": 0.9680968096809681, "step": 1320}, {"loss": 1.1071, "grad_norm": 0.49987098574638367, "learning_rate": 0.0002, "epoch": 0.9754308764209755, "step": 1330}, {"loss": 1.2259, "grad_norm": 0.3842747211456299, "learning_rate": 0.0002, "epoch": 0.9827649431609827, "step": 1340}, {"loss": 1.0862, "grad_norm": 0.6274653673171997, "learning_rate": 0.0002, "epoch": 0.9900990099009901, "step": 1350}, {"loss": 1.124, "grad_norm": 0.5239808559417725, "learning_rate": 0.0002, "epoch": 0.9974330766409975, "step": 1360}, {"eval_loss": 1.1822267770767212, "eval_runtime": 32.7389, "eval_samples_per_second": 13.165, "eval_steps_per_second": 1.649, "epoch": 0.9996332966629996, "step": 1363}, {"loss": 1.096, "grad_norm": 0.45311301946640015, "learning_rate": 0.0002, "epoch": 1.0047671433810048, "step": 1370}, {"loss": 1.0143, "grad_norm": 0.29685574769973755, "learning_rate": 0.0002, "epoch": 1.012101210121012, "step": 1380}, {"loss": 1.0302, "grad_norm": 0.3290937840938568, "learning_rate": 0.0002, "epoch": 1.0194352768610195, "step": 1390}, {"loss": 1.0295, "grad_norm": 0.3801758587360382, "learning_rate": 0.0002, "epoch": 1.0267693436010268, "step": 1400}, {"loss": 1.1226, "grad_norm": 0.794174313545227, "learning_rate": 0.0002, "epoch": 1.034103410341034, "step": 1410}, {"loss": 1.2232, "grad_norm": 0.3854154646396637, "learning_rate": 0.0002, "epoch": 1.0414374770810415, "step": 1420}, {"loss": 1.0652, "grad_norm": 0.32702451944351196, "learning_rate": 0.0002, "epoch": 1.0487715438210488, "step": 1430}, {"loss": 1.1144, "grad_norm": 0.7815203666687012, "learning_rate": 0.0002, "epoch": 1.056105610561056, "step": 1440}, {"loss": 1.1316, "grad_norm": 0.3087436854839325, "learning_rate": 0.0002, "epoch": 1.0634396773010635, "step": 1450}, {"loss": 1.1124, "grad_norm": 0.3847602903842926, "learning_rate": 0.0002, "epoch": 1.0707737440410707, "step": 1460}, {"loss": 1.1428, "grad_norm": 0.3693031370639801, "learning_rate": 0.0002, "epoch": 1.0781078107810782, "step": 1470}, {"loss": 1.0995, "grad_norm": 0.4111202359199524, "learning_rate": 0.0002, "epoch": 1.0854418775210855, "step": 1480}, {"loss": 1.0961, "grad_norm": 0.41452381014823914, "learning_rate": 0.0002, "epoch": 1.0927759442610927, "step": 1490}, {"loss": 1.1068, "grad_norm": 0.3336445093154907, "learning_rate": 0.0002, "epoch": 1.1001100110011002, "step": 1500}, {"loss": 1.0556, "grad_norm": 0.3923407793045044, "learning_rate": 0.0002, "epoch": 1.1074440777411074, "step": 1510}, {"loss": 1.1644, "grad_norm": 0.46215683221817017, "learning_rate": 0.0002, "epoch": 1.1147781444811147, "step": 1520}, {"loss": 1.1133, "grad_norm": 0.3592156767845154, "learning_rate": 0.0002, "epoch": 1.1221122112211221, "step": 1530}, {"loss": 1.0957, "grad_norm": 0.361110657453537, "learning_rate": 0.0002, "epoch": 1.1294462779611294, "step": 1540}, {"loss": 1.1553, "grad_norm": 0.5317131280899048, "learning_rate": 0.0002, "epoch": 1.1367803447011369, "step": 1550}, {"loss": 1.0368, "grad_norm": 0.3882388174533844, "learning_rate": 0.0002, "epoch": 1.1441144114411441, "step": 1560}, {"loss": 1.0805, "grad_norm": 0.3259428143501282, "learning_rate": 0.0002, "epoch": 1.1514484781811514, "step": 1570}, {"loss": 1.1819, "grad_norm": 0.410935640335083, "learning_rate": 0.0002, "epoch": 1.1587825449211588, "step": 1580}, {"loss": 1.1143, "grad_norm": 0.44940185546875, "learning_rate": 0.0002, "epoch": 1.166116611661166, "step": 1590}, {"loss": 1.0334, "grad_norm": 0.5106484293937683, "learning_rate": 0.0002, "epoch": 1.1734506784011733, "step": 1600}, {"loss": 1.2376, "grad_norm": 0.6603665947914124, "learning_rate": 0.0002, "epoch": 1.1807847451411808, "step": 1610}, {"loss": 1.1227, "grad_norm": 0.4799964129924774, "learning_rate": 0.0002, "epoch": 1.188118811881188, "step": 1620}, {"loss": 1.1191, "grad_norm": 0.4389883279800415, "learning_rate": 0.0002, "epoch": 1.1954528786211955, "step": 1630}, {"loss": 1.0667, "grad_norm": 0.4188813269138336, "learning_rate": 0.0002, "epoch": 1.2027869453612028, "step": 1640}, {"loss": 1.0605, "grad_norm": 0.7132157683372498, "learning_rate": 0.0002, "epoch": 1.21012101210121, "step": 1650}, {"loss": 1.0204, "grad_norm": 0.507480263710022, "learning_rate": 0.0002, "epoch": 1.2174550788412175, "step": 1660}, {"loss": 0.9948, "grad_norm": 0.9452332854270935, "learning_rate": 0.0002, "epoch": 1.2247891455812248, "step": 1670}, {"loss": 1.0228, "grad_norm": 0.4121614992618561, "learning_rate": 0.0002, "epoch": 1.2321232123212322, "step": 1680}, {"loss": 1.0366, "grad_norm": 0.34230247139930725, "learning_rate": 0.0002, "epoch": 1.2394572790612395, "step": 1690}, {"loss": 1.1289, "grad_norm": 0.4026208817958832, "learning_rate": 0.0002, "epoch": 1.2467913458012467, "step": 1700}, {"loss": 1.0206, "grad_norm": 0.46673697233200073, "learning_rate": 0.0002, "epoch": 1.2541254125412542, "step": 1710}, {"loss": 1.0827, "grad_norm": 0.38349825143814087, "learning_rate": 0.0002, "epoch": 1.2614594792812615, "step": 1720}, {"loss": 1.0356, "grad_norm": 0.4049997627735138, "learning_rate": 0.0002, "epoch": 1.2687935460212687, "step": 1730}, {"loss": 0.9504, "grad_norm": 0.3417615294456482, "learning_rate": 0.0002, "epoch": 1.2761276127612762, "step": 1740}, {"loss": 1.094, "grad_norm": 0.4277614951133728, "learning_rate": 0.0002, "epoch": 1.2834616795012834, "step": 1750}, {"loss": 0.9938, "grad_norm": 0.5864202976226807, "learning_rate": 0.0002, "epoch": 1.2907957462412907, "step": 1760}, {"loss": 1.1167, "grad_norm": 0.7097493410110474, "learning_rate": 0.0002, "epoch": 1.2981298129812981, "step": 1770}, {"loss": 1.1132, "grad_norm": 0.3145381212234497, "learning_rate": 0.0002, "epoch": 1.3054638797213054, "step": 1780}, {"loss": 1.1099, "grad_norm": 0.5116165280342102, "learning_rate": 0.0002, "epoch": 1.3127979464613129, "step": 1790}, {"loss": 1.0765, "grad_norm": 0.7469736337661743, "learning_rate": 0.0002, "epoch": 1.3201320132013201, "step": 1800}, {"loss": 1.0663, "grad_norm": 0.32272255420684814, "learning_rate": 0.0002, "epoch": 1.3274660799413276, "step": 1810}, {"loss": 0.9887, "grad_norm": 0.3534623086452484, "learning_rate": 0.0002, "epoch": 1.3348001466813348, "step": 1820}, {"loss": 1.1628, "grad_norm": 0.36127907037734985, "learning_rate": 0.0002, "epoch": 1.342134213421342, "step": 1830}, {"loss": 1.0972, "grad_norm": 0.4072401523590088, "learning_rate": 0.0002, "epoch": 1.3494682801613496, "step": 1840}, {"loss": 1.1267, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.3568023469013568, "step": 1850}, {"loss": 1.0173, "grad_norm": 0.412883460521698, "learning_rate": 0.0002, "epoch": 1.364136413641364, "step": 1860}, {"loss": 1.0265, "grad_norm": 0.3735875189304352, "learning_rate": 0.0002, "epoch": 1.3714704803813715, "step": 1870}, {"loss": 1.1061, "grad_norm": 0.39158159494400024, "learning_rate": 0.0002, "epoch": 1.3788045471213788, "step": 1880}, {"loss": 1.0433, "grad_norm": 0.44431769847869873, "learning_rate": 0.0002, "epoch": 1.386138613861386, "step": 1890}, {"loss": 1.0216, "grad_norm": 0.37772801518440247, "learning_rate": 0.0002, "epoch": 1.3934726806013935, "step": 1900}, {"loss": 1.0674, "grad_norm": 0.4056641757488251, "learning_rate": 0.0002, "epoch": 1.4008067473414008, "step": 1910}, {"loss": 1.0256, "grad_norm": 0.41612377762794495, "learning_rate": 0.0002, "epoch": 1.408140814081408, "step": 1920}, {"loss": 1.0467, "grad_norm": 0.41153013706207275, "learning_rate": 0.0002, "epoch": 1.4154748808214155, "step": 1930}, {"loss": 1.1062, "grad_norm": 0.387845516204834, "learning_rate": 0.0002, "epoch": 1.4228089475614227, "step": 1940}, {"loss": 1.1094, "grad_norm": 0.3809587061405182, "learning_rate": 0.0002, "epoch": 1.4301430143014302, "step": 1950}, {"loss": 1.0461, "grad_norm": 0.3625726103782654, "learning_rate": 0.0002, "epoch": 1.4374770810414375, "step": 1960}, {"loss": 0.9983, "grad_norm": 0.5294290781021118, "learning_rate": 0.0002, "epoch": 1.444811147781445, "step": 1970}, {"loss": 1.1114, "grad_norm": 0.39975494146347046, "learning_rate": 0.0002, "epoch": 1.4521452145214522, "step": 1980}, {"loss": 0.9704, "grad_norm": 0.4181167185306549, "learning_rate": 0.0002, "epoch": 1.4594792812614594, "step": 1990}, {"loss": 1.1146, "grad_norm": 0.42001503705978394, "learning_rate": 0.0002, "epoch": 1.466813348001467, "step": 2000}, {"loss": 1.1266, "grad_norm": 0.4877578616142273, "learning_rate": 0.0002, "epoch": 1.4741474147414741, "step": 2010}, {"loss": 1.1012, "grad_norm": 0.4050969183444977, "learning_rate": 0.0002, "epoch": 1.4814814814814814, "step": 2020}, {"loss": 1.0562, "grad_norm": 0.39068883657455444, "learning_rate": 0.0002, "epoch": 1.4888155482214889, "step": 2030}, {"loss": 1.0464, "grad_norm": 0.421282559633255, "learning_rate": 0.0002, "epoch": 1.4961496149614961, "step": 2040}, {"loss": 1.0532, "grad_norm": 0.47092297673225403, "learning_rate": 0.0002, "epoch": 1.5034836817015034, "step": 2050}, {"loss": 0.9348, "grad_norm": 0.39688974618911743, "learning_rate": 0.0002, "epoch": 1.5108177484415108, "step": 2060}, {"loss": 1.08, "grad_norm": 0.5529879331588745, "learning_rate": 0.0002, "epoch": 1.5181518151815183, "step": 2070}, {"loss": 1.1836, "grad_norm": 0.4879782199859619, "learning_rate": 0.0002, "epoch": 1.5254858819215253, "step": 2080}, {"loss": 1.0432, "grad_norm": 0.5517361164093018, "learning_rate": 0.0002, "epoch": 1.5328199486615328, "step": 2090}, {"loss": 1.0433, "grad_norm": 0.44015637040138245, "learning_rate": 0.0002, "epoch": 1.5401540154015403, "step": 2100}, {"loss": 1.1873, "grad_norm": 0.5435167551040649, "learning_rate": 0.0002, "epoch": 1.5474880821415475, "step": 2110}, {"loss": 1.1076, "grad_norm": 0.5714033246040344, "learning_rate": 0.0002, "epoch": 1.5548221488815548, "step": 2120}, {"loss": 1.1107, "grad_norm": 0.31732529401779175, "learning_rate": 0.0002, "epoch": 1.5621562156215623, "step": 2130}, {"loss": 1.0817, "grad_norm": 0.49068278074264526, "learning_rate": 0.0002, "epoch": 1.5694902823615695, "step": 2140}, {"loss": 1.0254, "grad_norm": 0.46851542592048645, "learning_rate": 0.0002, "epoch": 1.5768243491015768, "step": 2150}, {"loss": 1.0623, "grad_norm": 0.5083092451095581, "learning_rate": 0.0002, "epoch": 1.5841584158415842, "step": 2160}, {"loss": 1.0603, "grad_norm": 0.9822936058044434, "learning_rate": 0.0002, "epoch": 1.5914924825815915, "step": 2170}, {"loss": 0.9986, "grad_norm": 0.4575989246368408, "learning_rate": 0.0002, "epoch": 1.5988265493215987, "step": 2180}, {"loss": 1.1292, "grad_norm": 0.47444286942481995, "learning_rate": 0.0002, "epoch": 1.6061606160616062, "step": 2190}, {"loss": 1.0136, "grad_norm": 0.7208226919174194, "learning_rate": 0.0002, "epoch": 1.6134946828016135, "step": 2200}, {"loss": 1.15, "grad_norm": 0.43791481852531433, "learning_rate": 0.0002, "epoch": 1.6208287495416207, "step": 2210}, {"loss": 1.0961, "grad_norm": 0.5245792865753174, "learning_rate": 0.0002, "epoch": 1.6281628162816282, "step": 2220}, {"loss": 0.9957, "grad_norm": 0.39289429783821106, "learning_rate": 0.0002, "epoch": 1.6354968830216357, "step": 2230}, {"loss": 1.133, "grad_norm": 0.6106135845184326, "learning_rate": 0.0002, "epoch": 1.6428309497616427, "step": 2240}, {"loss": 1.0129, "grad_norm": 0.3722580671310425, "learning_rate": 0.0002, "epoch": 1.6501650165016502, "step": 2250}, {"loss": 1.0446, "grad_norm": 0.3649403750896454, "learning_rate": 0.0002, "epoch": 1.6574990832416576, "step": 2260}, {"loss": 1.0037, "grad_norm": 0.46514248847961426, "learning_rate": 0.0002, "epoch": 1.6648331499816649, "step": 2270}, {"loss": 1.0022, "grad_norm": 0.42034927010536194, "learning_rate": 0.0002, "epoch": 1.6721672167216721, "step": 2280}, {"loss": 1.1362, "grad_norm": 0.45202910900115967, "learning_rate": 0.0002, "epoch": 1.6795012834616796, "step": 2290}, {"loss": 1.0866, "grad_norm": 0.36257603764533997, "learning_rate": 0.0002, "epoch": 1.6868353502016868, "step": 2300}, {"loss": 1.0973, "grad_norm": 0.6340323090553284, "learning_rate": 0.0002, "epoch": 1.694169416941694, "step": 2310}, {"loss": 1.0615, "grad_norm": 0.4352878928184509, "learning_rate": 0.0002, "epoch": 1.7015034836817016, "step": 2320}, {"loss": 1.0629, "grad_norm": 0.45029792189598083, "learning_rate": 0.0002, "epoch": 1.7088375504217088, "step": 2330}, {"loss": 0.9621, "grad_norm": 0.3891315758228302, "learning_rate": 0.0002, "epoch": 1.716171617161716, "step": 2340}, {"loss": 0.9779, "grad_norm": 0.35180050134658813, "learning_rate": 0.0002, "epoch": 1.7235056839017235, "step": 2350}, {"loss": 1.0368, "grad_norm": 0.42367449402809143, "learning_rate": 0.0002, "epoch": 1.7308397506417308, "step": 2360}, {"loss": 1.0376, "grad_norm": 0.4553675353527069, "learning_rate": 0.0002, "epoch": 1.738173817381738, "step": 2370}, {"loss": 1.1467, "grad_norm": 0.5944654941558838, "learning_rate": 0.0002, "epoch": 1.7455078841217455, "step": 2380}, {"loss": 1.0548, "grad_norm": 0.3479664623737335, "learning_rate": 0.0002, "epoch": 1.752841950861753, "step": 2390}, {"loss": 1.0798, "grad_norm": 0.3585502505302429, "learning_rate": 0.0002, "epoch": 1.76017601760176, "step": 2400}, {"loss": 1.0983, "grad_norm": 0.4263346493244171, "learning_rate": 0.0002, "epoch": 1.7675100843417675, "step": 2410}, {"loss": 1.054, "grad_norm": 0.5476409196853638, "learning_rate": 0.0002, "epoch": 1.774844151081775, "step": 2420}, {"loss": 1.1615, "grad_norm": 0.3694186508655548, "learning_rate": 0.0002, "epoch": 1.7821782178217822, "step": 2430}, {"loss": 1.1343, "grad_norm": 0.9185658693313599, "learning_rate": 0.0002, "epoch": 1.7895122845617895, "step": 2440}, {"loss": 1.0764, "grad_norm": 0.7171908020973206, "learning_rate": 0.0002, "epoch": 1.796846351301797, "step": 2450}, {"loss": 1.1154, "grad_norm": 0.550658643245697, "learning_rate": 0.0002, "epoch": 1.8041804180418042, "step": 2460}, {"loss": 0.9975, "grad_norm": 0.4075568914413452, "learning_rate": 0.0002, "epoch": 1.8115144847818114, "step": 2470}, {"loss": 1.0935, "grad_norm": 0.3790127635002136, "learning_rate": 0.0002, "epoch": 1.818848551521819, "step": 2480}, {"loss": 0.9839, "grad_norm": 0.3576384484767914, "learning_rate": 0.0002, "epoch": 1.8261826182618262, "step": 2490}, {"loss": 1.1369, "grad_norm": 0.3919370770454407, "learning_rate": 0.0002, "epoch": 1.8335166850018334, "step": 2500}, {"loss": 0.9985, "grad_norm": 0.485083669424057, "learning_rate": 0.0002, "epoch": 1.8408507517418409, "step": 2510}, {"loss": 1.1585, "grad_norm": 0.4564347565174103, "learning_rate": 0.0002, "epoch": 1.8481848184818483, "step": 2520}, {"loss": 1.0944, "grad_norm": 0.3613106608390808, "learning_rate": 0.0002, "epoch": 1.8555188852218554, "step": 2530}, {"loss": 1.0819, "grad_norm": 0.39600759744644165, "learning_rate": 0.0002, "epoch": 1.8628529519618628, "step": 2540}, {"loss": 0.9453, "grad_norm": 1.123499870300293, "learning_rate": 0.0002, "epoch": 1.8701870187018703, "step": 2550}, {"loss": 1.0635, "grad_norm": 0.4612680673599243, "learning_rate": 0.0002, "epoch": 1.8775210854418776, "step": 2560}, {"loss": 1.0087, "grad_norm": 0.42745399475097656, "learning_rate": 0.0002, "epoch": 1.8848551521818848, "step": 2570}, {"loss": 1.0102, "grad_norm": 0.4055580198764801, "learning_rate": 0.0002, "epoch": 1.8921892189218923, "step": 2580}, {"loss": 1.0177, "grad_norm": 0.44174644351005554, "learning_rate": 0.0002, "epoch": 1.8995232856618995, "step": 2590}, {"loss": 0.9886, "grad_norm": 1.0228385925292969, "learning_rate": 0.0002, "epoch": 1.9068573524019068, "step": 2600}, {"loss": 1.0857, "grad_norm": 0.3496396243572235, "learning_rate": 0.0002, "epoch": 1.9141914191419143, "step": 2610}, {"loss": 1.0955, "grad_norm": 0.4191173017024994, "learning_rate": 0.0002, "epoch": 1.9215254858819215, "step": 2620}, {"loss": 1.0943, "grad_norm": 0.6778554916381836, "learning_rate": 0.0002, "epoch": 1.9288595526219288, "step": 2630}, {"loss": 1.0594, "grad_norm": 0.41992834210395813, "learning_rate": 0.0002, "epoch": 1.9361936193619362, "step": 2640}, {"loss": 1.1159, "grad_norm": 0.8760401010513306, "learning_rate": 0.0002, "epoch": 1.9435276861019435, "step": 2650}, {"loss": 1.0379, "grad_norm": 0.44049209356307983, "learning_rate": 0.0002, "epoch": 1.9508617528419507, "step": 2660}, {"loss": 1.1008, "grad_norm": 0.5651928782463074, "learning_rate": 0.0002, "epoch": 1.9581958195819582, "step": 2670}, {"loss": 1.1317, "grad_norm": 0.5292727947235107, "learning_rate": 0.0002, "epoch": 1.9655298863219657, "step": 2680}, {"loss": 1.1328, "grad_norm": 0.6012240648269653, "learning_rate": 0.0002, "epoch": 1.9728639530619727, "step": 2690}, {"loss": 1.0683, "grad_norm": 0.3945149779319763, "learning_rate": 0.0002, "epoch": 1.9801980198019802, "step": 2700}, {"loss": 1.0155, "grad_norm": 0.5732627511024475, "learning_rate": 0.0002, "epoch": 1.9875320865419877, "step": 2710}, {"loss": 0.9857, "grad_norm": 0.3963361084461212, "learning_rate": 0.0002, "epoch": 1.994866153281995, "step": 2720}, {"eval_loss": 1.1534006595611572, "eval_runtime": 32.7541, "eval_samples_per_second": 13.159, "eval_steps_per_second": 1.649, "epoch": 2.0, "step": 2727}, {"loss": 0.9624, "grad_norm": 0.48628315329551697, "learning_rate": 0.0002, "epoch": 2.002200220022002, "step": 2730}, {"loss": 0.9603, "grad_norm": 0.413875013589859, "learning_rate": 0.0002, "epoch": 2.0095342867620096, "step": 2740}, {"loss": 0.965, "grad_norm": 0.4988735616207123, "learning_rate": 0.0002, "epoch": 2.0168683535020167, "step": 2750}, {"loss": 0.9677, "grad_norm": 0.5634812712669373, "learning_rate": 0.0002, "epoch": 2.024202420242024, "step": 2760}, {"loss": 0.9547, "grad_norm": 0.48302653431892395, "learning_rate": 0.0002, "epoch": 2.0315364869820316, "step": 2770}, {"loss": 0.9346, "grad_norm": 0.49914175271987915, "learning_rate": 0.0002, "epoch": 2.038870553722039, "step": 2780}, {"loss": 0.904, "grad_norm": 1.14039945602417, "learning_rate": 0.0002, "epoch": 2.046204620462046, "step": 2790}, {"loss": 0.9588, "grad_norm": 0.6359720826148987, "learning_rate": 0.0002, "epoch": 2.0535386872020536, "step": 2800}, {"loss": 0.9031, "grad_norm": 0.4589158296585083, "learning_rate": 0.0002, "epoch": 2.060872753942061, "step": 2810}, {"loss": 0.9438, "grad_norm": 0.46255481243133545, "learning_rate": 0.0002, "epoch": 2.068206820682068, "step": 2820}, {"loss": 0.9464, "grad_norm": 0.6232137680053711, "learning_rate": 0.0002, "epoch": 2.0755408874220755, "step": 2830}, {"loss": 0.8978, "grad_norm": 0.41042178869247437, "learning_rate": 0.0002, "epoch": 2.082874954162083, "step": 2840}, {"loss": 0.8516, "grad_norm": 0.5334428548812866, "learning_rate": 0.0002, "epoch": 2.09020902090209, "step": 2850}, {"loss": 0.9313, "grad_norm": 0.8270058631896973, "learning_rate": 0.0002, "epoch": 2.0975430876420975, "step": 2860}, {"loss": 1.0064, "grad_norm": 0.6624533534049988, "learning_rate": 0.0002, "epoch": 2.104877154382105, "step": 2870}, {"loss": 0.9196, "grad_norm": 0.5448863506317139, "learning_rate": 0.0002, "epoch": 2.112211221122112, "step": 2880}, {"loss": 0.887, "grad_norm": 0.621482789516449, "learning_rate": 0.0002, "epoch": 2.1195452878621195, "step": 2890}, {"loss": 0.9702, "grad_norm": 0.4556255340576172, "learning_rate": 0.0002, "epoch": 2.126879354602127, "step": 2900}, {"loss": 0.9323, "grad_norm": 0.4620579183101654, "learning_rate": 0.0002, "epoch": 2.1342134213421344, "step": 2910}, {"loss": 0.836, "grad_norm": 0.9602415561676025, "learning_rate": 0.0002, "epoch": 2.1415474880821415, "step": 2920}, {"loss": 0.8826, "grad_norm": 0.587943971157074, "learning_rate": 0.0002, "epoch": 2.148881554822149, "step": 2930}, {"loss": 0.971, "grad_norm": 0.5121372938156128, "learning_rate": 0.0002, "epoch": 2.1562156215621564, "step": 2940}, {"loss": 0.8751, "grad_norm": 0.49424484372138977, "learning_rate": 0.0002, "epoch": 2.1635496883021634, "step": 2950}, {"loss": 0.8674, "grad_norm": 0.6312560439109802, "learning_rate": 0.0002, "epoch": 2.170883755042171, "step": 2960}, {"loss": 0.9791, "grad_norm": 0.5235576629638672, "learning_rate": 0.0002, "epoch": 2.1782178217821784, "step": 2970}, {"loss": 0.9706, "grad_norm": 0.5868439674377441, "learning_rate": 0.0002, "epoch": 2.1855518885221854, "step": 2980}, {"loss": 0.9338, "grad_norm": 0.42302873730659485, "learning_rate": 0.0002, "epoch": 2.192885955262193, "step": 2990}, {"loss": 0.9332, "grad_norm": 0.5097725987434387, "learning_rate": 0.0002, "epoch": 2.2002200220022003, "step": 3000}, {"loss": 0.9239, "grad_norm": 0.5091572403907776, "learning_rate": 0.0002, "epoch": 2.2075540887422074, "step": 3010}, {"loss": 0.8898, "grad_norm": 0.49433162808418274, "learning_rate": 0.0002, "epoch": 2.214888155482215, "step": 3020}, {"loss": 0.9734, "grad_norm": 0.5577368140220642, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3030}, {"loss": 0.9033, "grad_norm": 0.6177583932876587, "learning_rate": 0.0002, "epoch": 2.2295562889622293, "step": 3040}, {"loss": 0.9882, "grad_norm": 0.5256719589233398, "learning_rate": 0.0002, "epoch": 2.236890355702237, "step": 3050}, {"loss": 0.9439, "grad_norm": 0.5001118183135986, "learning_rate": 0.0002, "epoch": 2.2442244224422443, "step": 3060}, {"loss": 0.8718, "grad_norm": 0.5721249580383301, "learning_rate": 0.0002, "epoch": 2.2515584891822513, "step": 3070}, {"loss": 1.0648, "grad_norm": 0.5325384140014648, "learning_rate": 0.0002, "epoch": 2.258892555922259, "step": 3080}, {"loss": 0.9843, "grad_norm": 0.5719189047813416, "learning_rate": 0.0002, "epoch": 2.2662266226622663, "step": 3090}, {"loss": 0.8633, "grad_norm": 0.6337835788726807, "learning_rate": 0.0002, "epoch": 2.2735606894022737, "step": 3100}, {"loss": 0.9962, "grad_norm": 0.5381836891174316, "learning_rate": 0.0002, "epoch": 2.2808947561422808, "step": 3110}, {"loss": 0.8265, "grad_norm": 0.5408531427383423, "learning_rate": 0.0002, "epoch": 2.2882288228822882, "step": 3120}, {"loss": 1.0325, "grad_norm": 0.43705281615257263, "learning_rate": 0.0002, "epoch": 2.2955628896222957, "step": 3130}, {"loss": 0.9388, "grad_norm": 0.6454030275344849, "learning_rate": 0.0002, "epoch": 2.3028969563623027, "step": 3140}, {"loss": 0.954, "grad_norm": 0.686030387878418, "learning_rate": 0.0002, "epoch": 2.31023102310231, "step": 3150}, {"loss": 0.9403, "grad_norm": 0.5123633146286011, "learning_rate": 0.0002, "epoch": 2.3175650898423177, "step": 3160}, {"loss": 0.8834, "grad_norm": 0.842506468296051, "learning_rate": 0.0002, "epoch": 2.3248991565823247, "step": 3170}, {"loss": 1.0497, "grad_norm": 0.5193818807601929, "learning_rate": 0.0002, "epoch": 2.332233223322332, "step": 3180}, {"loss": 0.9473, "grad_norm": 0.5634409189224243, "learning_rate": 0.0002, "epoch": 2.3395672900623397, "step": 3190}, {"loss": 0.8499, "grad_norm": 0.6475534439086914, "learning_rate": 0.0002, "epoch": 2.3469013568023467, "step": 3200}, {"loss": 0.874, "grad_norm": 1.1503914594650269, "learning_rate": 0.0002, "epoch": 2.354235423542354, "step": 3210}, {"loss": 0.9762, "grad_norm": 0.7234905362129211, "learning_rate": 0.0002, "epoch": 2.3615694902823616, "step": 3220}, {"loss": 0.9007, "grad_norm": 0.664903461933136, "learning_rate": 0.0002, "epoch": 2.368903557022369, "step": 3230}, {"loss": 0.9987, "grad_norm": 0.5453006625175476, "learning_rate": 0.0002, "epoch": 2.376237623762376, "step": 3240}, {"loss": 0.9742, "grad_norm": 0.6256654262542725, "learning_rate": 0.0002, "epoch": 2.3835716905023836, "step": 3250}, {"loss": 0.9922, "grad_norm": 0.5166565179824829, "learning_rate": 0.0002, "epoch": 2.390905757242391, "step": 3260}, {"loss": 0.927, "grad_norm": 0.5699098110198975, "learning_rate": 0.0002, "epoch": 2.398239823982398, "step": 3270}, {"loss": 0.8878, "grad_norm": 0.4472540020942688, "learning_rate": 0.0002, "epoch": 2.4055738907224056, "step": 3280}, {"loss": 0.9439, "grad_norm": 0.6790403127670288, "learning_rate": 0.0002, "epoch": 2.412907957462413, "step": 3290}, {"loss": 0.972, "grad_norm": 0.5182185173034668, "learning_rate": 0.0002, "epoch": 2.42024202420242, "step": 3300}, {"loss": 0.9775, "grad_norm": 0.564647912979126, "learning_rate": 0.0002, "epoch": 2.4275760909424275, "step": 3310}, {"loss": 1.072, "grad_norm": 0.5625313520431519, "learning_rate": 0.0002, "epoch": 2.434910157682435, "step": 3320}, {"loss": 0.8798, "grad_norm": 0.7496559619903564, "learning_rate": 0.0002, "epoch": 2.442244224422442, "step": 3330}, {"loss": 0.868, "grad_norm": 0.4779128134250641, "learning_rate": 0.0002, "epoch": 2.4495782911624495, "step": 3340}, {"loss": 1.0316, "grad_norm": 0.578093409538269, "learning_rate": 0.0002, "epoch": 2.456912357902457, "step": 3350}, {"loss": 0.9282, "grad_norm": 0.5456080436706543, "learning_rate": 0.0002, "epoch": 2.4642464246424645, "step": 3360}, {"loss": 0.8409, "grad_norm": 0.4769273102283478, "learning_rate": 0.0002, "epoch": 2.4715804913824715, "step": 3370}, {"loss": 0.9312, "grad_norm": 0.5608189702033997, "learning_rate": 0.0002, "epoch": 2.478914558122479, "step": 3380}, {"loss": 0.9934, "grad_norm": 0.5590165853500366, "learning_rate": 0.0002, "epoch": 2.4862486248624864, "step": 3390}, {"loss": 1.025, "grad_norm": 0.801306962966919, "learning_rate": 0.0002, "epoch": 2.4935826916024935, "step": 3400}, {"loss": 0.9049, "grad_norm": 0.6045624613761902, "learning_rate": 0.0002, "epoch": 2.500916758342501, "step": 3410}, {"loss": 0.944, "grad_norm": 0.5735858082771301, "learning_rate": 0.0002, "epoch": 2.5082508250825084, "step": 3420}, {"loss": 0.9846, "grad_norm": 0.6827309131622314, "learning_rate": 0.0002, "epoch": 2.5155848918225154, "step": 3430}, {"loss": 0.9789, "grad_norm": 0.5702602863311768, "learning_rate": 0.0002, "epoch": 2.522918958562523, "step": 3440}, {"loss": 0.9127, "grad_norm": 0.6674721240997314, "learning_rate": 0.0002, "epoch": 2.5302530253025304, "step": 3450}, {"loss": 0.914, "grad_norm": 0.5635907649993896, "learning_rate": 0.0002, "epoch": 2.5375870920425374, "step": 3460}, {"loss": 0.8398, "grad_norm": 0.42737770080566406, "learning_rate": 0.0002, "epoch": 2.544921158782545, "step": 3470}, {"loss": 0.9474, "grad_norm": 0.6720691919326782, "learning_rate": 0.0002, "epoch": 2.5522552255225524, "step": 3480}, {"loss": 0.8637, "grad_norm": 0.8917084336280823, "learning_rate": 0.0002, "epoch": 2.55958929226256, "step": 3490}, {"loss": 0.9257, "grad_norm": 0.5134549140930176, "learning_rate": 0.0002, "epoch": 2.566923359002567, "step": 3500}, {"loss": 0.9362, "grad_norm": 0.4951367974281311, "learning_rate": 0.0002, "epoch": 2.5742574257425743, "step": 3510}, {"loss": 0.9184, "grad_norm": 0.9438204765319824, "learning_rate": 0.0002, "epoch": 2.5815914924825814, "step": 3520}, {"loss": 0.8939, "grad_norm": 0.6024714708328247, "learning_rate": 0.0002, "epoch": 2.588925559222589, "step": 3530}, {"loss": 0.9298, "grad_norm": 0.5248535871505737, "learning_rate": 0.0002, "epoch": 2.5962596259625963, "step": 3540}, {"loss": 0.941, "grad_norm": 0.8677568435668945, "learning_rate": 0.0002, "epoch": 2.6035936927026038, "step": 3550}, {"loss": 0.9253, "grad_norm": 0.82008296251297, "learning_rate": 0.0002, "epoch": 2.610927759442611, "step": 3560}, {"loss": 0.8429, "grad_norm": 0.4724634885787964, "learning_rate": 0.0002, "epoch": 2.6182618261826183, "step": 3570}, {"loss": 0.9058, "grad_norm": 0.5434244275093079, "learning_rate": 0.0002, "epoch": 2.6255958929226257, "step": 3580}, {"loss": 0.9379, "grad_norm": 0.4948740005493164, "learning_rate": 0.0002, "epoch": 2.6329299596626328, "step": 3590}, {"loss": 0.8718, "grad_norm": 0.42109328508377075, "learning_rate": 0.0002, "epoch": 2.6402640264026402, "step": 3600}, {"loss": 0.9809, "grad_norm": 0.7979786396026611, "learning_rate": 0.0002, "epoch": 2.6475980931426477, "step": 3610}, {"loss": 0.9229, "grad_norm": 0.6345919370651245, "learning_rate": 0.0002, "epoch": 2.654932159882655, "step": 3620}, {"loss": 0.8506, "grad_norm": 0.4971671402454376, "learning_rate": 0.0002, "epoch": 2.662266226622662, "step": 3630}, {"loss": 0.8054, "grad_norm": 0.6467748284339905, "learning_rate": 0.0002, "epoch": 2.6696002933626697, "step": 3640}, {"loss": 0.9277, "grad_norm": 0.4240160286426544, "learning_rate": 0.0002, "epoch": 2.6769343601026767, "step": 3650}, {"loss": 0.8213, "grad_norm": 0.5179754495620728, "learning_rate": 0.0002, "epoch": 2.684268426842684, "step": 3660}, {"loss": 0.9221, "grad_norm": 0.754012405872345, "learning_rate": 0.0002, "epoch": 2.6916024935826917, "step": 3670}, {"loss": 0.9194, "grad_norm": 0.5141299962997437, "learning_rate": 0.0002, "epoch": 2.698936560322699, "step": 3680}, {"loss": 0.9495, "grad_norm": 0.5737819075584412, "learning_rate": 0.0002, "epoch": 2.706270627062706, "step": 3690}, {"loss": 1.0162, "grad_norm": 0.5887577533721924, "learning_rate": 0.0002, "epoch": 2.7136046938027136, "step": 3700}, {"loss": 0.9169, "grad_norm": 0.6740471720695496, "learning_rate": 0.0002, "epoch": 2.720938760542721, "step": 3710}, {"loss": 0.9297, "grad_norm": 0.5879453420639038, "learning_rate": 0.0002, "epoch": 2.728272827282728, "step": 3720}, {"loss": 0.9358, "grad_norm": 0.4858354926109314, "learning_rate": 0.0002, "epoch": 2.7356068940227356, "step": 3730}, {"loss": 0.9308, "grad_norm": 0.5489001870155334, "learning_rate": 0.0002, "epoch": 2.742940960762743, "step": 3740}, {"loss": 0.894, "grad_norm": 0.8187092542648315, "learning_rate": 0.0002, "epoch": 2.7502750275027505, "step": 3750}, {"loss": 0.8954, "grad_norm": 0.5666626691818237, "learning_rate": 0.0002, "epoch": 2.7576090942427576, "step": 3760}, {"loss": 1.0059, "grad_norm": 0.5377066135406494, "learning_rate": 0.0002, "epoch": 2.764943160982765, "step": 3770}, {"loss": 0.9132, "grad_norm": 0.566330075263977, "learning_rate": 0.0002, "epoch": 2.772277227722772, "step": 3780}, {"loss": 0.9415, "grad_norm": 0.5522832870483398, "learning_rate": 0.0002, "epoch": 2.7796112944627795, "step": 3790}, {"loss": 0.8816, "grad_norm": 0.5668695569038391, "learning_rate": 0.0002, "epoch": 2.786945361202787, "step": 3800}, {"loss": 0.8885, "grad_norm": 0.7566602826118469, "learning_rate": 0.0002, "epoch": 2.7942794279427945, "step": 3810}, {"loss": 0.8598, "grad_norm": 0.5603684782981873, "learning_rate": 0.0002, "epoch": 2.8016134946828015, "step": 3820}, {"loss": 0.9602, "grad_norm": 0.49122217297554016, "learning_rate": 0.0002, "epoch": 2.808947561422809, "step": 3830}, {"loss": 0.9738, "grad_norm": 0.6798251867294312, "learning_rate": 0.0002, "epoch": 2.816281628162816, "step": 3840}, {"loss": 0.9533, "grad_norm": 0.6097991466522217, "learning_rate": 0.0002, "epoch": 2.8236156949028235, "step": 3850}, {"loss": 0.8672, "grad_norm": 0.6675726175308228, "learning_rate": 0.0002, "epoch": 2.830949761642831, "step": 3860}, {"loss": 0.9324, "grad_norm": 0.9223952889442444, "learning_rate": 0.0002, "epoch": 2.8382838283828384, "step": 3870}, {"loss": 0.8767, "grad_norm": 0.6020799875259399, "learning_rate": 0.0002, "epoch": 2.8456178951228455, "step": 3880}, {"loss": 0.9148, "grad_norm": 0.5206381678581238, "learning_rate": 0.0002, "epoch": 2.852951961862853, "step": 3890}, {"loss": 0.9479, "grad_norm": 0.6268777251243591, "learning_rate": 0.0002, "epoch": 2.8602860286028604, "step": 3900}, {"loss": 0.9409, "grad_norm": 1.1583497524261475, "learning_rate": 0.0002, "epoch": 2.8676200953428674, "step": 3910}, {"loss": 0.895, "grad_norm": 0.7263903021812439, "learning_rate": 0.0002, "epoch": 2.874954162082875, "step": 3920}, {"loss": 0.8786, "grad_norm": 0.5369910001754761, "learning_rate": 0.0002, "epoch": 2.8822882288228824, "step": 3930}, {"loss": 1.0015, "grad_norm": 0.7298350930213928, "learning_rate": 0.0002, "epoch": 2.88962229556289, "step": 3940}, {"loss": 0.979, "grad_norm": 0.577012836933136, "learning_rate": 0.0002, "epoch": 2.896956362302897, "step": 3950}, {"loss": 0.9716, "grad_norm": 0.5859594345092773, "learning_rate": 0.0002, "epoch": 2.9042904290429044, "step": 3960}, {"loss": 0.8772, "grad_norm": 0.47176122665405273, "learning_rate": 0.0002, "epoch": 2.9116244957829114, "step": 3970}, {"loss": 0.8997, "grad_norm": 0.9699620604515076, "learning_rate": 0.0002, "epoch": 2.918958562522919, "step": 3980}, {"loss": 0.9057, "grad_norm": 0.7908747792243958, "learning_rate": 0.0002, "epoch": 2.9262926292629263, "step": 3990}, {"loss": 0.9462, "grad_norm": 0.5777379274368286, "learning_rate": 0.0002, "epoch": 2.933626696002934, "step": 4000}, {"loss": 0.9358, "grad_norm": 0.599288284778595, "learning_rate": 0.0002, "epoch": 2.940960762742941, "step": 4010}, {"loss": 0.9812, "grad_norm": 0.5232274532318115, "learning_rate": 0.0002, "epoch": 2.9482948294829483, "step": 4020}, {"loss": 0.96, "grad_norm": 0.6395137310028076, "learning_rate": 0.0002, "epoch": 2.9556288962229558, "step": 4030}, {"loss": 0.9813, "grad_norm": 0.589260458946228, "learning_rate": 0.0002, "epoch": 2.962962962962963, "step": 4040}, {"loss": 0.9541, "grad_norm": 0.5699581503868103, "learning_rate": 0.0002, "epoch": 2.9702970297029703, "step": 4050}, {"loss": 0.9585, "grad_norm": 0.528468132019043, "learning_rate": 0.0002, "epoch": 2.9776310964429777, "step": 4060}, {"loss": 0.9164, "grad_norm": 0.4804670512676239, "learning_rate": 0.0002, "epoch": 2.984965163182985, "step": 4070}, {"loss": 0.9771, "grad_norm": 1.1918889284133911, "learning_rate": 0.0002, "epoch": 2.9922992299229922, "step": 4080}, {"loss": 0.9178, "grad_norm": 0.5479103326797485, "learning_rate": 0.0002, "epoch": 2.9996332966629997, "step": 4090}]}
+{"epoch": 4.0, "step": 5454, "epoch_duration": 1471.1494228839874, "total_accumulated_duration": 5905.44949555397, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7568.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 1.9722, "grad_norm": 0.47521963715553284, "learning_rate": 0.0002, "epoch": 0.007334066740007334, "step": 10}, {"loss": 1.4821, "grad_norm": 0.5395162105560303, "learning_rate": 0.0002, "epoch": 0.014668133480014669, "step": 20}, {"loss": 1.4202, "grad_norm": 0.4305780231952667, "learning_rate": 0.0002, "epoch": 0.022002200220022004, "step": 30}, {"loss": 1.4271, "grad_norm": 0.6938246488571167, "learning_rate": 0.0002, "epoch": 0.029336266960029337, "step": 40}, {"loss": 1.3112, "grad_norm": 1.5133819580078125, "learning_rate": 0.0002, "epoch": 0.03667033370003667, "step": 50}, {"loss": 1.3132, "grad_norm": 0.9173883199691772, "learning_rate": 0.0002, "epoch": 0.04400440044004401, "step": 60}, {"loss": 1.2844, "grad_norm": 0.4619861841201782, "learning_rate": 0.0002, "epoch": 0.05133846718005134, "step": 70}, {"loss": 1.2108, "grad_norm": 0.46118637919425964, "learning_rate": 0.0002, "epoch": 0.058672533920058674, "step": 80}, {"loss": 1.3441, "grad_norm": 0.4468648135662079, "learning_rate": 0.0002, "epoch": 0.066006600660066, "step": 90}, {"loss": 1.1863, "grad_norm": 0.46123769879341125, "learning_rate": 0.0002, "epoch": 0.07334066740007333, "step": 100}, {"loss": 1.2772, "grad_norm": 0.4859139025211334, "learning_rate": 0.0002, "epoch": 0.08067473414008068, "step": 110}, {"loss": 1.2087, "grad_norm": 0.4384922385215759, "learning_rate": 0.0002, "epoch": 0.08800880088008801, "step": 120}, {"loss": 1.2927, "grad_norm": 0.39519360661506653, "learning_rate": 0.0002, "epoch": 0.09534286762009535, "step": 130}, {"loss": 1.2349, "grad_norm": 0.4049859344959259, "learning_rate": 0.0002, "epoch": 0.10267693436010268, "step": 140}, {"loss": 1.293, "grad_norm": 0.4605638086795807, "learning_rate": 0.0002, "epoch": 0.11001100110011001, "step": 150}, {"loss": 1.2659, "grad_norm": 0.4201928377151489, "learning_rate": 0.0002, "epoch": 0.11734506784011735, "step": 160}, {"loss": 1.3961, "grad_norm": 0.5367777347564697, "learning_rate": 0.0002, "epoch": 0.12467913458012468, "step": 170}, {"loss": 1.2481, "grad_norm": 0.41752299666404724, "learning_rate": 0.0002, "epoch": 0.132013201320132, "step": 180}, {"loss": 1.207, "grad_norm": 0.31597763299942017, "learning_rate": 0.0002, "epoch": 0.13934726806013933, "step": 190}, {"loss": 1.2441, "grad_norm": 0.7468788623809814, "learning_rate": 0.0002, "epoch": 0.14668133480014667, "step": 200}, {"loss": 1.199, "grad_norm": 0.3403034508228302, "learning_rate": 0.0002, "epoch": 0.15401540154015403, "step": 210}, {"loss": 1.2439, "grad_norm": 0.34240293502807617, "learning_rate": 0.0002, "epoch": 0.16134946828016136, "step": 220}, {"loss": 1.2022, "grad_norm": 0.356158971786499, "learning_rate": 0.0002, "epoch": 0.1686835350201687, "step": 230}, {"loss": 1.207, "grad_norm": 0.3448857367038727, "learning_rate": 0.0002, "epoch": 0.17601760176017603, "step": 240}, {"loss": 1.2156, "grad_norm": 0.3475699722766876, "learning_rate": 0.0002, "epoch": 0.18335166850018336, "step": 250}, {"loss": 1.1551, "grad_norm": 0.2770358622074127, "learning_rate": 0.0002, "epoch": 0.1906857352401907, "step": 260}, {"loss": 1.2238, "grad_norm": 0.4310270845890045, "learning_rate": 0.0002, "epoch": 0.19801980198019803, "step": 270}, {"loss": 1.2917, "grad_norm": 0.335041880607605, "learning_rate": 0.0002, "epoch": 0.20535386872020536, "step": 280}, {"loss": 1.0959, "grad_norm": 0.3420602083206177, "learning_rate": 0.0002, "epoch": 0.2126879354602127, "step": 290}, {"loss": 1.1232, "grad_norm": 0.325001060962677, "learning_rate": 0.0002, "epoch": 0.22002200220022003, "step": 300}, {"loss": 1.2007, "grad_norm": 0.3027827739715576, "learning_rate": 0.0002, "epoch": 0.22735606894022736, "step": 310}, {"loss": 1.1803, "grad_norm": 0.435550719499588, "learning_rate": 0.0002, "epoch": 0.2346901356802347, "step": 320}, {"loss": 1.2045, "grad_norm": 0.3884522616863251, "learning_rate": 0.0002, "epoch": 0.24202420242024203, "step": 330}, {"loss": 1.2481, "grad_norm": 0.7736002206802368, "learning_rate": 0.0002, "epoch": 0.24935826916024936, "step": 340}, {"loss": 1.3606, "grad_norm": 0.35052821040153503, "learning_rate": 0.0002, "epoch": 0.2566923359002567, "step": 350}, {"loss": 1.2129, "grad_norm": 0.3311890959739685, "learning_rate": 0.0002, "epoch": 0.264026402640264, "step": 360}, {"loss": 1.2219, "grad_norm": 0.7473500370979309, "learning_rate": 0.0002, "epoch": 0.27136046938027136, "step": 370}, {"loss": 1.2712, "grad_norm": 0.3681875765323639, "learning_rate": 0.0002, "epoch": 0.27869453612027867, "step": 380}, {"loss": 1.2258, "grad_norm": 0.3764737844467163, "learning_rate": 0.0002, "epoch": 0.28602860286028603, "step": 390}, {"loss": 1.1917, "grad_norm": 0.4243989586830139, "learning_rate": 0.0002, "epoch": 0.29336266960029334, "step": 400}, {"loss": 1.199, "grad_norm": 0.2658531963825226, "learning_rate": 0.0002, "epoch": 0.3006967363403007, "step": 410}, {"loss": 1.1622, "grad_norm": 0.3436793386936188, "learning_rate": 0.0002, "epoch": 0.30803080308030806, "step": 420}, {"loss": 1.2953, "grad_norm": 0.5101129412651062, "learning_rate": 0.0002, "epoch": 0.31536486982031536, "step": 430}, {"loss": 1.1557, "grad_norm": 0.3319750726222992, "learning_rate": 0.0002, "epoch": 0.3226989365603227, "step": 440}, {"loss": 1.1804, "grad_norm": 0.385148286819458, "learning_rate": 0.0002, "epoch": 0.33003300330033003, "step": 450}, {"loss": 1.1808, "grad_norm": 0.3477935791015625, "learning_rate": 0.0002, "epoch": 0.3373670700403374, "step": 460}, {"loss": 1.1877, "grad_norm": 0.29748716950416565, "learning_rate": 0.0002, "epoch": 0.3447011367803447, "step": 470}, {"loss": 1.19, "grad_norm": 0.34083324670791626, "learning_rate": 0.0002, "epoch": 0.35203520352035206, "step": 480}, {"loss": 1.2, "grad_norm": 0.36904552578926086, "learning_rate": 0.0002, "epoch": 0.35936927026035936, "step": 490}, {"loss": 1.2223, "grad_norm": 0.315483033657074, "learning_rate": 0.0002, "epoch": 0.3667033370003667, "step": 500}, {"loss": 1.1461, "grad_norm": 0.44897955656051636, "learning_rate": 0.0002, "epoch": 0.37403740374037403, "step": 510}, {"loss": 1.3035, "grad_norm": 0.3160701394081116, "learning_rate": 0.0002, "epoch": 0.3813714704803814, "step": 520}, {"loss": 1.3197, "grad_norm": 0.29584741592407227, "learning_rate": 0.0002, "epoch": 0.3887055372203887, "step": 530}, {"loss": 1.2983, "grad_norm": 0.5430002808570862, "learning_rate": 0.0002, "epoch": 0.39603960396039606, "step": 540}, {"loss": 1.2459, "grad_norm": 0.2908070683479309, "learning_rate": 0.0002, "epoch": 0.40337367070040336, "step": 550}, {"loss": 1.2384, "grad_norm": 0.35066530108451843, "learning_rate": 0.0002, "epoch": 0.4107077374404107, "step": 560}, {"loss": 1.1784, "grad_norm": 0.37588003277778625, "learning_rate": 0.0002, "epoch": 0.41804180418041803, "step": 570}, {"loss": 1.2334, "grad_norm": 0.3112126886844635, "learning_rate": 0.0002, "epoch": 0.4253758709204254, "step": 580}, {"loss": 1.1439, "grad_norm": 0.35577139258384705, "learning_rate": 0.0002, "epoch": 0.4327099376604327, "step": 590}, {"loss": 1.184, "grad_norm": 0.31706422567367554, "learning_rate": 0.0002, "epoch": 0.44004400440044006, "step": 600}, {"loss": 1.2081, "grad_norm": 0.3249092102050781, "learning_rate": 0.0002, "epoch": 0.44737807114044736, "step": 610}, {"loss": 1.0824, "grad_norm": 0.3842705488204956, "learning_rate": 0.0002, "epoch": 0.4547121378804547, "step": 620}, {"loss": 1.2257, "grad_norm": 0.390991747379303, "learning_rate": 0.0002, "epoch": 0.46204620462046203, "step": 630}, {"loss": 1.1954, "grad_norm": 0.27532413601875305, "learning_rate": 0.0002, "epoch": 0.4693802713604694, "step": 640}, {"loss": 1.1058, "grad_norm": 0.31412816047668457, "learning_rate": 0.0002, "epoch": 0.4767143381004767, "step": 650}, {"loss": 1.1312, "grad_norm": 0.32117101550102234, "learning_rate": 0.0002, "epoch": 0.48404840484048406, "step": 660}, {"loss": 1.2423, "grad_norm": 0.3810010254383087, "learning_rate": 0.0002, "epoch": 0.49138247158049136, "step": 670}, {"loss": 1.1978, "grad_norm": 0.36289164423942566, "learning_rate": 0.0002, "epoch": 0.4987165383204987, "step": 680}, {"loss": 1.2034, "grad_norm": 0.34458720684051514, "learning_rate": 0.0002, "epoch": 0.506050605060506, "step": 690}, {"loss": 1.1756, "grad_norm": 0.32844600081443787, "learning_rate": 0.0002, "epoch": 0.5133846718005134, "step": 700}, {"loss": 1.0807, "grad_norm": 0.3144175708293915, "learning_rate": 0.0002, "epoch": 0.5207187385405208, "step": 710}, {"loss": 1.1952, "grad_norm": 0.3898887634277344, "learning_rate": 0.0002, "epoch": 0.528052805280528, "step": 720}, {"loss": 1.1244, "grad_norm": 1.3220758438110352, "learning_rate": 0.0002, "epoch": 0.5353868720205354, "step": 730}, {"loss": 1.227, "grad_norm": 0.3635874390602112, "learning_rate": 0.0002, "epoch": 0.5427209387605427, "step": 740}, {"loss": 1.2169, "grad_norm": 0.3138217628002167, "learning_rate": 0.0002, "epoch": 0.5500550055005501, "step": 750}, {"loss": 1.1516, "grad_norm": 0.4063207805156708, "learning_rate": 0.0002, "epoch": 0.5573890722405573, "step": 760}, {"loss": 1.1954, "grad_norm": 0.3926219940185547, "learning_rate": 0.0002, "epoch": 0.5647231389805647, "step": 770}, {"loss": 1.1726, "grad_norm": 0.31954652070999146, "learning_rate": 0.0002, "epoch": 0.5720572057205721, "step": 780}, {"loss": 1.2977, "grad_norm": 0.4248711168766022, "learning_rate": 0.0002, "epoch": 0.5793912724605794, "step": 790}, {"loss": 1.1728, "grad_norm": 0.643004834651947, "learning_rate": 0.0002, "epoch": 0.5867253392005867, "step": 800}, {"loss": 1.1793, "grad_norm": 0.3479592800140381, "learning_rate": 0.0002, "epoch": 0.594059405940594, "step": 810}, {"loss": 1.2426, "grad_norm": 0.4684754014015198, "learning_rate": 0.0002, "epoch": 0.6013934726806014, "step": 820}, {"loss": 1.2002, "grad_norm": 0.3739790916442871, "learning_rate": 0.0002, "epoch": 0.6087275394206088, "step": 830}, {"loss": 1.2139, "grad_norm": 0.40884748101234436, "learning_rate": 0.0002, "epoch": 0.6160616061606161, "step": 840}, {"loss": 1.1557, "grad_norm": 0.9722164273262024, "learning_rate": 0.0002, "epoch": 0.6233956729006234, "step": 850}, {"loss": 1.3069, "grad_norm": 0.42992347478866577, "learning_rate": 0.0002, "epoch": 0.6307297396406307, "step": 860}, {"loss": 1.1339, "grad_norm": 0.36654195189476013, "learning_rate": 0.0002, "epoch": 0.6380638063806381, "step": 870}, {"loss": 1.1932, "grad_norm": 0.4113832116127014, "learning_rate": 0.0002, "epoch": 0.6453978731206454, "step": 880}, {"loss": 1.2163, "grad_norm": 0.2948838770389557, "learning_rate": 0.0002, "epoch": 0.6527319398606527, "step": 890}, {"loss": 1.1081, "grad_norm": 0.38330280780792236, "learning_rate": 0.0002, "epoch": 0.6600660066006601, "step": 900}, {"loss": 1.1342, "grad_norm": 0.4428867697715759, "learning_rate": 0.0002, "epoch": 0.6674000733406674, "step": 910}, {"loss": 1.1021, "grad_norm": 0.23659265041351318, "learning_rate": 0.0002, "epoch": 0.6747341400806748, "step": 920}, {"loss": 1.1226, "grad_norm": 0.323685884475708, "learning_rate": 0.0002, "epoch": 0.682068206820682, "step": 930}, {"loss": 1.0853, "grad_norm": 0.39157727360725403, "learning_rate": 0.0002, "epoch": 0.6894022735606894, "step": 940}, {"loss": 1.1435, "grad_norm": 0.27189481258392334, "learning_rate": 0.0002, "epoch": 0.6967363403006968, "step": 950}, {"loss": 1.1033, "grad_norm": 0.529883861541748, "learning_rate": 0.0002, "epoch": 0.7040704070407041, "step": 960}, {"loss": 1.139, "grad_norm": 0.34758689999580383, "learning_rate": 0.0002, "epoch": 0.7114044737807114, "step": 970}, {"loss": 1.2197, "grad_norm": 0.831749439239502, "learning_rate": 0.0002, "epoch": 0.7187385405207187, "step": 980}, {"loss": 1.158, "grad_norm": 0.4438304007053375, "learning_rate": 0.0002, "epoch": 0.7260726072607261, "step": 990}, {"loss": 1.1021, "grad_norm": 0.33840006589889526, "learning_rate": 0.0002, "epoch": 0.7334066740007334, "step": 1000}, {"loss": 1.254, "grad_norm": 0.3454797863960266, "learning_rate": 0.0002, "epoch": 0.7407407407407407, "step": 1010}, {"loss": 1.106, "grad_norm": 0.38999441266059875, "learning_rate": 0.0002, "epoch": 0.7480748074807481, "step": 1020}, {"loss": 1.1428, "grad_norm": 0.2829911708831787, "learning_rate": 0.0002, "epoch": 0.7554088742207554, "step": 1030}, {"loss": 1.2123, "grad_norm": 0.36918163299560547, "learning_rate": 0.0002, "epoch": 0.7627429409607628, "step": 1040}, {"loss": 1.3028, "grad_norm": 0.3415680229663849, "learning_rate": 0.0002, "epoch": 0.77007700770077, "step": 1050}, {"loss": 1.1939, "grad_norm": 0.2974182963371277, "learning_rate": 0.0002, "epoch": 0.7774110744407774, "step": 1060}, {"loss": 1.194, "grad_norm": 0.3880919814109802, "learning_rate": 0.0002, "epoch": 0.7847451411807848, "step": 1070}, {"loss": 1.1095, "grad_norm": 0.33503302931785583, "learning_rate": 0.0002, "epoch": 0.7920792079207921, "step": 1080}, {"loss": 1.2111, "grad_norm": 0.3728407025337219, "learning_rate": 0.0002, "epoch": 0.7994132746607994, "step": 1090}, {"loss": 1.0835, "grad_norm": 0.3509373664855957, "learning_rate": 0.0002, "epoch": 0.8067473414008067, "step": 1100}, {"loss": 1.2661, "grad_norm": 0.42228564620018005, "learning_rate": 0.0002, "epoch": 0.8140814081408141, "step": 1110}, {"loss": 1.1788, "grad_norm": 0.313467800617218, "learning_rate": 0.0002, "epoch": 0.8214154748808215, "step": 1120}, {"loss": 1.1971, "grad_norm": 0.3378850817680359, "learning_rate": 0.0002, "epoch": 0.8287495416208287, "step": 1130}, {"loss": 1.1238, "grad_norm": 0.43200382590293884, "learning_rate": 0.0002, "epoch": 0.8360836083608361, "step": 1140}, {"loss": 1.3203, "grad_norm": 0.3309599459171295, "learning_rate": 0.0002, "epoch": 0.8434176751008434, "step": 1150}, {"loss": 1.1062, "grad_norm": 0.3526846170425415, "learning_rate": 0.0002, "epoch": 0.8507517418408508, "step": 1160}, {"loss": 1.0851, "grad_norm": 1.2722247838974, "learning_rate": 0.0002, "epoch": 0.858085808580858, "step": 1170}, {"loss": 1.0785, "grad_norm": 0.34142059087753296, "learning_rate": 0.0002, "epoch": 0.8654198753208654, "step": 1180}, {"loss": 1.2187, "grad_norm": 0.3805823028087616, "learning_rate": 0.0002, "epoch": 0.8727539420608728, "step": 1190}, {"loss": 1.1215, "grad_norm": 0.3931232690811157, "learning_rate": 0.0002, "epoch": 0.8800880088008801, "step": 1200}, {"loss": 1.0948, "grad_norm": 0.2937372624874115, "learning_rate": 0.0002, "epoch": 0.8874220755408874, "step": 1210}, {"loss": 1.1228, "grad_norm": 0.3757196366786957, "learning_rate": 0.0002, "epoch": 0.8947561422808947, "step": 1220}, {"loss": 1.1222, "grad_norm": 0.3502705991268158, "learning_rate": 0.0002, "epoch": 0.9020902090209021, "step": 1230}, {"loss": 1.2242, "grad_norm": 0.32758915424346924, "learning_rate": 0.0002, "epoch": 0.9094242757609095, "step": 1240}, {"loss": 1.215, "grad_norm": 0.37199416756629944, "learning_rate": 0.0002, "epoch": 0.9167583425009168, "step": 1250}, {"loss": 1.1225, "grad_norm": 0.3551490604877472, "learning_rate": 0.0002, "epoch": 0.9240924092409241, "step": 1260}, {"loss": 1.1966, "grad_norm": 0.2859550714492798, "learning_rate": 0.0002, "epoch": 0.9314264759809314, "step": 1270}, {"loss": 1.2186, "grad_norm": 0.427990585565567, "learning_rate": 0.0002, "epoch": 0.9387605427209388, "step": 1280}, {"loss": 1.2848, "grad_norm": 0.33717992901802063, "learning_rate": 0.0002, "epoch": 0.9460946094609461, "step": 1290}, {"loss": 1.1656, "grad_norm": 0.30225634574890137, "learning_rate": 0.0002, "epoch": 0.9534286762009534, "step": 1300}, {"loss": 1.2404, "grad_norm": 0.385821133852005, "learning_rate": 0.0002, "epoch": 0.9607627429409608, "step": 1310}, {"loss": 1.1932, "grad_norm": 0.35278066992759705, "learning_rate": 0.0002, "epoch": 0.9680968096809681, "step": 1320}, {"loss": 1.1071, "grad_norm": 0.49987098574638367, "learning_rate": 0.0002, "epoch": 0.9754308764209755, "step": 1330}, {"loss": 1.2259, "grad_norm": 0.3842747211456299, "learning_rate": 0.0002, "epoch": 0.9827649431609827, "step": 1340}, {"loss": 1.0862, "grad_norm": 0.6274653673171997, "learning_rate": 0.0002, "epoch": 0.9900990099009901, "step": 1350}, {"loss": 1.124, "grad_norm": 0.5239808559417725, "learning_rate": 0.0002, "epoch": 0.9974330766409975, "step": 1360}, {"eval_loss": 1.1822267770767212, "eval_runtime": 32.7389, "eval_samples_per_second": 13.165, "eval_steps_per_second": 1.649, "epoch": 0.9996332966629996, "step": 1363}, {"loss": 1.096, "grad_norm": 0.45311301946640015, "learning_rate": 0.0002, "epoch": 1.0047671433810048, "step": 1370}, {"loss": 1.0143, "grad_norm": 0.29685574769973755, "learning_rate": 0.0002, "epoch": 1.012101210121012, "step": 1380}, {"loss": 1.0302, "grad_norm": 0.3290937840938568, "learning_rate": 0.0002, "epoch": 1.0194352768610195, "step": 1390}, {"loss": 1.0295, "grad_norm": 0.3801758587360382, "learning_rate": 0.0002, "epoch": 1.0267693436010268, "step": 1400}, {"loss": 1.1226, "grad_norm": 0.794174313545227, "learning_rate": 0.0002, "epoch": 1.034103410341034, "step": 1410}, {"loss": 1.2232, "grad_norm": 0.3854154646396637, "learning_rate": 0.0002, "epoch": 1.0414374770810415, "step": 1420}, {"loss": 1.0652, "grad_norm": 0.32702451944351196, "learning_rate": 0.0002, "epoch": 1.0487715438210488, "step": 1430}, {"loss": 1.1144, "grad_norm": 0.7815203666687012, "learning_rate": 0.0002, "epoch": 1.056105610561056, "step": 1440}, {"loss": 1.1316, "grad_norm": 0.3087436854839325, "learning_rate": 0.0002, "epoch": 1.0634396773010635, "step": 1450}, {"loss": 1.1124, "grad_norm": 0.3847602903842926, "learning_rate": 0.0002, "epoch": 1.0707737440410707, "step": 1460}, {"loss": 1.1428, "grad_norm": 0.3693031370639801, "learning_rate": 0.0002, "epoch": 1.0781078107810782, "step": 1470}, {"loss": 1.0995, "grad_norm": 0.4111202359199524, "learning_rate": 0.0002, "epoch": 1.0854418775210855, "step": 1480}, {"loss": 1.0961, "grad_norm": 0.41452381014823914, "learning_rate": 0.0002, "epoch": 1.0927759442610927, "step": 1490}, {"loss": 1.1068, "grad_norm": 0.3336445093154907, "learning_rate": 0.0002, "epoch": 1.1001100110011002, "step": 1500}, {"loss": 1.0556, "grad_norm": 0.3923407793045044, "learning_rate": 0.0002, "epoch": 1.1074440777411074, "step": 1510}, {"loss": 1.1644, "grad_norm": 0.46215683221817017, "learning_rate": 0.0002, "epoch": 1.1147781444811147, "step": 1520}, {"loss": 1.1133, "grad_norm": 0.3592156767845154, "learning_rate": 0.0002, "epoch": 1.1221122112211221, "step": 1530}, {"loss": 1.0957, "grad_norm": 0.361110657453537, "learning_rate": 0.0002, "epoch": 1.1294462779611294, "step": 1540}, {"loss": 1.1553, "grad_norm": 0.5317131280899048, "learning_rate": 0.0002, "epoch": 1.1367803447011369, "step": 1550}, {"loss": 1.0368, "grad_norm": 0.3882388174533844, "learning_rate": 0.0002, "epoch": 1.1441144114411441, "step": 1560}, {"loss": 1.0805, "grad_norm": 0.3259428143501282, "learning_rate": 0.0002, "epoch": 1.1514484781811514, "step": 1570}, {"loss": 1.1819, "grad_norm": 0.410935640335083, "learning_rate": 0.0002, "epoch": 1.1587825449211588, "step": 1580}, {"loss": 1.1143, "grad_norm": 0.44940185546875, "learning_rate": 0.0002, "epoch": 1.166116611661166, "step": 1590}, {"loss": 1.0334, "grad_norm": 0.5106484293937683, "learning_rate": 0.0002, "epoch": 1.1734506784011733, "step": 1600}, {"loss": 1.2376, "grad_norm": 0.6603665947914124, "learning_rate": 0.0002, "epoch": 1.1807847451411808, "step": 1610}, {"loss": 1.1227, "grad_norm": 0.4799964129924774, "learning_rate": 0.0002, "epoch": 1.188118811881188, "step": 1620}, {"loss": 1.1191, "grad_norm": 0.4389883279800415, "learning_rate": 0.0002, "epoch": 1.1954528786211955, "step": 1630}, {"loss": 1.0667, "grad_norm": 0.4188813269138336, "learning_rate": 0.0002, "epoch": 1.2027869453612028, "step": 1640}, {"loss": 1.0605, "grad_norm": 0.7132157683372498, "learning_rate": 0.0002, "epoch": 1.21012101210121, "step": 1650}, {"loss": 1.0204, "grad_norm": 0.507480263710022, "learning_rate": 0.0002, "epoch": 1.2174550788412175, "step": 1660}, {"loss": 0.9948, "grad_norm": 0.9452332854270935, "learning_rate": 0.0002, "epoch": 1.2247891455812248, "step": 1670}, {"loss": 1.0228, "grad_norm": 0.4121614992618561, "learning_rate": 0.0002, "epoch": 1.2321232123212322, "step": 1680}, {"loss": 1.0366, "grad_norm": 0.34230247139930725, "learning_rate": 0.0002, "epoch": 1.2394572790612395, "step": 1690}, {"loss": 1.1289, "grad_norm": 0.4026208817958832, "learning_rate": 0.0002, "epoch": 1.2467913458012467, "step": 1700}, {"loss": 1.0206, "grad_norm": 0.46673697233200073, "learning_rate": 0.0002, "epoch": 1.2541254125412542, "step": 1710}, {"loss": 1.0827, "grad_norm": 0.38349825143814087, "learning_rate": 0.0002, "epoch": 1.2614594792812615, "step": 1720}, {"loss": 1.0356, "grad_norm": 0.4049997627735138, "learning_rate": 0.0002, "epoch": 1.2687935460212687, "step": 1730}, {"loss": 0.9504, "grad_norm": 0.3417615294456482, "learning_rate": 0.0002, "epoch": 1.2761276127612762, "step": 1740}, {"loss": 1.094, "grad_norm": 0.4277614951133728, "learning_rate": 0.0002, "epoch": 1.2834616795012834, "step": 1750}, {"loss": 0.9938, "grad_norm": 0.5864202976226807, "learning_rate": 0.0002, "epoch": 1.2907957462412907, "step": 1760}, {"loss": 1.1167, "grad_norm": 0.7097493410110474, "learning_rate": 0.0002, "epoch": 1.2981298129812981, "step": 1770}, {"loss": 1.1132, "grad_norm": 0.3145381212234497, "learning_rate": 0.0002, "epoch": 1.3054638797213054, "step": 1780}, {"loss": 1.1099, "grad_norm": 0.5116165280342102, "learning_rate": 0.0002, "epoch": 1.3127979464613129, "step": 1790}, {"loss": 1.0765, "grad_norm": 0.7469736337661743, "learning_rate": 0.0002, "epoch": 1.3201320132013201, "step": 1800}, {"loss": 1.0663, "grad_norm": 0.32272255420684814, "learning_rate": 0.0002, "epoch": 1.3274660799413276, "step": 1810}, {"loss": 0.9887, "grad_norm": 0.3534623086452484, "learning_rate": 0.0002, "epoch": 1.3348001466813348, "step": 1820}, {"loss": 1.1628, "grad_norm": 0.36127907037734985, "learning_rate": 0.0002, "epoch": 1.342134213421342, "step": 1830}, {"loss": 1.0972, "grad_norm": 0.4072401523590088, "learning_rate": 0.0002, "epoch": 1.3494682801613496, "step": 1840}, {"loss": 1.1267, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.3568023469013568, "step": 1850}, {"loss": 1.0173, "grad_norm": 0.412883460521698, "learning_rate": 0.0002, "epoch": 1.364136413641364, "step": 1860}, {"loss": 1.0265, "grad_norm": 0.3735875189304352, "learning_rate": 0.0002, "epoch": 1.3714704803813715, "step": 1870}, {"loss": 1.1061, "grad_norm": 0.39158159494400024, "learning_rate": 0.0002, "epoch": 1.3788045471213788, "step": 1880}, {"loss": 1.0433, "grad_norm": 0.44431769847869873, "learning_rate": 0.0002, "epoch": 1.386138613861386, "step": 1890}, {"loss": 1.0216, "grad_norm": 0.37772801518440247, "learning_rate": 0.0002, "epoch": 1.3934726806013935, "step": 1900}, {"loss": 1.0674, "grad_norm": 0.4056641757488251, "learning_rate": 0.0002, "epoch": 1.4008067473414008, "step": 1910}, {"loss": 1.0256, "grad_norm": 0.41612377762794495, "learning_rate": 0.0002, "epoch": 1.408140814081408, "step": 1920}, {"loss": 1.0467, "grad_norm": 0.41153013706207275, "learning_rate": 0.0002, "epoch": 1.4154748808214155, "step": 1930}, {"loss": 1.1062, "grad_norm": 0.387845516204834, "learning_rate": 0.0002, "epoch": 1.4228089475614227, "step": 1940}, {"loss": 1.1094, "grad_norm": 0.3809587061405182, "learning_rate": 0.0002, "epoch": 1.4301430143014302, "step": 1950}, {"loss": 1.0461, "grad_norm": 0.3625726103782654, "learning_rate": 0.0002, "epoch": 1.4374770810414375, "step": 1960}, {"loss": 0.9983, "grad_norm": 0.5294290781021118, "learning_rate": 0.0002, "epoch": 1.444811147781445, "step": 1970}, {"loss": 1.1114, "grad_norm": 0.39975494146347046, "learning_rate": 0.0002, "epoch": 1.4521452145214522, "step": 1980}, {"loss": 0.9704, "grad_norm": 0.4181167185306549, "learning_rate": 0.0002, "epoch": 1.4594792812614594, "step": 1990}, {"loss": 1.1146, "grad_norm": 0.42001503705978394, "learning_rate": 0.0002, "epoch": 1.466813348001467, "step": 2000}, {"loss": 1.1266, "grad_norm": 0.4877578616142273, "learning_rate": 0.0002, "epoch": 1.4741474147414741, "step": 2010}, {"loss": 1.1012, "grad_norm": 0.4050969183444977, "learning_rate": 0.0002, "epoch": 1.4814814814814814, "step": 2020}, {"loss": 1.0562, "grad_norm": 0.39068883657455444, "learning_rate": 0.0002, "epoch": 1.4888155482214889, "step": 2030}, {"loss": 1.0464, "grad_norm": 0.421282559633255, "learning_rate": 0.0002, "epoch": 1.4961496149614961, "step": 2040}, {"loss": 1.0532, "grad_norm": 0.47092297673225403, "learning_rate": 0.0002, "epoch": 1.5034836817015034, "step": 2050}, {"loss": 0.9348, "grad_norm": 0.39688974618911743, "learning_rate": 0.0002, "epoch": 1.5108177484415108, "step": 2060}, {"loss": 1.08, "grad_norm": 0.5529879331588745, "learning_rate": 0.0002, "epoch": 1.5181518151815183, "step": 2070}, {"loss": 1.1836, "grad_norm": 0.4879782199859619, "learning_rate": 0.0002, "epoch": 1.5254858819215253, "step": 2080}, {"loss": 1.0432, "grad_norm": 0.5517361164093018, "learning_rate": 0.0002, "epoch": 1.5328199486615328, "step": 2090}, {"loss": 1.0433, "grad_norm": 0.44015637040138245, "learning_rate": 0.0002, "epoch": 1.5401540154015403, "step": 2100}, {"loss": 1.1873, "grad_norm": 0.5435167551040649, "learning_rate": 0.0002, "epoch": 1.5474880821415475, "step": 2110}, {"loss": 1.1076, "grad_norm": 0.5714033246040344, "learning_rate": 0.0002, "epoch": 1.5548221488815548, "step": 2120}, {"loss": 1.1107, "grad_norm": 0.31732529401779175, "learning_rate": 0.0002, "epoch": 1.5621562156215623, "step": 2130}, {"loss": 1.0817, "grad_norm": 0.49068278074264526, "learning_rate": 0.0002, "epoch": 1.5694902823615695, "step": 2140}, {"loss": 1.0254, "grad_norm": 0.46851542592048645, "learning_rate": 0.0002, "epoch": 1.5768243491015768, "step": 2150}, {"loss": 1.0623, "grad_norm": 0.5083092451095581, "learning_rate": 0.0002, "epoch": 1.5841584158415842, "step": 2160}, {"loss": 1.0603, "grad_norm": 0.9822936058044434, "learning_rate": 0.0002, "epoch": 1.5914924825815915, "step": 2170}, {"loss": 0.9986, "grad_norm": 0.4575989246368408, "learning_rate": 0.0002, "epoch": 1.5988265493215987, "step": 2180}, {"loss": 1.1292, "grad_norm": 0.47444286942481995, "learning_rate": 0.0002, "epoch": 1.6061606160616062, "step": 2190}, {"loss": 1.0136, "grad_norm": 0.7208226919174194, "learning_rate": 0.0002, "epoch": 1.6134946828016135, "step": 2200}, {"loss": 1.15, "grad_norm": 0.43791481852531433, "learning_rate": 0.0002, "epoch": 1.6208287495416207, "step": 2210}, {"loss": 1.0961, "grad_norm": 0.5245792865753174, "learning_rate": 0.0002, "epoch": 1.6281628162816282, "step": 2220}, {"loss": 0.9957, "grad_norm": 0.39289429783821106, "learning_rate": 0.0002, "epoch": 1.6354968830216357, "step": 2230}, {"loss": 1.133, "grad_norm": 0.6106135845184326, "learning_rate": 0.0002, "epoch": 1.6428309497616427, "step": 2240}, {"loss": 1.0129, "grad_norm": 0.3722580671310425, "learning_rate": 0.0002, "epoch": 1.6501650165016502, "step": 2250}, {"loss": 1.0446, "grad_norm": 0.3649403750896454, "learning_rate": 0.0002, "epoch": 1.6574990832416576, "step": 2260}, {"loss": 1.0037, "grad_norm": 0.46514248847961426, "learning_rate": 0.0002, "epoch": 1.6648331499816649, "step": 2270}, {"loss": 1.0022, "grad_norm": 0.42034927010536194, "learning_rate": 0.0002, "epoch": 1.6721672167216721, "step": 2280}, {"loss": 1.1362, "grad_norm": 0.45202910900115967, "learning_rate": 0.0002, "epoch": 1.6795012834616796, "step": 2290}, {"loss": 1.0866, "grad_norm": 0.36257603764533997, "learning_rate": 0.0002, "epoch": 1.6868353502016868, "step": 2300}, {"loss": 1.0973, "grad_norm": 0.6340323090553284, "learning_rate": 0.0002, "epoch": 1.694169416941694, "step": 2310}, {"loss": 1.0615, "grad_norm": 0.4352878928184509, "learning_rate": 0.0002, "epoch": 1.7015034836817016, "step": 2320}, {"loss": 1.0629, "grad_norm": 0.45029792189598083, "learning_rate": 0.0002, "epoch": 1.7088375504217088, "step": 2330}, {"loss": 0.9621, "grad_norm": 0.3891315758228302, "learning_rate": 0.0002, "epoch": 1.716171617161716, "step": 2340}, {"loss": 0.9779, "grad_norm": 0.35180050134658813, "learning_rate": 0.0002, "epoch": 1.7235056839017235, "step": 2350}, {"loss": 1.0368, "grad_norm": 0.42367449402809143, "learning_rate": 0.0002, "epoch": 1.7308397506417308, "step": 2360}, {"loss": 1.0376, "grad_norm": 0.4553675353527069, "learning_rate": 0.0002, "epoch": 1.738173817381738, "step": 2370}, {"loss": 1.1467, "grad_norm": 0.5944654941558838, "learning_rate": 0.0002, "epoch": 1.7455078841217455, "step": 2380}, {"loss": 1.0548, "grad_norm": 0.3479664623737335, "learning_rate": 0.0002, "epoch": 1.752841950861753, "step": 2390}, {"loss": 1.0798, "grad_norm": 0.3585502505302429, "learning_rate": 0.0002, "epoch": 1.76017601760176, "step": 2400}, {"loss": 1.0983, "grad_norm": 0.4263346493244171, "learning_rate": 0.0002, "epoch": 1.7675100843417675, "step": 2410}, {"loss": 1.054, "grad_norm": 0.5476409196853638, "learning_rate": 0.0002, "epoch": 1.774844151081775, "step": 2420}, {"loss": 1.1615, "grad_norm": 0.3694186508655548, "learning_rate": 0.0002, "epoch": 1.7821782178217822, "step": 2430}, {"loss": 1.1343, "grad_norm": 0.9185658693313599, "learning_rate": 0.0002, "epoch": 1.7895122845617895, "step": 2440}, {"loss": 1.0764, "grad_norm": 0.7171908020973206, "learning_rate": 0.0002, "epoch": 1.796846351301797, "step": 2450}, {"loss": 1.1154, "grad_norm": 0.550658643245697, "learning_rate": 0.0002, "epoch": 1.8041804180418042, "step": 2460}, {"loss": 0.9975, "grad_norm": 0.4075568914413452, "learning_rate": 0.0002, "epoch": 1.8115144847818114, "step": 2470}, {"loss": 1.0935, "grad_norm": 0.3790127635002136, "learning_rate": 0.0002, "epoch": 1.818848551521819, "step": 2480}, {"loss": 0.9839, "grad_norm": 0.3576384484767914, "learning_rate": 0.0002, "epoch": 1.8261826182618262, "step": 2490}, {"loss": 1.1369, "grad_norm": 0.3919370770454407, "learning_rate": 0.0002, "epoch": 1.8335166850018334, "step": 2500}, {"loss": 0.9985, "grad_norm": 0.485083669424057, "learning_rate": 0.0002, "epoch": 1.8408507517418409, "step": 2510}, {"loss": 1.1585, "grad_norm": 0.4564347565174103, "learning_rate": 0.0002, "epoch": 1.8481848184818483, "step": 2520}, {"loss": 1.0944, "grad_norm": 0.3613106608390808, "learning_rate": 0.0002, "epoch": 1.8555188852218554, "step": 2530}, {"loss": 1.0819, "grad_norm": 0.39600759744644165, "learning_rate": 0.0002, "epoch": 1.8628529519618628, "step": 2540}, {"loss": 0.9453, "grad_norm": 1.123499870300293, "learning_rate": 0.0002, "epoch": 1.8701870187018703, "step": 2550}, {"loss": 1.0635, "grad_norm": 0.4612680673599243, "learning_rate": 0.0002, "epoch": 1.8775210854418776, "step": 2560}, {"loss": 1.0087, "grad_norm": 0.42745399475097656, "learning_rate": 0.0002, "epoch": 1.8848551521818848, "step": 2570}, {"loss": 1.0102, "grad_norm": 0.4055580198764801, "learning_rate": 0.0002, "epoch": 1.8921892189218923, "step": 2580}, {"loss": 1.0177, "grad_norm": 0.44174644351005554, "learning_rate": 0.0002, "epoch": 1.8995232856618995, "step": 2590}, {"loss": 0.9886, "grad_norm": 1.0228385925292969, "learning_rate": 0.0002, "epoch": 1.9068573524019068, "step": 2600}, {"loss": 1.0857, "grad_norm": 0.3496396243572235, "learning_rate": 0.0002, "epoch": 1.9141914191419143, "step": 2610}, {"loss": 1.0955, "grad_norm": 0.4191173017024994, "learning_rate": 0.0002, "epoch": 1.9215254858819215, "step": 2620}, {"loss": 1.0943, "grad_norm": 0.6778554916381836, "learning_rate": 0.0002, "epoch": 1.9288595526219288, "step": 2630}, {"loss": 1.0594, "grad_norm": 0.41992834210395813, "learning_rate": 0.0002, "epoch": 1.9361936193619362, "step": 2640}, {"loss": 1.1159, "grad_norm": 0.8760401010513306, "learning_rate": 0.0002, "epoch": 1.9435276861019435, "step": 2650}, {"loss": 1.0379, "grad_norm": 0.44049209356307983, "learning_rate": 0.0002, "epoch": 1.9508617528419507, "step": 2660}, {"loss": 1.1008, "grad_norm": 0.5651928782463074, "learning_rate": 0.0002, "epoch": 1.9581958195819582, "step": 2670}, {"loss": 1.1317, "grad_norm": 0.5292727947235107, "learning_rate": 0.0002, "epoch": 1.9655298863219657, "step": 2680}, {"loss": 1.1328, "grad_norm": 0.6012240648269653, "learning_rate": 0.0002, "epoch": 1.9728639530619727, "step": 2690}, {"loss": 1.0683, "grad_norm": 0.3945149779319763, "learning_rate": 0.0002, "epoch": 1.9801980198019802, "step": 2700}, {"loss": 1.0155, "grad_norm": 0.5732627511024475, "learning_rate": 0.0002, "epoch": 1.9875320865419877, "step": 2710}, {"loss": 0.9857, "grad_norm": 0.3963361084461212, "learning_rate": 0.0002, "epoch": 1.994866153281995, "step": 2720}, {"eval_loss": 1.1534006595611572, "eval_runtime": 32.7541, "eval_samples_per_second": 13.159, "eval_steps_per_second": 1.649, "epoch": 2.0, "step": 2727}, {"loss": 0.9624, "grad_norm": 0.48628315329551697, "learning_rate": 0.0002, "epoch": 2.002200220022002, "step": 2730}, {"loss": 0.9603, "grad_norm": 0.413875013589859, "learning_rate": 0.0002, "epoch": 2.0095342867620096, "step": 2740}, {"loss": 0.965, "grad_norm": 0.4988735616207123, "learning_rate": 0.0002, "epoch": 2.0168683535020167, "step": 2750}, {"loss": 0.9677, "grad_norm": 0.5634812712669373, "learning_rate": 0.0002, "epoch": 2.024202420242024, "step": 2760}, {"loss": 0.9547, "grad_norm": 0.48302653431892395, "learning_rate": 0.0002, "epoch": 2.0315364869820316, "step": 2770}, {"loss": 0.9346, "grad_norm": 0.49914175271987915, "learning_rate": 0.0002, "epoch": 2.038870553722039, "step": 2780}, {"loss": 0.904, "grad_norm": 1.14039945602417, "learning_rate": 0.0002, "epoch": 2.046204620462046, "step": 2790}, {"loss": 0.9588, "grad_norm": 0.6359720826148987, "learning_rate": 0.0002, "epoch": 2.0535386872020536, "step": 2800}, {"loss": 0.9031, "grad_norm": 0.4589158296585083, "learning_rate": 0.0002, "epoch": 2.060872753942061, "step": 2810}, {"loss": 0.9438, "grad_norm": 0.46255481243133545, "learning_rate": 0.0002, "epoch": 2.068206820682068, "step": 2820}, {"loss": 0.9464, "grad_norm": 0.6232137680053711, "learning_rate": 0.0002, "epoch": 2.0755408874220755, "step": 2830}, {"loss": 0.8978, "grad_norm": 0.41042178869247437, "learning_rate": 0.0002, "epoch": 2.082874954162083, "step": 2840}, {"loss": 0.8516, "grad_norm": 0.5334428548812866, "learning_rate": 0.0002, "epoch": 2.09020902090209, "step": 2850}, {"loss": 0.9313, "grad_norm": 0.8270058631896973, "learning_rate": 0.0002, "epoch": 2.0975430876420975, "step": 2860}, {"loss": 1.0064, "grad_norm": 0.6624533534049988, "learning_rate": 0.0002, "epoch": 2.104877154382105, "step": 2870}, {"loss": 0.9196, "grad_norm": 0.5448863506317139, "learning_rate": 0.0002, "epoch": 2.112211221122112, "step": 2880}, {"loss": 0.887, "grad_norm": 0.621482789516449, "learning_rate": 0.0002, "epoch": 2.1195452878621195, "step": 2890}, {"loss": 0.9702, "grad_norm": 0.4556255340576172, "learning_rate": 0.0002, "epoch": 2.126879354602127, "step": 2900}, {"loss": 0.9323, "grad_norm": 0.4620579183101654, "learning_rate": 0.0002, "epoch": 2.1342134213421344, "step": 2910}, {"loss": 0.836, "grad_norm": 0.9602415561676025, "learning_rate": 0.0002, "epoch": 2.1415474880821415, "step": 2920}, {"loss": 0.8826, "grad_norm": 0.587943971157074, "learning_rate": 0.0002, "epoch": 2.148881554822149, "step": 2930}, {"loss": 0.971, "grad_norm": 0.5121372938156128, "learning_rate": 0.0002, "epoch": 2.1562156215621564, "step": 2940}, {"loss": 0.8751, "grad_norm": 0.49424484372138977, "learning_rate": 0.0002, "epoch": 2.1635496883021634, "step": 2950}, {"loss": 0.8674, "grad_norm": 0.6312560439109802, "learning_rate": 0.0002, "epoch": 2.170883755042171, "step": 2960}, {"loss": 0.9791, "grad_norm": 0.5235576629638672, "learning_rate": 0.0002, "epoch": 2.1782178217821784, "step": 2970}, {"loss": 0.9706, "grad_norm": 0.5868439674377441, "learning_rate": 0.0002, "epoch": 2.1855518885221854, "step": 2980}, {"loss": 0.9338, "grad_norm": 0.42302873730659485, "learning_rate": 0.0002, "epoch": 2.192885955262193, "step": 2990}, {"loss": 0.9332, "grad_norm": 0.5097725987434387, "learning_rate": 0.0002, "epoch": 2.2002200220022003, "step": 3000}, {"loss": 0.9239, "grad_norm": 0.5091572403907776, "learning_rate": 0.0002, "epoch": 2.2075540887422074, "step": 3010}, {"loss": 0.8898, "grad_norm": 0.49433162808418274, "learning_rate": 0.0002, "epoch": 2.214888155482215, "step": 3020}, {"loss": 0.9734, "grad_norm": 0.5577368140220642, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3030}, {"loss": 0.9033, "grad_norm": 0.6177583932876587, "learning_rate": 0.0002, "epoch": 2.2295562889622293, "step": 3040}, {"loss": 0.9882, "grad_norm": 0.5256719589233398, "learning_rate": 0.0002, "epoch": 2.236890355702237, "step": 3050}, {"loss": 0.9439, "grad_norm": 0.5001118183135986, "learning_rate": 0.0002, "epoch": 2.2442244224422443, "step": 3060}, {"loss": 0.8718, "grad_norm": 0.5721249580383301, "learning_rate": 0.0002, "epoch": 2.2515584891822513, "step": 3070}, {"loss": 1.0648, "grad_norm": 0.5325384140014648, "learning_rate": 0.0002, "epoch": 2.258892555922259, "step": 3080}, {"loss": 0.9843, "grad_norm": 0.5719189047813416, "learning_rate": 0.0002, "epoch": 2.2662266226622663, "step": 3090}, {"loss": 0.8633, "grad_norm": 0.6337835788726807, "learning_rate": 0.0002, "epoch": 2.2735606894022737, "step": 3100}, {"loss": 0.9962, "grad_norm": 0.5381836891174316, "learning_rate": 0.0002, "epoch": 2.2808947561422808, "step": 3110}, {"loss": 0.8265, "grad_norm": 0.5408531427383423, "learning_rate": 0.0002, "epoch": 2.2882288228822882, "step": 3120}, {"loss": 1.0325, "grad_norm": 0.43705281615257263, "learning_rate": 0.0002, "epoch": 2.2955628896222957, "step": 3130}, {"loss": 0.9388, "grad_norm": 0.6454030275344849, "learning_rate": 0.0002, "epoch": 2.3028969563623027, "step": 3140}, {"loss": 0.954, "grad_norm": 0.686030387878418, "learning_rate": 0.0002, "epoch": 2.31023102310231, "step": 3150}, {"loss": 0.9403, "grad_norm": 0.5123633146286011, "learning_rate": 0.0002, "epoch": 2.3175650898423177, "step": 3160}, {"loss": 0.8834, "grad_norm": 0.842506468296051, "learning_rate": 0.0002, "epoch": 2.3248991565823247, "step": 3170}, {"loss": 1.0497, "grad_norm": 0.5193818807601929, "learning_rate": 0.0002, "epoch": 2.332233223322332, "step": 3180}, {"loss": 0.9473, "grad_norm": 0.5634409189224243, "learning_rate": 0.0002, "epoch": 2.3395672900623397, "step": 3190}, {"loss": 0.8499, "grad_norm": 0.6475534439086914, "learning_rate": 0.0002, "epoch": 2.3469013568023467, "step": 3200}, {"loss": 0.874, "grad_norm": 1.1503914594650269, "learning_rate": 0.0002, "epoch": 2.354235423542354, "step": 3210}, {"loss": 0.9762, "grad_norm": 0.7234905362129211, "learning_rate": 0.0002, "epoch": 2.3615694902823616, "step": 3220}, {"loss": 0.9007, "grad_norm": 0.664903461933136, "learning_rate": 0.0002, "epoch": 2.368903557022369, "step": 3230}, {"loss": 0.9987, "grad_norm": 0.5453006625175476, "learning_rate": 0.0002, "epoch": 2.376237623762376, "step": 3240}, {"loss": 0.9742, "grad_norm": 0.6256654262542725, "learning_rate": 0.0002, "epoch": 2.3835716905023836, "step": 3250}, {"loss": 0.9922, "grad_norm": 0.5166565179824829, "learning_rate": 0.0002, "epoch": 2.390905757242391, "step": 3260}, {"loss": 0.927, "grad_norm": 0.5699098110198975, "learning_rate": 0.0002, "epoch": 2.398239823982398, "step": 3270}, {"loss": 0.8878, "grad_norm": 0.4472540020942688, "learning_rate": 0.0002, "epoch": 2.4055738907224056, "step": 3280}, {"loss": 0.9439, "grad_norm": 0.6790403127670288, "learning_rate": 0.0002, "epoch": 2.412907957462413, "step": 3290}, {"loss": 0.972, "grad_norm": 0.5182185173034668, "learning_rate": 0.0002, "epoch": 2.42024202420242, "step": 3300}, {"loss": 0.9775, "grad_norm": 0.564647912979126, "learning_rate": 0.0002, "epoch": 2.4275760909424275, "step": 3310}, {"loss": 1.072, "grad_norm": 0.5625313520431519, "learning_rate": 0.0002, "epoch": 2.434910157682435, "step": 3320}, {"loss": 0.8798, "grad_norm": 0.7496559619903564, "learning_rate": 0.0002, "epoch": 2.442244224422442, "step": 3330}, {"loss": 0.868, "grad_norm": 0.4779128134250641, "learning_rate": 0.0002, "epoch": 2.4495782911624495, "step": 3340}, {"loss": 1.0316, "grad_norm": 0.578093409538269, "learning_rate": 0.0002, "epoch": 2.456912357902457, "step": 3350}, {"loss": 0.9282, "grad_norm": 0.5456080436706543, "learning_rate": 0.0002, "epoch": 2.4642464246424645, "step": 3360}, {"loss": 0.8409, "grad_norm": 0.4769273102283478, "learning_rate": 0.0002, "epoch": 2.4715804913824715, "step": 3370}, {"loss": 0.9312, "grad_norm": 0.5608189702033997, "learning_rate": 0.0002, "epoch": 2.478914558122479, "step": 3380}, {"loss": 0.9934, "grad_norm": 0.5590165853500366, "learning_rate": 0.0002, "epoch": 2.4862486248624864, "step": 3390}, {"loss": 1.025, "grad_norm": 0.801306962966919, "learning_rate": 0.0002, "epoch": 2.4935826916024935, "step": 3400}, {"loss": 0.9049, "grad_norm": 0.6045624613761902, "learning_rate": 0.0002, "epoch": 2.500916758342501, "step": 3410}, {"loss": 0.944, "grad_norm": 0.5735858082771301, "learning_rate": 0.0002, "epoch": 2.5082508250825084, "step": 3420}, {"loss": 0.9846, "grad_norm": 0.6827309131622314, "learning_rate": 0.0002, "epoch": 2.5155848918225154, "step": 3430}, {"loss": 0.9789, "grad_norm": 0.5702602863311768, "learning_rate": 0.0002, "epoch": 2.522918958562523, "step": 3440}, {"loss": 0.9127, "grad_norm": 0.6674721240997314, "learning_rate": 0.0002, "epoch": 2.5302530253025304, "step": 3450}, {"loss": 0.914, "grad_norm": 0.5635907649993896, "learning_rate": 0.0002, "epoch": 2.5375870920425374, "step": 3460}, {"loss": 0.8398, "grad_norm": 0.42737770080566406, "learning_rate": 0.0002, "epoch": 2.544921158782545, "step": 3470}, {"loss": 0.9474, "grad_norm": 0.6720691919326782, "learning_rate": 0.0002, "epoch": 2.5522552255225524, "step": 3480}, {"loss": 0.8637, "grad_norm": 0.8917084336280823, "learning_rate": 0.0002, "epoch": 2.55958929226256, "step": 3490}, {"loss": 0.9257, "grad_norm": 0.5134549140930176, "learning_rate": 0.0002, "epoch": 2.566923359002567, "step": 3500}, {"loss": 0.9362, "grad_norm": 0.4951367974281311, "learning_rate": 0.0002, "epoch": 2.5742574257425743, "step": 3510}, {"loss": 0.9184, "grad_norm": 0.9438204765319824, "learning_rate": 0.0002, "epoch": 2.5815914924825814, "step": 3520}, {"loss": 0.8939, "grad_norm": 0.6024714708328247, "learning_rate": 0.0002, "epoch": 2.588925559222589, "step": 3530}, {"loss": 0.9298, "grad_norm": 0.5248535871505737, "learning_rate": 0.0002, "epoch": 2.5962596259625963, "step": 3540}, {"loss": 0.941, "grad_norm": 0.8677568435668945, "learning_rate": 0.0002, "epoch": 2.6035936927026038, "step": 3550}, {"loss": 0.9253, "grad_norm": 0.82008296251297, "learning_rate": 0.0002, "epoch": 2.610927759442611, "step": 3560}, {"loss": 0.8429, "grad_norm": 0.4724634885787964, "learning_rate": 0.0002, "epoch": 2.6182618261826183, "step": 3570}, {"loss": 0.9058, "grad_norm": 0.5434244275093079, "learning_rate": 0.0002, "epoch": 2.6255958929226257, "step": 3580}, {"loss": 0.9379, "grad_norm": 0.4948740005493164, "learning_rate": 0.0002, "epoch": 2.6329299596626328, "step": 3590}, {"loss": 0.8718, "grad_norm": 0.42109328508377075, "learning_rate": 0.0002, "epoch": 2.6402640264026402, "step": 3600}, {"loss": 0.9809, "grad_norm": 0.7979786396026611, "learning_rate": 0.0002, "epoch": 2.6475980931426477, "step": 3610}, {"loss": 0.9229, "grad_norm": 0.6345919370651245, "learning_rate": 0.0002, "epoch": 2.654932159882655, "step": 3620}, {"loss": 0.8506, "grad_norm": 0.4971671402454376, "learning_rate": 0.0002, "epoch": 2.662266226622662, "step": 3630}, {"loss": 0.8054, "grad_norm": 0.6467748284339905, "learning_rate": 0.0002, "epoch": 2.6696002933626697, "step": 3640}, {"loss": 0.9277, "grad_norm": 0.4240160286426544, "learning_rate": 0.0002, "epoch": 2.6769343601026767, "step": 3650}, {"loss": 0.8213, "grad_norm": 0.5179754495620728, "learning_rate": 0.0002, "epoch": 2.684268426842684, "step": 3660}, {"loss": 0.9221, "grad_norm": 0.754012405872345, "learning_rate": 0.0002, "epoch": 2.6916024935826917, "step": 3670}, {"loss": 0.9194, "grad_norm": 0.5141299962997437, "learning_rate": 0.0002, "epoch": 2.698936560322699, "step": 3680}, {"loss": 0.9495, "grad_norm": 0.5737819075584412, "learning_rate": 0.0002, "epoch": 2.706270627062706, "step": 3690}, {"loss": 1.0162, "grad_norm": 0.5887577533721924, "learning_rate": 0.0002, "epoch": 2.7136046938027136, "step": 3700}, {"loss": 0.9169, "grad_norm": 0.6740471720695496, "learning_rate": 0.0002, "epoch": 2.720938760542721, "step": 3710}, {"loss": 0.9297, "grad_norm": 0.5879453420639038, "learning_rate": 0.0002, "epoch": 2.728272827282728, "step": 3720}, {"loss": 0.9358, "grad_norm": 0.4858354926109314, "learning_rate": 0.0002, "epoch": 2.7356068940227356, "step": 3730}, {"loss": 0.9308, "grad_norm": 0.5489001870155334, "learning_rate": 0.0002, "epoch": 2.742940960762743, "step": 3740}, {"loss": 0.894, "grad_norm": 0.8187092542648315, "learning_rate": 0.0002, "epoch": 2.7502750275027505, "step": 3750}, {"loss": 0.8954, "grad_norm": 0.5666626691818237, "learning_rate": 0.0002, "epoch": 2.7576090942427576, "step": 3760}, {"loss": 1.0059, "grad_norm": 0.5377066135406494, "learning_rate": 0.0002, "epoch": 2.764943160982765, "step": 3770}, {"loss": 0.9132, "grad_norm": 0.566330075263977, "learning_rate": 0.0002, "epoch": 2.772277227722772, "step": 3780}, {"loss": 0.9415, "grad_norm": 0.5522832870483398, "learning_rate": 0.0002, "epoch": 2.7796112944627795, "step": 3790}, {"loss": 0.8816, "grad_norm": 0.5668695569038391, "learning_rate": 0.0002, "epoch": 2.786945361202787, "step": 3800}, {"loss": 0.8885, "grad_norm": 0.7566602826118469, "learning_rate": 0.0002, "epoch": 2.7942794279427945, "step": 3810}, {"loss": 0.8598, "grad_norm": 0.5603684782981873, "learning_rate": 0.0002, "epoch": 2.8016134946828015, "step": 3820}, {"loss": 0.9602, "grad_norm": 0.49122217297554016, "learning_rate": 0.0002, "epoch": 2.808947561422809, "step": 3830}, {"loss": 0.9738, "grad_norm": 0.6798251867294312, "learning_rate": 0.0002, "epoch": 2.816281628162816, "step": 3840}, {"loss": 0.9533, "grad_norm": 0.6097991466522217, "learning_rate": 0.0002, "epoch": 2.8236156949028235, "step": 3850}, {"loss": 0.8672, "grad_norm": 0.6675726175308228, "learning_rate": 0.0002, "epoch": 2.830949761642831, "step": 3860}, {"loss": 0.9324, "grad_norm": 0.9223952889442444, "learning_rate": 0.0002, "epoch": 2.8382838283828384, "step": 3870}, {"loss": 0.8767, "grad_norm": 0.6020799875259399, "learning_rate": 0.0002, "epoch": 2.8456178951228455, "step": 3880}, {"loss": 0.9148, "grad_norm": 0.5206381678581238, "learning_rate": 0.0002, "epoch": 2.852951961862853, "step": 3890}, {"loss": 0.9479, "grad_norm": 0.6268777251243591, "learning_rate": 0.0002, "epoch": 2.8602860286028604, "step": 3900}, {"loss": 0.9409, "grad_norm": 1.1583497524261475, "learning_rate": 0.0002, "epoch": 2.8676200953428674, "step": 3910}, {"loss": 0.895, "grad_norm": 0.7263903021812439, "learning_rate": 0.0002, "epoch": 2.874954162082875, "step": 3920}, {"loss": 0.8786, "grad_norm": 0.5369910001754761, "learning_rate": 0.0002, "epoch": 2.8822882288228824, "step": 3930}, {"loss": 1.0015, "grad_norm": 0.7298350930213928, "learning_rate": 0.0002, "epoch": 2.88962229556289, "step": 3940}, {"loss": 0.979, "grad_norm": 0.577012836933136, "learning_rate": 0.0002, "epoch": 2.896956362302897, "step": 3950}, {"loss": 0.9716, "grad_norm": 0.5859594345092773, "learning_rate": 0.0002, "epoch": 2.9042904290429044, "step": 3960}, {"loss": 0.8772, "grad_norm": 0.47176122665405273, "learning_rate": 0.0002, "epoch": 2.9116244957829114, "step": 3970}, {"loss": 0.8997, "grad_norm": 0.9699620604515076, "learning_rate": 0.0002, "epoch": 2.918958562522919, "step": 3980}, {"loss": 0.9057, "grad_norm": 0.7908747792243958, "learning_rate": 0.0002, "epoch": 2.9262926292629263, "step": 3990}, {"loss": 0.9462, "grad_norm": 0.5777379274368286, "learning_rate": 0.0002, "epoch": 2.933626696002934, "step": 4000}, {"loss": 0.9358, "grad_norm": 0.599288284778595, "learning_rate": 0.0002, "epoch": 2.940960762742941, "step": 4010}, {"loss": 0.9812, "grad_norm": 0.5232274532318115, "learning_rate": 0.0002, "epoch": 2.9482948294829483, "step": 4020}, {"loss": 0.96, "grad_norm": 0.6395137310028076, "learning_rate": 0.0002, "epoch": 2.9556288962229558, "step": 4030}, {"loss": 0.9813, "grad_norm": 0.589260458946228, "learning_rate": 0.0002, "epoch": 2.962962962962963, "step": 4040}, {"loss": 0.9541, "grad_norm": 0.5699581503868103, "learning_rate": 0.0002, "epoch": 2.9702970297029703, "step": 4050}, {"loss": 0.9585, "grad_norm": 0.528468132019043, "learning_rate": 0.0002, "epoch": 2.9776310964429777, "step": 4060}, {"loss": 0.9164, "grad_norm": 0.4804670512676239, "learning_rate": 0.0002, "epoch": 2.984965163182985, "step": 4070}, {"loss": 0.9771, "grad_norm": 1.1918889284133911, "learning_rate": 0.0002, "epoch": 2.9922992299229922, "step": 4080}, {"loss": 0.9178, "grad_norm": 0.5479103326797485, "learning_rate": 0.0002, "epoch": 2.9996332966629997, "step": 4090}, {"eval_loss": 1.1642853021621704, "eval_runtime": 32.7511, "eval_samples_per_second": 13.16, "eval_steps_per_second": 1.649, "epoch": 2.9996332966629997, "step": 4090}, {"loss": 0.7981, "grad_norm": 0.7430027723312378, "learning_rate": 0.0002, "epoch": 3.006967363403007, "step": 4100}, {"loss": 0.7871, "grad_norm": 0.6293647289276123, "learning_rate": 0.0002, "epoch": 3.014301430143014, "step": 4110}, {"loss": 0.78, "grad_norm": 0.6191329956054688, "learning_rate": 0.0002, "epoch": 3.0216354968830217, "step": 4120}, {"loss": 0.7618, "grad_norm": 0.7959313988685608, "learning_rate": 0.0002, "epoch": 3.028969563623029, "step": 4130}, {"loss": 0.8039, "grad_norm": 0.5956351161003113, "learning_rate": 0.0002, "epoch": 3.036303630363036, "step": 4140}, {"loss": 0.7477, "grad_norm": 0.670383632183075, "learning_rate": 0.0002, "epoch": 3.0436376971030437, "step": 4150}, {"loss": 0.7984, "grad_norm": 0.6414518356323242, "learning_rate": 0.0002, "epoch": 3.050971763843051, "step": 4160}, {"loss": 0.7369, "grad_norm": 0.7928852438926697, "learning_rate": 0.0002, "epoch": 3.058305830583058, "step": 4170}, {"loss": 0.7914, "grad_norm": 0.6211121082305908, "learning_rate": 0.0002, "epoch": 3.0656398973230656, "step": 4180}, {"loss": 0.7365, "grad_norm": 0.6237057447433472, "learning_rate": 0.0002, "epoch": 3.072973964063073, "step": 4190}, {"loss": 0.702, "grad_norm": 0.6522233486175537, "learning_rate": 0.0002, "epoch": 3.08030803080308, "step": 4200}, {"loss": 0.7646, "grad_norm": 0.9396848678588867, "learning_rate": 0.0002, "epoch": 3.0876420975430876, "step": 4210}, {"loss": 0.7559, "grad_norm": 0.8003010749816895, "learning_rate": 0.0002, "epoch": 3.094976164283095, "step": 4220}, {"loss": 0.711, "grad_norm": 0.6733810305595398, "learning_rate": 0.0002, "epoch": 3.102310231023102, "step": 4230}, {"loss": 0.696, "grad_norm": 0.6365828514099121, "learning_rate": 0.0002, "epoch": 3.1096442977631096, "step": 4240}, {"loss": 0.8362, "grad_norm": 1.0805548429489136, "learning_rate": 0.0002, "epoch": 3.116978364503117, "step": 4250}, {"loss": 0.7651, "grad_norm": 0.7262141108512878, "learning_rate": 0.0002, "epoch": 3.1243124312431245, "step": 4260}, {"loss": 0.7304, "grad_norm": 0.5500539541244507, "learning_rate": 0.0002, "epoch": 3.1316464979831315, "step": 4270}, {"loss": 0.7721, "grad_norm": 0.793912947177887, "learning_rate": 0.0002, "epoch": 3.138980564723139, "step": 4280}, {"loss": 0.7708, "grad_norm": 1.2540518045425415, "learning_rate": 0.0002, "epoch": 3.1463146314631465, "step": 4290}, {"loss": 0.782, "grad_norm": 0.7020077705383301, "learning_rate": 0.0002, "epoch": 3.1536486982031535, "step": 4300}, {"loss": 0.7253, "grad_norm": 0.5111123323440552, "learning_rate": 0.0002, "epoch": 3.160982764943161, "step": 4310}, {"loss": 0.8159, "grad_norm": 0.7172090411186218, "learning_rate": 0.0002, "epoch": 3.1683168316831685, "step": 4320}, {"loss": 0.6962, "grad_norm": 0.6343168616294861, "learning_rate": 0.0002, "epoch": 3.1756508984231755, "step": 4330}, {"loss": 0.7938, "grad_norm": 0.9563672542572021, "learning_rate": 0.0002, "epoch": 3.182984965163183, "step": 4340}, {"loss": 0.7385, "grad_norm": 1.0225574970245361, "learning_rate": 0.0002, "epoch": 3.1903190319031904, "step": 4350}, {"loss": 0.8652, "grad_norm": 1.1633386611938477, "learning_rate": 0.0002, "epoch": 3.1976530986431975, "step": 4360}, {"loss": 0.7259, "grad_norm": 0.8915148973464966, "learning_rate": 0.0002, "epoch": 3.204987165383205, "step": 4370}, {"loss": 0.8061, "grad_norm": 0.9156812429428101, "learning_rate": 0.0002, "epoch": 3.2123212321232124, "step": 4380}, {"loss": 0.8189, "grad_norm": 0.6363258957862854, "learning_rate": 0.0002, "epoch": 3.21965529886322, "step": 4390}, {"loss": 0.7996, "grad_norm": 0.579099178314209, "learning_rate": 0.0002, "epoch": 3.226989365603227, "step": 4400}, {"loss": 0.8592, "grad_norm": 0.8778146505355835, "learning_rate": 0.0002, "epoch": 3.2343234323432344, "step": 4410}, {"loss": 0.8281, "grad_norm": 0.8356770873069763, "learning_rate": 0.0002, "epoch": 3.241657499083242, "step": 4420}, {"loss": 0.8484, "grad_norm": 0.702032208442688, "learning_rate": 0.0002, "epoch": 3.248991565823249, "step": 4430}, {"loss": 0.7227, "grad_norm": 0.6386539340019226, "learning_rate": 0.0002, "epoch": 3.2563256325632564, "step": 4440}, {"loss": 0.8374, "grad_norm": 0.7008408904075623, "learning_rate": 0.0002, "epoch": 3.263659699303264, "step": 4450}, {"loss": 0.7572, "grad_norm": 0.9556332230567932, "learning_rate": 0.0002, "epoch": 3.270993766043271, "step": 4460}, {"loss": 0.743, "grad_norm": 0.5667835474014282, "learning_rate": 0.0002, "epoch": 3.2783278327832783, "step": 4470}, {"loss": 0.8152, "grad_norm": 0.8239172697067261, "learning_rate": 0.0002, "epoch": 3.285661899523286, "step": 4480}, {"loss": 0.756, "grad_norm": 0.7045050859451294, "learning_rate": 0.0002, "epoch": 3.292995966263293, "step": 4490}, {"loss": 0.7655, "grad_norm": 0.7131434082984924, "learning_rate": 0.0002, "epoch": 3.3003300330033003, "step": 4500}, {"loss": 0.836, "grad_norm": 0.6924910545349121, "learning_rate": 0.0002, "epoch": 3.3076640997433078, "step": 4510}, {"loss": 0.736, "grad_norm": 0.8945356607437134, "learning_rate": 0.0002, "epoch": 3.3149981664833152, "step": 4520}, {"loss": 0.7575, "grad_norm": 0.6546903252601624, "learning_rate": 0.0002, "epoch": 3.3223322332233223, "step": 4530}, {"loss": 0.7893, "grad_norm": 0.8206679224967957, "learning_rate": 0.0002, "epoch": 3.3296662999633297, "step": 4540}, {"loss": 0.7502, "grad_norm": 0.6482203602790833, "learning_rate": 0.0002, "epoch": 3.3370003667033368, "step": 4550}, {"loss": 0.8172, "grad_norm": 0.7558760046958923, "learning_rate": 0.0002, "epoch": 3.3443344334433442, "step": 4560}, {"loss": 0.744, "grad_norm": 0.7794756889343262, "learning_rate": 0.0002, "epoch": 3.3516685001833517, "step": 4570}, {"loss": 0.7385, "grad_norm": 0.7382805943489075, "learning_rate": 0.0002, "epoch": 3.359002566923359, "step": 4580}, {"loss": 0.8511, "grad_norm": 0.5912511944770813, "learning_rate": 0.0002, "epoch": 3.366336633663366, "step": 4590}, {"loss": 0.8272, "grad_norm": 0.7444885969161987, "learning_rate": 0.0002, "epoch": 3.3736707004033737, "step": 4600}, {"loss": 0.7927, "grad_norm": 0.7354922890663147, "learning_rate": 0.0002, "epoch": 3.381004767143381, "step": 4610}, {"loss": 0.7183, "grad_norm": 0.7685934901237488, "learning_rate": 0.0002, "epoch": 3.388338833883388, "step": 4620}, {"loss": 0.7436, "grad_norm": 0.61041259765625, "learning_rate": 0.0002, "epoch": 3.3956729006233957, "step": 4630}, {"loss": 0.7661, "grad_norm": 0.6820451021194458, "learning_rate": 0.0002, "epoch": 3.403006967363403, "step": 4640}, {"loss": 0.8796, "grad_norm": 0.5819534063339233, "learning_rate": 0.0002, "epoch": 3.41034103410341, "step": 4650}, {"loss": 0.7314, "grad_norm": 0.705410897731781, "learning_rate": 0.0002, "epoch": 3.4176751008434176, "step": 4660}, {"loss": 0.7901, "grad_norm": 0.8052892088890076, "learning_rate": 0.0002, "epoch": 3.425009167583425, "step": 4670}, {"loss": 0.7298, "grad_norm": 0.7746483087539673, "learning_rate": 0.0002, "epoch": 3.432343234323432, "step": 4680}, {"loss": 0.7976, "grad_norm": 0.7713689804077148, "learning_rate": 0.0002, "epoch": 3.4396773010634396, "step": 4690}, {"loss": 0.7427, "grad_norm": 0.810371994972229, "learning_rate": 0.0002, "epoch": 3.447011367803447, "step": 4700}, {"loss": 0.7594, "grad_norm": 0.7702969312667847, "learning_rate": 0.0002, "epoch": 3.4543454345434546, "step": 4710}, {"loss": 0.7957, "grad_norm": 0.7069268822669983, "learning_rate": 0.0002, "epoch": 3.4616795012834616, "step": 4720}, {"loss": 0.8199, "grad_norm": 0.7640359401702881, "learning_rate": 0.0002, "epoch": 3.469013568023469, "step": 4730}, {"loss": 0.6875, "grad_norm": 0.8661707639694214, "learning_rate": 0.0002, "epoch": 3.4763476347634765, "step": 4740}, {"loss": 0.8528, "grad_norm": 0.9970282912254333, "learning_rate": 0.0002, "epoch": 3.4836817015034836, "step": 4750}, {"loss": 0.8462, "grad_norm": 0.5824355483055115, "learning_rate": 0.0002, "epoch": 3.491015768243491, "step": 4760}, {"loss": 0.851, "grad_norm": 1.3072649240493774, "learning_rate": 0.0002, "epoch": 3.4983498349834985, "step": 4770}, {"loss": 0.9101, "grad_norm": 0.873978316783905, "learning_rate": 0.0002, "epoch": 3.5056839017235055, "step": 4780}, {"loss": 0.7403, "grad_norm": 0.5526657104492188, "learning_rate": 0.0002, "epoch": 3.513017968463513, "step": 4790}, {"loss": 0.7921, "grad_norm": 0.790894627571106, "learning_rate": 0.0002, "epoch": 3.5203520352035205, "step": 4800}, {"loss": 0.831, "grad_norm": 0.8119630217552185, "learning_rate": 0.0002, "epoch": 3.5276861019435275, "step": 4810}, {"loss": 0.7351, "grad_norm": 0.633212149143219, "learning_rate": 0.0002, "epoch": 3.535020168683535, "step": 4820}, {"loss": 0.8505, "grad_norm": 0.703029990196228, "learning_rate": 0.0002, "epoch": 3.5423542354235424, "step": 4830}, {"loss": 0.7204, "grad_norm": 0.7603771686553955, "learning_rate": 0.0002, "epoch": 3.54968830216355, "step": 4840}, {"loss": 0.8868, "grad_norm": 0.6260480880737305, "learning_rate": 0.0002, "epoch": 3.557022368903557, "step": 4850}, {"loss": 0.8137, "grad_norm": 0.8203664422035217, "learning_rate": 0.0002, "epoch": 3.5643564356435644, "step": 4860}, {"loss": 0.8821, "grad_norm": 0.7793813347816467, "learning_rate": 0.0002, "epoch": 3.5716905023835714, "step": 4870}, {"loss": 0.8164, "grad_norm": 0.7667397260665894, "learning_rate": 0.0002, "epoch": 3.579024569123579, "step": 4880}, {"loss": 0.7597, "grad_norm": 0.8198829889297485, "learning_rate": 0.0002, "epoch": 3.5863586358635864, "step": 4890}, {"loss": 0.7027, "grad_norm": 0.7689233422279358, "learning_rate": 0.0002, "epoch": 3.593692702603594, "step": 4900}, {"loss": 0.804, "grad_norm": 0.7870983481407166, "learning_rate": 0.0002, "epoch": 3.601026769343601, "step": 4910}, {"loss": 0.8269, "grad_norm": 0.8133853077888489, "learning_rate": 0.0002, "epoch": 3.6083608360836084, "step": 4920}, {"loss": 0.8515, "grad_norm": 1.308401346206665, "learning_rate": 0.0002, "epoch": 3.615694902823616, "step": 4930}, {"loss": 0.8494, "grad_norm": 0.7131121754646301, "learning_rate": 0.0002, "epoch": 3.623028969563623, "step": 4940}, {"loss": 0.7235, "grad_norm": 0.6825910210609436, "learning_rate": 0.0002, "epoch": 3.6303630363036303, "step": 4950}, {"loss": 0.7824, "grad_norm": 0.7254678606987, "learning_rate": 0.0002, "epoch": 3.637697103043638, "step": 4960}, {"loss": 0.7983, "grad_norm": 0.8045085072517395, "learning_rate": 0.0002, "epoch": 3.6450311697836453, "step": 4970}, {"loss": 0.8223, "grad_norm": 0.6991777420043945, "learning_rate": 0.0002, "epoch": 3.6523652365236523, "step": 4980}, {"loss": 0.7806, "grad_norm": 0.7804713249206543, "learning_rate": 0.0002, "epoch": 3.6596993032636598, "step": 4990}, {"loss": 0.8402, "grad_norm": 0.8525708317756653, "learning_rate": 0.0002, "epoch": 3.667033370003667, "step": 5000}, {"loss": 0.8496, "grad_norm": 0.7959994673728943, "learning_rate": 0.0002, "epoch": 3.6743674367436743, "step": 5010}, {"loss": 0.8022, "grad_norm": 0.8103628158569336, "learning_rate": 0.0002, "epoch": 3.6817015034836817, "step": 5020}, {"loss": 0.7376, "grad_norm": 0.7517836093902588, "learning_rate": 0.0002, "epoch": 3.689035570223689, "step": 5030}, {"loss": 0.8375, "grad_norm": 0.6878514289855957, "learning_rate": 0.0002, "epoch": 3.6963696369636962, "step": 5040}, {"loss": 0.7998, "grad_norm": 1.2371820211410522, "learning_rate": 0.0002, "epoch": 3.7037037037037037, "step": 5050}, {"loss": 0.6941, "grad_norm": 0.6567103862762451, "learning_rate": 0.0002, "epoch": 3.711037770443711, "step": 5060}, {"loss": 0.8465, "grad_norm": 1.1254922151565552, "learning_rate": 0.0002, "epoch": 3.718371837183718, "step": 5070}, {"loss": 0.8365, "grad_norm": 0.6796132326126099, "learning_rate": 0.0002, "epoch": 3.7257059039237257, "step": 5080}, {"loss": 0.7818, "grad_norm": 0.7285300493240356, "learning_rate": 0.0002, "epoch": 3.733039970663733, "step": 5090}, {"loss": 0.8581, "grad_norm": 0.8931500911712646, "learning_rate": 0.0002, "epoch": 3.7403740374037406, "step": 5100}, {"loss": 0.8181, "grad_norm": 0.6256856918334961, "learning_rate": 0.0002, "epoch": 3.7477081041437477, "step": 5110}, {"loss": 0.743, "grad_norm": 0.79310142993927, "learning_rate": 0.0002, "epoch": 3.755042170883755, "step": 5120}, {"loss": 0.8235, "grad_norm": 0.6594041585922241, "learning_rate": 0.0002, "epoch": 3.762376237623762, "step": 5130}, {"loss": 0.6925, "grad_norm": 0.7029327750205994, "learning_rate": 0.0002, "epoch": 3.7697103043637696, "step": 5140}, {"loss": 0.7457, "grad_norm": 0.5880070328712463, "learning_rate": 0.0002, "epoch": 3.777044371103777, "step": 5150}, {"loss": 0.8716, "grad_norm": 0.7578945159912109, "learning_rate": 0.0002, "epoch": 3.7843784378437846, "step": 5160}, {"loss": 0.8819, "grad_norm": 0.8276378512382507, "learning_rate": 0.0002, "epoch": 3.7917125045837916, "step": 5170}, {"loss": 0.7559, "grad_norm": 0.7627953886985779, "learning_rate": 0.0002, "epoch": 3.799046571323799, "step": 5180}, {"loss": 0.7665, "grad_norm": 0.8169086575508118, "learning_rate": 0.0002, "epoch": 3.806380638063806, "step": 5190}, {"loss": 0.761, "grad_norm": 0.6605030298233032, "learning_rate": 0.0002, "epoch": 3.8137147048038136, "step": 5200}, {"loss": 0.8804, "grad_norm": 0.5837286114692688, "learning_rate": 0.0002, "epoch": 3.821048771543821, "step": 5210}, {"loss": 0.8369, "grad_norm": 1.2422157526016235, "learning_rate": 0.0002, "epoch": 3.8283828382838285, "step": 5220}, {"loss": 0.8431, "grad_norm": 0.6589220762252808, "learning_rate": 0.0002, "epoch": 3.8357169050238356, "step": 5230}, {"loss": 0.7686, "grad_norm": 0.8567556142807007, "learning_rate": 0.0002, "epoch": 3.843050971763843, "step": 5240}, {"loss": 0.8652, "grad_norm": 0.6490627527236938, "learning_rate": 0.0002, "epoch": 3.8503850385038505, "step": 5250}, {"loss": 0.7386, "grad_norm": 0.620232880115509, "learning_rate": 0.0002, "epoch": 3.8577191052438575, "step": 5260}, {"loss": 0.9192, "grad_norm": 0.7685128450393677, "learning_rate": 0.0002, "epoch": 3.865053171983865, "step": 5270}, {"loss": 0.872, "grad_norm": 0.8113296627998352, "learning_rate": 0.0002, "epoch": 3.8723872387238725, "step": 5280}, {"loss": 0.7156, "grad_norm": 0.8092675805091858, "learning_rate": 0.0002, "epoch": 3.87972130546388, "step": 5290}, {"loss": 0.7325, "grad_norm": 0.583570122718811, "learning_rate": 0.0002, "epoch": 3.887055372203887, "step": 5300}, {"loss": 0.9333, "grad_norm": 1.712363600730896, "learning_rate": 0.0002, "epoch": 3.8943894389438944, "step": 5310}, {"loss": 0.7537, "grad_norm": 0.6673534512519836, "learning_rate": 0.0002, "epoch": 3.9017235056839015, "step": 5320}, {"loss": 0.7035, "grad_norm": 1.9770312309265137, "learning_rate": 0.0002, "epoch": 3.909057572423909, "step": 5330}, {"loss": 0.8793, "grad_norm": 0.6430999636650085, "learning_rate": 0.0002, "epoch": 3.9163916391639164, "step": 5340}, {"loss": 0.839, "grad_norm": 1.0159571170806885, "learning_rate": 0.0002, "epoch": 3.923725705903924, "step": 5350}, {"loss": 0.9332, "grad_norm": 0.8607584834098816, "learning_rate": 0.0002, "epoch": 3.931059772643931, "step": 5360}, {"loss": 0.7261, "grad_norm": 0.6967900991439819, "learning_rate": 0.0002, "epoch": 3.9383938393839384, "step": 5370}, {"loss": 0.8456, "grad_norm": 0.7683077454566956, "learning_rate": 0.0002, "epoch": 3.945727906123946, "step": 5380}, {"loss": 0.7682, "grad_norm": 0.6805762648582458, "learning_rate": 0.0002, "epoch": 3.953061972863953, "step": 5390}, {"loss": 0.7746, "grad_norm": 0.7033619284629822, "learning_rate": 0.0002, "epoch": 3.9603960396039604, "step": 5400}, {"loss": 0.8393, "grad_norm": 0.966112494468689, "learning_rate": 0.0002, "epoch": 3.967730106343968, "step": 5410}, {"loss": 0.8316, "grad_norm": 0.8467881083488464, "learning_rate": 0.0002, "epoch": 3.9750641730839753, "step": 5420}, {"loss": 0.8084, "grad_norm": 0.8005317449569702, "learning_rate": 0.0002, "epoch": 3.9823982398239823, "step": 5430}, {"loss": 0.7168, "grad_norm": 1.1615241765975952, "learning_rate": 0.0002, "epoch": 3.98973230656399, "step": 5440}, {"loss": 0.8263, "grad_norm": 0.6121614575386047, "learning_rate": 0.0002, "epoch": 3.997066373303997, "step": 5450}]}
+{"epoch": 4.999633296662999, "step": 6817, "epoch_duration": 1471.577754497528, "total_accumulated_duration": 7377.027250051498, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7672.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 1.9722, "grad_norm": 0.47521963715553284, "learning_rate": 0.0002, "epoch": 0.007334066740007334, "step": 10}, {"loss": 1.4821, "grad_norm": 0.5395162105560303, "learning_rate": 0.0002, "epoch": 0.014668133480014669, "step": 20}, {"loss": 1.4202, "grad_norm": 0.4305780231952667, "learning_rate": 0.0002, "epoch": 0.022002200220022004, "step": 30}, {"loss": 1.4271, "grad_norm": 0.6938246488571167, "learning_rate": 0.0002, "epoch": 0.029336266960029337, "step": 40}, {"loss": 1.3112, "grad_norm": 1.5133819580078125, "learning_rate": 0.0002, "epoch": 0.03667033370003667, "step": 50}, {"loss": 1.3132, "grad_norm": 0.9173883199691772, "learning_rate": 0.0002, "epoch": 0.04400440044004401, "step": 60}, {"loss": 1.2844, "grad_norm": 0.4619861841201782, "learning_rate": 0.0002, "epoch": 0.05133846718005134, "step": 70}, {"loss": 1.2108, "grad_norm": 0.46118637919425964, "learning_rate": 0.0002, "epoch": 0.058672533920058674, "step": 80}, {"loss": 1.3441, "grad_norm": 0.4468648135662079, "learning_rate": 0.0002, "epoch": 0.066006600660066, "step": 90}, {"loss": 1.1863, "grad_norm": 0.46123769879341125, "learning_rate": 0.0002, "epoch": 0.07334066740007333, "step": 100}, {"loss": 1.2772, "grad_norm": 0.4859139025211334, "learning_rate": 0.0002, "epoch": 0.08067473414008068, "step": 110}, {"loss": 1.2087, "grad_norm": 0.4384922385215759, "learning_rate": 0.0002, "epoch": 0.08800880088008801, "step": 120}, {"loss": 1.2927, "grad_norm": 0.39519360661506653, "learning_rate": 0.0002, "epoch": 0.09534286762009535, "step": 130}, {"loss": 1.2349, "grad_norm": 0.4049859344959259, "learning_rate": 0.0002, "epoch": 0.10267693436010268, "step": 140}, {"loss": 1.293, "grad_norm": 0.4605638086795807, "learning_rate": 0.0002, "epoch": 0.11001100110011001, "step": 150}, {"loss": 1.2659, "grad_norm": 0.4201928377151489, "learning_rate": 0.0002, "epoch": 0.11734506784011735, "step": 160}, {"loss": 1.3961, "grad_norm": 0.5367777347564697, "learning_rate": 0.0002, "epoch": 0.12467913458012468, "step": 170}, {"loss": 1.2481, "grad_norm": 0.41752299666404724, "learning_rate": 0.0002, "epoch": 0.132013201320132, "step": 180}, {"loss": 1.207, "grad_norm": 0.31597763299942017, "learning_rate": 0.0002, "epoch": 0.13934726806013933, "step": 190}, {"loss": 1.2441, "grad_norm": 0.7468788623809814, "learning_rate": 0.0002, "epoch": 0.14668133480014667, "step": 200}, {"loss": 1.199, "grad_norm": 0.3403034508228302, "learning_rate": 0.0002, "epoch": 0.15401540154015403, "step": 210}, {"loss": 1.2439, "grad_norm": 0.34240293502807617, "learning_rate": 0.0002, "epoch": 0.16134946828016136, "step": 220}, {"loss": 1.2022, "grad_norm": 0.356158971786499, "learning_rate": 0.0002, "epoch": 0.1686835350201687, "step": 230}, {"loss": 1.207, "grad_norm": 0.3448857367038727, "learning_rate": 0.0002, "epoch": 0.17601760176017603, "step": 240}, {"loss": 1.2156, "grad_norm": 0.3475699722766876, "learning_rate": 0.0002, "epoch": 0.18335166850018336, "step": 250}, {"loss": 1.1551, "grad_norm": 0.2770358622074127, "learning_rate": 0.0002, "epoch": 0.1906857352401907, "step": 260}, {"loss": 1.2238, "grad_norm": 0.4310270845890045, "learning_rate": 0.0002, "epoch": 0.19801980198019803, "step": 270}, {"loss": 1.2917, "grad_norm": 0.335041880607605, "learning_rate": 0.0002, "epoch": 0.20535386872020536, "step": 280}, {"loss": 1.0959, "grad_norm": 0.3420602083206177, "learning_rate": 0.0002, "epoch": 0.2126879354602127, "step": 290}, {"loss": 1.1232, "grad_norm": 0.325001060962677, "learning_rate": 0.0002, "epoch": 0.22002200220022003, "step": 300}, {"loss": 1.2007, "grad_norm": 0.3027827739715576, "learning_rate": 0.0002, "epoch": 0.22735606894022736, "step": 310}, {"loss": 1.1803, "grad_norm": 0.435550719499588, "learning_rate": 0.0002, "epoch": 0.2346901356802347, "step": 320}, {"loss": 1.2045, "grad_norm": 0.3884522616863251, "learning_rate": 0.0002, "epoch": 0.24202420242024203, "step": 330}, {"loss": 1.2481, "grad_norm": 0.7736002206802368, "learning_rate": 0.0002, "epoch": 0.24935826916024936, "step": 340}, {"loss": 1.3606, "grad_norm": 0.35052821040153503, "learning_rate": 0.0002, "epoch": 0.2566923359002567, "step": 350}, {"loss": 1.2129, "grad_norm": 0.3311890959739685, "learning_rate": 0.0002, "epoch": 0.264026402640264, "step": 360}, {"loss": 1.2219, "grad_norm": 0.7473500370979309, "learning_rate": 0.0002, "epoch": 0.27136046938027136, "step": 370}, {"loss": 1.2712, "grad_norm": 0.3681875765323639, "learning_rate": 0.0002, "epoch": 0.27869453612027867, "step": 380}, {"loss": 1.2258, "grad_norm": 0.3764737844467163, "learning_rate": 0.0002, "epoch": 0.28602860286028603, "step": 390}, {"loss": 1.1917, "grad_norm": 0.4243989586830139, "learning_rate": 0.0002, "epoch": 0.29336266960029334, "step": 400}, {"loss": 1.199, "grad_norm": 0.2658531963825226, "learning_rate": 0.0002, "epoch": 0.3006967363403007, "step": 410}, {"loss": 1.1622, "grad_norm": 0.3436793386936188, "learning_rate": 0.0002, "epoch": 0.30803080308030806, "step": 420}, {"loss": 1.2953, "grad_norm": 0.5101129412651062, "learning_rate": 0.0002, "epoch": 0.31536486982031536, "step": 430}, {"loss": 1.1557, "grad_norm": 0.3319750726222992, "learning_rate": 0.0002, "epoch": 0.3226989365603227, "step": 440}, {"loss": 1.1804, "grad_norm": 0.385148286819458, "learning_rate": 0.0002, "epoch": 0.33003300330033003, "step": 450}, {"loss": 1.1808, "grad_norm": 0.3477935791015625, "learning_rate": 0.0002, "epoch": 0.3373670700403374, "step": 460}, {"loss": 1.1877, "grad_norm": 0.29748716950416565, "learning_rate": 0.0002, "epoch": 0.3447011367803447, "step": 470}, {"loss": 1.19, "grad_norm": 0.34083324670791626, "learning_rate": 0.0002, "epoch": 0.35203520352035206, "step": 480}, {"loss": 1.2, "grad_norm": 0.36904552578926086, "learning_rate": 0.0002, "epoch": 0.35936927026035936, "step": 490}, {"loss": 1.2223, "grad_norm": 0.315483033657074, "learning_rate": 0.0002, "epoch": 0.3667033370003667, "step": 500}, {"loss": 1.1461, "grad_norm": 0.44897955656051636, "learning_rate": 0.0002, "epoch": 0.37403740374037403, "step": 510}, {"loss": 1.3035, "grad_norm": 0.3160701394081116, "learning_rate": 0.0002, "epoch": 0.3813714704803814, "step": 520}, {"loss": 1.3197, "grad_norm": 0.29584741592407227, "learning_rate": 0.0002, "epoch": 0.3887055372203887, "step": 530}, {"loss": 1.2983, "grad_norm": 0.5430002808570862, "learning_rate": 0.0002, "epoch": 0.39603960396039606, "step": 540}, {"loss": 1.2459, "grad_norm": 0.2908070683479309, "learning_rate": 0.0002, "epoch": 0.40337367070040336, "step": 550}, {"loss": 1.2384, "grad_norm": 0.35066530108451843, "learning_rate": 0.0002, "epoch": 0.4107077374404107, "step": 560}, {"loss": 1.1784, "grad_norm": 0.37588003277778625, "learning_rate": 0.0002, "epoch": 0.41804180418041803, "step": 570}, {"loss": 1.2334, "grad_norm": 0.3112126886844635, "learning_rate": 0.0002, "epoch": 0.4253758709204254, "step": 580}, {"loss": 1.1439, "grad_norm": 0.35577139258384705, "learning_rate": 0.0002, "epoch": 0.4327099376604327, "step": 590}, {"loss": 1.184, "grad_norm": 0.31706422567367554, "learning_rate": 0.0002, "epoch": 0.44004400440044006, "step": 600}, {"loss": 1.2081, "grad_norm": 0.3249092102050781, "learning_rate": 0.0002, "epoch": 0.44737807114044736, "step": 610}, {"loss": 1.0824, "grad_norm": 0.3842705488204956, "learning_rate": 0.0002, "epoch": 0.4547121378804547, "step": 620}, {"loss": 1.2257, "grad_norm": 0.390991747379303, "learning_rate": 0.0002, "epoch": 0.46204620462046203, "step": 630}, {"loss": 1.1954, "grad_norm": 0.27532413601875305, "learning_rate": 0.0002, "epoch": 0.4693802713604694, "step": 640}, {"loss": 1.1058, "grad_norm": 0.31412816047668457, "learning_rate": 0.0002, "epoch": 0.4767143381004767, "step": 650}, {"loss": 1.1312, "grad_norm": 0.32117101550102234, "learning_rate": 0.0002, "epoch": 0.48404840484048406, "step": 660}, {"loss": 1.2423, "grad_norm": 0.3810010254383087, "learning_rate": 0.0002, "epoch": 0.49138247158049136, "step": 670}, {"loss": 1.1978, "grad_norm": 0.36289164423942566, "learning_rate": 0.0002, "epoch": 0.4987165383204987, "step": 680}, {"loss": 1.2034, "grad_norm": 0.34458720684051514, "learning_rate": 0.0002, "epoch": 0.506050605060506, "step": 690}, {"loss": 1.1756, "grad_norm": 0.32844600081443787, "learning_rate": 0.0002, "epoch": 0.5133846718005134, "step": 700}, {"loss": 1.0807, "grad_norm": 0.3144175708293915, "learning_rate": 0.0002, "epoch": 0.5207187385405208, "step": 710}, {"loss": 1.1952, "grad_norm": 0.3898887634277344, "learning_rate": 0.0002, "epoch": 0.528052805280528, "step": 720}, {"loss": 1.1244, "grad_norm": 1.3220758438110352, "learning_rate": 0.0002, "epoch": 0.5353868720205354, "step": 730}, {"loss": 1.227, "grad_norm": 0.3635874390602112, "learning_rate": 0.0002, "epoch": 0.5427209387605427, "step": 740}, {"loss": 1.2169, "grad_norm": 0.3138217628002167, "learning_rate": 0.0002, "epoch": 0.5500550055005501, "step": 750}, {"loss": 1.1516, "grad_norm": 0.4063207805156708, "learning_rate": 0.0002, "epoch": 0.5573890722405573, "step": 760}, {"loss": 1.1954, "grad_norm": 0.3926219940185547, "learning_rate": 0.0002, "epoch": 0.5647231389805647, "step": 770}, {"loss": 1.1726, "grad_norm": 0.31954652070999146, "learning_rate": 0.0002, "epoch": 0.5720572057205721, "step": 780}, {"loss": 1.2977, "grad_norm": 0.4248711168766022, "learning_rate": 0.0002, "epoch": 0.5793912724605794, "step": 790}, {"loss": 1.1728, "grad_norm": 0.643004834651947, "learning_rate": 0.0002, "epoch": 0.5867253392005867, "step": 800}, {"loss": 1.1793, "grad_norm": 0.3479592800140381, "learning_rate": 0.0002, "epoch": 0.594059405940594, "step": 810}, {"loss": 1.2426, "grad_norm": 0.4684754014015198, "learning_rate": 0.0002, "epoch": 0.6013934726806014, "step": 820}, {"loss": 1.2002, "grad_norm": 0.3739790916442871, "learning_rate": 0.0002, "epoch": 0.6087275394206088, "step": 830}, {"loss": 1.2139, "grad_norm": 0.40884748101234436, "learning_rate": 0.0002, "epoch": 0.6160616061606161, "step": 840}, {"loss": 1.1557, "grad_norm": 0.9722164273262024, "learning_rate": 0.0002, "epoch": 0.6233956729006234, "step": 850}, {"loss": 1.3069, "grad_norm": 0.42992347478866577, "learning_rate": 0.0002, "epoch": 0.6307297396406307, "step": 860}, {"loss": 1.1339, "grad_norm": 0.36654195189476013, "learning_rate": 0.0002, "epoch": 0.6380638063806381, "step": 870}, {"loss": 1.1932, "grad_norm": 0.4113832116127014, "learning_rate": 0.0002, "epoch": 0.6453978731206454, "step": 880}, {"loss": 1.2163, "grad_norm": 0.2948838770389557, "learning_rate": 0.0002, "epoch": 0.6527319398606527, "step": 890}, {"loss": 1.1081, "grad_norm": 0.38330280780792236, "learning_rate": 0.0002, "epoch": 0.6600660066006601, "step": 900}, {"loss": 1.1342, "grad_norm": 0.4428867697715759, "learning_rate": 0.0002, "epoch": 0.6674000733406674, "step": 910}, {"loss": 1.1021, "grad_norm": 0.23659265041351318, "learning_rate": 0.0002, "epoch": 0.6747341400806748, "step": 920}, {"loss": 1.1226, "grad_norm": 0.323685884475708, "learning_rate": 0.0002, "epoch": 0.682068206820682, "step": 930}, {"loss": 1.0853, "grad_norm": 0.39157727360725403, "learning_rate": 0.0002, "epoch": 0.6894022735606894, "step": 940}, {"loss": 1.1435, "grad_norm": 0.27189481258392334, "learning_rate": 0.0002, "epoch": 0.6967363403006968, "step": 950}, {"loss": 1.1033, "grad_norm": 0.529883861541748, "learning_rate": 0.0002, "epoch": 0.7040704070407041, "step": 960}, {"loss": 1.139, "grad_norm": 0.34758689999580383, "learning_rate": 0.0002, "epoch": 0.7114044737807114, "step": 970}, {"loss": 1.2197, "grad_norm": 0.831749439239502, "learning_rate": 0.0002, "epoch": 0.7187385405207187, "step": 980}, {"loss": 1.158, "grad_norm": 0.4438304007053375, "learning_rate": 0.0002, "epoch": 0.7260726072607261, "step": 990}, {"loss": 1.1021, "grad_norm": 0.33840006589889526, "learning_rate": 0.0002, "epoch": 0.7334066740007334, "step": 1000}, {"loss": 1.254, "grad_norm": 0.3454797863960266, "learning_rate": 0.0002, "epoch": 0.7407407407407407, "step": 1010}, {"loss": 1.106, "grad_norm": 0.38999441266059875, "learning_rate": 0.0002, "epoch": 0.7480748074807481, "step": 1020}, {"loss": 1.1428, "grad_norm": 0.2829911708831787, "learning_rate": 0.0002, "epoch": 0.7554088742207554, "step": 1030}, {"loss": 1.2123, "grad_norm": 0.36918163299560547, "learning_rate": 0.0002, "epoch": 0.7627429409607628, "step": 1040}, {"loss": 1.3028, "grad_norm": 0.3415680229663849, "learning_rate": 0.0002, "epoch": 0.77007700770077, "step": 1050}, {"loss": 1.1939, "grad_norm": 0.2974182963371277, "learning_rate": 0.0002, "epoch": 0.7774110744407774, "step": 1060}, {"loss": 1.194, "grad_norm": 0.3880919814109802, "learning_rate": 0.0002, "epoch": 0.7847451411807848, "step": 1070}, {"loss": 1.1095, "grad_norm": 0.33503302931785583, "learning_rate": 0.0002, "epoch": 0.7920792079207921, "step": 1080}, {"loss": 1.2111, "grad_norm": 0.3728407025337219, "learning_rate": 0.0002, "epoch": 0.7994132746607994, "step": 1090}, {"loss": 1.0835, "grad_norm": 0.3509373664855957, "learning_rate": 0.0002, "epoch": 0.8067473414008067, "step": 1100}, {"loss": 1.2661, "grad_norm": 0.42228564620018005, "learning_rate": 0.0002, "epoch": 0.8140814081408141, "step": 1110}, {"loss": 1.1788, "grad_norm": 0.313467800617218, "learning_rate": 0.0002, "epoch": 0.8214154748808215, "step": 1120}, {"loss": 1.1971, "grad_norm": 0.3378850817680359, "learning_rate": 0.0002, "epoch": 0.8287495416208287, "step": 1130}, {"loss": 1.1238, "grad_norm": 0.43200382590293884, "learning_rate": 0.0002, "epoch": 0.8360836083608361, "step": 1140}, {"loss": 1.3203, "grad_norm": 0.3309599459171295, "learning_rate": 0.0002, "epoch": 0.8434176751008434, "step": 1150}, {"loss": 1.1062, "grad_norm": 0.3526846170425415, "learning_rate": 0.0002, "epoch": 0.8507517418408508, "step": 1160}, {"loss": 1.0851, "grad_norm": 1.2722247838974, "learning_rate": 0.0002, "epoch": 0.858085808580858, "step": 1170}, {"loss": 1.0785, "grad_norm": 0.34142059087753296, "learning_rate": 0.0002, "epoch": 0.8654198753208654, "step": 1180}, {"loss": 1.2187, "grad_norm": 0.3805823028087616, "learning_rate": 0.0002, "epoch": 0.8727539420608728, "step": 1190}, {"loss": 1.1215, "grad_norm": 0.3931232690811157, "learning_rate": 0.0002, "epoch": 0.8800880088008801, "step": 1200}, {"loss": 1.0948, "grad_norm": 0.2937372624874115, "learning_rate": 0.0002, "epoch": 0.8874220755408874, "step": 1210}, {"loss": 1.1228, "grad_norm": 0.3757196366786957, "learning_rate": 0.0002, "epoch": 0.8947561422808947, "step": 1220}, {"loss": 1.1222, "grad_norm": 0.3502705991268158, "learning_rate": 0.0002, "epoch": 0.9020902090209021, "step": 1230}, {"loss": 1.2242, "grad_norm": 0.32758915424346924, "learning_rate": 0.0002, "epoch": 0.9094242757609095, "step": 1240}, {"loss": 1.215, "grad_norm": 0.37199416756629944, "learning_rate": 0.0002, "epoch": 0.9167583425009168, "step": 1250}, {"loss": 1.1225, "grad_norm": 0.3551490604877472, "learning_rate": 0.0002, "epoch": 0.9240924092409241, "step": 1260}, {"loss": 1.1966, "grad_norm": 0.2859550714492798, "learning_rate": 0.0002, "epoch": 0.9314264759809314, "step": 1270}, {"loss": 1.2186, "grad_norm": 0.427990585565567, "learning_rate": 0.0002, "epoch": 0.9387605427209388, "step": 1280}, {"loss": 1.2848, "grad_norm": 0.33717992901802063, "learning_rate": 0.0002, "epoch": 0.9460946094609461, "step": 1290}, {"loss": 1.1656, "grad_norm": 0.30225634574890137, "learning_rate": 0.0002, "epoch": 0.9534286762009534, "step": 1300}, {"loss": 1.2404, "grad_norm": 0.385821133852005, "learning_rate": 0.0002, "epoch": 0.9607627429409608, "step": 1310}, {"loss": 1.1932, "grad_norm": 0.35278066992759705, "learning_rate": 0.0002, "epoch": 0.9680968096809681, "step": 1320}, {"loss": 1.1071, "grad_norm": 0.49987098574638367, "learning_rate": 0.0002, "epoch": 0.9754308764209755, "step": 1330}, {"loss": 1.2259, "grad_norm": 0.3842747211456299, "learning_rate": 0.0002, "epoch": 0.9827649431609827, "step": 1340}, {"loss": 1.0862, "grad_norm": 0.6274653673171997, "learning_rate": 0.0002, "epoch": 0.9900990099009901, "step": 1350}, {"loss": 1.124, "grad_norm": 0.5239808559417725, "learning_rate": 0.0002, "epoch": 0.9974330766409975, "step": 1360}, {"eval_loss": 1.1822267770767212, "eval_runtime": 32.7389, "eval_samples_per_second": 13.165, "eval_steps_per_second": 1.649, "epoch": 0.9996332966629996, "step": 1363}, {"loss": 1.096, "grad_norm": 0.45311301946640015, "learning_rate": 0.0002, "epoch": 1.0047671433810048, "step": 1370}, {"loss": 1.0143, "grad_norm": 0.29685574769973755, "learning_rate": 0.0002, "epoch": 1.012101210121012, "step": 1380}, {"loss": 1.0302, "grad_norm": 0.3290937840938568, "learning_rate": 0.0002, "epoch": 1.0194352768610195, "step": 1390}, {"loss": 1.0295, "grad_norm": 0.3801758587360382, "learning_rate": 0.0002, "epoch": 1.0267693436010268, "step": 1400}, {"loss": 1.1226, "grad_norm": 0.794174313545227, "learning_rate": 0.0002, "epoch": 1.034103410341034, "step": 1410}, {"loss": 1.2232, "grad_norm": 0.3854154646396637, "learning_rate": 0.0002, "epoch": 1.0414374770810415, "step": 1420}, {"loss": 1.0652, "grad_norm": 0.32702451944351196, "learning_rate": 0.0002, "epoch": 1.0487715438210488, "step": 1430}, {"loss": 1.1144, "grad_norm": 0.7815203666687012, "learning_rate": 0.0002, "epoch": 1.056105610561056, "step": 1440}, {"loss": 1.1316, "grad_norm": 0.3087436854839325, "learning_rate": 0.0002, "epoch": 1.0634396773010635, "step": 1450}, {"loss": 1.1124, "grad_norm": 0.3847602903842926, "learning_rate": 0.0002, "epoch": 1.0707737440410707, "step": 1460}, {"loss": 1.1428, "grad_norm": 0.3693031370639801, "learning_rate": 0.0002, "epoch": 1.0781078107810782, "step": 1470}, {"loss": 1.0995, "grad_norm": 0.4111202359199524, "learning_rate": 0.0002, "epoch": 1.0854418775210855, "step": 1480}, {"loss": 1.0961, "grad_norm": 0.41452381014823914, "learning_rate": 0.0002, "epoch": 1.0927759442610927, "step": 1490}, {"loss": 1.1068, "grad_norm": 0.3336445093154907, "learning_rate": 0.0002, "epoch": 1.1001100110011002, "step": 1500}, {"loss": 1.0556, "grad_norm": 0.3923407793045044, "learning_rate": 0.0002, "epoch": 1.1074440777411074, "step": 1510}, {"loss": 1.1644, "grad_norm": 0.46215683221817017, "learning_rate": 0.0002, "epoch": 1.1147781444811147, "step": 1520}, {"loss": 1.1133, "grad_norm": 0.3592156767845154, "learning_rate": 0.0002, "epoch": 1.1221122112211221, "step": 1530}, {"loss": 1.0957, "grad_norm": 0.361110657453537, "learning_rate": 0.0002, "epoch": 1.1294462779611294, "step": 1540}, {"loss": 1.1553, "grad_norm": 0.5317131280899048, "learning_rate": 0.0002, "epoch": 1.1367803447011369, "step": 1550}, {"loss": 1.0368, "grad_norm": 0.3882388174533844, "learning_rate": 0.0002, "epoch": 1.1441144114411441, "step": 1560}, {"loss": 1.0805, "grad_norm": 0.3259428143501282, "learning_rate": 0.0002, "epoch": 1.1514484781811514, "step": 1570}, {"loss": 1.1819, "grad_norm": 0.410935640335083, "learning_rate": 0.0002, "epoch": 1.1587825449211588, "step": 1580}, {"loss": 1.1143, "grad_norm": 0.44940185546875, "learning_rate": 0.0002, "epoch": 1.166116611661166, "step": 1590}, {"loss": 1.0334, "grad_norm": 0.5106484293937683, "learning_rate": 0.0002, "epoch": 1.1734506784011733, "step": 1600}, {"loss": 1.2376, "grad_norm": 0.6603665947914124, "learning_rate": 0.0002, "epoch": 1.1807847451411808, "step": 1610}, {"loss": 1.1227, "grad_norm": 0.4799964129924774, "learning_rate": 0.0002, "epoch": 1.188118811881188, "step": 1620}, {"loss": 1.1191, "grad_norm": 0.4389883279800415, "learning_rate": 0.0002, "epoch": 1.1954528786211955, "step": 1630}, {"loss": 1.0667, "grad_norm": 0.4188813269138336, "learning_rate": 0.0002, "epoch": 1.2027869453612028, "step": 1640}, {"loss": 1.0605, "grad_norm": 0.7132157683372498, "learning_rate": 0.0002, "epoch": 1.21012101210121, "step": 1650}, {"loss": 1.0204, "grad_norm": 0.507480263710022, "learning_rate": 0.0002, "epoch": 1.2174550788412175, "step": 1660}, {"loss": 0.9948, "grad_norm": 0.9452332854270935, "learning_rate": 0.0002, "epoch": 1.2247891455812248, "step": 1670}, {"loss": 1.0228, "grad_norm": 0.4121614992618561, "learning_rate": 0.0002, "epoch": 1.2321232123212322, "step": 1680}, {"loss": 1.0366, "grad_norm": 0.34230247139930725, "learning_rate": 0.0002, "epoch": 1.2394572790612395, "step": 1690}, {"loss": 1.1289, "grad_norm": 0.4026208817958832, "learning_rate": 0.0002, "epoch": 1.2467913458012467, "step": 1700}, {"loss": 1.0206, "grad_norm": 0.46673697233200073, "learning_rate": 0.0002, "epoch": 1.2541254125412542, "step": 1710}, {"loss": 1.0827, "grad_norm": 0.38349825143814087, "learning_rate": 0.0002, "epoch": 1.2614594792812615, "step": 1720}, {"loss": 1.0356, "grad_norm": 0.4049997627735138, "learning_rate": 0.0002, "epoch": 1.2687935460212687, "step": 1730}, {"loss": 0.9504, "grad_norm": 0.3417615294456482, "learning_rate": 0.0002, "epoch": 1.2761276127612762, "step": 1740}, {"loss": 1.094, "grad_norm": 0.4277614951133728, "learning_rate": 0.0002, "epoch": 1.2834616795012834, "step": 1750}, {"loss": 0.9938, "grad_norm": 0.5864202976226807, "learning_rate": 0.0002, "epoch": 1.2907957462412907, "step": 1760}, {"loss": 1.1167, "grad_norm": 0.7097493410110474, "learning_rate": 0.0002, "epoch": 1.2981298129812981, "step": 1770}, {"loss": 1.1132, "grad_norm": 0.3145381212234497, "learning_rate": 0.0002, "epoch": 1.3054638797213054, "step": 1780}, {"loss": 1.1099, "grad_norm": 0.5116165280342102, "learning_rate": 0.0002, "epoch": 1.3127979464613129, "step": 1790}, {"loss": 1.0765, "grad_norm": 0.7469736337661743, "learning_rate": 0.0002, "epoch": 1.3201320132013201, "step": 1800}, {"loss": 1.0663, "grad_norm": 0.32272255420684814, "learning_rate": 0.0002, "epoch": 1.3274660799413276, "step": 1810}, {"loss": 0.9887, "grad_norm": 0.3534623086452484, "learning_rate": 0.0002, "epoch": 1.3348001466813348, "step": 1820}, {"loss": 1.1628, "grad_norm": 0.36127907037734985, "learning_rate": 0.0002, "epoch": 1.342134213421342, "step": 1830}, {"loss": 1.0972, "grad_norm": 0.4072401523590088, "learning_rate": 0.0002, "epoch": 1.3494682801613496, "step": 1840}, {"loss": 1.1267, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.3568023469013568, "step": 1850}, {"loss": 1.0173, "grad_norm": 0.412883460521698, "learning_rate": 0.0002, "epoch": 1.364136413641364, "step": 1860}, {"loss": 1.0265, "grad_norm": 0.3735875189304352, "learning_rate": 0.0002, "epoch": 1.3714704803813715, "step": 1870}, {"loss": 1.1061, "grad_norm": 0.39158159494400024, "learning_rate": 0.0002, "epoch": 1.3788045471213788, "step": 1880}, {"loss": 1.0433, "grad_norm": 0.44431769847869873, "learning_rate": 0.0002, "epoch": 1.386138613861386, "step": 1890}, {"loss": 1.0216, "grad_norm": 0.37772801518440247, "learning_rate": 0.0002, "epoch": 1.3934726806013935, "step": 1900}, {"loss": 1.0674, "grad_norm": 0.4056641757488251, "learning_rate": 0.0002, "epoch": 1.4008067473414008, "step": 1910}, {"loss": 1.0256, "grad_norm": 0.41612377762794495, "learning_rate": 0.0002, "epoch": 1.408140814081408, "step": 1920}, {"loss": 1.0467, "grad_norm": 0.41153013706207275, "learning_rate": 0.0002, "epoch": 1.4154748808214155, "step": 1930}, {"loss": 1.1062, "grad_norm": 0.387845516204834, "learning_rate": 0.0002, "epoch": 1.4228089475614227, "step": 1940}, {"loss": 1.1094, "grad_norm": 0.3809587061405182, "learning_rate": 0.0002, "epoch": 1.4301430143014302, "step": 1950}, {"loss": 1.0461, "grad_norm": 0.3625726103782654, "learning_rate": 0.0002, "epoch": 1.4374770810414375, "step": 1960}, {"loss": 0.9983, "grad_norm": 0.5294290781021118, "learning_rate": 0.0002, "epoch": 1.444811147781445, "step": 1970}, {"loss": 1.1114, "grad_norm": 0.39975494146347046, "learning_rate": 0.0002, "epoch": 1.4521452145214522, "step": 1980}, {"loss": 0.9704, "grad_norm": 0.4181167185306549, "learning_rate": 0.0002, "epoch": 1.4594792812614594, "step": 1990}, {"loss": 1.1146, "grad_norm": 0.42001503705978394, "learning_rate": 0.0002, "epoch": 1.466813348001467, "step": 2000}, {"loss": 1.1266, "grad_norm": 0.4877578616142273, "learning_rate": 0.0002, "epoch": 1.4741474147414741, "step": 2010}, {"loss": 1.1012, "grad_norm": 0.4050969183444977, "learning_rate": 0.0002, "epoch": 1.4814814814814814, "step": 2020}, {"loss": 1.0562, "grad_norm": 0.39068883657455444, "learning_rate": 0.0002, "epoch": 1.4888155482214889, "step": 2030}, {"loss": 1.0464, "grad_norm": 0.421282559633255, "learning_rate": 0.0002, "epoch": 1.4961496149614961, "step": 2040}, {"loss": 1.0532, "grad_norm": 0.47092297673225403, "learning_rate": 0.0002, "epoch": 1.5034836817015034, "step": 2050}, {"loss": 0.9348, "grad_norm": 0.39688974618911743, "learning_rate": 0.0002, "epoch": 1.5108177484415108, "step": 2060}, {"loss": 1.08, "grad_norm": 0.5529879331588745, "learning_rate": 0.0002, "epoch": 1.5181518151815183, "step": 2070}, {"loss": 1.1836, "grad_norm": 0.4879782199859619, "learning_rate": 0.0002, "epoch": 1.5254858819215253, "step": 2080}, {"loss": 1.0432, "grad_norm": 0.5517361164093018, "learning_rate": 0.0002, "epoch": 1.5328199486615328, "step": 2090}, {"loss": 1.0433, "grad_norm": 0.44015637040138245, "learning_rate": 0.0002, "epoch": 1.5401540154015403, "step": 2100}, {"loss": 1.1873, "grad_norm": 0.5435167551040649, "learning_rate": 0.0002, "epoch": 1.5474880821415475, "step": 2110}, {"loss": 1.1076, "grad_norm": 0.5714033246040344, "learning_rate": 0.0002, "epoch": 1.5548221488815548, "step": 2120}, {"loss": 1.1107, "grad_norm": 0.31732529401779175, "learning_rate": 0.0002, "epoch": 1.5621562156215623, "step": 2130}, {"loss": 1.0817, "grad_norm": 0.49068278074264526, "learning_rate": 0.0002, "epoch": 1.5694902823615695, "step": 2140}, {"loss": 1.0254, "grad_norm": 0.46851542592048645, "learning_rate": 0.0002, "epoch": 1.5768243491015768, "step": 2150}, {"loss": 1.0623, "grad_norm": 0.5083092451095581, "learning_rate": 0.0002, "epoch": 1.5841584158415842, "step": 2160}, {"loss": 1.0603, "grad_norm": 0.9822936058044434, "learning_rate": 0.0002, "epoch": 1.5914924825815915, "step": 2170}, {"loss": 0.9986, "grad_norm": 0.4575989246368408, "learning_rate": 0.0002, "epoch": 1.5988265493215987, "step": 2180}, {"loss": 1.1292, "grad_norm": 0.47444286942481995, "learning_rate": 0.0002, "epoch": 1.6061606160616062, "step": 2190}, {"loss": 1.0136, "grad_norm": 0.7208226919174194, "learning_rate": 0.0002, "epoch": 1.6134946828016135, "step": 2200}, {"loss": 1.15, "grad_norm": 0.43791481852531433, "learning_rate": 0.0002, "epoch": 1.6208287495416207, "step": 2210}, {"loss": 1.0961, "grad_norm": 0.5245792865753174, "learning_rate": 0.0002, "epoch": 1.6281628162816282, "step": 2220}, {"loss": 0.9957, "grad_norm": 0.39289429783821106, "learning_rate": 0.0002, "epoch": 1.6354968830216357, "step": 2230}, {"loss": 1.133, "grad_norm": 0.6106135845184326, "learning_rate": 0.0002, "epoch": 1.6428309497616427, "step": 2240}, {"loss": 1.0129, "grad_norm": 0.3722580671310425, "learning_rate": 0.0002, "epoch": 1.6501650165016502, "step": 2250}, {"loss": 1.0446, "grad_norm": 0.3649403750896454, "learning_rate": 0.0002, "epoch": 1.6574990832416576, "step": 2260}, {"loss": 1.0037, "grad_norm": 0.46514248847961426, "learning_rate": 0.0002, "epoch": 1.6648331499816649, "step": 2270}, {"loss": 1.0022, "grad_norm": 0.42034927010536194, "learning_rate": 0.0002, "epoch": 1.6721672167216721, "step": 2280}, {"loss": 1.1362, "grad_norm": 0.45202910900115967, "learning_rate": 0.0002, "epoch": 1.6795012834616796, "step": 2290}, {"loss": 1.0866, "grad_norm": 0.36257603764533997, "learning_rate": 0.0002, "epoch": 1.6868353502016868, "step": 2300}, {"loss": 1.0973, "grad_norm": 0.6340323090553284, "learning_rate": 0.0002, "epoch": 1.694169416941694, "step": 2310}, {"loss": 1.0615, "grad_norm": 0.4352878928184509, "learning_rate": 0.0002, "epoch": 1.7015034836817016, "step": 2320}, {"loss": 1.0629, "grad_norm": 0.45029792189598083, "learning_rate": 0.0002, "epoch": 1.7088375504217088, "step": 2330}, {"loss": 0.9621, "grad_norm": 0.3891315758228302, "learning_rate": 0.0002, "epoch": 1.716171617161716, "step": 2340}, {"loss": 0.9779, "grad_norm": 0.35180050134658813, "learning_rate": 0.0002, "epoch": 1.7235056839017235, "step": 2350}, {"loss": 1.0368, "grad_norm": 0.42367449402809143, "learning_rate": 0.0002, "epoch": 1.7308397506417308, "step": 2360}, {"loss": 1.0376, "grad_norm": 0.4553675353527069, "learning_rate": 0.0002, "epoch": 1.738173817381738, "step": 2370}, {"loss": 1.1467, "grad_norm": 0.5944654941558838, "learning_rate": 0.0002, "epoch": 1.7455078841217455, "step": 2380}, {"loss": 1.0548, "grad_norm": 0.3479664623737335, "learning_rate": 0.0002, "epoch": 1.752841950861753, "step": 2390}, {"loss": 1.0798, "grad_norm": 0.3585502505302429, "learning_rate": 0.0002, "epoch": 1.76017601760176, "step": 2400}, {"loss": 1.0983, "grad_norm": 0.4263346493244171, "learning_rate": 0.0002, "epoch": 1.7675100843417675, "step": 2410}, {"loss": 1.054, "grad_norm": 0.5476409196853638, "learning_rate": 0.0002, "epoch": 1.774844151081775, "step": 2420}, {"loss": 1.1615, "grad_norm": 0.3694186508655548, "learning_rate": 0.0002, "epoch": 1.7821782178217822, "step": 2430}, {"loss": 1.1343, "grad_norm": 0.9185658693313599, "learning_rate": 0.0002, "epoch": 1.7895122845617895, "step": 2440}, {"loss": 1.0764, "grad_norm": 0.7171908020973206, "learning_rate": 0.0002, "epoch": 1.796846351301797, "step": 2450}, {"loss": 1.1154, "grad_norm": 0.550658643245697, "learning_rate": 0.0002, "epoch": 1.8041804180418042, "step": 2460}, {"loss": 0.9975, "grad_norm": 0.4075568914413452, "learning_rate": 0.0002, "epoch": 1.8115144847818114, "step": 2470}, {"loss": 1.0935, "grad_norm": 0.3790127635002136, "learning_rate": 0.0002, "epoch": 1.818848551521819, "step": 2480}, {"loss": 0.9839, "grad_norm": 0.3576384484767914, "learning_rate": 0.0002, "epoch": 1.8261826182618262, "step": 2490}, {"loss": 1.1369, "grad_norm": 0.3919370770454407, "learning_rate": 0.0002, "epoch": 1.8335166850018334, "step": 2500}, {"loss": 0.9985, "grad_norm": 0.485083669424057, "learning_rate": 0.0002, "epoch": 1.8408507517418409, "step": 2510}, {"loss": 1.1585, "grad_norm": 0.4564347565174103, "learning_rate": 0.0002, "epoch": 1.8481848184818483, "step": 2520}, {"loss": 1.0944, "grad_norm": 0.3613106608390808, "learning_rate": 0.0002, "epoch": 1.8555188852218554, "step": 2530}, {"loss": 1.0819, "grad_norm": 0.39600759744644165, "learning_rate": 0.0002, "epoch": 1.8628529519618628, "step": 2540}, {"loss": 0.9453, "grad_norm": 1.123499870300293, "learning_rate": 0.0002, "epoch": 1.8701870187018703, "step": 2550}, {"loss": 1.0635, "grad_norm": 0.4612680673599243, "learning_rate": 0.0002, "epoch": 1.8775210854418776, "step": 2560}, {"loss": 1.0087, "grad_norm": 0.42745399475097656, "learning_rate": 0.0002, "epoch": 1.8848551521818848, "step": 2570}, {"loss": 1.0102, "grad_norm": 0.4055580198764801, "learning_rate": 0.0002, "epoch": 1.8921892189218923, "step": 2580}, {"loss": 1.0177, "grad_norm": 0.44174644351005554, "learning_rate": 0.0002, "epoch": 1.8995232856618995, "step": 2590}, {"loss": 0.9886, "grad_norm": 1.0228385925292969, "learning_rate": 0.0002, "epoch": 1.9068573524019068, "step": 2600}, {"loss": 1.0857, "grad_norm": 0.3496396243572235, "learning_rate": 0.0002, "epoch": 1.9141914191419143, "step": 2610}, {"loss": 1.0955, "grad_norm": 0.4191173017024994, "learning_rate": 0.0002, "epoch": 1.9215254858819215, "step": 2620}, {"loss": 1.0943, "grad_norm": 0.6778554916381836, "learning_rate": 0.0002, "epoch": 1.9288595526219288, "step": 2630}, {"loss": 1.0594, "grad_norm": 0.41992834210395813, "learning_rate": 0.0002, "epoch": 1.9361936193619362, "step": 2640}, {"loss": 1.1159, "grad_norm": 0.8760401010513306, "learning_rate": 0.0002, "epoch": 1.9435276861019435, "step": 2650}, {"loss": 1.0379, "grad_norm": 0.44049209356307983, "learning_rate": 0.0002, "epoch": 1.9508617528419507, "step": 2660}, {"loss": 1.1008, "grad_norm": 0.5651928782463074, "learning_rate": 0.0002, "epoch": 1.9581958195819582, "step": 2670}, {"loss": 1.1317, "grad_norm": 0.5292727947235107, "learning_rate": 0.0002, "epoch": 1.9655298863219657, "step": 2680}, {"loss": 1.1328, "grad_norm": 0.6012240648269653, "learning_rate": 0.0002, "epoch": 1.9728639530619727, "step": 2690}, {"loss": 1.0683, "grad_norm": 0.3945149779319763, "learning_rate": 0.0002, "epoch": 1.9801980198019802, "step": 2700}, {"loss": 1.0155, "grad_norm": 0.5732627511024475, "learning_rate": 0.0002, "epoch": 1.9875320865419877, "step": 2710}, {"loss": 0.9857, "grad_norm": 0.3963361084461212, "learning_rate": 0.0002, "epoch": 1.994866153281995, "step": 2720}, {"eval_loss": 1.1534006595611572, "eval_runtime": 32.7541, "eval_samples_per_second": 13.159, "eval_steps_per_second": 1.649, "epoch": 2.0, "step": 2727}, {"loss": 0.9624, "grad_norm": 0.48628315329551697, "learning_rate": 0.0002, "epoch": 2.002200220022002, "step": 2730}, {"loss": 0.9603, "grad_norm": 0.413875013589859, "learning_rate": 0.0002, "epoch": 2.0095342867620096, "step": 2740}, {"loss": 0.965, "grad_norm": 0.4988735616207123, "learning_rate": 0.0002, "epoch": 2.0168683535020167, "step": 2750}, {"loss": 0.9677, "grad_norm": 0.5634812712669373, "learning_rate": 0.0002, "epoch": 2.024202420242024, "step": 2760}, {"loss": 0.9547, "grad_norm": 0.48302653431892395, "learning_rate": 0.0002, "epoch": 2.0315364869820316, "step": 2770}, {"loss": 0.9346, "grad_norm": 0.49914175271987915, "learning_rate": 0.0002, "epoch": 2.038870553722039, "step": 2780}, {"loss": 0.904, "grad_norm": 1.14039945602417, "learning_rate": 0.0002, "epoch": 2.046204620462046, "step": 2790}, {"loss": 0.9588, "grad_norm": 0.6359720826148987, "learning_rate": 0.0002, "epoch": 2.0535386872020536, "step": 2800}, {"loss": 0.9031, "grad_norm": 0.4589158296585083, "learning_rate": 0.0002, "epoch": 2.060872753942061, "step": 2810}, {"loss": 0.9438, "grad_norm": 0.46255481243133545, "learning_rate": 0.0002, "epoch": 2.068206820682068, "step": 2820}, {"loss": 0.9464, "grad_norm": 0.6232137680053711, "learning_rate": 0.0002, "epoch": 2.0755408874220755, "step": 2830}, {"loss": 0.8978, "grad_norm": 0.41042178869247437, "learning_rate": 0.0002, "epoch": 2.082874954162083, "step": 2840}, {"loss": 0.8516, "grad_norm": 0.5334428548812866, "learning_rate": 0.0002, "epoch": 2.09020902090209, "step": 2850}, {"loss": 0.9313, "grad_norm": 0.8270058631896973, "learning_rate": 0.0002, "epoch": 2.0975430876420975, "step": 2860}, {"loss": 1.0064, "grad_norm": 0.6624533534049988, "learning_rate": 0.0002, "epoch": 2.104877154382105, "step": 2870}, {"loss": 0.9196, "grad_norm": 0.5448863506317139, "learning_rate": 0.0002, "epoch": 2.112211221122112, "step": 2880}, {"loss": 0.887, "grad_norm": 0.621482789516449, "learning_rate": 0.0002, "epoch": 2.1195452878621195, "step": 2890}, {"loss": 0.9702, "grad_norm": 0.4556255340576172, "learning_rate": 0.0002, "epoch": 2.126879354602127, "step": 2900}, {"loss": 0.9323, "grad_norm": 0.4620579183101654, "learning_rate": 0.0002, "epoch": 2.1342134213421344, "step": 2910}, {"loss": 0.836, "grad_norm": 0.9602415561676025, "learning_rate": 0.0002, "epoch": 2.1415474880821415, "step": 2920}, {"loss": 0.8826, "grad_norm": 0.587943971157074, "learning_rate": 0.0002, "epoch": 2.148881554822149, "step": 2930}, {"loss": 0.971, "grad_norm": 0.5121372938156128, "learning_rate": 0.0002, "epoch": 2.1562156215621564, "step": 2940}, {"loss": 0.8751, "grad_norm": 0.49424484372138977, "learning_rate": 0.0002, "epoch": 2.1635496883021634, "step": 2950}, {"loss": 0.8674, "grad_norm": 0.6312560439109802, "learning_rate": 0.0002, "epoch": 2.170883755042171, "step": 2960}, {"loss": 0.9791, "grad_norm": 0.5235576629638672, "learning_rate": 0.0002, "epoch": 2.1782178217821784, "step": 2970}, {"loss": 0.9706, "grad_norm": 0.5868439674377441, "learning_rate": 0.0002, "epoch": 2.1855518885221854, "step": 2980}, {"loss": 0.9338, "grad_norm": 0.42302873730659485, "learning_rate": 0.0002, "epoch": 2.192885955262193, "step": 2990}, {"loss": 0.9332, "grad_norm": 0.5097725987434387, "learning_rate": 0.0002, "epoch": 2.2002200220022003, "step": 3000}, {"loss": 0.9239, "grad_norm": 0.5091572403907776, "learning_rate": 0.0002, "epoch": 2.2075540887422074, "step": 3010}, {"loss": 0.8898, "grad_norm": 0.49433162808418274, "learning_rate": 0.0002, "epoch": 2.214888155482215, "step": 3020}, {"loss": 0.9734, "grad_norm": 0.5577368140220642, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3030}, {"loss": 0.9033, "grad_norm": 0.6177583932876587, "learning_rate": 0.0002, "epoch": 2.2295562889622293, "step": 3040}, {"loss": 0.9882, "grad_norm": 0.5256719589233398, "learning_rate": 0.0002, "epoch": 2.236890355702237, "step": 3050}, {"loss": 0.9439, "grad_norm": 0.5001118183135986, "learning_rate": 0.0002, "epoch": 2.2442244224422443, "step": 3060}, {"loss": 0.8718, "grad_norm": 0.5721249580383301, "learning_rate": 0.0002, "epoch": 2.2515584891822513, "step": 3070}, {"loss": 1.0648, "grad_norm": 0.5325384140014648, "learning_rate": 0.0002, "epoch": 2.258892555922259, "step": 3080}, {"loss": 0.9843, "grad_norm": 0.5719189047813416, "learning_rate": 0.0002, "epoch": 2.2662266226622663, "step": 3090}, {"loss": 0.8633, "grad_norm": 0.6337835788726807, "learning_rate": 0.0002, "epoch": 2.2735606894022737, "step": 3100}, {"loss": 0.9962, "grad_norm": 0.5381836891174316, "learning_rate": 0.0002, "epoch": 2.2808947561422808, "step": 3110}, {"loss": 0.8265, "grad_norm": 0.5408531427383423, "learning_rate": 0.0002, "epoch": 2.2882288228822882, "step": 3120}, {"loss": 1.0325, "grad_norm": 0.43705281615257263, "learning_rate": 0.0002, "epoch": 2.2955628896222957, "step": 3130}, {"loss": 0.9388, "grad_norm": 0.6454030275344849, "learning_rate": 0.0002, "epoch": 2.3028969563623027, "step": 3140}, {"loss": 0.954, "grad_norm": 0.686030387878418, "learning_rate": 0.0002, "epoch": 2.31023102310231, "step": 3150}, {"loss": 0.9403, "grad_norm": 0.5123633146286011, "learning_rate": 0.0002, "epoch": 2.3175650898423177, "step": 3160}, {"loss": 0.8834, "grad_norm": 0.842506468296051, "learning_rate": 0.0002, "epoch": 2.3248991565823247, "step": 3170}, {"loss": 1.0497, "grad_norm": 0.5193818807601929, "learning_rate": 0.0002, "epoch": 2.332233223322332, "step": 3180}, {"loss": 0.9473, "grad_norm": 0.5634409189224243, "learning_rate": 0.0002, "epoch": 2.3395672900623397, "step": 3190}, {"loss": 0.8499, "grad_norm": 0.6475534439086914, "learning_rate": 0.0002, "epoch": 2.3469013568023467, "step": 3200}, {"loss": 0.874, "grad_norm": 1.1503914594650269, "learning_rate": 0.0002, "epoch": 2.354235423542354, "step": 3210}, {"loss": 0.9762, "grad_norm": 0.7234905362129211, "learning_rate": 0.0002, "epoch": 2.3615694902823616, "step": 3220}, {"loss": 0.9007, "grad_norm": 0.664903461933136, "learning_rate": 0.0002, "epoch": 2.368903557022369, "step": 3230}, {"loss": 0.9987, "grad_norm": 0.5453006625175476, "learning_rate": 0.0002, "epoch": 2.376237623762376, "step": 3240}, {"loss": 0.9742, "grad_norm": 0.6256654262542725, "learning_rate": 0.0002, "epoch": 2.3835716905023836, "step": 3250}, {"loss": 0.9922, "grad_norm": 0.5166565179824829, "learning_rate": 0.0002, "epoch": 2.390905757242391, "step": 3260}, {"loss": 0.927, "grad_norm": 0.5699098110198975, "learning_rate": 0.0002, "epoch": 2.398239823982398, "step": 3270}, {"loss": 0.8878, "grad_norm": 0.4472540020942688, "learning_rate": 0.0002, "epoch": 2.4055738907224056, "step": 3280}, {"loss": 0.9439, "grad_norm": 0.6790403127670288, "learning_rate": 0.0002, "epoch": 2.412907957462413, "step": 3290}, {"loss": 0.972, "grad_norm": 0.5182185173034668, "learning_rate": 0.0002, "epoch": 2.42024202420242, "step": 3300}, {"loss": 0.9775, "grad_norm": 0.564647912979126, "learning_rate": 0.0002, "epoch": 2.4275760909424275, "step": 3310}, {"loss": 1.072, "grad_norm": 0.5625313520431519, "learning_rate": 0.0002, "epoch": 2.434910157682435, "step": 3320}, {"loss": 0.8798, "grad_norm": 0.7496559619903564, "learning_rate": 0.0002, "epoch": 2.442244224422442, "step": 3330}, {"loss": 0.868, "grad_norm": 0.4779128134250641, "learning_rate": 0.0002, "epoch": 2.4495782911624495, "step": 3340}, {"loss": 1.0316, "grad_norm": 0.578093409538269, "learning_rate": 0.0002, "epoch": 2.456912357902457, "step": 3350}, {"loss": 0.9282, "grad_norm": 0.5456080436706543, "learning_rate": 0.0002, "epoch": 2.4642464246424645, "step": 3360}, {"loss": 0.8409, "grad_norm": 0.4769273102283478, "learning_rate": 0.0002, "epoch": 2.4715804913824715, "step": 3370}, {"loss": 0.9312, "grad_norm": 0.5608189702033997, "learning_rate": 0.0002, "epoch": 2.478914558122479, "step": 3380}, {"loss": 0.9934, "grad_norm": 0.5590165853500366, "learning_rate": 0.0002, "epoch": 2.4862486248624864, "step": 3390}, {"loss": 1.025, "grad_norm": 0.801306962966919, "learning_rate": 0.0002, "epoch": 2.4935826916024935, "step": 3400}, {"loss": 0.9049, "grad_norm": 0.6045624613761902, "learning_rate": 0.0002, "epoch": 2.500916758342501, "step": 3410}, {"loss": 0.944, "grad_norm": 0.5735858082771301, "learning_rate": 0.0002, "epoch": 2.5082508250825084, "step": 3420}, {"loss": 0.9846, "grad_norm": 0.6827309131622314, "learning_rate": 0.0002, "epoch": 2.5155848918225154, "step": 3430}, {"loss": 0.9789, "grad_norm": 0.5702602863311768, "learning_rate": 0.0002, "epoch": 2.522918958562523, "step": 3440}, {"loss": 0.9127, "grad_norm": 0.6674721240997314, "learning_rate": 0.0002, "epoch": 2.5302530253025304, "step": 3450}, {"loss": 0.914, "grad_norm": 0.5635907649993896, "learning_rate": 0.0002, "epoch": 2.5375870920425374, "step": 3460}, {"loss": 0.8398, "grad_norm": 0.42737770080566406, "learning_rate": 0.0002, "epoch": 2.544921158782545, "step": 3470}, {"loss": 0.9474, "grad_norm": 0.6720691919326782, "learning_rate": 0.0002, "epoch": 2.5522552255225524, "step": 3480}, {"loss": 0.8637, "grad_norm": 0.8917084336280823, "learning_rate": 0.0002, "epoch": 2.55958929226256, "step": 3490}, {"loss": 0.9257, "grad_norm": 0.5134549140930176, "learning_rate": 0.0002, "epoch": 2.566923359002567, "step": 3500}, {"loss": 0.9362, "grad_norm": 0.4951367974281311, "learning_rate": 0.0002, "epoch": 2.5742574257425743, "step": 3510}, {"loss": 0.9184, "grad_norm": 0.9438204765319824, "learning_rate": 0.0002, "epoch": 2.5815914924825814, "step": 3520}, {"loss": 0.8939, "grad_norm": 0.6024714708328247, "learning_rate": 0.0002, "epoch": 2.588925559222589, "step": 3530}, {"loss": 0.9298, "grad_norm": 0.5248535871505737, "learning_rate": 0.0002, "epoch": 2.5962596259625963, "step": 3540}, {"loss": 0.941, "grad_norm": 0.8677568435668945, "learning_rate": 0.0002, "epoch": 2.6035936927026038, "step": 3550}, {"loss": 0.9253, "grad_norm": 0.82008296251297, "learning_rate": 0.0002, "epoch": 2.610927759442611, "step": 3560}, {"loss": 0.8429, "grad_norm": 0.4724634885787964, "learning_rate": 0.0002, "epoch": 2.6182618261826183, "step": 3570}, {"loss": 0.9058, "grad_norm": 0.5434244275093079, "learning_rate": 0.0002, "epoch": 2.6255958929226257, "step": 3580}, {"loss": 0.9379, "grad_norm": 0.4948740005493164, "learning_rate": 0.0002, "epoch": 2.6329299596626328, "step": 3590}, {"loss": 0.8718, "grad_norm": 0.42109328508377075, "learning_rate": 0.0002, "epoch": 2.6402640264026402, "step": 3600}, {"loss": 0.9809, "grad_norm": 0.7979786396026611, "learning_rate": 0.0002, "epoch": 2.6475980931426477, "step": 3610}, {"loss": 0.9229, "grad_norm": 0.6345919370651245, "learning_rate": 0.0002, "epoch": 2.654932159882655, "step": 3620}, {"loss": 0.8506, "grad_norm": 0.4971671402454376, "learning_rate": 0.0002, "epoch": 2.662266226622662, "step": 3630}, {"loss": 0.8054, "grad_norm": 0.6467748284339905, "learning_rate": 0.0002, "epoch": 2.6696002933626697, "step": 3640}, {"loss": 0.9277, "grad_norm": 0.4240160286426544, "learning_rate": 0.0002, "epoch": 2.6769343601026767, "step": 3650}, {"loss": 0.8213, "grad_norm": 0.5179754495620728, "learning_rate": 0.0002, "epoch": 2.684268426842684, "step": 3660}, {"loss": 0.9221, "grad_norm": 0.754012405872345, "learning_rate": 0.0002, "epoch": 2.6916024935826917, "step": 3670}, {"loss": 0.9194, "grad_norm": 0.5141299962997437, "learning_rate": 0.0002, "epoch": 2.698936560322699, "step": 3680}, {"loss": 0.9495, "grad_norm": 0.5737819075584412, "learning_rate": 0.0002, "epoch": 2.706270627062706, "step": 3690}, {"loss": 1.0162, "grad_norm": 0.5887577533721924, "learning_rate": 0.0002, "epoch": 2.7136046938027136, "step": 3700}, {"loss": 0.9169, "grad_norm": 0.6740471720695496, "learning_rate": 0.0002, "epoch": 2.720938760542721, "step": 3710}, {"loss": 0.9297, "grad_norm": 0.5879453420639038, "learning_rate": 0.0002, "epoch": 2.728272827282728, "step": 3720}, {"loss": 0.9358, "grad_norm": 0.4858354926109314, "learning_rate": 0.0002, "epoch": 2.7356068940227356, "step": 3730}, {"loss": 0.9308, "grad_norm": 0.5489001870155334, "learning_rate": 0.0002, "epoch": 2.742940960762743, "step": 3740}, {"loss": 0.894, "grad_norm": 0.8187092542648315, "learning_rate": 0.0002, "epoch": 2.7502750275027505, "step": 3750}, {"loss": 0.8954, "grad_norm": 0.5666626691818237, "learning_rate": 0.0002, "epoch": 2.7576090942427576, "step": 3760}, {"loss": 1.0059, "grad_norm": 0.5377066135406494, "learning_rate": 0.0002, "epoch": 2.764943160982765, "step": 3770}, {"loss": 0.9132, "grad_norm": 0.566330075263977, "learning_rate": 0.0002, "epoch": 2.772277227722772, "step": 3780}, {"loss": 0.9415, "grad_norm": 0.5522832870483398, "learning_rate": 0.0002, "epoch": 2.7796112944627795, "step": 3790}, {"loss": 0.8816, "grad_norm": 0.5668695569038391, "learning_rate": 0.0002, "epoch": 2.786945361202787, "step": 3800}, {"loss": 0.8885, "grad_norm": 0.7566602826118469, "learning_rate": 0.0002, "epoch": 2.7942794279427945, "step": 3810}, {"loss": 0.8598, "grad_norm": 0.5603684782981873, "learning_rate": 0.0002, "epoch": 2.8016134946828015, "step": 3820}, {"loss": 0.9602, "grad_norm": 0.49122217297554016, "learning_rate": 0.0002, "epoch": 2.808947561422809, "step": 3830}, {"loss": 0.9738, "grad_norm": 0.6798251867294312, "learning_rate": 0.0002, "epoch": 2.816281628162816, "step": 3840}, {"loss": 0.9533, "grad_norm": 0.6097991466522217, "learning_rate": 0.0002, "epoch": 2.8236156949028235, "step": 3850}, {"loss": 0.8672, "grad_norm": 0.6675726175308228, "learning_rate": 0.0002, "epoch": 2.830949761642831, "step": 3860}, {"loss": 0.9324, "grad_norm": 0.9223952889442444, "learning_rate": 0.0002, "epoch": 2.8382838283828384, "step": 3870}, {"loss": 0.8767, "grad_norm": 0.6020799875259399, "learning_rate": 0.0002, "epoch": 2.8456178951228455, "step": 3880}, {"loss": 0.9148, "grad_norm": 0.5206381678581238, "learning_rate": 0.0002, "epoch": 2.852951961862853, "step": 3890}, {"loss": 0.9479, "grad_norm": 0.6268777251243591, "learning_rate": 0.0002, "epoch": 2.8602860286028604, "step": 3900}, {"loss": 0.9409, "grad_norm": 1.1583497524261475, "learning_rate": 0.0002, "epoch": 2.8676200953428674, "step": 3910}, {"loss": 0.895, "grad_norm": 0.7263903021812439, "learning_rate": 0.0002, "epoch": 2.874954162082875, "step": 3920}, {"loss": 0.8786, "grad_norm": 0.5369910001754761, "learning_rate": 0.0002, "epoch": 2.8822882288228824, "step": 3930}, {"loss": 1.0015, "grad_norm": 0.7298350930213928, "learning_rate": 0.0002, "epoch": 2.88962229556289, "step": 3940}, {"loss": 0.979, "grad_norm": 0.577012836933136, "learning_rate": 0.0002, "epoch": 2.896956362302897, "step": 3950}, {"loss": 0.9716, "grad_norm": 0.5859594345092773, "learning_rate": 0.0002, "epoch": 2.9042904290429044, "step": 3960}, {"loss": 0.8772, "grad_norm": 0.47176122665405273, "learning_rate": 0.0002, "epoch": 2.9116244957829114, "step": 3970}, {"loss": 0.8997, "grad_norm": 0.9699620604515076, "learning_rate": 0.0002, "epoch": 2.918958562522919, "step": 3980}, {"loss": 0.9057, "grad_norm": 0.7908747792243958, "learning_rate": 0.0002, "epoch": 2.9262926292629263, "step": 3990}, {"loss": 0.9462, "grad_norm": 0.5777379274368286, "learning_rate": 0.0002, "epoch": 2.933626696002934, "step": 4000}, {"loss": 0.9358, "grad_norm": 0.599288284778595, "learning_rate": 0.0002, "epoch": 2.940960762742941, "step": 4010}, {"loss": 0.9812, "grad_norm": 0.5232274532318115, "learning_rate": 0.0002, "epoch": 2.9482948294829483, "step": 4020}, {"loss": 0.96, "grad_norm": 0.6395137310028076, "learning_rate": 0.0002, "epoch": 2.9556288962229558, "step": 4030}, {"loss": 0.9813, "grad_norm": 0.589260458946228, "learning_rate": 0.0002, "epoch": 2.962962962962963, "step": 4040}, {"loss": 0.9541, "grad_norm": 0.5699581503868103, "learning_rate": 0.0002, "epoch": 2.9702970297029703, "step": 4050}, {"loss": 0.9585, "grad_norm": 0.528468132019043, "learning_rate": 0.0002, "epoch": 2.9776310964429777, "step": 4060}, {"loss": 0.9164, "grad_norm": 0.4804670512676239, "learning_rate": 0.0002, "epoch": 2.984965163182985, "step": 4070}, {"loss": 0.9771, "grad_norm": 1.1918889284133911, "learning_rate": 0.0002, "epoch": 2.9922992299229922, "step": 4080}, {"loss": 0.9178, "grad_norm": 0.5479103326797485, "learning_rate": 0.0002, "epoch": 2.9996332966629997, "step": 4090}, {"eval_loss": 1.1642853021621704, "eval_runtime": 32.7511, "eval_samples_per_second": 13.16, "eval_steps_per_second": 1.649, "epoch": 2.9996332966629997, "step": 4090}, {"loss": 0.7981, "grad_norm": 0.7430027723312378, "learning_rate": 0.0002, "epoch": 3.006967363403007, "step": 4100}, {"loss": 0.7871, "grad_norm": 0.6293647289276123, "learning_rate": 0.0002, "epoch": 3.014301430143014, "step": 4110}, {"loss": 0.78, "grad_norm": 0.6191329956054688, "learning_rate": 0.0002, "epoch": 3.0216354968830217, "step": 4120}, {"loss": 0.7618, "grad_norm": 0.7959313988685608, "learning_rate": 0.0002, "epoch": 3.028969563623029, "step": 4130}, {"loss": 0.8039, "grad_norm": 0.5956351161003113, "learning_rate": 0.0002, "epoch": 3.036303630363036, "step": 4140}, {"loss": 0.7477, "grad_norm": 0.670383632183075, "learning_rate": 0.0002, "epoch": 3.0436376971030437, "step": 4150}, {"loss": 0.7984, "grad_norm": 0.6414518356323242, "learning_rate": 0.0002, "epoch": 3.050971763843051, "step": 4160}, {"loss": 0.7369, "grad_norm": 0.7928852438926697, "learning_rate": 0.0002, "epoch": 3.058305830583058, "step": 4170}, {"loss": 0.7914, "grad_norm": 0.6211121082305908, "learning_rate": 0.0002, "epoch": 3.0656398973230656, "step": 4180}, {"loss": 0.7365, "grad_norm": 0.6237057447433472, "learning_rate": 0.0002, "epoch": 3.072973964063073, "step": 4190}, {"loss": 0.702, "grad_norm": 0.6522233486175537, "learning_rate": 0.0002, "epoch": 3.08030803080308, "step": 4200}, {"loss": 0.7646, "grad_norm": 0.9396848678588867, "learning_rate": 0.0002, "epoch": 3.0876420975430876, "step": 4210}, {"loss": 0.7559, "grad_norm": 0.8003010749816895, "learning_rate": 0.0002, "epoch": 3.094976164283095, "step": 4220}, {"loss": 0.711, "grad_norm": 0.6733810305595398, "learning_rate": 0.0002, "epoch": 3.102310231023102, "step": 4230}, {"loss": 0.696, "grad_norm": 0.6365828514099121, "learning_rate": 0.0002, "epoch": 3.1096442977631096, "step": 4240}, {"loss": 0.8362, "grad_norm": 1.0805548429489136, "learning_rate": 0.0002, "epoch": 3.116978364503117, "step": 4250}, {"loss": 0.7651, "grad_norm": 0.7262141108512878, "learning_rate": 0.0002, "epoch": 3.1243124312431245, "step": 4260}, {"loss": 0.7304, "grad_norm": 0.5500539541244507, "learning_rate": 0.0002, "epoch": 3.1316464979831315, "step": 4270}, {"loss": 0.7721, "grad_norm": 0.793912947177887, "learning_rate": 0.0002, "epoch": 3.138980564723139, "step": 4280}, {"loss": 0.7708, "grad_norm": 1.2540518045425415, "learning_rate": 0.0002, "epoch": 3.1463146314631465, "step": 4290}, {"loss": 0.782, "grad_norm": 0.7020077705383301, "learning_rate": 0.0002, "epoch": 3.1536486982031535, "step": 4300}, {"loss": 0.7253, "grad_norm": 0.5111123323440552, "learning_rate": 0.0002, "epoch": 3.160982764943161, "step": 4310}, {"loss": 0.8159, "grad_norm": 0.7172090411186218, "learning_rate": 0.0002, "epoch": 3.1683168316831685, "step": 4320}, {"loss": 0.6962, "grad_norm": 0.6343168616294861, "learning_rate": 0.0002, "epoch": 3.1756508984231755, "step": 4330}, {"loss": 0.7938, "grad_norm": 0.9563672542572021, "learning_rate": 0.0002, "epoch": 3.182984965163183, "step": 4340}, {"loss": 0.7385, "grad_norm": 1.0225574970245361, "learning_rate": 0.0002, "epoch": 3.1903190319031904, "step": 4350}, {"loss": 0.8652, "grad_norm": 1.1633386611938477, "learning_rate": 0.0002, "epoch": 3.1976530986431975, "step": 4360}, {"loss": 0.7259, "grad_norm": 0.8915148973464966, "learning_rate": 0.0002, "epoch": 3.204987165383205, "step": 4370}, {"loss": 0.8061, "grad_norm": 0.9156812429428101, "learning_rate": 0.0002, "epoch": 3.2123212321232124, "step": 4380}, {"loss": 0.8189, "grad_norm": 0.6363258957862854, "learning_rate": 0.0002, "epoch": 3.21965529886322, "step": 4390}, {"loss": 0.7996, "grad_norm": 0.579099178314209, "learning_rate": 0.0002, "epoch": 3.226989365603227, "step": 4400}, {"loss": 0.8592, "grad_norm": 0.8778146505355835, "learning_rate": 0.0002, "epoch": 3.2343234323432344, "step": 4410}, {"loss": 0.8281, "grad_norm": 0.8356770873069763, "learning_rate": 0.0002, "epoch": 3.241657499083242, "step": 4420}, {"loss": 0.8484, "grad_norm": 0.702032208442688, "learning_rate": 0.0002, "epoch": 3.248991565823249, "step": 4430}, {"loss": 0.7227, "grad_norm": 0.6386539340019226, "learning_rate": 0.0002, "epoch": 3.2563256325632564, "step": 4440}, {"loss": 0.8374, "grad_norm": 0.7008408904075623, "learning_rate": 0.0002, "epoch": 3.263659699303264, "step": 4450}, {"loss": 0.7572, "grad_norm": 0.9556332230567932, "learning_rate": 0.0002, "epoch": 3.270993766043271, "step": 4460}, {"loss": 0.743, "grad_norm": 0.5667835474014282, "learning_rate": 0.0002, "epoch": 3.2783278327832783, "step": 4470}, {"loss": 0.8152, "grad_norm": 0.8239172697067261, "learning_rate": 0.0002, "epoch": 3.285661899523286, "step": 4480}, {"loss": 0.756, "grad_norm": 0.7045050859451294, "learning_rate": 0.0002, "epoch": 3.292995966263293, "step": 4490}, {"loss": 0.7655, "grad_norm": 0.7131434082984924, "learning_rate": 0.0002, "epoch": 3.3003300330033003, "step": 4500}, {"loss": 0.836, "grad_norm": 0.6924910545349121, "learning_rate": 0.0002, "epoch": 3.3076640997433078, "step": 4510}, {"loss": 0.736, "grad_norm": 0.8945356607437134, "learning_rate": 0.0002, "epoch": 3.3149981664833152, "step": 4520}, {"loss": 0.7575, "grad_norm": 0.6546903252601624, "learning_rate": 0.0002, "epoch": 3.3223322332233223, "step": 4530}, {"loss": 0.7893, "grad_norm": 0.8206679224967957, "learning_rate": 0.0002, "epoch": 3.3296662999633297, "step": 4540}, {"loss": 0.7502, "grad_norm": 0.6482203602790833, "learning_rate": 0.0002, "epoch": 3.3370003667033368, "step": 4550}, {"loss": 0.8172, "grad_norm": 0.7558760046958923, "learning_rate": 0.0002, "epoch": 3.3443344334433442, "step": 4560}, {"loss": 0.744, "grad_norm": 0.7794756889343262, "learning_rate": 0.0002, "epoch": 3.3516685001833517, "step": 4570}, {"loss": 0.7385, "grad_norm": 0.7382805943489075, "learning_rate": 0.0002, "epoch": 3.359002566923359, "step": 4580}, {"loss": 0.8511, "grad_norm": 0.5912511944770813, "learning_rate": 0.0002, "epoch": 3.366336633663366, "step": 4590}, {"loss": 0.8272, "grad_norm": 0.7444885969161987, "learning_rate": 0.0002, "epoch": 3.3736707004033737, "step": 4600}, {"loss": 0.7927, "grad_norm": 0.7354922890663147, "learning_rate": 0.0002, "epoch": 3.381004767143381, "step": 4610}, {"loss": 0.7183, "grad_norm": 0.7685934901237488, "learning_rate": 0.0002, "epoch": 3.388338833883388, "step": 4620}, {"loss": 0.7436, "grad_norm": 0.61041259765625, "learning_rate": 0.0002, "epoch": 3.3956729006233957, "step": 4630}, {"loss": 0.7661, "grad_norm": 0.6820451021194458, "learning_rate": 0.0002, "epoch": 3.403006967363403, "step": 4640}, {"loss": 0.8796, "grad_norm": 0.5819534063339233, "learning_rate": 0.0002, "epoch": 3.41034103410341, "step": 4650}, {"loss": 0.7314, "grad_norm": 0.705410897731781, "learning_rate": 0.0002, "epoch": 3.4176751008434176, "step": 4660}, {"loss": 0.7901, "grad_norm": 0.8052892088890076, "learning_rate": 0.0002, "epoch": 3.425009167583425, "step": 4670}, {"loss": 0.7298, "grad_norm": 0.7746483087539673, "learning_rate": 0.0002, "epoch": 3.432343234323432, "step": 4680}, {"loss": 0.7976, "grad_norm": 0.7713689804077148, "learning_rate": 0.0002, "epoch": 3.4396773010634396, "step": 4690}, {"loss": 0.7427, "grad_norm": 0.810371994972229, "learning_rate": 0.0002, "epoch": 3.447011367803447, "step": 4700}, {"loss": 0.7594, "grad_norm": 0.7702969312667847, "learning_rate": 0.0002, "epoch": 3.4543454345434546, "step": 4710}, {"loss": 0.7957, "grad_norm": 0.7069268822669983, "learning_rate": 0.0002, "epoch": 3.4616795012834616, "step": 4720}, {"loss": 0.8199, "grad_norm": 0.7640359401702881, "learning_rate": 0.0002, "epoch": 3.469013568023469, "step": 4730}, {"loss": 0.6875, "grad_norm": 0.8661707639694214, "learning_rate": 0.0002, "epoch": 3.4763476347634765, "step": 4740}, {"loss": 0.8528, "grad_norm": 0.9970282912254333, "learning_rate": 0.0002, "epoch": 3.4836817015034836, "step": 4750}, {"loss": 0.8462, "grad_norm": 0.5824355483055115, "learning_rate": 0.0002, "epoch": 3.491015768243491, "step": 4760}, {"loss": 0.851, "grad_norm": 1.3072649240493774, "learning_rate": 0.0002, "epoch": 3.4983498349834985, "step": 4770}, {"loss": 0.9101, "grad_norm": 0.873978316783905, "learning_rate": 0.0002, "epoch": 3.5056839017235055, "step": 4780}, {"loss": 0.7403, "grad_norm": 0.5526657104492188, "learning_rate": 0.0002, "epoch": 3.513017968463513, "step": 4790}, {"loss": 0.7921, "grad_norm": 0.790894627571106, "learning_rate": 0.0002, "epoch": 3.5203520352035205, "step": 4800}, {"loss": 0.831, "grad_norm": 0.8119630217552185, "learning_rate": 0.0002, "epoch": 3.5276861019435275, "step": 4810}, {"loss": 0.7351, "grad_norm": 0.633212149143219, "learning_rate": 0.0002, "epoch": 3.535020168683535, "step": 4820}, {"loss": 0.8505, "grad_norm": 0.703029990196228, "learning_rate": 0.0002, "epoch": 3.5423542354235424, "step": 4830}, {"loss": 0.7204, "grad_norm": 0.7603771686553955, "learning_rate": 0.0002, "epoch": 3.54968830216355, "step": 4840}, {"loss": 0.8868, "grad_norm": 0.6260480880737305, "learning_rate": 0.0002, "epoch": 3.557022368903557, "step": 4850}, {"loss": 0.8137, "grad_norm": 0.8203664422035217, "learning_rate": 0.0002, "epoch": 3.5643564356435644, "step": 4860}, {"loss": 0.8821, "grad_norm": 0.7793813347816467, "learning_rate": 0.0002, "epoch": 3.5716905023835714, "step": 4870}, {"loss": 0.8164, "grad_norm": 0.7667397260665894, "learning_rate": 0.0002, "epoch": 3.579024569123579, "step": 4880}, {"loss": 0.7597, "grad_norm": 0.8198829889297485, "learning_rate": 0.0002, "epoch": 3.5863586358635864, "step": 4890}, {"loss": 0.7027, "grad_norm": 0.7689233422279358, "learning_rate": 0.0002, "epoch": 3.593692702603594, "step": 4900}, {"loss": 0.804, "grad_norm": 0.7870983481407166, "learning_rate": 0.0002, "epoch": 3.601026769343601, "step": 4910}, {"loss": 0.8269, "grad_norm": 0.8133853077888489, "learning_rate": 0.0002, "epoch": 3.6083608360836084, "step": 4920}, {"loss": 0.8515, "grad_norm": 1.308401346206665, "learning_rate": 0.0002, "epoch": 3.615694902823616, "step": 4930}, {"loss": 0.8494, "grad_norm": 0.7131121754646301, "learning_rate": 0.0002, "epoch": 3.623028969563623, "step": 4940}, {"loss": 0.7235, "grad_norm": 0.6825910210609436, "learning_rate": 0.0002, "epoch": 3.6303630363036303, "step": 4950}, {"loss": 0.7824, "grad_norm": 0.7254678606987, "learning_rate": 0.0002, "epoch": 3.637697103043638, "step": 4960}, {"loss": 0.7983, "grad_norm": 0.8045085072517395, "learning_rate": 0.0002, "epoch": 3.6450311697836453, "step": 4970}, {"loss": 0.8223, "grad_norm": 0.6991777420043945, "learning_rate": 0.0002, "epoch": 3.6523652365236523, "step": 4980}, {"loss": 0.7806, "grad_norm": 0.7804713249206543, "learning_rate": 0.0002, "epoch": 3.6596993032636598, "step": 4990}, {"loss": 0.8402, "grad_norm": 0.8525708317756653, "learning_rate": 0.0002, "epoch": 3.667033370003667, "step": 5000}, {"loss": 0.8496, "grad_norm": 0.7959994673728943, "learning_rate": 0.0002, "epoch": 3.6743674367436743, "step": 5010}, {"loss": 0.8022, "grad_norm": 0.8103628158569336, "learning_rate": 0.0002, "epoch": 3.6817015034836817, "step": 5020}, {"loss": 0.7376, "grad_norm": 0.7517836093902588, "learning_rate": 0.0002, "epoch": 3.689035570223689, "step": 5030}, {"loss": 0.8375, "grad_norm": 0.6878514289855957, "learning_rate": 0.0002, "epoch": 3.6963696369636962, "step": 5040}, {"loss": 0.7998, "grad_norm": 1.2371820211410522, "learning_rate": 0.0002, "epoch": 3.7037037037037037, "step": 5050}, {"loss": 0.6941, "grad_norm": 0.6567103862762451, "learning_rate": 0.0002, "epoch": 3.711037770443711, "step": 5060}, {"loss": 0.8465, "grad_norm": 1.1254922151565552, "learning_rate": 0.0002, "epoch": 3.718371837183718, "step": 5070}, {"loss": 0.8365, "grad_norm": 0.6796132326126099, "learning_rate": 0.0002, "epoch": 3.7257059039237257, "step": 5080}, {"loss": 0.7818, "grad_norm": 0.7285300493240356, "learning_rate": 0.0002, "epoch": 3.733039970663733, "step": 5090}, {"loss": 0.8581, "grad_norm": 0.8931500911712646, "learning_rate": 0.0002, "epoch": 3.7403740374037406, "step": 5100}, {"loss": 0.8181, "grad_norm": 0.6256856918334961, "learning_rate": 0.0002, "epoch": 3.7477081041437477, "step": 5110}, {"loss": 0.743, "grad_norm": 0.79310142993927, "learning_rate": 0.0002, "epoch": 3.755042170883755, "step": 5120}, {"loss": 0.8235, "grad_norm": 0.6594041585922241, "learning_rate": 0.0002, "epoch": 3.762376237623762, "step": 5130}, {"loss": 0.6925, "grad_norm": 0.7029327750205994, "learning_rate": 0.0002, "epoch": 3.7697103043637696, "step": 5140}, {"loss": 0.7457, "grad_norm": 0.5880070328712463, "learning_rate": 0.0002, "epoch": 3.777044371103777, "step": 5150}, {"loss": 0.8716, "grad_norm": 0.7578945159912109, "learning_rate": 0.0002, "epoch": 3.7843784378437846, "step": 5160}, {"loss": 0.8819, "grad_norm": 0.8276378512382507, "learning_rate": 0.0002, "epoch": 3.7917125045837916, "step": 5170}, {"loss": 0.7559, "grad_norm": 0.7627953886985779, "learning_rate": 0.0002, "epoch": 3.799046571323799, "step": 5180}, {"loss": 0.7665, "grad_norm": 0.8169086575508118, "learning_rate": 0.0002, "epoch": 3.806380638063806, "step": 5190}, {"loss": 0.761, "grad_norm": 0.6605030298233032, "learning_rate": 0.0002, "epoch": 3.8137147048038136, "step": 5200}, {"loss": 0.8804, "grad_norm": 0.5837286114692688, "learning_rate": 0.0002, "epoch": 3.821048771543821, "step": 5210}, {"loss": 0.8369, "grad_norm": 1.2422157526016235, "learning_rate": 0.0002, "epoch": 3.8283828382838285, "step": 5220}, {"loss": 0.8431, "grad_norm": 0.6589220762252808, "learning_rate": 0.0002, "epoch": 3.8357169050238356, "step": 5230}, {"loss": 0.7686, "grad_norm": 0.8567556142807007, "learning_rate": 0.0002, "epoch": 3.843050971763843, "step": 5240}, {"loss": 0.8652, "grad_norm": 0.6490627527236938, "learning_rate": 0.0002, "epoch": 3.8503850385038505, "step": 5250}, {"loss": 0.7386, "grad_norm": 0.620232880115509, "learning_rate": 0.0002, "epoch": 3.8577191052438575, "step": 5260}, {"loss": 0.9192, "grad_norm": 0.7685128450393677, "learning_rate": 0.0002, "epoch": 3.865053171983865, "step": 5270}, {"loss": 0.872, "grad_norm": 0.8113296627998352, "learning_rate": 0.0002, "epoch": 3.8723872387238725, "step": 5280}, {"loss": 0.7156, "grad_norm": 0.8092675805091858, "learning_rate": 0.0002, "epoch": 3.87972130546388, "step": 5290}, {"loss": 0.7325, "grad_norm": 0.583570122718811, "learning_rate": 0.0002, "epoch": 3.887055372203887, "step": 5300}, {"loss": 0.9333, "grad_norm": 1.712363600730896, "learning_rate": 0.0002, "epoch": 3.8943894389438944, "step": 5310}, {"loss": 0.7537, "grad_norm": 0.6673534512519836, "learning_rate": 0.0002, "epoch": 3.9017235056839015, "step": 5320}, {"loss": 0.7035, "grad_norm": 1.9770312309265137, "learning_rate": 0.0002, "epoch": 3.909057572423909, "step": 5330}, {"loss": 0.8793, "grad_norm": 0.6430999636650085, "learning_rate": 0.0002, "epoch": 3.9163916391639164, "step": 5340}, {"loss": 0.839, "grad_norm": 1.0159571170806885, "learning_rate": 0.0002, "epoch": 3.923725705903924, "step": 5350}, {"loss": 0.9332, "grad_norm": 0.8607584834098816, "learning_rate": 0.0002, "epoch": 3.931059772643931, "step": 5360}, {"loss": 0.7261, "grad_norm": 0.6967900991439819, "learning_rate": 0.0002, "epoch": 3.9383938393839384, "step": 5370}, {"loss": 0.8456, "grad_norm": 0.7683077454566956, "learning_rate": 0.0002, "epoch": 3.945727906123946, "step": 5380}, {"loss": 0.7682, "grad_norm": 0.6805762648582458, "learning_rate": 0.0002, "epoch": 3.953061972863953, "step": 5390}, {"loss": 0.7746, "grad_norm": 0.7033619284629822, "learning_rate": 0.0002, "epoch": 3.9603960396039604, "step": 5400}, {"loss": 0.8393, "grad_norm": 0.966112494468689, "learning_rate": 0.0002, "epoch": 3.967730106343968, "step": 5410}, {"loss": 0.8316, "grad_norm": 0.8467881083488464, "learning_rate": 0.0002, "epoch": 3.9750641730839753, "step": 5420}, {"loss": 0.8084, "grad_norm": 0.8005317449569702, "learning_rate": 0.0002, "epoch": 3.9823982398239823, "step": 5430}, {"loss": 0.7168, "grad_norm": 1.1615241765975952, "learning_rate": 0.0002, "epoch": 3.98973230656399, "step": 5440}, {"loss": 0.8263, "grad_norm": 0.6121614575386047, "learning_rate": 0.0002, "epoch": 3.997066373303997, "step": 5450}, {"eval_loss": 1.1834222078323364, "eval_runtime": 32.7569, "eval_samples_per_second": 13.158, "eval_steps_per_second": 1.649, "epoch": 4.0, "step": 5454}, {"loss": 0.7267, "grad_norm": 0.6055727005004883, "learning_rate": 0.0002, "epoch": 4.004400440044004, "step": 5460}, {"loss": 0.5766, "grad_norm": 0.8232647180557251, "learning_rate": 0.0002, "epoch": 4.011734506784012, "step": 5470}, {"loss": 0.6489, "grad_norm": 0.7739192247390747, "learning_rate": 0.0002, "epoch": 4.019068573524019, "step": 5480}, {"loss": 0.5978, "grad_norm": 0.6264950633049011, "learning_rate": 0.0002, "epoch": 4.026402640264027, "step": 5490}, {"loss": 0.6392, "grad_norm": 1.4798702001571655, "learning_rate": 0.0002, "epoch": 4.033736707004033, "step": 5500}, {"loss": 0.6143, "grad_norm": 0.9538470506668091, "learning_rate": 0.0002, "epoch": 4.041070773744041, "step": 5510}, {"loss": 0.6056, "grad_norm": 0.834561288356781, "learning_rate": 0.0002, "epoch": 4.048404840484048, "step": 5520}, {"loss": 0.6077, "grad_norm": 0.6407850384712219, "learning_rate": 0.0002, "epoch": 4.055738907224056, "step": 5530}, {"loss": 0.6733, "grad_norm": 0.9035961627960205, "learning_rate": 0.0002, "epoch": 4.063072973964063, "step": 5540}, {"loss": 0.5854, "grad_norm": 0.842812716960907, "learning_rate": 0.0002, "epoch": 4.070407040704071, "step": 5550}, {"loss": 0.654, "grad_norm": 0.8197882175445557, "learning_rate": 0.0002, "epoch": 4.077741107444078, "step": 5560}, {"loss": 0.5919, "grad_norm": 0.8652673959732056, "learning_rate": 0.0002, "epoch": 4.085075174184085, "step": 5570}, {"loss": 0.6188, "grad_norm": 0.8048318028450012, "learning_rate": 0.0002, "epoch": 4.092409240924092, "step": 5580}, {"loss": 0.6487, "grad_norm": 0.9604969024658203, "learning_rate": 0.0002, "epoch": 4.0997433076641, "step": 5590}, {"loss": 0.6356, "grad_norm": 1.244756817817688, "learning_rate": 0.0002, "epoch": 4.107077374404107, "step": 5600}, {"loss": 0.6489, "grad_norm": 0.7975269556045532, "learning_rate": 0.0002, "epoch": 4.114411441144115, "step": 5610}, {"loss": 0.6445, "grad_norm": 0.6130099296569824, "learning_rate": 0.0002, "epoch": 4.121745507884122, "step": 5620}, {"loss": 0.6024, "grad_norm": 0.7793202996253967, "learning_rate": 0.0002, "epoch": 4.129079574624129, "step": 5630}, {"loss": 0.5723, "grad_norm": 1.187238335609436, "learning_rate": 0.0002, "epoch": 4.136413641364136, "step": 5640}, {"loss": 0.6385, "grad_norm": 0.8450375199317932, "learning_rate": 0.0002, "epoch": 4.143747708104144, "step": 5650}, {"loss": 0.6866, "grad_norm": 0.9006940126419067, "learning_rate": 0.0002, "epoch": 4.151081774844151, "step": 5660}, {"loss": 0.6179, "grad_norm": 0.9447154998779297, "learning_rate": 0.0002, "epoch": 4.158415841584159, "step": 5670}, {"loss": 0.6476, "grad_norm": 0.798032283782959, "learning_rate": 0.0002, "epoch": 4.165749908324166, "step": 5680}, {"loss": 0.6666, "grad_norm": 0.65578693151474, "learning_rate": 0.0002, "epoch": 4.1730839750641735, "step": 5690}, {"loss": 0.701, "grad_norm": 1.0864700078964233, "learning_rate": 0.0002, "epoch": 4.18041804180418, "step": 5700}, {"loss": 0.6895, "grad_norm": 0.7344121932983398, "learning_rate": 0.0002, "epoch": 4.187752108544188, "step": 5710}, {"loss": 0.6659, "grad_norm": 0.9722456932067871, "learning_rate": 0.0002, "epoch": 4.195086175284195, "step": 5720}, {"loss": 0.6887, "grad_norm": 1.263814926147461, "learning_rate": 0.0002, "epoch": 4.2024202420242025, "step": 5730}, {"loss": 0.608, "grad_norm": 0.9622581005096436, "learning_rate": 0.0002, "epoch": 4.20975430876421, "step": 5740}, {"loss": 0.6221, "grad_norm": 0.8497143387794495, "learning_rate": 0.0002, "epoch": 4.2170883755042174, "step": 5750}, {"loss": 0.6322, "grad_norm": 0.8248446583747864, "learning_rate": 0.0002, "epoch": 4.224422442244224, "step": 5760}, {"loss": 0.6045, "grad_norm": 1.2544798851013184, "learning_rate": 0.0002, "epoch": 4.2317565089842315, "step": 5770}, {"loss": 0.641, "grad_norm": 0.8224676251411438, "learning_rate": 0.0002, "epoch": 4.239090575724239, "step": 5780}, {"loss": 0.6399, "grad_norm": 0.8924877047538757, "learning_rate": 0.0002, "epoch": 4.2464246424642464, "step": 5790}, {"loss": 0.6845, "grad_norm": 0.8545848727226257, "learning_rate": 0.0002, "epoch": 4.253758709204254, "step": 5800}, {"loss": 0.6669, "grad_norm": 0.8081067800521851, "learning_rate": 0.0002, "epoch": 4.261092775944261, "step": 5810}, {"loss": 0.6149, "grad_norm": 0.7111002802848816, "learning_rate": 0.0002, "epoch": 4.268426842684269, "step": 5820}, {"loss": 0.6343, "grad_norm": 0.8696979880332947, "learning_rate": 0.0002, "epoch": 4.2757609094242754, "step": 5830}, {"loss": 0.6384, "grad_norm": 0.821401834487915, "learning_rate": 0.0002, "epoch": 4.283094976164283, "step": 5840}, {"loss": 0.6912, "grad_norm": 0.888908326625824, "learning_rate": 0.0002, "epoch": 4.29042904290429, "step": 5850}, {"loss": 0.6061, "grad_norm": 1.9380123615264893, "learning_rate": 0.0002, "epoch": 4.297763109644298, "step": 5860}, {"loss": 0.6766, "grad_norm": 1.121774435043335, "learning_rate": 0.0002, "epoch": 4.305097176384305, "step": 5870}, {"loss": 0.7205, "grad_norm": 0.9238282442092896, "learning_rate": 0.0002, "epoch": 4.312431243124313, "step": 5880}, {"loss": 0.6351, "grad_norm": 0.7321620583534241, "learning_rate": 0.0002, "epoch": 4.319765309864319, "step": 5890}, {"loss": 0.6404, "grad_norm": 0.8739548325538635, "learning_rate": 0.0002, "epoch": 4.327099376604327, "step": 5900}, {"loss": 0.5892, "grad_norm": 0.9686012268066406, "learning_rate": 0.0002, "epoch": 4.334433443344334, "step": 5910}, {"loss": 0.641, "grad_norm": 0.9033839106559753, "learning_rate": 0.0002, "epoch": 4.341767510084342, "step": 5920}, {"loss": 0.6456, "grad_norm": 0.8131115436553955, "learning_rate": 0.0002, "epoch": 4.349101576824349, "step": 5930}, {"loss": 0.5826, "grad_norm": 0.8942412734031677, "learning_rate": 0.0002, "epoch": 4.356435643564357, "step": 5940}, {"loss": 0.7336, "grad_norm": 0.8439112901687622, "learning_rate": 0.0002, "epoch": 4.363769710304364, "step": 5950}, {"loss": 0.6537, "grad_norm": 0.9176713228225708, "learning_rate": 0.0002, "epoch": 4.371103777044371, "step": 5960}, {"loss": 0.6792, "grad_norm": 0.6799634695053101, "learning_rate": 0.0002, "epoch": 4.378437843784378, "step": 5970}, {"loss": 0.7266, "grad_norm": 1.0435824394226074, "learning_rate": 0.0002, "epoch": 4.385771910524386, "step": 5980}, {"loss": 0.68, "grad_norm": 0.997937798500061, "learning_rate": 0.0002, "epoch": 4.393105977264393, "step": 5990}, {"loss": 0.6604, "grad_norm": 1.0308842658996582, "learning_rate": 0.0002, "epoch": 4.400440044004401, "step": 6000}, {"loss": 0.6402, "grad_norm": 1.3683775663375854, "learning_rate": 0.0002, "epoch": 4.407774110744408, "step": 6010}, {"loss": 0.7027, "grad_norm": 0.7569534182548523, "learning_rate": 0.0002, "epoch": 4.415108177484415, "step": 6020}, {"loss": 0.5949, "grad_norm": 1.089978575706482, "learning_rate": 0.0002, "epoch": 4.422442244224422, "step": 6030}, {"loss": 0.6353, "grad_norm": 0.7522459626197815, "learning_rate": 0.0002, "epoch": 4.42977631096443, "step": 6040}, {"loss": 0.5852, "grad_norm": 0.6709823608398438, "learning_rate": 0.0002, "epoch": 4.437110377704437, "step": 6050}, {"loss": 0.6718, "grad_norm": 0.6992089748382568, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 6060}, {"loss": 0.6933, "grad_norm": 1.0182931423187256, "learning_rate": 0.0002, "epoch": 4.451778511184452, "step": 6070}, {"loss": 0.6255, "grad_norm": 1.0685160160064697, "learning_rate": 0.0002, "epoch": 4.459112577924459, "step": 6080}, {"loss": 0.6086, "grad_norm": 0.8295124769210815, "learning_rate": 0.0002, "epoch": 4.466446644664466, "step": 6090}, {"loss": 0.6359, "grad_norm": 1.1862998008728027, "learning_rate": 0.0002, "epoch": 4.473780711404474, "step": 6100}, {"loss": 0.638, "grad_norm": 0.7400273084640503, "learning_rate": 0.0002, "epoch": 4.481114778144481, "step": 6110}, {"loss": 0.6854, "grad_norm": 0.7098417282104492, "learning_rate": 0.0002, "epoch": 4.488448844884489, "step": 6120}, {"loss": 0.6976, "grad_norm": 0.9745053648948669, "learning_rate": 0.0002, "epoch": 4.495782911624496, "step": 6130}, {"loss": 0.605, "grad_norm": 0.8638797998428345, "learning_rate": 0.0002, "epoch": 4.503116978364503, "step": 6140}, {"loss": 0.6491, "grad_norm": 0.8291046619415283, "learning_rate": 0.0002, "epoch": 4.51045104510451, "step": 6150}, {"loss": 0.6457, "grad_norm": 1.0301737785339355, "learning_rate": 0.0002, "epoch": 4.517785111844518, "step": 6160}, {"loss": 0.6742, "grad_norm": 1.1996512413024902, "learning_rate": 0.0002, "epoch": 4.525119178584525, "step": 6170}, {"loss": 0.6484, "grad_norm": 1.151038408279419, "learning_rate": 0.0002, "epoch": 4.5324532453245325, "step": 6180}, {"loss": 0.668, "grad_norm": 0.8385201096534729, "learning_rate": 0.0002, "epoch": 4.53978731206454, "step": 6190}, {"loss": 0.6381, "grad_norm": 0.8969188332557678, "learning_rate": 0.0002, "epoch": 4.5471213788045475, "step": 6200}, {"loss": 0.7141, "grad_norm": 1.60659658908844, "learning_rate": 0.0002, "epoch": 4.554455445544555, "step": 6210}, {"loss": 0.6388, "grad_norm": 0.9356731176376343, "learning_rate": 0.0002, "epoch": 4.5617895122845615, "step": 6220}, {"loss": 0.7393, "grad_norm": 0.95856773853302, "learning_rate": 0.0002, "epoch": 4.569123579024569, "step": 6230}, {"loss": 0.6554, "grad_norm": 1.1162524223327637, "learning_rate": 0.0002, "epoch": 4.5764576457645765, "step": 6240}, {"loss": 0.6012, "grad_norm": 0.8809238076210022, "learning_rate": 0.0002, "epoch": 4.583791712504584, "step": 6250}, {"loss": 0.648, "grad_norm": 0.890738844871521, "learning_rate": 0.0002, "epoch": 4.591125779244591, "step": 6260}, {"loss": 0.6663, "grad_norm": 0.918684720993042, "learning_rate": 0.0002, "epoch": 4.598459845984598, "step": 6270}, {"loss": 0.5992, "grad_norm": 0.8156296610832214, "learning_rate": 0.0002, "epoch": 4.6057939127246055, "step": 6280}, {"loss": 0.723, "grad_norm": 1.046634316444397, "learning_rate": 0.0002, "epoch": 4.613127979464613, "step": 6290}, {"loss": 0.7023, "grad_norm": 0.7725525498390198, "learning_rate": 0.0002, "epoch": 4.62046204620462, "step": 6300}, {"loss": 0.6414, "grad_norm": 0.9992046356201172, "learning_rate": 0.0002, "epoch": 4.627796112944628, "step": 6310}, {"loss": 0.6201, "grad_norm": 0.8480095267295837, "learning_rate": 0.0002, "epoch": 4.635130179684635, "step": 6320}, {"loss": 0.6869, "grad_norm": 0.7061955332756042, "learning_rate": 0.0002, "epoch": 4.642464246424643, "step": 6330}, {"loss": 0.6828, "grad_norm": 1.0354212522506714, "learning_rate": 0.0002, "epoch": 4.649798313164649, "step": 6340}, {"loss": 0.6651, "grad_norm": 1.0081377029418945, "learning_rate": 0.0002, "epoch": 4.657132379904657, "step": 6350}, {"loss": 0.726, "grad_norm": 1.2904249429702759, "learning_rate": 0.0002, "epoch": 4.664466446644664, "step": 6360}, {"loss": 0.7148, "grad_norm": 0.9248910546302795, "learning_rate": 0.0002, "epoch": 4.671800513384672, "step": 6370}, {"loss": 0.6961, "grad_norm": 0.9907804131507874, "learning_rate": 0.0002, "epoch": 4.679134580124679, "step": 6380}, {"loss": 0.6163, "grad_norm": 1.201143741607666, "learning_rate": 0.0002, "epoch": 4.686468646864687, "step": 6390}, {"loss": 0.6762, "grad_norm": 0.8709394335746765, "learning_rate": 0.0002, "epoch": 4.693802713604693, "step": 6400}, {"loss": 0.7217, "grad_norm": 0.7468608021736145, "learning_rate": 0.0002, "epoch": 4.701136780344701, "step": 6410}, {"loss": 0.6548, "grad_norm": 0.8607903718948364, "learning_rate": 0.0002, "epoch": 4.708470847084708, "step": 6420}, {"loss": 0.6449, "grad_norm": 0.9840512871742249, "learning_rate": 0.0002, "epoch": 4.715804913824716, "step": 6430}, {"loss": 0.685, "grad_norm": 0.8328204154968262, "learning_rate": 0.0002, "epoch": 4.723138980564723, "step": 6440}, {"loss": 0.697, "grad_norm": 0.924505352973938, "learning_rate": 0.0002, "epoch": 4.730473047304731, "step": 6450}, {"loss": 0.7422, "grad_norm": 0.8897685408592224, "learning_rate": 0.0002, "epoch": 4.737807114044738, "step": 6460}, {"loss": 0.6842, "grad_norm": 0.9605024456977844, "learning_rate": 0.0002, "epoch": 4.745141180784745, "step": 6470}, {"loss": 0.6488, "grad_norm": 0.8150759935379028, "learning_rate": 0.0002, "epoch": 4.752475247524752, "step": 6480}, {"loss": 0.6606, "grad_norm": 0.8128412961959839, "learning_rate": 0.0002, "epoch": 4.75980931426476, "step": 6490}, {"loss": 0.6729, "grad_norm": 0.7381404638290405, "learning_rate": 0.0002, "epoch": 4.767143381004767, "step": 6500}, {"loss": 0.6713, "grad_norm": 1.0565853118896484, "learning_rate": 0.0002, "epoch": 4.774477447744775, "step": 6510}, {"loss": 0.6496, "grad_norm": 0.9298134446144104, "learning_rate": 0.0002, "epoch": 4.781811514484782, "step": 6520}, {"loss": 0.7279, "grad_norm": 1.0145525932312012, "learning_rate": 0.0002, "epoch": 4.789145581224789, "step": 6530}, {"loss": 0.5986, "grad_norm": 0.92259681224823, "learning_rate": 0.0002, "epoch": 4.796479647964796, "step": 6540}, {"loss": 0.63, "grad_norm": 0.7881024479866028, "learning_rate": 0.0002, "epoch": 4.803813714704804, "step": 6550}, {"loss": 0.7134, "grad_norm": 1.4935206174850464, "learning_rate": 0.0002, "epoch": 4.811147781444811, "step": 6560}, {"loss": 0.6695, "grad_norm": 0.8612369298934937, "learning_rate": 0.0002, "epoch": 4.818481848184819, "step": 6570}, {"loss": 0.779, "grad_norm": 1.0118653774261475, "learning_rate": 0.0002, "epoch": 4.825815914924826, "step": 6580}, {"loss": 0.6991, "grad_norm": 1.1303809881210327, "learning_rate": 0.0002, "epoch": 4.833149981664834, "step": 6590}, {"loss": 0.7887, "grad_norm": 0.9112492203712463, "learning_rate": 0.0002, "epoch": 4.84048404840484, "step": 6600}, {"loss": 0.7699, "grad_norm": 0.864762544631958, "learning_rate": 0.0002, "epoch": 4.847818115144848, "step": 6610}, {"loss": 0.7347, "grad_norm": 0.9090572595596313, "learning_rate": 0.0002, "epoch": 4.855152181884855, "step": 6620}, {"loss": 0.6608, "grad_norm": 1.014953374862671, "learning_rate": 0.0002, "epoch": 4.862486248624863, "step": 6630}, {"loss": 0.6429, "grad_norm": 1.0702149868011475, "learning_rate": 0.0002, "epoch": 4.86982031536487, "step": 6640}, {"loss": 0.6943, "grad_norm": 1.002135157585144, "learning_rate": 0.0002, "epoch": 4.8771543821048775, "step": 6650}, {"loss": 0.7225, "grad_norm": 0.862545907497406, "learning_rate": 0.0002, "epoch": 4.884488448844884, "step": 6660}, {"loss": 0.6206, "grad_norm": 0.7302131056785583, "learning_rate": 0.0002, "epoch": 4.891822515584892, "step": 6670}, {"loss": 0.7175, "grad_norm": 0.8380730152130127, "learning_rate": 0.0002, "epoch": 4.899156582324899, "step": 6680}, {"loss": 0.645, "grad_norm": 0.7956018447875977, "learning_rate": 0.0002, "epoch": 4.9064906490649065, "step": 6690}, {"loss": 0.6431, "grad_norm": 0.6717583537101746, "learning_rate": 0.0002, "epoch": 4.913824715804914, "step": 6700}, {"loss": 0.6942, "grad_norm": 1.09099280834198, "learning_rate": 0.0002, "epoch": 4.9211587825449215, "step": 6710}, {"loss": 0.7533, "grad_norm": 0.8589889407157898, "learning_rate": 0.0002, "epoch": 4.928492849284929, "step": 6720}, {"loss": 0.66, "grad_norm": 1.0046314001083374, "learning_rate": 0.0002, "epoch": 4.9358269160249355, "step": 6730}, {"loss": 0.6864, "grad_norm": 0.8559659123420715, "learning_rate": 0.0002, "epoch": 4.943160982764943, "step": 6740}, {"loss": 0.6847, "grad_norm": 0.8588525652885437, "learning_rate": 0.0002, "epoch": 4.9504950495049505, "step": 6750}, {"loss": 0.6428, "grad_norm": 0.9192708134651184, "learning_rate": 0.0002, "epoch": 4.957829116244958, "step": 6760}, {"loss": 0.6873, "grad_norm": 1.051398754119873, "learning_rate": 0.0002, "epoch": 4.965163182984965, "step": 6770}, {"loss": 0.7249, "grad_norm": 0.9111362099647522, "learning_rate": 0.0002, "epoch": 4.972497249724973, "step": 6780}, {"loss": 0.7613, "grad_norm": 0.7305638194084167, "learning_rate": 0.0002, "epoch": 4.9798313164649795, "step": 6790}, {"loss": 0.6747, "grad_norm": 1.118837594985962, "learning_rate": 0.0002, "epoch": 4.987165383204987, "step": 6800}, {"loss": 0.6412, "grad_norm": 0.9075239300727844, "learning_rate": 0.0002, "epoch": 4.994499449944994, "step": 6810}]}
+{"epoch": 6.0, "step": 8181, "epoch_duration": 1471.2244851589203, "total_accumulated_duration": 8848.251735210419, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7568.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 1.9722, "grad_norm": 0.47521963715553284, "learning_rate": 0.0002, "epoch": 0.007334066740007334, "step": 10}, {"loss": 1.4821, "grad_norm": 0.5395162105560303, "learning_rate": 0.0002, "epoch": 0.014668133480014669, "step": 20}, {"loss": 1.4202, "grad_norm": 0.4305780231952667, "learning_rate": 0.0002, "epoch": 0.022002200220022004, "step": 30}, {"loss": 1.4271, "grad_norm": 0.6938246488571167, "learning_rate": 0.0002, "epoch": 0.029336266960029337, "step": 40}, {"loss": 1.3112, "grad_norm": 1.5133819580078125, "learning_rate": 0.0002, "epoch": 0.03667033370003667, "step": 50}, {"loss": 1.3132, "grad_norm": 0.9173883199691772, "learning_rate": 0.0002, "epoch": 0.04400440044004401, "step": 60}, {"loss": 1.2844, "grad_norm": 0.4619861841201782, "learning_rate": 0.0002, "epoch": 0.05133846718005134, "step": 70}, {"loss": 1.2108, "grad_norm": 0.46118637919425964, "learning_rate": 0.0002, "epoch": 0.058672533920058674, "step": 80}, {"loss": 1.3441, "grad_norm": 0.4468648135662079, "learning_rate": 0.0002, "epoch": 0.066006600660066, "step": 90}, {"loss": 1.1863, "grad_norm": 0.46123769879341125, "learning_rate": 0.0002, "epoch": 0.07334066740007333, "step": 100}, {"loss": 1.2772, "grad_norm": 0.4859139025211334, "learning_rate": 0.0002, "epoch": 0.08067473414008068, "step": 110}, {"loss": 1.2087, "grad_norm": 0.4384922385215759, "learning_rate": 0.0002, "epoch": 0.08800880088008801, "step": 120}, {"loss": 1.2927, "grad_norm": 0.39519360661506653, "learning_rate": 0.0002, "epoch": 0.09534286762009535, "step": 130}, {"loss": 1.2349, "grad_norm": 0.4049859344959259, "learning_rate": 0.0002, "epoch": 0.10267693436010268, "step": 140}, {"loss": 1.293, "grad_norm": 0.4605638086795807, "learning_rate": 0.0002, "epoch": 0.11001100110011001, "step": 150}, {"loss": 1.2659, "grad_norm": 0.4201928377151489, "learning_rate": 0.0002, "epoch": 0.11734506784011735, "step": 160}, {"loss": 1.3961, "grad_norm": 0.5367777347564697, "learning_rate": 0.0002, "epoch": 0.12467913458012468, "step": 170}, {"loss": 1.2481, "grad_norm": 0.41752299666404724, "learning_rate": 0.0002, "epoch": 0.132013201320132, "step": 180}, {"loss": 1.207, "grad_norm": 0.31597763299942017, "learning_rate": 0.0002, "epoch": 0.13934726806013933, "step": 190}, {"loss": 1.2441, "grad_norm": 0.7468788623809814, "learning_rate": 0.0002, "epoch": 0.14668133480014667, "step": 200}, {"loss": 1.199, "grad_norm": 0.3403034508228302, "learning_rate": 0.0002, "epoch": 0.15401540154015403, "step": 210}, {"loss": 1.2439, "grad_norm": 0.34240293502807617, "learning_rate": 0.0002, "epoch": 0.16134946828016136, "step": 220}, {"loss": 1.2022, "grad_norm": 0.356158971786499, "learning_rate": 0.0002, "epoch": 0.1686835350201687, "step": 230}, {"loss": 1.207, "grad_norm": 0.3448857367038727, "learning_rate": 0.0002, "epoch": 0.17601760176017603, "step": 240}, {"loss": 1.2156, "grad_norm": 0.3475699722766876, "learning_rate": 0.0002, "epoch": 0.18335166850018336, "step": 250}, {"loss": 1.1551, "grad_norm": 0.2770358622074127, "learning_rate": 0.0002, "epoch": 0.1906857352401907, "step": 260}, {"loss": 1.2238, "grad_norm": 0.4310270845890045, "learning_rate": 0.0002, "epoch": 0.19801980198019803, "step": 270}, {"loss": 1.2917, "grad_norm": 0.335041880607605, "learning_rate": 0.0002, "epoch": 0.20535386872020536, "step": 280}, {"loss": 1.0959, "grad_norm": 0.3420602083206177, "learning_rate": 0.0002, "epoch": 0.2126879354602127, "step": 290}, {"loss": 1.1232, "grad_norm": 0.325001060962677, "learning_rate": 0.0002, "epoch": 0.22002200220022003, "step": 300}, {"loss": 1.2007, "grad_norm": 0.3027827739715576, "learning_rate": 0.0002, "epoch": 0.22735606894022736, "step": 310}, {"loss": 1.1803, "grad_norm": 0.435550719499588, "learning_rate": 0.0002, "epoch": 0.2346901356802347, "step": 320}, {"loss": 1.2045, "grad_norm": 0.3884522616863251, "learning_rate": 0.0002, "epoch": 0.24202420242024203, "step": 330}, {"loss": 1.2481, "grad_norm": 0.7736002206802368, "learning_rate": 0.0002, "epoch": 0.24935826916024936, "step": 340}, {"loss": 1.3606, "grad_norm": 0.35052821040153503, "learning_rate": 0.0002, "epoch": 0.2566923359002567, "step": 350}, {"loss": 1.2129, "grad_norm": 0.3311890959739685, "learning_rate": 0.0002, "epoch": 0.264026402640264, "step": 360}, {"loss": 1.2219, "grad_norm": 0.7473500370979309, "learning_rate": 0.0002, "epoch": 0.27136046938027136, "step": 370}, {"loss": 1.2712, "grad_norm": 0.3681875765323639, "learning_rate": 0.0002, "epoch": 0.27869453612027867, "step": 380}, {"loss": 1.2258, "grad_norm": 0.3764737844467163, "learning_rate": 0.0002, "epoch": 0.28602860286028603, "step": 390}, {"loss": 1.1917, "grad_norm": 0.4243989586830139, "learning_rate": 0.0002, "epoch": 0.29336266960029334, "step": 400}, {"loss": 1.199, "grad_norm": 0.2658531963825226, "learning_rate": 0.0002, "epoch": 0.3006967363403007, "step": 410}, {"loss": 1.1622, "grad_norm": 0.3436793386936188, "learning_rate": 0.0002, "epoch": 0.30803080308030806, "step": 420}, {"loss": 1.2953, "grad_norm": 0.5101129412651062, "learning_rate": 0.0002, "epoch": 0.31536486982031536, "step": 430}, {"loss": 1.1557, "grad_norm": 0.3319750726222992, "learning_rate": 0.0002, "epoch": 0.3226989365603227, "step": 440}, {"loss": 1.1804, "grad_norm": 0.385148286819458, "learning_rate": 0.0002, "epoch": 0.33003300330033003, "step": 450}, {"loss": 1.1808, "grad_norm": 0.3477935791015625, "learning_rate": 0.0002, "epoch": 0.3373670700403374, "step": 460}, {"loss": 1.1877, "grad_norm": 0.29748716950416565, "learning_rate": 0.0002, "epoch": 0.3447011367803447, "step": 470}, {"loss": 1.19, "grad_norm": 0.34083324670791626, "learning_rate": 0.0002, "epoch": 0.35203520352035206, "step": 480}, {"loss": 1.2, "grad_norm": 0.36904552578926086, "learning_rate": 0.0002, "epoch": 0.35936927026035936, "step": 490}, {"loss": 1.2223, "grad_norm": 0.315483033657074, "learning_rate": 0.0002, "epoch": 0.3667033370003667, "step": 500}, {"loss": 1.1461, "grad_norm": 0.44897955656051636, "learning_rate": 0.0002, "epoch": 0.37403740374037403, "step": 510}, {"loss": 1.3035, "grad_norm": 0.3160701394081116, "learning_rate": 0.0002, "epoch": 0.3813714704803814, "step": 520}, {"loss": 1.3197, "grad_norm": 0.29584741592407227, "learning_rate": 0.0002, "epoch": 0.3887055372203887, "step": 530}, {"loss": 1.2983, "grad_norm": 0.5430002808570862, "learning_rate": 0.0002, "epoch": 0.39603960396039606, "step": 540}, {"loss": 1.2459, "grad_norm": 0.2908070683479309, "learning_rate": 0.0002, "epoch": 0.40337367070040336, "step": 550}, {"loss": 1.2384, "grad_norm": 0.35066530108451843, "learning_rate": 0.0002, "epoch": 0.4107077374404107, "step": 560}, {"loss": 1.1784, "grad_norm": 0.37588003277778625, "learning_rate": 0.0002, "epoch": 0.41804180418041803, "step": 570}, {"loss": 1.2334, "grad_norm": 0.3112126886844635, "learning_rate": 0.0002, "epoch": 0.4253758709204254, "step": 580}, {"loss": 1.1439, "grad_norm": 0.35577139258384705, "learning_rate": 0.0002, "epoch": 0.4327099376604327, "step": 590}, {"loss": 1.184, "grad_norm": 0.31706422567367554, "learning_rate": 0.0002, "epoch": 0.44004400440044006, "step": 600}, {"loss": 1.2081, "grad_norm": 0.3249092102050781, "learning_rate": 0.0002, "epoch": 0.44737807114044736, "step": 610}, {"loss": 1.0824, "grad_norm": 0.3842705488204956, "learning_rate": 0.0002, "epoch": 0.4547121378804547, "step": 620}, {"loss": 1.2257, "grad_norm": 0.390991747379303, "learning_rate": 0.0002, "epoch": 0.46204620462046203, "step": 630}, {"loss": 1.1954, "grad_norm": 0.27532413601875305, "learning_rate": 0.0002, "epoch": 0.4693802713604694, "step": 640}, {"loss": 1.1058, "grad_norm": 0.31412816047668457, "learning_rate": 0.0002, "epoch": 0.4767143381004767, "step": 650}, {"loss": 1.1312, "grad_norm": 0.32117101550102234, "learning_rate": 0.0002, "epoch": 0.48404840484048406, "step": 660}, {"loss": 1.2423, "grad_norm": 0.3810010254383087, "learning_rate": 0.0002, "epoch": 0.49138247158049136, "step": 670}, {"loss": 1.1978, "grad_norm": 0.36289164423942566, "learning_rate": 0.0002, "epoch": 0.4987165383204987, "step": 680}, {"loss": 1.2034, "grad_norm": 0.34458720684051514, "learning_rate": 0.0002, "epoch": 0.506050605060506, "step": 690}, {"loss": 1.1756, "grad_norm": 0.32844600081443787, "learning_rate": 0.0002, "epoch": 0.5133846718005134, "step": 700}, {"loss": 1.0807, "grad_norm": 0.3144175708293915, "learning_rate": 0.0002, "epoch": 0.5207187385405208, "step": 710}, {"loss": 1.1952, "grad_norm": 0.3898887634277344, "learning_rate": 0.0002, "epoch": 0.528052805280528, "step": 720}, {"loss": 1.1244, "grad_norm": 1.3220758438110352, "learning_rate": 0.0002, "epoch": 0.5353868720205354, "step": 730}, {"loss": 1.227, "grad_norm": 0.3635874390602112, "learning_rate": 0.0002, "epoch": 0.5427209387605427, "step": 740}, {"loss": 1.2169, "grad_norm": 0.3138217628002167, "learning_rate": 0.0002, "epoch": 0.5500550055005501, "step": 750}, {"loss": 1.1516, "grad_norm": 0.4063207805156708, "learning_rate": 0.0002, "epoch": 0.5573890722405573, "step": 760}, {"loss": 1.1954, "grad_norm": 0.3926219940185547, "learning_rate": 0.0002, "epoch": 0.5647231389805647, "step": 770}, {"loss": 1.1726, "grad_norm": 0.31954652070999146, "learning_rate": 0.0002, "epoch": 0.5720572057205721, "step": 780}, {"loss": 1.2977, "grad_norm": 0.4248711168766022, "learning_rate": 0.0002, "epoch": 0.5793912724605794, "step": 790}, {"loss": 1.1728, "grad_norm": 0.643004834651947, "learning_rate": 0.0002, "epoch": 0.5867253392005867, "step": 800}, {"loss": 1.1793, "grad_norm": 0.3479592800140381, "learning_rate": 0.0002, "epoch": 0.594059405940594, "step": 810}, {"loss": 1.2426, "grad_norm": 0.4684754014015198, "learning_rate": 0.0002, "epoch": 0.6013934726806014, "step": 820}, {"loss": 1.2002, "grad_norm": 0.3739790916442871, "learning_rate": 0.0002, "epoch": 0.6087275394206088, "step": 830}, {"loss": 1.2139, "grad_norm": 0.40884748101234436, "learning_rate": 0.0002, "epoch": 0.6160616061606161, "step": 840}, {"loss": 1.1557, "grad_norm": 0.9722164273262024, "learning_rate": 0.0002, "epoch": 0.6233956729006234, "step": 850}, {"loss": 1.3069, "grad_norm": 0.42992347478866577, "learning_rate": 0.0002, "epoch": 0.6307297396406307, "step": 860}, {"loss": 1.1339, "grad_norm": 0.36654195189476013, "learning_rate": 0.0002, "epoch": 0.6380638063806381, "step": 870}, {"loss": 1.1932, "grad_norm": 0.4113832116127014, "learning_rate": 0.0002, "epoch": 0.6453978731206454, "step": 880}, {"loss": 1.2163, "grad_norm": 0.2948838770389557, "learning_rate": 0.0002, "epoch": 0.6527319398606527, "step": 890}, {"loss": 1.1081, "grad_norm": 0.38330280780792236, "learning_rate": 0.0002, "epoch": 0.6600660066006601, "step": 900}, {"loss": 1.1342, "grad_norm": 0.4428867697715759, "learning_rate": 0.0002, "epoch": 0.6674000733406674, "step": 910}, {"loss": 1.1021, "grad_norm": 0.23659265041351318, "learning_rate": 0.0002, "epoch": 0.6747341400806748, "step": 920}, {"loss": 1.1226, "grad_norm": 0.323685884475708, "learning_rate": 0.0002, "epoch": 0.682068206820682, "step": 930}, {"loss": 1.0853, "grad_norm": 0.39157727360725403, "learning_rate": 0.0002, "epoch": 0.6894022735606894, "step": 940}, {"loss": 1.1435, "grad_norm": 0.27189481258392334, "learning_rate": 0.0002, "epoch": 0.6967363403006968, "step": 950}, {"loss": 1.1033, "grad_norm": 0.529883861541748, "learning_rate": 0.0002, "epoch": 0.7040704070407041, "step": 960}, {"loss": 1.139, "grad_norm": 0.34758689999580383, "learning_rate": 0.0002, "epoch": 0.7114044737807114, "step": 970}, {"loss": 1.2197, "grad_norm": 0.831749439239502, "learning_rate": 0.0002, "epoch": 0.7187385405207187, "step": 980}, {"loss": 1.158, "grad_norm": 0.4438304007053375, "learning_rate": 0.0002, "epoch": 0.7260726072607261, "step": 990}, {"loss": 1.1021, "grad_norm": 0.33840006589889526, "learning_rate": 0.0002, "epoch": 0.7334066740007334, "step": 1000}, {"loss": 1.254, "grad_norm": 0.3454797863960266, "learning_rate": 0.0002, "epoch": 0.7407407407407407, "step": 1010}, {"loss": 1.106, "grad_norm": 0.38999441266059875, "learning_rate": 0.0002, "epoch": 0.7480748074807481, "step": 1020}, {"loss": 1.1428, "grad_norm": 0.2829911708831787, "learning_rate": 0.0002, "epoch": 0.7554088742207554, "step": 1030}, {"loss": 1.2123, "grad_norm": 0.36918163299560547, "learning_rate": 0.0002, "epoch": 0.7627429409607628, "step": 1040}, {"loss": 1.3028, "grad_norm": 0.3415680229663849, "learning_rate": 0.0002, "epoch": 0.77007700770077, "step": 1050}, {"loss": 1.1939, "grad_norm": 0.2974182963371277, "learning_rate": 0.0002, "epoch": 0.7774110744407774, "step": 1060}, {"loss": 1.194, "grad_norm": 0.3880919814109802, "learning_rate": 0.0002, "epoch": 0.7847451411807848, "step": 1070}, {"loss": 1.1095, "grad_norm": 0.33503302931785583, "learning_rate": 0.0002, "epoch": 0.7920792079207921, "step": 1080}, {"loss": 1.2111, "grad_norm": 0.3728407025337219, "learning_rate": 0.0002, "epoch": 0.7994132746607994, "step": 1090}, {"loss": 1.0835, "grad_norm": 0.3509373664855957, "learning_rate": 0.0002, "epoch": 0.8067473414008067, "step": 1100}, {"loss": 1.2661, "grad_norm": 0.42228564620018005, "learning_rate": 0.0002, "epoch": 0.8140814081408141, "step": 1110}, {"loss": 1.1788, "grad_norm": 0.313467800617218, "learning_rate": 0.0002, "epoch": 0.8214154748808215, "step": 1120}, {"loss": 1.1971, "grad_norm": 0.3378850817680359, "learning_rate": 0.0002, "epoch": 0.8287495416208287, "step": 1130}, {"loss": 1.1238, "grad_norm": 0.43200382590293884, "learning_rate": 0.0002, "epoch": 0.8360836083608361, "step": 1140}, {"loss": 1.3203, "grad_norm": 0.3309599459171295, "learning_rate": 0.0002, "epoch": 0.8434176751008434, "step": 1150}, {"loss": 1.1062, "grad_norm": 0.3526846170425415, "learning_rate": 0.0002, "epoch": 0.8507517418408508, "step": 1160}, {"loss": 1.0851, "grad_norm": 1.2722247838974, "learning_rate": 0.0002, "epoch": 0.858085808580858, "step": 1170}, {"loss": 1.0785, "grad_norm": 0.34142059087753296, "learning_rate": 0.0002, "epoch": 0.8654198753208654, "step": 1180}, {"loss": 1.2187, "grad_norm": 0.3805823028087616, "learning_rate": 0.0002, "epoch": 0.8727539420608728, "step": 1190}, {"loss": 1.1215, "grad_norm": 0.3931232690811157, "learning_rate": 0.0002, "epoch": 0.8800880088008801, "step": 1200}, {"loss": 1.0948, "grad_norm": 0.2937372624874115, "learning_rate": 0.0002, "epoch": 0.8874220755408874, "step": 1210}, {"loss": 1.1228, "grad_norm": 0.3757196366786957, "learning_rate": 0.0002, "epoch": 0.8947561422808947, "step": 1220}, {"loss": 1.1222, "grad_norm": 0.3502705991268158, "learning_rate": 0.0002, "epoch": 0.9020902090209021, "step": 1230}, {"loss": 1.2242, "grad_norm": 0.32758915424346924, "learning_rate": 0.0002, "epoch": 0.9094242757609095, "step": 1240}, {"loss": 1.215, "grad_norm": 0.37199416756629944, "learning_rate": 0.0002, "epoch": 0.9167583425009168, "step": 1250}, {"loss": 1.1225, "grad_norm": 0.3551490604877472, "learning_rate": 0.0002, "epoch": 0.9240924092409241, "step": 1260}, {"loss": 1.1966, "grad_norm": 0.2859550714492798, "learning_rate": 0.0002, "epoch": 0.9314264759809314, "step": 1270}, {"loss": 1.2186, "grad_norm": 0.427990585565567, "learning_rate": 0.0002, "epoch": 0.9387605427209388, "step": 1280}, {"loss": 1.2848, "grad_norm": 0.33717992901802063, "learning_rate": 0.0002, "epoch": 0.9460946094609461, "step": 1290}, {"loss": 1.1656, "grad_norm": 0.30225634574890137, "learning_rate": 0.0002, "epoch": 0.9534286762009534, "step": 1300}, {"loss": 1.2404, "grad_norm": 0.385821133852005, "learning_rate": 0.0002, "epoch": 0.9607627429409608, "step": 1310}, {"loss": 1.1932, "grad_norm": 0.35278066992759705, "learning_rate": 0.0002, "epoch": 0.9680968096809681, "step": 1320}, {"loss": 1.1071, "grad_norm": 0.49987098574638367, "learning_rate": 0.0002, "epoch": 0.9754308764209755, "step": 1330}, {"loss": 1.2259, "grad_norm": 0.3842747211456299, "learning_rate": 0.0002, "epoch": 0.9827649431609827, "step": 1340}, {"loss": 1.0862, "grad_norm": 0.6274653673171997, "learning_rate": 0.0002, "epoch": 0.9900990099009901, "step": 1350}, {"loss": 1.124, "grad_norm": 0.5239808559417725, "learning_rate": 0.0002, "epoch": 0.9974330766409975, "step": 1360}, {"eval_loss": 1.1822267770767212, "eval_runtime": 32.7389, "eval_samples_per_second": 13.165, "eval_steps_per_second": 1.649, "epoch": 0.9996332966629996, "step": 1363}, {"loss": 1.096, "grad_norm": 0.45311301946640015, "learning_rate": 0.0002, "epoch": 1.0047671433810048, "step": 1370}, {"loss": 1.0143, "grad_norm": 0.29685574769973755, "learning_rate": 0.0002, "epoch": 1.012101210121012, "step": 1380}, {"loss": 1.0302, "grad_norm": 0.3290937840938568, "learning_rate": 0.0002, "epoch": 1.0194352768610195, "step": 1390}, {"loss": 1.0295, "grad_norm": 0.3801758587360382, "learning_rate": 0.0002, "epoch": 1.0267693436010268, "step": 1400}, {"loss": 1.1226, "grad_norm": 0.794174313545227, "learning_rate": 0.0002, "epoch": 1.034103410341034, "step": 1410}, {"loss": 1.2232, "grad_norm": 0.3854154646396637, "learning_rate": 0.0002, "epoch": 1.0414374770810415, "step": 1420}, {"loss": 1.0652, "grad_norm": 0.32702451944351196, "learning_rate": 0.0002, "epoch": 1.0487715438210488, "step": 1430}, {"loss": 1.1144, "grad_norm": 0.7815203666687012, "learning_rate": 0.0002, "epoch": 1.056105610561056, "step": 1440}, {"loss": 1.1316, "grad_norm": 0.3087436854839325, "learning_rate": 0.0002, "epoch": 1.0634396773010635, "step": 1450}, {"loss": 1.1124, "grad_norm": 0.3847602903842926, "learning_rate": 0.0002, "epoch": 1.0707737440410707, "step": 1460}, {"loss": 1.1428, "grad_norm": 0.3693031370639801, "learning_rate": 0.0002, "epoch": 1.0781078107810782, "step": 1470}, {"loss": 1.0995, "grad_norm": 0.4111202359199524, "learning_rate": 0.0002, "epoch": 1.0854418775210855, "step": 1480}, {"loss": 1.0961, "grad_norm": 0.41452381014823914, "learning_rate": 0.0002, "epoch": 1.0927759442610927, "step": 1490}, {"loss": 1.1068, "grad_norm": 0.3336445093154907, "learning_rate": 0.0002, "epoch": 1.1001100110011002, "step": 1500}, {"loss": 1.0556, "grad_norm": 0.3923407793045044, "learning_rate": 0.0002, "epoch": 1.1074440777411074, "step": 1510}, {"loss": 1.1644, "grad_norm": 0.46215683221817017, "learning_rate": 0.0002, "epoch": 1.1147781444811147, "step": 1520}, {"loss": 1.1133, "grad_norm": 0.3592156767845154, "learning_rate": 0.0002, "epoch": 1.1221122112211221, "step": 1530}, {"loss": 1.0957, "grad_norm": 0.361110657453537, "learning_rate": 0.0002, "epoch": 1.1294462779611294, "step": 1540}, {"loss": 1.1553, "grad_norm": 0.5317131280899048, "learning_rate": 0.0002, "epoch": 1.1367803447011369, "step": 1550}, {"loss": 1.0368, "grad_norm": 0.3882388174533844, "learning_rate": 0.0002, "epoch": 1.1441144114411441, "step": 1560}, {"loss": 1.0805, "grad_norm": 0.3259428143501282, "learning_rate": 0.0002, "epoch": 1.1514484781811514, "step": 1570}, {"loss": 1.1819, "grad_norm": 0.410935640335083, "learning_rate": 0.0002, "epoch": 1.1587825449211588, "step": 1580}, {"loss": 1.1143, "grad_norm": 0.44940185546875, "learning_rate": 0.0002, "epoch": 1.166116611661166, "step": 1590}, {"loss": 1.0334, "grad_norm": 0.5106484293937683, "learning_rate": 0.0002, "epoch": 1.1734506784011733, "step": 1600}, {"loss": 1.2376, "grad_norm": 0.6603665947914124, "learning_rate": 0.0002, "epoch": 1.1807847451411808, "step": 1610}, {"loss": 1.1227, "grad_norm": 0.4799964129924774, "learning_rate": 0.0002, "epoch": 1.188118811881188, "step": 1620}, {"loss": 1.1191, "grad_norm": 0.4389883279800415, "learning_rate": 0.0002, "epoch": 1.1954528786211955, "step": 1630}, {"loss": 1.0667, "grad_norm": 0.4188813269138336, "learning_rate": 0.0002, "epoch": 1.2027869453612028, "step": 1640}, {"loss": 1.0605, "grad_norm": 0.7132157683372498, "learning_rate": 0.0002, "epoch": 1.21012101210121, "step": 1650}, {"loss": 1.0204, "grad_norm": 0.507480263710022, "learning_rate": 0.0002, "epoch": 1.2174550788412175, "step": 1660}, {"loss": 0.9948, "grad_norm": 0.9452332854270935, "learning_rate": 0.0002, "epoch": 1.2247891455812248, "step": 1670}, {"loss": 1.0228, "grad_norm": 0.4121614992618561, "learning_rate": 0.0002, "epoch": 1.2321232123212322, "step": 1680}, {"loss": 1.0366, "grad_norm": 0.34230247139930725, "learning_rate": 0.0002, "epoch": 1.2394572790612395, "step": 1690}, {"loss": 1.1289, "grad_norm": 0.4026208817958832, "learning_rate": 0.0002, "epoch": 1.2467913458012467, "step": 1700}, {"loss": 1.0206, "grad_norm": 0.46673697233200073, "learning_rate": 0.0002, "epoch": 1.2541254125412542, "step": 1710}, {"loss": 1.0827, "grad_norm": 0.38349825143814087, "learning_rate": 0.0002, "epoch": 1.2614594792812615, "step": 1720}, {"loss": 1.0356, "grad_norm": 0.4049997627735138, "learning_rate": 0.0002, "epoch": 1.2687935460212687, "step": 1730}, {"loss": 0.9504, "grad_norm": 0.3417615294456482, "learning_rate": 0.0002, "epoch": 1.2761276127612762, "step": 1740}, {"loss": 1.094, "grad_norm": 0.4277614951133728, "learning_rate": 0.0002, "epoch": 1.2834616795012834, "step": 1750}, {"loss": 0.9938, "grad_norm": 0.5864202976226807, "learning_rate": 0.0002, "epoch": 1.2907957462412907, "step": 1760}, {"loss": 1.1167, "grad_norm": 0.7097493410110474, "learning_rate": 0.0002, "epoch": 1.2981298129812981, "step": 1770}, {"loss": 1.1132, "grad_norm": 0.3145381212234497, "learning_rate": 0.0002, "epoch": 1.3054638797213054, "step": 1780}, {"loss": 1.1099, "grad_norm": 0.5116165280342102, "learning_rate": 0.0002, "epoch": 1.3127979464613129, "step": 1790}, {"loss": 1.0765, "grad_norm": 0.7469736337661743, "learning_rate": 0.0002, "epoch": 1.3201320132013201, "step": 1800}, {"loss": 1.0663, "grad_norm": 0.32272255420684814, "learning_rate": 0.0002, "epoch": 1.3274660799413276, "step": 1810}, {"loss": 0.9887, "grad_norm": 0.3534623086452484, "learning_rate": 0.0002, "epoch": 1.3348001466813348, "step": 1820}, {"loss": 1.1628, "grad_norm": 0.36127907037734985, "learning_rate": 0.0002, "epoch": 1.342134213421342, "step": 1830}, {"loss": 1.0972, "grad_norm": 0.4072401523590088, "learning_rate": 0.0002, "epoch": 1.3494682801613496, "step": 1840}, {"loss": 1.1267, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.3568023469013568, "step": 1850}, {"loss": 1.0173, "grad_norm": 0.412883460521698, "learning_rate": 0.0002, "epoch": 1.364136413641364, "step": 1860}, {"loss": 1.0265, "grad_norm": 0.3735875189304352, "learning_rate": 0.0002, "epoch": 1.3714704803813715, "step": 1870}, {"loss": 1.1061, "grad_norm": 0.39158159494400024, "learning_rate": 0.0002, "epoch": 1.3788045471213788, "step": 1880}, {"loss": 1.0433, "grad_norm": 0.44431769847869873, "learning_rate": 0.0002, "epoch": 1.386138613861386, "step": 1890}, {"loss": 1.0216, "grad_norm": 0.37772801518440247, "learning_rate": 0.0002, "epoch": 1.3934726806013935, "step": 1900}, {"loss": 1.0674, "grad_norm": 0.4056641757488251, "learning_rate": 0.0002, "epoch": 1.4008067473414008, "step": 1910}, {"loss": 1.0256, "grad_norm": 0.41612377762794495, "learning_rate": 0.0002, "epoch": 1.408140814081408, "step": 1920}, {"loss": 1.0467, "grad_norm": 0.41153013706207275, "learning_rate": 0.0002, "epoch": 1.4154748808214155, "step": 1930}, {"loss": 1.1062, "grad_norm": 0.387845516204834, "learning_rate": 0.0002, "epoch": 1.4228089475614227, "step": 1940}, {"loss": 1.1094, "grad_norm": 0.3809587061405182, "learning_rate": 0.0002, "epoch": 1.4301430143014302, "step": 1950}, {"loss": 1.0461, "grad_norm": 0.3625726103782654, "learning_rate": 0.0002, "epoch": 1.4374770810414375, "step": 1960}, {"loss": 0.9983, "grad_norm": 0.5294290781021118, "learning_rate": 0.0002, "epoch": 1.444811147781445, "step": 1970}, {"loss": 1.1114, "grad_norm": 0.39975494146347046, "learning_rate": 0.0002, "epoch": 1.4521452145214522, "step": 1980}, {"loss": 0.9704, "grad_norm": 0.4181167185306549, "learning_rate": 0.0002, "epoch": 1.4594792812614594, "step": 1990}, {"loss": 1.1146, "grad_norm": 0.42001503705978394, "learning_rate": 0.0002, "epoch": 1.466813348001467, "step": 2000}, {"loss": 1.1266, "grad_norm": 0.4877578616142273, "learning_rate": 0.0002, "epoch": 1.4741474147414741, "step": 2010}, {"loss": 1.1012, "grad_norm": 0.4050969183444977, "learning_rate": 0.0002, "epoch": 1.4814814814814814, "step": 2020}, {"loss": 1.0562, "grad_norm": 0.39068883657455444, "learning_rate": 0.0002, "epoch": 1.4888155482214889, "step": 2030}, {"loss": 1.0464, "grad_norm": 0.421282559633255, "learning_rate": 0.0002, "epoch": 1.4961496149614961, "step": 2040}, {"loss": 1.0532, "grad_norm": 0.47092297673225403, "learning_rate": 0.0002, "epoch": 1.5034836817015034, "step": 2050}, {"loss": 0.9348, "grad_norm": 0.39688974618911743, "learning_rate": 0.0002, "epoch": 1.5108177484415108, "step": 2060}, {"loss": 1.08, "grad_norm": 0.5529879331588745, "learning_rate": 0.0002, "epoch": 1.5181518151815183, "step": 2070}, {"loss": 1.1836, "grad_norm": 0.4879782199859619, "learning_rate": 0.0002, "epoch": 1.5254858819215253, "step": 2080}, {"loss": 1.0432, "grad_norm": 0.5517361164093018, "learning_rate": 0.0002, "epoch": 1.5328199486615328, "step": 2090}, {"loss": 1.0433, "grad_norm": 0.44015637040138245, "learning_rate": 0.0002, "epoch": 1.5401540154015403, "step": 2100}, {"loss": 1.1873, "grad_norm": 0.5435167551040649, "learning_rate": 0.0002, "epoch": 1.5474880821415475, "step": 2110}, {"loss": 1.1076, "grad_norm": 0.5714033246040344, "learning_rate": 0.0002, "epoch": 1.5548221488815548, "step": 2120}, {"loss": 1.1107, "grad_norm": 0.31732529401779175, "learning_rate": 0.0002, "epoch": 1.5621562156215623, "step": 2130}, {"loss": 1.0817, "grad_norm": 0.49068278074264526, "learning_rate": 0.0002, "epoch": 1.5694902823615695, "step": 2140}, {"loss": 1.0254, "grad_norm": 0.46851542592048645, "learning_rate": 0.0002, "epoch": 1.5768243491015768, "step": 2150}, {"loss": 1.0623, "grad_norm": 0.5083092451095581, "learning_rate": 0.0002, "epoch": 1.5841584158415842, "step": 2160}, {"loss": 1.0603, "grad_norm": 0.9822936058044434, "learning_rate": 0.0002, "epoch": 1.5914924825815915, "step": 2170}, {"loss": 0.9986, "grad_norm": 0.4575989246368408, "learning_rate": 0.0002, "epoch": 1.5988265493215987, "step": 2180}, {"loss": 1.1292, "grad_norm": 0.47444286942481995, "learning_rate": 0.0002, "epoch": 1.6061606160616062, "step": 2190}, {"loss": 1.0136, "grad_norm": 0.7208226919174194, "learning_rate": 0.0002, "epoch": 1.6134946828016135, "step": 2200}, {"loss": 1.15, "grad_norm": 0.43791481852531433, "learning_rate": 0.0002, "epoch": 1.6208287495416207, "step": 2210}, {"loss": 1.0961, "grad_norm": 0.5245792865753174, "learning_rate": 0.0002, "epoch": 1.6281628162816282, "step": 2220}, {"loss": 0.9957, "grad_norm": 0.39289429783821106, "learning_rate": 0.0002, "epoch": 1.6354968830216357, "step": 2230}, {"loss": 1.133, "grad_norm": 0.6106135845184326, "learning_rate": 0.0002, "epoch": 1.6428309497616427, "step": 2240}, {"loss": 1.0129, "grad_norm": 0.3722580671310425, "learning_rate": 0.0002, "epoch": 1.6501650165016502, "step": 2250}, {"loss": 1.0446, "grad_norm": 0.3649403750896454, "learning_rate": 0.0002, "epoch": 1.6574990832416576, "step": 2260}, {"loss": 1.0037, "grad_norm": 0.46514248847961426, "learning_rate": 0.0002, "epoch": 1.6648331499816649, "step": 2270}, {"loss": 1.0022, "grad_norm": 0.42034927010536194, "learning_rate": 0.0002, "epoch": 1.6721672167216721, "step": 2280}, {"loss": 1.1362, "grad_norm": 0.45202910900115967, "learning_rate": 0.0002, "epoch": 1.6795012834616796, "step": 2290}, {"loss": 1.0866, "grad_norm": 0.36257603764533997, "learning_rate": 0.0002, "epoch": 1.6868353502016868, "step": 2300}, {"loss": 1.0973, "grad_norm": 0.6340323090553284, "learning_rate": 0.0002, "epoch": 1.694169416941694, "step": 2310}, {"loss": 1.0615, "grad_norm": 0.4352878928184509, "learning_rate": 0.0002, "epoch": 1.7015034836817016, "step": 2320}, {"loss": 1.0629, "grad_norm": 0.45029792189598083, "learning_rate": 0.0002, "epoch": 1.7088375504217088, "step": 2330}, {"loss": 0.9621, "grad_norm": 0.3891315758228302, "learning_rate": 0.0002, "epoch": 1.716171617161716, "step": 2340}, {"loss": 0.9779, "grad_norm": 0.35180050134658813, "learning_rate": 0.0002, "epoch": 1.7235056839017235, "step": 2350}, {"loss": 1.0368, "grad_norm": 0.42367449402809143, "learning_rate": 0.0002, "epoch": 1.7308397506417308, "step": 2360}, {"loss": 1.0376, "grad_norm": 0.4553675353527069, "learning_rate": 0.0002, "epoch": 1.738173817381738, "step": 2370}, {"loss": 1.1467, "grad_norm": 0.5944654941558838, "learning_rate": 0.0002, "epoch": 1.7455078841217455, "step": 2380}, {"loss": 1.0548, "grad_norm": 0.3479664623737335, "learning_rate": 0.0002, "epoch": 1.752841950861753, "step": 2390}, {"loss": 1.0798, "grad_norm": 0.3585502505302429, "learning_rate": 0.0002, "epoch": 1.76017601760176, "step": 2400}, {"loss": 1.0983, "grad_norm": 0.4263346493244171, "learning_rate": 0.0002, "epoch": 1.7675100843417675, "step": 2410}, {"loss": 1.054, "grad_norm": 0.5476409196853638, "learning_rate": 0.0002, "epoch": 1.774844151081775, "step": 2420}, {"loss": 1.1615, "grad_norm": 0.3694186508655548, "learning_rate": 0.0002, "epoch": 1.7821782178217822, "step": 2430}, {"loss": 1.1343, "grad_norm": 0.9185658693313599, "learning_rate": 0.0002, "epoch": 1.7895122845617895, "step": 2440}, {"loss": 1.0764, "grad_norm": 0.7171908020973206, "learning_rate": 0.0002, "epoch": 1.796846351301797, "step": 2450}, {"loss": 1.1154, "grad_norm": 0.550658643245697, "learning_rate": 0.0002, "epoch": 1.8041804180418042, "step": 2460}, {"loss": 0.9975, "grad_norm": 0.4075568914413452, "learning_rate": 0.0002, "epoch": 1.8115144847818114, "step": 2470}, {"loss": 1.0935, "grad_norm": 0.3790127635002136, "learning_rate": 0.0002, "epoch": 1.818848551521819, "step": 2480}, {"loss": 0.9839, "grad_norm": 0.3576384484767914, "learning_rate": 0.0002, "epoch": 1.8261826182618262, "step": 2490}, {"loss": 1.1369, "grad_norm": 0.3919370770454407, "learning_rate": 0.0002, "epoch": 1.8335166850018334, "step": 2500}, {"loss": 0.9985, "grad_norm": 0.485083669424057, "learning_rate": 0.0002, "epoch": 1.8408507517418409, "step": 2510}, {"loss": 1.1585, "grad_norm": 0.4564347565174103, "learning_rate": 0.0002, "epoch": 1.8481848184818483, "step": 2520}, {"loss": 1.0944, "grad_norm": 0.3613106608390808, "learning_rate": 0.0002, "epoch": 1.8555188852218554, "step": 2530}, {"loss": 1.0819, "grad_norm": 0.39600759744644165, "learning_rate": 0.0002, "epoch": 1.8628529519618628, "step": 2540}, {"loss": 0.9453, "grad_norm": 1.123499870300293, "learning_rate": 0.0002, "epoch": 1.8701870187018703, "step": 2550}, {"loss": 1.0635, "grad_norm": 0.4612680673599243, "learning_rate": 0.0002, "epoch": 1.8775210854418776, "step": 2560}, {"loss": 1.0087, "grad_norm": 0.42745399475097656, "learning_rate": 0.0002, "epoch": 1.8848551521818848, "step": 2570}, {"loss": 1.0102, "grad_norm": 0.4055580198764801, "learning_rate": 0.0002, "epoch": 1.8921892189218923, "step": 2580}, {"loss": 1.0177, "grad_norm": 0.44174644351005554, "learning_rate": 0.0002, "epoch": 1.8995232856618995, "step": 2590}, {"loss": 0.9886, "grad_norm": 1.0228385925292969, "learning_rate": 0.0002, "epoch": 1.9068573524019068, "step": 2600}, {"loss": 1.0857, "grad_norm": 0.3496396243572235, "learning_rate": 0.0002, "epoch": 1.9141914191419143, "step": 2610}, {"loss": 1.0955, "grad_norm": 0.4191173017024994, "learning_rate": 0.0002, "epoch": 1.9215254858819215, "step": 2620}, {"loss": 1.0943, "grad_norm": 0.6778554916381836, "learning_rate": 0.0002, "epoch": 1.9288595526219288, "step": 2630}, {"loss": 1.0594, "grad_norm": 0.41992834210395813, "learning_rate": 0.0002, "epoch": 1.9361936193619362, "step": 2640}, {"loss": 1.1159, "grad_norm": 0.8760401010513306, "learning_rate": 0.0002, "epoch": 1.9435276861019435, "step": 2650}, {"loss": 1.0379, "grad_norm": 0.44049209356307983, "learning_rate": 0.0002, "epoch": 1.9508617528419507, "step": 2660}, {"loss": 1.1008, "grad_norm": 0.5651928782463074, "learning_rate": 0.0002, "epoch": 1.9581958195819582, "step": 2670}, {"loss": 1.1317, "grad_norm": 0.5292727947235107, "learning_rate": 0.0002, "epoch": 1.9655298863219657, "step": 2680}, {"loss": 1.1328, "grad_norm": 0.6012240648269653, "learning_rate": 0.0002, "epoch": 1.9728639530619727, "step": 2690}, {"loss": 1.0683, "grad_norm": 0.3945149779319763, "learning_rate": 0.0002, "epoch": 1.9801980198019802, "step": 2700}, {"loss": 1.0155, "grad_norm": 0.5732627511024475, "learning_rate": 0.0002, "epoch": 1.9875320865419877, "step": 2710}, {"loss": 0.9857, "grad_norm": 0.3963361084461212, "learning_rate": 0.0002, "epoch": 1.994866153281995, "step": 2720}, {"eval_loss": 1.1534006595611572, "eval_runtime": 32.7541, "eval_samples_per_second": 13.159, "eval_steps_per_second": 1.649, "epoch": 2.0, "step": 2727}, {"loss": 0.9624, "grad_norm": 0.48628315329551697, "learning_rate": 0.0002, "epoch": 2.002200220022002, "step": 2730}, {"loss": 0.9603, "grad_norm": 0.413875013589859, "learning_rate": 0.0002, "epoch": 2.0095342867620096, "step": 2740}, {"loss": 0.965, "grad_norm": 0.4988735616207123, "learning_rate": 0.0002, "epoch": 2.0168683535020167, "step": 2750}, {"loss": 0.9677, "grad_norm": 0.5634812712669373, "learning_rate": 0.0002, "epoch": 2.024202420242024, "step": 2760}, {"loss": 0.9547, "grad_norm": 0.48302653431892395, "learning_rate": 0.0002, "epoch": 2.0315364869820316, "step": 2770}, {"loss": 0.9346, "grad_norm": 0.49914175271987915, "learning_rate": 0.0002, "epoch": 2.038870553722039, "step": 2780}, {"loss": 0.904, "grad_norm": 1.14039945602417, "learning_rate": 0.0002, "epoch": 2.046204620462046, "step": 2790}, {"loss": 0.9588, "grad_norm": 0.6359720826148987, "learning_rate": 0.0002, "epoch": 2.0535386872020536, "step": 2800}, {"loss": 0.9031, "grad_norm": 0.4589158296585083, "learning_rate": 0.0002, "epoch": 2.060872753942061, "step": 2810}, {"loss": 0.9438, "grad_norm": 0.46255481243133545, "learning_rate": 0.0002, "epoch": 2.068206820682068, "step": 2820}, {"loss": 0.9464, "grad_norm": 0.6232137680053711, "learning_rate": 0.0002, "epoch": 2.0755408874220755, "step": 2830}, {"loss": 0.8978, "grad_norm": 0.41042178869247437, "learning_rate": 0.0002, "epoch": 2.082874954162083, "step": 2840}, {"loss": 0.8516, "grad_norm": 0.5334428548812866, "learning_rate": 0.0002, "epoch": 2.09020902090209, "step": 2850}, {"loss": 0.9313, "grad_norm": 0.8270058631896973, "learning_rate": 0.0002, "epoch": 2.0975430876420975, "step": 2860}, {"loss": 1.0064, "grad_norm": 0.6624533534049988, "learning_rate": 0.0002, "epoch": 2.104877154382105, "step": 2870}, {"loss": 0.9196, "grad_norm": 0.5448863506317139, "learning_rate": 0.0002, "epoch": 2.112211221122112, "step": 2880}, {"loss": 0.887, "grad_norm": 0.621482789516449, "learning_rate": 0.0002, "epoch": 2.1195452878621195, "step": 2890}, {"loss": 0.9702, "grad_norm": 0.4556255340576172, "learning_rate": 0.0002, "epoch": 2.126879354602127, "step": 2900}, {"loss": 0.9323, "grad_norm": 0.4620579183101654, "learning_rate": 0.0002, "epoch": 2.1342134213421344, "step": 2910}, {"loss": 0.836, "grad_norm": 0.9602415561676025, "learning_rate": 0.0002, "epoch": 2.1415474880821415, "step": 2920}, {"loss": 0.8826, "grad_norm": 0.587943971157074, "learning_rate": 0.0002, "epoch": 2.148881554822149, "step": 2930}, {"loss": 0.971, "grad_norm": 0.5121372938156128, "learning_rate": 0.0002, "epoch": 2.1562156215621564, "step": 2940}, {"loss": 0.8751, "grad_norm": 0.49424484372138977, "learning_rate": 0.0002, "epoch": 2.1635496883021634, "step": 2950}, {"loss": 0.8674, "grad_norm": 0.6312560439109802, "learning_rate": 0.0002, "epoch": 2.170883755042171, "step": 2960}, {"loss": 0.9791, "grad_norm": 0.5235576629638672, "learning_rate": 0.0002, "epoch": 2.1782178217821784, "step": 2970}, {"loss": 0.9706, "grad_norm": 0.5868439674377441, "learning_rate": 0.0002, "epoch": 2.1855518885221854, "step": 2980}, {"loss": 0.9338, "grad_norm": 0.42302873730659485, "learning_rate": 0.0002, "epoch": 2.192885955262193, "step": 2990}, {"loss": 0.9332, "grad_norm": 0.5097725987434387, "learning_rate": 0.0002, "epoch": 2.2002200220022003, "step": 3000}, {"loss": 0.9239, "grad_norm": 0.5091572403907776, "learning_rate": 0.0002, "epoch": 2.2075540887422074, "step": 3010}, {"loss": 0.8898, "grad_norm": 0.49433162808418274, "learning_rate": 0.0002, "epoch": 2.214888155482215, "step": 3020}, {"loss": 0.9734, "grad_norm": 0.5577368140220642, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3030}, {"loss": 0.9033, "grad_norm": 0.6177583932876587, "learning_rate": 0.0002, "epoch": 2.2295562889622293, "step": 3040}, {"loss": 0.9882, "grad_norm": 0.5256719589233398, "learning_rate": 0.0002, "epoch": 2.236890355702237, "step": 3050}, {"loss": 0.9439, "grad_norm": 0.5001118183135986, "learning_rate": 0.0002, "epoch": 2.2442244224422443, "step": 3060}, {"loss": 0.8718, "grad_norm": 0.5721249580383301, "learning_rate": 0.0002, "epoch": 2.2515584891822513, "step": 3070}, {"loss": 1.0648, "grad_norm": 0.5325384140014648, "learning_rate": 0.0002, "epoch": 2.258892555922259, "step": 3080}, {"loss": 0.9843, "grad_norm": 0.5719189047813416, "learning_rate": 0.0002, "epoch": 2.2662266226622663, "step": 3090}, {"loss": 0.8633, "grad_norm": 0.6337835788726807, "learning_rate": 0.0002, "epoch": 2.2735606894022737, "step": 3100}, {"loss": 0.9962, "grad_norm": 0.5381836891174316, "learning_rate": 0.0002, "epoch": 2.2808947561422808, "step": 3110}, {"loss": 0.8265, "grad_norm": 0.5408531427383423, "learning_rate": 0.0002, "epoch": 2.2882288228822882, "step": 3120}, {"loss": 1.0325, "grad_norm": 0.43705281615257263, "learning_rate": 0.0002, "epoch": 2.2955628896222957, "step": 3130}, {"loss": 0.9388, "grad_norm": 0.6454030275344849, "learning_rate": 0.0002, "epoch": 2.3028969563623027, "step": 3140}, {"loss": 0.954, "grad_norm": 0.686030387878418, "learning_rate": 0.0002, "epoch": 2.31023102310231, "step": 3150}, {"loss": 0.9403, "grad_norm": 0.5123633146286011, "learning_rate": 0.0002, "epoch": 2.3175650898423177, "step": 3160}, {"loss": 0.8834, "grad_norm": 0.842506468296051, "learning_rate": 0.0002, "epoch": 2.3248991565823247, "step": 3170}, {"loss": 1.0497, "grad_norm": 0.5193818807601929, "learning_rate": 0.0002, "epoch": 2.332233223322332, "step": 3180}, {"loss": 0.9473, "grad_norm": 0.5634409189224243, "learning_rate": 0.0002, "epoch": 2.3395672900623397, "step": 3190}, {"loss": 0.8499, "grad_norm": 0.6475534439086914, "learning_rate": 0.0002, "epoch": 2.3469013568023467, "step": 3200}, {"loss": 0.874, "grad_norm": 1.1503914594650269, "learning_rate": 0.0002, "epoch": 2.354235423542354, "step": 3210}, {"loss": 0.9762, "grad_norm": 0.7234905362129211, "learning_rate": 0.0002, "epoch": 2.3615694902823616, "step": 3220}, {"loss": 0.9007, "grad_norm": 0.664903461933136, "learning_rate": 0.0002, "epoch": 2.368903557022369, "step": 3230}, {"loss": 0.9987, "grad_norm": 0.5453006625175476, "learning_rate": 0.0002, "epoch": 2.376237623762376, "step": 3240}, {"loss": 0.9742, "grad_norm": 0.6256654262542725, "learning_rate": 0.0002, "epoch": 2.3835716905023836, "step": 3250}, {"loss": 0.9922, "grad_norm": 0.5166565179824829, "learning_rate": 0.0002, "epoch": 2.390905757242391, "step": 3260}, {"loss": 0.927, "grad_norm": 0.5699098110198975, "learning_rate": 0.0002, "epoch": 2.398239823982398, "step": 3270}, {"loss": 0.8878, "grad_norm": 0.4472540020942688, "learning_rate": 0.0002, "epoch": 2.4055738907224056, "step": 3280}, {"loss": 0.9439, "grad_norm": 0.6790403127670288, "learning_rate": 0.0002, "epoch": 2.412907957462413, "step": 3290}, {"loss": 0.972, "grad_norm": 0.5182185173034668, "learning_rate": 0.0002, "epoch": 2.42024202420242, "step": 3300}, {"loss": 0.9775, "grad_norm": 0.564647912979126, "learning_rate": 0.0002, "epoch": 2.4275760909424275, "step": 3310}, {"loss": 1.072, "grad_norm": 0.5625313520431519, "learning_rate": 0.0002, "epoch": 2.434910157682435, "step": 3320}, {"loss": 0.8798, "grad_norm": 0.7496559619903564, "learning_rate": 0.0002, "epoch": 2.442244224422442, "step": 3330}, {"loss": 0.868, "grad_norm": 0.4779128134250641, "learning_rate": 0.0002, "epoch": 2.4495782911624495, "step": 3340}, {"loss": 1.0316, "grad_norm": 0.578093409538269, "learning_rate": 0.0002, "epoch": 2.456912357902457, "step": 3350}, {"loss": 0.9282, "grad_norm": 0.5456080436706543, "learning_rate": 0.0002, "epoch": 2.4642464246424645, "step": 3360}, {"loss": 0.8409, "grad_norm": 0.4769273102283478, "learning_rate": 0.0002, "epoch": 2.4715804913824715, "step": 3370}, {"loss": 0.9312, "grad_norm": 0.5608189702033997, "learning_rate": 0.0002, "epoch": 2.478914558122479, "step": 3380}, {"loss": 0.9934, "grad_norm": 0.5590165853500366, "learning_rate": 0.0002, "epoch": 2.4862486248624864, "step": 3390}, {"loss": 1.025, "grad_norm": 0.801306962966919, "learning_rate": 0.0002, "epoch": 2.4935826916024935, "step": 3400}, {"loss": 0.9049, "grad_norm": 0.6045624613761902, "learning_rate": 0.0002, "epoch": 2.500916758342501, "step": 3410}, {"loss": 0.944, "grad_norm": 0.5735858082771301, "learning_rate": 0.0002, "epoch": 2.5082508250825084, "step": 3420}, {"loss": 0.9846, "grad_norm": 0.6827309131622314, "learning_rate": 0.0002, "epoch": 2.5155848918225154, "step": 3430}, {"loss": 0.9789, "grad_norm": 0.5702602863311768, "learning_rate": 0.0002, "epoch": 2.522918958562523, "step": 3440}, {"loss": 0.9127, "grad_norm": 0.6674721240997314, "learning_rate": 0.0002, "epoch": 2.5302530253025304, "step": 3450}, {"loss": 0.914, "grad_norm": 0.5635907649993896, "learning_rate": 0.0002, "epoch": 2.5375870920425374, "step": 3460}, {"loss": 0.8398, "grad_norm": 0.42737770080566406, "learning_rate": 0.0002, "epoch": 2.544921158782545, "step": 3470}, {"loss": 0.9474, "grad_norm": 0.6720691919326782, "learning_rate": 0.0002, "epoch": 2.5522552255225524, "step": 3480}, {"loss": 0.8637, "grad_norm": 0.8917084336280823, "learning_rate": 0.0002, "epoch": 2.55958929226256, "step": 3490}, {"loss": 0.9257, "grad_norm": 0.5134549140930176, "learning_rate": 0.0002, "epoch": 2.566923359002567, "step": 3500}, {"loss": 0.9362, "grad_norm": 0.4951367974281311, "learning_rate": 0.0002, "epoch": 2.5742574257425743, "step": 3510}, {"loss": 0.9184, "grad_norm": 0.9438204765319824, "learning_rate": 0.0002, "epoch": 2.5815914924825814, "step": 3520}, {"loss": 0.8939, "grad_norm": 0.6024714708328247, "learning_rate": 0.0002, "epoch": 2.588925559222589, "step": 3530}, {"loss": 0.9298, "grad_norm": 0.5248535871505737, "learning_rate": 0.0002, "epoch": 2.5962596259625963, "step": 3540}, {"loss": 0.941, "grad_norm": 0.8677568435668945, "learning_rate": 0.0002, "epoch": 2.6035936927026038, "step": 3550}, {"loss": 0.9253, "grad_norm": 0.82008296251297, "learning_rate": 0.0002, "epoch": 2.610927759442611, "step": 3560}, {"loss": 0.8429, "grad_norm": 0.4724634885787964, "learning_rate": 0.0002, "epoch": 2.6182618261826183, "step": 3570}, {"loss": 0.9058, "grad_norm": 0.5434244275093079, "learning_rate": 0.0002, "epoch": 2.6255958929226257, "step": 3580}, {"loss": 0.9379, "grad_norm": 0.4948740005493164, "learning_rate": 0.0002, "epoch": 2.6329299596626328, "step": 3590}, {"loss": 0.8718, "grad_norm": 0.42109328508377075, "learning_rate": 0.0002, "epoch": 2.6402640264026402, "step": 3600}, {"loss": 0.9809, "grad_norm": 0.7979786396026611, "learning_rate": 0.0002, "epoch": 2.6475980931426477, "step": 3610}, {"loss": 0.9229, "grad_norm": 0.6345919370651245, "learning_rate": 0.0002, "epoch": 2.654932159882655, "step": 3620}, {"loss": 0.8506, "grad_norm": 0.4971671402454376, "learning_rate": 0.0002, "epoch": 2.662266226622662, "step": 3630}, {"loss": 0.8054, "grad_norm": 0.6467748284339905, "learning_rate": 0.0002, "epoch": 2.6696002933626697, "step": 3640}, {"loss": 0.9277, "grad_norm": 0.4240160286426544, "learning_rate": 0.0002, "epoch": 2.6769343601026767, "step": 3650}, {"loss": 0.8213, "grad_norm": 0.5179754495620728, "learning_rate": 0.0002, "epoch": 2.684268426842684, "step": 3660}, {"loss": 0.9221, "grad_norm": 0.754012405872345, "learning_rate": 0.0002, "epoch": 2.6916024935826917, "step": 3670}, {"loss": 0.9194, "grad_norm": 0.5141299962997437, "learning_rate": 0.0002, "epoch": 2.698936560322699, "step": 3680}, {"loss": 0.9495, "grad_norm": 0.5737819075584412, "learning_rate": 0.0002, "epoch": 2.706270627062706, "step": 3690}, {"loss": 1.0162, "grad_norm": 0.5887577533721924, "learning_rate": 0.0002, "epoch": 2.7136046938027136, "step": 3700}, {"loss": 0.9169, "grad_norm": 0.6740471720695496, "learning_rate": 0.0002, "epoch": 2.720938760542721, "step": 3710}, {"loss": 0.9297, "grad_norm": 0.5879453420639038, "learning_rate": 0.0002, "epoch": 2.728272827282728, "step": 3720}, {"loss": 0.9358, "grad_norm": 0.4858354926109314, "learning_rate": 0.0002, "epoch": 2.7356068940227356, "step": 3730}, {"loss": 0.9308, "grad_norm": 0.5489001870155334, "learning_rate": 0.0002, "epoch": 2.742940960762743, "step": 3740}, {"loss": 0.894, "grad_norm": 0.8187092542648315, "learning_rate": 0.0002, "epoch": 2.7502750275027505, "step": 3750}, {"loss": 0.8954, "grad_norm": 0.5666626691818237, "learning_rate": 0.0002, "epoch": 2.7576090942427576, "step": 3760}, {"loss": 1.0059, "grad_norm": 0.5377066135406494, "learning_rate": 0.0002, "epoch": 2.764943160982765, "step": 3770}, {"loss": 0.9132, "grad_norm": 0.566330075263977, "learning_rate": 0.0002, "epoch": 2.772277227722772, "step": 3780}, {"loss": 0.9415, "grad_norm": 0.5522832870483398, "learning_rate": 0.0002, "epoch": 2.7796112944627795, "step": 3790}, {"loss": 0.8816, "grad_norm": 0.5668695569038391, "learning_rate": 0.0002, "epoch": 2.786945361202787, "step": 3800}, {"loss": 0.8885, "grad_norm": 0.7566602826118469, "learning_rate": 0.0002, "epoch": 2.7942794279427945, "step": 3810}, {"loss": 0.8598, "grad_norm": 0.5603684782981873, "learning_rate": 0.0002, "epoch": 2.8016134946828015, "step": 3820}, {"loss": 0.9602, "grad_norm": 0.49122217297554016, "learning_rate": 0.0002, "epoch": 2.808947561422809, "step": 3830}, {"loss": 0.9738, "grad_norm": 0.6798251867294312, "learning_rate": 0.0002, "epoch": 2.816281628162816, "step": 3840}, {"loss": 0.9533, "grad_norm": 0.6097991466522217, "learning_rate": 0.0002, "epoch": 2.8236156949028235, "step": 3850}, {"loss": 0.8672, "grad_norm": 0.6675726175308228, "learning_rate": 0.0002, "epoch": 2.830949761642831, "step": 3860}, {"loss": 0.9324, "grad_norm": 0.9223952889442444, "learning_rate": 0.0002, "epoch": 2.8382838283828384, "step": 3870}, {"loss": 0.8767, "grad_norm": 0.6020799875259399, "learning_rate": 0.0002, "epoch": 2.8456178951228455, "step": 3880}, {"loss": 0.9148, "grad_norm": 0.5206381678581238, "learning_rate": 0.0002, "epoch": 2.852951961862853, "step": 3890}, {"loss": 0.9479, "grad_norm": 0.6268777251243591, "learning_rate": 0.0002, "epoch": 2.8602860286028604, "step": 3900}, {"loss": 0.9409, "grad_norm": 1.1583497524261475, "learning_rate": 0.0002, "epoch": 2.8676200953428674, "step": 3910}, {"loss": 0.895, "grad_norm": 0.7263903021812439, "learning_rate": 0.0002, "epoch": 2.874954162082875, "step": 3920}, {"loss": 0.8786, "grad_norm": 0.5369910001754761, "learning_rate": 0.0002, "epoch": 2.8822882288228824, "step": 3930}, {"loss": 1.0015, "grad_norm": 0.7298350930213928, "learning_rate": 0.0002, "epoch": 2.88962229556289, "step": 3940}, {"loss": 0.979, "grad_norm": 0.577012836933136, "learning_rate": 0.0002, "epoch": 2.896956362302897, "step": 3950}, {"loss": 0.9716, "grad_norm": 0.5859594345092773, "learning_rate": 0.0002, "epoch": 2.9042904290429044, "step": 3960}, {"loss": 0.8772, "grad_norm": 0.47176122665405273, "learning_rate": 0.0002, "epoch": 2.9116244957829114, "step": 3970}, {"loss": 0.8997, "grad_norm": 0.9699620604515076, "learning_rate": 0.0002, "epoch": 2.918958562522919, "step": 3980}, {"loss": 0.9057, "grad_norm": 0.7908747792243958, "learning_rate": 0.0002, "epoch": 2.9262926292629263, "step": 3990}, {"loss": 0.9462, "grad_norm": 0.5777379274368286, "learning_rate": 0.0002, "epoch": 2.933626696002934, "step": 4000}, {"loss": 0.9358, "grad_norm": 0.599288284778595, "learning_rate": 0.0002, "epoch": 2.940960762742941, "step": 4010}, {"loss": 0.9812, "grad_norm": 0.5232274532318115, "learning_rate": 0.0002, "epoch": 2.9482948294829483, "step": 4020}, {"loss": 0.96, "grad_norm": 0.6395137310028076, "learning_rate": 0.0002, "epoch": 2.9556288962229558, "step": 4030}, {"loss": 0.9813, "grad_norm": 0.589260458946228, "learning_rate": 0.0002, "epoch": 2.962962962962963, "step": 4040}, {"loss": 0.9541, "grad_norm": 0.5699581503868103, "learning_rate": 0.0002, "epoch": 2.9702970297029703, "step": 4050}, {"loss": 0.9585, "grad_norm": 0.528468132019043, "learning_rate": 0.0002, "epoch": 2.9776310964429777, "step": 4060}, {"loss": 0.9164, "grad_norm": 0.4804670512676239, "learning_rate": 0.0002, "epoch": 2.984965163182985, "step": 4070}, {"loss": 0.9771, "grad_norm": 1.1918889284133911, "learning_rate": 0.0002, "epoch": 2.9922992299229922, "step": 4080}, {"loss": 0.9178, "grad_norm": 0.5479103326797485, "learning_rate": 0.0002, "epoch": 2.9996332966629997, "step": 4090}, {"eval_loss": 1.1642853021621704, "eval_runtime": 32.7511, "eval_samples_per_second": 13.16, "eval_steps_per_second": 1.649, "epoch": 2.9996332966629997, "step": 4090}, {"loss": 0.7981, "grad_norm": 0.7430027723312378, "learning_rate": 0.0002, "epoch": 3.006967363403007, "step": 4100}, {"loss": 0.7871, "grad_norm": 0.6293647289276123, "learning_rate": 0.0002, "epoch": 3.014301430143014, "step": 4110}, {"loss": 0.78, "grad_norm": 0.6191329956054688, "learning_rate": 0.0002, "epoch": 3.0216354968830217, "step": 4120}, {"loss": 0.7618, "grad_norm": 0.7959313988685608, "learning_rate": 0.0002, "epoch": 3.028969563623029, "step": 4130}, {"loss": 0.8039, "grad_norm": 0.5956351161003113, "learning_rate": 0.0002, "epoch": 3.036303630363036, "step": 4140}, {"loss": 0.7477, "grad_norm": 0.670383632183075, "learning_rate": 0.0002, "epoch": 3.0436376971030437, "step": 4150}, {"loss": 0.7984, "grad_norm": 0.6414518356323242, "learning_rate": 0.0002, "epoch": 3.050971763843051, "step": 4160}, {"loss": 0.7369, "grad_norm": 0.7928852438926697, "learning_rate": 0.0002, "epoch": 3.058305830583058, "step": 4170}, {"loss": 0.7914, "grad_norm": 0.6211121082305908, "learning_rate": 0.0002, "epoch": 3.0656398973230656, "step": 4180}, {"loss": 0.7365, "grad_norm": 0.6237057447433472, "learning_rate": 0.0002, "epoch": 3.072973964063073, "step": 4190}, {"loss": 0.702, "grad_norm": 0.6522233486175537, "learning_rate": 0.0002, "epoch": 3.08030803080308, "step": 4200}, {"loss": 0.7646, "grad_norm": 0.9396848678588867, "learning_rate": 0.0002, "epoch": 3.0876420975430876, "step": 4210}, {"loss": 0.7559, "grad_norm": 0.8003010749816895, "learning_rate": 0.0002, "epoch": 3.094976164283095, "step": 4220}, {"loss": 0.711, "grad_norm": 0.6733810305595398, "learning_rate": 0.0002, "epoch": 3.102310231023102, "step": 4230}, {"loss": 0.696, "grad_norm": 0.6365828514099121, "learning_rate": 0.0002, "epoch": 3.1096442977631096, "step": 4240}, {"loss": 0.8362, "grad_norm": 1.0805548429489136, "learning_rate": 0.0002, "epoch": 3.116978364503117, "step": 4250}, {"loss": 0.7651, "grad_norm": 0.7262141108512878, "learning_rate": 0.0002, "epoch": 3.1243124312431245, "step": 4260}, {"loss": 0.7304, "grad_norm": 0.5500539541244507, "learning_rate": 0.0002, "epoch": 3.1316464979831315, "step": 4270}, {"loss": 0.7721, "grad_norm": 0.793912947177887, "learning_rate": 0.0002, "epoch": 3.138980564723139, "step": 4280}, {"loss": 0.7708, "grad_norm": 1.2540518045425415, "learning_rate": 0.0002, "epoch": 3.1463146314631465, "step": 4290}, {"loss": 0.782, "grad_norm": 0.7020077705383301, "learning_rate": 0.0002, "epoch": 3.1536486982031535, "step": 4300}, {"loss": 0.7253, "grad_norm": 0.5111123323440552, "learning_rate": 0.0002, "epoch": 3.160982764943161, "step": 4310}, {"loss": 0.8159, "grad_norm": 0.7172090411186218, "learning_rate": 0.0002, "epoch": 3.1683168316831685, "step": 4320}, {"loss": 0.6962, "grad_norm": 0.6343168616294861, "learning_rate": 0.0002, "epoch": 3.1756508984231755, "step": 4330}, {"loss": 0.7938, "grad_norm": 0.9563672542572021, "learning_rate": 0.0002, "epoch": 3.182984965163183, "step": 4340}, {"loss": 0.7385, "grad_norm": 1.0225574970245361, "learning_rate": 0.0002, "epoch": 3.1903190319031904, "step": 4350}, {"loss": 0.8652, "grad_norm": 1.1633386611938477, "learning_rate": 0.0002, "epoch": 3.1976530986431975, "step": 4360}, {"loss": 0.7259, "grad_norm": 0.8915148973464966, "learning_rate": 0.0002, "epoch": 3.204987165383205, "step": 4370}, {"loss": 0.8061, "grad_norm": 0.9156812429428101, "learning_rate": 0.0002, "epoch": 3.2123212321232124, "step": 4380}, {"loss": 0.8189, "grad_norm": 0.6363258957862854, "learning_rate": 0.0002, "epoch": 3.21965529886322, "step": 4390}, {"loss": 0.7996, "grad_norm": 0.579099178314209, "learning_rate": 0.0002, "epoch": 3.226989365603227, "step": 4400}, {"loss": 0.8592, "grad_norm": 0.8778146505355835, "learning_rate": 0.0002, "epoch": 3.2343234323432344, "step": 4410}, {"loss": 0.8281, "grad_norm": 0.8356770873069763, "learning_rate": 0.0002, "epoch": 3.241657499083242, "step": 4420}, {"loss": 0.8484, "grad_norm": 0.702032208442688, "learning_rate": 0.0002, "epoch": 3.248991565823249, "step": 4430}, {"loss": 0.7227, "grad_norm": 0.6386539340019226, "learning_rate": 0.0002, "epoch": 3.2563256325632564, "step": 4440}, {"loss": 0.8374, "grad_norm": 0.7008408904075623, "learning_rate": 0.0002, "epoch": 3.263659699303264, "step": 4450}, {"loss": 0.7572, "grad_norm": 0.9556332230567932, "learning_rate": 0.0002, "epoch": 3.270993766043271, "step": 4460}, {"loss": 0.743, "grad_norm": 0.5667835474014282, "learning_rate": 0.0002, "epoch": 3.2783278327832783, "step": 4470}, {"loss": 0.8152, "grad_norm": 0.8239172697067261, "learning_rate": 0.0002, "epoch": 3.285661899523286, "step": 4480}, {"loss": 0.756, "grad_norm": 0.7045050859451294, "learning_rate": 0.0002, "epoch": 3.292995966263293, "step": 4490}, {"loss": 0.7655, "grad_norm": 0.7131434082984924, "learning_rate": 0.0002, "epoch": 3.3003300330033003, "step": 4500}, {"loss": 0.836, "grad_norm": 0.6924910545349121, "learning_rate": 0.0002, "epoch": 3.3076640997433078, "step": 4510}, {"loss": 0.736, "grad_norm": 0.8945356607437134, "learning_rate": 0.0002, "epoch": 3.3149981664833152, "step": 4520}, {"loss": 0.7575, "grad_norm": 0.6546903252601624, "learning_rate": 0.0002, "epoch": 3.3223322332233223, "step": 4530}, {"loss": 0.7893, "grad_norm": 0.8206679224967957, "learning_rate": 0.0002, "epoch": 3.3296662999633297, "step": 4540}, {"loss": 0.7502, "grad_norm": 0.6482203602790833, "learning_rate": 0.0002, "epoch": 3.3370003667033368, "step": 4550}, {"loss": 0.8172, "grad_norm": 0.7558760046958923, "learning_rate": 0.0002, "epoch": 3.3443344334433442, "step": 4560}, {"loss": 0.744, "grad_norm": 0.7794756889343262, "learning_rate": 0.0002, "epoch": 3.3516685001833517, "step": 4570}, {"loss": 0.7385, "grad_norm": 0.7382805943489075, "learning_rate": 0.0002, "epoch": 3.359002566923359, "step": 4580}, {"loss": 0.8511, "grad_norm": 0.5912511944770813, "learning_rate": 0.0002, "epoch": 3.366336633663366, "step": 4590}, {"loss": 0.8272, "grad_norm": 0.7444885969161987, "learning_rate": 0.0002, "epoch": 3.3736707004033737, "step": 4600}, {"loss": 0.7927, "grad_norm": 0.7354922890663147, "learning_rate": 0.0002, "epoch": 3.381004767143381, "step": 4610}, {"loss": 0.7183, "grad_norm": 0.7685934901237488, "learning_rate": 0.0002, "epoch": 3.388338833883388, "step": 4620}, {"loss": 0.7436, "grad_norm": 0.61041259765625, "learning_rate": 0.0002, "epoch": 3.3956729006233957, "step": 4630}, {"loss": 0.7661, "grad_norm": 0.6820451021194458, "learning_rate": 0.0002, "epoch": 3.403006967363403, "step": 4640}, {"loss": 0.8796, "grad_norm": 0.5819534063339233, "learning_rate": 0.0002, "epoch": 3.41034103410341, "step": 4650}, {"loss": 0.7314, "grad_norm": 0.705410897731781, "learning_rate": 0.0002, "epoch": 3.4176751008434176, "step": 4660}, {"loss": 0.7901, "grad_norm": 0.8052892088890076, "learning_rate": 0.0002, "epoch": 3.425009167583425, "step": 4670}, {"loss": 0.7298, "grad_norm": 0.7746483087539673, "learning_rate": 0.0002, "epoch": 3.432343234323432, "step": 4680}, {"loss": 0.7976, "grad_norm": 0.7713689804077148, "learning_rate": 0.0002, "epoch": 3.4396773010634396, "step": 4690}, {"loss": 0.7427, "grad_norm": 0.810371994972229, "learning_rate": 0.0002, "epoch": 3.447011367803447, "step": 4700}, {"loss": 0.7594, "grad_norm": 0.7702969312667847, "learning_rate": 0.0002, "epoch": 3.4543454345434546, "step": 4710}, {"loss": 0.7957, "grad_norm": 0.7069268822669983, "learning_rate": 0.0002, "epoch": 3.4616795012834616, "step": 4720}, {"loss": 0.8199, "grad_norm": 0.7640359401702881, "learning_rate": 0.0002, "epoch": 3.469013568023469, "step": 4730}, {"loss": 0.6875, "grad_norm": 0.8661707639694214, "learning_rate": 0.0002, "epoch": 3.4763476347634765, "step": 4740}, {"loss": 0.8528, "grad_norm": 0.9970282912254333, "learning_rate": 0.0002, "epoch": 3.4836817015034836, "step": 4750}, {"loss": 0.8462, "grad_norm": 0.5824355483055115, "learning_rate": 0.0002, "epoch": 3.491015768243491, "step": 4760}, {"loss": 0.851, "grad_norm": 1.3072649240493774, "learning_rate": 0.0002, "epoch": 3.4983498349834985, "step": 4770}, {"loss": 0.9101, "grad_norm": 0.873978316783905, "learning_rate": 0.0002, "epoch": 3.5056839017235055, "step": 4780}, {"loss": 0.7403, "grad_norm": 0.5526657104492188, "learning_rate": 0.0002, "epoch": 3.513017968463513, "step": 4790}, {"loss": 0.7921, "grad_norm": 0.790894627571106, "learning_rate": 0.0002, "epoch": 3.5203520352035205, "step": 4800}, {"loss": 0.831, "grad_norm": 0.8119630217552185, "learning_rate": 0.0002, "epoch": 3.5276861019435275, "step": 4810}, {"loss": 0.7351, "grad_norm": 0.633212149143219, "learning_rate": 0.0002, "epoch": 3.535020168683535, "step": 4820}, {"loss": 0.8505, "grad_norm": 0.703029990196228, "learning_rate": 0.0002, "epoch": 3.5423542354235424, "step": 4830}, {"loss": 0.7204, "grad_norm": 0.7603771686553955, "learning_rate": 0.0002, "epoch": 3.54968830216355, "step": 4840}, {"loss": 0.8868, "grad_norm": 0.6260480880737305, "learning_rate": 0.0002, "epoch": 3.557022368903557, "step": 4850}, {"loss": 0.8137, "grad_norm": 0.8203664422035217, "learning_rate": 0.0002, "epoch": 3.5643564356435644, "step": 4860}, {"loss": 0.8821, "grad_norm": 0.7793813347816467, "learning_rate": 0.0002, "epoch": 3.5716905023835714, "step": 4870}, {"loss": 0.8164, "grad_norm": 0.7667397260665894, "learning_rate": 0.0002, "epoch": 3.579024569123579, "step": 4880}, {"loss": 0.7597, "grad_norm": 0.8198829889297485, "learning_rate": 0.0002, "epoch": 3.5863586358635864, "step": 4890}, {"loss": 0.7027, "grad_norm": 0.7689233422279358, "learning_rate": 0.0002, "epoch": 3.593692702603594, "step": 4900}, {"loss": 0.804, "grad_norm": 0.7870983481407166, "learning_rate": 0.0002, "epoch": 3.601026769343601, "step": 4910}, {"loss": 0.8269, "grad_norm": 0.8133853077888489, "learning_rate": 0.0002, "epoch": 3.6083608360836084, "step": 4920}, {"loss": 0.8515, "grad_norm": 1.308401346206665, "learning_rate": 0.0002, "epoch": 3.615694902823616, "step": 4930}, {"loss": 0.8494, "grad_norm": 0.7131121754646301, "learning_rate": 0.0002, "epoch": 3.623028969563623, "step": 4940}, {"loss": 0.7235, "grad_norm": 0.6825910210609436, "learning_rate": 0.0002, "epoch": 3.6303630363036303, "step": 4950}, {"loss": 0.7824, "grad_norm": 0.7254678606987, "learning_rate": 0.0002, "epoch": 3.637697103043638, "step": 4960}, {"loss": 0.7983, "grad_norm": 0.8045085072517395, "learning_rate": 0.0002, "epoch": 3.6450311697836453, "step": 4970}, {"loss": 0.8223, "grad_norm": 0.6991777420043945, "learning_rate": 0.0002, "epoch": 3.6523652365236523, "step": 4980}, {"loss": 0.7806, "grad_norm": 0.7804713249206543, "learning_rate": 0.0002, "epoch": 3.6596993032636598, "step": 4990}, {"loss": 0.8402, "grad_norm": 0.8525708317756653, "learning_rate": 0.0002, "epoch": 3.667033370003667, "step": 5000}, {"loss": 0.8496, "grad_norm": 0.7959994673728943, "learning_rate": 0.0002, "epoch": 3.6743674367436743, "step": 5010}, {"loss": 0.8022, "grad_norm": 0.8103628158569336, "learning_rate": 0.0002, "epoch": 3.6817015034836817, "step": 5020}, {"loss": 0.7376, "grad_norm": 0.7517836093902588, "learning_rate": 0.0002, "epoch": 3.689035570223689, "step": 5030}, {"loss": 0.8375, "grad_norm": 0.6878514289855957, "learning_rate": 0.0002, "epoch": 3.6963696369636962, "step": 5040}, {"loss": 0.7998, "grad_norm": 1.2371820211410522, "learning_rate": 0.0002, "epoch": 3.7037037037037037, "step": 5050}, {"loss": 0.6941, "grad_norm": 0.6567103862762451, "learning_rate": 0.0002, "epoch": 3.711037770443711, "step": 5060}, {"loss": 0.8465, "grad_norm": 1.1254922151565552, "learning_rate": 0.0002, "epoch": 3.718371837183718, "step": 5070}, {"loss": 0.8365, "grad_norm": 0.6796132326126099, "learning_rate": 0.0002, "epoch": 3.7257059039237257, "step": 5080}, {"loss": 0.7818, "grad_norm": 0.7285300493240356, "learning_rate": 0.0002, "epoch": 3.733039970663733, "step": 5090}, {"loss": 0.8581, "grad_norm": 0.8931500911712646, "learning_rate": 0.0002, "epoch": 3.7403740374037406, "step": 5100}, {"loss": 0.8181, "grad_norm": 0.6256856918334961, "learning_rate": 0.0002, "epoch": 3.7477081041437477, "step": 5110}, {"loss": 0.743, "grad_norm": 0.79310142993927, "learning_rate": 0.0002, "epoch": 3.755042170883755, "step": 5120}, {"loss": 0.8235, "grad_norm": 0.6594041585922241, "learning_rate": 0.0002, "epoch": 3.762376237623762, "step": 5130}, {"loss": 0.6925, "grad_norm": 0.7029327750205994, "learning_rate": 0.0002, "epoch": 3.7697103043637696, "step": 5140}, {"loss": 0.7457, "grad_norm": 0.5880070328712463, "learning_rate": 0.0002, "epoch": 3.777044371103777, "step": 5150}, {"loss": 0.8716, "grad_norm": 0.7578945159912109, "learning_rate": 0.0002, "epoch": 3.7843784378437846, "step": 5160}, {"loss": 0.8819, "grad_norm": 0.8276378512382507, "learning_rate": 0.0002, "epoch": 3.7917125045837916, "step": 5170}, {"loss": 0.7559, "grad_norm": 0.7627953886985779, "learning_rate": 0.0002, "epoch": 3.799046571323799, "step": 5180}, {"loss": 0.7665, "grad_norm": 0.8169086575508118, "learning_rate": 0.0002, "epoch": 3.806380638063806, "step": 5190}, {"loss": 0.761, "grad_norm": 0.6605030298233032, "learning_rate": 0.0002, "epoch": 3.8137147048038136, "step": 5200}, {"loss": 0.8804, "grad_norm": 0.5837286114692688, "learning_rate": 0.0002, "epoch": 3.821048771543821, "step": 5210}, {"loss": 0.8369, "grad_norm": 1.2422157526016235, "learning_rate": 0.0002, "epoch": 3.8283828382838285, "step": 5220}, {"loss": 0.8431, "grad_norm": 0.6589220762252808, "learning_rate": 0.0002, "epoch": 3.8357169050238356, "step": 5230}, {"loss": 0.7686, "grad_norm": 0.8567556142807007, "learning_rate": 0.0002, "epoch": 3.843050971763843, "step": 5240}, {"loss": 0.8652, "grad_norm": 0.6490627527236938, "learning_rate": 0.0002, "epoch": 3.8503850385038505, "step": 5250}, {"loss": 0.7386, "grad_norm": 0.620232880115509, "learning_rate": 0.0002, "epoch": 3.8577191052438575, "step": 5260}, {"loss": 0.9192, "grad_norm": 0.7685128450393677, "learning_rate": 0.0002, "epoch": 3.865053171983865, "step": 5270}, {"loss": 0.872, "grad_norm": 0.8113296627998352, "learning_rate": 0.0002, "epoch": 3.8723872387238725, "step": 5280}, {"loss": 0.7156, "grad_norm": 0.8092675805091858, "learning_rate": 0.0002, "epoch": 3.87972130546388, "step": 5290}, {"loss": 0.7325, "grad_norm": 0.583570122718811, "learning_rate": 0.0002, "epoch": 3.887055372203887, "step": 5300}, {"loss": 0.9333, "grad_norm": 1.712363600730896, "learning_rate": 0.0002, "epoch": 3.8943894389438944, "step": 5310}, {"loss": 0.7537, "grad_norm": 0.6673534512519836, "learning_rate": 0.0002, "epoch": 3.9017235056839015, "step": 5320}, {"loss": 0.7035, "grad_norm": 1.9770312309265137, "learning_rate": 0.0002, "epoch": 3.909057572423909, "step": 5330}, {"loss": 0.8793, "grad_norm": 0.6430999636650085, "learning_rate": 0.0002, "epoch": 3.9163916391639164, "step": 5340}, {"loss": 0.839, "grad_norm": 1.0159571170806885, "learning_rate": 0.0002, "epoch": 3.923725705903924, "step": 5350}, {"loss": 0.9332, "grad_norm": 0.8607584834098816, "learning_rate": 0.0002, "epoch": 3.931059772643931, "step": 5360}, {"loss": 0.7261, "grad_norm": 0.6967900991439819, "learning_rate": 0.0002, "epoch": 3.9383938393839384, "step": 5370}, {"loss": 0.8456, "grad_norm": 0.7683077454566956, "learning_rate": 0.0002, "epoch": 3.945727906123946, "step": 5380}, {"loss": 0.7682, "grad_norm": 0.6805762648582458, "learning_rate": 0.0002, "epoch": 3.953061972863953, "step": 5390}, {"loss": 0.7746, "grad_norm": 0.7033619284629822, "learning_rate": 0.0002, "epoch": 3.9603960396039604, "step": 5400}, {"loss": 0.8393, "grad_norm": 0.966112494468689, "learning_rate": 0.0002, "epoch": 3.967730106343968, "step": 5410}, {"loss": 0.8316, "grad_norm": 0.8467881083488464, "learning_rate": 0.0002, "epoch": 3.9750641730839753, "step": 5420}, {"loss": 0.8084, "grad_norm": 0.8005317449569702, "learning_rate": 0.0002, "epoch": 3.9823982398239823, "step": 5430}, {"loss": 0.7168, "grad_norm": 1.1615241765975952, "learning_rate": 0.0002, "epoch": 3.98973230656399, "step": 5440}, {"loss": 0.8263, "grad_norm": 0.6121614575386047, "learning_rate": 0.0002, "epoch": 3.997066373303997, "step": 5450}, {"eval_loss": 1.1834222078323364, "eval_runtime": 32.7569, "eval_samples_per_second": 13.158, "eval_steps_per_second": 1.649, "epoch": 4.0, "step": 5454}, {"loss": 0.7267, "grad_norm": 0.6055727005004883, "learning_rate": 0.0002, "epoch": 4.004400440044004, "step": 5460}, {"loss": 0.5766, "grad_norm": 0.8232647180557251, "learning_rate": 0.0002, "epoch": 4.011734506784012, "step": 5470}, {"loss": 0.6489, "grad_norm": 0.7739192247390747, "learning_rate": 0.0002, "epoch": 4.019068573524019, "step": 5480}, {"loss": 0.5978, "grad_norm": 0.6264950633049011, "learning_rate": 0.0002, "epoch": 4.026402640264027, "step": 5490}, {"loss": 0.6392, "grad_norm": 1.4798702001571655, "learning_rate": 0.0002, "epoch": 4.033736707004033, "step": 5500}, {"loss": 0.6143, "grad_norm": 0.9538470506668091, "learning_rate": 0.0002, "epoch": 4.041070773744041, "step": 5510}, {"loss": 0.6056, "grad_norm": 0.834561288356781, "learning_rate": 0.0002, "epoch": 4.048404840484048, "step": 5520}, {"loss": 0.6077, "grad_norm": 0.6407850384712219, "learning_rate": 0.0002, "epoch": 4.055738907224056, "step": 5530}, {"loss": 0.6733, "grad_norm": 0.9035961627960205, "learning_rate": 0.0002, "epoch": 4.063072973964063, "step": 5540}, {"loss": 0.5854, "grad_norm": 0.842812716960907, "learning_rate": 0.0002, "epoch": 4.070407040704071, "step": 5550}, {"loss": 0.654, "grad_norm": 0.8197882175445557, "learning_rate": 0.0002, "epoch": 4.077741107444078, "step": 5560}, {"loss": 0.5919, "grad_norm": 0.8652673959732056, "learning_rate": 0.0002, "epoch": 4.085075174184085, "step": 5570}, {"loss": 0.6188, "grad_norm": 0.8048318028450012, "learning_rate": 0.0002, "epoch": 4.092409240924092, "step": 5580}, {"loss": 0.6487, "grad_norm": 0.9604969024658203, "learning_rate": 0.0002, "epoch": 4.0997433076641, "step": 5590}, {"loss": 0.6356, "grad_norm": 1.244756817817688, "learning_rate": 0.0002, "epoch": 4.107077374404107, "step": 5600}, {"loss": 0.6489, "grad_norm": 0.7975269556045532, "learning_rate": 0.0002, "epoch": 4.114411441144115, "step": 5610}, {"loss": 0.6445, "grad_norm": 0.6130099296569824, "learning_rate": 0.0002, "epoch": 4.121745507884122, "step": 5620}, {"loss": 0.6024, "grad_norm": 0.7793202996253967, "learning_rate": 0.0002, "epoch": 4.129079574624129, "step": 5630}, {"loss": 0.5723, "grad_norm": 1.187238335609436, "learning_rate": 0.0002, "epoch": 4.136413641364136, "step": 5640}, {"loss": 0.6385, "grad_norm": 0.8450375199317932, "learning_rate": 0.0002, "epoch": 4.143747708104144, "step": 5650}, {"loss": 0.6866, "grad_norm": 0.9006940126419067, "learning_rate": 0.0002, "epoch": 4.151081774844151, "step": 5660}, {"loss": 0.6179, "grad_norm": 0.9447154998779297, "learning_rate": 0.0002, "epoch": 4.158415841584159, "step": 5670}, {"loss": 0.6476, "grad_norm": 0.798032283782959, "learning_rate": 0.0002, "epoch": 4.165749908324166, "step": 5680}, {"loss": 0.6666, "grad_norm": 0.65578693151474, "learning_rate": 0.0002, "epoch": 4.1730839750641735, "step": 5690}, {"loss": 0.701, "grad_norm": 1.0864700078964233, "learning_rate": 0.0002, "epoch": 4.18041804180418, "step": 5700}, {"loss": 0.6895, "grad_norm": 0.7344121932983398, "learning_rate": 0.0002, "epoch": 4.187752108544188, "step": 5710}, {"loss": 0.6659, "grad_norm": 0.9722456932067871, "learning_rate": 0.0002, "epoch": 4.195086175284195, "step": 5720}, {"loss": 0.6887, "grad_norm": 1.263814926147461, "learning_rate": 0.0002, "epoch": 4.2024202420242025, "step": 5730}, {"loss": 0.608, "grad_norm": 0.9622581005096436, "learning_rate": 0.0002, "epoch": 4.20975430876421, "step": 5740}, {"loss": 0.6221, "grad_norm": 0.8497143387794495, "learning_rate": 0.0002, "epoch": 4.2170883755042174, "step": 5750}, {"loss": 0.6322, "grad_norm": 0.8248446583747864, "learning_rate": 0.0002, "epoch": 4.224422442244224, "step": 5760}, {"loss": 0.6045, "grad_norm": 1.2544798851013184, "learning_rate": 0.0002, "epoch": 4.2317565089842315, "step": 5770}, {"loss": 0.641, "grad_norm": 0.8224676251411438, "learning_rate": 0.0002, "epoch": 4.239090575724239, "step": 5780}, {"loss": 0.6399, "grad_norm": 0.8924877047538757, "learning_rate": 0.0002, "epoch": 4.2464246424642464, "step": 5790}, {"loss": 0.6845, "grad_norm": 0.8545848727226257, "learning_rate": 0.0002, "epoch": 4.253758709204254, "step": 5800}, {"loss": 0.6669, "grad_norm": 0.8081067800521851, "learning_rate": 0.0002, "epoch": 4.261092775944261, "step": 5810}, {"loss": 0.6149, "grad_norm": 0.7111002802848816, "learning_rate": 0.0002, "epoch": 4.268426842684269, "step": 5820}, {"loss": 0.6343, "grad_norm": 0.8696979880332947, "learning_rate": 0.0002, "epoch": 4.2757609094242754, "step": 5830}, {"loss": 0.6384, "grad_norm": 0.821401834487915, "learning_rate": 0.0002, "epoch": 4.283094976164283, "step": 5840}, {"loss": 0.6912, "grad_norm": 0.888908326625824, "learning_rate": 0.0002, "epoch": 4.29042904290429, "step": 5850}, {"loss": 0.6061, "grad_norm": 1.9380123615264893, "learning_rate": 0.0002, "epoch": 4.297763109644298, "step": 5860}, {"loss": 0.6766, "grad_norm": 1.121774435043335, "learning_rate": 0.0002, "epoch": 4.305097176384305, "step": 5870}, {"loss": 0.7205, "grad_norm": 0.9238282442092896, "learning_rate": 0.0002, "epoch": 4.312431243124313, "step": 5880}, {"loss": 0.6351, "grad_norm": 0.7321620583534241, "learning_rate": 0.0002, "epoch": 4.319765309864319, "step": 5890}, {"loss": 0.6404, "grad_norm": 0.8739548325538635, "learning_rate": 0.0002, "epoch": 4.327099376604327, "step": 5900}, {"loss": 0.5892, "grad_norm": 0.9686012268066406, "learning_rate": 0.0002, "epoch": 4.334433443344334, "step": 5910}, {"loss": 0.641, "grad_norm": 0.9033839106559753, "learning_rate": 0.0002, "epoch": 4.341767510084342, "step": 5920}, {"loss": 0.6456, "grad_norm": 0.8131115436553955, "learning_rate": 0.0002, "epoch": 4.349101576824349, "step": 5930}, {"loss": 0.5826, "grad_norm": 0.8942412734031677, "learning_rate": 0.0002, "epoch": 4.356435643564357, "step": 5940}, {"loss": 0.7336, "grad_norm": 0.8439112901687622, "learning_rate": 0.0002, "epoch": 4.363769710304364, "step": 5950}, {"loss": 0.6537, "grad_norm": 0.9176713228225708, "learning_rate": 0.0002, "epoch": 4.371103777044371, "step": 5960}, {"loss": 0.6792, "grad_norm": 0.6799634695053101, "learning_rate": 0.0002, "epoch": 4.378437843784378, "step": 5970}, {"loss": 0.7266, "grad_norm": 1.0435824394226074, "learning_rate": 0.0002, "epoch": 4.385771910524386, "step": 5980}, {"loss": 0.68, "grad_norm": 0.997937798500061, "learning_rate": 0.0002, "epoch": 4.393105977264393, "step": 5990}, {"loss": 0.6604, "grad_norm": 1.0308842658996582, "learning_rate": 0.0002, "epoch": 4.400440044004401, "step": 6000}, {"loss": 0.6402, "grad_norm": 1.3683775663375854, "learning_rate": 0.0002, "epoch": 4.407774110744408, "step": 6010}, {"loss": 0.7027, "grad_norm": 0.7569534182548523, "learning_rate": 0.0002, "epoch": 4.415108177484415, "step": 6020}, {"loss": 0.5949, "grad_norm": 1.089978575706482, "learning_rate": 0.0002, "epoch": 4.422442244224422, "step": 6030}, {"loss": 0.6353, "grad_norm": 0.7522459626197815, "learning_rate": 0.0002, "epoch": 4.42977631096443, "step": 6040}, {"loss": 0.5852, "grad_norm": 0.6709823608398438, "learning_rate": 0.0002, "epoch": 4.437110377704437, "step": 6050}, {"loss": 0.6718, "grad_norm": 0.6992089748382568, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 6060}, {"loss": 0.6933, "grad_norm": 1.0182931423187256, "learning_rate": 0.0002, "epoch": 4.451778511184452, "step": 6070}, {"loss": 0.6255, "grad_norm": 1.0685160160064697, "learning_rate": 0.0002, "epoch": 4.459112577924459, "step": 6080}, {"loss": 0.6086, "grad_norm": 0.8295124769210815, "learning_rate": 0.0002, "epoch": 4.466446644664466, "step": 6090}, {"loss": 0.6359, "grad_norm": 1.1862998008728027, "learning_rate": 0.0002, "epoch": 4.473780711404474, "step": 6100}, {"loss": 0.638, "grad_norm": 0.7400273084640503, "learning_rate": 0.0002, "epoch": 4.481114778144481, "step": 6110}, {"loss": 0.6854, "grad_norm": 0.7098417282104492, "learning_rate": 0.0002, "epoch": 4.488448844884489, "step": 6120}, {"loss": 0.6976, "grad_norm": 0.9745053648948669, "learning_rate": 0.0002, "epoch": 4.495782911624496, "step": 6130}, {"loss": 0.605, "grad_norm": 0.8638797998428345, "learning_rate": 0.0002, "epoch": 4.503116978364503, "step": 6140}, {"loss": 0.6491, "grad_norm": 0.8291046619415283, "learning_rate": 0.0002, "epoch": 4.51045104510451, "step": 6150}, {"loss": 0.6457, "grad_norm": 1.0301737785339355, "learning_rate": 0.0002, "epoch": 4.517785111844518, "step": 6160}, {"loss": 0.6742, "grad_norm": 1.1996512413024902, "learning_rate": 0.0002, "epoch": 4.525119178584525, "step": 6170}, {"loss": 0.6484, "grad_norm": 1.151038408279419, "learning_rate": 0.0002, "epoch": 4.5324532453245325, "step": 6180}, {"loss": 0.668, "grad_norm": 0.8385201096534729, "learning_rate": 0.0002, "epoch": 4.53978731206454, "step": 6190}, {"loss": 0.6381, "grad_norm": 0.8969188332557678, "learning_rate": 0.0002, "epoch": 4.5471213788045475, "step": 6200}, {"loss": 0.7141, "grad_norm": 1.60659658908844, "learning_rate": 0.0002, "epoch": 4.554455445544555, "step": 6210}, {"loss": 0.6388, "grad_norm": 0.9356731176376343, "learning_rate": 0.0002, "epoch": 4.5617895122845615, "step": 6220}, {"loss": 0.7393, "grad_norm": 0.95856773853302, "learning_rate": 0.0002, "epoch": 4.569123579024569, "step": 6230}, {"loss": 0.6554, "grad_norm": 1.1162524223327637, "learning_rate": 0.0002, "epoch": 4.5764576457645765, "step": 6240}, {"loss": 0.6012, "grad_norm": 0.8809238076210022, "learning_rate": 0.0002, "epoch": 4.583791712504584, "step": 6250}, {"loss": 0.648, "grad_norm": 0.890738844871521, "learning_rate": 0.0002, "epoch": 4.591125779244591, "step": 6260}, {"loss": 0.6663, "grad_norm": 0.918684720993042, "learning_rate": 0.0002, "epoch": 4.598459845984598, "step": 6270}, {"loss": 0.5992, "grad_norm": 0.8156296610832214, "learning_rate": 0.0002, "epoch": 4.6057939127246055, "step": 6280}, {"loss": 0.723, "grad_norm": 1.046634316444397, "learning_rate": 0.0002, "epoch": 4.613127979464613, "step": 6290}, {"loss": 0.7023, "grad_norm": 0.7725525498390198, "learning_rate": 0.0002, "epoch": 4.62046204620462, "step": 6300}, {"loss": 0.6414, "grad_norm": 0.9992046356201172, "learning_rate": 0.0002, "epoch": 4.627796112944628, "step": 6310}, {"loss": 0.6201, "grad_norm": 0.8480095267295837, "learning_rate": 0.0002, "epoch": 4.635130179684635, "step": 6320}, {"loss": 0.6869, "grad_norm": 0.7061955332756042, "learning_rate": 0.0002, "epoch": 4.642464246424643, "step": 6330}, {"loss": 0.6828, "grad_norm": 1.0354212522506714, "learning_rate": 0.0002, "epoch": 4.649798313164649, "step": 6340}, {"loss": 0.6651, "grad_norm": 1.0081377029418945, "learning_rate": 0.0002, "epoch": 4.657132379904657, "step": 6350}, {"loss": 0.726, "grad_norm": 1.2904249429702759, "learning_rate": 0.0002, "epoch": 4.664466446644664, "step": 6360}, {"loss": 0.7148, "grad_norm": 0.9248910546302795, "learning_rate": 0.0002, "epoch": 4.671800513384672, "step": 6370}, {"loss": 0.6961, "grad_norm": 0.9907804131507874, "learning_rate": 0.0002, "epoch": 4.679134580124679, "step": 6380}, {"loss": 0.6163, "grad_norm": 1.201143741607666, "learning_rate": 0.0002, "epoch": 4.686468646864687, "step": 6390}, {"loss": 0.6762, "grad_norm": 0.8709394335746765, "learning_rate": 0.0002, "epoch": 4.693802713604693, "step": 6400}, {"loss": 0.7217, "grad_norm": 0.7468608021736145, "learning_rate": 0.0002, "epoch": 4.701136780344701, "step": 6410}, {"loss": 0.6548, "grad_norm": 0.8607903718948364, "learning_rate": 0.0002, "epoch": 4.708470847084708, "step": 6420}, {"loss": 0.6449, "grad_norm": 0.9840512871742249, "learning_rate": 0.0002, "epoch": 4.715804913824716, "step": 6430}, {"loss": 0.685, "grad_norm": 0.8328204154968262, "learning_rate": 0.0002, "epoch": 4.723138980564723, "step": 6440}, {"loss": 0.697, "grad_norm": 0.924505352973938, "learning_rate": 0.0002, "epoch": 4.730473047304731, "step": 6450}, {"loss": 0.7422, "grad_norm": 0.8897685408592224, "learning_rate": 0.0002, "epoch": 4.737807114044738, "step": 6460}, {"loss": 0.6842, "grad_norm": 0.9605024456977844, "learning_rate": 0.0002, "epoch": 4.745141180784745, "step": 6470}, {"loss": 0.6488, "grad_norm": 0.8150759935379028, "learning_rate": 0.0002, "epoch": 4.752475247524752, "step": 6480}, {"loss": 0.6606, "grad_norm": 0.8128412961959839, "learning_rate": 0.0002, "epoch": 4.75980931426476, "step": 6490}, {"loss": 0.6729, "grad_norm": 0.7381404638290405, "learning_rate": 0.0002, "epoch": 4.767143381004767, "step": 6500}, {"loss": 0.6713, "grad_norm": 1.0565853118896484, "learning_rate": 0.0002, "epoch": 4.774477447744775, "step": 6510}, {"loss": 0.6496, "grad_norm": 0.9298134446144104, "learning_rate": 0.0002, "epoch": 4.781811514484782, "step": 6520}, {"loss": 0.7279, "grad_norm": 1.0145525932312012, "learning_rate": 0.0002, "epoch": 4.789145581224789, "step": 6530}, {"loss": 0.5986, "grad_norm": 0.92259681224823, "learning_rate": 0.0002, "epoch": 4.796479647964796, "step": 6540}, {"loss": 0.63, "grad_norm": 0.7881024479866028, "learning_rate": 0.0002, "epoch": 4.803813714704804, "step": 6550}, {"loss": 0.7134, "grad_norm": 1.4935206174850464, "learning_rate": 0.0002, "epoch": 4.811147781444811, "step": 6560}, {"loss": 0.6695, "grad_norm": 0.8612369298934937, "learning_rate": 0.0002, "epoch": 4.818481848184819, "step": 6570}, {"loss": 0.779, "grad_norm": 1.0118653774261475, "learning_rate": 0.0002, "epoch": 4.825815914924826, "step": 6580}, {"loss": 0.6991, "grad_norm": 1.1303809881210327, "learning_rate": 0.0002, "epoch": 4.833149981664834, "step": 6590}, {"loss": 0.7887, "grad_norm": 0.9112492203712463, "learning_rate": 0.0002, "epoch": 4.84048404840484, "step": 6600}, {"loss": 0.7699, "grad_norm": 0.864762544631958, "learning_rate": 0.0002, "epoch": 4.847818115144848, "step": 6610}, {"loss": 0.7347, "grad_norm": 0.9090572595596313, "learning_rate": 0.0002, "epoch": 4.855152181884855, "step": 6620}, {"loss": 0.6608, "grad_norm": 1.014953374862671, "learning_rate": 0.0002, "epoch": 4.862486248624863, "step": 6630}, {"loss": 0.6429, "grad_norm": 1.0702149868011475, "learning_rate": 0.0002, "epoch": 4.86982031536487, "step": 6640}, {"loss": 0.6943, "grad_norm": 1.002135157585144, "learning_rate": 0.0002, "epoch": 4.8771543821048775, "step": 6650}, {"loss": 0.7225, "grad_norm": 0.862545907497406, "learning_rate": 0.0002, "epoch": 4.884488448844884, "step": 6660}, {"loss": 0.6206, "grad_norm": 0.7302131056785583, "learning_rate": 0.0002, "epoch": 4.891822515584892, "step": 6670}, {"loss": 0.7175, "grad_norm": 0.8380730152130127, "learning_rate": 0.0002, "epoch": 4.899156582324899, "step": 6680}, {"loss": 0.645, "grad_norm": 0.7956018447875977, "learning_rate": 0.0002, "epoch": 4.9064906490649065, "step": 6690}, {"loss": 0.6431, "grad_norm": 0.6717583537101746, "learning_rate": 0.0002, "epoch": 4.913824715804914, "step": 6700}, {"loss": 0.6942, "grad_norm": 1.09099280834198, "learning_rate": 0.0002, "epoch": 4.9211587825449215, "step": 6710}, {"loss": 0.7533, "grad_norm": 0.8589889407157898, "learning_rate": 0.0002, "epoch": 4.928492849284929, "step": 6720}, {"loss": 0.66, "grad_norm": 1.0046314001083374, "learning_rate": 0.0002, "epoch": 4.9358269160249355, "step": 6730}, {"loss": 0.6864, "grad_norm": 0.8559659123420715, "learning_rate": 0.0002, "epoch": 4.943160982764943, "step": 6740}, {"loss": 0.6847, "grad_norm": 0.8588525652885437, "learning_rate": 0.0002, "epoch": 4.9504950495049505, "step": 6750}, {"loss": 0.6428, "grad_norm": 0.9192708134651184, "learning_rate": 0.0002, "epoch": 4.957829116244958, "step": 6760}, {"loss": 0.6873, "grad_norm": 1.051398754119873, "learning_rate": 0.0002, "epoch": 4.965163182984965, "step": 6770}, {"loss": 0.7249, "grad_norm": 0.9111362099647522, "learning_rate": 0.0002, "epoch": 4.972497249724973, "step": 6780}, {"loss": 0.7613, "grad_norm": 0.7305638194084167, "learning_rate": 0.0002, "epoch": 4.9798313164649795, "step": 6790}, {"loss": 0.6747, "grad_norm": 1.118837594985962, "learning_rate": 0.0002, "epoch": 4.987165383204987, "step": 6800}, {"loss": 0.6412, "grad_norm": 0.9075239300727844, "learning_rate": 0.0002, "epoch": 4.994499449944994, "step": 6810}, {"eval_loss": 1.2361247539520264, "eval_runtime": 32.7325, "eval_samples_per_second": 13.167, "eval_steps_per_second": 1.65, "epoch": 4.999633296662999, "step": 6817}, {"loss": 0.7091, "grad_norm": 1.0541315078735352, "learning_rate": 0.0002, "epoch": 5.001833516685002, "step": 6820}, {"loss": 0.4882, "grad_norm": 0.9750140905380249, "learning_rate": 0.0002, "epoch": 5.009167583425009, "step": 6830}, {"loss": 0.6022, "grad_norm": 0.931838870048523, "learning_rate": 0.0002, "epoch": 5.016501650165017, "step": 6840}, {"loss": 0.5194, "grad_norm": 1.110278844833374, "learning_rate": 0.0002, "epoch": 5.023835716905023, "step": 6850}, {"loss": 0.4676, "grad_norm": 1.0670180320739746, "learning_rate": 0.0002, "epoch": 5.031169783645031, "step": 6860}, {"loss": 0.4374, "grad_norm": 0.8762092590332031, "learning_rate": 0.0002, "epoch": 5.038503850385038, "step": 6870}, {"loss": 0.505, "grad_norm": 1.1169432401657104, "learning_rate": 0.0002, "epoch": 5.045837917125046, "step": 6880}, {"loss": 0.5114, "grad_norm": 1.005491018295288, "learning_rate": 0.0002, "epoch": 5.053171983865053, "step": 6890}, {"loss": 0.5221, "grad_norm": 1.1751841306686401, "learning_rate": 0.0002, "epoch": 5.060506050605061, "step": 6900}, {"loss": 0.451, "grad_norm": 0.8501367568969727, "learning_rate": 0.0002, "epoch": 5.067840117345068, "step": 6910}, {"loss": 0.5292, "grad_norm": 0.9795131683349609, "learning_rate": 0.0002, "epoch": 5.075174184085075, "step": 6920}, {"loss": 0.5234, "grad_norm": 0.8929879665374756, "learning_rate": 0.0002, "epoch": 5.082508250825082, "step": 6930}, {"loss": 0.5378, "grad_norm": 1.0156651735305786, "learning_rate": 0.0002, "epoch": 5.08984231756509, "step": 6940}, {"loss": 0.5241, "grad_norm": 1.0974335670471191, "learning_rate": 0.0002, "epoch": 5.097176384305097, "step": 6950}, {"loss": 0.5705, "grad_norm": 1.7015666961669922, "learning_rate": 0.0002, "epoch": 5.104510451045105, "step": 6960}, {"loss": 0.523, "grad_norm": 1.0343226194381714, "learning_rate": 0.0002, "epoch": 5.111844517785112, "step": 6970}, {"loss": 0.4616, "grad_norm": 1.3072983026504517, "learning_rate": 0.0002, "epoch": 5.119178584525119, "step": 6980}, {"loss": 0.4813, "grad_norm": 1.038986086845398, "learning_rate": 0.0002, "epoch": 5.126512651265126, "step": 6990}, {"loss": 0.4616, "grad_norm": 0.8638386130332947, "learning_rate": 0.0002, "epoch": 5.133846718005134, "step": 7000}, {"loss": 0.5294, "grad_norm": 0.8326523900032043, "learning_rate": 0.0002, "epoch": 5.141180784745141, "step": 7010}, {"loss": 0.5021, "grad_norm": 1.0976895093917847, "learning_rate": 0.0002, "epoch": 5.148514851485149, "step": 7020}, {"loss": 0.4677, "grad_norm": 1.0077873468399048, "learning_rate": 0.0002, "epoch": 5.155848918225156, "step": 7030}, {"loss": 0.5262, "grad_norm": 1.0662257671356201, "learning_rate": 0.0002, "epoch": 5.163182984965164, "step": 7040}, {"loss": 0.5484, "grad_norm": 1.206271767616272, "learning_rate": 0.0002, "epoch": 5.17051705170517, "step": 7050}, {"loss": 0.4817, "grad_norm": 1.1990262269973755, "learning_rate": 0.0002, "epoch": 5.177851118445178, "step": 7060}, {"loss": 0.6048, "grad_norm": 1.0207163095474243, "learning_rate": 0.0002, "epoch": 5.185185185185185, "step": 7070}, {"loss": 0.4816, "grad_norm": 1.2783987522125244, "learning_rate": 0.0002, "epoch": 5.192519251925193, "step": 7080}, {"loss": 0.5322, "grad_norm": 1.1592512130737305, "learning_rate": 0.0002, "epoch": 5.1998533186652, "step": 7090}, {"loss": 0.5472, "grad_norm": 1.1053160429000854, "learning_rate": 0.0002, "epoch": 5.2071873854052075, "step": 7100}, {"loss": 0.4986, "grad_norm": 1.1925510168075562, "learning_rate": 0.0002, "epoch": 5.214521452145214, "step": 7110}, {"loss": 0.5065, "grad_norm": 1.0714877843856812, "learning_rate": 0.0002, "epoch": 5.221855518885222, "step": 7120}, {"loss": 0.5209, "grad_norm": 0.9451011419296265, "learning_rate": 0.0002, "epoch": 5.229189585625229, "step": 7130}, {"loss": 0.5298, "grad_norm": 1.03838050365448, "learning_rate": 0.0002, "epoch": 5.2365236523652365, "step": 7140}, {"loss": 0.4848, "grad_norm": 0.9204146265983582, "learning_rate": 0.0002, "epoch": 5.243857719105244, "step": 7150}, {"loss": 0.5164, "grad_norm": 1.0142229795455933, "learning_rate": 0.0002, "epoch": 5.2511917858452515, "step": 7160}, {"loss": 0.5092, "grad_norm": 1.4432005882263184, "learning_rate": 0.0002, "epoch": 5.258525852585258, "step": 7170}, {"loss": 0.5133, "grad_norm": 1.1239633560180664, "learning_rate": 0.0002, "epoch": 5.2658599193252655, "step": 7180}, {"loss": 0.4969, "grad_norm": 0.7012821435928345, "learning_rate": 0.0002, "epoch": 5.273193986065273, "step": 7190}, {"loss": 0.5466, "grad_norm": 1.3499128818511963, "learning_rate": 0.0002, "epoch": 5.2805280528052805, "step": 7200}, {"loss": 0.5282, "grad_norm": 0.9498730897903442, "learning_rate": 0.0002, "epoch": 5.287862119545288, "step": 7210}, {"loss": 0.5051, "grad_norm": 0.9552369117736816, "learning_rate": 0.0002, "epoch": 5.295196186285295, "step": 7220}, {"loss": 0.5329, "grad_norm": 0.7610348463058472, "learning_rate": 0.0002, "epoch": 5.302530253025303, "step": 7230}, {"loss": 0.468, "grad_norm": 1.0314512252807617, "learning_rate": 0.0002, "epoch": 5.3098643197653095, "step": 7240}, {"loss": 0.5367, "grad_norm": 1.0534334182739258, "learning_rate": 0.0002, "epoch": 5.317198386505317, "step": 7250}, {"loss": 0.5491, "grad_norm": 1.2553406953811646, "learning_rate": 0.0002, "epoch": 5.324532453245324, "step": 7260}, {"loss": 0.5218, "grad_norm": 0.7061691880226135, "learning_rate": 0.0002, "epoch": 5.331866519985332, "step": 7270}, {"loss": 0.5625, "grad_norm": 0.9652578830718994, "learning_rate": 0.0002, "epoch": 5.339200586725339, "step": 7280}, {"loss": 0.5608, "grad_norm": 1.114788293838501, "learning_rate": 0.0002, "epoch": 5.346534653465347, "step": 7290}, {"loss": 0.578, "grad_norm": 1.0940049886703491, "learning_rate": 0.0002, "epoch": 5.353868720205353, "step": 7300}, {"loss": 0.5256, "grad_norm": 1.0151008367538452, "learning_rate": 0.0002, "epoch": 5.361202786945361, "step": 7310}, {"loss": 0.5377, "grad_norm": 1.0369552373886108, "learning_rate": 0.0002, "epoch": 5.368536853685368, "step": 7320}, {"loss": 0.5028, "grad_norm": 0.8489866256713867, "learning_rate": 0.0002, "epoch": 5.375870920425376, "step": 7330}, {"loss": 0.5937, "grad_norm": 1.1031713485717773, "learning_rate": 0.0002, "epoch": 5.383204987165383, "step": 7340}, {"loss": 0.5355, "grad_norm": 0.9094716310501099, "learning_rate": 0.0002, "epoch": 5.390539053905391, "step": 7350}, {"loss": 0.5406, "grad_norm": 0.9530431032180786, "learning_rate": 0.0002, "epoch": 5.397873120645398, "step": 7360}, {"loss": 0.529, "grad_norm": 0.9633604884147644, "learning_rate": 0.0002, "epoch": 5.405207187385405, "step": 7370}, {"loss": 0.5315, "grad_norm": 0.9541662335395813, "learning_rate": 0.0002, "epoch": 5.412541254125412, "step": 7380}, {"loss": 0.6774, "grad_norm": 1.0459771156311035, "learning_rate": 0.0002, "epoch": 5.41987532086542, "step": 7390}, {"loss": 0.5737, "grad_norm": 1.027388334274292, "learning_rate": 0.0002, "epoch": 5.427209387605427, "step": 7400}, {"loss": 0.556, "grad_norm": 0.7267653346061707, "learning_rate": 0.0002, "epoch": 5.434543454345435, "step": 7410}, {"loss": 0.4581, "grad_norm": 1.020142674446106, "learning_rate": 0.0002, "epoch": 5.441877521085442, "step": 7420}, {"loss": 0.4853, "grad_norm": 1.044754147529602, "learning_rate": 0.0002, "epoch": 5.449211587825449, "step": 7430}, {"loss": 0.5666, "grad_norm": 1.5476195812225342, "learning_rate": 0.0002, "epoch": 5.456545654565456, "step": 7440}, {"loss": 0.5302, "grad_norm": 0.9879506826400757, "learning_rate": 0.0002, "epoch": 5.463879721305464, "step": 7450}, {"loss": 0.591, "grad_norm": 1.2562980651855469, "learning_rate": 0.0002, "epoch": 5.471213788045471, "step": 7460}, {"loss": 0.5188, "grad_norm": 1.3051384687423706, "learning_rate": 0.0002, "epoch": 5.478547854785479, "step": 7470}, {"loss": 0.5658, "grad_norm": 1.0511597394943237, "learning_rate": 0.0002, "epoch": 5.485881921525486, "step": 7480}, {"loss": 0.6327, "grad_norm": 1.0380817651748657, "learning_rate": 0.0002, "epoch": 5.493215988265494, "step": 7490}, {"loss": 0.5356, "grad_norm": 1.170274257659912, "learning_rate": 0.0002, "epoch": 5.5005500550055, "step": 7500}, {"loss": 0.5405, "grad_norm": 1.3356517553329468, "learning_rate": 0.0002, "epoch": 5.507884121745508, "step": 7510}, {"loss": 0.5305, "grad_norm": 1.0727124214172363, "learning_rate": 0.0002, "epoch": 5.515218188485515, "step": 7520}, {"loss": 0.5543, "grad_norm": 1.0110199451446533, "learning_rate": 0.0002, "epoch": 5.522552255225523, "step": 7530}, {"loss": 0.5962, "grad_norm": 1.3086743354797363, "learning_rate": 0.0002, "epoch": 5.52988632196553, "step": 7540}, {"loss": 0.5512, "grad_norm": 1.1904916763305664, "learning_rate": 0.0002, "epoch": 5.537220388705538, "step": 7550}, {"loss": 0.5915, "grad_norm": 0.9466280937194824, "learning_rate": 0.0002, "epoch": 5.544554455445544, "step": 7560}, {"loss": 0.5573, "grad_norm": 1.1237901449203491, "learning_rate": 0.0002, "epoch": 5.551888522185552, "step": 7570}, {"loss": 0.5383, "grad_norm": 0.9590660333633423, "learning_rate": 0.0002, "epoch": 5.559222588925559, "step": 7580}, {"loss": 0.5594, "grad_norm": 1.0890778303146362, "learning_rate": 0.0002, "epoch": 5.566556655665567, "step": 7590}, {"loss": 0.5698, "grad_norm": 0.7206931114196777, "learning_rate": 0.0002, "epoch": 5.573890722405574, "step": 7600}, {"loss": 0.5511, "grad_norm": 1.2884514331817627, "learning_rate": 0.0002, "epoch": 5.5812247891455815, "step": 7610}, {"loss": 0.5279, "grad_norm": 0.7798039317131042, "learning_rate": 0.0002, "epoch": 5.588558855885589, "step": 7620}, {"loss": 0.4847, "grad_norm": 1.166046142578125, "learning_rate": 0.0002, "epoch": 5.595892922625596, "step": 7630}, {"loss": 0.5821, "grad_norm": 1.0150201320648193, "learning_rate": 0.0002, "epoch": 5.603226989365603, "step": 7640}, {"loss": 0.5296, "grad_norm": 1.0449682474136353, "learning_rate": 0.0002, "epoch": 5.6105610561056105, "step": 7650}, {"loss": 0.5431, "grad_norm": 0.9310530424118042, "learning_rate": 0.0002, "epoch": 5.617895122845618, "step": 7660}, {"loss": 0.5234, "grad_norm": 0.9117933511734009, "learning_rate": 0.0002, "epoch": 5.6252291895856255, "step": 7670}, {"loss": 0.5807, "grad_norm": 1.1475164890289307, "learning_rate": 0.0002, "epoch": 5.632563256325633, "step": 7680}, {"loss": 0.5816, "grad_norm": 1.066809058189392, "learning_rate": 0.0002, "epoch": 5.6398973230656395, "step": 7690}, {"loss": 0.551, "grad_norm": 1.2834991216659546, "learning_rate": 0.0002, "epoch": 5.647231389805647, "step": 7700}, {"loss": 0.5914, "grad_norm": 1.2245112657546997, "learning_rate": 0.0002, "epoch": 5.6545654565456545, "step": 7710}, {"loss": 0.5552, "grad_norm": 1.1424106359481812, "learning_rate": 0.0002, "epoch": 5.661899523285662, "step": 7720}, {"loss": 0.559, "grad_norm": 1.0673892498016357, "learning_rate": 0.0002, "epoch": 5.669233590025669, "step": 7730}, {"loss": 0.544, "grad_norm": 1.4312121868133545, "learning_rate": 0.0002, "epoch": 5.676567656765677, "step": 7740}, {"loss": 0.5576, "grad_norm": 0.9976982474327087, "learning_rate": 0.0002, "epoch": 5.683901723505684, "step": 7750}, {"loss": 0.4855, "grad_norm": 0.9464678168296814, "learning_rate": 0.0002, "epoch": 5.691235790245691, "step": 7760}, {"loss": 0.5363, "grad_norm": 1.010995626449585, "learning_rate": 0.0002, "epoch": 5.698569856985698, "step": 7770}, {"loss": 0.5873, "grad_norm": 1.3787750005722046, "learning_rate": 0.0002, "epoch": 5.705903923725706, "step": 7780}, {"loss": 0.6234, "grad_norm": 1.020922303199768, "learning_rate": 0.0002, "epoch": 5.713237990465713, "step": 7790}, {"loss": 0.5337, "grad_norm": 0.9748636484146118, "learning_rate": 0.0002, "epoch": 5.720572057205721, "step": 7800}, {"loss": 0.5507, "grad_norm": 1.3077744245529175, "learning_rate": 0.0002, "epoch": 5.727906123945728, "step": 7810}, {"loss": 0.558, "grad_norm": 1.4770057201385498, "learning_rate": 0.0002, "epoch": 5.735240190685735, "step": 7820}, {"loss": 0.5571, "grad_norm": 1.6349090337753296, "learning_rate": 0.0002, "epoch": 5.742574257425742, "step": 7830}, {"loss": 0.5056, "grad_norm": 0.9818630814552307, "learning_rate": 0.0002, "epoch": 5.74990832416575, "step": 7840}, {"loss": 0.5495, "grad_norm": 0.9659715890884399, "learning_rate": 0.0002, "epoch": 5.757242390905757, "step": 7850}, {"loss": 0.5628, "grad_norm": 0.9269950985908508, "learning_rate": 0.0002, "epoch": 5.764576457645765, "step": 7860}, {"loss": 0.5594, "grad_norm": 1.0099073648452759, "learning_rate": 0.0002, "epoch": 5.771910524385772, "step": 7870}, {"loss": 0.5912, "grad_norm": 0.9123615026473999, "learning_rate": 0.0002, "epoch": 5.77924459112578, "step": 7880}, {"loss": 0.6054, "grad_norm": 1.1542246341705322, "learning_rate": 0.0002, "epoch": 5.786578657865786, "step": 7890}, {"loss": 0.5829, "grad_norm": 1.0792022943496704, "learning_rate": 0.0002, "epoch": 5.793912724605794, "step": 7900}, {"loss": 0.504, "grad_norm": 0.95615553855896, "learning_rate": 0.0002, "epoch": 5.801246791345801, "step": 7910}, {"loss": 0.5918, "grad_norm": 1.2471332550048828, "learning_rate": 0.0002, "epoch": 5.808580858085809, "step": 7920}, {"loss": 0.5719, "grad_norm": 1.0189851522445679, "learning_rate": 0.0002, "epoch": 5.815914924825816, "step": 7930}, {"loss": 0.5958, "grad_norm": 1.3309742212295532, "learning_rate": 0.0002, "epoch": 5.823248991565823, "step": 7940}, {"loss": 0.6255, "grad_norm": 1.2930549383163452, "learning_rate": 0.0002, "epoch": 5.83058305830583, "step": 7950}, {"loss": 0.5301, "grad_norm": 0.8216308951377869, "learning_rate": 0.0002, "epoch": 5.837917125045838, "step": 7960}, {"loss": 0.5397, "grad_norm": 1.1205775737762451, "learning_rate": 0.0002, "epoch": 5.845251191785845, "step": 7970}, {"loss": 0.5903, "grad_norm": 0.851298451423645, "learning_rate": 0.0002, "epoch": 5.852585258525853, "step": 7980}, {"loss": 0.5981, "grad_norm": 0.8797095417976379, "learning_rate": 0.0002, "epoch": 5.85991932526586, "step": 7990}, {"loss": 0.6106, "grad_norm": 1.5784614086151123, "learning_rate": 0.0002, "epoch": 5.867253392005868, "step": 8000}, {"loss": 0.5956, "grad_norm": 1.1531187295913696, "learning_rate": 0.0002, "epoch": 5.874587458745875, "step": 8010}, {"loss": 0.6289, "grad_norm": 1.2469146251678467, "learning_rate": 0.0002, "epoch": 5.881921525485882, "step": 8020}, {"loss": 0.5827, "grad_norm": 1.0784350633621216, "learning_rate": 0.0002, "epoch": 5.889255592225889, "step": 8030}, {"loss": 0.6339, "grad_norm": 1.1311599016189575, "learning_rate": 0.0002, "epoch": 5.896589658965897, "step": 8040}, {"loss": 0.5815, "grad_norm": 0.9654512405395508, "learning_rate": 0.0002, "epoch": 5.903923725705904, "step": 8050}, {"loss": 0.6198, "grad_norm": 1.3288270235061646, "learning_rate": 0.0002, "epoch": 5.9112577924459115, "step": 8060}, {"loss": 0.6515, "grad_norm": 1.12800931930542, "learning_rate": 0.0002, "epoch": 5.918591859185918, "step": 8070}, {"loss": 0.5684, "grad_norm": 0.9449917674064636, "learning_rate": 0.0002, "epoch": 5.925925925925926, "step": 8080}, {"loss": 0.6063, "grad_norm": 1.1532357931137085, "learning_rate": 0.0002, "epoch": 5.933259992665933, "step": 8090}, {"loss": 0.5318, "grad_norm": 1.2211151123046875, "learning_rate": 0.0002, "epoch": 5.9405940594059405, "step": 8100}, {"loss": 0.6512, "grad_norm": 1.3459105491638184, "learning_rate": 0.0002, "epoch": 5.947928126145948, "step": 8110}, {"loss": 0.5952, "grad_norm": 1.251999855041504, "learning_rate": 0.0002, "epoch": 5.9552621928859555, "step": 8120}, {"loss": 0.6203, "grad_norm": 1.5682506561279297, "learning_rate": 0.0002, "epoch": 5.962596259625963, "step": 8130}, {"loss": 0.6253, "grad_norm": 0.926075279712677, "learning_rate": 0.0002, "epoch": 5.9699303263659695, "step": 8140}, {"loss": 0.5545, "grad_norm": 0.9622511863708496, "learning_rate": 0.0002, "epoch": 5.977264393105977, "step": 8150}, {"loss": 0.5518, "grad_norm": 0.9633373618125916, "learning_rate": 0.0002, "epoch": 5.9845984598459845, "step": 8160}, {"loss": 0.5831, "grad_norm": 0.8960476517677307, "learning_rate": 0.0002, "epoch": 5.991932526585992, "step": 8170}, {"loss": 0.5442, "grad_norm": 0.9372805953025818, "learning_rate": 0.0002, "epoch": 5.999266593325999, "step": 8180}]}
+{"epoch": 6.999633296662999, "step": 9544, "epoch_duration": 1554.9810886383057, "total_accumulated_duration": 10403.232823848724, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7672.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 1.9722, "grad_norm": 0.47521963715553284, "learning_rate": 0.0002, "epoch": 0.007334066740007334, "step": 10}, {"loss": 1.4821, "grad_norm": 0.5395162105560303, "learning_rate": 0.0002, "epoch": 0.014668133480014669, "step": 20}, {"loss": 1.4202, "grad_norm": 0.4305780231952667, "learning_rate": 0.0002, "epoch": 0.022002200220022004, "step": 30}, {"loss": 1.4271, "grad_norm": 0.6938246488571167, "learning_rate": 0.0002, "epoch": 0.029336266960029337, "step": 40}, {"loss": 1.3112, "grad_norm": 1.5133819580078125, "learning_rate": 0.0002, "epoch": 0.03667033370003667, "step": 50}, {"loss": 1.3132, "grad_norm": 0.9173883199691772, "learning_rate": 0.0002, "epoch": 0.04400440044004401, "step": 60}, {"loss": 1.2844, "grad_norm": 0.4619861841201782, "learning_rate": 0.0002, "epoch": 0.05133846718005134, "step": 70}, {"loss": 1.2108, "grad_norm": 0.46118637919425964, "learning_rate": 0.0002, "epoch": 0.058672533920058674, "step": 80}, {"loss": 1.3441, "grad_norm": 0.4468648135662079, "learning_rate": 0.0002, "epoch": 0.066006600660066, "step": 90}, {"loss": 1.1863, "grad_norm": 0.46123769879341125, "learning_rate": 0.0002, "epoch": 0.07334066740007333, "step": 100}, {"loss": 1.2772, "grad_norm": 0.4859139025211334, "learning_rate": 0.0002, "epoch": 0.08067473414008068, "step": 110}, {"loss": 1.2087, "grad_norm": 0.4384922385215759, "learning_rate": 0.0002, "epoch": 0.08800880088008801, "step": 120}, {"loss": 1.2927, "grad_norm": 0.39519360661506653, "learning_rate": 0.0002, "epoch": 0.09534286762009535, "step": 130}, {"loss": 1.2349, "grad_norm": 0.4049859344959259, "learning_rate": 0.0002, "epoch": 0.10267693436010268, "step": 140}, {"loss": 1.293, "grad_norm": 0.4605638086795807, "learning_rate": 0.0002, "epoch": 0.11001100110011001, "step": 150}, {"loss": 1.2659, "grad_norm": 0.4201928377151489, "learning_rate": 0.0002, "epoch": 0.11734506784011735, "step": 160}, {"loss": 1.3961, "grad_norm": 0.5367777347564697, "learning_rate": 0.0002, "epoch": 0.12467913458012468, "step": 170}, {"loss": 1.2481, "grad_norm": 0.41752299666404724, "learning_rate": 0.0002, "epoch": 0.132013201320132, "step": 180}, {"loss": 1.207, "grad_norm": 0.31597763299942017, "learning_rate": 0.0002, "epoch": 0.13934726806013933, "step": 190}, {"loss": 1.2441, "grad_norm": 0.7468788623809814, "learning_rate": 0.0002, "epoch": 0.14668133480014667, "step": 200}, {"loss": 1.199, "grad_norm": 0.3403034508228302, "learning_rate": 0.0002, "epoch": 0.15401540154015403, "step": 210}, {"loss": 1.2439, "grad_norm": 0.34240293502807617, "learning_rate": 0.0002, "epoch": 0.16134946828016136, "step": 220}, {"loss": 1.2022, "grad_norm": 0.356158971786499, "learning_rate": 0.0002, "epoch": 0.1686835350201687, "step": 230}, {"loss": 1.207, "grad_norm": 0.3448857367038727, "learning_rate": 0.0002, "epoch": 0.17601760176017603, "step": 240}, {"loss": 1.2156, "grad_norm": 0.3475699722766876, "learning_rate": 0.0002, "epoch": 0.18335166850018336, "step": 250}, {"loss": 1.1551, "grad_norm": 0.2770358622074127, "learning_rate": 0.0002, "epoch": 0.1906857352401907, "step": 260}, {"loss": 1.2238, "grad_norm": 0.4310270845890045, "learning_rate": 0.0002, "epoch": 0.19801980198019803, "step": 270}, {"loss": 1.2917, "grad_norm": 0.335041880607605, "learning_rate": 0.0002, "epoch": 0.20535386872020536, "step": 280}, {"loss": 1.0959, "grad_norm": 0.3420602083206177, "learning_rate": 0.0002, "epoch": 0.2126879354602127, "step": 290}, {"loss": 1.1232, "grad_norm": 0.325001060962677, "learning_rate": 0.0002, "epoch": 0.22002200220022003, "step": 300}, {"loss": 1.2007, "grad_norm": 0.3027827739715576, "learning_rate": 0.0002, "epoch": 0.22735606894022736, "step": 310}, {"loss": 1.1803, "grad_norm": 0.435550719499588, "learning_rate": 0.0002, "epoch": 0.2346901356802347, "step": 320}, {"loss": 1.2045, "grad_norm": 0.3884522616863251, "learning_rate": 0.0002, "epoch": 0.24202420242024203, "step": 330}, {"loss": 1.2481, "grad_norm": 0.7736002206802368, "learning_rate": 0.0002, "epoch": 0.24935826916024936, "step": 340}, {"loss": 1.3606, "grad_norm": 0.35052821040153503, "learning_rate": 0.0002, "epoch": 0.2566923359002567, "step": 350}, {"loss": 1.2129, "grad_norm": 0.3311890959739685, "learning_rate": 0.0002, "epoch": 0.264026402640264, "step": 360}, {"loss": 1.2219, "grad_norm": 0.7473500370979309, "learning_rate": 0.0002, "epoch": 0.27136046938027136, "step": 370}, {"loss": 1.2712, "grad_norm": 0.3681875765323639, "learning_rate": 0.0002, "epoch": 0.27869453612027867, "step": 380}, {"loss": 1.2258, "grad_norm": 0.3764737844467163, "learning_rate": 0.0002, "epoch": 0.28602860286028603, "step": 390}, {"loss": 1.1917, "grad_norm": 0.4243989586830139, "learning_rate": 0.0002, "epoch": 0.29336266960029334, "step": 400}, {"loss": 1.199, "grad_norm": 0.2658531963825226, "learning_rate": 0.0002, "epoch": 0.3006967363403007, "step": 410}, {"loss": 1.1622, "grad_norm": 0.3436793386936188, "learning_rate": 0.0002, "epoch": 0.30803080308030806, "step": 420}, {"loss": 1.2953, "grad_norm": 0.5101129412651062, "learning_rate": 0.0002, "epoch": 0.31536486982031536, "step": 430}, {"loss": 1.1557, "grad_norm": 0.3319750726222992, "learning_rate": 0.0002, "epoch": 0.3226989365603227, "step": 440}, {"loss": 1.1804, "grad_norm": 0.385148286819458, "learning_rate": 0.0002, "epoch": 0.33003300330033003, "step": 450}, {"loss": 1.1808, "grad_norm": 0.3477935791015625, "learning_rate": 0.0002, "epoch": 0.3373670700403374, "step": 460}, {"loss": 1.1877, "grad_norm": 0.29748716950416565, "learning_rate": 0.0002, "epoch": 0.3447011367803447, "step": 470}, {"loss": 1.19, "grad_norm": 0.34083324670791626, "learning_rate": 0.0002, "epoch": 0.35203520352035206, "step": 480}, {"loss": 1.2, "grad_norm": 0.36904552578926086, "learning_rate": 0.0002, "epoch": 0.35936927026035936, "step": 490}, {"loss": 1.2223, "grad_norm": 0.315483033657074, "learning_rate": 0.0002, "epoch": 0.3667033370003667, "step": 500}, {"loss": 1.1461, "grad_norm": 0.44897955656051636, "learning_rate": 0.0002, "epoch": 0.37403740374037403, "step": 510}, {"loss": 1.3035, "grad_norm": 0.3160701394081116, "learning_rate": 0.0002, "epoch": 0.3813714704803814, "step": 520}, {"loss": 1.3197, "grad_norm": 0.29584741592407227, "learning_rate": 0.0002, "epoch": 0.3887055372203887, "step": 530}, {"loss": 1.2983, "grad_norm": 0.5430002808570862, "learning_rate": 0.0002, "epoch": 0.39603960396039606, "step": 540}, {"loss": 1.2459, "grad_norm": 0.2908070683479309, "learning_rate": 0.0002, "epoch": 0.40337367070040336, "step": 550}, {"loss": 1.2384, "grad_norm": 0.35066530108451843, "learning_rate": 0.0002, "epoch": 0.4107077374404107, "step": 560}, {"loss": 1.1784, "grad_norm": 0.37588003277778625, "learning_rate": 0.0002, "epoch": 0.41804180418041803, "step": 570}, {"loss": 1.2334, "grad_norm": 0.3112126886844635, "learning_rate": 0.0002, "epoch": 0.4253758709204254, "step": 580}, {"loss": 1.1439, "grad_norm": 0.35577139258384705, "learning_rate": 0.0002, "epoch": 0.4327099376604327, "step": 590}, {"loss": 1.184, "grad_norm": 0.31706422567367554, "learning_rate": 0.0002, "epoch": 0.44004400440044006, "step": 600}, {"loss": 1.2081, "grad_norm": 0.3249092102050781, "learning_rate": 0.0002, "epoch": 0.44737807114044736, "step": 610}, {"loss": 1.0824, "grad_norm": 0.3842705488204956, "learning_rate": 0.0002, "epoch": 0.4547121378804547, "step": 620}, {"loss": 1.2257, "grad_norm": 0.390991747379303, "learning_rate": 0.0002, "epoch": 0.46204620462046203, "step": 630}, {"loss": 1.1954, "grad_norm": 0.27532413601875305, "learning_rate": 0.0002, "epoch": 0.4693802713604694, "step": 640}, {"loss": 1.1058, "grad_norm": 0.31412816047668457, "learning_rate": 0.0002, "epoch": 0.4767143381004767, "step": 650}, {"loss": 1.1312, "grad_norm": 0.32117101550102234, "learning_rate": 0.0002, "epoch": 0.48404840484048406, "step": 660}, {"loss": 1.2423, "grad_norm": 0.3810010254383087, "learning_rate": 0.0002, "epoch": 0.49138247158049136, "step": 670}, {"loss": 1.1978, "grad_norm": 0.36289164423942566, "learning_rate": 0.0002, "epoch": 0.4987165383204987, "step": 680}, {"loss": 1.2034, "grad_norm": 0.34458720684051514, "learning_rate": 0.0002, "epoch": 0.506050605060506, "step": 690}, {"loss": 1.1756, "grad_norm": 0.32844600081443787, "learning_rate": 0.0002, "epoch": 0.5133846718005134, "step": 700}, {"loss": 1.0807, "grad_norm": 0.3144175708293915, "learning_rate": 0.0002, "epoch": 0.5207187385405208, "step": 710}, {"loss": 1.1952, "grad_norm": 0.3898887634277344, "learning_rate": 0.0002, "epoch": 0.528052805280528, "step": 720}, {"loss": 1.1244, "grad_norm": 1.3220758438110352, "learning_rate": 0.0002, "epoch": 0.5353868720205354, "step": 730}, {"loss": 1.227, "grad_norm": 0.3635874390602112, "learning_rate": 0.0002, "epoch": 0.5427209387605427, "step": 740}, {"loss": 1.2169, "grad_norm": 0.3138217628002167, "learning_rate": 0.0002, "epoch": 0.5500550055005501, "step": 750}, {"loss": 1.1516, "grad_norm": 0.4063207805156708, "learning_rate": 0.0002, "epoch": 0.5573890722405573, "step": 760}, {"loss": 1.1954, "grad_norm": 0.3926219940185547, "learning_rate": 0.0002, "epoch": 0.5647231389805647, "step": 770}, {"loss": 1.1726, "grad_norm": 0.31954652070999146, "learning_rate": 0.0002, "epoch": 0.5720572057205721, "step": 780}, {"loss": 1.2977, "grad_norm": 0.4248711168766022, "learning_rate": 0.0002, "epoch": 0.5793912724605794, "step": 790}, {"loss": 1.1728, "grad_norm": 0.643004834651947, "learning_rate": 0.0002, "epoch": 0.5867253392005867, "step": 800}, {"loss": 1.1793, "grad_norm": 0.3479592800140381, "learning_rate": 0.0002, "epoch": 0.594059405940594, "step": 810}, {"loss": 1.2426, "grad_norm": 0.4684754014015198, "learning_rate": 0.0002, "epoch": 0.6013934726806014, "step": 820}, {"loss": 1.2002, "grad_norm": 0.3739790916442871, "learning_rate": 0.0002, "epoch": 0.6087275394206088, "step": 830}, {"loss": 1.2139, "grad_norm": 0.40884748101234436, "learning_rate": 0.0002, "epoch": 0.6160616061606161, "step": 840}, {"loss": 1.1557, "grad_norm": 0.9722164273262024, "learning_rate": 0.0002, "epoch": 0.6233956729006234, "step": 850}, {"loss": 1.3069, "grad_norm": 0.42992347478866577, "learning_rate": 0.0002, "epoch": 0.6307297396406307, "step": 860}, {"loss": 1.1339, "grad_norm": 0.36654195189476013, "learning_rate": 0.0002, "epoch": 0.6380638063806381, "step": 870}, {"loss": 1.1932, "grad_norm": 0.4113832116127014, "learning_rate": 0.0002, "epoch": 0.6453978731206454, "step": 880}, {"loss": 1.2163, "grad_norm": 0.2948838770389557, "learning_rate": 0.0002, "epoch": 0.6527319398606527, "step": 890}, {"loss": 1.1081, "grad_norm": 0.38330280780792236, "learning_rate": 0.0002, "epoch": 0.6600660066006601, "step": 900}, {"loss": 1.1342, "grad_norm": 0.4428867697715759, "learning_rate": 0.0002, "epoch": 0.6674000733406674, "step": 910}, {"loss": 1.1021, "grad_norm": 0.23659265041351318, "learning_rate": 0.0002, "epoch": 0.6747341400806748, "step": 920}, {"loss": 1.1226, "grad_norm": 0.323685884475708, "learning_rate": 0.0002, "epoch": 0.682068206820682, "step": 930}, {"loss": 1.0853, "grad_norm": 0.39157727360725403, "learning_rate": 0.0002, "epoch": 0.6894022735606894, "step": 940}, {"loss": 1.1435, "grad_norm": 0.27189481258392334, "learning_rate": 0.0002, "epoch": 0.6967363403006968, "step": 950}, {"loss": 1.1033, "grad_norm": 0.529883861541748, "learning_rate": 0.0002, "epoch": 0.7040704070407041, "step": 960}, {"loss": 1.139, "grad_norm": 0.34758689999580383, "learning_rate": 0.0002, "epoch": 0.7114044737807114, "step": 970}, {"loss": 1.2197, "grad_norm": 0.831749439239502, "learning_rate": 0.0002, "epoch": 0.7187385405207187, "step": 980}, {"loss": 1.158, "grad_norm": 0.4438304007053375, "learning_rate": 0.0002, "epoch": 0.7260726072607261, "step": 990}, {"loss": 1.1021, "grad_norm": 0.33840006589889526, "learning_rate": 0.0002, "epoch": 0.7334066740007334, "step": 1000}, {"loss": 1.254, "grad_norm": 0.3454797863960266, "learning_rate": 0.0002, "epoch": 0.7407407407407407, "step": 1010}, {"loss": 1.106, "grad_norm": 0.38999441266059875, "learning_rate": 0.0002, "epoch": 0.7480748074807481, "step": 1020}, {"loss": 1.1428, "grad_norm": 0.2829911708831787, "learning_rate": 0.0002, "epoch": 0.7554088742207554, "step": 1030}, {"loss": 1.2123, "grad_norm": 0.36918163299560547, "learning_rate": 0.0002, "epoch": 0.7627429409607628, "step": 1040}, {"loss": 1.3028, "grad_norm": 0.3415680229663849, "learning_rate": 0.0002, "epoch": 0.77007700770077, "step": 1050}, {"loss": 1.1939, "grad_norm": 0.2974182963371277, "learning_rate": 0.0002, "epoch": 0.7774110744407774, "step": 1060}, {"loss": 1.194, "grad_norm": 0.3880919814109802, "learning_rate": 0.0002, "epoch": 0.7847451411807848, "step": 1070}, {"loss": 1.1095, "grad_norm": 0.33503302931785583, "learning_rate": 0.0002, "epoch": 0.7920792079207921, "step": 1080}, {"loss": 1.2111, "grad_norm": 0.3728407025337219, "learning_rate": 0.0002, "epoch": 0.7994132746607994, "step": 1090}, {"loss": 1.0835, "grad_norm": 0.3509373664855957, "learning_rate": 0.0002, "epoch": 0.8067473414008067, "step": 1100}, {"loss": 1.2661, "grad_norm": 0.42228564620018005, "learning_rate": 0.0002, "epoch": 0.8140814081408141, "step": 1110}, {"loss": 1.1788, "grad_norm": 0.313467800617218, "learning_rate": 0.0002, "epoch": 0.8214154748808215, "step": 1120}, {"loss": 1.1971, "grad_norm": 0.3378850817680359, "learning_rate": 0.0002, "epoch": 0.8287495416208287, "step": 1130}, {"loss": 1.1238, "grad_norm": 0.43200382590293884, "learning_rate": 0.0002, "epoch": 0.8360836083608361, "step": 1140}, {"loss": 1.3203, "grad_norm": 0.3309599459171295, "learning_rate": 0.0002, "epoch": 0.8434176751008434, "step": 1150}, {"loss": 1.1062, "grad_norm": 0.3526846170425415, "learning_rate": 0.0002, "epoch": 0.8507517418408508, "step": 1160}, {"loss": 1.0851, "grad_norm": 1.2722247838974, "learning_rate": 0.0002, "epoch": 0.858085808580858, "step": 1170}, {"loss": 1.0785, "grad_norm": 0.34142059087753296, "learning_rate": 0.0002, "epoch": 0.8654198753208654, "step": 1180}, {"loss": 1.2187, "grad_norm": 0.3805823028087616, "learning_rate": 0.0002, "epoch": 0.8727539420608728, "step": 1190}, {"loss": 1.1215, "grad_norm": 0.3931232690811157, "learning_rate": 0.0002, "epoch": 0.8800880088008801, "step": 1200}, {"loss": 1.0948, "grad_norm": 0.2937372624874115, "learning_rate": 0.0002, "epoch": 0.8874220755408874, "step": 1210}, {"loss": 1.1228, "grad_norm": 0.3757196366786957, "learning_rate": 0.0002, "epoch": 0.8947561422808947, "step": 1220}, {"loss": 1.1222, "grad_norm": 0.3502705991268158, "learning_rate": 0.0002, "epoch": 0.9020902090209021, "step": 1230}, {"loss": 1.2242, "grad_norm": 0.32758915424346924, "learning_rate": 0.0002, "epoch": 0.9094242757609095, "step": 1240}, {"loss": 1.215, "grad_norm": 0.37199416756629944, "learning_rate": 0.0002, "epoch": 0.9167583425009168, "step": 1250}, {"loss": 1.1225, "grad_norm": 0.3551490604877472, "learning_rate": 0.0002, "epoch": 0.9240924092409241, "step": 1260}, {"loss": 1.1966, "grad_norm": 0.2859550714492798, "learning_rate": 0.0002, "epoch": 0.9314264759809314, "step": 1270}, {"loss": 1.2186, "grad_norm": 0.427990585565567, "learning_rate": 0.0002, "epoch": 0.9387605427209388, "step": 1280}, {"loss": 1.2848, "grad_norm": 0.33717992901802063, "learning_rate": 0.0002, "epoch": 0.9460946094609461, "step": 1290}, {"loss": 1.1656, "grad_norm": 0.30225634574890137, "learning_rate": 0.0002, "epoch": 0.9534286762009534, "step": 1300}, {"loss": 1.2404, "grad_norm": 0.385821133852005, "learning_rate": 0.0002, "epoch": 0.9607627429409608, "step": 1310}, {"loss": 1.1932, "grad_norm": 0.35278066992759705, "learning_rate": 0.0002, "epoch": 0.9680968096809681, "step": 1320}, {"loss": 1.1071, "grad_norm": 0.49987098574638367, "learning_rate": 0.0002, "epoch": 0.9754308764209755, "step": 1330}, {"loss": 1.2259, "grad_norm": 0.3842747211456299, "learning_rate": 0.0002, "epoch": 0.9827649431609827, "step": 1340}, {"loss": 1.0862, "grad_norm": 0.6274653673171997, "learning_rate": 0.0002, "epoch": 0.9900990099009901, "step": 1350}, {"loss": 1.124, "grad_norm": 0.5239808559417725, "learning_rate": 0.0002, "epoch": 0.9974330766409975, "step": 1360}, {"eval_loss": 1.1822267770767212, "eval_runtime": 32.7389, "eval_samples_per_second": 13.165, "eval_steps_per_second": 1.649, "epoch": 0.9996332966629996, "step": 1363}, {"loss": 1.096, "grad_norm": 0.45311301946640015, "learning_rate": 0.0002, "epoch": 1.0047671433810048, "step": 1370}, {"loss": 1.0143, "grad_norm": 0.29685574769973755, "learning_rate": 0.0002, "epoch": 1.012101210121012, "step": 1380}, {"loss": 1.0302, "grad_norm": 0.3290937840938568, "learning_rate": 0.0002, "epoch": 1.0194352768610195, "step": 1390}, {"loss": 1.0295, "grad_norm": 0.3801758587360382, "learning_rate": 0.0002, "epoch": 1.0267693436010268, "step": 1400}, {"loss": 1.1226, "grad_norm": 0.794174313545227, "learning_rate": 0.0002, "epoch": 1.034103410341034, "step": 1410}, {"loss": 1.2232, "grad_norm": 0.3854154646396637, "learning_rate": 0.0002, "epoch": 1.0414374770810415, "step": 1420}, {"loss": 1.0652, "grad_norm": 0.32702451944351196, "learning_rate": 0.0002, "epoch": 1.0487715438210488, "step": 1430}, {"loss": 1.1144, "grad_norm": 0.7815203666687012, "learning_rate": 0.0002, "epoch": 1.056105610561056, "step": 1440}, {"loss": 1.1316, "grad_norm": 0.3087436854839325, "learning_rate": 0.0002, "epoch": 1.0634396773010635, "step": 1450}, {"loss": 1.1124, "grad_norm": 0.3847602903842926, "learning_rate": 0.0002, "epoch": 1.0707737440410707, "step": 1460}, {"loss": 1.1428, "grad_norm": 0.3693031370639801, "learning_rate": 0.0002, "epoch": 1.0781078107810782, "step": 1470}, {"loss": 1.0995, "grad_norm": 0.4111202359199524, "learning_rate": 0.0002, "epoch": 1.0854418775210855, "step": 1480}, {"loss": 1.0961, "grad_norm": 0.41452381014823914, "learning_rate": 0.0002, "epoch": 1.0927759442610927, "step": 1490}, {"loss": 1.1068, "grad_norm": 0.3336445093154907, "learning_rate": 0.0002, "epoch": 1.1001100110011002, "step": 1500}, {"loss": 1.0556, "grad_norm": 0.3923407793045044, "learning_rate": 0.0002, "epoch": 1.1074440777411074, "step": 1510}, {"loss": 1.1644, "grad_norm": 0.46215683221817017, "learning_rate": 0.0002, "epoch": 1.1147781444811147, "step": 1520}, {"loss": 1.1133, "grad_norm": 0.3592156767845154, "learning_rate": 0.0002, "epoch": 1.1221122112211221, "step": 1530}, {"loss": 1.0957, "grad_norm": 0.361110657453537, "learning_rate": 0.0002, "epoch": 1.1294462779611294, "step": 1540}, {"loss": 1.1553, "grad_norm": 0.5317131280899048, "learning_rate": 0.0002, "epoch": 1.1367803447011369, "step": 1550}, {"loss": 1.0368, "grad_norm": 0.3882388174533844, "learning_rate": 0.0002, "epoch": 1.1441144114411441, "step": 1560}, {"loss": 1.0805, "grad_norm": 0.3259428143501282, "learning_rate": 0.0002, "epoch": 1.1514484781811514, "step": 1570}, {"loss": 1.1819, "grad_norm": 0.410935640335083, "learning_rate": 0.0002, "epoch": 1.1587825449211588, "step": 1580}, {"loss": 1.1143, "grad_norm": 0.44940185546875, "learning_rate": 0.0002, "epoch": 1.166116611661166, "step": 1590}, {"loss": 1.0334, "grad_norm": 0.5106484293937683, "learning_rate": 0.0002, "epoch": 1.1734506784011733, "step": 1600}, {"loss": 1.2376, "grad_norm": 0.6603665947914124, "learning_rate": 0.0002, "epoch": 1.1807847451411808, "step": 1610}, {"loss": 1.1227, "grad_norm": 0.4799964129924774, "learning_rate": 0.0002, "epoch": 1.188118811881188, "step": 1620}, {"loss": 1.1191, "grad_norm": 0.4389883279800415, "learning_rate": 0.0002, "epoch": 1.1954528786211955, "step": 1630}, {"loss": 1.0667, "grad_norm": 0.4188813269138336, "learning_rate": 0.0002, "epoch": 1.2027869453612028, "step": 1640}, {"loss": 1.0605, "grad_norm": 0.7132157683372498, "learning_rate": 0.0002, "epoch": 1.21012101210121, "step": 1650}, {"loss": 1.0204, "grad_norm": 0.507480263710022, "learning_rate": 0.0002, "epoch": 1.2174550788412175, "step": 1660}, {"loss": 0.9948, "grad_norm": 0.9452332854270935, "learning_rate": 0.0002, "epoch": 1.2247891455812248, "step": 1670}, {"loss": 1.0228, "grad_norm": 0.4121614992618561, "learning_rate": 0.0002, "epoch": 1.2321232123212322, "step": 1680}, {"loss": 1.0366, "grad_norm": 0.34230247139930725, "learning_rate": 0.0002, "epoch": 1.2394572790612395, "step": 1690}, {"loss": 1.1289, "grad_norm": 0.4026208817958832, "learning_rate": 0.0002, "epoch": 1.2467913458012467, "step": 1700}, {"loss": 1.0206, "grad_norm": 0.46673697233200073, "learning_rate": 0.0002, "epoch": 1.2541254125412542, "step": 1710}, {"loss": 1.0827, "grad_norm": 0.38349825143814087, "learning_rate": 0.0002, "epoch": 1.2614594792812615, "step": 1720}, {"loss": 1.0356, "grad_norm": 0.4049997627735138, "learning_rate": 0.0002, "epoch": 1.2687935460212687, "step": 1730}, {"loss": 0.9504, "grad_norm": 0.3417615294456482, "learning_rate": 0.0002, "epoch": 1.2761276127612762, "step": 1740}, {"loss": 1.094, "grad_norm": 0.4277614951133728, "learning_rate": 0.0002, "epoch": 1.2834616795012834, "step": 1750}, {"loss": 0.9938, "grad_norm": 0.5864202976226807, "learning_rate": 0.0002, "epoch": 1.2907957462412907, "step": 1760}, {"loss": 1.1167, "grad_norm": 0.7097493410110474, "learning_rate": 0.0002, "epoch": 1.2981298129812981, "step": 1770}, {"loss": 1.1132, "grad_norm": 0.3145381212234497, "learning_rate": 0.0002, "epoch": 1.3054638797213054, "step": 1780}, {"loss": 1.1099, "grad_norm": 0.5116165280342102, "learning_rate": 0.0002, "epoch": 1.3127979464613129, "step": 1790}, {"loss": 1.0765, "grad_norm": 0.7469736337661743, "learning_rate": 0.0002, "epoch": 1.3201320132013201, "step": 1800}, {"loss": 1.0663, "grad_norm": 0.32272255420684814, "learning_rate": 0.0002, "epoch": 1.3274660799413276, "step": 1810}, {"loss": 0.9887, "grad_norm": 0.3534623086452484, "learning_rate": 0.0002, "epoch": 1.3348001466813348, "step": 1820}, {"loss": 1.1628, "grad_norm": 0.36127907037734985, "learning_rate": 0.0002, "epoch": 1.342134213421342, "step": 1830}, {"loss": 1.0972, "grad_norm": 0.4072401523590088, "learning_rate": 0.0002, "epoch": 1.3494682801613496, "step": 1840}, {"loss": 1.1267, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.3568023469013568, "step": 1850}, {"loss": 1.0173, "grad_norm": 0.412883460521698, "learning_rate": 0.0002, "epoch": 1.364136413641364, "step": 1860}, {"loss": 1.0265, "grad_norm": 0.3735875189304352, "learning_rate": 0.0002, "epoch": 1.3714704803813715, "step": 1870}, {"loss": 1.1061, "grad_norm": 0.39158159494400024, "learning_rate": 0.0002, "epoch": 1.3788045471213788, "step": 1880}, {"loss": 1.0433, "grad_norm": 0.44431769847869873, "learning_rate": 0.0002, "epoch": 1.386138613861386, "step": 1890}, {"loss": 1.0216, "grad_norm": 0.37772801518440247, "learning_rate": 0.0002, "epoch": 1.3934726806013935, "step": 1900}, {"loss": 1.0674, "grad_norm": 0.4056641757488251, "learning_rate": 0.0002, "epoch": 1.4008067473414008, "step": 1910}, {"loss": 1.0256, "grad_norm": 0.41612377762794495, "learning_rate": 0.0002, "epoch": 1.408140814081408, "step": 1920}, {"loss": 1.0467, "grad_norm": 0.41153013706207275, "learning_rate": 0.0002, "epoch": 1.4154748808214155, "step": 1930}, {"loss": 1.1062, "grad_norm": 0.387845516204834, "learning_rate": 0.0002, "epoch": 1.4228089475614227, "step": 1940}, {"loss": 1.1094, "grad_norm": 0.3809587061405182, "learning_rate": 0.0002, "epoch": 1.4301430143014302, "step": 1950}, {"loss": 1.0461, "grad_norm": 0.3625726103782654, "learning_rate": 0.0002, "epoch": 1.4374770810414375, "step": 1960}, {"loss": 0.9983, "grad_norm": 0.5294290781021118, "learning_rate": 0.0002, "epoch": 1.444811147781445, "step": 1970}, {"loss": 1.1114, "grad_norm": 0.39975494146347046, "learning_rate": 0.0002, "epoch": 1.4521452145214522, "step": 1980}, {"loss": 0.9704, "grad_norm": 0.4181167185306549, "learning_rate": 0.0002, "epoch": 1.4594792812614594, "step": 1990}, {"loss": 1.1146, "grad_norm": 0.42001503705978394, "learning_rate": 0.0002, "epoch": 1.466813348001467, "step": 2000}, {"loss": 1.1266, "grad_norm": 0.4877578616142273, "learning_rate": 0.0002, "epoch": 1.4741474147414741, "step": 2010}, {"loss": 1.1012, "grad_norm": 0.4050969183444977, "learning_rate": 0.0002, "epoch": 1.4814814814814814, "step": 2020}, {"loss": 1.0562, "grad_norm": 0.39068883657455444, "learning_rate": 0.0002, "epoch": 1.4888155482214889, "step": 2030}, {"loss": 1.0464, "grad_norm": 0.421282559633255, "learning_rate": 0.0002, "epoch": 1.4961496149614961, "step": 2040}, {"loss": 1.0532, "grad_norm": 0.47092297673225403, "learning_rate": 0.0002, "epoch": 1.5034836817015034, "step": 2050}, {"loss": 0.9348, "grad_norm": 0.39688974618911743, "learning_rate": 0.0002, "epoch": 1.5108177484415108, "step": 2060}, {"loss": 1.08, "grad_norm": 0.5529879331588745, "learning_rate": 0.0002, "epoch": 1.5181518151815183, "step": 2070}, {"loss": 1.1836, "grad_norm": 0.4879782199859619, "learning_rate": 0.0002, "epoch": 1.5254858819215253, "step": 2080}, {"loss": 1.0432, "grad_norm": 0.5517361164093018, "learning_rate": 0.0002, "epoch": 1.5328199486615328, "step": 2090}, {"loss": 1.0433, "grad_norm": 0.44015637040138245, "learning_rate": 0.0002, "epoch": 1.5401540154015403, "step": 2100}, {"loss": 1.1873, "grad_norm": 0.5435167551040649, "learning_rate": 0.0002, "epoch": 1.5474880821415475, "step": 2110}, {"loss": 1.1076, "grad_norm": 0.5714033246040344, "learning_rate": 0.0002, "epoch": 1.5548221488815548, "step": 2120}, {"loss": 1.1107, "grad_norm": 0.31732529401779175, "learning_rate": 0.0002, "epoch": 1.5621562156215623, "step": 2130}, {"loss": 1.0817, "grad_norm": 0.49068278074264526, "learning_rate": 0.0002, "epoch": 1.5694902823615695, "step": 2140}, {"loss": 1.0254, "grad_norm": 0.46851542592048645, "learning_rate": 0.0002, "epoch": 1.5768243491015768, "step": 2150}, {"loss": 1.0623, "grad_norm": 0.5083092451095581, "learning_rate": 0.0002, "epoch": 1.5841584158415842, "step": 2160}, {"loss": 1.0603, "grad_norm": 0.9822936058044434, "learning_rate": 0.0002, "epoch": 1.5914924825815915, "step": 2170}, {"loss": 0.9986, "grad_norm": 0.4575989246368408, "learning_rate": 0.0002, "epoch": 1.5988265493215987, "step": 2180}, {"loss": 1.1292, "grad_norm": 0.47444286942481995, "learning_rate": 0.0002, "epoch": 1.6061606160616062, "step": 2190}, {"loss": 1.0136, "grad_norm": 0.7208226919174194, "learning_rate": 0.0002, "epoch": 1.6134946828016135, "step": 2200}, {"loss": 1.15, "grad_norm": 0.43791481852531433, "learning_rate": 0.0002, "epoch": 1.6208287495416207, "step": 2210}, {"loss": 1.0961, "grad_norm": 0.5245792865753174, "learning_rate": 0.0002, "epoch": 1.6281628162816282, "step": 2220}, {"loss": 0.9957, "grad_norm": 0.39289429783821106, "learning_rate": 0.0002, "epoch": 1.6354968830216357, "step": 2230}, {"loss": 1.133, "grad_norm": 0.6106135845184326, "learning_rate": 0.0002, "epoch": 1.6428309497616427, "step": 2240}, {"loss": 1.0129, "grad_norm": 0.3722580671310425, "learning_rate": 0.0002, "epoch": 1.6501650165016502, "step": 2250}, {"loss": 1.0446, "grad_norm": 0.3649403750896454, "learning_rate": 0.0002, "epoch": 1.6574990832416576, "step": 2260}, {"loss": 1.0037, "grad_norm": 0.46514248847961426, "learning_rate": 0.0002, "epoch": 1.6648331499816649, "step": 2270}, {"loss": 1.0022, "grad_norm": 0.42034927010536194, "learning_rate": 0.0002, "epoch": 1.6721672167216721, "step": 2280}, {"loss": 1.1362, "grad_norm": 0.45202910900115967, "learning_rate": 0.0002, "epoch": 1.6795012834616796, "step": 2290}, {"loss": 1.0866, "grad_norm": 0.36257603764533997, "learning_rate": 0.0002, "epoch": 1.6868353502016868, "step": 2300}, {"loss": 1.0973, "grad_norm": 0.6340323090553284, "learning_rate": 0.0002, "epoch": 1.694169416941694, "step": 2310}, {"loss": 1.0615, "grad_norm": 0.4352878928184509, "learning_rate": 0.0002, "epoch": 1.7015034836817016, "step": 2320}, {"loss": 1.0629, "grad_norm": 0.45029792189598083, "learning_rate": 0.0002, "epoch": 1.7088375504217088, "step": 2330}, {"loss": 0.9621, "grad_norm": 0.3891315758228302, "learning_rate": 0.0002, "epoch": 1.716171617161716, "step": 2340}, {"loss": 0.9779, "grad_norm": 0.35180050134658813, "learning_rate": 0.0002, "epoch": 1.7235056839017235, "step": 2350}, {"loss": 1.0368, "grad_norm": 0.42367449402809143, "learning_rate": 0.0002, "epoch": 1.7308397506417308, "step": 2360}, {"loss": 1.0376, "grad_norm": 0.4553675353527069, "learning_rate": 0.0002, "epoch": 1.738173817381738, "step": 2370}, {"loss": 1.1467, "grad_norm": 0.5944654941558838, "learning_rate": 0.0002, "epoch": 1.7455078841217455, "step": 2380}, {"loss": 1.0548, "grad_norm": 0.3479664623737335, "learning_rate": 0.0002, "epoch": 1.752841950861753, "step": 2390}, {"loss": 1.0798, "grad_norm": 0.3585502505302429, "learning_rate": 0.0002, "epoch": 1.76017601760176, "step": 2400}, {"loss": 1.0983, "grad_norm": 0.4263346493244171, "learning_rate": 0.0002, "epoch": 1.7675100843417675, "step": 2410}, {"loss": 1.054, "grad_norm": 0.5476409196853638, "learning_rate": 0.0002, "epoch": 1.774844151081775, "step": 2420}, {"loss": 1.1615, "grad_norm": 0.3694186508655548, "learning_rate": 0.0002, "epoch": 1.7821782178217822, "step": 2430}, {"loss": 1.1343, "grad_norm": 0.9185658693313599, "learning_rate": 0.0002, "epoch": 1.7895122845617895, "step": 2440}, {"loss": 1.0764, "grad_norm": 0.7171908020973206, "learning_rate": 0.0002, "epoch": 1.796846351301797, "step": 2450}, {"loss": 1.1154, "grad_norm": 0.550658643245697, "learning_rate": 0.0002, "epoch": 1.8041804180418042, "step": 2460}, {"loss": 0.9975, "grad_norm": 0.4075568914413452, "learning_rate": 0.0002, "epoch": 1.8115144847818114, "step": 2470}, {"loss": 1.0935, "grad_norm": 0.3790127635002136, "learning_rate": 0.0002, "epoch": 1.818848551521819, "step": 2480}, {"loss": 0.9839, "grad_norm": 0.3576384484767914, "learning_rate": 0.0002, "epoch": 1.8261826182618262, "step": 2490}, {"loss": 1.1369, "grad_norm": 0.3919370770454407, "learning_rate": 0.0002, "epoch": 1.8335166850018334, "step": 2500}, {"loss": 0.9985, "grad_norm": 0.485083669424057, "learning_rate": 0.0002, "epoch": 1.8408507517418409, "step": 2510}, {"loss": 1.1585, "grad_norm": 0.4564347565174103, "learning_rate": 0.0002, "epoch": 1.8481848184818483, "step": 2520}, {"loss": 1.0944, "grad_norm": 0.3613106608390808, "learning_rate": 0.0002, "epoch": 1.8555188852218554, "step": 2530}, {"loss": 1.0819, "grad_norm": 0.39600759744644165, "learning_rate": 0.0002, "epoch": 1.8628529519618628, "step": 2540}, {"loss": 0.9453, "grad_norm": 1.123499870300293, "learning_rate": 0.0002, "epoch": 1.8701870187018703, "step": 2550}, {"loss": 1.0635, "grad_norm": 0.4612680673599243, "learning_rate": 0.0002, "epoch": 1.8775210854418776, "step": 2560}, {"loss": 1.0087, "grad_norm": 0.42745399475097656, "learning_rate": 0.0002, "epoch": 1.8848551521818848, "step": 2570}, {"loss": 1.0102, "grad_norm": 0.4055580198764801, "learning_rate": 0.0002, "epoch": 1.8921892189218923, "step": 2580}, {"loss": 1.0177, "grad_norm": 0.44174644351005554, "learning_rate": 0.0002, "epoch": 1.8995232856618995, "step": 2590}, {"loss": 0.9886, "grad_norm": 1.0228385925292969, "learning_rate": 0.0002, "epoch": 1.9068573524019068, "step": 2600}, {"loss": 1.0857, "grad_norm": 0.3496396243572235, "learning_rate": 0.0002, "epoch": 1.9141914191419143, "step": 2610}, {"loss": 1.0955, "grad_norm": 0.4191173017024994, "learning_rate": 0.0002, "epoch": 1.9215254858819215, "step": 2620}, {"loss": 1.0943, "grad_norm": 0.6778554916381836, "learning_rate": 0.0002, "epoch": 1.9288595526219288, "step": 2630}, {"loss": 1.0594, "grad_norm": 0.41992834210395813, "learning_rate": 0.0002, "epoch": 1.9361936193619362, "step": 2640}, {"loss": 1.1159, "grad_norm": 0.8760401010513306, "learning_rate": 0.0002, "epoch": 1.9435276861019435, "step": 2650}, {"loss": 1.0379, "grad_norm": 0.44049209356307983, "learning_rate": 0.0002, "epoch": 1.9508617528419507, "step": 2660}, {"loss": 1.1008, "grad_norm": 0.5651928782463074, "learning_rate": 0.0002, "epoch": 1.9581958195819582, "step": 2670}, {"loss": 1.1317, "grad_norm": 0.5292727947235107, "learning_rate": 0.0002, "epoch": 1.9655298863219657, "step": 2680}, {"loss": 1.1328, "grad_norm": 0.6012240648269653, "learning_rate": 0.0002, "epoch": 1.9728639530619727, "step": 2690}, {"loss": 1.0683, "grad_norm": 0.3945149779319763, "learning_rate": 0.0002, "epoch": 1.9801980198019802, "step": 2700}, {"loss": 1.0155, "grad_norm": 0.5732627511024475, "learning_rate": 0.0002, "epoch": 1.9875320865419877, "step": 2710}, {"loss": 0.9857, "grad_norm": 0.3963361084461212, "learning_rate": 0.0002, "epoch": 1.994866153281995, "step": 2720}, {"eval_loss": 1.1534006595611572, "eval_runtime": 32.7541, "eval_samples_per_second": 13.159, "eval_steps_per_second": 1.649, "epoch": 2.0, "step": 2727}, {"loss": 0.9624, "grad_norm": 0.48628315329551697, "learning_rate": 0.0002, "epoch": 2.002200220022002, "step": 2730}, {"loss": 0.9603, "grad_norm": 0.413875013589859, "learning_rate": 0.0002, "epoch": 2.0095342867620096, "step": 2740}, {"loss": 0.965, "grad_norm": 0.4988735616207123, "learning_rate": 0.0002, "epoch": 2.0168683535020167, "step": 2750}, {"loss": 0.9677, "grad_norm": 0.5634812712669373, "learning_rate": 0.0002, "epoch": 2.024202420242024, "step": 2760}, {"loss": 0.9547, "grad_norm": 0.48302653431892395, "learning_rate": 0.0002, "epoch": 2.0315364869820316, "step": 2770}, {"loss": 0.9346, "grad_norm": 0.49914175271987915, "learning_rate": 0.0002, "epoch": 2.038870553722039, "step": 2780}, {"loss": 0.904, "grad_norm": 1.14039945602417, "learning_rate": 0.0002, "epoch": 2.046204620462046, "step": 2790}, {"loss": 0.9588, "grad_norm": 0.6359720826148987, "learning_rate": 0.0002, "epoch": 2.0535386872020536, "step": 2800}, {"loss": 0.9031, "grad_norm": 0.4589158296585083, "learning_rate": 0.0002, "epoch": 2.060872753942061, "step": 2810}, {"loss": 0.9438, "grad_norm": 0.46255481243133545, "learning_rate": 0.0002, "epoch": 2.068206820682068, "step": 2820}, {"loss": 0.9464, "grad_norm": 0.6232137680053711, "learning_rate": 0.0002, "epoch": 2.0755408874220755, "step": 2830}, {"loss": 0.8978, "grad_norm": 0.41042178869247437, "learning_rate": 0.0002, "epoch": 2.082874954162083, "step": 2840}, {"loss": 0.8516, "grad_norm": 0.5334428548812866, "learning_rate": 0.0002, "epoch": 2.09020902090209, "step": 2850}, {"loss": 0.9313, "grad_norm": 0.8270058631896973, "learning_rate": 0.0002, "epoch": 2.0975430876420975, "step": 2860}, {"loss": 1.0064, "grad_norm": 0.6624533534049988, "learning_rate": 0.0002, "epoch": 2.104877154382105, "step": 2870}, {"loss": 0.9196, "grad_norm": 0.5448863506317139, "learning_rate": 0.0002, "epoch": 2.112211221122112, "step": 2880}, {"loss": 0.887, "grad_norm": 0.621482789516449, "learning_rate": 0.0002, "epoch": 2.1195452878621195, "step": 2890}, {"loss": 0.9702, "grad_norm": 0.4556255340576172, "learning_rate": 0.0002, "epoch": 2.126879354602127, "step": 2900}, {"loss": 0.9323, "grad_norm": 0.4620579183101654, "learning_rate": 0.0002, "epoch": 2.1342134213421344, "step": 2910}, {"loss": 0.836, "grad_norm": 0.9602415561676025, "learning_rate": 0.0002, "epoch": 2.1415474880821415, "step": 2920}, {"loss": 0.8826, "grad_norm": 0.587943971157074, "learning_rate": 0.0002, "epoch": 2.148881554822149, "step": 2930}, {"loss": 0.971, "grad_norm": 0.5121372938156128, "learning_rate": 0.0002, "epoch": 2.1562156215621564, "step": 2940}, {"loss": 0.8751, "grad_norm": 0.49424484372138977, "learning_rate": 0.0002, "epoch": 2.1635496883021634, "step": 2950}, {"loss": 0.8674, "grad_norm": 0.6312560439109802, "learning_rate": 0.0002, "epoch": 2.170883755042171, "step": 2960}, {"loss": 0.9791, "grad_norm": 0.5235576629638672, "learning_rate": 0.0002, "epoch": 2.1782178217821784, "step": 2970}, {"loss": 0.9706, "grad_norm": 0.5868439674377441, "learning_rate": 0.0002, "epoch": 2.1855518885221854, "step": 2980}, {"loss": 0.9338, "grad_norm": 0.42302873730659485, "learning_rate": 0.0002, "epoch": 2.192885955262193, "step": 2990}, {"loss": 0.9332, "grad_norm": 0.5097725987434387, "learning_rate": 0.0002, "epoch": 2.2002200220022003, "step": 3000}, {"loss": 0.9239, "grad_norm": 0.5091572403907776, "learning_rate": 0.0002, "epoch": 2.2075540887422074, "step": 3010}, {"loss": 0.8898, "grad_norm": 0.49433162808418274, "learning_rate": 0.0002, "epoch": 2.214888155482215, "step": 3020}, {"loss": 0.9734, "grad_norm": 0.5577368140220642, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3030}, {"loss": 0.9033, "grad_norm": 0.6177583932876587, "learning_rate": 0.0002, "epoch": 2.2295562889622293, "step": 3040}, {"loss": 0.9882, "grad_norm": 0.5256719589233398, "learning_rate": 0.0002, "epoch": 2.236890355702237, "step": 3050}, {"loss": 0.9439, "grad_norm": 0.5001118183135986, "learning_rate": 0.0002, "epoch": 2.2442244224422443, "step": 3060}, {"loss": 0.8718, "grad_norm": 0.5721249580383301, "learning_rate": 0.0002, "epoch": 2.2515584891822513, "step": 3070}, {"loss": 1.0648, "grad_norm": 0.5325384140014648, "learning_rate": 0.0002, "epoch": 2.258892555922259, "step": 3080}, {"loss": 0.9843, "grad_norm": 0.5719189047813416, "learning_rate": 0.0002, "epoch": 2.2662266226622663, "step": 3090}, {"loss": 0.8633, "grad_norm": 0.6337835788726807, "learning_rate": 0.0002, "epoch": 2.2735606894022737, "step": 3100}, {"loss": 0.9962, "grad_norm": 0.5381836891174316, "learning_rate": 0.0002, "epoch": 2.2808947561422808, "step": 3110}, {"loss": 0.8265, "grad_norm": 0.5408531427383423, "learning_rate": 0.0002, "epoch": 2.2882288228822882, "step": 3120}, {"loss": 1.0325, "grad_norm": 0.43705281615257263, "learning_rate": 0.0002, "epoch": 2.2955628896222957, "step": 3130}, {"loss": 0.9388, "grad_norm": 0.6454030275344849, "learning_rate": 0.0002, "epoch": 2.3028969563623027, "step": 3140}, {"loss": 0.954, "grad_norm": 0.686030387878418, "learning_rate": 0.0002, "epoch": 2.31023102310231, "step": 3150}, {"loss": 0.9403, "grad_norm": 0.5123633146286011, "learning_rate": 0.0002, "epoch": 2.3175650898423177, "step": 3160}, {"loss": 0.8834, "grad_norm": 0.842506468296051, "learning_rate": 0.0002, "epoch": 2.3248991565823247, "step": 3170}, {"loss": 1.0497, "grad_norm": 0.5193818807601929, "learning_rate": 0.0002, "epoch": 2.332233223322332, "step": 3180}, {"loss": 0.9473, "grad_norm": 0.5634409189224243, "learning_rate": 0.0002, "epoch": 2.3395672900623397, "step": 3190}, {"loss": 0.8499, "grad_norm": 0.6475534439086914, "learning_rate": 0.0002, "epoch": 2.3469013568023467, "step": 3200}, {"loss": 0.874, "grad_norm": 1.1503914594650269, "learning_rate": 0.0002, "epoch": 2.354235423542354, "step": 3210}, {"loss": 0.9762, "grad_norm": 0.7234905362129211, "learning_rate": 0.0002, "epoch": 2.3615694902823616, "step": 3220}, {"loss": 0.9007, "grad_norm": 0.664903461933136, "learning_rate": 0.0002, "epoch": 2.368903557022369, "step": 3230}, {"loss": 0.9987, "grad_norm": 0.5453006625175476, "learning_rate": 0.0002, "epoch": 2.376237623762376, "step": 3240}, {"loss": 0.9742, "grad_norm": 0.6256654262542725, "learning_rate": 0.0002, "epoch": 2.3835716905023836, "step": 3250}, {"loss": 0.9922, "grad_norm": 0.5166565179824829, "learning_rate": 0.0002, "epoch": 2.390905757242391, "step": 3260}, {"loss": 0.927, "grad_norm": 0.5699098110198975, "learning_rate": 0.0002, "epoch": 2.398239823982398, "step": 3270}, {"loss": 0.8878, "grad_norm": 0.4472540020942688, "learning_rate": 0.0002, "epoch": 2.4055738907224056, "step": 3280}, {"loss": 0.9439, "grad_norm": 0.6790403127670288, "learning_rate": 0.0002, "epoch": 2.412907957462413, "step": 3290}, {"loss": 0.972, "grad_norm": 0.5182185173034668, "learning_rate": 0.0002, "epoch": 2.42024202420242, "step": 3300}, {"loss": 0.9775, "grad_norm": 0.564647912979126, "learning_rate": 0.0002, "epoch": 2.4275760909424275, "step": 3310}, {"loss": 1.072, "grad_norm": 0.5625313520431519, "learning_rate": 0.0002, "epoch": 2.434910157682435, "step": 3320}, {"loss": 0.8798, "grad_norm": 0.7496559619903564, "learning_rate": 0.0002, "epoch": 2.442244224422442, "step": 3330}, {"loss": 0.868, "grad_norm": 0.4779128134250641, "learning_rate": 0.0002, "epoch": 2.4495782911624495, "step": 3340}, {"loss": 1.0316, "grad_norm": 0.578093409538269, "learning_rate": 0.0002, "epoch": 2.456912357902457, "step": 3350}, {"loss": 0.9282, "grad_norm": 0.5456080436706543, "learning_rate": 0.0002, "epoch": 2.4642464246424645, "step": 3360}, {"loss": 0.8409, "grad_norm": 0.4769273102283478, "learning_rate": 0.0002, "epoch": 2.4715804913824715, "step": 3370}, {"loss": 0.9312, "grad_norm": 0.5608189702033997, "learning_rate": 0.0002, "epoch": 2.478914558122479, "step": 3380}, {"loss": 0.9934, "grad_norm": 0.5590165853500366, "learning_rate": 0.0002, "epoch": 2.4862486248624864, "step": 3390}, {"loss": 1.025, "grad_norm": 0.801306962966919, "learning_rate": 0.0002, "epoch": 2.4935826916024935, "step": 3400}, {"loss": 0.9049, "grad_norm": 0.6045624613761902, "learning_rate": 0.0002, "epoch": 2.500916758342501, "step": 3410}, {"loss": 0.944, "grad_norm": 0.5735858082771301, "learning_rate": 0.0002, "epoch": 2.5082508250825084, "step": 3420}, {"loss": 0.9846, "grad_norm": 0.6827309131622314, "learning_rate": 0.0002, "epoch": 2.5155848918225154, "step": 3430}, {"loss": 0.9789, "grad_norm": 0.5702602863311768, "learning_rate": 0.0002, "epoch": 2.522918958562523, "step": 3440}, {"loss": 0.9127, "grad_norm": 0.6674721240997314, "learning_rate": 0.0002, "epoch": 2.5302530253025304, "step": 3450}, {"loss": 0.914, "grad_norm": 0.5635907649993896, "learning_rate": 0.0002, "epoch": 2.5375870920425374, "step": 3460}, {"loss": 0.8398, "grad_norm": 0.42737770080566406, "learning_rate": 0.0002, "epoch": 2.544921158782545, "step": 3470}, {"loss": 0.9474, "grad_norm": 0.6720691919326782, "learning_rate": 0.0002, "epoch": 2.5522552255225524, "step": 3480}, {"loss": 0.8637, "grad_norm": 0.8917084336280823, "learning_rate": 0.0002, "epoch": 2.55958929226256, "step": 3490}, {"loss": 0.9257, "grad_norm": 0.5134549140930176, "learning_rate": 0.0002, "epoch": 2.566923359002567, "step": 3500}, {"loss": 0.9362, "grad_norm": 0.4951367974281311, "learning_rate": 0.0002, "epoch": 2.5742574257425743, "step": 3510}, {"loss": 0.9184, "grad_norm": 0.9438204765319824, "learning_rate": 0.0002, "epoch": 2.5815914924825814, "step": 3520}, {"loss": 0.8939, "grad_norm": 0.6024714708328247, "learning_rate": 0.0002, "epoch": 2.588925559222589, "step": 3530}, {"loss": 0.9298, "grad_norm": 0.5248535871505737, "learning_rate": 0.0002, "epoch": 2.5962596259625963, "step": 3540}, {"loss": 0.941, "grad_norm": 0.8677568435668945, "learning_rate": 0.0002, "epoch": 2.6035936927026038, "step": 3550}, {"loss": 0.9253, "grad_norm": 0.82008296251297, "learning_rate": 0.0002, "epoch": 2.610927759442611, "step": 3560}, {"loss": 0.8429, "grad_norm": 0.4724634885787964, "learning_rate": 0.0002, "epoch": 2.6182618261826183, "step": 3570}, {"loss": 0.9058, "grad_norm": 0.5434244275093079, "learning_rate": 0.0002, "epoch": 2.6255958929226257, "step": 3580}, {"loss": 0.9379, "grad_norm": 0.4948740005493164, "learning_rate": 0.0002, "epoch": 2.6329299596626328, "step": 3590}, {"loss": 0.8718, "grad_norm": 0.42109328508377075, "learning_rate": 0.0002, "epoch": 2.6402640264026402, "step": 3600}, {"loss": 0.9809, "grad_norm": 0.7979786396026611, "learning_rate": 0.0002, "epoch": 2.6475980931426477, "step": 3610}, {"loss": 0.9229, "grad_norm": 0.6345919370651245, "learning_rate": 0.0002, "epoch": 2.654932159882655, "step": 3620}, {"loss": 0.8506, "grad_norm": 0.4971671402454376, "learning_rate": 0.0002, "epoch": 2.662266226622662, "step": 3630}, {"loss": 0.8054, "grad_norm": 0.6467748284339905, "learning_rate": 0.0002, "epoch": 2.6696002933626697, "step": 3640}, {"loss": 0.9277, "grad_norm": 0.4240160286426544, "learning_rate": 0.0002, "epoch": 2.6769343601026767, "step": 3650}, {"loss": 0.8213, "grad_norm": 0.5179754495620728, "learning_rate": 0.0002, "epoch": 2.684268426842684, "step": 3660}, {"loss": 0.9221, "grad_norm": 0.754012405872345, "learning_rate": 0.0002, "epoch": 2.6916024935826917, "step": 3670}, {"loss": 0.9194, "grad_norm": 0.5141299962997437, "learning_rate": 0.0002, "epoch": 2.698936560322699, "step": 3680}, {"loss": 0.9495, "grad_norm": 0.5737819075584412, "learning_rate": 0.0002, "epoch": 2.706270627062706, "step": 3690}, {"loss": 1.0162, "grad_norm": 0.5887577533721924, "learning_rate": 0.0002, "epoch": 2.7136046938027136, "step": 3700}, {"loss": 0.9169, "grad_norm": 0.6740471720695496, "learning_rate": 0.0002, "epoch": 2.720938760542721, "step": 3710}, {"loss": 0.9297, "grad_norm": 0.5879453420639038, "learning_rate": 0.0002, "epoch": 2.728272827282728, "step": 3720}, {"loss": 0.9358, "grad_norm": 0.4858354926109314, "learning_rate": 0.0002, "epoch": 2.7356068940227356, "step": 3730}, {"loss": 0.9308, "grad_norm": 0.5489001870155334, "learning_rate": 0.0002, "epoch": 2.742940960762743, "step": 3740}, {"loss": 0.894, "grad_norm": 0.8187092542648315, "learning_rate": 0.0002, "epoch": 2.7502750275027505, "step": 3750}, {"loss": 0.8954, "grad_norm": 0.5666626691818237, "learning_rate": 0.0002, "epoch": 2.7576090942427576, "step": 3760}, {"loss": 1.0059, "grad_norm": 0.5377066135406494, "learning_rate": 0.0002, "epoch": 2.764943160982765, "step": 3770}, {"loss": 0.9132, "grad_norm": 0.566330075263977, "learning_rate": 0.0002, "epoch": 2.772277227722772, "step": 3780}, {"loss": 0.9415, "grad_norm": 0.5522832870483398, "learning_rate": 0.0002, "epoch": 2.7796112944627795, "step": 3790}, {"loss": 0.8816, "grad_norm": 0.5668695569038391, "learning_rate": 0.0002, "epoch": 2.786945361202787, "step": 3800}, {"loss": 0.8885, "grad_norm": 0.7566602826118469, "learning_rate": 0.0002, "epoch": 2.7942794279427945, "step": 3810}, {"loss": 0.8598, "grad_norm": 0.5603684782981873, "learning_rate": 0.0002, "epoch": 2.8016134946828015, "step": 3820}, {"loss": 0.9602, "grad_norm": 0.49122217297554016, "learning_rate": 0.0002, "epoch": 2.808947561422809, "step": 3830}, {"loss": 0.9738, "grad_norm": 0.6798251867294312, "learning_rate": 0.0002, "epoch": 2.816281628162816, "step": 3840}, {"loss": 0.9533, "grad_norm": 0.6097991466522217, "learning_rate": 0.0002, "epoch": 2.8236156949028235, "step": 3850}, {"loss": 0.8672, "grad_norm": 0.6675726175308228, "learning_rate": 0.0002, "epoch": 2.830949761642831, "step": 3860}, {"loss": 0.9324, "grad_norm": 0.9223952889442444, "learning_rate": 0.0002, "epoch": 2.8382838283828384, "step": 3870}, {"loss": 0.8767, "grad_norm": 0.6020799875259399, "learning_rate": 0.0002, "epoch": 2.8456178951228455, "step": 3880}, {"loss": 0.9148, "grad_norm": 0.5206381678581238, "learning_rate": 0.0002, "epoch": 2.852951961862853, "step": 3890}, {"loss": 0.9479, "grad_norm": 0.6268777251243591, "learning_rate": 0.0002, "epoch": 2.8602860286028604, "step": 3900}, {"loss": 0.9409, "grad_norm": 1.1583497524261475, "learning_rate": 0.0002, "epoch": 2.8676200953428674, "step": 3910}, {"loss": 0.895, "grad_norm": 0.7263903021812439, "learning_rate": 0.0002, "epoch": 2.874954162082875, "step": 3920}, {"loss": 0.8786, "grad_norm": 0.5369910001754761, "learning_rate": 0.0002, "epoch": 2.8822882288228824, "step": 3930}, {"loss": 1.0015, "grad_norm": 0.7298350930213928, "learning_rate": 0.0002, "epoch": 2.88962229556289, "step": 3940}, {"loss": 0.979, "grad_norm": 0.577012836933136, "learning_rate": 0.0002, "epoch": 2.896956362302897, "step": 3950}, {"loss": 0.9716, "grad_norm": 0.5859594345092773, "learning_rate": 0.0002, "epoch": 2.9042904290429044, "step": 3960}, {"loss": 0.8772, "grad_norm": 0.47176122665405273, "learning_rate": 0.0002, "epoch": 2.9116244957829114, "step": 3970}, {"loss": 0.8997, "grad_norm": 0.9699620604515076, "learning_rate": 0.0002, "epoch": 2.918958562522919, "step": 3980}, {"loss": 0.9057, "grad_norm": 0.7908747792243958, "learning_rate": 0.0002, "epoch": 2.9262926292629263, "step": 3990}, {"loss": 0.9462, "grad_norm": 0.5777379274368286, "learning_rate": 0.0002, "epoch": 2.933626696002934, "step": 4000}, {"loss": 0.9358, "grad_norm": 0.599288284778595, "learning_rate": 0.0002, "epoch": 2.940960762742941, "step": 4010}, {"loss": 0.9812, "grad_norm": 0.5232274532318115, "learning_rate": 0.0002, "epoch": 2.9482948294829483, "step": 4020}, {"loss": 0.96, "grad_norm": 0.6395137310028076, "learning_rate": 0.0002, "epoch": 2.9556288962229558, "step": 4030}, {"loss": 0.9813, "grad_norm": 0.589260458946228, "learning_rate": 0.0002, "epoch": 2.962962962962963, "step": 4040}, {"loss": 0.9541, "grad_norm": 0.5699581503868103, "learning_rate": 0.0002, "epoch": 2.9702970297029703, "step": 4050}, {"loss": 0.9585, "grad_norm": 0.528468132019043, "learning_rate": 0.0002, "epoch": 2.9776310964429777, "step": 4060}, {"loss": 0.9164, "grad_norm": 0.4804670512676239, "learning_rate": 0.0002, "epoch": 2.984965163182985, "step": 4070}, {"loss": 0.9771, "grad_norm": 1.1918889284133911, "learning_rate": 0.0002, "epoch": 2.9922992299229922, "step": 4080}, {"loss": 0.9178, "grad_norm": 0.5479103326797485, "learning_rate": 0.0002, "epoch": 2.9996332966629997, "step": 4090}, {"eval_loss": 1.1642853021621704, "eval_runtime": 32.7511, "eval_samples_per_second": 13.16, "eval_steps_per_second": 1.649, "epoch": 2.9996332966629997, "step": 4090}, {"loss": 0.7981, "grad_norm": 0.7430027723312378, "learning_rate": 0.0002, "epoch": 3.006967363403007, "step": 4100}, {"loss": 0.7871, "grad_norm": 0.6293647289276123, "learning_rate": 0.0002, "epoch": 3.014301430143014, "step": 4110}, {"loss": 0.78, "grad_norm": 0.6191329956054688, "learning_rate": 0.0002, "epoch": 3.0216354968830217, "step": 4120}, {"loss": 0.7618, "grad_norm": 0.7959313988685608, "learning_rate": 0.0002, "epoch": 3.028969563623029, "step": 4130}, {"loss": 0.8039, "grad_norm": 0.5956351161003113, "learning_rate": 0.0002, "epoch": 3.036303630363036, "step": 4140}, {"loss": 0.7477, "grad_norm": 0.670383632183075, "learning_rate": 0.0002, "epoch": 3.0436376971030437, "step": 4150}, {"loss": 0.7984, "grad_norm": 0.6414518356323242, "learning_rate": 0.0002, "epoch": 3.050971763843051, "step": 4160}, {"loss": 0.7369, "grad_norm": 0.7928852438926697, "learning_rate": 0.0002, "epoch": 3.058305830583058, "step": 4170}, {"loss": 0.7914, "grad_norm": 0.6211121082305908, "learning_rate": 0.0002, "epoch": 3.0656398973230656, "step": 4180}, {"loss": 0.7365, "grad_norm": 0.6237057447433472, "learning_rate": 0.0002, "epoch": 3.072973964063073, "step": 4190}, {"loss": 0.702, "grad_norm": 0.6522233486175537, "learning_rate": 0.0002, "epoch": 3.08030803080308, "step": 4200}, {"loss": 0.7646, "grad_norm": 0.9396848678588867, "learning_rate": 0.0002, "epoch": 3.0876420975430876, "step": 4210}, {"loss": 0.7559, "grad_norm": 0.8003010749816895, "learning_rate": 0.0002, "epoch": 3.094976164283095, "step": 4220}, {"loss": 0.711, "grad_norm": 0.6733810305595398, "learning_rate": 0.0002, "epoch": 3.102310231023102, "step": 4230}, {"loss": 0.696, "grad_norm": 0.6365828514099121, "learning_rate": 0.0002, "epoch": 3.1096442977631096, "step": 4240}, {"loss": 0.8362, "grad_norm": 1.0805548429489136, "learning_rate": 0.0002, "epoch": 3.116978364503117, "step": 4250}, {"loss": 0.7651, "grad_norm": 0.7262141108512878, "learning_rate": 0.0002, "epoch": 3.1243124312431245, "step": 4260}, {"loss": 0.7304, "grad_norm": 0.5500539541244507, "learning_rate": 0.0002, "epoch": 3.1316464979831315, "step": 4270}, {"loss": 0.7721, "grad_norm": 0.793912947177887, "learning_rate": 0.0002, "epoch": 3.138980564723139, "step": 4280}, {"loss": 0.7708, "grad_norm": 1.2540518045425415, "learning_rate": 0.0002, "epoch": 3.1463146314631465, "step": 4290}, {"loss": 0.782, "grad_norm": 0.7020077705383301, "learning_rate": 0.0002, "epoch": 3.1536486982031535, "step": 4300}, {"loss": 0.7253, "grad_norm": 0.5111123323440552, "learning_rate": 0.0002, "epoch": 3.160982764943161, "step": 4310}, {"loss": 0.8159, "grad_norm": 0.7172090411186218, "learning_rate": 0.0002, "epoch": 3.1683168316831685, "step": 4320}, {"loss": 0.6962, "grad_norm": 0.6343168616294861, "learning_rate": 0.0002, "epoch": 3.1756508984231755, "step": 4330}, {"loss": 0.7938, "grad_norm": 0.9563672542572021, "learning_rate": 0.0002, "epoch": 3.182984965163183, "step": 4340}, {"loss": 0.7385, "grad_norm": 1.0225574970245361, "learning_rate": 0.0002, "epoch": 3.1903190319031904, "step": 4350}, {"loss": 0.8652, "grad_norm": 1.1633386611938477, "learning_rate": 0.0002, "epoch": 3.1976530986431975, "step": 4360}, {"loss": 0.7259, "grad_norm": 0.8915148973464966, "learning_rate": 0.0002, "epoch": 3.204987165383205, "step": 4370}, {"loss": 0.8061, "grad_norm": 0.9156812429428101, "learning_rate": 0.0002, "epoch": 3.2123212321232124, "step": 4380}, {"loss": 0.8189, "grad_norm": 0.6363258957862854, "learning_rate": 0.0002, "epoch": 3.21965529886322, "step": 4390}, {"loss": 0.7996, "grad_norm": 0.579099178314209, "learning_rate": 0.0002, "epoch": 3.226989365603227, "step": 4400}, {"loss": 0.8592, "grad_norm": 0.8778146505355835, "learning_rate": 0.0002, "epoch": 3.2343234323432344, "step": 4410}, {"loss": 0.8281, "grad_norm": 0.8356770873069763, "learning_rate": 0.0002, "epoch": 3.241657499083242, "step": 4420}, {"loss": 0.8484, "grad_norm": 0.702032208442688, "learning_rate": 0.0002, "epoch": 3.248991565823249, "step": 4430}, {"loss": 0.7227, "grad_norm": 0.6386539340019226, "learning_rate": 0.0002, "epoch": 3.2563256325632564, "step": 4440}, {"loss": 0.8374, "grad_norm": 0.7008408904075623, "learning_rate": 0.0002, "epoch": 3.263659699303264, "step": 4450}, {"loss": 0.7572, "grad_norm": 0.9556332230567932, "learning_rate": 0.0002, "epoch": 3.270993766043271, "step": 4460}, {"loss": 0.743, "grad_norm": 0.5667835474014282, "learning_rate": 0.0002, "epoch": 3.2783278327832783, "step": 4470}, {"loss": 0.8152, "grad_norm": 0.8239172697067261, "learning_rate": 0.0002, "epoch": 3.285661899523286, "step": 4480}, {"loss": 0.756, "grad_norm": 0.7045050859451294, "learning_rate": 0.0002, "epoch": 3.292995966263293, "step": 4490}, {"loss": 0.7655, "grad_norm": 0.7131434082984924, "learning_rate": 0.0002, "epoch": 3.3003300330033003, "step": 4500}, {"loss": 0.836, "grad_norm": 0.6924910545349121, "learning_rate": 0.0002, "epoch": 3.3076640997433078, "step": 4510}, {"loss": 0.736, "grad_norm": 0.8945356607437134, "learning_rate": 0.0002, "epoch": 3.3149981664833152, "step": 4520}, {"loss": 0.7575, "grad_norm": 0.6546903252601624, "learning_rate": 0.0002, "epoch": 3.3223322332233223, "step": 4530}, {"loss": 0.7893, "grad_norm": 0.8206679224967957, "learning_rate": 0.0002, "epoch": 3.3296662999633297, "step": 4540}, {"loss": 0.7502, "grad_norm": 0.6482203602790833, "learning_rate": 0.0002, "epoch": 3.3370003667033368, "step": 4550}, {"loss": 0.8172, "grad_norm": 0.7558760046958923, "learning_rate": 0.0002, "epoch": 3.3443344334433442, "step": 4560}, {"loss": 0.744, "grad_norm": 0.7794756889343262, "learning_rate": 0.0002, "epoch": 3.3516685001833517, "step": 4570}, {"loss": 0.7385, "grad_norm": 0.7382805943489075, "learning_rate": 0.0002, "epoch": 3.359002566923359, "step": 4580}, {"loss": 0.8511, "grad_norm": 0.5912511944770813, "learning_rate": 0.0002, "epoch": 3.366336633663366, "step": 4590}, {"loss": 0.8272, "grad_norm": 0.7444885969161987, "learning_rate": 0.0002, "epoch": 3.3736707004033737, "step": 4600}, {"loss": 0.7927, "grad_norm": 0.7354922890663147, "learning_rate": 0.0002, "epoch": 3.381004767143381, "step": 4610}, {"loss": 0.7183, "grad_norm": 0.7685934901237488, "learning_rate": 0.0002, "epoch": 3.388338833883388, "step": 4620}, {"loss": 0.7436, "grad_norm": 0.61041259765625, "learning_rate": 0.0002, "epoch": 3.3956729006233957, "step": 4630}, {"loss": 0.7661, "grad_norm": 0.6820451021194458, "learning_rate": 0.0002, "epoch": 3.403006967363403, "step": 4640}, {"loss": 0.8796, "grad_norm": 0.5819534063339233, "learning_rate": 0.0002, "epoch": 3.41034103410341, "step": 4650}, {"loss": 0.7314, "grad_norm": 0.705410897731781, "learning_rate": 0.0002, "epoch": 3.4176751008434176, "step": 4660}, {"loss": 0.7901, "grad_norm": 0.8052892088890076, "learning_rate": 0.0002, "epoch": 3.425009167583425, "step": 4670}, {"loss": 0.7298, "grad_norm": 0.7746483087539673, "learning_rate": 0.0002, "epoch": 3.432343234323432, "step": 4680}, {"loss": 0.7976, "grad_norm": 0.7713689804077148, "learning_rate": 0.0002, "epoch": 3.4396773010634396, "step": 4690}, {"loss": 0.7427, "grad_norm": 0.810371994972229, "learning_rate": 0.0002, "epoch": 3.447011367803447, "step": 4700}, {"loss": 0.7594, "grad_norm": 0.7702969312667847, "learning_rate": 0.0002, "epoch": 3.4543454345434546, "step": 4710}, {"loss": 0.7957, "grad_norm": 0.7069268822669983, "learning_rate": 0.0002, "epoch": 3.4616795012834616, "step": 4720}, {"loss": 0.8199, "grad_norm": 0.7640359401702881, "learning_rate": 0.0002, "epoch": 3.469013568023469, "step": 4730}, {"loss": 0.6875, "grad_norm": 0.8661707639694214, "learning_rate": 0.0002, "epoch": 3.4763476347634765, "step": 4740}, {"loss": 0.8528, "grad_norm": 0.9970282912254333, "learning_rate": 0.0002, "epoch": 3.4836817015034836, "step": 4750}, {"loss": 0.8462, "grad_norm": 0.5824355483055115, "learning_rate": 0.0002, "epoch": 3.491015768243491, "step": 4760}, {"loss": 0.851, "grad_norm": 1.3072649240493774, "learning_rate": 0.0002, "epoch": 3.4983498349834985, "step": 4770}, {"loss": 0.9101, "grad_norm": 0.873978316783905, "learning_rate": 0.0002, "epoch": 3.5056839017235055, "step": 4780}, {"loss": 0.7403, "grad_norm": 0.5526657104492188, "learning_rate": 0.0002, "epoch": 3.513017968463513, "step": 4790}, {"loss": 0.7921, "grad_norm": 0.790894627571106, "learning_rate": 0.0002, "epoch": 3.5203520352035205, "step": 4800}, {"loss": 0.831, "grad_norm": 0.8119630217552185, "learning_rate": 0.0002, "epoch": 3.5276861019435275, "step": 4810}, {"loss": 0.7351, "grad_norm": 0.633212149143219, "learning_rate": 0.0002, "epoch": 3.535020168683535, "step": 4820}, {"loss": 0.8505, "grad_norm": 0.703029990196228, "learning_rate": 0.0002, "epoch": 3.5423542354235424, "step": 4830}, {"loss": 0.7204, "grad_norm": 0.7603771686553955, "learning_rate": 0.0002, "epoch": 3.54968830216355, "step": 4840}, {"loss": 0.8868, "grad_norm": 0.6260480880737305, "learning_rate": 0.0002, "epoch": 3.557022368903557, "step": 4850}, {"loss": 0.8137, "grad_norm": 0.8203664422035217, "learning_rate": 0.0002, "epoch": 3.5643564356435644, "step": 4860}, {"loss": 0.8821, "grad_norm": 0.7793813347816467, "learning_rate": 0.0002, "epoch": 3.5716905023835714, "step": 4870}, {"loss": 0.8164, "grad_norm": 0.7667397260665894, "learning_rate": 0.0002, "epoch": 3.579024569123579, "step": 4880}, {"loss": 0.7597, "grad_norm": 0.8198829889297485, "learning_rate": 0.0002, "epoch": 3.5863586358635864, "step": 4890}, {"loss": 0.7027, "grad_norm": 0.7689233422279358, "learning_rate": 0.0002, "epoch": 3.593692702603594, "step": 4900}, {"loss": 0.804, "grad_norm": 0.7870983481407166, "learning_rate": 0.0002, "epoch": 3.601026769343601, "step": 4910}, {"loss": 0.8269, "grad_norm": 0.8133853077888489, "learning_rate": 0.0002, "epoch": 3.6083608360836084, "step": 4920}, {"loss": 0.8515, "grad_norm": 1.308401346206665, "learning_rate": 0.0002, "epoch": 3.615694902823616, "step": 4930}, {"loss": 0.8494, "grad_norm": 0.7131121754646301, "learning_rate": 0.0002, "epoch": 3.623028969563623, "step": 4940}, {"loss": 0.7235, "grad_norm": 0.6825910210609436, "learning_rate": 0.0002, "epoch": 3.6303630363036303, "step": 4950}, {"loss": 0.7824, "grad_norm": 0.7254678606987, "learning_rate": 0.0002, "epoch": 3.637697103043638, "step": 4960}, {"loss": 0.7983, "grad_norm": 0.8045085072517395, "learning_rate": 0.0002, "epoch": 3.6450311697836453, "step": 4970}, {"loss": 0.8223, "grad_norm": 0.6991777420043945, "learning_rate": 0.0002, "epoch": 3.6523652365236523, "step": 4980}, {"loss": 0.7806, "grad_norm": 0.7804713249206543, "learning_rate": 0.0002, "epoch": 3.6596993032636598, "step": 4990}, {"loss": 0.8402, "grad_norm": 0.8525708317756653, "learning_rate": 0.0002, "epoch": 3.667033370003667, "step": 5000}, {"loss": 0.8496, "grad_norm": 0.7959994673728943, "learning_rate": 0.0002, "epoch": 3.6743674367436743, "step": 5010}, {"loss": 0.8022, "grad_norm": 0.8103628158569336, "learning_rate": 0.0002, "epoch": 3.6817015034836817, "step": 5020}, {"loss": 0.7376, "grad_norm": 0.7517836093902588, "learning_rate": 0.0002, "epoch": 3.689035570223689, "step": 5030}, {"loss": 0.8375, "grad_norm": 0.6878514289855957, "learning_rate": 0.0002, "epoch": 3.6963696369636962, "step": 5040}, {"loss": 0.7998, "grad_norm": 1.2371820211410522, "learning_rate": 0.0002, "epoch": 3.7037037037037037, "step": 5050}, {"loss": 0.6941, "grad_norm": 0.6567103862762451, "learning_rate": 0.0002, "epoch": 3.711037770443711, "step": 5060}, {"loss": 0.8465, "grad_norm": 1.1254922151565552, "learning_rate": 0.0002, "epoch": 3.718371837183718, "step": 5070}, {"loss": 0.8365, "grad_norm": 0.6796132326126099, "learning_rate": 0.0002, "epoch": 3.7257059039237257, "step": 5080}, {"loss": 0.7818, "grad_norm": 0.7285300493240356, "learning_rate": 0.0002, "epoch": 3.733039970663733, "step": 5090}, {"loss": 0.8581, "grad_norm": 0.8931500911712646, "learning_rate": 0.0002, "epoch": 3.7403740374037406, "step": 5100}, {"loss": 0.8181, "grad_norm": 0.6256856918334961, "learning_rate": 0.0002, "epoch": 3.7477081041437477, "step": 5110}, {"loss": 0.743, "grad_norm": 0.79310142993927, "learning_rate": 0.0002, "epoch": 3.755042170883755, "step": 5120}, {"loss": 0.8235, "grad_norm": 0.6594041585922241, "learning_rate": 0.0002, "epoch": 3.762376237623762, "step": 5130}, {"loss": 0.6925, "grad_norm": 0.7029327750205994, "learning_rate": 0.0002, "epoch": 3.7697103043637696, "step": 5140}, {"loss": 0.7457, "grad_norm": 0.5880070328712463, "learning_rate": 0.0002, "epoch": 3.777044371103777, "step": 5150}, {"loss": 0.8716, "grad_norm": 0.7578945159912109, "learning_rate": 0.0002, "epoch": 3.7843784378437846, "step": 5160}, {"loss": 0.8819, "grad_norm": 0.8276378512382507, "learning_rate": 0.0002, "epoch": 3.7917125045837916, "step": 5170}, {"loss": 0.7559, "grad_norm": 0.7627953886985779, "learning_rate": 0.0002, "epoch": 3.799046571323799, "step": 5180}, {"loss": 0.7665, "grad_norm": 0.8169086575508118, "learning_rate": 0.0002, "epoch": 3.806380638063806, "step": 5190}, {"loss": 0.761, "grad_norm": 0.6605030298233032, "learning_rate": 0.0002, "epoch": 3.8137147048038136, "step": 5200}, {"loss": 0.8804, "grad_norm": 0.5837286114692688, "learning_rate": 0.0002, "epoch": 3.821048771543821, "step": 5210}, {"loss": 0.8369, "grad_norm": 1.2422157526016235, "learning_rate": 0.0002, "epoch": 3.8283828382838285, "step": 5220}, {"loss": 0.8431, "grad_norm": 0.6589220762252808, "learning_rate": 0.0002, "epoch": 3.8357169050238356, "step": 5230}, {"loss": 0.7686, "grad_norm": 0.8567556142807007, "learning_rate": 0.0002, "epoch": 3.843050971763843, "step": 5240}, {"loss": 0.8652, "grad_norm": 0.6490627527236938, "learning_rate": 0.0002, "epoch": 3.8503850385038505, "step": 5250}, {"loss": 0.7386, "grad_norm": 0.620232880115509, "learning_rate": 0.0002, "epoch": 3.8577191052438575, "step": 5260}, {"loss": 0.9192, "grad_norm": 0.7685128450393677, "learning_rate": 0.0002, "epoch": 3.865053171983865, "step": 5270}, {"loss": 0.872, "grad_norm": 0.8113296627998352, "learning_rate": 0.0002, "epoch": 3.8723872387238725, "step": 5280}, {"loss": 0.7156, "grad_norm": 0.8092675805091858, "learning_rate": 0.0002, "epoch": 3.87972130546388, "step": 5290}, {"loss": 0.7325, "grad_norm": 0.583570122718811, "learning_rate": 0.0002, "epoch": 3.887055372203887, "step": 5300}, {"loss": 0.9333, "grad_norm": 1.712363600730896, "learning_rate": 0.0002, "epoch": 3.8943894389438944, "step": 5310}, {"loss": 0.7537, "grad_norm": 0.6673534512519836, "learning_rate": 0.0002, "epoch": 3.9017235056839015, "step": 5320}, {"loss": 0.7035, "grad_norm": 1.9770312309265137, "learning_rate": 0.0002, "epoch": 3.909057572423909, "step": 5330}, {"loss": 0.8793, "grad_norm": 0.6430999636650085, "learning_rate": 0.0002, "epoch": 3.9163916391639164, "step": 5340}, {"loss": 0.839, "grad_norm": 1.0159571170806885, "learning_rate": 0.0002, "epoch": 3.923725705903924, "step": 5350}, {"loss": 0.9332, "grad_norm": 0.8607584834098816, "learning_rate": 0.0002, "epoch": 3.931059772643931, "step": 5360}, {"loss": 0.7261, "grad_norm": 0.6967900991439819, "learning_rate": 0.0002, "epoch": 3.9383938393839384, "step": 5370}, {"loss": 0.8456, "grad_norm": 0.7683077454566956, "learning_rate": 0.0002, "epoch": 3.945727906123946, "step": 5380}, {"loss": 0.7682, "grad_norm": 0.6805762648582458, "learning_rate": 0.0002, "epoch": 3.953061972863953, "step": 5390}, {"loss": 0.7746, "grad_norm": 0.7033619284629822, "learning_rate": 0.0002, "epoch": 3.9603960396039604, "step": 5400}, {"loss": 0.8393, "grad_norm": 0.966112494468689, "learning_rate": 0.0002, "epoch": 3.967730106343968, "step": 5410}, {"loss": 0.8316, "grad_norm": 0.8467881083488464, "learning_rate": 0.0002, "epoch": 3.9750641730839753, "step": 5420}, {"loss": 0.8084, "grad_norm": 0.8005317449569702, "learning_rate": 0.0002, "epoch": 3.9823982398239823, "step": 5430}, {"loss": 0.7168, "grad_norm": 1.1615241765975952, "learning_rate": 0.0002, "epoch": 3.98973230656399, "step": 5440}, {"loss": 0.8263, "grad_norm": 0.6121614575386047, "learning_rate": 0.0002, "epoch": 3.997066373303997, "step": 5450}, {"eval_loss": 1.1834222078323364, "eval_runtime": 32.7569, "eval_samples_per_second": 13.158, "eval_steps_per_second": 1.649, "epoch": 4.0, "step": 5454}, {"loss": 0.7267, "grad_norm": 0.6055727005004883, "learning_rate": 0.0002, "epoch": 4.004400440044004, "step": 5460}, {"loss": 0.5766, "grad_norm": 0.8232647180557251, "learning_rate": 0.0002, "epoch": 4.011734506784012, "step": 5470}, {"loss": 0.6489, "grad_norm": 0.7739192247390747, "learning_rate": 0.0002, "epoch": 4.019068573524019, "step": 5480}, {"loss": 0.5978, "grad_norm": 0.6264950633049011, "learning_rate": 0.0002, "epoch": 4.026402640264027, "step": 5490}, {"loss": 0.6392, "grad_norm": 1.4798702001571655, "learning_rate": 0.0002, "epoch": 4.033736707004033, "step": 5500}, {"loss": 0.6143, "grad_norm": 0.9538470506668091, "learning_rate": 0.0002, "epoch": 4.041070773744041, "step": 5510}, {"loss": 0.6056, "grad_norm": 0.834561288356781, "learning_rate": 0.0002, "epoch": 4.048404840484048, "step": 5520}, {"loss": 0.6077, "grad_norm": 0.6407850384712219, "learning_rate": 0.0002, "epoch": 4.055738907224056, "step": 5530}, {"loss": 0.6733, "grad_norm": 0.9035961627960205, "learning_rate": 0.0002, "epoch": 4.063072973964063, "step": 5540}, {"loss": 0.5854, "grad_norm": 0.842812716960907, "learning_rate": 0.0002, "epoch": 4.070407040704071, "step": 5550}, {"loss": 0.654, "grad_norm": 0.8197882175445557, "learning_rate": 0.0002, "epoch": 4.077741107444078, "step": 5560}, {"loss": 0.5919, "grad_norm": 0.8652673959732056, "learning_rate": 0.0002, "epoch": 4.085075174184085, "step": 5570}, {"loss": 0.6188, "grad_norm": 0.8048318028450012, "learning_rate": 0.0002, "epoch": 4.092409240924092, "step": 5580}, {"loss": 0.6487, "grad_norm": 0.9604969024658203, "learning_rate": 0.0002, "epoch": 4.0997433076641, "step": 5590}, {"loss": 0.6356, "grad_norm": 1.244756817817688, "learning_rate": 0.0002, "epoch": 4.107077374404107, "step": 5600}, {"loss": 0.6489, "grad_norm": 0.7975269556045532, "learning_rate": 0.0002, "epoch": 4.114411441144115, "step": 5610}, {"loss": 0.6445, "grad_norm": 0.6130099296569824, "learning_rate": 0.0002, "epoch": 4.121745507884122, "step": 5620}, {"loss": 0.6024, "grad_norm": 0.7793202996253967, "learning_rate": 0.0002, "epoch": 4.129079574624129, "step": 5630}, {"loss": 0.5723, "grad_norm": 1.187238335609436, "learning_rate": 0.0002, "epoch": 4.136413641364136, "step": 5640}, {"loss": 0.6385, "grad_norm": 0.8450375199317932, "learning_rate": 0.0002, "epoch": 4.143747708104144, "step": 5650}, {"loss": 0.6866, "grad_norm": 0.9006940126419067, "learning_rate": 0.0002, "epoch": 4.151081774844151, "step": 5660}, {"loss": 0.6179, "grad_norm": 0.9447154998779297, "learning_rate": 0.0002, "epoch": 4.158415841584159, "step": 5670}, {"loss": 0.6476, "grad_norm": 0.798032283782959, "learning_rate": 0.0002, "epoch": 4.165749908324166, "step": 5680}, {"loss": 0.6666, "grad_norm": 0.65578693151474, "learning_rate": 0.0002, "epoch": 4.1730839750641735, "step": 5690}, {"loss": 0.701, "grad_norm": 1.0864700078964233, "learning_rate": 0.0002, "epoch": 4.18041804180418, "step": 5700}, {"loss": 0.6895, "grad_norm": 0.7344121932983398, "learning_rate": 0.0002, "epoch": 4.187752108544188, "step": 5710}, {"loss": 0.6659, "grad_norm": 0.9722456932067871, "learning_rate": 0.0002, "epoch": 4.195086175284195, "step": 5720}, {"loss": 0.6887, "grad_norm": 1.263814926147461, "learning_rate": 0.0002, "epoch": 4.2024202420242025, "step": 5730}, {"loss": 0.608, "grad_norm": 0.9622581005096436, "learning_rate": 0.0002, "epoch": 4.20975430876421, "step": 5740}, {"loss": 0.6221, "grad_norm": 0.8497143387794495, "learning_rate": 0.0002, "epoch": 4.2170883755042174, "step": 5750}, {"loss": 0.6322, "grad_norm": 0.8248446583747864, "learning_rate": 0.0002, "epoch": 4.224422442244224, "step": 5760}, {"loss": 0.6045, "grad_norm": 1.2544798851013184, "learning_rate": 0.0002, "epoch": 4.2317565089842315, "step": 5770}, {"loss": 0.641, "grad_norm": 0.8224676251411438, "learning_rate": 0.0002, "epoch": 4.239090575724239, "step": 5780}, {"loss": 0.6399, "grad_norm": 0.8924877047538757, "learning_rate": 0.0002, "epoch": 4.2464246424642464, "step": 5790}, {"loss": 0.6845, "grad_norm": 0.8545848727226257, "learning_rate": 0.0002, "epoch": 4.253758709204254, "step": 5800}, {"loss": 0.6669, "grad_norm": 0.8081067800521851, "learning_rate": 0.0002, "epoch": 4.261092775944261, "step": 5810}, {"loss": 0.6149, "grad_norm": 0.7111002802848816, "learning_rate": 0.0002, "epoch": 4.268426842684269, "step": 5820}, {"loss": 0.6343, "grad_norm": 0.8696979880332947, "learning_rate": 0.0002, "epoch": 4.2757609094242754, "step": 5830}, {"loss": 0.6384, "grad_norm": 0.821401834487915, "learning_rate": 0.0002, "epoch": 4.283094976164283, "step": 5840}, {"loss": 0.6912, "grad_norm": 0.888908326625824, "learning_rate": 0.0002, "epoch": 4.29042904290429, "step": 5850}, {"loss": 0.6061, "grad_norm": 1.9380123615264893, "learning_rate": 0.0002, "epoch": 4.297763109644298, "step": 5860}, {"loss": 0.6766, "grad_norm": 1.121774435043335, "learning_rate": 0.0002, "epoch": 4.305097176384305, "step": 5870}, {"loss": 0.7205, "grad_norm": 0.9238282442092896, "learning_rate": 0.0002, "epoch": 4.312431243124313, "step": 5880}, {"loss": 0.6351, "grad_norm": 0.7321620583534241, "learning_rate": 0.0002, "epoch": 4.319765309864319, "step": 5890}, {"loss": 0.6404, "grad_norm": 0.8739548325538635, "learning_rate": 0.0002, "epoch": 4.327099376604327, "step": 5900}, {"loss": 0.5892, "grad_norm": 0.9686012268066406, "learning_rate": 0.0002, "epoch": 4.334433443344334, "step": 5910}, {"loss": 0.641, "grad_norm": 0.9033839106559753, "learning_rate": 0.0002, "epoch": 4.341767510084342, "step": 5920}, {"loss": 0.6456, "grad_norm": 0.8131115436553955, "learning_rate": 0.0002, "epoch": 4.349101576824349, "step": 5930}, {"loss": 0.5826, "grad_norm": 0.8942412734031677, "learning_rate": 0.0002, "epoch": 4.356435643564357, "step": 5940}, {"loss": 0.7336, "grad_norm": 0.8439112901687622, "learning_rate": 0.0002, "epoch": 4.363769710304364, "step": 5950}, {"loss": 0.6537, "grad_norm": 0.9176713228225708, "learning_rate": 0.0002, "epoch": 4.371103777044371, "step": 5960}, {"loss": 0.6792, "grad_norm": 0.6799634695053101, "learning_rate": 0.0002, "epoch": 4.378437843784378, "step": 5970}, {"loss": 0.7266, "grad_norm": 1.0435824394226074, "learning_rate": 0.0002, "epoch": 4.385771910524386, "step": 5980}, {"loss": 0.68, "grad_norm": 0.997937798500061, "learning_rate": 0.0002, "epoch": 4.393105977264393, "step": 5990}, {"loss": 0.6604, "grad_norm": 1.0308842658996582, "learning_rate": 0.0002, "epoch": 4.400440044004401, "step": 6000}, {"loss": 0.6402, "grad_norm": 1.3683775663375854, "learning_rate": 0.0002, "epoch": 4.407774110744408, "step": 6010}, {"loss": 0.7027, "grad_norm": 0.7569534182548523, "learning_rate": 0.0002, "epoch": 4.415108177484415, "step": 6020}, {"loss": 0.5949, "grad_norm": 1.089978575706482, "learning_rate": 0.0002, "epoch": 4.422442244224422, "step": 6030}, {"loss": 0.6353, "grad_norm": 0.7522459626197815, "learning_rate": 0.0002, "epoch": 4.42977631096443, "step": 6040}, {"loss": 0.5852, "grad_norm": 0.6709823608398438, "learning_rate": 0.0002, "epoch": 4.437110377704437, "step": 6050}, {"loss": 0.6718, "grad_norm": 0.6992089748382568, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 6060}, {"loss": 0.6933, "grad_norm": 1.0182931423187256, "learning_rate": 0.0002, "epoch": 4.451778511184452, "step": 6070}, {"loss": 0.6255, "grad_norm": 1.0685160160064697, "learning_rate": 0.0002, "epoch": 4.459112577924459, "step": 6080}, {"loss": 0.6086, "grad_norm": 0.8295124769210815, "learning_rate": 0.0002, "epoch": 4.466446644664466, "step": 6090}, {"loss": 0.6359, "grad_norm": 1.1862998008728027, "learning_rate": 0.0002, "epoch": 4.473780711404474, "step": 6100}, {"loss": 0.638, "grad_norm": 0.7400273084640503, "learning_rate": 0.0002, "epoch": 4.481114778144481, "step": 6110}, {"loss": 0.6854, "grad_norm": 0.7098417282104492, "learning_rate": 0.0002, "epoch": 4.488448844884489, "step": 6120}, {"loss": 0.6976, "grad_norm": 0.9745053648948669, "learning_rate": 0.0002, "epoch": 4.495782911624496, "step": 6130}, {"loss": 0.605, "grad_norm": 0.8638797998428345, "learning_rate": 0.0002, "epoch": 4.503116978364503, "step": 6140}, {"loss": 0.6491, "grad_norm": 0.8291046619415283, "learning_rate": 0.0002, "epoch": 4.51045104510451, "step": 6150}, {"loss": 0.6457, "grad_norm": 1.0301737785339355, "learning_rate": 0.0002, "epoch": 4.517785111844518, "step": 6160}, {"loss": 0.6742, "grad_norm": 1.1996512413024902, "learning_rate": 0.0002, "epoch": 4.525119178584525, "step": 6170}, {"loss": 0.6484, "grad_norm": 1.151038408279419, "learning_rate": 0.0002, "epoch": 4.5324532453245325, "step": 6180}, {"loss": 0.668, "grad_norm": 0.8385201096534729, "learning_rate": 0.0002, "epoch": 4.53978731206454, "step": 6190}, {"loss": 0.6381, "grad_norm": 0.8969188332557678, "learning_rate": 0.0002, "epoch": 4.5471213788045475, "step": 6200}, {"loss": 0.7141, "grad_norm": 1.60659658908844, "learning_rate": 0.0002, "epoch": 4.554455445544555, "step": 6210}, {"loss": 0.6388, "grad_norm": 0.9356731176376343, "learning_rate": 0.0002, "epoch": 4.5617895122845615, "step": 6220}, {"loss": 0.7393, "grad_norm": 0.95856773853302, "learning_rate": 0.0002, "epoch": 4.569123579024569, "step": 6230}, {"loss": 0.6554, "grad_norm": 1.1162524223327637, "learning_rate": 0.0002, "epoch": 4.5764576457645765, "step": 6240}, {"loss": 0.6012, "grad_norm": 0.8809238076210022, "learning_rate": 0.0002, "epoch": 4.583791712504584, "step": 6250}, {"loss": 0.648, "grad_norm": 0.890738844871521, "learning_rate": 0.0002, "epoch": 4.591125779244591, "step": 6260}, {"loss": 0.6663, "grad_norm": 0.918684720993042, "learning_rate": 0.0002, "epoch": 4.598459845984598, "step": 6270}, {"loss": 0.5992, "grad_norm": 0.8156296610832214, "learning_rate": 0.0002, "epoch": 4.6057939127246055, "step": 6280}, {"loss": 0.723, "grad_norm": 1.046634316444397, "learning_rate": 0.0002, "epoch": 4.613127979464613, "step": 6290}, {"loss": 0.7023, "grad_norm": 0.7725525498390198, "learning_rate": 0.0002, "epoch": 4.62046204620462, "step": 6300}, {"loss": 0.6414, "grad_norm": 0.9992046356201172, "learning_rate": 0.0002, "epoch": 4.627796112944628, "step": 6310}, {"loss": 0.6201, "grad_norm": 0.8480095267295837, "learning_rate": 0.0002, "epoch": 4.635130179684635, "step": 6320}, {"loss": 0.6869, "grad_norm": 0.7061955332756042, "learning_rate": 0.0002, "epoch": 4.642464246424643, "step": 6330}, {"loss": 0.6828, "grad_norm": 1.0354212522506714, "learning_rate": 0.0002, "epoch": 4.649798313164649, "step": 6340}, {"loss": 0.6651, "grad_norm": 1.0081377029418945, "learning_rate": 0.0002, "epoch": 4.657132379904657, "step": 6350}, {"loss": 0.726, "grad_norm": 1.2904249429702759, "learning_rate": 0.0002, "epoch": 4.664466446644664, "step": 6360}, {"loss": 0.7148, "grad_norm": 0.9248910546302795, "learning_rate": 0.0002, "epoch": 4.671800513384672, "step": 6370}, {"loss": 0.6961, "grad_norm": 0.9907804131507874, "learning_rate": 0.0002, "epoch": 4.679134580124679, "step": 6380}, {"loss": 0.6163, "grad_norm": 1.201143741607666, "learning_rate": 0.0002, "epoch": 4.686468646864687, "step": 6390}, {"loss": 0.6762, "grad_norm": 0.8709394335746765, "learning_rate": 0.0002, "epoch": 4.693802713604693, "step": 6400}, {"loss": 0.7217, "grad_norm": 0.7468608021736145, "learning_rate": 0.0002, "epoch": 4.701136780344701, "step": 6410}, {"loss": 0.6548, "grad_norm": 0.8607903718948364, "learning_rate": 0.0002, "epoch": 4.708470847084708, "step": 6420}, {"loss": 0.6449, "grad_norm": 0.9840512871742249, "learning_rate": 0.0002, "epoch": 4.715804913824716, "step": 6430}, {"loss": 0.685, "grad_norm": 0.8328204154968262, "learning_rate": 0.0002, "epoch": 4.723138980564723, "step": 6440}, {"loss": 0.697, "grad_norm": 0.924505352973938, "learning_rate": 0.0002, "epoch": 4.730473047304731, "step": 6450}, {"loss": 0.7422, "grad_norm": 0.8897685408592224, "learning_rate": 0.0002, "epoch": 4.737807114044738, "step": 6460}, {"loss": 0.6842, "grad_norm": 0.9605024456977844, "learning_rate": 0.0002, "epoch": 4.745141180784745, "step": 6470}, {"loss": 0.6488, "grad_norm": 0.8150759935379028, "learning_rate": 0.0002, "epoch": 4.752475247524752, "step": 6480}, {"loss": 0.6606, "grad_norm": 0.8128412961959839, "learning_rate": 0.0002, "epoch": 4.75980931426476, "step": 6490}, {"loss": 0.6729, "grad_norm": 0.7381404638290405, "learning_rate": 0.0002, "epoch": 4.767143381004767, "step": 6500}, {"loss": 0.6713, "grad_norm": 1.0565853118896484, "learning_rate": 0.0002, "epoch": 4.774477447744775, "step": 6510}, {"loss": 0.6496, "grad_norm": 0.9298134446144104, "learning_rate": 0.0002, "epoch": 4.781811514484782, "step": 6520}, {"loss": 0.7279, "grad_norm": 1.0145525932312012, "learning_rate": 0.0002, "epoch": 4.789145581224789, "step": 6530}, {"loss": 0.5986, "grad_norm": 0.92259681224823, "learning_rate": 0.0002, "epoch": 4.796479647964796, "step": 6540}, {"loss": 0.63, "grad_norm": 0.7881024479866028, "learning_rate": 0.0002, "epoch": 4.803813714704804, "step": 6550}, {"loss": 0.7134, "grad_norm": 1.4935206174850464, "learning_rate": 0.0002, "epoch": 4.811147781444811, "step": 6560}, {"loss": 0.6695, "grad_norm": 0.8612369298934937, "learning_rate": 0.0002, "epoch": 4.818481848184819, "step": 6570}, {"loss": 0.779, "grad_norm": 1.0118653774261475, "learning_rate": 0.0002, "epoch": 4.825815914924826, "step": 6580}, {"loss": 0.6991, "grad_norm": 1.1303809881210327, "learning_rate": 0.0002, "epoch": 4.833149981664834, "step": 6590}, {"loss": 0.7887, "grad_norm": 0.9112492203712463, "learning_rate": 0.0002, "epoch": 4.84048404840484, "step": 6600}, {"loss": 0.7699, "grad_norm": 0.864762544631958, "learning_rate": 0.0002, "epoch": 4.847818115144848, "step": 6610}, {"loss": 0.7347, "grad_norm": 0.9090572595596313, "learning_rate": 0.0002, "epoch": 4.855152181884855, "step": 6620}, {"loss": 0.6608, "grad_norm": 1.014953374862671, "learning_rate": 0.0002, "epoch": 4.862486248624863, "step": 6630}, {"loss": 0.6429, "grad_norm": 1.0702149868011475, "learning_rate": 0.0002, "epoch": 4.86982031536487, "step": 6640}, {"loss": 0.6943, "grad_norm": 1.002135157585144, "learning_rate": 0.0002, "epoch": 4.8771543821048775, "step": 6650}, {"loss": 0.7225, "grad_norm": 0.862545907497406, "learning_rate": 0.0002, "epoch": 4.884488448844884, "step": 6660}, {"loss": 0.6206, "grad_norm": 0.7302131056785583, "learning_rate": 0.0002, "epoch": 4.891822515584892, "step": 6670}, {"loss": 0.7175, "grad_norm": 0.8380730152130127, "learning_rate": 0.0002, "epoch": 4.899156582324899, "step": 6680}, {"loss": 0.645, "grad_norm": 0.7956018447875977, "learning_rate": 0.0002, "epoch": 4.9064906490649065, "step": 6690}, {"loss": 0.6431, "grad_norm": 0.6717583537101746, "learning_rate": 0.0002, "epoch": 4.913824715804914, "step": 6700}, {"loss": 0.6942, "grad_norm": 1.09099280834198, "learning_rate": 0.0002, "epoch": 4.9211587825449215, "step": 6710}, {"loss": 0.7533, "grad_norm": 0.8589889407157898, "learning_rate": 0.0002, "epoch": 4.928492849284929, "step": 6720}, {"loss": 0.66, "grad_norm": 1.0046314001083374, "learning_rate": 0.0002, "epoch": 4.9358269160249355, "step": 6730}, {"loss": 0.6864, "grad_norm": 0.8559659123420715, "learning_rate": 0.0002, "epoch": 4.943160982764943, "step": 6740}, {"loss": 0.6847, "grad_norm": 0.8588525652885437, "learning_rate": 0.0002, "epoch": 4.9504950495049505, "step": 6750}, {"loss": 0.6428, "grad_norm": 0.9192708134651184, "learning_rate": 0.0002, "epoch": 4.957829116244958, "step": 6760}, {"loss": 0.6873, "grad_norm": 1.051398754119873, "learning_rate": 0.0002, "epoch": 4.965163182984965, "step": 6770}, {"loss": 0.7249, "grad_norm": 0.9111362099647522, "learning_rate": 0.0002, "epoch": 4.972497249724973, "step": 6780}, {"loss": 0.7613, "grad_norm": 0.7305638194084167, "learning_rate": 0.0002, "epoch": 4.9798313164649795, "step": 6790}, {"loss": 0.6747, "grad_norm": 1.118837594985962, "learning_rate": 0.0002, "epoch": 4.987165383204987, "step": 6800}, {"loss": 0.6412, "grad_norm": 0.9075239300727844, "learning_rate": 0.0002, "epoch": 4.994499449944994, "step": 6810}, {"eval_loss": 1.2361247539520264, "eval_runtime": 32.7325, "eval_samples_per_second": 13.167, "eval_steps_per_second": 1.65, "epoch": 4.999633296662999, "step": 6817}, {"loss": 0.7091, "grad_norm": 1.0541315078735352, "learning_rate": 0.0002, "epoch": 5.001833516685002, "step": 6820}, {"loss": 0.4882, "grad_norm": 0.9750140905380249, "learning_rate": 0.0002, "epoch": 5.009167583425009, "step": 6830}, {"loss": 0.6022, "grad_norm": 0.931838870048523, "learning_rate": 0.0002, "epoch": 5.016501650165017, "step": 6840}, {"loss": 0.5194, "grad_norm": 1.110278844833374, "learning_rate": 0.0002, "epoch": 5.023835716905023, "step": 6850}, {"loss": 0.4676, "grad_norm": 1.0670180320739746, "learning_rate": 0.0002, "epoch": 5.031169783645031, "step": 6860}, {"loss": 0.4374, "grad_norm": 0.8762092590332031, "learning_rate": 0.0002, "epoch": 5.038503850385038, "step": 6870}, {"loss": 0.505, "grad_norm": 1.1169432401657104, "learning_rate": 0.0002, "epoch": 5.045837917125046, "step": 6880}, {"loss": 0.5114, "grad_norm": 1.005491018295288, "learning_rate": 0.0002, "epoch": 5.053171983865053, "step": 6890}, {"loss": 0.5221, "grad_norm": 1.1751841306686401, "learning_rate": 0.0002, "epoch": 5.060506050605061, "step": 6900}, {"loss": 0.451, "grad_norm": 0.8501367568969727, "learning_rate": 0.0002, "epoch": 5.067840117345068, "step": 6910}, {"loss": 0.5292, "grad_norm": 0.9795131683349609, "learning_rate": 0.0002, "epoch": 5.075174184085075, "step": 6920}, {"loss": 0.5234, "grad_norm": 0.8929879665374756, "learning_rate": 0.0002, "epoch": 5.082508250825082, "step": 6930}, {"loss": 0.5378, "grad_norm": 1.0156651735305786, "learning_rate": 0.0002, "epoch": 5.08984231756509, "step": 6940}, {"loss": 0.5241, "grad_norm": 1.0974335670471191, "learning_rate": 0.0002, "epoch": 5.097176384305097, "step": 6950}, {"loss": 0.5705, "grad_norm": 1.7015666961669922, "learning_rate": 0.0002, "epoch": 5.104510451045105, "step": 6960}, {"loss": 0.523, "grad_norm": 1.0343226194381714, "learning_rate": 0.0002, "epoch": 5.111844517785112, "step": 6970}, {"loss": 0.4616, "grad_norm": 1.3072983026504517, "learning_rate": 0.0002, "epoch": 5.119178584525119, "step": 6980}, {"loss": 0.4813, "grad_norm": 1.038986086845398, "learning_rate": 0.0002, "epoch": 5.126512651265126, "step": 6990}, {"loss": 0.4616, "grad_norm": 0.8638386130332947, "learning_rate": 0.0002, "epoch": 5.133846718005134, "step": 7000}, {"loss": 0.5294, "grad_norm": 0.8326523900032043, "learning_rate": 0.0002, "epoch": 5.141180784745141, "step": 7010}, {"loss": 0.5021, "grad_norm": 1.0976895093917847, "learning_rate": 0.0002, "epoch": 5.148514851485149, "step": 7020}, {"loss": 0.4677, "grad_norm": 1.0077873468399048, "learning_rate": 0.0002, "epoch": 5.155848918225156, "step": 7030}, {"loss": 0.5262, "grad_norm": 1.0662257671356201, "learning_rate": 0.0002, "epoch": 5.163182984965164, "step": 7040}, {"loss": 0.5484, "grad_norm": 1.206271767616272, "learning_rate": 0.0002, "epoch": 5.17051705170517, "step": 7050}, {"loss": 0.4817, "grad_norm": 1.1990262269973755, "learning_rate": 0.0002, "epoch": 5.177851118445178, "step": 7060}, {"loss": 0.6048, "grad_norm": 1.0207163095474243, "learning_rate": 0.0002, "epoch": 5.185185185185185, "step": 7070}, {"loss": 0.4816, "grad_norm": 1.2783987522125244, "learning_rate": 0.0002, "epoch": 5.192519251925193, "step": 7080}, {"loss": 0.5322, "grad_norm": 1.1592512130737305, "learning_rate": 0.0002, "epoch": 5.1998533186652, "step": 7090}, {"loss": 0.5472, "grad_norm": 1.1053160429000854, "learning_rate": 0.0002, "epoch": 5.2071873854052075, "step": 7100}, {"loss": 0.4986, "grad_norm": 1.1925510168075562, "learning_rate": 0.0002, "epoch": 5.214521452145214, "step": 7110}, {"loss": 0.5065, "grad_norm": 1.0714877843856812, "learning_rate": 0.0002, "epoch": 5.221855518885222, "step": 7120}, {"loss": 0.5209, "grad_norm": 0.9451011419296265, "learning_rate": 0.0002, "epoch": 5.229189585625229, "step": 7130}, {"loss": 0.5298, "grad_norm": 1.03838050365448, "learning_rate": 0.0002, "epoch": 5.2365236523652365, "step": 7140}, {"loss": 0.4848, "grad_norm": 0.9204146265983582, "learning_rate": 0.0002, "epoch": 5.243857719105244, "step": 7150}, {"loss": 0.5164, "grad_norm": 1.0142229795455933, "learning_rate": 0.0002, "epoch": 5.2511917858452515, "step": 7160}, {"loss": 0.5092, "grad_norm": 1.4432005882263184, "learning_rate": 0.0002, "epoch": 5.258525852585258, "step": 7170}, {"loss": 0.5133, "grad_norm": 1.1239633560180664, "learning_rate": 0.0002, "epoch": 5.2658599193252655, "step": 7180}, {"loss": 0.4969, "grad_norm": 0.7012821435928345, "learning_rate": 0.0002, "epoch": 5.273193986065273, "step": 7190}, {"loss": 0.5466, "grad_norm": 1.3499128818511963, "learning_rate": 0.0002, "epoch": 5.2805280528052805, "step": 7200}, {"loss": 0.5282, "grad_norm": 0.9498730897903442, "learning_rate": 0.0002, "epoch": 5.287862119545288, "step": 7210}, {"loss": 0.5051, "grad_norm": 0.9552369117736816, "learning_rate": 0.0002, "epoch": 5.295196186285295, "step": 7220}, {"loss": 0.5329, "grad_norm": 0.7610348463058472, "learning_rate": 0.0002, "epoch": 5.302530253025303, "step": 7230}, {"loss": 0.468, "grad_norm": 1.0314512252807617, "learning_rate": 0.0002, "epoch": 5.3098643197653095, "step": 7240}, {"loss": 0.5367, "grad_norm": 1.0534334182739258, "learning_rate": 0.0002, "epoch": 5.317198386505317, "step": 7250}, {"loss": 0.5491, "grad_norm": 1.2553406953811646, "learning_rate": 0.0002, "epoch": 5.324532453245324, "step": 7260}, {"loss": 0.5218, "grad_norm": 0.7061691880226135, "learning_rate": 0.0002, "epoch": 5.331866519985332, "step": 7270}, {"loss": 0.5625, "grad_norm": 0.9652578830718994, "learning_rate": 0.0002, "epoch": 5.339200586725339, "step": 7280}, {"loss": 0.5608, "grad_norm": 1.114788293838501, "learning_rate": 0.0002, "epoch": 5.346534653465347, "step": 7290}, {"loss": 0.578, "grad_norm": 1.0940049886703491, "learning_rate": 0.0002, "epoch": 5.353868720205353, "step": 7300}, {"loss": 0.5256, "grad_norm": 1.0151008367538452, "learning_rate": 0.0002, "epoch": 5.361202786945361, "step": 7310}, {"loss": 0.5377, "grad_norm": 1.0369552373886108, "learning_rate": 0.0002, "epoch": 5.368536853685368, "step": 7320}, {"loss": 0.5028, "grad_norm": 0.8489866256713867, "learning_rate": 0.0002, "epoch": 5.375870920425376, "step": 7330}, {"loss": 0.5937, "grad_norm": 1.1031713485717773, "learning_rate": 0.0002, "epoch": 5.383204987165383, "step": 7340}, {"loss": 0.5355, "grad_norm": 0.9094716310501099, "learning_rate": 0.0002, "epoch": 5.390539053905391, "step": 7350}, {"loss": 0.5406, "grad_norm": 0.9530431032180786, "learning_rate": 0.0002, "epoch": 5.397873120645398, "step": 7360}, {"loss": 0.529, "grad_norm": 0.9633604884147644, "learning_rate": 0.0002, "epoch": 5.405207187385405, "step": 7370}, {"loss": 0.5315, "grad_norm": 0.9541662335395813, "learning_rate": 0.0002, "epoch": 5.412541254125412, "step": 7380}, {"loss": 0.6774, "grad_norm": 1.0459771156311035, "learning_rate": 0.0002, "epoch": 5.41987532086542, "step": 7390}, {"loss": 0.5737, "grad_norm": 1.027388334274292, "learning_rate": 0.0002, "epoch": 5.427209387605427, "step": 7400}, {"loss": 0.556, "grad_norm": 0.7267653346061707, "learning_rate": 0.0002, "epoch": 5.434543454345435, "step": 7410}, {"loss": 0.4581, "grad_norm": 1.020142674446106, "learning_rate": 0.0002, "epoch": 5.441877521085442, "step": 7420}, {"loss": 0.4853, "grad_norm": 1.044754147529602, "learning_rate": 0.0002, "epoch": 5.449211587825449, "step": 7430}, {"loss": 0.5666, "grad_norm": 1.5476195812225342, "learning_rate": 0.0002, "epoch": 5.456545654565456, "step": 7440}, {"loss": 0.5302, "grad_norm": 0.9879506826400757, "learning_rate": 0.0002, "epoch": 5.463879721305464, "step": 7450}, {"loss": 0.591, "grad_norm": 1.2562980651855469, "learning_rate": 0.0002, "epoch": 5.471213788045471, "step": 7460}, {"loss": 0.5188, "grad_norm": 1.3051384687423706, "learning_rate": 0.0002, "epoch": 5.478547854785479, "step": 7470}, {"loss": 0.5658, "grad_norm": 1.0511597394943237, "learning_rate": 0.0002, "epoch": 5.485881921525486, "step": 7480}, {"loss": 0.6327, "grad_norm": 1.0380817651748657, "learning_rate": 0.0002, "epoch": 5.493215988265494, "step": 7490}, {"loss": 0.5356, "grad_norm": 1.170274257659912, "learning_rate": 0.0002, "epoch": 5.5005500550055, "step": 7500}, {"loss": 0.5405, "grad_norm": 1.3356517553329468, "learning_rate": 0.0002, "epoch": 5.507884121745508, "step": 7510}, {"loss": 0.5305, "grad_norm": 1.0727124214172363, "learning_rate": 0.0002, "epoch": 5.515218188485515, "step": 7520}, {"loss": 0.5543, "grad_norm": 1.0110199451446533, "learning_rate": 0.0002, "epoch": 5.522552255225523, "step": 7530}, {"loss": 0.5962, "grad_norm": 1.3086743354797363, "learning_rate": 0.0002, "epoch": 5.52988632196553, "step": 7540}, {"loss": 0.5512, "grad_norm": 1.1904916763305664, "learning_rate": 0.0002, "epoch": 5.537220388705538, "step": 7550}, {"loss": 0.5915, "grad_norm": 0.9466280937194824, "learning_rate": 0.0002, "epoch": 5.544554455445544, "step": 7560}, {"loss": 0.5573, "grad_norm": 1.1237901449203491, "learning_rate": 0.0002, "epoch": 5.551888522185552, "step": 7570}, {"loss": 0.5383, "grad_norm": 0.9590660333633423, "learning_rate": 0.0002, "epoch": 5.559222588925559, "step": 7580}, {"loss": 0.5594, "grad_norm": 1.0890778303146362, "learning_rate": 0.0002, "epoch": 5.566556655665567, "step": 7590}, {"loss": 0.5698, "grad_norm": 0.7206931114196777, "learning_rate": 0.0002, "epoch": 5.573890722405574, "step": 7600}, {"loss": 0.5511, "grad_norm": 1.2884514331817627, "learning_rate": 0.0002, "epoch": 5.5812247891455815, "step": 7610}, {"loss": 0.5279, "grad_norm": 0.7798039317131042, "learning_rate": 0.0002, "epoch": 5.588558855885589, "step": 7620}, {"loss": 0.4847, "grad_norm": 1.166046142578125, "learning_rate": 0.0002, "epoch": 5.595892922625596, "step": 7630}, {"loss": 0.5821, "grad_norm": 1.0150201320648193, "learning_rate": 0.0002, "epoch": 5.603226989365603, "step": 7640}, {"loss": 0.5296, "grad_norm": 1.0449682474136353, "learning_rate": 0.0002, "epoch": 5.6105610561056105, "step": 7650}, {"loss": 0.5431, "grad_norm": 0.9310530424118042, "learning_rate": 0.0002, "epoch": 5.617895122845618, "step": 7660}, {"loss": 0.5234, "grad_norm": 0.9117933511734009, "learning_rate": 0.0002, "epoch": 5.6252291895856255, "step": 7670}, {"loss": 0.5807, "grad_norm": 1.1475164890289307, "learning_rate": 0.0002, "epoch": 5.632563256325633, "step": 7680}, {"loss": 0.5816, "grad_norm": 1.066809058189392, "learning_rate": 0.0002, "epoch": 5.6398973230656395, "step": 7690}, {"loss": 0.551, "grad_norm": 1.2834991216659546, "learning_rate": 0.0002, "epoch": 5.647231389805647, "step": 7700}, {"loss": 0.5914, "grad_norm": 1.2245112657546997, "learning_rate": 0.0002, "epoch": 5.6545654565456545, "step": 7710}, {"loss": 0.5552, "grad_norm": 1.1424106359481812, "learning_rate": 0.0002, "epoch": 5.661899523285662, "step": 7720}, {"loss": 0.559, "grad_norm": 1.0673892498016357, "learning_rate": 0.0002, "epoch": 5.669233590025669, "step": 7730}, {"loss": 0.544, "grad_norm": 1.4312121868133545, "learning_rate": 0.0002, "epoch": 5.676567656765677, "step": 7740}, {"loss": 0.5576, "grad_norm": 0.9976982474327087, "learning_rate": 0.0002, "epoch": 5.683901723505684, "step": 7750}, {"loss": 0.4855, "grad_norm": 0.9464678168296814, "learning_rate": 0.0002, "epoch": 5.691235790245691, "step": 7760}, {"loss": 0.5363, "grad_norm": 1.010995626449585, "learning_rate": 0.0002, "epoch": 5.698569856985698, "step": 7770}, {"loss": 0.5873, "grad_norm": 1.3787750005722046, "learning_rate": 0.0002, "epoch": 5.705903923725706, "step": 7780}, {"loss": 0.6234, "grad_norm": 1.020922303199768, "learning_rate": 0.0002, "epoch": 5.713237990465713, "step": 7790}, {"loss": 0.5337, "grad_norm": 0.9748636484146118, "learning_rate": 0.0002, "epoch": 5.720572057205721, "step": 7800}, {"loss": 0.5507, "grad_norm": 1.3077744245529175, "learning_rate": 0.0002, "epoch": 5.727906123945728, "step": 7810}, {"loss": 0.558, "grad_norm": 1.4770057201385498, "learning_rate": 0.0002, "epoch": 5.735240190685735, "step": 7820}, {"loss": 0.5571, "grad_norm": 1.6349090337753296, "learning_rate": 0.0002, "epoch": 5.742574257425742, "step": 7830}, {"loss": 0.5056, "grad_norm": 0.9818630814552307, "learning_rate": 0.0002, "epoch": 5.74990832416575, "step": 7840}, {"loss": 0.5495, "grad_norm": 0.9659715890884399, "learning_rate": 0.0002, "epoch": 5.757242390905757, "step": 7850}, {"loss": 0.5628, "grad_norm": 0.9269950985908508, "learning_rate": 0.0002, "epoch": 5.764576457645765, "step": 7860}, {"loss": 0.5594, "grad_norm": 1.0099073648452759, "learning_rate": 0.0002, "epoch": 5.771910524385772, "step": 7870}, {"loss": 0.5912, "grad_norm": 0.9123615026473999, "learning_rate": 0.0002, "epoch": 5.77924459112578, "step": 7880}, {"loss": 0.6054, "grad_norm": 1.1542246341705322, "learning_rate": 0.0002, "epoch": 5.786578657865786, "step": 7890}, {"loss": 0.5829, "grad_norm": 1.0792022943496704, "learning_rate": 0.0002, "epoch": 5.793912724605794, "step": 7900}, {"loss": 0.504, "grad_norm": 0.95615553855896, "learning_rate": 0.0002, "epoch": 5.801246791345801, "step": 7910}, {"loss": 0.5918, "grad_norm": 1.2471332550048828, "learning_rate": 0.0002, "epoch": 5.808580858085809, "step": 7920}, {"loss": 0.5719, "grad_norm": 1.0189851522445679, "learning_rate": 0.0002, "epoch": 5.815914924825816, "step": 7930}, {"loss": 0.5958, "grad_norm": 1.3309742212295532, "learning_rate": 0.0002, "epoch": 5.823248991565823, "step": 7940}, {"loss": 0.6255, "grad_norm": 1.2930549383163452, "learning_rate": 0.0002, "epoch": 5.83058305830583, "step": 7950}, {"loss": 0.5301, "grad_norm": 0.8216308951377869, "learning_rate": 0.0002, "epoch": 5.837917125045838, "step": 7960}, {"loss": 0.5397, "grad_norm": 1.1205775737762451, "learning_rate": 0.0002, "epoch": 5.845251191785845, "step": 7970}, {"loss": 0.5903, "grad_norm": 0.851298451423645, "learning_rate": 0.0002, "epoch": 5.852585258525853, "step": 7980}, {"loss": 0.5981, "grad_norm": 0.8797095417976379, "learning_rate": 0.0002, "epoch": 5.85991932526586, "step": 7990}, {"loss": 0.6106, "grad_norm": 1.5784614086151123, "learning_rate": 0.0002, "epoch": 5.867253392005868, "step": 8000}, {"loss": 0.5956, "grad_norm": 1.1531187295913696, "learning_rate": 0.0002, "epoch": 5.874587458745875, "step": 8010}, {"loss": 0.6289, "grad_norm": 1.2469146251678467, "learning_rate": 0.0002, "epoch": 5.881921525485882, "step": 8020}, {"loss": 0.5827, "grad_norm": 1.0784350633621216, "learning_rate": 0.0002, "epoch": 5.889255592225889, "step": 8030}, {"loss": 0.6339, "grad_norm": 1.1311599016189575, "learning_rate": 0.0002, "epoch": 5.896589658965897, "step": 8040}, {"loss": 0.5815, "grad_norm": 0.9654512405395508, "learning_rate": 0.0002, "epoch": 5.903923725705904, "step": 8050}, {"loss": 0.6198, "grad_norm": 1.3288270235061646, "learning_rate": 0.0002, "epoch": 5.9112577924459115, "step": 8060}, {"loss": 0.6515, "grad_norm": 1.12800931930542, "learning_rate": 0.0002, "epoch": 5.918591859185918, "step": 8070}, {"loss": 0.5684, "grad_norm": 0.9449917674064636, "learning_rate": 0.0002, "epoch": 5.925925925925926, "step": 8080}, {"loss": 0.6063, "grad_norm": 1.1532357931137085, "learning_rate": 0.0002, "epoch": 5.933259992665933, "step": 8090}, {"loss": 0.5318, "grad_norm": 1.2211151123046875, "learning_rate": 0.0002, "epoch": 5.9405940594059405, "step": 8100}, {"loss": 0.6512, "grad_norm": 1.3459105491638184, "learning_rate": 0.0002, "epoch": 5.947928126145948, "step": 8110}, {"loss": 0.5952, "grad_norm": 1.251999855041504, "learning_rate": 0.0002, "epoch": 5.9552621928859555, "step": 8120}, {"loss": 0.6203, "grad_norm": 1.5682506561279297, "learning_rate": 0.0002, "epoch": 5.962596259625963, "step": 8130}, {"loss": 0.6253, "grad_norm": 0.926075279712677, "learning_rate": 0.0002, "epoch": 5.9699303263659695, "step": 8140}, {"loss": 0.5545, "grad_norm": 0.9622511863708496, "learning_rate": 0.0002, "epoch": 5.977264393105977, "step": 8150}, {"loss": 0.5518, "grad_norm": 0.9633373618125916, "learning_rate": 0.0002, "epoch": 5.9845984598459845, "step": 8160}, {"loss": 0.5831, "grad_norm": 0.8960476517677307, "learning_rate": 0.0002, "epoch": 5.991932526585992, "step": 8170}, {"loss": 0.5442, "grad_norm": 0.9372805953025818, "learning_rate": 0.0002, "epoch": 5.999266593325999, "step": 8180}, {"eval_loss": 1.3233846426010132, "eval_runtime": 32.7419, "eval_samples_per_second": 13.164, "eval_steps_per_second": 1.649, "epoch": 6.0, "step": 8181}, {"loss": 0.4644, "grad_norm": 1.1900787353515625, "learning_rate": 0.0002, "epoch": 6.006600660066007, "step": 8190}, {"loss": 0.4509, "grad_norm": 1.1448326110839844, "learning_rate": 0.0002, "epoch": 6.013934726806014, "step": 8200}, {"loss": 0.3667, "grad_norm": 1.1848368644714355, "learning_rate": 0.0002, "epoch": 6.021268793546021, "step": 8210}, {"loss": 0.4315, "grad_norm": 1.2315572500228882, "learning_rate": 0.0002, "epoch": 6.028602860286028, "step": 8220}, {"loss": 0.3541, "grad_norm": 1.2214244604110718, "learning_rate": 0.0002, "epoch": 6.035936927026036, "step": 8230}, {"loss": 0.4025, "grad_norm": 0.9455513954162598, "learning_rate": 0.0002, "epoch": 6.043270993766043, "step": 8240}, {"loss": 0.4448, "grad_norm": 0.9574248790740967, "learning_rate": 0.0002, "epoch": 6.050605060506051, "step": 8250}, {"loss": 0.4271, "grad_norm": 1.1022400856018066, "learning_rate": 0.0002, "epoch": 6.057939127246058, "step": 8260}, {"loss": 0.3603, "grad_norm": 0.9555122256278992, "learning_rate": 0.0002, "epoch": 6.065273193986065, "step": 8270}, {"loss": 0.4324, "grad_norm": 1.1956106424331665, "learning_rate": 0.0002, "epoch": 6.072607260726072, "step": 8280}, {"loss": 0.3924, "grad_norm": 1.3110876083374023, "learning_rate": 0.0002, "epoch": 6.07994132746608, "step": 8290}, {"loss": 0.3664, "grad_norm": 1.1293374300003052, "learning_rate": 0.0002, "epoch": 6.087275394206087, "step": 8300}, {"loss": 0.385, "grad_norm": 0.9176164269447327, "learning_rate": 0.0002, "epoch": 6.094609460946095, "step": 8310}, {"loss": 0.4142, "grad_norm": 0.9751231670379639, "learning_rate": 0.0002, "epoch": 6.101943527686102, "step": 8320}, {"loss": 0.4356, "grad_norm": 1.0536044836044312, "learning_rate": 0.0002, "epoch": 6.109277594426109, "step": 8330}, {"loss": 0.409, "grad_norm": 1.289342999458313, "learning_rate": 0.0002, "epoch": 6.116611661166116, "step": 8340}, {"loss": 0.4121, "grad_norm": 1.1773661375045776, "learning_rate": 0.0002, "epoch": 6.123945727906124, "step": 8350}, {"loss": 0.4499, "grad_norm": 1.2450661659240723, "learning_rate": 0.0002, "epoch": 6.131279794646131, "step": 8360}, {"loss": 0.4467, "grad_norm": 1.3965914249420166, "learning_rate": 0.0002, "epoch": 6.138613861386139, "step": 8370}, {"loss": 0.4024, "grad_norm": 1.3530808687210083, "learning_rate": 0.0002, "epoch": 6.145947928126146, "step": 8380}, {"loss": 0.4658, "grad_norm": 1.296276330947876, "learning_rate": 0.0002, "epoch": 6.153281994866154, "step": 8390}, {"loss": 0.5073, "grad_norm": 0.9759053587913513, "learning_rate": 0.0002, "epoch": 6.16061606160616, "step": 8400}, {"loss": 0.4718, "grad_norm": 1.2110707759857178, "learning_rate": 0.0002, "epoch": 6.167950128346168, "step": 8410}, {"loss": 0.4453, "grad_norm": 1.312226414680481, "learning_rate": 0.0002, "epoch": 6.175284195086175, "step": 8420}, {"loss": 0.4183, "grad_norm": 1.1696736812591553, "learning_rate": 0.0002, "epoch": 6.182618261826183, "step": 8430}, {"loss": 0.4546, "grad_norm": 1.260304570198059, "learning_rate": 0.0002, "epoch": 6.18995232856619, "step": 8440}, {"loss": 0.4137, "grad_norm": 1.472961187362671, "learning_rate": 0.0002, "epoch": 6.197286395306198, "step": 8450}, {"loss": 0.42, "grad_norm": 1.3618475198745728, "learning_rate": 0.0002, "epoch": 6.204620462046204, "step": 8460}, {"loss": 0.415, "grad_norm": 1.2544318437576294, "learning_rate": 0.0002, "epoch": 6.211954528786212, "step": 8470}, {"loss": 0.3907, "grad_norm": 1.205898642539978, "learning_rate": 0.0002, "epoch": 6.219288595526219, "step": 8480}, {"loss": 0.4431, "grad_norm": 0.9984724521636963, "learning_rate": 0.0002, "epoch": 6.226622662266227, "step": 8490}, {"loss": 0.4768, "grad_norm": 1.3184109926223755, "learning_rate": 0.0002, "epoch": 6.233956729006234, "step": 8500}, {"loss": 0.3859, "grad_norm": 1.135520100593567, "learning_rate": 0.0002, "epoch": 6.241290795746242, "step": 8510}, {"loss": 0.4159, "grad_norm": 1.4528400897979736, "learning_rate": 0.0002, "epoch": 6.248624862486249, "step": 8520}, {"loss": 0.4347, "grad_norm": 1.1222716569900513, "learning_rate": 0.0002, "epoch": 6.255958929226256, "step": 8530}, {"loss": 0.4581, "grad_norm": 1.7878046035766602, "learning_rate": 0.0002, "epoch": 6.263292995966263, "step": 8540}, {"loss": 0.4298, "grad_norm": 0.9789481163024902, "learning_rate": 0.0002, "epoch": 6.270627062706271, "step": 8550}, {"loss": 0.4316, "grad_norm": 1.151977300643921, "learning_rate": 0.0002, "epoch": 6.277961129446278, "step": 8560}, {"loss": 0.428, "grad_norm": 1.389968752861023, "learning_rate": 0.0002, "epoch": 6.2852951961862855, "step": 8570}, {"loss": 0.3903, "grad_norm": 0.884211003780365, "learning_rate": 0.0002, "epoch": 6.292629262926293, "step": 8580}, {"loss": 0.4611, "grad_norm": 1.3604296445846558, "learning_rate": 0.0002, "epoch": 6.2999633296663, "step": 8590}, {"loss": 0.4183, "grad_norm": 1.1845694780349731, "learning_rate": 0.0002, "epoch": 6.307297396406307, "step": 8600}, {"loss": 0.472, "grad_norm": 1.3231550455093384, "learning_rate": 0.0002, "epoch": 6.3146314631463145, "step": 8610}, {"loss": 0.3922, "grad_norm": 0.9546721577644348, "learning_rate": 0.0002, "epoch": 6.321965529886322, "step": 8620}, {"loss": 0.4395, "grad_norm": 1.2329787015914917, "learning_rate": 0.0002, "epoch": 6.3292995966263295, "step": 8630}, {"loss": 0.4344, "grad_norm": 1.0240199565887451, "learning_rate": 0.0002, "epoch": 6.336633663366337, "step": 8640}, {"loss": 0.4529, "grad_norm": 1.1866962909698486, "learning_rate": 0.0002, "epoch": 6.343967730106344, "step": 8650}, {"loss": 0.4575, "grad_norm": 1.2819687128067017, "learning_rate": 0.0002, "epoch": 6.351301796846351, "step": 8660}, {"loss": 0.455, "grad_norm": 0.9654944539070129, "learning_rate": 0.0002, "epoch": 6.3586358635863585, "step": 8670}, {"loss": 0.4739, "grad_norm": 0.9443874955177307, "learning_rate": 0.0002, "epoch": 6.365969930326366, "step": 8680}, {"loss": 0.435, "grad_norm": 1.2914115190505981, "learning_rate": 0.0002, "epoch": 6.373303997066373, "step": 8690}, {"loss": 0.4392, "grad_norm": 1.4558709859848022, "learning_rate": 0.0002, "epoch": 6.380638063806381, "step": 8700}, {"loss": 0.4398, "grad_norm": 1.3255952596664429, "learning_rate": 0.0002, "epoch": 6.387972130546388, "step": 8710}, {"loss": 0.4451, "grad_norm": 1.348742961883545, "learning_rate": 0.0002, "epoch": 6.395306197286395, "step": 8720}, {"loss": 0.41, "grad_norm": 1.0096025466918945, "learning_rate": 0.0002, "epoch": 6.402640264026402, "step": 8730}, {"loss": 0.4459, "grad_norm": 1.1720590591430664, "learning_rate": 0.0002, "epoch": 6.40997433076641, "step": 8740}, {"loss": 0.5059, "grad_norm": 1.1803077459335327, "learning_rate": 0.0002, "epoch": 6.417308397506417, "step": 8750}, {"loss": 0.4539, "grad_norm": 1.3649998903274536, "learning_rate": 0.0002, "epoch": 6.424642464246425, "step": 8760}, {"loss": 0.4171, "grad_norm": 1.1503992080688477, "learning_rate": 0.0002, "epoch": 6.431976530986432, "step": 8770}, {"loss": 0.488, "grad_norm": 1.1537176370620728, "learning_rate": 0.0002, "epoch": 6.43931059772644, "step": 8780}, {"loss": 0.4167, "grad_norm": 0.9743003845214844, "learning_rate": 0.0002, "epoch": 6.446644664466446, "step": 8790}, {"loss": 0.4813, "grad_norm": 0.9097744822502136, "learning_rate": 0.0002, "epoch": 6.453978731206454, "step": 8800}, {"loss": 0.4809, "grad_norm": 2.0174002647399902, "learning_rate": 0.0002, "epoch": 6.461312797946461, "step": 8810}, {"loss": 0.4879, "grad_norm": 1.0809309482574463, "learning_rate": 0.0002, "epoch": 6.468646864686469, "step": 8820}, {"loss": 0.4235, "grad_norm": 1.100294828414917, "learning_rate": 0.0002, "epoch": 6.475980931426476, "step": 8830}, {"loss": 0.4251, "grad_norm": 1.3707489967346191, "learning_rate": 0.0002, "epoch": 6.483314998166484, "step": 8840}, {"loss": 0.4533, "grad_norm": 1.1304761171340942, "learning_rate": 0.0002, "epoch": 6.49064906490649, "step": 8850}, {"loss": 0.4596, "grad_norm": 1.2171573638916016, "learning_rate": 0.0002, "epoch": 6.497983131646498, "step": 8860}, {"loss": 0.4694, "grad_norm": 1.0452901124954224, "learning_rate": 0.0002, "epoch": 6.505317198386505, "step": 8870}, {"loss": 0.4855, "grad_norm": 1.197298526763916, "learning_rate": 0.0002, "epoch": 6.512651265126513, "step": 8880}, {"loss": 0.4167, "grad_norm": 0.9179880619049072, "learning_rate": 0.0002, "epoch": 6.51998533186652, "step": 8890}, {"loss": 0.445, "grad_norm": 1.415079951286316, "learning_rate": 0.0002, "epoch": 6.527319398606528, "step": 8900}, {"loss": 0.424, "grad_norm": 1.1032487154006958, "learning_rate": 0.0002, "epoch": 6.534653465346535, "step": 8910}, {"loss": 0.4496, "grad_norm": 1.2295007705688477, "learning_rate": 0.0002, "epoch": 6.541987532086542, "step": 8920}, {"loss": 0.4755, "grad_norm": 1.4223219156265259, "learning_rate": 0.0002, "epoch": 6.549321598826549, "step": 8930}, {"loss": 0.4597, "grad_norm": 1.2785786390304565, "learning_rate": 0.0002, "epoch": 6.556655665566557, "step": 8940}, {"loss": 0.4651, "grad_norm": 1.3514775037765503, "learning_rate": 0.0002, "epoch": 6.563989732306564, "step": 8950}, {"loss": 0.4961, "grad_norm": 1.107937216758728, "learning_rate": 0.0002, "epoch": 6.571323799046572, "step": 8960}, {"loss": 0.4954, "grad_norm": 1.2839902639389038, "learning_rate": 0.0002, "epoch": 6.578657865786578, "step": 8970}, {"loss": 0.4207, "grad_norm": 0.9793244004249573, "learning_rate": 0.0002, "epoch": 6.585991932526586, "step": 8980}, {"loss": 0.4989, "grad_norm": 1.3403126001358032, "learning_rate": 0.0002, "epoch": 6.593325999266593, "step": 8990}, {"loss": 0.465, "grad_norm": 1.2612813711166382, "learning_rate": 0.0002, "epoch": 6.600660066006601, "step": 9000}, {"loss": 0.4589, "grad_norm": 1.4347625970840454, "learning_rate": 0.0002, "epoch": 6.607994132746608, "step": 9010}, {"loss": 0.4864, "grad_norm": 1.225921869277954, "learning_rate": 0.0002, "epoch": 6.6153281994866155, "step": 9020}, {"loss": 0.4364, "grad_norm": 1.033644676208496, "learning_rate": 0.0002, "epoch": 6.622662266226623, "step": 9030}, {"loss": 0.4698, "grad_norm": 1.1791894435882568, "learning_rate": 0.0002, "epoch": 6.6299963329666305, "step": 9040}, {"loss": 0.4908, "grad_norm": 1.0968137979507446, "learning_rate": 0.0002, "epoch": 6.637330399706637, "step": 9050}, {"loss": 0.4346, "grad_norm": 1.5639140605926514, "learning_rate": 0.0002, "epoch": 6.6446644664466445, "step": 9060}, {"loss": 0.4627, "grad_norm": 1.4158905744552612, "learning_rate": 0.0002, "epoch": 6.651998533186652, "step": 9070}, {"loss": 0.4619, "grad_norm": 1.2120254039764404, "learning_rate": 0.0002, "epoch": 6.6593325999266595, "step": 9080}, {"loss": 0.4564, "grad_norm": 1.1866531372070312, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 9090}, {"loss": 0.5175, "grad_norm": 1.2704026699066162, "learning_rate": 0.0002, "epoch": 6.6740007334066735, "step": 9100}, {"loss": 0.4859, "grad_norm": 1.1878353357315063, "learning_rate": 0.0002, "epoch": 6.681334800146681, "step": 9110}, {"loss": 0.4657, "grad_norm": 1.193995475769043, "learning_rate": 0.0002, "epoch": 6.6886688668866885, "step": 9120}, {"loss": 0.4939, "grad_norm": 1.2927545309066772, "learning_rate": 0.0002, "epoch": 6.696002933626696, "step": 9130}, {"loss": 0.4157, "grad_norm": 1.0770703554153442, "learning_rate": 0.0002, "epoch": 6.703337000366703, "step": 9140}, {"loss": 0.4571, "grad_norm": 1.2200851440429688, "learning_rate": 0.0002, "epoch": 6.710671067106711, "step": 9150}, {"loss": 0.4605, "grad_norm": 1.293891191482544, "learning_rate": 0.0002, "epoch": 6.718005133846718, "step": 9160}, {"loss": 0.5328, "grad_norm": 1.9376052618026733, "learning_rate": 0.0002, "epoch": 6.725339200586725, "step": 9170}, {"loss": 0.4861, "grad_norm": 1.0353254079818726, "learning_rate": 0.0002, "epoch": 6.732673267326732, "step": 9180}, {"loss": 0.5668, "grad_norm": 1.1274057626724243, "learning_rate": 0.0002, "epoch": 6.74000733406674, "step": 9190}, {"loss": 0.4486, "grad_norm": 1.3344064950942993, "learning_rate": 0.0002, "epoch": 6.747341400806747, "step": 9200}, {"loss": 0.49, "grad_norm": 1.303621768951416, "learning_rate": 0.0002, "epoch": 6.754675467546755, "step": 9210}, {"loss": 0.5059, "grad_norm": 1.2327780723571777, "learning_rate": 0.0002, "epoch": 6.762009534286762, "step": 9220}, {"loss": 0.486, "grad_norm": 1.3513109683990479, "learning_rate": 0.0002, "epoch": 6.769343601026769, "step": 9230}, {"loss": 0.5254, "grad_norm": 1.4762850999832153, "learning_rate": 0.0002, "epoch": 6.776677667766776, "step": 9240}, {"loss": 0.4181, "grad_norm": 1.0967189073562622, "learning_rate": 0.0002, "epoch": 6.784011734506784, "step": 9250}, {"loss": 0.4862, "grad_norm": 0.933936357498169, "learning_rate": 0.0002, "epoch": 6.791345801246791, "step": 9260}, {"loss": 0.4667, "grad_norm": 1.065553903579712, "learning_rate": 0.0002, "epoch": 6.798679867986799, "step": 9270}, {"loss": 0.5164, "grad_norm": 1.2044163942337036, "learning_rate": 0.0002, "epoch": 6.806013934726806, "step": 9280}, {"loss": 0.4648, "grad_norm": 1.404137134552002, "learning_rate": 0.0002, "epoch": 6.813348001466814, "step": 9290}, {"loss": 0.4442, "grad_norm": 1.4005582332611084, "learning_rate": 0.0002, "epoch": 6.82068206820682, "step": 9300}, {"loss": 0.459, "grad_norm": 1.1771104335784912, "learning_rate": 0.0002, "epoch": 6.828016134946828, "step": 9310}, {"loss": 0.5059, "grad_norm": 1.191933035850525, "learning_rate": 0.0002, "epoch": 6.835350201686835, "step": 9320}, {"loss": 0.4733, "grad_norm": 1.3395432233810425, "learning_rate": 0.0002, "epoch": 6.842684268426843, "step": 9330}, {"loss": 0.4882, "grad_norm": 1.4145503044128418, "learning_rate": 0.0002, "epoch": 6.85001833516685, "step": 9340}, {"loss": 0.4872, "grad_norm": 1.1128839254379272, "learning_rate": 0.0002, "epoch": 6.857352401906858, "step": 9350}, {"loss": 0.4909, "grad_norm": 1.0771174430847168, "learning_rate": 0.0002, "epoch": 6.864686468646864, "step": 9360}, {"loss": 0.4739, "grad_norm": 1.1089814901351929, "learning_rate": 0.0002, "epoch": 6.872020535386872, "step": 9370}, {"loss": 0.4854, "grad_norm": 1.078444004058838, "learning_rate": 0.0002, "epoch": 6.879354602126879, "step": 9380}, {"loss": 0.4904, "grad_norm": 1.3676636219024658, "learning_rate": 0.0002, "epoch": 6.886688668866887, "step": 9390}, {"loss": 0.4854, "grad_norm": 0.8973749876022339, "learning_rate": 0.0002, "epoch": 6.894022735606894, "step": 9400}, {"loss": 0.4274, "grad_norm": 1.141552448272705, "learning_rate": 0.0002, "epoch": 6.901356802346902, "step": 9410}, {"loss": 0.4972, "grad_norm": 0.8345359563827515, "learning_rate": 0.0002, "epoch": 6.908690869086909, "step": 9420}, {"loss": 0.5218, "grad_norm": 1.1602197885513306, "learning_rate": 0.0002, "epoch": 6.916024935826916, "step": 9430}, {"loss": 0.4911, "grad_norm": 1.275466799736023, "learning_rate": 0.0002, "epoch": 6.923359002566923, "step": 9440}, {"loss": 0.4904, "grad_norm": 0.9186071157455444, "learning_rate": 0.0002, "epoch": 6.930693069306931, "step": 9450}, {"loss": 0.4604, "grad_norm": 0.9069198966026306, "learning_rate": 0.0002, "epoch": 6.938027136046938, "step": 9460}, {"loss": 0.4363, "grad_norm": 1.2331899404525757, "learning_rate": 0.0002, "epoch": 6.945361202786946, "step": 9470}, {"loss": 0.4815, "grad_norm": 0.8685150742530823, "learning_rate": 0.0002, "epoch": 6.952695269526953, "step": 9480}, {"loss": 0.4424, "grad_norm": 1.4067939519882202, "learning_rate": 0.0002, "epoch": 6.96002933626696, "step": 9490}, {"loss": 0.5089, "grad_norm": 1.1864029169082642, "learning_rate": 0.0002, "epoch": 6.967363403006967, "step": 9500}, {"loss": 0.4906, "grad_norm": 1.3697725534439087, "learning_rate": 0.0002, "epoch": 6.974697469746975, "step": 9510}, {"loss": 0.4797, "grad_norm": 1.1632893085479736, "learning_rate": 0.0002, "epoch": 6.982031536486982, "step": 9520}, {"loss": 0.4526, "grad_norm": 1.1447268724441528, "learning_rate": 0.0002, "epoch": 6.9893656032269895, "step": 9530}, {"loss": 0.4627, "grad_norm": 1.5017213821411133, "learning_rate": 0.0002, "epoch": 6.996699669966997, "step": 9540}]}
+{"epoch": 7.997066373303997, "step": 10904, "epoch_duration": 1482.7493734359741, "total_accumulated_duration": 11885.982197284698, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7568.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-4/checkpoint-2727", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 1.9722, "grad_norm": 0.47521963715553284, "learning_rate": 0.0002, "epoch": 0.007334066740007334, "step": 10}, {"loss": 1.4821, "grad_norm": 0.5395162105560303, "learning_rate": 0.0002, "epoch": 0.014668133480014669, "step": 20}, {"loss": 1.4202, "grad_norm": 0.4305780231952667, "learning_rate": 0.0002, "epoch": 0.022002200220022004, "step": 30}, {"loss": 1.4271, "grad_norm": 0.6938246488571167, "learning_rate": 0.0002, "epoch": 0.029336266960029337, "step": 40}, {"loss": 1.3112, "grad_norm": 1.5133819580078125, "learning_rate": 0.0002, "epoch": 0.03667033370003667, "step": 50}, {"loss": 1.3132, "grad_norm": 0.9173883199691772, "learning_rate": 0.0002, "epoch": 0.04400440044004401, "step": 60}, {"loss": 1.2844, "grad_norm": 0.4619861841201782, "learning_rate": 0.0002, "epoch": 0.05133846718005134, "step": 70}, {"loss": 1.2108, "grad_norm": 0.46118637919425964, "learning_rate": 0.0002, "epoch": 0.058672533920058674, "step": 80}, {"loss": 1.3441, "grad_norm": 0.4468648135662079, "learning_rate": 0.0002, "epoch": 0.066006600660066, "step": 90}, {"loss": 1.1863, "grad_norm": 0.46123769879341125, "learning_rate": 0.0002, "epoch": 0.07334066740007333, "step": 100}, {"loss": 1.2772, "grad_norm": 0.4859139025211334, "learning_rate": 0.0002, "epoch": 0.08067473414008068, "step": 110}, {"loss": 1.2087, "grad_norm": 0.4384922385215759, "learning_rate": 0.0002, "epoch": 0.08800880088008801, "step": 120}, {"loss": 1.2927, "grad_norm": 0.39519360661506653, "learning_rate": 0.0002, "epoch": 0.09534286762009535, "step": 130}, {"loss": 1.2349, "grad_norm": 0.4049859344959259, "learning_rate": 0.0002, "epoch": 0.10267693436010268, "step": 140}, {"loss": 1.293, "grad_norm": 0.4605638086795807, "learning_rate": 0.0002, "epoch": 0.11001100110011001, "step": 150}, {"loss": 1.2659, "grad_norm": 0.4201928377151489, "learning_rate": 0.0002, "epoch": 0.11734506784011735, "step": 160}, {"loss": 1.3961, "grad_norm": 0.5367777347564697, "learning_rate": 0.0002, "epoch": 0.12467913458012468, "step": 170}, {"loss": 1.2481, "grad_norm": 0.41752299666404724, "learning_rate": 0.0002, "epoch": 0.132013201320132, "step": 180}, {"loss": 1.207, "grad_norm": 0.31597763299942017, "learning_rate": 0.0002, "epoch": 0.13934726806013933, "step": 190}, {"loss": 1.2441, "grad_norm": 0.7468788623809814, "learning_rate": 0.0002, "epoch": 0.14668133480014667, "step": 200}, {"loss": 1.199, "grad_norm": 0.3403034508228302, "learning_rate": 0.0002, "epoch": 0.15401540154015403, "step": 210}, {"loss": 1.2439, "grad_norm": 0.34240293502807617, "learning_rate": 0.0002, "epoch": 0.16134946828016136, "step": 220}, {"loss": 1.2022, "grad_norm": 0.356158971786499, "learning_rate": 0.0002, "epoch": 0.1686835350201687, "step": 230}, {"loss": 1.207, "grad_norm": 0.3448857367038727, "learning_rate": 0.0002, "epoch": 0.17601760176017603, "step": 240}, {"loss": 1.2156, "grad_norm": 0.3475699722766876, "learning_rate": 0.0002, "epoch": 0.18335166850018336, "step": 250}, {"loss": 1.1551, "grad_norm": 0.2770358622074127, "learning_rate": 0.0002, "epoch": 0.1906857352401907, "step": 260}, {"loss": 1.2238, "grad_norm": 0.4310270845890045, "learning_rate": 0.0002, "epoch": 0.19801980198019803, "step": 270}, {"loss": 1.2917, "grad_norm": 0.335041880607605, "learning_rate": 0.0002, "epoch": 0.20535386872020536, "step": 280}, {"loss": 1.0959, "grad_norm": 0.3420602083206177, "learning_rate": 0.0002, "epoch": 0.2126879354602127, "step": 290}, {"loss": 1.1232, "grad_norm": 0.325001060962677, "learning_rate": 0.0002, "epoch": 0.22002200220022003, "step": 300}, {"loss": 1.2007, "grad_norm": 0.3027827739715576, "learning_rate": 0.0002, "epoch": 0.22735606894022736, "step": 310}, {"loss": 1.1803, "grad_norm": 0.435550719499588, "learning_rate": 0.0002, "epoch": 0.2346901356802347, "step": 320}, {"loss": 1.2045, "grad_norm": 0.3884522616863251, "learning_rate": 0.0002, "epoch": 0.24202420242024203, "step": 330}, {"loss": 1.2481, "grad_norm": 0.7736002206802368, "learning_rate": 0.0002, "epoch": 0.24935826916024936, "step": 340}, {"loss": 1.3606, "grad_norm": 0.35052821040153503, "learning_rate": 0.0002, "epoch": 0.2566923359002567, "step": 350}, {"loss": 1.2129, "grad_norm": 0.3311890959739685, "learning_rate": 0.0002, "epoch": 0.264026402640264, "step": 360}, {"loss": 1.2219, "grad_norm": 0.7473500370979309, "learning_rate": 0.0002, "epoch": 0.27136046938027136, "step": 370}, {"loss": 1.2712, "grad_norm": 0.3681875765323639, "learning_rate": 0.0002, "epoch": 0.27869453612027867, "step": 380}, {"loss": 1.2258, "grad_norm": 0.3764737844467163, "learning_rate": 0.0002, "epoch": 0.28602860286028603, "step": 390}, {"loss": 1.1917, "grad_norm": 0.4243989586830139, "learning_rate": 0.0002, "epoch": 0.29336266960029334, "step": 400}, {"loss": 1.199, "grad_norm": 0.2658531963825226, "learning_rate": 0.0002, "epoch": 0.3006967363403007, "step": 410}, {"loss": 1.1622, "grad_norm": 0.3436793386936188, "learning_rate": 0.0002, "epoch": 0.30803080308030806, "step": 420}, {"loss": 1.2953, "grad_norm": 0.5101129412651062, "learning_rate": 0.0002, "epoch": 0.31536486982031536, "step": 430}, {"loss": 1.1557, "grad_norm": 0.3319750726222992, "learning_rate": 0.0002, "epoch": 0.3226989365603227, "step": 440}, {"loss": 1.1804, "grad_norm": 0.385148286819458, "learning_rate": 0.0002, "epoch": 0.33003300330033003, "step": 450}, {"loss": 1.1808, "grad_norm": 0.3477935791015625, "learning_rate": 0.0002, "epoch": 0.3373670700403374, "step": 460}, {"loss": 1.1877, "grad_norm": 0.29748716950416565, "learning_rate": 0.0002, "epoch": 0.3447011367803447, "step": 470}, {"loss": 1.19, "grad_norm": 0.34083324670791626, "learning_rate": 0.0002, "epoch": 0.35203520352035206, "step": 480}, {"loss": 1.2, "grad_norm": 0.36904552578926086, "learning_rate": 0.0002, "epoch": 0.35936927026035936, "step": 490}, {"loss": 1.2223, "grad_norm": 0.315483033657074, "learning_rate": 0.0002, "epoch": 0.3667033370003667, "step": 500}, {"loss": 1.1461, "grad_norm": 0.44897955656051636, "learning_rate": 0.0002, "epoch": 0.37403740374037403, "step": 510}, {"loss": 1.3035, "grad_norm": 0.3160701394081116, "learning_rate": 0.0002, "epoch": 0.3813714704803814, "step": 520}, {"loss": 1.3197, "grad_norm": 0.29584741592407227, "learning_rate": 0.0002, "epoch": 0.3887055372203887, "step": 530}, {"loss": 1.2983, "grad_norm": 0.5430002808570862, "learning_rate": 0.0002, "epoch": 0.39603960396039606, "step": 540}, {"loss": 1.2459, "grad_norm": 0.2908070683479309, "learning_rate": 0.0002, "epoch": 0.40337367070040336, "step": 550}, {"loss": 1.2384, "grad_norm": 0.35066530108451843, "learning_rate": 0.0002, "epoch": 0.4107077374404107, "step": 560}, {"loss": 1.1784, "grad_norm": 0.37588003277778625, "learning_rate": 0.0002, "epoch": 0.41804180418041803, "step": 570}, {"loss": 1.2334, "grad_norm": 0.3112126886844635, "learning_rate": 0.0002, "epoch": 0.4253758709204254, "step": 580}, {"loss": 1.1439, "grad_norm": 0.35577139258384705, "learning_rate": 0.0002, "epoch": 0.4327099376604327, "step": 590}, {"loss": 1.184, "grad_norm": 0.31706422567367554, "learning_rate": 0.0002, "epoch": 0.44004400440044006, "step": 600}, {"loss": 1.2081, "grad_norm": 0.3249092102050781, "learning_rate": 0.0002, "epoch": 0.44737807114044736, "step": 610}, {"loss": 1.0824, "grad_norm": 0.3842705488204956, "learning_rate": 0.0002, "epoch": 0.4547121378804547, "step": 620}, {"loss": 1.2257, "grad_norm": 0.390991747379303, "learning_rate": 0.0002, "epoch": 0.46204620462046203, "step": 630}, {"loss": 1.1954, "grad_norm": 0.27532413601875305, "learning_rate": 0.0002, "epoch": 0.4693802713604694, "step": 640}, {"loss": 1.1058, "grad_norm": 0.31412816047668457, "learning_rate": 0.0002, "epoch": 0.4767143381004767, "step": 650}, {"loss": 1.1312, "grad_norm": 0.32117101550102234, "learning_rate": 0.0002, "epoch": 0.48404840484048406, "step": 660}, {"loss": 1.2423, "grad_norm": 0.3810010254383087, "learning_rate": 0.0002, "epoch": 0.49138247158049136, "step": 670}, {"loss": 1.1978, "grad_norm": 0.36289164423942566, "learning_rate": 0.0002, "epoch": 0.4987165383204987, "step": 680}, {"loss": 1.2034, "grad_norm": 0.34458720684051514, "learning_rate": 0.0002, "epoch": 0.506050605060506, "step": 690}, {"loss": 1.1756, "grad_norm": 0.32844600081443787, "learning_rate": 0.0002, "epoch": 0.5133846718005134, "step": 700}, {"loss": 1.0807, "grad_norm": 0.3144175708293915, "learning_rate": 0.0002, "epoch": 0.5207187385405208, "step": 710}, {"loss": 1.1952, "grad_norm": 0.3898887634277344, "learning_rate": 0.0002, "epoch": 0.528052805280528, "step": 720}, {"loss": 1.1244, "grad_norm": 1.3220758438110352, "learning_rate": 0.0002, "epoch": 0.5353868720205354, "step": 730}, {"loss": 1.227, "grad_norm": 0.3635874390602112, "learning_rate": 0.0002, "epoch": 0.5427209387605427, "step": 740}, {"loss": 1.2169, "grad_norm": 0.3138217628002167, "learning_rate": 0.0002, "epoch": 0.5500550055005501, "step": 750}, {"loss": 1.1516, "grad_norm": 0.4063207805156708, "learning_rate": 0.0002, "epoch": 0.5573890722405573, "step": 760}, {"loss": 1.1954, "grad_norm": 0.3926219940185547, "learning_rate": 0.0002, "epoch": 0.5647231389805647, "step": 770}, {"loss": 1.1726, "grad_norm": 0.31954652070999146, "learning_rate": 0.0002, "epoch": 0.5720572057205721, "step": 780}, {"loss": 1.2977, "grad_norm": 0.4248711168766022, "learning_rate": 0.0002, "epoch": 0.5793912724605794, "step": 790}, {"loss": 1.1728, "grad_norm": 0.643004834651947, "learning_rate": 0.0002, "epoch": 0.5867253392005867, "step": 800}, {"loss": 1.1793, "grad_norm": 0.3479592800140381, "learning_rate": 0.0002, "epoch": 0.594059405940594, "step": 810}, {"loss": 1.2426, "grad_norm": 0.4684754014015198, "learning_rate": 0.0002, "epoch": 0.6013934726806014, "step": 820}, {"loss": 1.2002, "grad_norm": 0.3739790916442871, "learning_rate": 0.0002, "epoch": 0.6087275394206088, "step": 830}, {"loss": 1.2139, "grad_norm": 0.40884748101234436, "learning_rate": 0.0002, "epoch": 0.6160616061606161, "step": 840}, {"loss": 1.1557, "grad_norm": 0.9722164273262024, "learning_rate": 0.0002, "epoch": 0.6233956729006234, "step": 850}, {"loss": 1.3069, "grad_norm": 0.42992347478866577, "learning_rate": 0.0002, "epoch": 0.6307297396406307, "step": 860}, {"loss": 1.1339, "grad_norm": 0.36654195189476013, "learning_rate": 0.0002, "epoch": 0.6380638063806381, "step": 870}, {"loss": 1.1932, "grad_norm": 0.4113832116127014, "learning_rate": 0.0002, "epoch": 0.6453978731206454, "step": 880}, {"loss": 1.2163, "grad_norm": 0.2948838770389557, "learning_rate": 0.0002, "epoch": 0.6527319398606527, "step": 890}, {"loss": 1.1081, "grad_norm": 0.38330280780792236, "learning_rate": 0.0002, "epoch": 0.6600660066006601, "step": 900}, {"loss": 1.1342, "grad_norm": 0.4428867697715759, "learning_rate": 0.0002, "epoch": 0.6674000733406674, "step": 910}, {"loss": 1.1021, "grad_norm": 0.23659265041351318, "learning_rate": 0.0002, "epoch": 0.6747341400806748, "step": 920}, {"loss": 1.1226, "grad_norm": 0.323685884475708, "learning_rate": 0.0002, "epoch": 0.682068206820682, "step": 930}, {"loss": 1.0853, "grad_norm": 0.39157727360725403, "learning_rate": 0.0002, "epoch": 0.6894022735606894, "step": 940}, {"loss": 1.1435, "grad_norm": 0.27189481258392334, "learning_rate": 0.0002, "epoch": 0.6967363403006968, "step": 950}, {"loss": 1.1033, "grad_norm": 0.529883861541748, "learning_rate": 0.0002, "epoch": 0.7040704070407041, "step": 960}, {"loss": 1.139, "grad_norm": 0.34758689999580383, "learning_rate": 0.0002, "epoch": 0.7114044737807114, "step": 970}, {"loss": 1.2197, "grad_norm": 0.831749439239502, "learning_rate": 0.0002, "epoch": 0.7187385405207187, "step": 980}, {"loss": 1.158, "grad_norm": 0.4438304007053375, "learning_rate": 0.0002, "epoch": 0.7260726072607261, "step": 990}, {"loss": 1.1021, "grad_norm": 0.33840006589889526, "learning_rate": 0.0002, "epoch": 0.7334066740007334, "step": 1000}, {"loss": 1.254, "grad_norm": 0.3454797863960266, "learning_rate": 0.0002, "epoch": 0.7407407407407407, "step": 1010}, {"loss": 1.106, "grad_norm": 0.38999441266059875, "learning_rate": 0.0002, "epoch": 0.7480748074807481, "step": 1020}, {"loss": 1.1428, "grad_norm": 0.2829911708831787, "learning_rate": 0.0002, "epoch": 0.7554088742207554, "step": 1030}, {"loss": 1.2123, "grad_norm": 0.36918163299560547, "learning_rate": 0.0002, "epoch": 0.7627429409607628, "step": 1040}, {"loss": 1.3028, "grad_norm": 0.3415680229663849, "learning_rate": 0.0002, "epoch": 0.77007700770077, "step": 1050}, {"loss": 1.1939, "grad_norm": 0.2974182963371277, "learning_rate": 0.0002, "epoch": 0.7774110744407774, "step": 1060}, {"loss": 1.194, "grad_norm": 0.3880919814109802, "learning_rate": 0.0002, "epoch": 0.7847451411807848, "step": 1070}, {"loss": 1.1095, "grad_norm": 0.33503302931785583, "learning_rate": 0.0002, "epoch": 0.7920792079207921, "step": 1080}, {"loss": 1.2111, "grad_norm": 0.3728407025337219, "learning_rate": 0.0002, "epoch": 0.7994132746607994, "step": 1090}, {"loss": 1.0835, "grad_norm": 0.3509373664855957, "learning_rate": 0.0002, "epoch": 0.8067473414008067, "step": 1100}, {"loss": 1.2661, "grad_norm": 0.42228564620018005, "learning_rate": 0.0002, "epoch": 0.8140814081408141, "step": 1110}, {"loss": 1.1788, "grad_norm": 0.313467800617218, "learning_rate": 0.0002, "epoch": 0.8214154748808215, "step": 1120}, {"loss": 1.1971, "grad_norm": 0.3378850817680359, "learning_rate": 0.0002, "epoch": 0.8287495416208287, "step": 1130}, {"loss": 1.1238, "grad_norm": 0.43200382590293884, "learning_rate": 0.0002, "epoch": 0.8360836083608361, "step": 1140}, {"loss": 1.3203, "grad_norm": 0.3309599459171295, "learning_rate": 0.0002, "epoch": 0.8434176751008434, "step": 1150}, {"loss": 1.1062, "grad_norm": 0.3526846170425415, "learning_rate": 0.0002, "epoch": 0.8507517418408508, "step": 1160}, {"loss": 1.0851, "grad_norm": 1.2722247838974, "learning_rate": 0.0002, "epoch": 0.858085808580858, "step": 1170}, {"loss": 1.0785, "grad_norm": 0.34142059087753296, "learning_rate": 0.0002, "epoch": 0.8654198753208654, "step": 1180}, {"loss": 1.2187, "grad_norm": 0.3805823028087616, "learning_rate": 0.0002, "epoch": 0.8727539420608728, "step": 1190}, {"loss": 1.1215, "grad_norm": 0.3931232690811157, "learning_rate": 0.0002, "epoch": 0.8800880088008801, "step": 1200}, {"loss": 1.0948, "grad_norm": 0.2937372624874115, "learning_rate": 0.0002, "epoch": 0.8874220755408874, "step": 1210}, {"loss": 1.1228, "grad_norm": 0.3757196366786957, "learning_rate": 0.0002, "epoch": 0.8947561422808947, "step": 1220}, {"loss": 1.1222, "grad_norm": 0.3502705991268158, "learning_rate": 0.0002, "epoch": 0.9020902090209021, "step": 1230}, {"loss": 1.2242, "grad_norm": 0.32758915424346924, "learning_rate": 0.0002, "epoch": 0.9094242757609095, "step": 1240}, {"loss": 1.215, "grad_norm": 0.37199416756629944, "learning_rate": 0.0002, "epoch": 0.9167583425009168, "step": 1250}, {"loss": 1.1225, "grad_norm": 0.3551490604877472, "learning_rate": 0.0002, "epoch": 0.9240924092409241, "step": 1260}, {"loss": 1.1966, "grad_norm": 0.2859550714492798, "learning_rate": 0.0002, "epoch": 0.9314264759809314, "step": 1270}, {"loss": 1.2186, "grad_norm": 0.427990585565567, "learning_rate": 0.0002, "epoch": 0.9387605427209388, "step": 1280}, {"loss": 1.2848, "grad_norm": 0.33717992901802063, "learning_rate": 0.0002, "epoch": 0.9460946094609461, "step": 1290}, {"loss": 1.1656, "grad_norm": 0.30225634574890137, "learning_rate": 0.0002, "epoch": 0.9534286762009534, "step": 1300}, {"loss": 1.2404, "grad_norm": 0.385821133852005, "learning_rate": 0.0002, "epoch": 0.9607627429409608, "step": 1310}, {"loss": 1.1932, "grad_norm": 0.35278066992759705, "learning_rate": 0.0002, "epoch": 0.9680968096809681, "step": 1320}, {"loss": 1.1071, "grad_norm": 0.49987098574638367, "learning_rate": 0.0002, "epoch": 0.9754308764209755, "step": 1330}, {"loss": 1.2259, "grad_norm": 0.3842747211456299, "learning_rate": 0.0002, "epoch": 0.9827649431609827, "step": 1340}, {"loss": 1.0862, "grad_norm": 0.6274653673171997, "learning_rate": 0.0002, "epoch": 0.9900990099009901, "step": 1350}, {"loss": 1.124, "grad_norm": 0.5239808559417725, "learning_rate": 0.0002, "epoch": 0.9974330766409975, "step": 1360}, {"eval_loss": 1.1822267770767212, "eval_runtime": 32.7389, "eval_samples_per_second": 13.165, "eval_steps_per_second": 1.649, "epoch": 0.9996332966629996, "step": 1363}, {"loss": 1.096, "grad_norm": 0.45311301946640015, "learning_rate": 0.0002, "epoch": 1.0047671433810048, "step": 1370}, {"loss": 1.0143, "grad_norm": 0.29685574769973755, "learning_rate": 0.0002, "epoch": 1.012101210121012, "step": 1380}, {"loss": 1.0302, "grad_norm": 0.3290937840938568, "learning_rate": 0.0002, "epoch": 1.0194352768610195, "step": 1390}, {"loss": 1.0295, "grad_norm": 0.3801758587360382, "learning_rate": 0.0002, "epoch": 1.0267693436010268, "step": 1400}, {"loss": 1.1226, "grad_norm": 0.794174313545227, "learning_rate": 0.0002, "epoch": 1.034103410341034, "step": 1410}, {"loss": 1.2232, "grad_norm": 0.3854154646396637, "learning_rate": 0.0002, "epoch": 1.0414374770810415, "step": 1420}, {"loss": 1.0652, "grad_norm": 0.32702451944351196, "learning_rate": 0.0002, "epoch": 1.0487715438210488, "step": 1430}, {"loss": 1.1144, "grad_norm": 0.7815203666687012, "learning_rate": 0.0002, "epoch": 1.056105610561056, "step": 1440}, {"loss": 1.1316, "grad_norm": 0.3087436854839325, "learning_rate": 0.0002, "epoch": 1.0634396773010635, "step": 1450}, {"loss": 1.1124, "grad_norm": 0.3847602903842926, "learning_rate": 0.0002, "epoch": 1.0707737440410707, "step": 1460}, {"loss": 1.1428, "grad_norm": 0.3693031370639801, "learning_rate": 0.0002, "epoch": 1.0781078107810782, "step": 1470}, {"loss": 1.0995, "grad_norm": 0.4111202359199524, "learning_rate": 0.0002, "epoch": 1.0854418775210855, "step": 1480}, {"loss": 1.0961, "grad_norm": 0.41452381014823914, "learning_rate": 0.0002, "epoch": 1.0927759442610927, "step": 1490}, {"loss": 1.1068, "grad_norm": 0.3336445093154907, "learning_rate": 0.0002, "epoch": 1.1001100110011002, "step": 1500}, {"loss": 1.0556, "grad_norm": 0.3923407793045044, "learning_rate": 0.0002, "epoch": 1.1074440777411074, "step": 1510}, {"loss": 1.1644, "grad_norm": 0.46215683221817017, "learning_rate": 0.0002, "epoch": 1.1147781444811147, "step": 1520}, {"loss": 1.1133, "grad_norm": 0.3592156767845154, "learning_rate": 0.0002, "epoch": 1.1221122112211221, "step": 1530}, {"loss": 1.0957, "grad_norm": 0.361110657453537, "learning_rate": 0.0002, "epoch": 1.1294462779611294, "step": 1540}, {"loss": 1.1553, "grad_norm": 0.5317131280899048, "learning_rate": 0.0002, "epoch": 1.1367803447011369, "step": 1550}, {"loss": 1.0368, "grad_norm": 0.3882388174533844, "learning_rate": 0.0002, "epoch": 1.1441144114411441, "step": 1560}, {"loss": 1.0805, "grad_norm": 0.3259428143501282, "learning_rate": 0.0002, "epoch": 1.1514484781811514, "step": 1570}, {"loss": 1.1819, "grad_norm": 0.410935640335083, "learning_rate": 0.0002, "epoch": 1.1587825449211588, "step": 1580}, {"loss": 1.1143, "grad_norm": 0.44940185546875, "learning_rate": 0.0002, "epoch": 1.166116611661166, "step": 1590}, {"loss": 1.0334, "grad_norm": 0.5106484293937683, "learning_rate": 0.0002, "epoch": 1.1734506784011733, "step": 1600}, {"loss": 1.2376, "grad_norm": 0.6603665947914124, "learning_rate": 0.0002, "epoch": 1.1807847451411808, "step": 1610}, {"loss": 1.1227, "grad_norm": 0.4799964129924774, "learning_rate": 0.0002, "epoch": 1.188118811881188, "step": 1620}, {"loss": 1.1191, "grad_norm": 0.4389883279800415, "learning_rate": 0.0002, "epoch": 1.1954528786211955, "step": 1630}, {"loss": 1.0667, "grad_norm": 0.4188813269138336, "learning_rate": 0.0002, "epoch": 1.2027869453612028, "step": 1640}, {"loss": 1.0605, "grad_norm": 0.7132157683372498, "learning_rate": 0.0002, "epoch": 1.21012101210121, "step": 1650}, {"loss": 1.0204, "grad_norm": 0.507480263710022, "learning_rate": 0.0002, "epoch": 1.2174550788412175, "step": 1660}, {"loss": 0.9948, "grad_norm": 0.9452332854270935, "learning_rate": 0.0002, "epoch": 1.2247891455812248, "step": 1670}, {"loss": 1.0228, "grad_norm": 0.4121614992618561, "learning_rate": 0.0002, "epoch": 1.2321232123212322, "step": 1680}, {"loss": 1.0366, "grad_norm": 0.34230247139930725, "learning_rate": 0.0002, "epoch": 1.2394572790612395, "step": 1690}, {"loss": 1.1289, "grad_norm": 0.4026208817958832, "learning_rate": 0.0002, "epoch": 1.2467913458012467, "step": 1700}, {"loss": 1.0206, "grad_norm": 0.46673697233200073, "learning_rate": 0.0002, "epoch": 1.2541254125412542, "step": 1710}, {"loss": 1.0827, "grad_norm": 0.38349825143814087, "learning_rate": 0.0002, "epoch": 1.2614594792812615, "step": 1720}, {"loss": 1.0356, "grad_norm": 0.4049997627735138, "learning_rate": 0.0002, "epoch": 1.2687935460212687, "step": 1730}, {"loss": 0.9504, "grad_norm": 0.3417615294456482, "learning_rate": 0.0002, "epoch": 1.2761276127612762, "step": 1740}, {"loss": 1.094, "grad_norm": 0.4277614951133728, "learning_rate": 0.0002, "epoch": 1.2834616795012834, "step": 1750}, {"loss": 0.9938, "grad_norm": 0.5864202976226807, "learning_rate": 0.0002, "epoch": 1.2907957462412907, "step": 1760}, {"loss": 1.1167, "grad_norm": 0.7097493410110474, "learning_rate": 0.0002, "epoch": 1.2981298129812981, "step": 1770}, {"loss": 1.1132, "grad_norm": 0.3145381212234497, "learning_rate": 0.0002, "epoch": 1.3054638797213054, "step": 1780}, {"loss": 1.1099, "grad_norm": 0.5116165280342102, "learning_rate": 0.0002, "epoch": 1.3127979464613129, "step": 1790}, {"loss": 1.0765, "grad_norm": 0.7469736337661743, "learning_rate": 0.0002, "epoch": 1.3201320132013201, "step": 1800}, {"loss": 1.0663, "grad_norm": 0.32272255420684814, "learning_rate": 0.0002, "epoch": 1.3274660799413276, "step": 1810}, {"loss": 0.9887, "grad_norm": 0.3534623086452484, "learning_rate": 0.0002, "epoch": 1.3348001466813348, "step": 1820}, {"loss": 1.1628, "grad_norm": 0.36127907037734985, "learning_rate": 0.0002, "epoch": 1.342134213421342, "step": 1830}, {"loss": 1.0972, "grad_norm": 0.4072401523590088, "learning_rate": 0.0002, "epoch": 1.3494682801613496, "step": 1840}, {"loss": 1.1267, "grad_norm": 0.3769161105155945, "learning_rate": 0.0002, "epoch": 1.3568023469013568, "step": 1850}, {"loss": 1.0173, "grad_norm": 0.412883460521698, "learning_rate": 0.0002, "epoch": 1.364136413641364, "step": 1860}, {"loss": 1.0265, "grad_norm": 0.3735875189304352, "learning_rate": 0.0002, "epoch": 1.3714704803813715, "step": 1870}, {"loss": 1.1061, "grad_norm": 0.39158159494400024, "learning_rate": 0.0002, "epoch": 1.3788045471213788, "step": 1880}, {"loss": 1.0433, "grad_norm": 0.44431769847869873, "learning_rate": 0.0002, "epoch": 1.386138613861386, "step": 1890}, {"loss": 1.0216, "grad_norm": 0.37772801518440247, "learning_rate": 0.0002, "epoch": 1.3934726806013935, "step": 1900}, {"loss": 1.0674, "grad_norm": 0.4056641757488251, "learning_rate": 0.0002, "epoch": 1.4008067473414008, "step": 1910}, {"loss": 1.0256, "grad_norm": 0.41612377762794495, "learning_rate": 0.0002, "epoch": 1.408140814081408, "step": 1920}, {"loss": 1.0467, "grad_norm": 0.41153013706207275, "learning_rate": 0.0002, "epoch": 1.4154748808214155, "step": 1930}, {"loss": 1.1062, "grad_norm": 0.387845516204834, "learning_rate": 0.0002, "epoch": 1.4228089475614227, "step": 1940}, {"loss": 1.1094, "grad_norm": 0.3809587061405182, "learning_rate": 0.0002, "epoch": 1.4301430143014302, "step": 1950}, {"loss": 1.0461, "grad_norm": 0.3625726103782654, "learning_rate": 0.0002, "epoch": 1.4374770810414375, "step": 1960}, {"loss": 0.9983, "grad_norm": 0.5294290781021118, "learning_rate": 0.0002, "epoch": 1.444811147781445, "step": 1970}, {"loss": 1.1114, "grad_norm": 0.39975494146347046, "learning_rate": 0.0002, "epoch": 1.4521452145214522, "step": 1980}, {"loss": 0.9704, "grad_norm": 0.4181167185306549, "learning_rate": 0.0002, "epoch": 1.4594792812614594, "step": 1990}, {"loss": 1.1146, "grad_norm": 0.42001503705978394, "learning_rate": 0.0002, "epoch": 1.466813348001467, "step": 2000}, {"loss": 1.1266, "grad_norm": 0.4877578616142273, "learning_rate": 0.0002, "epoch": 1.4741474147414741, "step": 2010}, {"loss": 1.1012, "grad_norm": 0.4050969183444977, "learning_rate": 0.0002, "epoch": 1.4814814814814814, "step": 2020}, {"loss": 1.0562, "grad_norm": 0.39068883657455444, "learning_rate": 0.0002, "epoch": 1.4888155482214889, "step": 2030}, {"loss": 1.0464, "grad_norm": 0.421282559633255, "learning_rate": 0.0002, "epoch": 1.4961496149614961, "step": 2040}, {"loss": 1.0532, "grad_norm": 0.47092297673225403, "learning_rate": 0.0002, "epoch": 1.5034836817015034, "step": 2050}, {"loss": 0.9348, "grad_norm": 0.39688974618911743, "learning_rate": 0.0002, "epoch": 1.5108177484415108, "step": 2060}, {"loss": 1.08, "grad_norm": 0.5529879331588745, "learning_rate": 0.0002, "epoch": 1.5181518151815183, "step": 2070}, {"loss": 1.1836, "grad_norm": 0.4879782199859619, "learning_rate": 0.0002, "epoch": 1.5254858819215253, "step": 2080}, {"loss": 1.0432, "grad_norm": 0.5517361164093018, "learning_rate": 0.0002, "epoch": 1.5328199486615328, "step": 2090}, {"loss": 1.0433, "grad_norm": 0.44015637040138245, "learning_rate": 0.0002, "epoch": 1.5401540154015403, "step": 2100}, {"loss": 1.1873, "grad_norm": 0.5435167551040649, "learning_rate": 0.0002, "epoch": 1.5474880821415475, "step": 2110}, {"loss": 1.1076, "grad_norm": 0.5714033246040344, "learning_rate": 0.0002, "epoch": 1.5548221488815548, "step": 2120}, {"loss": 1.1107, "grad_norm": 0.31732529401779175, "learning_rate": 0.0002, "epoch": 1.5621562156215623, "step": 2130}, {"loss": 1.0817, "grad_norm": 0.49068278074264526, "learning_rate": 0.0002, "epoch": 1.5694902823615695, "step": 2140}, {"loss": 1.0254, "grad_norm": 0.46851542592048645, "learning_rate": 0.0002, "epoch": 1.5768243491015768, "step": 2150}, {"loss": 1.0623, "grad_norm": 0.5083092451095581, "learning_rate": 0.0002, "epoch": 1.5841584158415842, "step": 2160}, {"loss": 1.0603, "grad_norm": 0.9822936058044434, "learning_rate": 0.0002, "epoch": 1.5914924825815915, "step": 2170}, {"loss": 0.9986, "grad_norm": 0.4575989246368408, "learning_rate": 0.0002, "epoch": 1.5988265493215987, "step": 2180}, {"loss": 1.1292, "grad_norm": 0.47444286942481995, "learning_rate": 0.0002, "epoch": 1.6061606160616062, "step": 2190}, {"loss": 1.0136, "grad_norm": 0.7208226919174194, "learning_rate": 0.0002, "epoch": 1.6134946828016135, "step": 2200}, {"loss": 1.15, "grad_norm": 0.43791481852531433, "learning_rate": 0.0002, "epoch": 1.6208287495416207, "step": 2210}, {"loss": 1.0961, "grad_norm": 0.5245792865753174, "learning_rate": 0.0002, "epoch": 1.6281628162816282, "step": 2220}, {"loss": 0.9957, "grad_norm": 0.39289429783821106, "learning_rate": 0.0002, "epoch": 1.6354968830216357, "step": 2230}, {"loss": 1.133, "grad_norm": 0.6106135845184326, "learning_rate": 0.0002, "epoch": 1.6428309497616427, "step": 2240}, {"loss": 1.0129, "grad_norm": 0.3722580671310425, "learning_rate": 0.0002, "epoch": 1.6501650165016502, "step": 2250}, {"loss": 1.0446, "grad_norm": 0.3649403750896454, "learning_rate": 0.0002, "epoch": 1.6574990832416576, "step": 2260}, {"loss": 1.0037, "grad_norm": 0.46514248847961426, "learning_rate": 0.0002, "epoch": 1.6648331499816649, "step": 2270}, {"loss": 1.0022, "grad_norm": 0.42034927010536194, "learning_rate": 0.0002, "epoch": 1.6721672167216721, "step": 2280}, {"loss": 1.1362, "grad_norm": 0.45202910900115967, "learning_rate": 0.0002, "epoch": 1.6795012834616796, "step": 2290}, {"loss": 1.0866, "grad_norm": 0.36257603764533997, "learning_rate": 0.0002, "epoch": 1.6868353502016868, "step": 2300}, {"loss": 1.0973, "grad_norm": 0.6340323090553284, "learning_rate": 0.0002, "epoch": 1.694169416941694, "step": 2310}, {"loss": 1.0615, "grad_norm": 0.4352878928184509, "learning_rate": 0.0002, "epoch": 1.7015034836817016, "step": 2320}, {"loss": 1.0629, "grad_norm": 0.45029792189598083, "learning_rate": 0.0002, "epoch": 1.7088375504217088, "step": 2330}, {"loss": 0.9621, "grad_norm": 0.3891315758228302, "learning_rate": 0.0002, "epoch": 1.716171617161716, "step": 2340}, {"loss": 0.9779, "grad_norm": 0.35180050134658813, "learning_rate": 0.0002, "epoch": 1.7235056839017235, "step": 2350}, {"loss": 1.0368, "grad_norm": 0.42367449402809143, "learning_rate": 0.0002, "epoch": 1.7308397506417308, "step": 2360}, {"loss": 1.0376, "grad_norm": 0.4553675353527069, "learning_rate": 0.0002, "epoch": 1.738173817381738, "step": 2370}, {"loss": 1.1467, "grad_norm": 0.5944654941558838, "learning_rate": 0.0002, "epoch": 1.7455078841217455, "step": 2380}, {"loss": 1.0548, "grad_norm": 0.3479664623737335, "learning_rate": 0.0002, "epoch": 1.752841950861753, "step": 2390}, {"loss": 1.0798, "grad_norm": 0.3585502505302429, "learning_rate": 0.0002, "epoch": 1.76017601760176, "step": 2400}, {"loss": 1.0983, "grad_norm": 0.4263346493244171, "learning_rate": 0.0002, "epoch": 1.7675100843417675, "step": 2410}, {"loss": 1.054, "grad_norm": 0.5476409196853638, "learning_rate": 0.0002, "epoch": 1.774844151081775, "step": 2420}, {"loss": 1.1615, "grad_norm": 0.3694186508655548, "learning_rate": 0.0002, "epoch": 1.7821782178217822, "step": 2430}, {"loss": 1.1343, "grad_norm": 0.9185658693313599, "learning_rate": 0.0002, "epoch": 1.7895122845617895, "step": 2440}, {"loss": 1.0764, "grad_norm": 0.7171908020973206, "learning_rate": 0.0002, "epoch": 1.796846351301797, "step": 2450}, {"loss": 1.1154, "grad_norm": 0.550658643245697, "learning_rate": 0.0002, "epoch": 1.8041804180418042, "step": 2460}, {"loss": 0.9975, "grad_norm": 0.4075568914413452, "learning_rate": 0.0002, "epoch": 1.8115144847818114, "step": 2470}, {"loss": 1.0935, "grad_norm": 0.3790127635002136, "learning_rate": 0.0002, "epoch": 1.818848551521819, "step": 2480}, {"loss": 0.9839, "grad_norm": 0.3576384484767914, "learning_rate": 0.0002, "epoch": 1.8261826182618262, "step": 2490}, {"loss": 1.1369, "grad_norm": 0.3919370770454407, "learning_rate": 0.0002, "epoch": 1.8335166850018334, "step": 2500}, {"loss": 0.9985, "grad_norm": 0.485083669424057, "learning_rate": 0.0002, "epoch": 1.8408507517418409, "step": 2510}, {"loss": 1.1585, "grad_norm": 0.4564347565174103, "learning_rate": 0.0002, "epoch": 1.8481848184818483, "step": 2520}, {"loss": 1.0944, "grad_norm": 0.3613106608390808, "learning_rate": 0.0002, "epoch": 1.8555188852218554, "step": 2530}, {"loss": 1.0819, "grad_norm": 0.39600759744644165, "learning_rate": 0.0002, "epoch": 1.8628529519618628, "step": 2540}, {"loss": 0.9453, "grad_norm": 1.123499870300293, "learning_rate": 0.0002, "epoch": 1.8701870187018703, "step": 2550}, {"loss": 1.0635, "grad_norm": 0.4612680673599243, "learning_rate": 0.0002, "epoch": 1.8775210854418776, "step": 2560}, {"loss": 1.0087, "grad_norm": 0.42745399475097656, "learning_rate": 0.0002, "epoch": 1.8848551521818848, "step": 2570}, {"loss": 1.0102, "grad_norm": 0.4055580198764801, "learning_rate": 0.0002, "epoch": 1.8921892189218923, "step": 2580}, {"loss": 1.0177, "grad_norm": 0.44174644351005554, "learning_rate": 0.0002, "epoch": 1.8995232856618995, "step": 2590}, {"loss": 0.9886, "grad_norm": 1.0228385925292969, "learning_rate": 0.0002, "epoch": 1.9068573524019068, "step": 2600}, {"loss": 1.0857, "grad_norm": 0.3496396243572235, "learning_rate": 0.0002, "epoch": 1.9141914191419143, "step": 2610}, {"loss": 1.0955, "grad_norm": 0.4191173017024994, "learning_rate": 0.0002, "epoch": 1.9215254858819215, "step": 2620}, {"loss": 1.0943, "grad_norm": 0.6778554916381836, "learning_rate": 0.0002, "epoch": 1.9288595526219288, "step": 2630}, {"loss": 1.0594, "grad_norm": 0.41992834210395813, "learning_rate": 0.0002, "epoch": 1.9361936193619362, "step": 2640}, {"loss": 1.1159, "grad_norm": 0.8760401010513306, "learning_rate": 0.0002, "epoch": 1.9435276861019435, "step": 2650}, {"loss": 1.0379, "grad_norm": 0.44049209356307983, "learning_rate": 0.0002, "epoch": 1.9508617528419507, "step": 2660}, {"loss": 1.1008, "grad_norm": 0.5651928782463074, "learning_rate": 0.0002, "epoch": 1.9581958195819582, "step": 2670}, {"loss": 1.1317, "grad_norm": 0.5292727947235107, "learning_rate": 0.0002, "epoch": 1.9655298863219657, "step": 2680}, {"loss": 1.1328, "grad_norm": 0.6012240648269653, "learning_rate": 0.0002, "epoch": 1.9728639530619727, "step": 2690}, {"loss": 1.0683, "grad_norm": 0.3945149779319763, "learning_rate": 0.0002, "epoch": 1.9801980198019802, "step": 2700}, {"loss": 1.0155, "grad_norm": 0.5732627511024475, "learning_rate": 0.0002, "epoch": 1.9875320865419877, "step": 2710}, {"loss": 0.9857, "grad_norm": 0.3963361084461212, "learning_rate": 0.0002, "epoch": 1.994866153281995, "step": 2720}, {"eval_loss": 1.1534006595611572, "eval_runtime": 32.7541, "eval_samples_per_second": 13.159, "eval_steps_per_second": 1.649, "epoch": 2.0, "step": 2727}, {"loss": 0.9624, "grad_norm": 0.48628315329551697, "learning_rate": 0.0002, "epoch": 2.002200220022002, "step": 2730}, {"loss": 0.9603, "grad_norm": 0.413875013589859, "learning_rate": 0.0002, "epoch": 2.0095342867620096, "step": 2740}, {"loss": 0.965, "grad_norm": 0.4988735616207123, "learning_rate": 0.0002, "epoch": 2.0168683535020167, "step": 2750}, {"loss": 0.9677, "grad_norm": 0.5634812712669373, "learning_rate": 0.0002, "epoch": 2.024202420242024, "step": 2760}, {"loss": 0.9547, "grad_norm": 0.48302653431892395, "learning_rate": 0.0002, "epoch": 2.0315364869820316, "step": 2770}, {"loss": 0.9346, "grad_norm": 0.49914175271987915, "learning_rate": 0.0002, "epoch": 2.038870553722039, "step": 2780}, {"loss": 0.904, "grad_norm": 1.14039945602417, "learning_rate": 0.0002, "epoch": 2.046204620462046, "step": 2790}, {"loss": 0.9588, "grad_norm": 0.6359720826148987, "learning_rate": 0.0002, "epoch": 2.0535386872020536, "step": 2800}, {"loss": 0.9031, "grad_norm": 0.4589158296585083, "learning_rate": 0.0002, "epoch": 2.060872753942061, "step": 2810}, {"loss": 0.9438, "grad_norm": 0.46255481243133545, "learning_rate": 0.0002, "epoch": 2.068206820682068, "step": 2820}, {"loss": 0.9464, "grad_norm": 0.6232137680053711, "learning_rate": 0.0002, "epoch": 2.0755408874220755, "step": 2830}, {"loss": 0.8978, "grad_norm": 0.41042178869247437, "learning_rate": 0.0002, "epoch": 2.082874954162083, "step": 2840}, {"loss": 0.8516, "grad_norm": 0.5334428548812866, "learning_rate": 0.0002, "epoch": 2.09020902090209, "step": 2850}, {"loss": 0.9313, "grad_norm": 0.8270058631896973, "learning_rate": 0.0002, "epoch": 2.0975430876420975, "step": 2860}, {"loss": 1.0064, "grad_norm": 0.6624533534049988, "learning_rate": 0.0002, "epoch": 2.104877154382105, "step": 2870}, {"loss": 0.9196, "grad_norm": 0.5448863506317139, "learning_rate": 0.0002, "epoch": 2.112211221122112, "step": 2880}, {"loss": 0.887, "grad_norm": 0.621482789516449, "learning_rate": 0.0002, "epoch": 2.1195452878621195, "step": 2890}, {"loss": 0.9702, "grad_norm": 0.4556255340576172, "learning_rate": 0.0002, "epoch": 2.126879354602127, "step": 2900}, {"loss": 0.9323, "grad_norm": 0.4620579183101654, "learning_rate": 0.0002, "epoch": 2.1342134213421344, "step": 2910}, {"loss": 0.836, "grad_norm": 0.9602415561676025, "learning_rate": 0.0002, "epoch": 2.1415474880821415, "step": 2920}, {"loss": 0.8826, "grad_norm": 0.587943971157074, "learning_rate": 0.0002, "epoch": 2.148881554822149, "step": 2930}, {"loss": 0.971, "grad_norm": 0.5121372938156128, "learning_rate": 0.0002, "epoch": 2.1562156215621564, "step": 2940}, {"loss": 0.8751, "grad_norm": 0.49424484372138977, "learning_rate": 0.0002, "epoch": 2.1635496883021634, "step": 2950}, {"loss": 0.8674, "grad_norm": 0.6312560439109802, "learning_rate": 0.0002, "epoch": 2.170883755042171, "step": 2960}, {"loss": 0.9791, "grad_norm": 0.5235576629638672, "learning_rate": 0.0002, "epoch": 2.1782178217821784, "step": 2970}, {"loss": 0.9706, "grad_norm": 0.5868439674377441, "learning_rate": 0.0002, "epoch": 2.1855518885221854, "step": 2980}, {"loss": 0.9338, "grad_norm": 0.42302873730659485, "learning_rate": 0.0002, "epoch": 2.192885955262193, "step": 2990}, {"loss": 0.9332, "grad_norm": 0.5097725987434387, "learning_rate": 0.0002, "epoch": 2.2002200220022003, "step": 3000}, {"loss": 0.9239, "grad_norm": 0.5091572403907776, "learning_rate": 0.0002, "epoch": 2.2075540887422074, "step": 3010}, {"loss": 0.8898, "grad_norm": 0.49433162808418274, "learning_rate": 0.0002, "epoch": 2.214888155482215, "step": 3020}, {"loss": 0.9734, "grad_norm": 0.5577368140220642, "learning_rate": 0.0002, "epoch": 2.2222222222222223, "step": 3030}, {"loss": 0.9033, "grad_norm": 0.6177583932876587, "learning_rate": 0.0002, "epoch": 2.2295562889622293, "step": 3040}, {"loss": 0.9882, "grad_norm": 0.5256719589233398, "learning_rate": 0.0002, "epoch": 2.236890355702237, "step": 3050}, {"loss": 0.9439, "grad_norm": 0.5001118183135986, "learning_rate": 0.0002, "epoch": 2.2442244224422443, "step": 3060}, {"loss": 0.8718, "grad_norm": 0.5721249580383301, "learning_rate": 0.0002, "epoch": 2.2515584891822513, "step": 3070}, {"loss": 1.0648, "grad_norm": 0.5325384140014648, "learning_rate": 0.0002, "epoch": 2.258892555922259, "step": 3080}, {"loss": 0.9843, "grad_norm": 0.5719189047813416, "learning_rate": 0.0002, "epoch": 2.2662266226622663, "step": 3090}, {"loss": 0.8633, "grad_norm": 0.6337835788726807, "learning_rate": 0.0002, "epoch": 2.2735606894022737, "step": 3100}, {"loss": 0.9962, "grad_norm": 0.5381836891174316, "learning_rate": 0.0002, "epoch": 2.2808947561422808, "step": 3110}, {"loss": 0.8265, "grad_norm": 0.5408531427383423, "learning_rate": 0.0002, "epoch": 2.2882288228822882, "step": 3120}, {"loss": 1.0325, "grad_norm": 0.43705281615257263, "learning_rate": 0.0002, "epoch": 2.2955628896222957, "step": 3130}, {"loss": 0.9388, "grad_norm": 0.6454030275344849, "learning_rate": 0.0002, "epoch": 2.3028969563623027, "step": 3140}, {"loss": 0.954, "grad_norm": 0.686030387878418, "learning_rate": 0.0002, "epoch": 2.31023102310231, "step": 3150}, {"loss": 0.9403, "grad_norm": 0.5123633146286011, "learning_rate": 0.0002, "epoch": 2.3175650898423177, "step": 3160}, {"loss": 0.8834, "grad_norm": 0.842506468296051, "learning_rate": 0.0002, "epoch": 2.3248991565823247, "step": 3170}, {"loss": 1.0497, "grad_norm": 0.5193818807601929, "learning_rate": 0.0002, "epoch": 2.332233223322332, "step": 3180}, {"loss": 0.9473, "grad_norm": 0.5634409189224243, "learning_rate": 0.0002, "epoch": 2.3395672900623397, "step": 3190}, {"loss": 0.8499, "grad_norm": 0.6475534439086914, "learning_rate": 0.0002, "epoch": 2.3469013568023467, "step": 3200}, {"loss": 0.874, "grad_norm": 1.1503914594650269, "learning_rate": 0.0002, "epoch": 2.354235423542354, "step": 3210}, {"loss": 0.9762, "grad_norm": 0.7234905362129211, "learning_rate": 0.0002, "epoch": 2.3615694902823616, "step": 3220}, {"loss": 0.9007, "grad_norm": 0.664903461933136, "learning_rate": 0.0002, "epoch": 2.368903557022369, "step": 3230}, {"loss": 0.9987, "grad_norm": 0.5453006625175476, "learning_rate": 0.0002, "epoch": 2.376237623762376, "step": 3240}, {"loss": 0.9742, "grad_norm": 0.6256654262542725, "learning_rate": 0.0002, "epoch": 2.3835716905023836, "step": 3250}, {"loss": 0.9922, "grad_norm": 0.5166565179824829, "learning_rate": 0.0002, "epoch": 2.390905757242391, "step": 3260}, {"loss": 0.927, "grad_norm": 0.5699098110198975, "learning_rate": 0.0002, "epoch": 2.398239823982398, "step": 3270}, {"loss": 0.8878, "grad_norm": 0.4472540020942688, "learning_rate": 0.0002, "epoch": 2.4055738907224056, "step": 3280}, {"loss": 0.9439, "grad_norm": 0.6790403127670288, "learning_rate": 0.0002, "epoch": 2.412907957462413, "step": 3290}, {"loss": 0.972, "grad_norm": 0.5182185173034668, "learning_rate": 0.0002, "epoch": 2.42024202420242, "step": 3300}, {"loss": 0.9775, "grad_norm": 0.564647912979126, "learning_rate": 0.0002, "epoch": 2.4275760909424275, "step": 3310}, {"loss": 1.072, "grad_norm": 0.5625313520431519, "learning_rate": 0.0002, "epoch": 2.434910157682435, "step": 3320}, {"loss": 0.8798, "grad_norm": 0.7496559619903564, "learning_rate": 0.0002, "epoch": 2.442244224422442, "step": 3330}, {"loss": 0.868, "grad_norm": 0.4779128134250641, "learning_rate": 0.0002, "epoch": 2.4495782911624495, "step": 3340}, {"loss": 1.0316, "grad_norm": 0.578093409538269, "learning_rate": 0.0002, "epoch": 2.456912357902457, "step": 3350}, {"loss": 0.9282, "grad_norm": 0.5456080436706543, "learning_rate": 0.0002, "epoch": 2.4642464246424645, "step": 3360}, {"loss": 0.8409, "grad_norm": 0.4769273102283478, "learning_rate": 0.0002, "epoch": 2.4715804913824715, "step": 3370}, {"loss": 0.9312, "grad_norm": 0.5608189702033997, "learning_rate": 0.0002, "epoch": 2.478914558122479, "step": 3380}, {"loss": 0.9934, "grad_norm": 0.5590165853500366, "learning_rate": 0.0002, "epoch": 2.4862486248624864, "step": 3390}, {"loss": 1.025, "grad_norm": 0.801306962966919, "learning_rate": 0.0002, "epoch": 2.4935826916024935, "step": 3400}, {"loss": 0.9049, "grad_norm": 0.6045624613761902, "learning_rate": 0.0002, "epoch": 2.500916758342501, "step": 3410}, {"loss": 0.944, "grad_norm": 0.5735858082771301, "learning_rate": 0.0002, "epoch": 2.5082508250825084, "step": 3420}, {"loss": 0.9846, "grad_norm": 0.6827309131622314, "learning_rate": 0.0002, "epoch": 2.5155848918225154, "step": 3430}, {"loss": 0.9789, "grad_norm": 0.5702602863311768, "learning_rate": 0.0002, "epoch": 2.522918958562523, "step": 3440}, {"loss": 0.9127, "grad_norm": 0.6674721240997314, "learning_rate": 0.0002, "epoch": 2.5302530253025304, "step": 3450}, {"loss": 0.914, "grad_norm": 0.5635907649993896, "learning_rate": 0.0002, "epoch": 2.5375870920425374, "step": 3460}, {"loss": 0.8398, "grad_norm": 0.42737770080566406, "learning_rate": 0.0002, "epoch": 2.544921158782545, "step": 3470}, {"loss": 0.9474, "grad_norm": 0.6720691919326782, "learning_rate": 0.0002, "epoch": 2.5522552255225524, "step": 3480}, {"loss": 0.8637, "grad_norm": 0.8917084336280823, "learning_rate": 0.0002, "epoch": 2.55958929226256, "step": 3490}, {"loss": 0.9257, "grad_norm": 0.5134549140930176, "learning_rate": 0.0002, "epoch": 2.566923359002567, "step": 3500}, {"loss": 0.9362, "grad_norm": 0.4951367974281311, "learning_rate": 0.0002, "epoch": 2.5742574257425743, "step": 3510}, {"loss": 0.9184, "grad_norm": 0.9438204765319824, "learning_rate": 0.0002, "epoch": 2.5815914924825814, "step": 3520}, {"loss": 0.8939, "grad_norm": 0.6024714708328247, "learning_rate": 0.0002, "epoch": 2.588925559222589, "step": 3530}, {"loss": 0.9298, "grad_norm": 0.5248535871505737, "learning_rate": 0.0002, "epoch": 2.5962596259625963, "step": 3540}, {"loss": 0.941, "grad_norm": 0.8677568435668945, "learning_rate": 0.0002, "epoch": 2.6035936927026038, "step": 3550}, {"loss": 0.9253, "grad_norm": 0.82008296251297, "learning_rate": 0.0002, "epoch": 2.610927759442611, "step": 3560}, {"loss": 0.8429, "grad_norm": 0.4724634885787964, "learning_rate": 0.0002, "epoch": 2.6182618261826183, "step": 3570}, {"loss": 0.9058, "grad_norm": 0.5434244275093079, "learning_rate": 0.0002, "epoch": 2.6255958929226257, "step": 3580}, {"loss": 0.9379, "grad_norm": 0.4948740005493164, "learning_rate": 0.0002, "epoch": 2.6329299596626328, "step": 3590}, {"loss": 0.8718, "grad_norm": 0.42109328508377075, "learning_rate": 0.0002, "epoch": 2.6402640264026402, "step": 3600}, {"loss": 0.9809, "grad_norm": 0.7979786396026611, "learning_rate": 0.0002, "epoch": 2.6475980931426477, "step": 3610}, {"loss": 0.9229, "grad_norm": 0.6345919370651245, "learning_rate": 0.0002, "epoch": 2.654932159882655, "step": 3620}, {"loss": 0.8506, "grad_norm": 0.4971671402454376, "learning_rate": 0.0002, "epoch": 2.662266226622662, "step": 3630}, {"loss": 0.8054, "grad_norm": 0.6467748284339905, "learning_rate": 0.0002, "epoch": 2.6696002933626697, "step": 3640}, {"loss": 0.9277, "grad_norm": 0.4240160286426544, "learning_rate": 0.0002, "epoch": 2.6769343601026767, "step": 3650}, {"loss": 0.8213, "grad_norm": 0.5179754495620728, "learning_rate": 0.0002, "epoch": 2.684268426842684, "step": 3660}, {"loss": 0.9221, "grad_norm": 0.754012405872345, "learning_rate": 0.0002, "epoch": 2.6916024935826917, "step": 3670}, {"loss": 0.9194, "grad_norm": 0.5141299962997437, "learning_rate": 0.0002, "epoch": 2.698936560322699, "step": 3680}, {"loss": 0.9495, "grad_norm": 0.5737819075584412, "learning_rate": 0.0002, "epoch": 2.706270627062706, "step": 3690}, {"loss": 1.0162, "grad_norm": 0.5887577533721924, "learning_rate": 0.0002, "epoch": 2.7136046938027136, "step": 3700}, {"loss": 0.9169, "grad_norm": 0.6740471720695496, "learning_rate": 0.0002, "epoch": 2.720938760542721, "step": 3710}, {"loss": 0.9297, "grad_norm": 0.5879453420639038, "learning_rate": 0.0002, "epoch": 2.728272827282728, "step": 3720}, {"loss": 0.9358, "grad_norm": 0.4858354926109314, "learning_rate": 0.0002, "epoch": 2.7356068940227356, "step": 3730}, {"loss": 0.9308, "grad_norm": 0.5489001870155334, "learning_rate": 0.0002, "epoch": 2.742940960762743, "step": 3740}, {"loss": 0.894, "grad_norm": 0.8187092542648315, "learning_rate": 0.0002, "epoch": 2.7502750275027505, "step": 3750}, {"loss": 0.8954, "grad_norm": 0.5666626691818237, "learning_rate": 0.0002, "epoch": 2.7576090942427576, "step": 3760}, {"loss": 1.0059, "grad_norm": 0.5377066135406494, "learning_rate": 0.0002, "epoch": 2.764943160982765, "step": 3770}, {"loss": 0.9132, "grad_norm": 0.566330075263977, "learning_rate": 0.0002, "epoch": 2.772277227722772, "step": 3780}, {"loss": 0.9415, "grad_norm": 0.5522832870483398, "learning_rate": 0.0002, "epoch": 2.7796112944627795, "step": 3790}, {"loss": 0.8816, "grad_norm": 0.5668695569038391, "learning_rate": 0.0002, "epoch": 2.786945361202787, "step": 3800}, {"loss": 0.8885, "grad_norm": 0.7566602826118469, "learning_rate": 0.0002, "epoch": 2.7942794279427945, "step": 3810}, {"loss": 0.8598, "grad_norm": 0.5603684782981873, "learning_rate": 0.0002, "epoch": 2.8016134946828015, "step": 3820}, {"loss": 0.9602, "grad_norm": 0.49122217297554016, "learning_rate": 0.0002, "epoch": 2.808947561422809, "step": 3830}, {"loss": 0.9738, "grad_norm": 0.6798251867294312, "learning_rate": 0.0002, "epoch": 2.816281628162816, "step": 3840}, {"loss": 0.9533, "grad_norm": 0.6097991466522217, "learning_rate": 0.0002, "epoch": 2.8236156949028235, "step": 3850}, {"loss": 0.8672, "grad_norm": 0.6675726175308228, "learning_rate": 0.0002, "epoch": 2.830949761642831, "step": 3860}, {"loss": 0.9324, "grad_norm": 0.9223952889442444, "learning_rate": 0.0002, "epoch": 2.8382838283828384, "step": 3870}, {"loss": 0.8767, "grad_norm": 0.6020799875259399, "learning_rate": 0.0002, "epoch": 2.8456178951228455, "step": 3880}, {"loss": 0.9148, "grad_norm": 0.5206381678581238, "learning_rate": 0.0002, "epoch": 2.852951961862853, "step": 3890}, {"loss": 0.9479, "grad_norm": 0.6268777251243591, "learning_rate": 0.0002, "epoch": 2.8602860286028604, "step": 3900}, {"loss": 0.9409, "grad_norm": 1.1583497524261475, "learning_rate": 0.0002, "epoch": 2.8676200953428674, "step": 3910}, {"loss": 0.895, "grad_norm": 0.7263903021812439, "learning_rate": 0.0002, "epoch": 2.874954162082875, "step": 3920}, {"loss": 0.8786, "grad_norm": 0.5369910001754761, "learning_rate": 0.0002, "epoch": 2.8822882288228824, "step": 3930}, {"loss": 1.0015, "grad_norm": 0.7298350930213928, "learning_rate": 0.0002, "epoch": 2.88962229556289, "step": 3940}, {"loss": 0.979, "grad_norm": 0.577012836933136, "learning_rate": 0.0002, "epoch": 2.896956362302897, "step": 3950}, {"loss": 0.9716, "grad_norm": 0.5859594345092773, "learning_rate": 0.0002, "epoch": 2.9042904290429044, "step": 3960}, {"loss": 0.8772, "grad_norm": 0.47176122665405273, "learning_rate": 0.0002, "epoch": 2.9116244957829114, "step": 3970}, {"loss": 0.8997, "grad_norm": 0.9699620604515076, "learning_rate": 0.0002, "epoch": 2.918958562522919, "step": 3980}, {"loss": 0.9057, "grad_norm": 0.7908747792243958, "learning_rate": 0.0002, "epoch": 2.9262926292629263, "step": 3990}, {"loss": 0.9462, "grad_norm": 0.5777379274368286, "learning_rate": 0.0002, "epoch": 2.933626696002934, "step": 4000}, {"loss": 0.9358, "grad_norm": 0.599288284778595, "learning_rate": 0.0002, "epoch": 2.940960762742941, "step": 4010}, {"loss": 0.9812, "grad_norm": 0.5232274532318115, "learning_rate": 0.0002, "epoch": 2.9482948294829483, "step": 4020}, {"loss": 0.96, "grad_norm": 0.6395137310028076, "learning_rate": 0.0002, "epoch": 2.9556288962229558, "step": 4030}, {"loss": 0.9813, "grad_norm": 0.589260458946228, "learning_rate": 0.0002, "epoch": 2.962962962962963, "step": 4040}, {"loss": 0.9541, "grad_norm": 0.5699581503868103, "learning_rate": 0.0002, "epoch": 2.9702970297029703, "step": 4050}, {"loss": 0.9585, "grad_norm": 0.528468132019043, "learning_rate": 0.0002, "epoch": 2.9776310964429777, "step": 4060}, {"loss": 0.9164, "grad_norm": 0.4804670512676239, "learning_rate": 0.0002, "epoch": 2.984965163182985, "step": 4070}, {"loss": 0.9771, "grad_norm": 1.1918889284133911, "learning_rate": 0.0002, "epoch": 2.9922992299229922, "step": 4080}, {"loss": 0.9178, "grad_norm": 0.5479103326797485, "learning_rate": 0.0002, "epoch": 2.9996332966629997, "step": 4090}, {"eval_loss": 1.1642853021621704, "eval_runtime": 32.7511, "eval_samples_per_second": 13.16, "eval_steps_per_second": 1.649, "epoch": 2.9996332966629997, "step": 4090}, {"loss": 0.7981, "grad_norm": 0.7430027723312378, "learning_rate": 0.0002, "epoch": 3.006967363403007, "step": 4100}, {"loss": 0.7871, "grad_norm": 0.6293647289276123, "learning_rate": 0.0002, "epoch": 3.014301430143014, "step": 4110}, {"loss": 0.78, "grad_norm": 0.6191329956054688, "learning_rate": 0.0002, "epoch": 3.0216354968830217, "step": 4120}, {"loss": 0.7618, "grad_norm": 0.7959313988685608, "learning_rate": 0.0002, "epoch": 3.028969563623029, "step": 4130}, {"loss": 0.8039, "grad_norm": 0.5956351161003113, "learning_rate": 0.0002, "epoch": 3.036303630363036, "step": 4140}, {"loss": 0.7477, "grad_norm": 0.670383632183075, "learning_rate": 0.0002, "epoch": 3.0436376971030437, "step": 4150}, {"loss": 0.7984, "grad_norm": 0.6414518356323242, "learning_rate": 0.0002, "epoch": 3.050971763843051, "step": 4160}, {"loss": 0.7369, "grad_norm": 0.7928852438926697, "learning_rate": 0.0002, "epoch": 3.058305830583058, "step": 4170}, {"loss": 0.7914, "grad_norm": 0.6211121082305908, "learning_rate": 0.0002, "epoch": 3.0656398973230656, "step": 4180}, {"loss": 0.7365, "grad_norm": 0.6237057447433472, "learning_rate": 0.0002, "epoch": 3.072973964063073, "step": 4190}, {"loss": 0.702, "grad_norm": 0.6522233486175537, "learning_rate": 0.0002, "epoch": 3.08030803080308, "step": 4200}, {"loss": 0.7646, "grad_norm": 0.9396848678588867, "learning_rate": 0.0002, "epoch": 3.0876420975430876, "step": 4210}, {"loss": 0.7559, "grad_norm": 0.8003010749816895, "learning_rate": 0.0002, "epoch": 3.094976164283095, "step": 4220}, {"loss": 0.711, "grad_norm": 0.6733810305595398, "learning_rate": 0.0002, "epoch": 3.102310231023102, "step": 4230}, {"loss": 0.696, "grad_norm": 0.6365828514099121, "learning_rate": 0.0002, "epoch": 3.1096442977631096, "step": 4240}, {"loss": 0.8362, "grad_norm": 1.0805548429489136, "learning_rate": 0.0002, "epoch": 3.116978364503117, "step": 4250}, {"loss": 0.7651, "grad_norm": 0.7262141108512878, "learning_rate": 0.0002, "epoch": 3.1243124312431245, "step": 4260}, {"loss": 0.7304, "grad_norm": 0.5500539541244507, "learning_rate": 0.0002, "epoch": 3.1316464979831315, "step": 4270}, {"loss": 0.7721, "grad_norm": 0.793912947177887, "learning_rate": 0.0002, "epoch": 3.138980564723139, "step": 4280}, {"loss": 0.7708, "grad_norm": 1.2540518045425415, "learning_rate": 0.0002, "epoch": 3.1463146314631465, "step": 4290}, {"loss": 0.782, "grad_norm": 0.7020077705383301, "learning_rate": 0.0002, "epoch": 3.1536486982031535, "step": 4300}, {"loss": 0.7253, "grad_norm": 0.5111123323440552, "learning_rate": 0.0002, "epoch": 3.160982764943161, "step": 4310}, {"loss": 0.8159, "grad_norm": 0.7172090411186218, "learning_rate": 0.0002, "epoch": 3.1683168316831685, "step": 4320}, {"loss": 0.6962, "grad_norm": 0.6343168616294861, "learning_rate": 0.0002, "epoch": 3.1756508984231755, "step": 4330}, {"loss": 0.7938, "grad_norm": 0.9563672542572021, "learning_rate": 0.0002, "epoch": 3.182984965163183, "step": 4340}, {"loss": 0.7385, "grad_norm": 1.0225574970245361, "learning_rate": 0.0002, "epoch": 3.1903190319031904, "step": 4350}, {"loss": 0.8652, "grad_norm": 1.1633386611938477, "learning_rate": 0.0002, "epoch": 3.1976530986431975, "step": 4360}, {"loss": 0.7259, "grad_norm": 0.8915148973464966, "learning_rate": 0.0002, "epoch": 3.204987165383205, "step": 4370}, {"loss": 0.8061, "grad_norm": 0.9156812429428101, "learning_rate": 0.0002, "epoch": 3.2123212321232124, "step": 4380}, {"loss": 0.8189, "grad_norm": 0.6363258957862854, "learning_rate": 0.0002, "epoch": 3.21965529886322, "step": 4390}, {"loss": 0.7996, "grad_norm": 0.579099178314209, "learning_rate": 0.0002, "epoch": 3.226989365603227, "step": 4400}, {"loss": 0.8592, "grad_norm": 0.8778146505355835, "learning_rate": 0.0002, "epoch": 3.2343234323432344, "step": 4410}, {"loss": 0.8281, "grad_norm": 0.8356770873069763, "learning_rate": 0.0002, "epoch": 3.241657499083242, "step": 4420}, {"loss": 0.8484, "grad_norm": 0.702032208442688, "learning_rate": 0.0002, "epoch": 3.248991565823249, "step": 4430}, {"loss": 0.7227, "grad_norm": 0.6386539340019226, "learning_rate": 0.0002, "epoch": 3.2563256325632564, "step": 4440}, {"loss": 0.8374, "grad_norm": 0.7008408904075623, "learning_rate": 0.0002, "epoch": 3.263659699303264, "step": 4450}, {"loss": 0.7572, "grad_norm": 0.9556332230567932, "learning_rate": 0.0002, "epoch": 3.270993766043271, "step": 4460}, {"loss": 0.743, "grad_norm": 0.5667835474014282, "learning_rate": 0.0002, "epoch": 3.2783278327832783, "step": 4470}, {"loss": 0.8152, "grad_norm": 0.8239172697067261, "learning_rate": 0.0002, "epoch": 3.285661899523286, "step": 4480}, {"loss": 0.756, "grad_norm": 0.7045050859451294, "learning_rate": 0.0002, "epoch": 3.292995966263293, "step": 4490}, {"loss": 0.7655, "grad_norm": 0.7131434082984924, "learning_rate": 0.0002, "epoch": 3.3003300330033003, "step": 4500}, {"loss": 0.836, "grad_norm": 0.6924910545349121, "learning_rate": 0.0002, "epoch": 3.3076640997433078, "step": 4510}, {"loss": 0.736, "grad_norm": 0.8945356607437134, "learning_rate": 0.0002, "epoch": 3.3149981664833152, "step": 4520}, {"loss": 0.7575, "grad_norm": 0.6546903252601624, "learning_rate": 0.0002, "epoch": 3.3223322332233223, "step": 4530}, {"loss": 0.7893, "grad_norm": 0.8206679224967957, "learning_rate": 0.0002, "epoch": 3.3296662999633297, "step": 4540}, {"loss": 0.7502, "grad_norm": 0.6482203602790833, "learning_rate": 0.0002, "epoch": 3.3370003667033368, "step": 4550}, {"loss": 0.8172, "grad_norm": 0.7558760046958923, "learning_rate": 0.0002, "epoch": 3.3443344334433442, "step": 4560}, {"loss": 0.744, "grad_norm": 0.7794756889343262, "learning_rate": 0.0002, "epoch": 3.3516685001833517, "step": 4570}, {"loss": 0.7385, "grad_norm": 0.7382805943489075, "learning_rate": 0.0002, "epoch": 3.359002566923359, "step": 4580}, {"loss": 0.8511, "grad_norm": 0.5912511944770813, "learning_rate": 0.0002, "epoch": 3.366336633663366, "step": 4590}, {"loss": 0.8272, "grad_norm": 0.7444885969161987, "learning_rate": 0.0002, "epoch": 3.3736707004033737, "step": 4600}, {"loss": 0.7927, "grad_norm": 0.7354922890663147, "learning_rate": 0.0002, "epoch": 3.381004767143381, "step": 4610}, {"loss": 0.7183, "grad_norm": 0.7685934901237488, "learning_rate": 0.0002, "epoch": 3.388338833883388, "step": 4620}, {"loss": 0.7436, "grad_norm": 0.61041259765625, "learning_rate": 0.0002, "epoch": 3.3956729006233957, "step": 4630}, {"loss": 0.7661, "grad_norm": 0.6820451021194458, "learning_rate": 0.0002, "epoch": 3.403006967363403, "step": 4640}, {"loss": 0.8796, "grad_norm": 0.5819534063339233, "learning_rate": 0.0002, "epoch": 3.41034103410341, "step": 4650}, {"loss": 0.7314, "grad_norm": 0.705410897731781, "learning_rate": 0.0002, "epoch": 3.4176751008434176, "step": 4660}, {"loss": 0.7901, "grad_norm": 0.8052892088890076, "learning_rate": 0.0002, "epoch": 3.425009167583425, "step": 4670}, {"loss": 0.7298, "grad_norm": 0.7746483087539673, "learning_rate": 0.0002, "epoch": 3.432343234323432, "step": 4680}, {"loss": 0.7976, "grad_norm": 0.7713689804077148, "learning_rate": 0.0002, "epoch": 3.4396773010634396, "step": 4690}, {"loss": 0.7427, "grad_norm": 0.810371994972229, "learning_rate": 0.0002, "epoch": 3.447011367803447, "step": 4700}, {"loss": 0.7594, "grad_norm": 0.7702969312667847, "learning_rate": 0.0002, "epoch": 3.4543454345434546, "step": 4710}, {"loss": 0.7957, "grad_norm": 0.7069268822669983, "learning_rate": 0.0002, "epoch": 3.4616795012834616, "step": 4720}, {"loss": 0.8199, "grad_norm": 0.7640359401702881, "learning_rate": 0.0002, "epoch": 3.469013568023469, "step": 4730}, {"loss": 0.6875, "grad_norm": 0.8661707639694214, "learning_rate": 0.0002, "epoch": 3.4763476347634765, "step": 4740}, {"loss": 0.8528, "grad_norm": 0.9970282912254333, "learning_rate": 0.0002, "epoch": 3.4836817015034836, "step": 4750}, {"loss": 0.8462, "grad_norm": 0.5824355483055115, "learning_rate": 0.0002, "epoch": 3.491015768243491, "step": 4760}, {"loss": 0.851, "grad_norm": 1.3072649240493774, "learning_rate": 0.0002, "epoch": 3.4983498349834985, "step": 4770}, {"loss": 0.9101, "grad_norm": 0.873978316783905, "learning_rate": 0.0002, "epoch": 3.5056839017235055, "step": 4780}, {"loss": 0.7403, "grad_norm": 0.5526657104492188, "learning_rate": 0.0002, "epoch": 3.513017968463513, "step": 4790}, {"loss": 0.7921, "grad_norm": 0.790894627571106, "learning_rate": 0.0002, "epoch": 3.5203520352035205, "step": 4800}, {"loss": 0.831, "grad_norm": 0.8119630217552185, "learning_rate": 0.0002, "epoch": 3.5276861019435275, "step": 4810}, {"loss": 0.7351, "grad_norm": 0.633212149143219, "learning_rate": 0.0002, "epoch": 3.535020168683535, "step": 4820}, {"loss": 0.8505, "grad_norm": 0.703029990196228, "learning_rate": 0.0002, "epoch": 3.5423542354235424, "step": 4830}, {"loss": 0.7204, "grad_norm": 0.7603771686553955, "learning_rate": 0.0002, "epoch": 3.54968830216355, "step": 4840}, {"loss": 0.8868, "grad_norm": 0.6260480880737305, "learning_rate": 0.0002, "epoch": 3.557022368903557, "step": 4850}, {"loss": 0.8137, "grad_norm": 0.8203664422035217, "learning_rate": 0.0002, "epoch": 3.5643564356435644, "step": 4860}, {"loss": 0.8821, "grad_norm": 0.7793813347816467, "learning_rate": 0.0002, "epoch": 3.5716905023835714, "step": 4870}, {"loss": 0.8164, "grad_norm": 0.7667397260665894, "learning_rate": 0.0002, "epoch": 3.579024569123579, "step": 4880}, {"loss": 0.7597, "grad_norm": 0.8198829889297485, "learning_rate": 0.0002, "epoch": 3.5863586358635864, "step": 4890}, {"loss": 0.7027, "grad_norm": 0.7689233422279358, "learning_rate": 0.0002, "epoch": 3.593692702603594, "step": 4900}, {"loss": 0.804, "grad_norm": 0.7870983481407166, "learning_rate": 0.0002, "epoch": 3.601026769343601, "step": 4910}, {"loss": 0.8269, "grad_norm": 0.8133853077888489, "learning_rate": 0.0002, "epoch": 3.6083608360836084, "step": 4920}, {"loss": 0.8515, "grad_norm": 1.308401346206665, "learning_rate": 0.0002, "epoch": 3.615694902823616, "step": 4930}, {"loss": 0.8494, "grad_norm": 0.7131121754646301, "learning_rate": 0.0002, "epoch": 3.623028969563623, "step": 4940}, {"loss": 0.7235, "grad_norm": 0.6825910210609436, "learning_rate": 0.0002, "epoch": 3.6303630363036303, "step": 4950}, {"loss": 0.7824, "grad_norm": 0.7254678606987, "learning_rate": 0.0002, "epoch": 3.637697103043638, "step": 4960}, {"loss": 0.7983, "grad_norm": 0.8045085072517395, "learning_rate": 0.0002, "epoch": 3.6450311697836453, "step": 4970}, {"loss": 0.8223, "grad_norm": 0.6991777420043945, "learning_rate": 0.0002, "epoch": 3.6523652365236523, "step": 4980}, {"loss": 0.7806, "grad_norm": 0.7804713249206543, "learning_rate": 0.0002, "epoch": 3.6596993032636598, "step": 4990}, {"loss": 0.8402, "grad_norm": 0.8525708317756653, "learning_rate": 0.0002, "epoch": 3.667033370003667, "step": 5000}, {"loss": 0.8496, "grad_norm": 0.7959994673728943, "learning_rate": 0.0002, "epoch": 3.6743674367436743, "step": 5010}, {"loss": 0.8022, "grad_norm": 0.8103628158569336, "learning_rate": 0.0002, "epoch": 3.6817015034836817, "step": 5020}, {"loss": 0.7376, "grad_norm": 0.7517836093902588, "learning_rate": 0.0002, "epoch": 3.689035570223689, "step": 5030}, {"loss": 0.8375, "grad_norm": 0.6878514289855957, "learning_rate": 0.0002, "epoch": 3.6963696369636962, "step": 5040}, {"loss": 0.7998, "grad_norm": 1.2371820211410522, "learning_rate": 0.0002, "epoch": 3.7037037037037037, "step": 5050}, {"loss": 0.6941, "grad_norm": 0.6567103862762451, "learning_rate": 0.0002, "epoch": 3.711037770443711, "step": 5060}, {"loss": 0.8465, "grad_norm": 1.1254922151565552, "learning_rate": 0.0002, "epoch": 3.718371837183718, "step": 5070}, {"loss": 0.8365, "grad_norm": 0.6796132326126099, "learning_rate": 0.0002, "epoch": 3.7257059039237257, "step": 5080}, {"loss": 0.7818, "grad_norm": 0.7285300493240356, "learning_rate": 0.0002, "epoch": 3.733039970663733, "step": 5090}, {"loss": 0.8581, "grad_norm": 0.8931500911712646, "learning_rate": 0.0002, "epoch": 3.7403740374037406, "step": 5100}, {"loss": 0.8181, "grad_norm": 0.6256856918334961, "learning_rate": 0.0002, "epoch": 3.7477081041437477, "step": 5110}, {"loss": 0.743, "grad_norm": 0.79310142993927, "learning_rate": 0.0002, "epoch": 3.755042170883755, "step": 5120}, {"loss": 0.8235, "grad_norm": 0.6594041585922241, "learning_rate": 0.0002, "epoch": 3.762376237623762, "step": 5130}, {"loss": 0.6925, "grad_norm": 0.7029327750205994, "learning_rate": 0.0002, "epoch": 3.7697103043637696, "step": 5140}, {"loss": 0.7457, "grad_norm": 0.5880070328712463, "learning_rate": 0.0002, "epoch": 3.777044371103777, "step": 5150}, {"loss": 0.8716, "grad_norm": 0.7578945159912109, "learning_rate": 0.0002, "epoch": 3.7843784378437846, "step": 5160}, {"loss": 0.8819, "grad_norm": 0.8276378512382507, "learning_rate": 0.0002, "epoch": 3.7917125045837916, "step": 5170}, {"loss": 0.7559, "grad_norm": 0.7627953886985779, "learning_rate": 0.0002, "epoch": 3.799046571323799, "step": 5180}, {"loss": 0.7665, "grad_norm": 0.8169086575508118, "learning_rate": 0.0002, "epoch": 3.806380638063806, "step": 5190}, {"loss": 0.761, "grad_norm": 0.6605030298233032, "learning_rate": 0.0002, "epoch": 3.8137147048038136, "step": 5200}, {"loss": 0.8804, "grad_norm": 0.5837286114692688, "learning_rate": 0.0002, "epoch": 3.821048771543821, "step": 5210}, {"loss": 0.8369, "grad_norm": 1.2422157526016235, "learning_rate": 0.0002, "epoch": 3.8283828382838285, "step": 5220}, {"loss": 0.8431, "grad_norm": 0.6589220762252808, "learning_rate": 0.0002, "epoch": 3.8357169050238356, "step": 5230}, {"loss": 0.7686, "grad_norm": 0.8567556142807007, "learning_rate": 0.0002, "epoch": 3.843050971763843, "step": 5240}, {"loss": 0.8652, "grad_norm": 0.6490627527236938, "learning_rate": 0.0002, "epoch": 3.8503850385038505, "step": 5250}, {"loss": 0.7386, "grad_norm": 0.620232880115509, "learning_rate": 0.0002, "epoch": 3.8577191052438575, "step": 5260}, {"loss": 0.9192, "grad_norm": 0.7685128450393677, "learning_rate": 0.0002, "epoch": 3.865053171983865, "step": 5270}, {"loss": 0.872, "grad_norm": 0.8113296627998352, "learning_rate": 0.0002, "epoch": 3.8723872387238725, "step": 5280}, {"loss": 0.7156, "grad_norm": 0.8092675805091858, "learning_rate": 0.0002, "epoch": 3.87972130546388, "step": 5290}, {"loss": 0.7325, "grad_norm": 0.583570122718811, "learning_rate": 0.0002, "epoch": 3.887055372203887, "step": 5300}, {"loss": 0.9333, "grad_norm": 1.712363600730896, "learning_rate": 0.0002, "epoch": 3.8943894389438944, "step": 5310}, {"loss": 0.7537, "grad_norm": 0.6673534512519836, "learning_rate": 0.0002, "epoch": 3.9017235056839015, "step": 5320}, {"loss": 0.7035, "grad_norm": 1.9770312309265137, "learning_rate": 0.0002, "epoch": 3.909057572423909, "step": 5330}, {"loss": 0.8793, "grad_norm": 0.6430999636650085, "learning_rate": 0.0002, "epoch": 3.9163916391639164, "step": 5340}, {"loss": 0.839, "grad_norm": 1.0159571170806885, "learning_rate": 0.0002, "epoch": 3.923725705903924, "step": 5350}, {"loss": 0.9332, "grad_norm": 0.8607584834098816, "learning_rate": 0.0002, "epoch": 3.931059772643931, "step": 5360}, {"loss": 0.7261, "grad_norm": 0.6967900991439819, "learning_rate": 0.0002, "epoch": 3.9383938393839384, "step": 5370}, {"loss": 0.8456, "grad_norm": 0.7683077454566956, "learning_rate": 0.0002, "epoch": 3.945727906123946, "step": 5380}, {"loss": 0.7682, "grad_norm": 0.6805762648582458, "learning_rate": 0.0002, "epoch": 3.953061972863953, "step": 5390}, {"loss": 0.7746, "grad_norm": 0.7033619284629822, "learning_rate": 0.0002, "epoch": 3.9603960396039604, "step": 5400}, {"loss": 0.8393, "grad_norm": 0.966112494468689, "learning_rate": 0.0002, "epoch": 3.967730106343968, "step": 5410}, {"loss": 0.8316, "grad_norm": 0.8467881083488464, "learning_rate": 0.0002, "epoch": 3.9750641730839753, "step": 5420}, {"loss": 0.8084, "grad_norm": 0.8005317449569702, "learning_rate": 0.0002, "epoch": 3.9823982398239823, "step": 5430}, {"loss": 0.7168, "grad_norm": 1.1615241765975952, "learning_rate": 0.0002, "epoch": 3.98973230656399, "step": 5440}, {"loss": 0.8263, "grad_norm": 0.6121614575386047, "learning_rate": 0.0002, "epoch": 3.997066373303997, "step": 5450}, {"eval_loss": 1.1834222078323364, "eval_runtime": 32.7569, "eval_samples_per_second": 13.158, "eval_steps_per_second": 1.649, "epoch": 4.0, "step": 5454}, {"loss": 0.7267, "grad_norm": 0.6055727005004883, "learning_rate": 0.0002, "epoch": 4.004400440044004, "step": 5460}, {"loss": 0.5766, "grad_norm": 0.8232647180557251, "learning_rate": 0.0002, "epoch": 4.011734506784012, "step": 5470}, {"loss": 0.6489, "grad_norm": 0.7739192247390747, "learning_rate": 0.0002, "epoch": 4.019068573524019, "step": 5480}, {"loss": 0.5978, "grad_norm": 0.6264950633049011, "learning_rate": 0.0002, "epoch": 4.026402640264027, "step": 5490}, {"loss": 0.6392, "grad_norm": 1.4798702001571655, "learning_rate": 0.0002, "epoch": 4.033736707004033, "step": 5500}, {"loss": 0.6143, "grad_norm": 0.9538470506668091, "learning_rate": 0.0002, "epoch": 4.041070773744041, "step": 5510}, {"loss": 0.6056, "grad_norm": 0.834561288356781, "learning_rate": 0.0002, "epoch": 4.048404840484048, "step": 5520}, {"loss": 0.6077, "grad_norm": 0.6407850384712219, "learning_rate": 0.0002, "epoch": 4.055738907224056, "step": 5530}, {"loss": 0.6733, "grad_norm": 0.9035961627960205, "learning_rate": 0.0002, "epoch": 4.063072973964063, "step": 5540}, {"loss": 0.5854, "grad_norm": 0.842812716960907, "learning_rate": 0.0002, "epoch": 4.070407040704071, "step": 5550}, {"loss": 0.654, "grad_norm": 0.8197882175445557, "learning_rate": 0.0002, "epoch": 4.077741107444078, "step": 5560}, {"loss": 0.5919, "grad_norm": 0.8652673959732056, "learning_rate": 0.0002, "epoch": 4.085075174184085, "step": 5570}, {"loss": 0.6188, "grad_norm": 0.8048318028450012, "learning_rate": 0.0002, "epoch": 4.092409240924092, "step": 5580}, {"loss": 0.6487, "grad_norm": 0.9604969024658203, "learning_rate": 0.0002, "epoch": 4.0997433076641, "step": 5590}, {"loss": 0.6356, "grad_norm": 1.244756817817688, "learning_rate": 0.0002, "epoch": 4.107077374404107, "step": 5600}, {"loss": 0.6489, "grad_norm": 0.7975269556045532, "learning_rate": 0.0002, "epoch": 4.114411441144115, "step": 5610}, {"loss": 0.6445, "grad_norm": 0.6130099296569824, "learning_rate": 0.0002, "epoch": 4.121745507884122, "step": 5620}, {"loss": 0.6024, "grad_norm": 0.7793202996253967, "learning_rate": 0.0002, "epoch": 4.129079574624129, "step": 5630}, {"loss": 0.5723, "grad_norm": 1.187238335609436, "learning_rate": 0.0002, "epoch": 4.136413641364136, "step": 5640}, {"loss": 0.6385, "grad_norm": 0.8450375199317932, "learning_rate": 0.0002, "epoch": 4.143747708104144, "step": 5650}, {"loss": 0.6866, "grad_norm": 0.9006940126419067, "learning_rate": 0.0002, "epoch": 4.151081774844151, "step": 5660}, {"loss": 0.6179, "grad_norm": 0.9447154998779297, "learning_rate": 0.0002, "epoch": 4.158415841584159, "step": 5670}, {"loss": 0.6476, "grad_norm": 0.798032283782959, "learning_rate": 0.0002, "epoch": 4.165749908324166, "step": 5680}, {"loss": 0.6666, "grad_norm": 0.65578693151474, "learning_rate": 0.0002, "epoch": 4.1730839750641735, "step": 5690}, {"loss": 0.701, "grad_norm": 1.0864700078964233, "learning_rate": 0.0002, "epoch": 4.18041804180418, "step": 5700}, {"loss": 0.6895, "grad_norm": 0.7344121932983398, "learning_rate": 0.0002, "epoch": 4.187752108544188, "step": 5710}, {"loss": 0.6659, "grad_norm": 0.9722456932067871, "learning_rate": 0.0002, "epoch": 4.195086175284195, "step": 5720}, {"loss": 0.6887, "grad_norm": 1.263814926147461, "learning_rate": 0.0002, "epoch": 4.2024202420242025, "step": 5730}, {"loss": 0.608, "grad_norm": 0.9622581005096436, "learning_rate": 0.0002, "epoch": 4.20975430876421, "step": 5740}, {"loss": 0.6221, "grad_norm": 0.8497143387794495, "learning_rate": 0.0002, "epoch": 4.2170883755042174, "step": 5750}, {"loss": 0.6322, "grad_norm": 0.8248446583747864, "learning_rate": 0.0002, "epoch": 4.224422442244224, "step": 5760}, {"loss": 0.6045, "grad_norm": 1.2544798851013184, "learning_rate": 0.0002, "epoch": 4.2317565089842315, "step": 5770}, {"loss": 0.641, "grad_norm": 0.8224676251411438, "learning_rate": 0.0002, "epoch": 4.239090575724239, "step": 5780}, {"loss": 0.6399, "grad_norm": 0.8924877047538757, "learning_rate": 0.0002, "epoch": 4.2464246424642464, "step": 5790}, {"loss": 0.6845, "grad_norm": 0.8545848727226257, "learning_rate": 0.0002, "epoch": 4.253758709204254, "step": 5800}, {"loss": 0.6669, "grad_norm": 0.8081067800521851, "learning_rate": 0.0002, "epoch": 4.261092775944261, "step": 5810}, {"loss": 0.6149, "grad_norm": 0.7111002802848816, "learning_rate": 0.0002, "epoch": 4.268426842684269, "step": 5820}, {"loss": 0.6343, "grad_norm": 0.8696979880332947, "learning_rate": 0.0002, "epoch": 4.2757609094242754, "step": 5830}, {"loss": 0.6384, "grad_norm": 0.821401834487915, "learning_rate": 0.0002, "epoch": 4.283094976164283, "step": 5840}, {"loss": 0.6912, "grad_norm": 0.888908326625824, "learning_rate": 0.0002, "epoch": 4.29042904290429, "step": 5850}, {"loss": 0.6061, "grad_norm": 1.9380123615264893, "learning_rate": 0.0002, "epoch": 4.297763109644298, "step": 5860}, {"loss": 0.6766, "grad_norm": 1.121774435043335, "learning_rate": 0.0002, "epoch": 4.305097176384305, "step": 5870}, {"loss": 0.7205, "grad_norm": 0.9238282442092896, "learning_rate": 0.0002, "epoch": 4.312431243124313, "step": 5880}, {"loss": 0.6351, "grad_norm": 0.7321620583534241, "learning_rate": 0.0002, "epoch": 4.319765309864319, "step": 5890}, {"loss": 0.6404, "grad_norm": 0.8739548325538635, "learning_rate": 0.0002, "epoch": 4.327099376604327, "step": 5900}, {"loss": 0.5892, "grad_norm": 0.9686012268066406, "learning_rate": 0.0002, "epoch": 4.334433443344334, "step": 5910}, {"loss": 0.641, "grad_norm": 0.9033839106559753, "learning_rate": 0.0002, "epoch": 4.341767510084342, "step": 5920}, {"loss": 0.6456, "grad_norm": 0.8131115436553955, "learning_rate": 0.0002, "epoch": 4.349101576824349, "step": 5930}, {"loss": 0.5826, "grad_norm": 0.8942412734031677, "learning_rate": 0.0002, "epoch": 4.356435643564357, "step": 5940}, {"loss": 0.7336, "grad_norm": 0.8439112901687622, "learning_rate": 0.0002, "epoch": 4.363769710304364, "step": 5950}, {"loss": 0.6537, "grad_norm": 0.9176713228225708, "learning_rate": 0.0002, "epoch": 4.371103777044371, "step": 5960}, {"loss": 0.6792, "grad_norm": 0.6799634695053101, "learning_rate": 0.0002, "epoch": 4.378437843784378, "step": 5970}, {"loss": 0.7266, "grad_norm": 1.0435824394226074, "learning_rate": 0.0002, "epoch": 4.385771910524386, "step": 5980}, {"loss": 0.68, "grad_norm": 0.997937798500061, "learning_rate": 0.0002, "epoch": 4.393105977264393, "step": 5990}, {"loss": 0.6604, "grad_norm": 1.0308842658996582, "learning_rate": 0.0002, "epoch": 4.400440044004401, "step": 6000}, {"loss": 0.6402, "grad_norm": 1.3683775663375854, "learning_rate": 0.0002, "epoch": 4.407774110744408, "step": 6010}, {"loss": 0.7027, "grad_norm": 0.7569534182548523, "learning_rate": 0.0002, "epoch": 4.415108177484415, "step": 6020}, {"loss": 0.5949, "grad_norm": 1.089978575706482, "learning_rate": 0.0002, "epoch": 4.422442244224422, "step": 6030}, {"loss": 0.6353, "grad_norm": 0.7522459626197815, "learning_rate": 0.0002, "epoch": 4.42977631096443, "step": 6040}, {"loss": 0.5852, "grad_norm": 0.6709823608398438, "learning_rate": 0.0002, "epoch": 4.437110377704437, "step": 6050}, {"loss": 0.6718, "grad_norm": 0.6992089748382568, "learning_rate": 0.0002, "epoch": 4.444444444444445, "step": 6060}, {"loss": 0.6933, "grad_norm": 1.0182931423187256, "learning_rate": 0.0002, "epoch": 4.451778511184452, "step": 6070}, {"loss": 0.6255, "grad_norm": 1.0685160160064697, "learning_rate": 0.0002, "epoch": 4.459112577924459, "step": 6080}, {"loss": 0.6086, "grad_norm": 0.8295124769210815, "learning_rate": 0.0002, "epoch": 4.466446644664466, "step": 6090}, {"loss": 0.6359, "grad_norm": 1.1862998008728027, "learning_rate": 0.0002, "epoch": 4.473780711404474, "step": 6100}, {"loss": 0.638, "grad_norm": 0.7400273084640503, "learning_rate": 0.0002, "epoch": 4.481114778144481, "step": 6110}, {"loss": 0.6854, "grad_norm": 0.7098417282104492, "learning_rate": 0.0002, "epoch": 4.488448844884489, "step": 6120}, {"loss": 0.6976, "grad_norm": 0.9745053648948669, "learning_rate": 0.0002, "epoch": 4.495782911624496, "step": 6130}, {"loss": 0.605, "grad_norm": 0.8638797998428345, "learning_rate": 0.0002, "epoch": 4.503116978364503, "step": 6140}, {"loss": 0.6491, "grad_norm": 0.8291046619415283, "learning_rate": 0.0002, "epoch": 4.51045104510451, "step": 6150}, {"loss": 0.6457, "grad_norm": 1.0301737785339355, "learning_rate": 0.0002, "epoch": 4.517785111844518, "step": 6160}, {"loss": 0.6742, "grad_norm": 1.1996512413024902, "learning_rate": 0.0002, "epoch": 4.525119178584525, "step": 6170}, {"loss": 0.6484, "grad_norm": 1.151038408279419, "learning_rate": 0.0002, "epoch": 4.5324532453245325, "step": 6180}, {"loss": 0.668, "grad_norm": 0.8385201096534729, "learning_rate": 0.0002, "epoch": 4.53978731206454, "step": 6190}, {"loss": 0.6381, "grad_norm": 0.8969188332557678, "learning_rate": 0.0002, "epoch": 4.5471213788045475, "step": 6200}, {"loss": 0.7141, "grad_norm": 1.60659658908844, "learning_rate": 0.0002, "epoch": 4.554455445544555, "step": 6210}, {"loss": 0.6388, "grad_norm": 0.9356731176376343, "learning_rate": 0.0002, "epoch": 4.5617895122845615, "step": 6220}, {"loss": 0.7393, "grad_norm": 0.95856773853302, "learning_rate": 0.0002, "epoch": 4.569123579024569, "step": 6230}, {"loss": 0.6554, "grad_norm": 1.1162524223327637, "learning_rate": 0.0002, "epoch": 4.5764576457645765, "step": 6240}, {"loss": 0.6012, "grad_norm": 0.8809238076210022, "learning_rate": 0.0002, "epoch": 4.583791712504584, "step": 6250}, {"loss": 0.648, "grad_norm": 0.890738844871521, "learning_rate": 0.0002, "epoch": 4.591125779244591, "step": 6260}, {"loss": 0.6663, "grad_norm": 0.918684720993042, "learning_rate": 0.0002, "epoch": 4.598459845984598, "step": 6270}, {"loss": 0.5992, "grad_norm": 0.8156296610832214, "learning_rate": 0.0002, "epoch": 4.6057939127246055, "step": 6280}, {"loss": 0.723, "grad_norm": 1.046634316444397, "learning_rate": 0.0002, "epoch": 4.613127979464613, "step": 6290}, {"loss": 0.7023, "grad_norm": 0.7725525498390198, "learning_rate": 0.0002, "epoch": 4.62046204620462, "step": 6300}, {"loss": 0.6414, "grad_norm": 0.9992046356201172, "learning_rate": 0.0002, "epoch": 4.627796112944628, "step": 6310}, {"loss": 0.6201, "grad_norm": 0.8480095267295837, "learning_rate": 0.0002, "epoch": 4.635130179684635, "step": 6320}, {"loss": 0.6869, "grad_norm": 0.7061955332756042, "learning_rate": 0.0002, "epoch": 4.642464246424643, "step": 6330}, {"loss": 0.6828, "grad_norm": 1.0354212522506714, "learning_rate": 0.0002, "epoch": 4.649798313164649, "step": 6340}, {"loss": 0.6651, "grad_norm": 1.0081377029418945, "learning_rate": 0.0002, "epoch": 4.657132379904657, "step": 6350}, {"loss": 0.726, "grad_norm": 1.2904249429702759, "learning_rate": 0.0002, "epoch": 4.664466446644664, "step": 6360}, {"loss": 0.7148, "grad_norm": 0.9248910546302795, "learning_rate": 0.0002, "epoch": 4.671800513384672, "step": 6370}, {"loss": 0.6961, "grad_norm": 0.9907804131507874, "learning_rate": 0.0002, "epoch": 4.679134580124679, "step": 6380}, {"loss": 0.6163, "grad_norm": 1.201143741607666, "learning_rate": 0.0002, "epoch": 4.686468646864687, "step": 6390}, {"loss": 0.6762, "grad_norm": 0.8709394335746765, "learning_rate": 0.0002, "epoch": 4.693802713604693, "step": 6400}, {"loss": 0.7217, "grad_norm": 0.7468608021736145, "learning_rate": 0.0002, "epoch": 4.701136780344701, "step": 6410}, {"loss": 0.6548, "grad_norm": 0.8607903718948364, "learning_rate": 0.0002, "epoch": 4.708470847084708, "step": 6420}, {"loss": 0.6449, "grad_norm": 0.9840512871742249, "learning_rate": 0.0002, "epoch": 4.715804913824716, "step": 6430}, {"loss": 0.685, "grad_norm": 0.8328204154968262, "learning_rate": 0.0002, "epoch": 4.723138980564723, "step": 6440}, {"loss": 0.697, "grad_norm": 0.924505352973938, "learning_rate": 0.0002, "epoch": 4.730473047304731, "step": 6450}, {"loss": 0.7422, "grad_norm": 0.8897685408592224, "learning_rate": 0.0002, "epoch": 4.737807114044738, "step": 6460}, {"loss": 0.6842, "grad_norm": 0.9605024456977844, "learning_rate": 0.0002, "epoch": 4.745141180784745, "step": 6470}, {"loss": 0.6488, "grad_norm": 0.8150759935379028, "learning_rate": 0.0002, "epoch": 4.752475247524752, "step": 6480}, {"loss": 0.6606, "grad_norm": 0.8128412961959839, "learning_rate": 0.0002, "epoch": 4.75980931426476, "step": 6490}, {"loss": 0.6729, "grad_norm": 0.7381404638290405, "learning_rate": 0.0002, "epoch": 4.767143381004767, "step": 6500}, {"loss": 0.6713, "grad_norm": 1.0565853118896484, "learning_rate": 0.0002, "epoch": 4.774477447744775, "step": 6510}, {"loss": 0.6496, "grad_norm": 0.9298134446144104, "learning_rate": 0.0002, "epoch": 4.781811514484782, "step": 6520}, {"loss": 0.7279, "grad_norm": 1.0145525932312012, "learning_rate": 0.0002, "epoch": 4.789145581224789, "step": 6530}, {"loss": 0.5986, "grad_norm": 0.92259681224823, "learning_rate": 0.0002, "epoch": 4.796479647964796, "step": 6540}, {"loss": 0.63, "grad_norm": 0.7881024479866028, "learning_rate": 0.0002, "epoch": 4.803813714704804, "step": 6550}, {"loss": 0.7134, "grad_norm": 1.4935206174850464, "learning_rate": 0.0002, "epoch": 4.811147781444811, "step": 6560}, {"loss": 0.6695, "grad_norm": 0.8612369298934937, "learning_rate": 0.0002, "epoch": 4.818481848184819, "step": 6570}, {"loss": 0.779, "grad_norm": 1.0118653774261475, "learning_rate": 0.0002, "epoch": 4.825815914924826, "step": 6580}, {"loss": 0.6991, "grad_norm": 1.1303809881210327, "learning_rate": 0.0002, "epoch": 4.833149981664834, "step": 6590}, {"loss": 0.7887, "grad_norm": 0.9112492203712463, "learning_rate": 0.0002, "epoch": 4.84048404840484, "step": 6600}, {"loss": 0.7699, "grad_norm": 0.864762544631958, "learning_rate": 0.0002, "epoch": 4.847818115144848, "step": 6610}, {"loss": 0.7347, "grad_norm": 0.9090572595596313, "learning_rate": 0.0002, "epoch": 4.855152181884855, "step": 6620}, {"loss": 0.6608, "grad_norm": 1.014953374862671, "learning_rate": 0.0002, "epoch": 4.862486248624863, "step": 6630}, {"loss": 0.6429, "grad_norm": 1.0702149868011475, "learning_rate": 0.0002, "epoch": 4.86982031536487, "step": 6640}, {"loss": 0.6943, "grad_norm": 1.002135157585144, "learning_rate": 0.0002, "epoch": 4.8771543821048775, "step": 6650}, {"loss": 0.7225, "grad_norm": 0.862545907497406, "learning_rate": 0.0002, "epoch": 4.884488448844884, "step": 6660}, {"loss": 0.6206, "grad_norm": 0.7302131056785583, "learning_rate": 0.0002, "epoch": 4.891822515584892, "step": 6670}, {"loss": 0.7175, "grad_norm": 0.8380730152130127, "learning_rate": 0.0002, "epoch": 4.899156582324899, "step": 6680}, {"loss": 0.645, "grad_norm": 0.7956018447875977, "learning_rate": 0.0002, "epoch": 4.9064906490649065, "step": 6690}, {"loss": 0.6431, "grad_norm": 0.6717583537101746, "learning_rate": 0.0002, "epoch": 4.913824715804914, "step": 6700}, {"loss": 0.6942, "grad_norm": 1.09099280834198, "learning_rate": 0.0002, "epoch": 4.9211587825449215, "step": 6710}, {"loss": 0.7533, "grad_norm": 0.8589889407157898, "learning_rate": 0.0002, "epoch": 4.928492849284929, "step": 6720}, {"loss": 0.66, "grad_norm": 1.0046314001083374, "learning_rate": 0.0002, "epoch": 4.9358269160249355, "step": 6730}, {"loss": 0.6864, "grad_norm": 0.8559659123420715, "learning_rate": 0.0002, "epoch": 4.943160982764943, "step": 6740}, {"loss": 0.6847, "grad_norm": 0.8588525652885437, "learning_rate": 0.0002, "epoch": 4.9504950495049505, "step": 6750}, {"loss": 0.6428, "grad_norm": 0.9192708134651184, "learning_rate": 0.0002, "epoch": 4.957829116244958, "step": 6760}, {"loss": 0.6873, "grad_norm": 1.051398754119873, "learning_rate": 0.0002, "epoch": 4.965163182984965, "step": 6770}, {"loss": 0.7249, "grad_norm": 0.9111362099647522, "learning_rate": 0.0002, "epoch": 4.972497249724973, "step": 6780}, {"loss": 0.7613, "grad_norm": 0.7305638194084167, "learning_rate": 0.0002, "epoch": 4.9798313164649795, "step": 6790}, {"loss": 0.6747, "grad_norm": 1.118837594985962, "learning_rate": 0.0002, "epoch": 4.987165383204987, "step": 6800}, {"loss": 0.6412, "grad_norm": 0.9075239300727844, "learning_rate": 0.0002, "epoch": 4.994499449944994, "step": 6810}, {"eval_loss": 1.2361247539520264, "eval_runtime": 32.7325, "eval_samples_per_second": 13.167, "eval_steps_per_second": 1.65, "epoch": 4.999633296662999, "step": 6817}, {"loss": 0.7091, "grad_norm": 1.0541315078735352, "learning_rate": 0.0002, "epoch": 5.001833516685002, "step": 6820}, {"loss": 0.4882, "grad_norm": 0.9750140905380249, "learning_rate": 0.0002, "epoch": 5.009167583425009, "step": 6830}, {"loss": 0.6022, "grad_norm": 0.931838870048523, "learning_rate": 0.0002, "epoch": 5.016501650165017, "step": 6840}, {"loss": 0.5194, "grad_norm": 1.110278844833374, "learning_rate": 0.0002, "epoch": 5.023835716905023, "step": 6850}, {"loss": 0.4676, "grad_norm": 1.0670180320739746, "learning_rate": 0.0002, "epoch": 5.031169783645031, "step": 6860}, {"loss": 0.4374, "grad_norm": 0.8762092590332031, "learning_rate": 0.0002, "epoch": 5.038503850385038, "step": 6870}, {"loss": 0.505, "grad_norm": 1.1169432401657104, "learning_rate": 0.0002, "epoch": 5.045837917125046, "step": 6880}, {"loss": 0.5114, "grad_norm": 1.005491018295288, "learning_rate": 0.0002, "epoch": 5.053171983865053, "step": 6890}, {"loss": 0.5221, "grad_norm": 1.1751841306686401, "learning_rate": 0.0002, "epoch": 5.060506050605061, "step": 6900}, {"loss": 0.451, "grad_norm": 0.8501367568969727, "learning_rate": 0.0002, "epoch": 5.067840117345068, "step": 6910}, {"loss": 0.5292, "grad_norm": 0.9795131683349609, "learning_rate": 0.0002, "epoch": 5.075174184085075, "step": 6920}, {"loss": 0.5234, "grad_norm": 0.8929879665374756, "learning_rate": 0.0002, "epoch": 5.082508250825082, "step": 6930}, {"loss": 0.5378, "grad_norm": 1.0156651735305786, "learning_rate": 0.0002, "epoch": 5.08984231756509, "step": 6940}, {"loss": 0.5241, "grad_norm": 1.0974335670471191, "learning_rate": 0.0002, "epoch": 5.097176384305097, "step": 6950}, {"loss": 0.5705, "grad_norm": 1.7015666961669922, "learning_rate": 0.0002, "epoch": 5.104510451045105, "step": 6960}, {"loss": 0.523, "grad_norm": 1.0343226194381714, "learning_rate": 0.0002, "epoch": 5.111844517785112, "step": 6970}, {"loss": 0.4616, "grad_norm": 1.3072983026504517, "learning_rate": 0.0002, "epoch": 5.119178584525119, "step": 6980}, {"loss": 0.4813, "grad_norm": 1.038986086845398, "learning_rate": 0.0002, "epoch": 5.126512651265126, "step": 6990}, {"loss": 0.4616, "grad_norm": 0.8638386130332947, "learning_rate": 0.0002, "epoch": 5.133846718005134, "step": 7000}, {"loss": 0.5294, "grad_norm": 0.8326523900032043, "learning_rate": 0.0002, "epoch": 5.141180784745141, "step": 7010}, {"loss": 0.5021, "grad_norm": 1.0976895093917847, "learning_rate": 0.0002, "epoch": 5.148514851485149, "step": 7020}, {"loss": 0.4677, "grad_norm": 1.0077873468399048, "learning_rate": 0.0002, "epoch": 5.155848918225156, "step": 7030}, {"loss": 0.5262, "grad_norm": 1.0662257671356201, "learning_rate": 0.0002, "epoch": 5.163182984965164, "step": 7040}, {"loss": 0.5484, "grad_norm": 1.206271767616272, "learning_rate": 0.0002, "epoch": 5.17051705170517, "step": 7050}, {"loss": 0.4817, "grad_norm": 1.1990262269973755, "learning_rate": 0.0002, "epoch": 5.177851118445178, "step": 7060}, {"loss": 0.6048, "grad_norm": 1.0207163095474243, "learning_rate": 0.0002, "epoch": 5.185185185185185, "step": 7070}, {"loss": 0.4816, "grad_norm": 1.2783987522125244, "learning_rate": 0.0002, "epoch": 5.192519251925193, "step": 7080}, {"loss": 0.5322, "grad_norm": 1.1592512130737305, "learning_rate": 0.0002, "epoch": 5.1998533186652, "step": 7090}, {"loss": 0.5472, "grad_norm": 1.1053160429000854, "learning_rate": 0.0002, "epoch": 5.2071873854052075, "step": 7100}, {"loss": 0.4986, "grad_norm": 1.1925510168075562, "learning_rate": 0.0002, "epoch": 5.214521452145214, "step": 7110}, {"loss": 0.5065, "grad_norm": 1.0714877843856812, "learning_rate": 0.0002, "epoch": 5.221855518885222, "step": 7120}, {"loss": 0.5209, "grad_norm": 0.9451011419296265, "learning_rate": 0.0002, "epoch": 5.229189585625229, "step": 7130}, {"loss": 0.5298, "grad_norm": 1.03838050365448, "learning_rate": 0.0002, "epoch": 5.2365236523652365, "step": 7140}, {"loss": 0.4848, "grad_norm": 0.9204146265983582, "learning_rate": 0.0002, "epoch": 5.243857719105244, "step": 7150}, {"loss": 0.5164, "grad_norm": 1.0142229795455933, "learning_rate": 0.0002, "epoch": 5.2511917858452515, "step": 7160}, {"loss": 0.5092, "grad_norm": 1.4432005882263184, "learning_rate": 0.0002, "epoch": 5.258525852585258, "step": 7170}, {"loss": 0.5133, "grad_norm": 1.1239633560180664, "learning_rate": 0.0002, "epoch": 5.2658599193252655, "step": 7180}, {"loss": 0.4969, "grad_norm": 0.7012821435928345, "learning_rate": 0.0002, "epoch": 5.273193986065273, "step": 7190}, {"loss": 0.5466, "grad_norm": 1.3499128818511963, "learning_rate": 0.0002, "epoch": 5.2805280528052805, "step": 7200}, {"loss": 0.5282, "grad_norm": 0.9498730897903442, "learning_rate": 0.0002, "epoch": 5.287862119545288, "step": 7210}, {"loss": 0.5051, "grad_norm": 0.9552369117736816, "learning_rate": 0.0002, "epoch": 5.295196186285295, "step": 7220}, {"loss": 0.5329, "grad_norm": 0.7610348463058472, "learning_rate": 0.0002, "epoch": 5.302530253025303, "step": 7230}, {"loss": 0.468, "grad_norm": 1.0314512252807617, "learning_rate": 0.0002, "epoch": 5.3098643197653095, "step": 7240}, {"loss": 0.5367, "grad_norm": 1.0534334182739258, "learning_rate": 0.0002, "epoch": 5.317198386505317, "step": 7250}, {"loss": 0.5491, "grad_norm": 1.2553406953811646, "learning_rate": 0.0002, "epoch": 5.324532453245324, "step": 7260}, {"loss": 0.5218, "grad_norm": 0.7061691880226135, "learning_rate": 0.0002, "epoch": 5.331866519985332, "step": 7270}, {"loss": 0.5625, "grad_norm": 0.9652578830718994, "learning_rate": 0.0002, "epoch": 5.339200586725339, "step": 7280}, {"loss": 0.5608, "grad_norm": 1.114788293838501, "learning_rate": 0.0002, "epoch": 5.346534653465347, "step": 7290}, {"loss": 0.578, "grad_norm": 1.0940049886703491, "learning_rate": 0.0002, "epoch": 5.353868720205353, "step": 7300}, {"loss": 0.5256, "grad_norm": 1.0151008367538452, "learning_rate": 0.0002, "epoch": 5.361202786945361, "step": 7310}, {"loss": 0.5377, "grad_norm": 1.0369552373886108, "learning_rate": 0.0002, "epoch": 5.368536853685368, "step": 7320}, {"loss": 0.5028, "grad_norm": 0.8489866256713867, "learning_rate": 0.0002, "epoch": 5.375870920425376, "step": 7330}, {"loss": 0.5937, "grad_norm": 1.1031713485717773, "learning_rate": 0.0002, "epoch": 5.383204987165383, "step": 7340}, {"loss": 0.5355, "grad_norm": 0.9094716310501099, "learning_rate": 0.0002, "epoch": 5.390539053905391, "step": 7350}, {"loss": 0.5406, "grad_norm": 0.9530431032180786, "learning_rate": 0.0002, "epoch": 5.397873120645398, "step": 7360}, {"loss": 0.529, "grad_norm": 0.9633604884147644, "learning_rate": 0.0002, "epoch": 5.405207187385405, "step": 7370}, {"loss": 0.5315, "grad_norm": 0.9541662335395813, "learning_rate": 0.0002, "epoch": 5.412541254125412, "step": 7380}, {"loss": 0.6774, "grad_norm": 1.0459771156311035, "learning_rate": 0.0002, "epoch": 5.41987532086542, "step": 7390}, {"loss": 0.5737, "grad_norm": 1.027388334274292, "learning_rate": 0.0002, "epoch": 5.427209387605427, "step": 7400}, {"loss": 0.556, "grad_norm": 0.7267653346061707, "learning_rate": 0.0002, "epoch": 5.434543454345435, "step": 7410}, {"loss": 0.4581, "grad_norm": 1.020142674446106, "learning_rate": 0.0002, "epoch": 5.441877521085442, "step": 7420}, {"loss": 0.4853, "grad_norm": 1.044754147529602, "learning_rate": 0.0002, "epoch": 5.449211587825449, "step": 7430}, {"loss": 0.5666, "grad_norm": 1.5476195812225342, "learning_rate": 0.0002, "epoch": 5.456545654565456, "step": 7440}, {"loss": 0.5302, "grad_norm": 0.9879506826400757, "learning_rate": 0.0002, "epoch": 5.463879721305464, "step": 7450}, {"loss": 0.591, "grad_norm": 1.2562980651855469, "learning_rate": 0.0002, "epoch": 5.471213788045471, "step": 7460}, {"loss": 0.5188, "grad_norm": 1.3051384687423706, "learning_rate": 0.0002, "epoch": 5.478547854785479, "step": 7470}, {"loss": 0.5658, "grad_norm": 1.0511597394943237, "learning_rate": 0.0002, "epoch": 5.485881921525486, "step": 7480}, {"loss": 0.6327, "grad_norm": 1.0380817651748657, "learning_rate": 0.0002, "epoch": 5.493215988265494, "step": 7490}, {"loss": 0.5356, "grad_norm": 1.170274257659912, "learning_rate": 0.0002, "epoch": 5.5005500550055, "step": 7500}, {"loss": 0.5405, "grad_norm": 1.3356517553329468, "learning_rate": 0.0002, "epoch": 5.507884121745508, "step": 7510}, {"loss": 0.5305, "grad_norm": 1.0727124214172363, "learning_rate": 0.0002, "epoch": 5.515218188485515, "step": 7520}, {"loss": 0.5543, "grad_norm": 1.0110199451446533, "learning_rate": 0.0002, "epoch": 5.522552255225523, "step": 7530}, {"loss": 0.5962, "grad_norm": 1.3086743354797363, "learning_rate": 0.0002, "epoch": 5.52988632196553, "step": 7540}, {"loss": 0.5512, "grad_norm": 1.1904916763305664, "learning_rate": 0.0002, "epoch": 5.537220388705538, "step": 7550}, {"loss": 0.5915, "grad_norm": 0.9466280937194824, "learning_rate": 0.0002, "epoch": 5.544554455445544, "step": 7560}, {"loss": 0.5573, "grad_norm": 1.1237901449203491, "learning_rate": 0.0002, "epoch": 5.551888522185552, "step": 7570}, {"loss": 0.5383, "grad_norm": 0.9590660333633423, "learning_rate": 0.0002, "epoch": 5.559222588925559, "step": 7580}, {"loss": 0.5594, "grad_norm": 1.0890778303146362, "learning_rate": 0.0002, "epoch": 5.566556655665567, "step": 7590}, {"loss": 0.5698, "grad_norm": 0.7206931114196777, "learning_rate": 0.0002, "epoch": 5.573890722405574, "step": 7600}, {"loss": 0.5511, "grad_norm": 1.2884514331817627, "learning_rate": 0.0002, "epoch": 5.5812247891455815, "step": 7610}, {"loss": 0.5279, "grad_norm": 0.7798039317131042, "learning_rate": 0.0002, "epoch": 5.588558855885589, "step": 7620}, {"loss": 0.4847, "grad_norm": 1.166046142578125, "learning_rate": 0.0002, "epoch": 5.595892922625596, "step": 7630}, {"loss": 0.5821, "grad_norm": 1.0150201320648193, "learning_rate": 0.0002, "epoch": 5.603226989365603, "step": 7640}, {"loss": 0.5296, "grad_norm": 1.0449682474136353, "learning_rate": 0.0002, "epoch": 5.6105610561056105, "step": 7650}, {"loss": 0.5431, "grad_norm": 0.9310530424118042, "learning_rate": 0.0002, "epoch": 5.617895122845618, "step": 7660}, {"loss": 0.5234, "grad_norm": 0.9117933511734009, "learning_rate": 0.0002, "epoch": 5.6252291895856255, "step": 7670}, {"loss": 0.5807, "grad_norm": 1.1475164890289307, "learning_rate": 0.0002, "epoch": 5.632563256325633, "step": 7680}, {"loss": 0.5816, "grad_norm": 1.066809058189392, "learning_rate": 0.0002, "epoch": 5.6398973230656395, "step": 7690}, {"loss": 0.551, "grad_norm": 1.2834991216659546, "learning_rate": 0.0002, "epoch": 5.647231389805647, "step": 7700}, {"loss": 0.5914, "grad_norm": 1.2245112657546997, "learning_rate": 0.0002, "epoch": 5.6545654565456545, "step": 7710}, {"loss": 0.5552, "grad_norm": 1.1424106359481812, "learning_rate": 0.0002, "epoch": 5.661899523285662, "step": 7720}, {"loss": 0.559, "grad_norm": 1.0673892498016357, "learning_rate": 0.0002, "epoch": 5.669233590025669, "step": 7730}, {"loss": 0.544, "grad_norm": 1.4312121868133545, "learning_rate": 0.0002, "epoch": 5.676567656765677, "step": 7740}, {"loss": 0.5576, "grad_norm": 0.9976982474327087, "learning_rate": 0.0002, "epoch": 5.683901723505684, "step": 7750}, {"loss": 0.4855, "grad_norm": 0.9464678168296814, "learning_rate": 0.0002, "epoch": 5.691235790245691, "step": 7760}, {"loss": 0.5363, "grad_norm": 1.010995626449585, "learning_rate": 0.0002, "epoch": 5.698569856985698, "step": 7770}, {"loss": 0.5873, "grad_norm": 1.3787750005722046, "learning_rate": 0.0002, "epoch": 5.705903923725706, "step": 7780}, {"loss": 0.6234, "grad_norm": 1.020922303199768, "learning_rate": 0.0002, "epoch": 5.713237990465713, "step": 7790}, {"loss": 0.5337, "grad_norm": 0.9748636484146118, "learning_rate": 0.0002, "epoch": 5.720572057205721, "step": 7800}, {"loss": 0.5507, "grad_norm": 1.3077744245529175, "learning_rate": 0.0002, "epoch": 5.727906123945728, "step": 7810}, {"loss": 0.558, "grad_norm": 1.4770057201385498, "learning_rate": 0.0002, "epoch": 5.735240190685735, "step": 7820}, {"loss": 0.5571, "grad_norm": 1.6349090337753296, "learning_rate": 0.0002, "epoch": 5.742574257425742, "step": 7830}, {"loss": 0.5056, "grad_norm": 0.9818630814552307, "learning_rate": 0.0002, "epoch": 5.74990832416575, "step": 7840}, {"loss": 0.5495, "grad_norm": 0.9659715890884399, "learning_rate": 0.0002, "epoch": 5.757242390905757, "step": 7850}, {"loss": 0.5628, "grad_norm": 0.9269950985908508, "learning_rate": 0.0002, "epoch": 5.764576457645765, "step": 7860}, {"loss": 0.5594, "grad_norm": 1.0099073648452759, "learning_rate": 0.0002, "epoch": 5.771910524385772, "step": 7870}, {"loss": 0.5912, "grad_norm": 0.9123615026473999, "learning_rate": 0.0002, "epoch": 5.77924459112578, "step": 7880}, {"loss": 0.6054, "grad_norm": 1.1542246341705322, "learning_rate": 0.0002, "epoch": 5.786578657865786, "step": 7890}, {"loss": 0.5829, "grad_norm": 1.0792022943496704, "learning_rate": 0.0002, "epoch": 5.793912724605794, "step": 7900}, {"loss": 0.504, "grad_norm": 0.95615553855896, "learning_rate": 0.0002, "epoch": 5.801246791345801, "step": 7910}, {"loss": 0.5918, "grad_norm": 1.2471332550048828, "learning_rate": 0.0002, "epoch": 5.808580858085809, "step": 7920}, {"loss": 0.5719, "grad_norm": 1.0189851522445679, "learning_rate": 0.0002, "epoch": 5.815914924825816, "step": 7930}, {"loss": 0.5958, "grad_norm": 1.3309742212295532, "learning_rate": 0.0002, "epoch": 5.823248991565823, "step": 7940}, {"loss": 0.6255, "grad_norm": 1.2930549383163452, "learning_rate": 0.0002, "epoch": 5.83058305830583, "step": 7950}, {"loss": 0.5301, "grad_norm": 0.8216308951377869, "learning_rate": 0.0002, "epoch": 5.837917125045838, "step": 7960}, {"loss": 0.5397, "grad_norm": 1.1205775737762451, "learning_rate": 0.0002, "epoch": 5.845251191785845, "step": 7970}, {"loss": 0.5903, "grad_norm": 0.851298451423645, "learning_rate": 0.0002, "epoch": 5.852585258525853, "step": 7980}, {"loss": 0.5981, "grad_norm": 0.8797095417976379, "learning_rate": 0.0002, "epoch": 5.85991932526586, "step": 7990}, {"loss": 0.6106, "grad_norm": 1.5784614086151123, "learning_rate": 0.0002, "epoch": 5.867253392005868, "step": 8000}, {"loss": 0.5956, "grad_norm": 1.1531187295913696, "learning_rate": 0.0002, "epoch": 5.874587458745875, "step": 8010}, {"loss": 0.6289, "grad_norm": 1.2469146251678467, "learning_rate": 0.0002, "epoch": 5.881921525485882, "step": 8020}, {"loss": 0.5827, "grad_norm": 1.0784350633621216, "learning_rate": 0.0002, "epoch": 5.889255592225889, "step": 8030}, {"loss": 0.6339, "grad_norm": 1.1311599016189575, "learning_rate": 0.0002, "epoch": 5.896589658965897, "step": 8040}, {"loss": 0.5815, "grad_norm": 0.9654512405395508, "learning_rate": 0.0002, "epoch": 5.903923725705904, "step": 8050}, {"loss": 0.6198, "grad_norm": 1.3288270235061646, "learning_rate": 0.0002, "epoch": 5.9112577924459115, "step": 8060}, {"loss": 0.6515, "grad_norm": 1.12800931930542, "learning_rate": 0.0002, "epoch": 5.918591859185918, "step": 8070}, {"loss": 0.5684, "grad_norm": 0.9449917674064636, "learning_rate": 0.0002, "epoch": 5.925925925925926, "step": 8080}, {"loss": 0.6063, "grad_norm": 1.1532357931137085, "learning_rate": 0.0002, "epoch": 5.933259992665933, "step": 8090}, {"loss": 0.5318, "grad_norm": 1.2211151123046875, "learning_rate": 0.0002, "epoch": 5.9405940594059405, "step": 8100}, {"loss": 0.6512, "grad_norm": 1.3459105491638184, "learning_rate": 0.0002, "epoch": 5.947928126145948, "step": 8110}, {"loss": 0.5952, "grad_norm": 1.251999855041504, "learning_rate": 0.0002, "epoch": 5.9552621928859555, "step": 8120}, {"loss": 0.6203, "grad_norm": 1.5682506561279297, "learning_rate": 0.0002, "epoch": 5.962596259625963, "step": 8130}, {"loss": 0.6253, "grad_norm": 0.926075279712677, "learning_rate": 0.0002, "epoch": 5.9699303263659695, "step": 8140}, {"loss": 0.5545, "grad_norm": 0.9622511863708496, "learning_rate": 0.0002, "epoch": 5.977264393105977, "step": 8150}, {"loss": 0.5518, "grad_norm": 0.9633373618125916, "learning_rate": 0.0002, "epoch": 5.9845984598459845, "step": 8160}, {"loss": 0.5831, "grad_norm": 0.8960476517677307, "learning_rate": 0.0002, "epoch": 5.991932526585992, "step": 8170}, {"loss": 0.5442, "grad_norm": 0.9372805953025818, "learning_rate": 0.0002, "epoch": 5.999266593325999, "step": 8180}, {"eval_loss": 1.3233846426010132, "eval_runtime": 32.7419, "eval_samples_per_second": 13.164, "eval_steps_per_second": 1.649, "epoch": 6.0, "step": 8181}, {"loss": 0.4644, "grad_norm": 1.1900787353515625, "learning_rate": 0.0002, "epoch": 6.006600660066007, "step": 8190}, {"loss": 0.4509, "grad_norm": 1.1448326110839844, "learning_rate": 0.0002, "epoch": 6.013934726806014, "step": 8200}, {"loss": 0.3667, "grad_norm": 1.1848368644714355, "learning_rate": 0.0002, "epoch": 6.021268793546021, "step": 8210}, {"loss": 0.4315, "grad_norm": 1.2315572500228882, "learning_rate": 0.0002, "epoch": 6.028602860286028, "step": 8220}, {"loss": 0.3541, "grad_norm": 1.2214244604110718, "learning_rate": 0.0002, "epoch": 6.035936927026036, "step": 8230}, {"loss": 0.4025, "grad_norm": 0.9455513954162598, "learning_rate": 0.0002, "epoch": 6.043270993766043, "step": 8240}, {"loss": 0.4448, "grad_norm": 0.9574248790740967, "learning_rate": 0.0002, "epoch": 6.050605060506051, "step": 8250}, {"loss": 0.4271, "grad_norm": 1.1022400856018066, "learning_rate": 0.0002, "epoch": 6.057939127246058, "step": 8260}, {"loss": 0.3603, "grad_norm": 0.9555122256278992, "learning_rate": 0.0002, "epoch": 6.065273193986065, "step": 8270}, {"loss": 0.4324, "grad_norm": 1.1956106424331665, "learning_rate": 0.0002, "epoch": 6.072607260726072, "step": 8280}, {"loss": 0.3924, "grad_norm": 1.3110876083374023, "learning_rate": 0.0002, "epoch": 6.07994132746608, "step": 8290}, {"loss": 0.3664, "grad_norm": 1.1293374300003052, "learning_rate": 0.0002, "epoch": 6.087275394206087, "step": 8300}, {"loss": 0.385, "grad_norm": 0.9176164269447327, "learning_rate": 0.0002, "epoch": 6.094609460946095, "step": 8310}, {"loss": 0.4142, "grad_norm": 0.9751231670379639, "learning_rate": 0.0002, "epoch": 6.101943527686102, "step": 8320}, {"loss": 0.4356, "grad_norm": 1.0536044836044312, "learning_rate": 0.0002, "epoch": 6.109277594426109, "step": 8330}, {"loss": 0.409, "grad_norm": 1.289342999458313, "learning_rate": 0.0002, "epoch": 6.116611661166116, "step": 8340}, {"loss": 0.4121, "grad_norm": 1.1773661375045776, "learning_rate": 0.0002, "epoch": 6.123945727906124, "step": 8350}, {"loss": 0.4499, "grad_norm": 1.2450661659240723, "learning_rate": 0.0002, "epoch": 6.131279794646131, "step": 8360}, {"loss": 0.4467, "grad_norm": 1.3965914249420166, "learning_rate": 0.0002, "epoch": 6.138613861386139, "step": 8370}, {"loss": 0.4024, "grad_norm": 1.3530808687210083, "learning_rate": 0.0002, "epoch": 6.145947928126146, "step": 8380}, {"loss": 0.4658, "grad_norm": 1.296276330947876, "learning_rate": 0.0002, "epoch": 6.153281994866154, "step": 8390}, {"loss": 0.5073, "grad_norm": 0.9759053587913513, "learning_rate": 0.0002, "epoch": 6.16061606160616, "step": 8400}, {"loss": 0.4718, "grad_norm": 1.2110707759857178, "learning_rate": 0.0002, "epoch": 6.167950128346168, "step": 8410}, {"loss": 0.4453, "grad_norm": 1.312226414680481, "learning_rate": 0.0002, "epoch": 6.175284195086175, "step": 8420}, {"loss": 0.4183, "grad_norm": 1.1696736812591553, "learning_rate": 0.0002, "epoch": 6.182618261826183, "step": 8430}, {"loss": 0.4546, "grad_norm": 1.260304570198059, "learning_rate": 0.0002, "epoch": 6.18995232856619, "step": 8440}, {"loss": 0.4137, "grad_norm": 1.472961187362671, "learning_rate": 0.0002, "epoch": 6.197286395306198, "step": 8450}, {"loss": 0.42, "grad_norm": 1.3618475198745728, "learning_rate": 0.0002, "epoch": 6.204620462046204, "step": 8460}, {"loss": 0.415, "grad_norm": 1.2544318437576294, "learning_rate": 0.0002, "epoch": 6.211954528786212, "step": 8470}, {"loss": 0.3907, "grad_norm": 1.205898642539978, "learning_rate": 0.0002, "epoch": 6.219288595526219, "step": 8480}, {"loss": 0.4431, "grad_norm": 0.9984724521636963, "learning_rate": 0.0002, "epoch": 6.226622662266227, "step": 8490}, {"loss": 0.4768, "grad_norm": 1.3184109926223755, "learning_rate": 0.0002, "epoch": 6.233956729006234, "step": 8500}, {"loss": 0.3859, "grad_norm": 1.135520100593567, "learning_rate": 0.0002, "epoch": 6.241290795746242, "step": 8510}, {"loss": 0.4159, "grad_norm": 1.4528400897979736, "learning_rate": 0.0002, "epoch": 6.248624862486249, "step": 8520}, {"loss": 0.4347, "grad_norm": 1.1222716569900513, "learning_rate": 0.0002, "epoch": 6.255958929226256, "step": 8530}, {"loss": 0.4581, "grad_norm": 1.7878046035766602, "learning_rate": 0.0002, "epoch": 6.263292995966263, "step": 8540}, {"loss": 0.4298, "grad_norm": 0.9789481163024902, "learning_rate": 0.0002, "epoch": 6.270627062706271, "step": 8550}, {"loss": 0.4316, "grad_norm": 1.151977300643921, "learning_rate": 0.0002, "epoch": 6.277961129446278, "step": 8560}, {"loss": 0.428, "grad_norm": 1.389968752861023, "learning_rate": 0.0002, "epoch": 6.2852951961862855, "step": 8570}, {"loss": 0.3903, "grad_norm": 0.884211003780365, "learning_rate": 0.0002, "epoch": 6.292629262926293, "step": 8580}, {"loss": 0.4611, "grad_norm": 1.3604296445846558, "learning_rate": 0.0002, "epoch": 6.2999633296663, "step": 8590}, {"loss": 0.4183, "grad_norm": 1.1845694780349731, "learning_rate": 0.0002, "epoch": 6.307297396406307, "step": 8600}, {"loss": 0.472, "grad_norm": 1.3231550455093384, "learning_rate": 0.0002, "epoch": 6.3146314631463145, "step": 8610}, {"loss": 0.3922, "grad_norm": 0.9546721577644348, "learning_rate": 0.0002, "epoch": 6.321965529886322, "step": 8620}, {"loss": 0.4395, "grad_norm": 1.2329787015914917, "learning_rate": 0.0002, "epoch": 6.3292995966263295, "step": 8630}, {"loss": 0.4344, "grad_norm": 1.0240199565887451, "learning_rate": 0.0002, "epoch": 6.336633663366337, "step": 8640}, {"loss": 0.4529, "grad_norm": 1.1866962909698486, "learning_rate": 0.0002, "epoch": 6.343967730106344, "step": 8650}, {"loss": 0.4575, "grad_norm": 1.2819687128067017, "learning_rate": 0.0002, "epoch": 6.351301796846351, "step": 8660}, {"loss": 0.455, "grad_norm": 0.9654944539070129, "learning_rate": 0.0002, "epoch": 6.3586358635863585, "step": 8670}, {"loss": 0.4739, "grad_norm": 0.9443874955177307, "learning_rate": 0.0002, "epoch": 6.365969930326366, "step": 8680}, {"loss": 0.435, "grad_norm": 1.2914115190505981, "learning_rate": 0.0002, "epoch": 6.373303997066373, "step": 8690}, {"loss": 0.4392, "grad_norm": 1.4558709859848022, "learning_rate": 0.0002, "epoch": 6.380638063806381, "step": 8700}, {"loss": 0.4398, "grad_norm": 1.3255952596664429, "learning_rate": 0.0002, "epoch": 6.387972130546388, "step": 8710}, {"loss": 0.4451, "grad_norm": 1.348742961883545, "learning_rate": 0.0002, "epoch": 6.395306197286395, "step": 8720}, {"loss": 0.41, "grad_norm": 1.0096025466918945, "learning_rate": 0.0002, "epoch": 6.402640264026402, "step": 8730}, {"loss": 0.4459, "grad_norm": 1.1720590591430664, "learning_rate": 0.0002, "epoch": 6.40997433076641, "step": 8740}, {"loss": 0.5059, "grad_norm": 1.1803077459335327, "learning_rate": 0.0002, "epoch": 6.417308397506417, "step": 8750}, {"loss": 0.4539, "grad_norm": 1.3649998903274536, "learning_rate": 0.0002, "epoch": 6.424642464246425, "step": 8760}, {"loss": 0.4171, "grad_norm": 1.1503992080688477, "learning_rate": 0.0002, "epoch": 6.431976530986432, "step": 8770}, {"loss": 0.488, "grad_norm": 1.1537176370620728, "learning_rate": 0.0002, "epoch": 6.43931059772644, "step": 8780}, {"loss": 0.4167, "grad_norm": 0.9743003845214844, "learning_rate": 0.0002, "epoch": 6.446644664466446, "step": 8790}, {"loss": 0.4813, "grad_norm": 0.9097744822502136, "learning_rate": 0.0002, "epoch": 6.453978731206454, "step": 8800}, {"loss": 0.4809, "grad_norm": 2.0174002647399902, "learning_rate": 0.0002, "epoch": 6.461312797946461, "step": 8810}, {"loss": 0.4879, "grad_norm": 1.0809309482574463, "learning_rate": 0.0002, "epoch": 6.468646864686469, "step": 8820}, {"loss": 0.4235, "grad_norm": 1.100294828414917, "learning_rate": 0.0002, "epoch": 6.475980931426476, "step": 8830}, {"loss": 0.4251, "grad_norm": 1.3707489967346191, "learning_rate": 0.0002, "epoch": 6.483314998166484, "step": 8840}, {"loss": 0.4533, "grad_norm": 1.1304761171340942, "learning_rate": 0.0002, "epoch": 6.49064906490649, "step": 8850}, {"loss": 0.4596, "grad_norm": 1.2171573638916016, "learning_rate": 0.0002, "epoch": 6.497983131646498, "step": 8860}, {"loss": 0.4694, "grad_norm": 1.0452901124954224, "learning_rate": 0.0002, "epoch": 6.505317198386505, "step": 8870}, {"loss": 0.4855, "grad_norm": 1.197298526763916, "learning_rate": 0.0002, "epoch": 6.512651265126513, "step": 8880}, {"loss": 0.4167, "grad_norm": 0.9179880619049072, "learning_rate": 0.0002, "epoch": 6.51998533186652, "step": 8890}, {"loss": 0.445, "grad_norm": 1.415079951286316, "learning_rate": 0.0002, "epoch": 6.527319398606528, "step": 8900}, {"loss": 0.424, "grad_norm": 1.1032487154006958, "learning_rate": 0.0002, "epoch": 6.534653465346535, "step": 8910}, {"loss": 0.4496, "grad_norm": 1.2295007705688477, "learning_rate": 0.0002, "epoch": 6.541987532086542, "step": 8920}, {"loss": 0.4755, "grad_norm": 1.4223219156265259, "learning_rate": 0.0002, "epoch": 6.549321598826549, "step": 8930}, {"loss": 0.4597, "grad_norm": 1.2785786390304565, "learning_rate": 0.0002, "epoch": 6.556655665566557, "step": 8940}, {"loss": 0.4651, "grad_norm": 1.3514775037765503, "learning_rate": 0.0002, "epoch": 6.563989732306564, "step": 8950}, {"loss": 0.4961, "grad_norm": 1.107937216758728, "learning_rate": 0.0002, "epoch": 6.571323799046572, "step": 8960}, {"loss": 0.4954, "grad_norm": 1.2839902639389038, "learning_rate": 0.0002, "epoch": 6.578657865786578, "step": 8970}, {"loss": 0.4207, "grad_norm": 0.9793244004249573, "learning_rate": 0.0002, "epoch": 6.585991932526586, "step": 8980}, {"loss": 0.4989, "grad_norm": 1.3403126001358032, "learning_rate": 0.0002, "epoch": 6.593325999266593, "step": 8990}, {"loss": 0.465, "grad_norm": 1.2612813711166382, "learning_rate": 0.0002, "epoch": 6.600660066006601, "step": 9000}, {"loss": 0.4589, "grad_norm": 1.4347625970840454, "learning_rate": 0.0002, "epoch": 6.607994132746608, "step": 9010}, {"loss": 0.4864, "grad_norm": 1.225921869277954, "learning_rate": 0.0002, "epoch": 6.6153281994866155, "step": 9020}, {"loss": 0.4364, "grad_norm": 1.033644676208496, "learning_rate": 0.0002, "epoch": 6.622662266226623, "step": 9030}, {"loss": 0.4698, "grad_norm": 1.1791894435882568, "learning_rate": 0.0002, "epoch": 6.6299963329666305, "step": 9040}, {"loss": 0.4908, "grad_norm": 1.0968137979507446, "learning_rate": 0.0002, "epoch": 6.637330399706637, "step": 9050}, {"loss": 0.4346, "grad_norm": 1.5639140605926514, "learning_rate": 0.0002, "epoch": 6.6446644664466445, "step": 9060}, {"loss": 0.4627, "grad_norm": 1.4158905744552612, "learning_rate": 0.0002, "epoch": 6.651998533186652, "step": 9070}, {"loss": 0.4619, "grad_norm": 1.2120254039764404, "learning_rate": 0.0002, "epoch": 6.6593325999266595, "step": 9080}, {"loss": 0.4564, "grad_norm": 1.1866531372070312, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 9090}, {"loss": 0.5175, "grad_norm": 1.2704026699066162, "learning_rate": 0.0002, "epoch": 6.6740007334066735, "step": 9100}, {"loss": 0.4859, "grad_norm": 1.1878353357315063, "learning_rate": 0.0002, "epoch": 6.681334800146681, "step": 9110}, {"loss": 0.4657, "grad_norm": 1.193995475769043, "learning_rate": 0.0002, "epoch": 6.6886688668866885, "step": 9120}, {"loss": 0.4939, "grad_norm": 1.2927545309066772, "learning_rate": 0.0002, "epoch": 6.696002933626696, "step": 9130}, {"loss": 0.4157, "grad_norm": 1.0770703554153442, "learning_rate": 0.0002, "epoch": 6.703337000366703, "step": 9140}, {"loss": 0.4571, "grad_norm": 1.2200851440429688, "learning_rate": 0.0002, "epoch": 6.710671067106711, "step": 9150}, {"loss": 0.4605, "grad_norm": 1.293891191482544, "learning_rate": 0.0002, "epoch": 6.718005133846718, "step": 9160}, {"loss": 0.5328, "grad_norm": 1.9376052618026733, "learning_rate": 0.0002, "epoch": 6.725339200586725, "step": 9170}, {"loss": 0.4861, "grad_norm": 1.0353254079818726, "learning_rate": 0.0002, "epoch": 6.732673267326732, "step": 9180}, {"loss": 0.5668, "grad_norm": 1.1274057626724243, "learning_rate": 0.0002, "epoch": 6.74000733406674, "step": 9190}, {"loss": 0.4486, "grad_norm": 1.3344064950942993, "learning_rate": 0.0002, "epoch": 6.747341400806747, "step": 9200}, {"loss": 0.49, "grad_norm": 1.303621768951416, "learning_rate": 0.0002, "epoch": 6.754675467546755, "step": 9210}, {"loss": 0.5059, "grad_norm": 1.2327780723571777, "learning_rate": 0.0002, "epoch": 6.762009534286762, "step": 9220}, {"loss": 0.486, "grad_norm": 1.3513109683990479, "learning_rate": 0.0002, "epoch": 6.769343601026769, "step": 9230}, {"loss": 0.5254, "grad_norm": 1.4762850999832153, "learning_rate": 0.0002, "epoch": 6.776677667766776, "step": 9240}, {"loss": 0.4181, "grad_norm": 1.0967189073562622, "learning_rate": 0.0002, "epoch": 6.784011734506784, "step": 9250}, {"loss": 0.4862, "grad_norm": 0.933936357498169, "learning_rate": 0.0002, "epoch": 6.791345801246791, "step": 9260}, {"loss": 0.4667, "grad_norm": 1.065553903579712, "learning_rate": 0.0002, "epoch": 6.798679867986799, "step": 9270}, {"loss": 0.5164, "grad_norm": 1.2044163942337036, "learning_rate": 0.0002, "epoch": 6.806013934726806, "step": 9280}, {"loss": 0.4648, "grad_norm": 1.404137134552002, "learning_rate": 0.0002, "epoch": 6.813348001466814, "step": 9290}, {"loss": 0.4442, "grad_norm": 1.4005582332611084, "learning_rate": 0.0002, "epoch": 6.82068206820682, "step": 9300}, {"loss": 0.459, "grad_norm": 1.1771104335784912, "learning_rate": 0.0002, "epoch": 6.828016134946828, "step": 9310}, {"loss": 0.5059, "grad_norm": 1.191933035850525, "learning_rate": 0.0002, "epoch": 6.835350201686835, "step": 9320}, {"loss": 0.4733, "grad_norm": 1.3395432233810425, "learning_rate": 0.0002, "epoch": 6.842684268426843, "step": 9330}, {"loss": 0.4882, "grad_norm": 1.4145503044128418, "learning_rate": 0.0002, "epoch": 6.85001833516685, "step": 9340}, {"loss": 0.4872, "grad_norm": 1.1128839254379272, "learning_rate": 0.0002, "epoch": 6.857352401906858, "step": 9350}, {"loss": 0.4909, "grad_norm": 1.0771174430847168, "learning_rate": 0.0002, "epoch": 6.864686468646864, "step": 9360}, {"loss": 0.4739, "grad_norm": 1.1089814901351929, "learning_rate": 0.0002, "epoch": 6.872020535386872, "step": 9370}, {"loss": 0.4854, "grad_norm": 1.078444004058838, "learning_rate": 0.0002, "epoch": 6.879354602126879, "step": 9380}, {"loss": 0.4904, "grad_norm": 1.3676636219024658, "learning_rate": 0.0002, "epoch": 6.886688668866887, "step": 9390}, {"loss": 0.4854, "grad_norm": 0.8973749876022339, "learning_rate": 0.0002, "epoch": 6.894022735606894, "step": 9400}, {"loss": 0.4274, "grad_norm": 1.141552448272705, "learning_rate": 0.0002, "epoch": 6.901356802346902, "step": 9410}, {"loss": 0.4972, "grad_norm": 0.8345359563827515, "learning_rate": 0.0002, "epoch": 6.908690869086909, "step": 9420}, {"loss": 0.5218, "grad_norm": 1.1602197885513306, "learning_rate": 0.0002, "epoch": 6.916024935826916, "step": 9430}, {"loss": 0.4911, "grad_norm": 1.275466799736023, "learning_rate": 0.0002, "epoch": 6.923359002566923, "step": 9440}, {"loss": 0.4904, "grad_norm": 0.9186071157455444, "learning_rate": 0.0002, "epoch": 6.930693069306931, "step": 9450}, {"loss": 0.4604, "grad_norm": 0.9069198966026306, "learning_rate": 0.0002, "epoch": 6.938027136046938, "step": 9460}, {"loss": 0.4363, "grad_norm": 1.2331899404525757, "learning_rate": 0.0002, "epoch": 6.945361202786946, "step": 9470}, {"loss": 0.4815, "grad_norm": 0.8685150742530823, "learning_rate": 0.0002, "epoch": 6.952695269526953, "step": 9480}, {"loss": 0.4424, "grad_norm": 1.4067939519882202, "learning_rate": 0.0002, "epoch": 6.96002933626696, "step": 9490}, {"loss": 0.5089, "grad_norm": 1.1864029169082642, "learning_rate": 0.0002, "epoch": 6.967363403006967, "step": 9500}, {"loss": 0.4906, "grad_norm": 1.3697725534439087, "learning_rate": 0.0002, "epoch": 6.974697469746975, "step": 9510}, {"loss": 0.4797, "grad_norm": 1.1632893085479736, "learning_rate": 0.0002, "epoch": 6.982031536486982, "step": 9520}, {"loss": 0.4526, "grad_norm": 1.1447268724441528, "learning_rate": 0.0002, "epoch": 6.9893656032269895, "step": 9530}, {"loss": 0.4627, "grad_norm": 1.5017213821411133, "learning_rate": 0.0002, "epoch": 6.996699669966997, "step": 9540}, {"eval_loss": 1.4178194999694824, "eval_runtime": 32.7488, "eval_samples_per_second": 13.161, "eval_steps_per_second": 1.649, "epoch": 6.999633296662999, "step": 9544}, {"loss": 0.4396, "grad_norm": 1.110981822013855, "learning_rate": 0.0002, "epoch": 7.0040337367070045, "step": 9550}, {"loss": 0.3475, "grad_norm": 1.2793253660202026, "learning_rate": 0.0002, "epoch": 7.011367803447011, "step": 9560}, {"loss": 0.4022, "grad_norm": 1.1258823871612549, "learning_rate": 0.0002, "epoch": 7.0187018701870185, "step": 9570}, {"loss": 0.3008, "grad_norm": 1.410486102104187, "learning_rate": 0.0002, "epoch": 7.026035936927026, "step": 9580}, {"loss": 0.3716, "grad_norm": 1.2088500261306763, "learning_rate": 0.0002, "epoch": 7.0333700036670335, "step": 9590}, {"loss": 0.3379, "grad_norm": 0.8303650617599487, "learning_rate": 0.0002, "epoch": 7.040704070407041, "step": 9600}, {"loss": 0.3376, "grad_norm": 0.9813525080680847, "learning_rate": 0.0002, "epoch": 7.048038137147048, "step": 9610}, {"loss": 0.3202, "grad_norm": 0.9679017066955566, "learning_rate": 0.0002, "epoch": 7.055372203887055, "step": 9620}, {"loss": 0.3287, "grad_norm": 1.1532220840454102, "learning_rate": 0.0002, "epoch": 7.0627062706270625, "step": 9630}, {"loss": 0.3639, "grad_norm": 1.312053918838501, "learning_rate": 0.0002, "epoch": 7.07004033736707, "step": 9640}, {"loss": 0.3278, "grad_norm": 1.0594364404678345, "learning_rate": 0.0002, "epoch": 7.077374404107077, "step": 9650}, {"loss": 0.3259, "grad_norm": 1.545080542564392, "learning_rate": 0.0002, "epoch": 7.084708470847085, "step": 9660}, {"loss": 0.328, "grad_norm": 1.1748381853103638, "learning_rate": 0.0002, "epoch": 7.092042537587092, "step": 9670}, {"loss": 0.3313, "grad_norm": 1.6107453107833862, "learning_rate": 0.0002, "epoch": 7.0993766043271, "step": 9680}, {"loss": 0.3469, "grad_norm": 0.9478244185447693, "learning_rate": 0.0002, "epoch": 7.106710671067106, "step": 9690}, {"loss": 0.3289, "grad_norm": 1.508410930633545, "learning_rate": 0.0002, "epoch": 7.114044737807114, "step": 9700}, {"loss": 0.3077, "grad_norm": 1.3175169229507446, "learning_rate": 0.0002, "epoch": 7.121378804547121, "step": 9710}, {"loss": 0.3241, "grad_norm": 1.2631924152374268, "learning_rate": 0.0002, "epoch": 7.128712871287129, "step": 9720}, {"loss": 0.3806, "grad_norm": 1.0640755891799927, "learning_rate": 0.0002, "epoch": 7.136046938027136, "step": 9730}, {"loss": 0.3418, "grad_norm": 1.247279405593872, "learning_rate": 0.0002, "epoch": 7.143381004767144, "step": 9740}, {"loss": 0.3385, "grad_norm": 1.2538974285125732, "learning_rate": 0.0002, "epoch": 7.15071507150715, "step": 9750}, {"loss": 0.3445, "grad_norm": 1.3157252073287964, "learning_rate": 0.0002, "epoch": 7.158049138247158, "step": 9760}, {"loss": 0.3518, "grad_norm": 1.5254220962524414, "learning_rate": 0.0002, "epoch": 7.165383204987165, "step": 9770}, {"loss": 0.3575, "grad_norm": 1.0063719749450684, "learning_rate": 0.0002, "epoch": 7.172717271727173, "step": 9780}, {"loss": 0.3701, "grad_norm": 0.8030351996421814, "learning_rate": 0.0002, "epoch": 7.18005133846718, "step": 9790}, {"loss": 0.3645, "grad_norm": 1.2086257934570312, "learning_rate": 0.0002, "epoch": 7.187385405207188, "step": 9800}, {"loss": 0.354, "grad_norm": 1.7020413875579834, "learning_rate": 0.0002, "epoch": 7.194719471947194, "step": 9810}, {"loss": 0.3242, "grad_norm": 1.2517976760864258, "learning_rate": 0.0002, "epoch": 7.202053538687202, "step": 9820}, {"loss": 0.3503, "grad_norm": 1.330505132675171, "learning_rate": 0.0002, "epoch": 7.209387605427209, "step": 9830}, {"loss": 0.3349, "grad_norm": 1.1273366212844849, "learning_rate": 0.0002, "epoch": 7.216721672167217, "step": 9840}, {"loss": 0.3562, "grad_norm": 1.3738148212432861, "learning_rate": 0.0002, "epoch": 7.224055738907224, "step": 9850}, {"loss": 0.3622, "grad_norm": 1.2162928581237793, "learning_rate": 0.0002, "epoch": 7.231389805647232, "step": 9860}, {"loss": 0.3913, "grad_norm": 1.743969440460205, "learning_rate": 0.0002, "epoch": 7.238723872387239, "step": 9870}, {"loss": 0.3855, "grad_norm": 1.5357484817504883, "learning_rate": 0.0002, "epoch": 7.246057939127246, "step": 9880}, {"loss": 0.3556, "grad_norm": 1.342976450920105, "learning_rate": 0.0002, "epoch": 7.253392005867253, "step": 9890}, {"loss": 0.3787, "grad_norm": 1.428523302078247, "learning_rate": 0.0002, "epoch": 7.260726072607261, "step": 9900}, {"loss": 0.343, "grad_norm": 1.5631695985794067, "learning_rate": 0.0002, "epoch": 7.268060139347268, "step": 9910}, {"loss": 0.3292, "grad_norm": 1.192564606666565, "learning_rate": 0.0002, "epoch": 7.275394206087276, "step": 9920}, {"loss": 0.324, "grad_norm": 1.1428006887435913, "learning_rate": 0.0002, "epoch": 7.282728272827283, "step": 9930}, {"loss": 0.2868, "grad_norm": 1.1959515810012817, "learning_rate": 0.0002, "epoch": 7.29006233956729, "step": 9940}, {"loss": 0.3623, "grad_norm": 0.984326958656311, "learning_rate": 0.0002, "epoch": 7.297396406307297, "step": 9950}, {"loss": 0.3442, "grad_norm": 1.1258848905563354, "learning_rate": 0.0002, "epoch": 7.304730473047305, "step": 9960}, {"loss": 0.3682, "grad_norm": 1.2445521354675293, "learning_rate": 0.0002, "epoch": 7.312064539787312, "step": 9970}, {"loss": 0.4032, "grad_norm": 1.066351294517517, "learning_rate": 0.0002, "epoch": 7.3193986065273196, "step": 9980}, {"loss": 0.3689, "grad_norm": 1.3111763000488281, "learning_rate": 0.0002, "epoch": 7.326732673267327, "step": 9990}, {"loss": 0.4251, "grad_norm": 1.5113508701324463, "learning_rate": 0.0002, "epoch": 7.334066740007334, "step": 10000}, {"loss": 0.3668, "grad_norm": 1.2499724626541138, "learning_rate": 0.0002, "epoch": 7.341400806747341, "step": 10010}, {"loss": 0.392, "grad_norm": 1.1003477573394775, "learning_rate": 0.0002, "epoch": 7.3487348734873486, "step": 10020}, {"loss": 0.4045, "grad_norm": 1.4911425113677979, "learning_rate": 0.0002, "epoch": 7.356068940227356, "step": 10030}, {"loss": 0.3892, "grad_norm": 1.291712999343872, "learning_rate": 0.0002, "epoch": 7.3634030069673635, "step": 10040}, {"loss": 0.3817, "grad_norm": 1.4001942873001099, "learning_rate": 0.0002, "epoch": 7.370737073707371, "step": 10050}, {"loss": 0.404, "grad_norm": 2.015535593032837, "learning_rate": 0.0002, "epoch": 7.378071140447378, "step": 10060}, {"loss": 0.3758, "grad_norm": 1.3355735540390015, "learning_rate": 0.0002, "epoch": 7.385405207187385, "step": 10070}, {"loss": 0.3764, "grad_norm": 1.1258678436279297, "learning_rate": 0.0002, "epoch": 7.3927392739273925, "step": 10080}, {"loss": 0.3827, "grad_norm": 1.3883707523345947, "learning_rate": 0.0002, "epoch": 7.4000733406674, "step": 10090}, {"loss": 0.3623, "grad_norm": 1.144474744796753, "learning_rate": 0.0002, "epoch": 7.407407407407407, "step": 10100}, {"loss": 0.4024, "grad_norm": 1.636843204498291, "learning_rate": 0.0002, "epoch": 7.414741474147415, "step": 10110}, {"loss": 0.3924, "grad_norm": 1.6167247295379639, "learning_rate": 0.0002, "epoch": 7.422075540887422, "step": 10120}, {"loss": 0.4233, "grad_norm": 1.3800078630447388, "learning_rate": 0.0002, "epoch": 7.429409607627429, "step": 10130}, {"loss": 0.3859, "grad_norm": 1.2631969451904297, "learning_rate": 0.0002, "epoch": 7.436743674367436, "step": 10140}, {"loss": 0.3523, "grad_norm": 1.32834792137146, "learning_rate": 0.0002, "epoch": 7.444077741107444, "step": 10150}, {"loss": 0.3945, "grad_norm": 1.370316982269287, "learning_rate": 0.0002, "epoch": 7.451411807847451, "step": 10160}, {"loss": 0.3695, "grad_norm": 1.6096234321594238, "learning_rate": 0.0002, "epoch": 7.458745874587459, "step": 10170}, {"loss": 0.3378, "grad_norm": 1.3638662099838257, "learning_rate": 0.0002, "epoch": 7.466079941327466, "step": 10180}, {"loss": 0.4015, "grad_norm": 1.3508107662200928, "learning_rate": 0.0002, "epoch": 7.473414008067474, "step": 10190}, {"loss": 0.4169, "grad_norm": 1.5599194765090942, "learning_rate": 0.0002, "epoch": 7.48074807480748, "step": 10200}, {"loss": 0.4071, "grad_norm": 1.4922538995742798, "learning_rate": 0.0002, "epoch": 7.488082141547488, "step": 10210}, {"loss": 0.419, "grad_norm": 1.485437273979187, "learning_rate": 0.0002, "epoch": 7.495416208287495, "step": 10220}, {"loss": 0.382, "grad_norm": 0.9040785431861877, "learning_rate": 0.0002, "epoch": 7.502750275027503, "step": 10230}, {"loss": 0.3437, "grad_norm": 1.2453011274337769, "learning_rate": 0.0002, "epoch": 7.51008434176751, "step": 10240}, {"loss": 0.4063, "grad_norm": 1.4167460203170776, "learning_rate": 0.0002, "epoch": 7.517418408507518, "step": 10250}, {"loss": 0.402, "grad_norm": 1.396972894668579, "learning_rate": 0.0002, "epoch": 7.524752475247524, "step": 10260}, {"loss": 0.3658, "grad_norm": 1.384286880493164, "learning_rate": 0.0002, "epoch": 7.532086541987532, "step": 10270}, {"loss": 0.3813, "grad_norm": 1.478095531463623, "learning_rate": 0.0002, "epoch": 7.539420608727539, "step": 10280}, {"loss": 0.3813, "grad_norm": 1.2642205953598022, "learning_rate": 0.0002, "epoch": 7.546754675467547, "step": 10290}, {"loss": 0.4084, "grad_norm": 1.1110541820526123, "learning_rate": 0.0002, "epoch": 7.554088742207554, "step": 10300}, {"loss": 0.418, "grad_norm": 1.1147890090942383, "learning_rate": 0.0002, "epoch": 7.561422808947562, "step": 10310}, {"loss": 0.4148, "grad_norm": 1.5677998065948486, "learning_rate": 0.0002, "epoch": 7.568756875687569, "step": 10320}, {"loss": 0.3675, "grad_norm": 0.9347636699676514, "learning_rate": 0.0002, "epoch": 7.576090942427576, "step": 10330}, {"loss": 0.3534, "grad_norm": 1.1126737594604492, "learning_rate": 0.0002, "epoch": 7.583425009167583, "step": 10340}, {"loss": 0.418, "grad_norm": 1.462611198425293, "learning_rate": 0.0002, "epoch": 7.590759075907591, "step": 10350}, {"loss": 0.3755, "grad_norm": 0.9907522201538086, "learning_rate": 0.0002, "epoch": 7.598093142647598, "step": 10360}, {"loss": 0.4133, "grad_norm": 1.306152582168579, "learning_rate": 0.0002, "epoch": 7.605427209387606, "step": 10370}, {"loss": 0.3644, "grad_norm": 1.11135995388031, "learning_rate": 0.0002, "epoch": 7.612761276127613, "step": 10380}, {"loss": 0.3659, "grad_norm": 1.0825806856155396, "learning_rate": 0.0002, "epoch": 7.62009534286762, "step": 10390}, {"loss": 0.3952, "grad_norm": 1.5346975326538086, "learning_rate": 0.0002, "epoch": 7.627429409607627, "step": 10400}, {"loss": 0.3807, "grad_norm": 1.5885388851165771, "learning_rate": 0.0002, "epoch": 7.634763476347635, "step": 10410}, {"loss": 0.3751, "grad_norm": 1.130261778831482, "learning_rate": 0.0002, "epoch": 7.642097543087642, "step": 10420}, {"loss": 0.3598, "grad_norm": 1.2318342924118042, "learning_rate": 0.0002, "epoch": 7.64943160982765, "step": 10430}, {"loss": 0.3823, "grad_norm": 1.07103431224823, "learning_rate": 0.0002, "epoch": 7.656765676567657, "step": 10440}, {"loss": 0.4707, "grad_norm": 1.3836923837661743, "learning_rate": 0.0002, "epoch": 7.6640997433076645, "step": 10450}, {"loss": 0.3829, "grad_norm": 1.2110271453857422, "learning_rate": 0.0002, "epoch": 7.671433810047671, "step": 10460}, {"loss": 0.3747, "grad_norm": 1.2304844856262207, "learning_rate": 0.0002, "epoch": 7.678767876787679, "step": 10470}, {"loss": 0.3761, "grad_norm": 1.3444706201553345, "learning_rate": 0.0002, "epoch": 7.686101943527686, "step": 10480}, {"loss": 0.3772, "grad_norm": 1.151705026626587, "learning_rate": 0.0002, "epoch": 7.6934360102676935, "step": 10490}, {"loss": 0.3351, "grad_norm": 1.4373983144760132, "learning_rate": 0.0002, "epoch": 7.700770077007701, "step": 10500}, {"loss": 0.389, "grad_norm": 1.5898514986038208, "learning_rate": 0.0002, "epoch": 7.7081041437477085, "step": 10510}, {"loss": 0.3679, "grad_norm": 1.3767389059066772, "learning_rate": 0.0002, "epoch": 7.715438210487715, "step": 10520}, {"loss": 0.4023, "grad_norm": 1.3730027675628662, "learning_rate": 0.0002, "epoch": 7.7227722772277225, "step": 10530}, {"loss": 0.4291, "grad_norm": 1.3917304277420044, "learning_rate": 0.0002, "epoch": 7.73010634396773, "step": 10540}, {"loss": 0.4378, "grad_norm": 1.230663776397705, "learning_rate": 0.0002, "epoch": 7.7374404107077375, "step": 10550}, {"loss": 0.4289, "grad_norm": 0.9922441244125366, "learning_rate": 0.0002, "epoch": 7.744774477447745, "step": 10560}, {"loss": 0.4118, "grad_norm": 1.4576551914215088, "learning_rate": 0.0002, "epoch": 7.752108544187752, "step": 10570}, {"loss": 0.3793, "grad_norm": 1.061685562133789, "learning_rate": 0.0002, "epoch": 7.75944261092776, "step": 10580}, {"loss": 0.3748, "grad_norm": 1.1523276567459106, "learning_rate": 0.0002, "epoch": 7.7667766776677665, "step": 10590}, {"loss": 0.3982, "grad_norm": 1.3917267322540283, "learning_rate": 0.0002, "epoch": 7.774110744407774, "step": 10600}, {"loss": 0.4305, "grad_norm": 1.1314283609390259, "learning_rate": 0.0002, "epoch": 7.781444811147781, "step": 10610}, {"loss": 0.4244, "grad_norm": 1.624324083328247, "learning_rate": 0.0002, "epoch": 7.788778877887789, "step": 10620}, {"loss": 0.4129, "grad_norm": 1.5369168519973755, "learning_rate": 0.0002, "epoch": 7.796112944627796, "step": 10630}, {"loss": 0.3661, "grad_norm": 1.082222819328308, "learning_rate": 0.0002, "epoch": 7.803447011367804, "step": 10640}, {"loss": 0.4341, "grad_norm": 1.281540870666504, "learning_rate": 0.0002, "epoch": 7.81078107810781, "step": 10650}, {"loss": 0.3939, "grad_norm": 1.1889171600341797, "learning_rate": 0.0002, "epoch": 7.818115144847818, "step": 10660}, {"loss": 0.3923, "grad_norm": 1.5310896635055542, "learning_rate": 0.0002, "epoch": 7.825449211587825, "step": 10670}, {"loss": 0.4148, "grad_norm": 1.6724708080291748, "learning_rate": 0.0002, "epoch": 7.832783278327833, "step": 10680}, {"loss": 0.363, "grad_norm": 1.3742409944534302, "learning_rate": 0.0002, "epoch": 7.84011734506784, "step": 10690}, {"loss": 0.3599, "grad_norm": 1.2421947717666626, "learning_rate": 0.0002, "epoch": 7.847451411807848, "step": 10700}, {"loss": 0.3829, "grad_norm": 1.0233848094940186, "learning_rate": 0.0002, "epoch": 7.854785478547855, "step": 10710}, {"loss": 0.4265, "grad_norm": 1.640842080116272, "learning_rate": 0.0002, "epoch": 7.862119545287862, "step": 10720}, {"loss": 0.4257, "grad_norm": 1.3571926355361938, "learning_rate": 0.0002, "epoch": 7.869453612027869, "step": 10730}, {"loss": 0.4454, "grad_norm": 1.459564208984375, "learning_rate": 0.0002, "epoch": 7.876787678767877, "step": 10740}, {"loss": 0.3903, "grad_norm": 0.9202831387519836, "learning_rate": 0.0002, "epoch": 7.884121745507884, "step": 10750}, {"loss": 0.4149, "grad_norm": 1.3509176969528198, "learning_rate": 0.0002, "epoch": 7.891455812247892, "step": 10760}, {"loss": 0.4001, "grad_norm": 1.5858603715896606, "learning_rate": 0.0002, "epoch": 7.898789878987898, "step": 10770}, {"loss": 0.3753, "grad_norm": 1.2391952276229858, "learning_rate": 0.0002, "epoch": 7.906123945727906, "step": 10780}, {"loss": 0.4085, "grad_norm": 1.3442552089691162, "learning_rate": 0.0002, "epoch": 7.913458012467913, "step": 10790}, {"loss": 0.4377, "grad_norm": 1.7327884435653687, "learning_rate": 0.0002, "epoch": 7.920792079207921, "step": 10800}, {"loss": 0.376, "grad_norm": 1.4246922731399536, "learning_rate": 0.0002, "epoch": 7.928126145947928, "step": 10810}, {"loss": 0.4158, "grad_norm": 1.4421411752700806, "learning_rate": 0.0002, "epoch": 7.935460212687936, "step": 10820}, {"loss": 0.4084, "grad_norm": 1.3445014953613281, "learning_rate": 0.0002, "epoch": 7.942794279427943, "step": 10830}, {"loss": 0.3986, "grad_norm": 1.2219295501708984, "learning_rate": 0.0002, "epoch": 7.950128346167951, "step": 10840}, {"loss": 0.428, "grad_norm": 1.241843342781067, "learning_rate": 0.0002, "epoch": 7.957462412907957, "step": 10850}, {"loss": 0.3776, "grad_norm": 0.9814007878303528, "learning_rate": 0.0002, "epoch": 7.964796479647965, "step": 10860}, {"loss": 0.4866, "grad_norm": 1.4015462398529053, "learning_rate": 0.0002, "epoch": 7.972130546387972, "step": 10870}, {"loss": 0.437, "grad_norm": 1.4638406038284302, "learning_rate": 0.0002, "epoch": 7.97946461312798, "step": 10880}, {"loss": 0.4191, "grad_norm": 1.585194706916809, "learning_rate": 0.0002, "epoch": 7.986798679867987, "step": 10890}, {"loss": 0.3969, "grad_norm": 1.197031855583191, "learning_rate": 0.0002, "epoch": 7.994132746607994, "step": 10900}]}