diff --git a/.gitattributes b/.gitattributes
index b35593da2da780c0c3ac827a66630d73015a0af8..0c46ef5d938b208667e5cbe7a3abd1b269b94f2c 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -884,3 +884,12 @@ gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-
 gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-1/checkpoint-8289/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-1/checkpoint-9670/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-1/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-0.8-num-5623-sd-42/checkpoint-1188/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-0.8-num-5623-sd-42/checkpoint-2377/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-0.8-num-5623-sd-42/checkpoint-3565/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-0.8-num-5623-sd-42/checkpoint-4754/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-0.8-num-5623-sd-42/checkpoint-5942/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-0.8-num-5623-sd-42/checkpoint-7131/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-0.8-num-5623-sd-42/checkpoint-8319/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-0.8-num-5623-sd-42/checkpoint-9504/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-0.8-num-5623-sd-42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/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-0.8-num-5623-sd-42/README.md b/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-0.8-num-5623-sd-42/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/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_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-0.8-num-5623-sd-42/adapter_config.json b/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-0.8-num-5623-sd-42/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/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_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-0.8-num-5623-sd-42/adapter_model.safetensors b/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-0.8-num-5623-sd-42/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b9eca8e044620be3ce058a1783f5a421a1736fd6
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e67bfcb11314afe23d52fe35c156ccfcb48ab2e93bd10c7abf7c1376d9b958d
+size 109069176
diff --git a/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-0.8-num-5623-sd-42/checkpoint-1188/README.md b/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-0.8-num-5623-sd-42/checkpoint-1188/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-1188/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_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-0.8-num-5623-sd-42/checkpoint-1188/adapter_config.json b/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-0.8-num-5623-sd-42/checkpoint-1188/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-1188/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_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-0.8-num-5623-sd-42/checkpoint-1188/adapter_model.safetensors b/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-0.8-num-5623-sd-42/checkpoint-1188/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..04756492035a96f74ab026f558aac7c7c1c3752d
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-1188/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6a766b5a8cb310431a6373ef78d5c7816e3b99ec1bf6134964a443595e21c12
+size 109069176
diff --git a/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-0.8-num-5623-sd-42/checkpoint-1188/optimizer.pt b/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-0.8-num-5623-sd-42/checkpoint-1188/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fa865bfb191dd0da150fdcd2e9df1dfb7c750bff
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-1188/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cdbc159bd3df1d681732bb87a4f18efebbbc6ee634b52296157df759110522a
+size 55532666
diff --git a/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-0.8-num-5623-sd-42/checkpoint-1188/rng_state.pth b/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-0.8-num-5623-sd-42/checkpoint-1188/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..1c3441bb7d6925db65a31929aecd1a111b60f5d8
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-1188/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8b9f092d336f37a81d1ef105078f4a0c705a9b85b218a3c1b473f24caa0e8e0
+size 14244
diff --git a/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-0.8-num-5623-sd-42/checkpoint-1188/scheduler.pt b/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-0.8-num-5623-sd-42/checkpoint-1188/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c6d6ca61a8d538d11e10a20ed03a220172d4bf5f
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-1188/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:894f660531cb1239606766f6559fac4e84c1220ce4349a19c8858d608371b1b7
+size 1064
diff --git a/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-0.8-num-5623-sd-42/checkpoint-1188/special_tokens_map.json b/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-0.8-num-5623-sd-42/checkpoint-1188/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-1188/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_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-0.8-num-5623-sd-42/checkpoint-1188/tokenizer.json b/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-0.8-num-5623-sd-42/checkpoint-1188/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-1188/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_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-0.8-num-5623-sd-42/checkpoint-1188/tokenizer_config.json b/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-0.8-num-5623-sd-42/checkpoint-1188/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-1188/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_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-0.8-num-5623-sd-42/checkpoint-1188/trainer_state.json b/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-0.8-num-5623-sd-42/checkpoint-1188/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..d1ca1fe0503185b38b78622e51fce93e6bf730e5
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-1188/trainer_state.json
@@ -0,0 +1,867 @@
+{
+  "best_metric": 1.8077726364135742,
+  "best_model_checkpoint": "outputs-001/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-0.8-num-5623-sd-42/checkpoint-1188",
+  "epoch": 0.9995793016407236,
+  "eval_steps": 10,
+  "global_step": 1188,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008413967185527976,
+      "grad_norm": 0.5458821654319763,
+      "learning_rate": 0.0002,
+      "loss": 2.56,
+      "step": 10
+    },
+    {
+      "epoch": 0.016827934371055953,
+      "grad_norm": 0.7293308973312378,
+      "learning_rate": 0.0002,
+      "loss": 2.3235,
+      "step": 20
+    },
+    {
+      "epoch": 0.02524190155658393,
+      "grad_norm": 0.47792306542396545,
+      "learning_rate": 0.0002,
+      "loss": 2.0815,
+      "step": 30
+    },
+    {
+      "epoch": 0.033655868742111905,
+      "grad_norm": 0.5944402813911438,
+      "learning_rate": 0.0002,
+      "loss": 1.9718,
+      "step": 40
+    },
+    {
+      "epoch": 0.04206983592763988,
+      "grad_norm": 0.5415359735488892,
+      "learning_rate": 0.0002,
+      "loss": 1.8848,
+      "step": 50
+    },
+    {
+      "epoch": 0.05048380311316786,
+      "grad_norm": 0.535713791847229,
+      "learning_rate": 0.0002,
+      "loss": 1.8953,
+      "step": 60
+    },
+    {
+      "epoch": 0.058897770298695834,
+      "grad_norm": 0.5184146761894226,
+      "learning_rate": 0.0002,
+      "loss": 1.937,
+      "step": 70
+    },
+    {
+      "epoch": 0.06731173748422381,
+      "grad_norm": 0.458926796913147,
+      "learning_rate": 0.0002,
+      "loss": 1.8396,
+      "step": 80
+    },
+    {
+      "epoch": 0.07572570466975179,
+      "grad_norm": 0.4780142307281494,
+      "learning_rate": 0.0002,
+      "loss": 1.8677,
+      "step": 90
+    },
+    {
+      "epoch": 0.08413967185527976,
+      "grad_norm": 0.79965740442276,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 100
+    },
+    {
+      "epoch": 0.09255363904080774,
+      "grad_norm": 0.4498862028121948,
+      "learning_rate": 0.0002,
+      "loss": 1.9081,
+      "step": 110
+    },
+    {
+      "epoch": 0.10096760622633572,
+      "grad_norm": 0.39338430762290955,
+      "learning_rate": 0.0002,
+      "loss": 1.8503,
+      "step": 120
+    },
+    {
+      "epoch": 0.10938157341186369,
+      "grad_norm": 0.9588953852653503,
+      "learning_rate": 0.0002,
+      "loss": 1.8637,
+      "step": 130
+    },
+    {
+      "epoch": 0.11779554059739167,
+      "grad_norm": 0.41675639152526855,
+      "learning_rate": 0.0002,
+      "loss": 1.8676,
+      "step": 140
+    },
+    {
+      "epoch": 0.12620950778291964,
+      "grad_norm": 0.44519832730293274,
+      "learning_rate": 0.0002,
+      "loss": 1.8904,
+      "step": 150
+    },
+    {
+      "epoch": 0.13462347496844762,
+      "grad_norm": 0.4176260530948639,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 160
+    },
+    {
+      "epoch": 0.1430374421539756,
+      "grad_norm": 0.35840365290641785,
+      "learning_rate": 0.0002,
+      "loss": 1.8398,
+      "step": 170
+    },
+    {
+      "epoch": 0.15145140933950357,
+      "grad_norm": 0.3794495463371277,
+      "learning_rate": 0.0002,
+      "loss": 1.8666,
+      "step": 180
+    },
+    {
+      "epoch": 0.15986537652503155,
+      "grad_norm": 0.4563522934913635,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 190
+    },
+    {
+      "epoch": 0.16827934371055953,
+      "grad_norm": 0.37057486176490784,
+      "learning_rate": 0.0002,
+      "loss": 1.8893,
+      "step": 200
+    },
+    {
+      "epoch": 0.1766933108960875,
+      "grad_norm": 0.44081518054008484,
+      "learning_rate": 0.0002,
+      "loss": 1.7995,
+      "step": 210
+    },
+    {
+      "epoch": 0.18510727808161548,
+      "grad_norm": 0.46078577637672424,
+      "learning_rate": 0.0002,
+      "loss": 1.9048,
+      "step": 220
+    },
+    {
+      "epoch": 0.19352124526714345,
+      "grad_norm": 0.36132094264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.8403,
+      "step": 230
+    },
+    {
+      "epoch": 0.20193521245267143,
+      "grad_norm": 0.3747289180755615,
+      "learning_rate": 0.0002,
+      "loss": 1.8827,
+      "step": 240
+    },
+    {
+      "epoch": 0.2103491796381994,
+      "grad_norm": 0.3540179133415222,
+      "learning_rate": 0.0002,
+      "loss": 1.8382,
+      "step": 250
+    },
+    {
+      "epoch": 0.21876314682372738,
+      "grad_norm": 0.3461375832557678,
+      "learning_rate": 0.0002,
+      "loss": 1.8196,
+      "step": 260
+    },
+    {
+      "epoch": 0.22717711400925536,
+      "grad_norm": 0.3436960279941559,
+      "learning_rate": 0.0002,
+      "loss": 1.8509,
+      "step": 270
+    },
+    {
+      "epoch": 0.23559108119478334,
+      "grad_norm": 0.35403719544410706,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 280
+    },
+    {
+      "epoch": 0.2440050483803113,
+      "grad_norm": 0.37142616510391235,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 290
+    },
+    {
+      "epoch": 0.2524190155658393,
+      "grad_norm": 0.3307955861091614,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 300
+    },
+    {
+      "epoch": 0.2608329827513673,
+      "grad_norm": 0.32855314016342163,
+      "learning_rate": 0.0002,
+      "loss": 1.817,
+      "step": 310
+    },
+    {
+      "epoch": 0.26924694993689524,
+      "grad_norm": 0.3299003839492798,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 320
+    },
+    {
+      "epoch": 0.27766091712242325,
+      "grad_norm": 0.44311287999153137,
+      "learning_rate": 0.0002,
+      "loss": 1.8129,
+      "step": 330
+    },
+    {
+      "epoch": 0.2860748843079512,
+      "grad_norm": 0.32989758253097534,
+      "learning_rate": 0.0002,
+      "loss": 1.8232,
+      "step": 340
+    },
+    {
+      "epoch": 0.2944888514934792,
+      "grad_norm": 0.34400200843811035,
+      "learning_rate": 0.0002,
+      "loss": 1.7716,
+      "step": 350
+    },
+    {
+      "epoch": 0.30290281867900715,
+      "grad_norm": 0.36286211013793945,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 360
+    },
+    {
+      "epoch": 0.31131678586453515,
+      "grad_norm": 0.406827837228775,
+      "learning_rate": 0.0002,
+      "loss": 1.8025,
+      "step": 370
+    },
+    {
+      "epoch": 0.3197307530500631,
+      "grad_norm": 0.36299195885658264,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 380
+    },
+    {
+      "epoch": 0.3281447202355911,
+      "grad_norm": 0.3477257192134857,
+      "learning_rate": 0.0002,
+      "loss": 1.837,
+      "step": 390
+    },
+    {
+      "epoch": 0.33655868742111905,
+      "grad_norm": 0.3730369210243225,
+      "learning_rate": 0.0002,
+      "loss": 1.7767,
+      "step": 400
+    },
+    {
+      "epoch": 0.34497265460664706,
+      "grad_norm": 0.4644559919834137,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 410
+    },
+    {
+      "epoch": 0.353386621792175,
+      "grad_norm": 0.406576544046402,
+      "learning_rate": 0.0002,
+      "loss": 1.7538,
+      "step": 420
+    },
+    {
+      "epoch": 0.361800588977703,
+      "grad_norm": 0.3612699508666992,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 430
+    },
+    {
+      "epoch": 0.37021455616323096,
+      "grad_norm": 0.3243742287158966,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 440
+    },
+    {
+      "epoch": 0.37862852334875896,
+      "grad_norm": 0.36671221256256104,
+      "learning_rate": 0.0002,
+      "loss": 1.8851,
+      "step": 450
+    },
+    {
+      "epoch": 0.3870424905342869,
+      "grad_norm": 0.3565002381801605,
+      "learning_rate": 0.0002,
+      "loss": 1.8853,
+      "step": 460
+    },
+    {
+      "epoch": 0.3954564577198149,
+      "grad_norm": 0.34630221128463745,
+      "learning_rate": 0.0002,
+      "loss": 1.8923,
+      "step": 470
+    },
+    {
+      "epoch": 0.40387042490534286,
+      "grad_norm": 0.3353537321090698,
+      "learning_rate": 0.0002,
+      "loss": 1.8234,
+      "step": 480
+    },
+    {
+      "epoch": 0.41228439209087087,
+      "grad_norm": 0.4015921950340271,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 490
+    },
+    {
+      "epoch": 0.4206983592763988,
+      "grad_norm": 0.5489419102668762,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 500
+    },
+    {
+      "epoch": 0.4291123264619268,
+      "grad_norm": 0.4193589985370636,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 510
+    },
+    {
+      "epoch": 0.43752629364745477,
+      "grad_norm": 0.3418922424316406,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 520
+    },
+    {
+      "epoch": 0.44594026083298277,
+      "grad_norm": 0.32668185234069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 530
+    },
+    {
+      "epoch": 0.4543542280185107,
+      "grad_norm": 0.3094325661659241,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 540
+    },
+    {
+      "epoch": 0.4627681952040387,
+      "grad_norm": 0.3743017315864563,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 550
+    },
+    {
+      "epoch": 0.47118216238956667,
+      "grad_norm": 0.3295630216598511,
+      "learning_rate": 0.0002,
+      "loss": 1.8451,
+      "step": 560
+    },
+    {
+      "epoch": 0.4795961295750947,
+      "grad_norm": 1.6124513149261475,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 570
+    },
+    {
+      "epoch": 0.4880100967606226,
+      "grad_norm": 0.3245585858821869,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 580
+    },
+    {
+      "epoch": 0.49642406394615063,
+      "grad_norm": 0.3332934081554413,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 590
+    },
+    {
+      "epoch": 0.5048380311316786,
+      "grad_norm": 0.3836138844490051,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 600
+    },
+    {
+      "epoch": 0.5132519983172066,
+      "grad_norm": 0.32953888177871704,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 610
+    },
+    {
+      "epoch": 0.5216659655027346,
+      "grad_norm": 0.36291512846946716,
+      "learning_rate": 0.0002,
+      "loss": 1.7729,
+      "step": 620
+    },
+    {
+      "epoch": 0.5300799326882625,
+      "grad_norm": 0.3237783908843994,
+      "learning_rate": 0.0002,
+      "loss": 1.7758,
+      "step": 630
+    },
+    {
+      "epoch": 0.5384938998737905,
+      "grad_norm": 0.38882696628570557,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 640
+    },
+    {
+      "epoch": 0.5469078670593185,
+      "grad_norm": 0.37821972370147705,
+      "learning_rate": 0.0002,
+      "loss": 1.8624,
+      "step": 650
+    },
+    {
+      "epoch": 0.5553218342448465,
+      "grad_norm": 0.3556285500526428,
+      "learning_rate": 0.0002,
+      "loss": 1.8075,
+      "step": 660
+    },
+    {
+      "epoch": 0.5637358014303744,
+      "grad_norm": 0.347499281167984,
+      "learning_rate": 0.0002,
+      "loss": 1.778,
+      "step": 670
+    },
+    {
+      "epoch": 0.5721497686159024,
+      "grad_norm": 0.3176489472389221,
+      "learning_rate": 0.0002,
+      "loss": 1.8066,
+      "step": 680
+    },
+    {
+      "epoch": 0.5805637358014304,
+      "grad_norm": 0.30220088362693787,
+      "learning_rate": 0.0002,
+      "loss": 1.7257,
+      "step": 690
+    },
+    {
+      "epoch": 0.5889777029869584,
+      "grad_norm": 0.3711601793766022,
+      "learning_rate": 0.0002,
+      "loss": 1.8415,
+      "step": 700
+    },
+    {
+      "epoch": 0.5973916701724863,
+      "grad_norm": 0.3311759829521179,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 710
+    },
+    {
+      "epoch": 0.6058056373580143,
+      "grad_norm": 0.34824270009994507,
+      "learning_rate": 0.0002,
+      "loss": 1.7712,
+      "step": 720
+    },
+    {
+      "epoch": 0.6142196045435423,
+      "grad_norm": 0.29668381810188293,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 730
+    },
+    {
+      "epoch": 0.6226335717290703,
+      "grad_norm": 0.36087489128112793,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 740
+    },
+    {
+      "epoch": 0.6310475389145982,
+      "grad_norm": 0.31590089201927185,
+      "learning_rate": 0.0002,
+      "loss": 1.7956,
+      "step": 750
+    },
+    {
+      "epoch": 0.6394615061001262,
+      "grad_norm": 0.37632957100868225,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 760
+    },
+    {
+      "epoch": 0.6478754732856542,
+      "grad_norm": 0.3360748589038849,
+      "learning_rate": 0.0002,
+      "loss": 1.8499,
+      "step": 770
+    },
+    {
+      "epoch": 0.6562894404711822,
+      "grad_norm": 0.3420640528202057,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 780
+    },
+    {
+      "epoch": 0.6647034076567101,
+      "grad_norm": 0.5734959244728088,
+      "learning_rate": 0.0002,
+      "loss": 1.8353,
+      "step": 790
+    },
+    {
+      "epoch": 0.6731173748422381,
+      "grad_norm": 0.36440837383270264,
+      "learning_rate": 0.0002,
+      "loss": 1.7746,
+      "step": 800
+    },
+    {
+      "epoch": 0.6815313420277661,
+      "grad_norm": 0.3179708421230316,
+      "learning_rate": 0.0002,
+      "loss": 1.7532,
+      "step": 810
+    },
+    {
+      "epoch": 0.6899453092132941,
+      "grad_norm": 0.34122881293296814,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 820
+    },
+    {
+      "epoch": 0.698359276398822,
+      "grad_norm": 0.31886112689971924,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 830
+    },
+    {
+      "epoch": 0.70677324358435,
+      "grad_norm": 0.31782326102256775,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 840
+    },
+    {
+      "epoch": 0.715187210769878,
+      "grad_norm": 0.36052989959716797,
+      "learning_rate": 0.0002,
+      "loss": 1.7588,
+      "step": 850
+    },
+    {
+      "epoch": 0.723601177955406,
+      "grad_norm": 0.28946155309677124,
+      "learning_rate": 0.0002,
+      "loss": 1.7891,
+      "step": 860
+    },
+    {
+      "epoch": 0.7320151451409339,
+      "grad_norm": 0.3095663785934448,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 870
+    },
+    {
+      "epoch": 0.7404291123264619,
+      "grad_norm": 0.3317491412162781,
+      "learning_rate": 0.0002,
+      "loss": 1.785,
+      "step": 880
+    },
+    {
+      "epoch": 0.7488430795119899,
+      "grad_norm": 0.31324660778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 890
+    },
+    {
+      "epoch": 0.7572570466975179,
+      "grad_norm": 0.3290475606918335,
+      "learning_rate": 0.0002,
+      "loss": 1.8753,
+      "step": 900
+    },
+    {
+      "epoch": 0.7656710138830458,
+      "grad_norm": 0.35690343379974365,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 910
+    },
+    {
+      "epoch": 0.7740849810685738,
+      "grad_norm": 0.39558273553848267,
+      "learning_rate": 0.0002,
+      "loss": 1.826,
+      "step": 920
+    },
+    {
+      "epoch": 0.7824989482541018,
+      "grad_norm": 0.34254348278045654,
+      "learning_rate": 0.0002,
+      "loss": 1.8722,
+      "step": 930
+    },
+    {
+      "epoch": 0.7909129154396298,
+      "grad_norm": 0.3560165464878082,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 940
+    },
+    {
+      "epoch": 0.7993268826251577,
+      "grad_norm": 0.30693164467811584,
+      "learning_rate": 0.0002,
+      "loss": 1.7992,
+      "step": 950
+    },
+    {
+      "epoch": 0.8077408498106857,
+      "grad_norm": 0.3394823372364044,
+      "learning_rate": 0.0002,
+      "loss": 1.8029,
+      "step": 960
+    },
+    {
+      "epoch": 0.8161548169962137,
+      "grad_norm": 0.3741514980792999,
+      "learning_rate": 0.0002,
+      "loss": 1.8105,
+      "step": 970
+    },
+    {
+      "epoch": 0.8245687841817417,
+      "grad_norm": 0.3655228316783905,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 980
+    },
+    {
+      "epoch": 0.8329827513672696,
+      "grad_norm": 0.3586033880710602,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 990
+    },
+    {
+      "epoch": 0.8413967185527976,
+      "grad_norm": 0.3459678888320923,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8498106857383256,
+      "grad_norm": 0.3184349834918976,
+      "learning_rate": 0.0002,
+      "loss": 1.8498,
+      "step": 1010
+    },
+    {
+      "epoch": 0.8582246529238536,
+      "grad_norm": 0.3099786043167114,
+      "learning_rate": 0.0002,
+      "loss": 1.7632,
+      "step": 1020
+    },
+    {
+      "epoch": 0.8666386201093815,
+      "grad_norm": 0.30300915241241455,
+      "learning_rate": 0.0002,
+      "loss": 1.8067,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8750525872949095,
+      "grad_norm": 0.3128705620765686,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8834665544804375,
+      "grad_norm": 0.3336263597011566,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8918805216659655,
+      "grad_norm": 0.3801328241825104,
+      "learning_rate": 0.0002,
+      "loss": 1.8375,
+      "step": 1060
+    },
+    {
+      "epoch": 0.9002944888514934,
+      "grad_norm": 0.3122096359729767,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 1070
+    },
+    {
+      "epoch": 0.9087084560370214,
+      "grad_norm": 0.35990869998931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 1080
+    },
+    {
+      "epoch": 0.9171224232225494,
+      "grad_norm": 0.3321819305419922,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1090
+    },
+    {
+      "epoch": 0.9255363904080774,
+      "grad_norm": 0.4202139377593994,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9339503575936053,
+      "grad_norm": 0.32559722661972046,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 1110
+    },
+    {
+      "epoch": 0.9423643247791333,
+      "grad_norm": 0.3098459839820862,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 1120
+    },
+    {
+      "epoch": 0.9507782919646613,
+      "grad_norm": 0.33917108178138733,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1130
+    },
+    {
+      "epoch": 0.9591922591501894,
+      "grad_norm": 0.4055837094783783,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1140
+    },
+    {
+      "epoch": 0.9676062263357172,
+      "grad_norm": 0.32508623600006104,
+      "learning_rate": 0.0002,
+      "loss": 1.8259,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9760201935212452,
+      "grad_norm": 0.30150601267814636,
+      "learning_rate": 0.0002,
+      "loss": 1.782,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9844341607067733,
+      "grad_norm": 0.3042563199996948,
+      "learning_rate": 0.0002,
+      "loss": 1.8291,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9928481278923013,
+      "grad_norm": 0.33254584670066833,
+      "learning_rate": 0.0002,
+      "loss": 1.7847,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9995793016407236,
+      "eval_loss": 1.8077726364135742,
+      "eval_runtime": 38.4359,
+      "eval_samples_per_second": 13.399,
+      "eval_steps_per_second": 1.691,
+      "step": 1188
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 9504,
+  "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": 5.500111884071731e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/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-0.8-num-5623-sd-42/checkpoint-1188/training_args.bin b/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-0.8-num-5623-sd-42/checkpoint-1188/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f1502d478cfbb1424f707352d007b740bde5e373
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-1188/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df2b79d3acefeedef5a0229881de39ec68ef9b40046a60d7976a49f7e6b3b936
+size 5560
diff --git a/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-0.8-num-5623-sd-42/checkpoint-2377/README.md b/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-0.8-num-5623-sd-42/checkpoint-2377/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-2377/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_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-0.8-num-5623-sd-42/checkpoint-2377/adapter_config.json b/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-0.8-num-5623-sd-42/checkpoint-2377/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-2377/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_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-0.8-num-5623-sd-42/checkpoint-2377/adapter_model.safetensors b/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-0.8-num-5623-sd-42/checkpoint-2377/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b9eca8e044620be3ce058a1783f5a421a1736fd6
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-2377/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e67bfcb11314afe23d52fe35c156ccfcb48ab2e93bd10c7abf7c1376d9b958d
+size 109069176
diff --git a/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-0.8-num-5623-sd-42/checkpoint-2377/optimizer.pt b/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-0.8-num-5623-sd-42/checkpoint-2377/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b692e373cc096a7cdb6ea39962d7ac1d446bfdb1
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-2377/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0af334a78bf7194c5014e3070ce3111162e24cd28d906ba20333107b38b9932
+size 55532666
diff --git a/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-0.8-num-5623-sd-42/checkpoint-2377/rng_state.pth b/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-0.8-num-5623-sd-42/checkpoint-2377/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..1f95d736f5723087a478fbcbb62288c39adc449a
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-2377/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72c34d90a4f85cd0c9aa48d0b81fd5edad3e1427338f0942568ff10c9f0e75f8
+size 14244
diff --git a/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-0.8-num-5623-sd-42/checkpoint-2377/scheduler.pt b/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-0.8-num-5623-sd-42/checkpoint-2377/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5d6f390c6996e4c0fd28834642189b9c340eb0b1
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-2377/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38384347c3fbdf548f3429703d7a725557ee363e767816165662ae294c9e82ec
+size 1064
diff --git a/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-0.8-num-5623-sd-42/checkpoint-2377/special_tokens_map.json b/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-0.8-num-5623-sd-42/checkpoint-2377/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-2377/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_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-0.8-num-5623-sd-42/checkpoint-2377/tokenizer.json b/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-0.8-num-5623-sd-42/checkpoint-2377/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-2377/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_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-0.8-num-5623-sd-42/checkpoint-2377/tokenizer_config.json b/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-0.8-num-5623-sd-42/checkpoint-2377/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-2377/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_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-0.8-num-5623-sd-42/checkpoint-2377/trainer_state.json b/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-0.8-num-5623-sd-42/checkpoint-2377/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..0d106069a7038427e703fd55a559f0be01bf4f2b
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-2377/trainer_state.json
@@ -0,0 +1,1708 @@
+{
+  "best_metric": 1.8055059909820557,
+  "best_model_checkpoint": "outputs-001/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-0.8-num-5623-sd-42/checkpoint-2377",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 2377,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008413967185527976,
+      "grad_norm": 0.5458821654319763,
+      "learning_rate": 0.0002,
+      "loss": 2.56,
+      "step": 10
+    },
+    {
+      "epoch": 0.016827934371055953,
+      "grad_norm": 0.7293308973312378,
+      "learning_rate": 0.0002,
+      "loss": 2.3235,
+      "step": 20
+    },
+    {
+      "epoch": 0.02524190155658393,
+      "grad_norm": 0.47792306542396545,
+      "learning_rate": 0.0002,
+      "loss": 2.0815,
+      "step": 30
+    },
+    {
+      "epoch": 0.033655868742111905,
+      "grad_norm": 0.5944402813911438,
+      "learning_rate": 0.0002,
+      "loss": 1.9718,
+      "step": 40
+    },
+    {
+      "epoch": 0.04206983592763988,
+      "grad_norm": 0.5415359735488892,
+      "learning_rate": 0.0002,
+      "loss": 1.8848,
+      "step": 50
+    },
+    {
+      "epoch": 0.05048380311316786,
+      "grad_norm": 0.535713791847229,
+      "learning_rate": 0.0002,
+      "loss": 1.8953,
+      "step": 60
+    },
+    {
+      "epoch": 0.058897770298695834,
+      "grad_norm": 0.5184146761894226,
+      "learning_rate": 0.0002,
+      "loss": 1.937,
+      "step": 70
+    },
+    {
+      "epoch": 0.06731173748422381,
+      "grad_norm": 0.458926796913147,
+      "learning_rate": 0.0002,
+      "loss": 1.8396,
+      "step": 80
+    },
+    {
+      "epoch": 0.07572570466975179,
+      "grad_norm": 0.4780142307281494,
+      "learning_rate": 0.0002,
+      "loss": 1.8677,
+      "step": 90
+    },
+    {
+      "epoch": 0.08413967185527976,
+      "grad_norm": 0.79965740442276,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 100
+    },
+    {
+      "epoch": 0.09255363904080774,
+      "grad_norm": 0.4498862028121948,
+      "learning_rate": 0.0002,
+      "loss": 1.9081,
+      "step": 110
+    },
+    {
+      "epoch": 0.10096760622633572,
+      "grad_norm": 0.39338430762290955,
+      "learning_rate": 0.0002,
+      "loss": 1.8503,
+      "step": 120
+    },
+    {
+      "epoch": 0.10938157341186369,
+      "grad_norm": 0.9588953852653503,
+      "learning_rate": 0.0002,
+      "loss": 1.8637,
+      "step": 130
+    },
+    {
+      "epoch": 0.11779554059739167,
+      "grad_norm": 0.41675639152526855,
+      "learning_rate": 0.0002,
+      "loss": 1.8676,
+      "step": 140
+    },
+    {
+      "epoch": 0.12620950778291964,
+      "grad_norm": 0.44519832730293274,
+      "learning_rate": 0.0002,
+      "loss": 1.8904,
+      "step": 150
+    },
+    {
+      "epoch": 0.13462347496844762,
+      "grad_norm": 0.4176260530948639,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 160
+    },
+    {
+      "epoch": 0.1430374421539756,
+      "grad_norm": 0.35840365290641785,
+      "learning_rate": 0.0002,
+      "loss": 1.8398,
+      "step": 170
+    },
+    {
+      "epoch": 0.15145140933950357,
+      "grad_norm": 0.3794495463371277,
+      "learning_rate": 0.0002,
+      "loss": 1.8666,
+      "step": 180
+    },
+    {
+      "epoch": 0.15986537652503155,
+      "grad_norm": 0.4563522934913635,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 190
+    },
+    {
+      "epoch": 0.16827934371055953,
+      "grad_norm": 0.37057486176490784,
+      "learning_rate": 0.0002,
+      "loss": 1.8893,
+      "step": 200
+    },
+    {
+      "epoch": 0.1766933108960875,
+      "grad_norm": 0.44081518054008484,
+      "learning_rate": 0.0002,
+      "loss": 1.7995,
+      "step": 210
+    },
+    {
+      "epoch": 0.18510727808161548,
+      "grad_norm": 0.46078577637672424,
+      "learning_rate": 0.0002,
+      "loss": 1.9048,
+      "step": 220
+    },
+    {
+      "epoch": 0.19352124526714345,
+      "grad_norm": 0.36132094264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.8403,
+      "step": 230
+    },
+    {
+      "epoch": 0.20193521245267143,
+      "grad_norm": 0.3747289180755615,
+      "learning_rate": 0.0002,
+      "loss": 1.8827,
+      "step": 240
+    },
+    {
+      "epoch": 0.2103491796381994,
+      "grad_norm": 0.3540179133415222,
+      "learning_rate": 0.0002,
+      "loss": 1.8382,
+      "step": 250
+    },
+    {
+      "epoch": 0.21876314682372738,
+      "grad_norm": 0.3461375832557678,
+      "learning_rate": 0.0002,
+      "loss": 1.8196,
+      "step": 260
+    },
+    {
+      "epoch": 0.22717711400925536,
+      "grad_norm": 0.3436960279941559,
+      "learning_rate": 0.0002,
+      "loss": 1.8509,
+      "step": 270
+    },
+    {
+      "epoch": 0.23559108119478334,
+      "grad_norm": 0.35403719544410706,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 280
+    },
+    {
+      "epoch": 0.2440050483803113,
+      "grad_norm": 0.37142616510391235,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 290
+    },
+    {
+      "epoch": 0.2524190155658393,
+      "grad_norm": 0.3307955861091614,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 300
+    },
+    {
+      "epoch": 0.2608329827513673,
+      "grad_norm": 0.32855314016342163,
+      "learning_rate": 0.0002,
+      "loss": 1.817,
+      "step": 310
+    },
+    {
+      "epoch": 0.26924694993689524,
+      "grad_norm": 0.3299003839492798,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 320
+    },
+    {
+      "epoch": 0.27766091712242325,
+      "grad_norm": 0.44311287999153137,
+      "learning_rate": 0.0002,
+      "loss": 1.8129,
+      "step": 330
+    },
+    {
+      "epoch": 0.2860748843079512,
+      "grad_norm": 0.32989758253097534,
+      "learning_rate": 0.0002,
+      "loss": 1.8232,
+      "step": 340
+    },
+    {
+      "epoch": 0.2944888514934792,
+      "grad_norm": 0.34400200843811035,
+      "learning_rate": 0.0002,
+      "loss": 1.7716,
+      "step": 350
+    },
+    {
+      "epoch": 0.30290281867900715,
+      "grad_norm": 0.36286211013793945,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 360
+    },
+    {
+      "epoch": 0.31131678586453515,
+      "grad_norm": 0.406827837228775,
+      "learning_rate": 0.0002,
+      "loss": 1.8025,
+      "step": 370
+    },
+    {
+      "epoch": 0.3197307530500631,
+      "grad_norm": 0.36299195885658264,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 380
+    },
+    {
+      "epoch": 0.3281447202355911,
+      "grad_norm": 0.3477257192134857,
+      "learning_rate": 0.0002,
+      "loss": 1.837,
+      "step": 390
+    },
+    {
+      "epoch": 0.33655868742111905,
+      "grad_norm": 0.3730369210243225,
+      "learning_rate": 0.0002,
+      "loss": 1.7767,
+      "step": 400
+    },
+    {
+      "epoch": 0.34497265460664706,
+      "grad_norm": 0.4644559919834137,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 410
+    },
+    {
+      "epoch": 0.353386621792175,
+      "grad_norm": 0.406576544046402,
+      "learning_rate": 0.0002,
+      "loss": 1.7538,
+      "step": 420
+    },
+    {
+      "epoch": 0.361800588977703,
+      "grad_norm": 0.3612699508666992,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 430
+    },
+    {
+      "epoch": 0.37021455616323096,
+      "grad_norm": 0.3243742287158966,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 440
+    },
+    {
+      "epoch": 0.37862852334875896,
+      "grad_norm": 0.36671221256256104,
+      "learning_rate": 0.0002,
+      "loss": 1.8851,
+      "step": 450
+    },
+    {
+      "epoch": 0.3870424905342869,
+      "grad_norm": 0.3565002381801605,
+      "learning_rate": 0.0002,
+      "loss": 1.8853,
+      "step": 460
+    },
+    {
+      "epoch": 0.3954564577198149,
+      "grad_norm": 0.34630221128463745,
+      "learning_rate": 0.0002,
+      "loss": 1.8923,
+      "step": 470
+    },
+    {
+      "epoch": 0.40387042490534286,
+      "grad_norm": 0.3353537321090698,
+      "learning_rate": 0.0002,
+      "loss": 1.8234,
+      "step": 480
+    },
+    {
+      "epoch": 0.41228439209087087,
+      "grad_norm": 0.4015921950340271,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 490
+    },
+    {
+      "epoch": 0.4206983592763988,
+      "grad_norm": 0.5489419102668762,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 500
+    },
+    {
+      "epoch": 0.4291123264619268,
+      "grad_norm": 0.4193589985370636,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 510
+    },
+    {
+      "epoch": 0.43752629364745477,
+      "grad_norm": 0.3418922424316406,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 520
+    },
+    {
+      "epoch": 0.44594026083298277,
+      "grad_norm": 0.32668185234069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 530
+    },
+    {
+      "epoch": 0.4543542280185107,
+      "grad_norm": 0.3094325661659241,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 540
+    },
+    {
+      "epoch": 0.4627681952040387,
+      "grad_norm": 0.3743017315864563,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 550
+    },
+    {
+      "epoch": 0.47118216238956667,
+      "grad_norm": 0.3295630216598511,
+      "learning_rate": 0.0002,
+      "loss": 1.8451,
+      "step": 560
+    },
+    {
+      "epoch": 0.4795961295750947,
+      "grad_norm": 1.6124513149261475,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 570
+    },
+    {
+      "epoch": 0.4880100967606226,
+      "grad_norm": 0.3245585858821869,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 580
+    },
+    {
+      "epoch": 0.49642406394615063,
+      "grad_norm": 0.3332934081554413,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 590
+    },
+    {
+      "epoch": 0.5048380311316786,
+      "grad_norm": 0.3836138844490051,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 600
+    },
+    {
+      "epoch": 0.5132519983172066,
+      "grad_norm": 0.32953888177871704,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 610
+    },
+    {
+      "epoch": 0.5216659655027346,
+      "grad_norm": 0.36291512846946716,
+      "learning_rate": 0.0002,
+      "loss": 1.7729,
+      "step": 620
+    },
+    {
+      "epoch": 0.5300799326882625,
+      "grad_norm": 0.3237783908843994,
+      "learning_rate": 0.0002,
+      "loss": 1.7758,
+      "step": 630
+    },
+    {
+      "epoch": 0.5384938998737905,
+      "grad_norm": 0.38882696628570557,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 640
+    },
+    {
+      "epoch": 0.5469078670593185,
+      "grad_norm": 0.37821972370147705,
+      "learning_rate": 0.0002,
+      "loss": 1.8624,
+      "step": 650
+    },
+    {
+      "epoch": 0.5553218342448465,
+      "grad_norm": 0.3556285500526428,
+      "learning_rate": 0.0002,
+      "loss": 1.8075,
+      "step": 660
+    },
+    {
+      "epoch": 0.5637358014303744,
+      "grad_norm": 0.347499281167984,
+      "learning_rate": 0.0002,
+      "loss": 1.778,
+      "step": 670
+    },
+    {
+      "epoch": 0.5721497686159024,
+      "grad_norm": 0.3176489472389221,
+      "learning_rate": 0.0002,
+      "loss": 1.8066,
+      "step": 680
+    },
+    {
+      "epoch": 0.5805637358014304,
+      "grad_norm": 0.30220088362693787,
+      "learning_rate": 0.0002,
+      "loss": 1.7257,
+      "step": 690
+    },
+    {
+      "epoch": 0.5889777029869584,
+      "grad_norm": 0.3711601793766022,
+      "learning_rate": 0.0002,
+      "loss": 1.8415,
+      "step": 700
+    },
+    {
+      "epoch": 0.5973916701724863,
+      "grad_norm": 0.3311759829521179,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 710
+    },
+    {
+      "epoch": 0.6058056373580143,
+      "grad_norm": 0.34824270009994507,
+      "learning_rate": 0.0002,
+      "loss": 1.7712,
+      "step": 720
+    },
+    {
+      "epoch": 0.6142196045435423,
+      "grad_norm": 0.29668381810188293,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 730
+    },
+    {
+      "epoch": 0.6226335717290703,
+      "grad_norm": 0.36087489128112793,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 740
+    },
+    {
+      "epoch": 0.6310475389145982,
+      "grad_norm": 0.31590089201927185,
+      "learning_rate": 0.0002,
+      "loss": 1.7956,
+      "step": 750
+    },
+    {
+      "epoch": 0.6394615061001262,
+      "grad_norm": 0.37632957100868225,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 760
+    },
+    {
+      "epoch": 0.6478754732856542,
+      "grad_norm": 0.3360748589038849,
+      "learning_rate": 0.0002,
+      "loss": 1.8499,
+      "step": 770
+    },
+    {
+      "epoch": 0.6562894404711822,
+      "grad_norm": 0.3420640528202057,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 780
+    },
+    {
+      "epoch": 0.6647034076567101,
+      "grad_norm": 0.5734959244728088,
+      "learning_rate": 0.0002,
+      "loss": 1.8353,
+      "step": 790
+    },
+    {
+      "epoch": 0.6731173748422381,
+      "grad_norm": 0.36440837383270264,
+      "learning_rate": 0.0002,
+      "loss": 1.7746,
+      "step": 800
+    },
+    {
+      "epoch": 0.6815313420277661,
+      "grad_norm": 0.3179708421230316,
+      "learning_rate": 0.0002,
+      "loss": 1.7532,
+      "step": 810
+    },
+    {
+      "epoch": 0.6899453092132941,
+      "grad_norm": 0.34122881293296814,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 820
+    },
+    {
+      "epoch": 0.698359276398822,
+      "grad_norm": 0.31886112689971924,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 830
+    },
+    {
+      "epoch": 0.70677324358435,
+      "grad_norm": 0.31782326102256775,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 840
+    },
+    {
+      "epoch": 0.715187210769878,
+      "grad_norm": 0.36052989959716797,
+      "learning_rate": 0.0002,
+      "loss": 1.7588,
+      "step": 850
+    },
+    {
+      "epoch": 0.723601177955406,
+      "grad_norm": 0.28946155309677124,
+      "learning_rate": 0.0002,
+      "loss": 1.7891,
+      "step": 860
+    },
+    {
+      "epoch": 0.7320151451409339,
+      "grad_norm": 0.3095663785934448,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 870
+    },
+    {
+      "epoch": 0.7404291123264619,
+      "grad_norm": 0.3317491412162781,
+      "learning_rate": 0.0002,
+      "loss": 1.785,
+      "step": 880
+    },
+    {
+      "epoch": 0.7488430795119899,
+      "grad_norm": 0.31324660778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 890
+    },
+    {
+      "epoch": 0.7572570466975179,
+      "grad_norm": 0.3290475606918335,
+      "learning_rate": 0.0002,
+      "loss": 1.8753,
+      "step": 900
+    },
+    {
+      "epoch": 0.7656710138830458,
+      "grad_norm": 0.35690343379974365,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 910
+    },
+    {
+      "epoch": 0.7740849810685738,
+      "grad_norm": 0.39558273553848267,
+      "learning_rate": 0.0002,
+      "loss": 1.826,
+      "step": 920
+    },
+    {
+      "epoch": 0.7824989482541018,
+      "grad_norm": 0.34254348278045654,
+      "learning_rate": 0.0002,
+      "loss": 1.8722,
+      "step": 930
+    },
+    {
+      "epoch": 0.7909129154396298,
+      "grad_norm": 0.3560165464878082,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 940
+    },
+    {
+      "epoch": 0.7993268826251577,
+      "grad_norm": 0.30693164467811584,
+      "learning_rate": 0.0002,
+      "loss": 1.7992,
+      "step": 950
+    },
+    {
+      "epoch": 0.8077408498106857,
+      "grad_norm": 0.3394823372364044,
+      "learning_rate": 0.0002,
+      "loss": 1.8029,
+      "step": 960
+    },
+    {
+      "epoch": 0.8161548169962137,
+      "grad_norm": 0.3741514980792999,
+      "learning_rate": 0.0002,
+      "loss": 1.8105,
+      "step": 970
+    },
+    {
+      "epoch": 0.8245687841817417,
+      "grad_norm": 0.3655228316783905,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 980
+    },
+    {
+      "epoch": 0.8329827513672696,
+      "grad_norm": 0.3586033880710602,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 990
+    },
+    {
+      "epoch": 0.8413967185527976,
+      "grad_norm": 0.3459678888320923,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8498106857383256,
+      "grad_norm": 0.3184349834918976,
+      "learning_rate": 0.0002,
+      "loss": 1.8498,
+      "step": 1010
+    },
+    {
+      "epoch": 0.8582246529238536,
+      "grad_norm": 0.3099786043167114,
+      "learning_rate": 0.0002,
+      "loss": 1.7632,
+      "step": 1020
+    },
+    {
+      "epoch": 0.8666386201093815,
+      "grad_norm": 0.30300915241241455,
+      "learning_rate": 0.0002,
+      "loss": 1.8067,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8750525872949095,
+      "grad_norm": 0.3128705620765686,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8834665544804375,
+      "grad_norm": 0.3336263597011566,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8918805216659655,
+      "grad_norm": 0.3801328241825104,
+      "learning_rate": 0.0002,
+      "loss": 1.8375,
+      "step": 1060
+    },
+    {
+      "epoch": 0.9002944888514934,
+      "grad_norm": 0.3122096359729767,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 1070
+    },
+    {
+      "epoch": 0.9087084560370214,
+      "grad_norm": 0.35990869998931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 1080
+    },
+    {
+      "epoch": 0.9171224232225494,
+      "grad_norm": 0.3321819305419922,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1090
+    },
+    {
+      "epoch": 0.9255363904080774,
+      "grad_norm": 0.4202139377593994,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9339503575936053,
+      "grad_norm": 0.32559722661972046,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 1110
+    },
+    {
+      "epoch": 0.9423643247791333,
+      "grad_norm": 0.3098459839820862,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 1120
+    },
+    {
+      "epoch": 0.9507782919646613,
+      "grad_norm": 0.33917108178138733,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1130
+    },
+    {
+      "epoch": 0.9591922591501894,
+      "grad_norm": 0.4055837094783783,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1140
+    },
+    {
+      "epoch": 0.9676062263357172,
+      "grad_norm": 0.32508623600006104,
+      "learning_rate": 0.0002,
+      "loss": 1.8259,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9760201935212452,
+      "grad_norm": 0.30150601267814636,
+      "learning_rate": 0.0002,
+      "loss": 1.782,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9844341607067733,
+      "grad_norm": 0.3042563199996948,
+      "learning_rate": 0.0002,
+      "loss": 1.8291,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9928481278923013,
+      "grad_norm": 0.33254584670066833,
+      "learning_rate": 0.0002,
+      "loss": 1.7847,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9995793016407236,
+      "eval_loss": 1.8077726364135742,
+      "eval_runtime": 38.4359,
+      "eval_samples_per_second": 13.399,
+      "eval_steps_per_second": 1.691,
+      "step": 1188
+    },
+    {
+      "epoch": 1.0012620950778293,
+      "grad_norm": 0.35073035955429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 1190
+    },
+    {
+      "epoch": 1.0096760622633572,
+      "grad_norm": 0.3217269778251648,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1200
+    },
+    {
+      "epoch": 1.018090029448885,
+      "grad_norm": 0.3635033369064331,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1210
+    },
+    {
+      "epoch": 1.0265039966344132,
+      "grad_norm": 0.32468414306640625,
+      "learning_rate": 0.0002,
+      "loss": 1.6949,
+      "step": 1220
+    },
+    {
+      "epoch": 1.034917963819941,
+      "grad_norm": 0.3307163417339325,
+      "learning_rate": 0.0002,
+      "loss": 1.711,
+      "step": 1230
+    },
+    {
+      "epoch": 1.0433319310054692,
+      "grad_norm": 0.34381359815597534,
+      "learning_rate": 0.0002,
+      "loss": 1.7881,
+      "step": 1240
+    },
+    {
+      "epoch": 1.051745898190997,
+      "grad_norm": 0.35874804854393005,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 1250
+    },
+    {
+      "epoch": 1.060159865376525,
+      "grad_norm": 0.3615919351577759,
+      "learning_rate": 0.0002,
+      "loss": 1.7314,
+      "step": 1260
+    },
+    {
+      "epoch": 1.068573832562053,
+      "grad_norm": 0.32835808396339417,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1270
+    },
+    {
+      "epoch": 1.076987799747581,
+      "grad_norm": 0.3876388370990753,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0854017669331089,
+      "grad_norm": 0.39895930886268616,
+      "learning_rate": 0.0002,
+      "loss": 1.7442,
+      "step": 1290
+    },
+    {
+      "epoch": 1.093815734118637,
+      "grad_norm": 0.39081698656082153,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 1300
+    },
+    {
+      "epoch": 1.1022297013041649,
+      "grad_norm": 0.39974215626716614,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1310
+    },
+    {
+      "epoch": 1.110643668489693,
+      "grad_norm": 0.3887332081794739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1320
+    },
+    {
+      "epoch": 1.1190576356752209,
+      "grad_norm": 0.36216408014297485,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 1330
+    },
+    {
+      "epoch": 1.1274716028607488,
+      "grad_norm": 0.36979028582572937,
+      "learning_rate": 0.0002,
+      "loss": 1.762,
+      "step": 1340
+    },
+    {
+      "epoch": 1.1358855700462769,
+      "grad_norm": 0.34052133560180664,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 1350
+    },
+    {
+      "epoch": 1.1442995372318048,
+      "grad_norm": 0.3467716574668884,
+      "learning_rate": 0.0002,
+      "loss": 1.7513,
+      "step": 1360
+    },
+    {
+      "epoch": 1.1527135044173327,
+      "grad_norm": 0.35528799891471863,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1370
+    },
+    {
+      "epoch": 1.1611274716028608,
+      "grad_norm": 0.36282262206077576,
+      "learning_rate": 0.0002,
+      "loss": 1.794,
+      "step": 1380
+    },
+    {
+      "epoch": 1.1695414387883887,
+      "grad_norm": 0.37355899810791016,
+      "learning_rate": 0.0002,
+      "loss": 1.7731,
+      "step": 1390
+    },
+    {
+      "epoch": 1.1779554059739168,
+      "grad_norm": 0.37292736768722534,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1400
+    },
+    {
+      "epoch": 1.1863693731594447,
+      "grad_norm": 0.5892812013626099,
+      "learning_rate": 0.0002,
+      "loss": 1.6916,
+      "step": 1410
+    },
+    {
+      "epoch": 1.1947833403449726,
+      "grad_norm": 0.3712292015552521,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 1420
+    },
+    {
+      "epoch": 1.2031973075305007,
+      "grad_norm": 0.3349577486515045,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1430
+    },
+    {
+      "epoch": 1.2116112747160286,
+      "grad_norm": 0.32591062784194946,
+      "learning_rate": 0.0002,
+      "loss": 1.7412,
+      "step": 1440
+    },
+    {
+      "epoch": 1.2200252419015567,
+      "grad_norm": 0.3840635418891907,
+      "learning_rate": 0.0002,
+      "loss": 1.7406,
+      "step": 1450
+    },
+    {
+      "epoch": 1.2284392090870846,
+      "grad_norm": 0.37238365411758423,
+      "learning_rate": 0.0002,
+      "loss": 1.7276,
+      "step": 1460
+    },
+    {
+      "epoch": 1.2368531762726125,
+      "grad_norm": 0.3731217682361603,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 1470
+    },
+    {
+      "epoch": 1.2452671434581406,
+      "grad_norm": 0.3318967819213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7255,
+      "step": 1480
+    },
+    {
+      "epoch": 1.2536811106436685,
+      "grad_norm": 0.3784034848213196,
+      "learning_rate": 0.0002,
+      "loss": 1.7463,
+      "step": 1490
+    },
+    {
+      "epoch": 1.2620950778291964,
+      "grad_norm": 0.3541383147239685,
+      "learning_rate": 0.0002,
+      "loss": 1.6862,
+      "step": 1500
+    },
+    {
+      "epoch": 1.2705090450147245,
+      "grad_norm": 0.35312485694885254,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1510
+    },
+    {
+      "epoch": 1.2789230122002524,
+      "grad_norm": 0.35272929072380066,
+      "learning_rate": 0.0002,
+      "loss": 1.7029,
+      "step": 1520
+    },
+    {
+      "epoch": 1.2873369793857803,
+      "grad_norm": 0.40988272428512573,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1530
+    },
+    {
+      "epoch": 1.2957509465713084,
+      "grad_norm": 0.3543946146965027,
+      "learning_rate": 0.0002,
+      "loss": 1.6912,
+      "step": 1540
+    },
+    {
+      "epoch": 1.3041649137568363,
+      "grad_norm": 0.35639145970344543,
+      "learning_rate": 0.0002,
+      "loss": 1.6757,
+      "step": 1550
+    },
+    {
+      "epoch": 1.3125788809423642,
+      "grad_norm": 0.3290826678276062,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 1560
+    },
+    {
+      "epoch": 1.3209928481278923,
+      "grad_norm": 0.39264336228370667,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 1570
+    },
+    {
+      "epoch": 1.3294068153134202,
+      "grad_norm": 0.5390415191650391,
+      "learning_rate": 0.0002,
+      "loss": 1.6804,
+      "step": 1580
+    },
+    {
+      "epoch": 1.3378207824989483,
+      "grad_norm": 0.5188116431236267,
+      "learning_rate": 0.0002,
+      "loss": 1.708,
+      "step": 1590
+    },
+    {
+      "epoch": 1.3462347496844762,
+      "grad_norm": 0.37445148825645447,
+      "learning_rate": 0.0002,
+      "loss": 1.6763,
+      "step": 1600
+    },
+    {
+      "epoch": 1.3546487168700043,
+      "grad_norm": 0.3296085298061371,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1610
+    },
+    {
+      "epoch": 1.3630626840555322,
+      "grad_norm": 0.39879581332206726,
+      "learning_rate": 0.0002,
+      "loss": 1.8107,
+      "step": 1620
+    },
+    {
+      "epoch": 1.37147665124106,
+      "grad_norm": 0.36092764139175415,
+      "learning_rate": 0.0002,
+      "loss": 1.6744,
+      "step": 1630
+    },
+    {
+      "epoch": 1.3798906184265882,
+      "grad_norm": 0.37011823058128357,
+      "learning_rate": 0.0002,
+      "loss": 1.7144,
+      "step": 1640
+    },
+    {
+      "epoch": 1.3883045856121161,
+      "grad_norm": 0.40863534808158875,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1650
+    },
+    {
+      "epoch": 1.396718552797644,
+      "grad_norm": 0.337001770734787,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 1660
+    },
+    {
+      "epoch": 1.4051325199831721,
+      "grad_norm": 0.35596707463264465,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1670
+    },
+    {
+      "epoch": 1.4135464871687,
+      "grad_norm": 0.3857671916484833,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 1680
+    },
+    {
+      "epoch": 1.421960454354228,
+      "grad_norm": 0.419502317905426,
+      "learning_rate": 0.0002,
+      "loss": 1.7015,
+      "step": 1690
+    },
+    {
+      "epoch": 1.430374421539756,
+      "grad_norm": 0.35459452867507935,
+      "learning_rate": 0.0002,
+      "loss": 1.7261,
+      "step": 1700
+    },
+    {
+      "epoch": 1.438788388725284,
+      "grad_norm": 0.37246978282928467,
+      "learning_rate": 0.0002,
+      "loss": 1.7361,
+      "step": 1710
+    },
+    {
+      "epoch": 1.4472023559108118,
+      "grad_norm": 0.33091893792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 1720
+    },
+    {
+      "epoch": 1.45561632309634,
+      "grad_norm": 0.37029674649238586,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1730
+    },
+    {
+      "epoch": 1.4640302902818678,
+      "grad_norm": 0.374025821685791,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1740
+    },
+    {
+      "epoch": 1.472444257467396,
+      "grad_norm": 0.3416315019130707,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1750
+    },
+    {
+      "epoch": 1.4808582246529238,
+      "grad_norm": 0.36502841114997864,
+      "learning_rate": 0.0002,
+      "loss": 1.7093,
+      "step": 1760
+    },
+    {
+      "epoch": 1.489272191838452,
+      "grad_norm": 0.35458803176879883,
+      "learning_rate": 0.0002,
+      "loss": 1.6597,
+      "step": 1770
+    },
+    {
+      "epoch": 1.4976861590239798,
+      "grad_norm": 0.4462839663028717,
+      "learning_rate": 0.0002,
+      "loss": 1.675,
+      "step": 1780
+    },
+    {
+      "epoch": 1.5061001262095077,
+      "grad_norm": 0.34836092591285706,
+      "learning_rate": 0.0002,
+      "loss": 1.7267,
+      "step": 1790
+    },
+    {
+      "epoch": 1.5145140933950358,
+      "grad_norm": 0.3445749282836914,
+      "learning_rate": 0.0002,
+      "loss": 1.7295,
+      "step": 1800
+    },
+    {
+      "epoch": 1.5229280605805637,
+      "grad_norm": 0.36012160778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1810
+    },
+    {
+      "epoch": 1.5313420277660916,
+      "grad_norm": 0.4052616059780121,
+      "learning_rate": 0.0002,
+      "loss": 1.6594,
+      "step": 1820
+    },
+    {
+      "epoch": 1.5397559949516197,
+      "grad_norm": 0.3966905474662781,
+      "learning_rate": 0.0002,
+      "loss": 1.72,
+      "step": 1830
+    },
+    {
+      "epoch": 1.5481699621371476,
+      "grad_norm": 0.35028719902038574,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1840
+    },
+    {
+      "epoch": 1.5565839293226755,
+      "grad_norm": 0.3936742842197418,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1850
+    },
+    {
+      "epoch": 1.5649978965082036,
+      "grad_norm": 0.34473296999931335,
+      "learning_rate": 0.0002,
+      "loss": 1.7579,
+      "step": 1860
+    },
+    {
+      "epoch": 1.5734118636937318,
+      "grad_norm": 0.4328365623950958,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1870
+    },
+    {
+      "epoch": 1.5818258308792594,
+      "grad_norm": 0.3566315472126007,
+      "learning_rate": 0.0002,
+      "loss": 1.7098,
+      "step": 1880
+    },
+    {
+      "epoch": 1.5902397980647875,
+      "grad_norm": 0.3301256597042084,
+      "learning_rate": 0.0002,
+      "loss": 1.6095,
+      "step": 1890
+    },
+    {
+      "epoch": 1.5986537652503157,
+      "grad_norm": 0.3743041455745697,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 1900
+    },
+    {
+      "epoch": 1.6070677324358436,
+      "grad_norm": 0.3735344707965851,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 1910
+    },
+    {
+      "epoch": 1.6154816996213714,
+      "grad_norm": 0.42191144824028015,
+      "learning_rate": 0.0002,
+      "loss": 1.7445,
+      "step": 1920
+    },
+    {
+      "epoch": 1.6238956668068996,
+      "grad_norm": 0.3787207305431366,
+      "learning_rate": 0.0002,
+      "loss": 1.6978,
+      "step": 1930
+    },
+    {
+      "epoch": 1.6323096339924275,
+      "grad_norm": 0.35647350549697876,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 1940
+    },
+    {
+      "epoch": 1.6407236011779553,
+      "grad_norm": 0.39791446924209595,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 1950
+    },
+    {
+      "epoch": 1.6491375683634835,
+      "grad_norm": 0.37341275811195374,
+      "learning_rate": 0.0002,
+      "loss": 1.7293,
+      "step": 1960
+    },
+    {
+      "epoch": 1.6575515355490114,
+      "grad_norm": 0.3722686469554901,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1970
+    },
+    {
+      "epoch": 1.6659655027345392,
+      "grad_norm": 0.37467387318611145,
+      "learning_rate": 0.0002,
+      "loss": 1.6383,
+      "step": 1980
+    },
+    {
+      "epoch": 1.6743794699200674,
+      "grad_norm": 0.37109461426734924,
+      "learning_rate": 0.0002,
+      "loss": 1.7439,
+      "step": 1990
+    },
+    {
+      "epoch": 1.6827934371055953,
+      "grad_norm": 0.4008837044239044,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 2000
+    },
+    {
+      "epoch": 1.6912074042911232,
+      "grad_norm": 0.3316999673843384,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2010
+    },
+    {
+      "epoch": 1.6996213714766513,
+      "grad_norm": 0.3683805465698242,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 2020
+    },
+    {
+      "epoch": 1.7080353386621794,
+      "grad_norm": 0.4163658320903778,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 2030
+    },
+    {
+      "epoch": 1.716449305847707,
+      "grad_norm": 0.4245431125164032,
+      "learning_rate": 0.0002,
+      "loss": 1.741,
+      "step": 2040
+    },
+    {
+      "epoch": 1.7248632730332352,
+      "grad_norm": 0.36732038855552673,
+      "learning_rate": 0.0002,
+      "loss": 1.7184,
+      "step": 2050
+    },
+    {
+      "epoch": 1.7332772402187633,
+      "grad_norm": 0.34981656074523926,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 2060
+    },
+    {
+      "epoch": 1.7416912074042912,
+      "grad_norm": 0.38588812947273254,
+      "learning_rate": 0.0002,
+      "loss": 1.7545,
+      "step": 2070
+    },
+    {
+      "epoch": 1.750105174589819,
+      "grad_norm": 0.39914557337760925,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 2080
+    },
+    {
+      "epoch": 1.7585191417753472,
+      "grad_norm": 0.36068692803382874,
+      "learning_rate": 0.0002,
+      "loss": 1.7049,
+      "step": 2090
+    },
+    {
+      "epoch": 1.766933108960875,
+      "grad_norm": 0.3983287215232849,
+      "learning_rate": 0.0002,
+      "loss": 1.7537,
+      "step": 2100
+    },
+    {
+      "epoch": 1.775347076146403,
+      "grad_norm": 0.45008400082588196,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2110
+    },
+    {
+      "epoch": 1.783761043331931,
+      "grad_norm": 0.3618052303791046,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 2120
+    },
+    {
+      "epoch": 1.792175010517459,
+      "grad_norm": 0.38745400309562683,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 2130
+    },
+    {
+      "epoch": 1.8005889777029869,
+      "grad_norm": 0.3413826525211334,
+      "learning_rate": 0.0002,
+      "loss": 1.7387,
+      "step": 2140
+    },
+    {
+      "epoch": 1.809002944888515,
+      "grad_norm": 0.35983747243881226,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 2150
+    },
+    {
+      "epoch": 1.8174169120740429,
+      "grad_norm": 0.40926849842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 2160
+    },
+    {
+      "epoch": 1.8258308792595708,
+      "grad_norm": 0.3543093800544739,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2170
+    },
+    {
+      "epoch": 1.8342448464450989,
+      "grad_norm": 0.42690935730934143,
+      "learning_rate": 0.0002,
+      "loss": 1.7812,
+      "step": 2180
+    },
+    {
+      "epoch": 1.842658813630627,
+      "grad_norm": 0.40282756090164185,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 2190
+    },
+    {
+      "epoch": 1.8510727808161547,
+      "grad_norm": 0.36568400263786316,
+      "learning_rate": 0.0002,
+      "loss": 1.7411,
+      "step": 2200
+    },
+    {
+      "epoch": 1.8594867480016828,
+      "grad_norm": 0.43159013986587524,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 2210
+    },
+    {
+      "epoch": 1.867900715187211,
+      "grad_norm": 0.3554118573665619,
+      "learning_rate": 0.0002,
+      "loss": 1.7298,
+      "step": 2220
+    },
+    {
+      "epoch": 1.8763146823727388,
+      "grad_norm": 0.43349072337150574,
+      "learning_rate": 0.0002,
+      "loss": 1.7157,
+      "step": 2230
+    },
+    {
+      "epoch": 1.8847286495582667,
+      "grad_norm": 0.36486536264419556,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 2240
+    },
+    {
+      "epoch": 1.8931426167437948,
+      "grad_norm": 0.39260047674179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6901,
+      "step": 2250
+    },
+    {
+      "epoch": 1.9015565839293227,
+      "grad_norm": 0.3741776943206787,
+      "learning_rate": 0.0002,
+      "loss": 1.6691,
+      "step": 2260
+    },
+    {
+      "epoch": 1.9099705511148506,
+      "grad_norm": 0.3961946964263916,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2270
+    },
+    {
+      "epoch": 1.9183845183003787,
+      "grad_norm": 0.3659731149673462,
+      "learning_rate": 0.0002,
+      "loss": 1.737,
+      "step": 2280
+    },
+    {
+      "epoch": 1.9267984854859066,
+      "grad_norm": 0.34744107723236084,
+      "learning_rate": 0.0002,
+      "loss": 1.7342,
+      "step": 2290
+    },
+    {
+      "epoch": 1.9352124526714345,
+      "grad_norm": 0.3607442378997803,
+      "learning_rate": 0.0002,
+      "loss": 1.7162,
+      "step": 2300
+    },
+    {
+      "epoch": 1.9436264198569626,
+      "grad_norm": 0.331464558839798,
+      "learning_rate": 0.0002,
+      "loss": 1.6673,
+      "step": 2310
+    },
+    {
+      "epoch": 1.9520403870424905,
+      "grad_norm": 0.3904414474964142,
+      "learning_rate": 0.0002,
+      "loss": 1.7101,
+      "step": 2320
+    },
+    {
+      "epoch": 1.9604543542280184,
+      "grad_norm": 0.37584832310676575,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 2330
+    },
+    {
+      "epoch": 1.9688683214135465,
+      "grad_norm": 0.3698684275150299,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 2340
+    },
+    {
+      "epoch": 1.9772822885990746,
+      "grad_norm": 0.40571412444114685,
+      "learning_rate": 0.0002,
+      "loss": 1.7764,
+      "step": 2350
+    },
+    {
+      "epoch": 1.9856962557846023,
+      "grad_norm": 0.40059587359428406,
+      "learning_rate": 0.0002,
+      "loss": 1.744,
+      "step": 2360
+    },
+    {
+      "epoch": 1.9941102229701304,
+      "grad_norm": 0.4168248474597931,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2370
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8055059909820557,
+      "eval_runtime": 38.422,
+      "eval_samples_per_second": 13.404,
+      "eval_steps_per_second": 1.692,
+      "step": 2377
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 9504,
+  "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": 1.1000223768143462e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/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-0.8-num-5623-sd-42/checkpoint-2377/training_args.bin b/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-0.8-num-5623-sd-42/checkpoint-2377/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f1502d478cfbb1424f707352d007b740bde5e373
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-2377/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df2b79d3acefeedef5a0229881de39ec68ef9b40046a60d7976a49f7e6b3b936
+size 5560
diff --git a/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-0.8-num-5623-sd-42/checkpoint-3565/README.md b/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-0.8-num-5623-sd-42/checkpoint-3565/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-3565/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_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-0.8-num-5623-sd-42/checkpoint-3565/adapter_config.json b/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-0.8-num-5623-sd-42/checkpoint-3565/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-3565/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_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-0.8-num-5623-sd-42/checkpoint-3565/adapter_model.safetensors b/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-0.8-num-5623-sd-42/checkpoint-3565/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f4f410418768a7b413f16514274c7bad022e7c28
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-3565/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3da09c9b7c982009a6041e24fa2892e5e37d74df4f04aaeb442193dbdfdda14
+size 109069176
diff --git a/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-0.8-num-5623-sd-42/checkpoint-3565/optimizer.pt b/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-0.8-num-5623-sd-42/checkpoint-3565/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1ca02432285bc8788092b94c9075679d552e1e09
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-3565/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7fdfc4dca404e4173ddda2b81c3a0777828734ce1f8a4fb95728878ddd8e176
+size 55532666
diff --git a/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-0.8-num-5623-sd-42/checkpoint-3565/rng_state.pth b/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-0.8-num-5623-sd-42/checkpoint-3565/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..5a9c8e92ef5359d5c5fe05b86784ee2d1aa70e08
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-3565/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7d31c1acccbc5ba723c396072530ed85853ea77c94cb42addb003ff299ebdd9
+size 14244
diff --git a/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-0.8-num-5623-sd-42/checkpoint-3565/scheduler.pt b/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-0.8-num-5623-sd-42/checkpoint-3565/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ddb4edffd36393473f879801e11c72eab1e7bcc7
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-3565/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0a1123053227ea54b69507b378d69d72b7ad22b2d8e1e92614975b757ee988f
+size 1064
diff --git a/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-0.8-num-5623-sd-42/checkpoint-3565/special_tokens_map.json b/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-0.8-num-5623-sd-42/checkpoint-3565/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-3565/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_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-0.8-num-5623-sd-42/checkpoint-3565/tokenizer.json b/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-0.8-num-5623-sd-42/checkpoint-3565/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-3565/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_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-0.8-num-5623-sd-42/checkpoint-3565/tokenizer_config.json b/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-0.8-num-5623-sd-42/checkpoint-3565/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-3565/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_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-0.8-num-5623-sd-42/checkpoint-3565/trainer_state.json b/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-0.8-num-5623-sd-42/checkpoint-3565/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..8ad93920afd45593d9d92e25c6612506f4212e9e
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-3565/trainer_state.json
@@ -0,0 +1,2549 @@
+{
+  "best_metric": 1.8055059909820557,
+  "best_model_checkpoint": "outputs-001/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-0.8-num-5623-sd-42/checkpoint-2377",
+  "epoch": 2.9995793016407237,
+  "eval_steps": 10,
+  "global_step": 3565,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008413967185527976,
+      "grad_norm": 0.5458821654319763,
+      "learning_rate": 0.0002,
+      "loss": 2.56,
+      "step": 10
+    },
+    {
+      "epoch": 0.016827934371055953,
+      "grad_norm": 0.7293308973312378,
+      "learning_rate": 0.0002,
+      "loss": 2.3235,
+      "step": 20
+    },
+    {
+      "epoch": 0.02524190155658393,
+      "grad_norm": 0.47792306542396545,
+      "learning_rate": 0.0002,
+      "loss": 2.0815,
+      "step": 30
+    },
+    {
+      "epoch": 0.033655868742111905,
+      "grad_norm": 0.5944402813911438,
+      "learning_rate": 0.0002,
+      "loss": 1.9718,
+      "step": 40
+    },
+    {
+      "epoch": 0.04206983592763988,
+      "grad_norm": 0.5415359735488892,
+      "learning_rate": 0.0002,
+      "loss": 1.8848,
+      "step": 50
+    },
+    {
+      "epoch": 0.05048380311316786,
+      "grad_norm": 0.535713791847229,
+      "learning_rate": 0.0002,
+      "loss": 1.8953,
+      "step": 60
+    },
+    {
+      "epoch": 0.058897770298695834,
+      "grad_norm": 0.5184146761894226,
+      "learning_rate": 0.0002,
+      "loss": 1.937,
+      "step": 70
+    },
+    {
+      "epoch": 0.06731173748422381,
+      "grad_norm": 0.458926796913147,
+      "learning_rate": 0.0002,
+      "loss": 1.8396,
+      "step": 80
+    },
+    {
+      "epoch": 0.07572570466975179,
+      "grad_norm": 0.4780142307281494,
+      "learning_rate": 0.0002,
+      "loss": 1.8677,
+      "step": 90
+    },
+    {
+      "epoch": 0.08413967185527976,
+      "grad_norm": 0.79965740442276,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 100
+    },
+    {
+      "epoch": 0.09255363904080774,
+      "grad_norm": 0.4498862028121948,
+      "learning_rate": 0.0002,
+      "loss": 1.9081,
+      "step": 110
+    },
+    {
+      "epoch": 0.10096760622633572,
+      "grad_norm": 0.39338430762290955,
+      "learning_rate": 0.0002,
+      "loss": 1.8503,
+      "step": 120
+    },
+    {
+      "epoch": 0.10938157341186369,
+      "grad_norm": 0.9588953852653503,
+      "learning_rate": 0.0002,
+      "loss": 1.8637,
+      "step": 130
+    },
+    {
+      "epoch": 0.11779554059739167,
+      "grad_norm": 0.41675639152526855,
+      "learning_rate": 0.0002,
+      "loss": 1.8676,
+      "step": 140
+    },
+    {
+      "epoch": 0.12620950778291964,
+      "grad_norm": 0.44519832730293274,
+      "learning_rate": 0.0002,
+      "loss": 1.8904,
+      "step": 150
+    },
+    {
+      "epoch": 0.13462347496844762,
+      "grad_norm": 0.4176260530948639,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 160
+    },
+    {
+      "epoch": 0.1430374421539756,
+      "grad_norm": 0.35840365290641785,
+      "learning_rate": 0.0002,
+      "loss": 1.8398,
+      "step": 170
+    },
+    {
+      "epoch": 0.15145140933950357,
+      "grad_norm": 0.3794495463371277,
+      "learning_rate": 0.0002,
+      "loss": 1.8666,
+      "step": 180
+    },
+    {
+      "epoch": 0.15986537652503155,
+      "grad_norm": 0.4563522934913635,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 190
+    },
+    {
+      "epoch": 0.16827934371055953,
+      "grad_norm": 0.37057486176490784,
+      "learning_rate": 0.0002,
+      "loss": 1.8893,
+      "step": 200
+    },
+    {
+      "epoch": 0.1766933108960875,
+      "grad_norm": 0.44081518054008484,
+      "learning_rate": 0.0002,
+      "loss": 1.7995,
+      "step": 210
+    },
+    {
+      "epoch": 0.18510727808161548,
+      "grad_norm": 0.46078577637672424,
+      "learning_rate": 0.0002,
+      "loss": 1.9048,
+      "step": 220
+    },
+    {
+      "epoch": 0.19352124526714345,
+      "grad_norm": 0.36132094264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.8403,
+      "step": 230
+    },
+    {
+      "epoch": 0.20193521245267143,
+      "grad_norm": 0.3747289180755615,
+      "learning_rate": 0.0002,
+      "loss": 1.8827,
+      "step": 240
+    },
+    {
+      "epoch": 0.2103491796381994,
+      "grad_norm": 0.3540179133415222,
+      "learning_rate": 0.0002,
+      "loss": 1.8382,
+      "step": 250
+    },
+    {
+      "epoch": 0.21876314682372738,
+      "grad_norm": 0.3461375832557678,
+      "learning_rate": 0.0002,
+      "loss": 1.8196,
+      "step": 260
+    },
+    {
+      "epoch": 0.22717711400925536,
+      "grad_norm": 0.3436960279941559,
+      "learning_rate": 0.0002,
+      "loss": 1.8509,
+      "step": 270
+    },
+    {
+      "epoch": 0.23559108119478334,
+      "grad_norm": 0.35403719544410706,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 280
+    },
+    {
+      "epoch": 0.2440050483803113,
+      "grad_norm": 0.37142616510391235,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 290
+    },
+    {
+      "epoch": 0.2524190155658393,
+      "grad_norm": 0.3307955861091614,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 300
+    },
+    {
+      "epoch": 0.2608329827513673,
+      "grad_norm": 0.32855314016342163,
+      "learning_rate": 0.0002,
+      "loss": 1.817,
+      "step": 310
+    },
+    {
+      "epoch": 0.26924694993689524,
+      "grad_norm": 0.3299003839492798,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 320
+    },
+    {
+      "epoch": 0.27766091712242325,
+      "grad_norm": 0.44311287999153137,
+      "learning_rate": 0.0002,
+      "loss": 1.8129,
+      "step": 330
+    },
+    {
+      "epoch": 0.2860748843079512,
+      "grad_norm": 0.32989758253097534,
+      "learning_rate": 0.0002,
+      "loss": 1.8232,
+      "step": 340
+    },
+    {
+      "epoch": 0.2944888514934792,
+      "grad_norm": 0.34400200843811035,
+      "learning_rate": 0.0002,
+      "loss": 1.7716,
+      "step": 350
+    },
+    {
+      "epoch": 0.30290281867900715,
+      "grad_norm": 0.36286211013793945,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 360
+    },
+    {
+      "epoch": 0.31131678586453515,
+      "grad_norm": 0.406827837228775,
+      "learning_rate": 0.0002,
+      "loss": 1.8025,
+      "step": 370
+    },
+    {
+      "epoch": 0.3197307530500631,
+      "grad_norm": 0.36299195885658264,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 380
+    },
+    {
+      "epoch": 0.3281447202355911,
+      "grad_norm": 0.3477257192134857,
+      "learning_rate": 0.0002,
+      "loss": 1.837,
+      "step": 390
+    },
+    {
+      "epoch": 0.33655868742111905,
+      "grad_norm": 0.3730369210243225,
+      "learning_rate": 0.0002,
+      "loss": 1.7767,
+      "step": 400
+    },
+    {
+      "epoch": 0.34497265460664706,
+      "grad_norm": 0.4644559919834137,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 410
+    },
+    {
+      "epoch": 0.353386621792175,
+      "grad_norm": 0.406576544046402,
+      "learning_rate": 0.0002,
+      "loss": 1.7538,
+      "step": 420
+    },
+    {
+      "epoch": 0.361800588977703,
+      "grad_norm": 0.3612699508666992,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 430
+    },
+    {
+      "epoch": 0.37021455616323096,
+      "grad_norm": 0.3243742287158966,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 440
+    },
+    {
+      "epoch": 0.37862852334875896,
+      "grad_norm": 0.36671221256256104,
+      "learning_rate": 0.0002,
+      "loss": 1.8851,
+      "step": 450
+    },
+    {
+      "epoch": 0.3870424905342869,
+      "grad_norm": 0.3565002381801605,
+      "learning_rate": 0.0002,
+      "loss": 1.8853,
+      "step": 460
+    },
+    {
+      "epoch": 0.3954564577198149,
+      "grad_norm": 0.34630221128463745,
+      "learning_rate": 0.0002,
+      "loss": 1.8923,
+      "step": 470
+    },
+    {
+      "epoch": 0.40387042490534286,
+      "grad_norm": 0.3353537321090698,
+      "learning_rate": 0.0002,
+      "loss": 1.8234,
+      "step": 480
+    },
+    {
+      "epoch": 0.41228439209087087,
+      "grad_norm": 0.4015921950340271,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 490
+    },
+    {
+      "epoch": 0.4206983592763988,
+      "grad_norm": 0.5489419102668762,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 500
+    },
+    {
+      "epoch": 0.4291123264619268,
+      "grad_norm": 0.4193589985370636,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 510
+    },
+    {
+      "epoch": 0.43752629364745477,
+      "grad_norm": 0.3418922424316406,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 520
+    },
+    {
+      "epoch": 0.44594026083298277,
+      "grad_norm": 0.32668185234069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 530
+    },
+    {
+      "epoch": 0.4543542280185107,
+      "grad_norm": 0.3094325661659241,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 540
+    },
+    {
+      "epoch": 0.4627681952040387,
+      "grad_norm": 0.3743017315864563,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 550
+    },
+    {
+      "epoch": 0.47118216238956667,
+      "grad_norm": 0.3295630216598511,
+      "learning_rate": 0.0002,
+      "loss": 1.8451,
+      "step": 560
+    },
+    {
+      "epoch": 0.4795961295750947,
+      "grad_norm": 1.6124513149261475,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 570
+    },
+    {
+      "epoch": 0.4880100967606226,
+      "grad_norm": 0.3245585858821869,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 580
+    },
+    {
+      "epoch": 0.49642406394615063,
+      "grad_norm": 0.3332934081554413,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 590
+    },
+    {
+      "epoch": 0.5048380311316786,
+      "grad_norm": 0.3836138844490051,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 600
+    },
+    {
+      "epoch": 0.5132519983172066,
+      "grad_norm": 0.32953888177871704,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 610
+    },
+    {
+      "epoch": 0.5216659655027346,
+      "grad_norm": 0.36291512846946716,
+      "learning_rate": 0.0002,
+      "loss": 1.7729,
+      "step": 620
+    },
+    {
+      "epoch": 0.5300799326882625,
+      "grad_norm": 0.3237783908843994,
+      "learning_rate": 0.0002,
+      "loss": 1.7758,
+      "step": 630
+    },
+    {
+      "epoch": 0.5384938998737905,
+      "grad_norm": 0.38882696628570557,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 640
+    },
+    {
+      "epoch": 0.5469078670593185,
+      "grad_norm": 0.37821972370147705,
+      "learning_rate": 0.0002,
+      "loss": 1.8624,
+      "step": 650
+    },
+    {
+      "epoch": 0.5553218342448465,
+      "grad_norm": 0.3556285500526428,
+      "learning_rate": 0.0002,
+      "loss": 1.8075,
+      "step": 660
+    },
+    {
+      "epoch": 0.5637358014303744,
+      "grad_norm": 0.347499281167984,
+      "learning_rate": 0.0002,
+      "loss": 1.778,
+      "step": 670
+    },
+    {
+      "epoch": 0.5721497686159024,
+      "grad_norm": 0.3176489472389221,
+      "learning_rate": 0.0002,
+      "loss": 1.8066,
+      "step": 680
+    },
+    {
+      "epoch": 0.5805637358014304,
+      "grad_norm": 0.30220088362693787,
+      "learning_rate": 0.0002,
+      "loss": 1.7257,
+      "step": 690
+    },
+    {
+      "epoch": 0.5889777029869584,
+      "grad_norm": 0.3711601793766022,
+      "learning_rate": 0.0002,
+      "loss": 1.8415,
+      "step": 700
+    },
+    {
+      "epoch": 0.5973916701724863,
+      "grad_norm": 0.3311759829521179,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 710
+    },
+    {
+      "epoch": 0.6058056373580143,
+      "grad_norm": 0.34824270009994507,
+      "learning_rate": 0.0002,
+      "loss": 1.7712,
+      "step": 720
+    },
+    {
+      "epoch": 0.6142196045435423,
+      "grad_norm": 0.29668381810188293,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 730
+    },
+    {
+      "epoch": 0.6226335717290703,
+      "grad_norm": 0.36087489128112793,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 740
+    },
+    {
+      "epoch": 0.6310475389145982,
+      "grad_norm": 0.31590089201927185,
+      "learning_rate": 0.0002,
+      "loss": 1.7956,
+      "step": 750
+    },
+    {
+      "epoch": 0.6394615061001262,
+      "grad_norm": 0.37632957100868225,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 760
+    },
+    {
+      "epoch": 0.6478754732856542,
+      "grad_norm": 0.3360748589038849,
+      "learning_rate": 0.0002,
+      "loss": 1.8499,
+      "step": 770
+    },
+    {
+      "epoch": 0.6562894404711822,
+      "grad_norm": 0.3420640528202057,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 780
+    },
+    {
+      "epoch": 0.6647034076567101,
+      "grad_norm": 0.5734959244728088,
+      "learning_rate": 0.0002,
+      "loss": 1.8353,
+      "step": 790
+    },
+    {
+      "epoch": 0.6731173748422381,
+      "grad_norm": 0.36440837383270264,
+      "learning_rate": 0.0002,
+      "loss": 1.7746,
+      "step": 800
+    },
+    {
+      "epoch": 0.6815313420277661,
+      "grad_norm": 0.3179708421230316,
+      "learning_rate": 0.0002,
+      "loss": 1.7532,
+      "step": 810
+    },
+    {
+      "epoch": 0.6899453092132941,
+      "grad_norm": 0.34122881293296814,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 820
+    },
+    {
+      "epoch": 0.698359276398822,
+      "grad_norm": 0.31886112689971924,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 830
+    },
+    {
+      "epoch": 0.70677324358435,
+      "grad_norm": 0.31782326102256775,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 840
+    },
+    {
+      "epoch": 0.715187210769878,
+      "grad_norm": 0.36052989959716797,
+      "learning_rate": 0.0002,
+      "loss": 1.7588,
+      "step": 850
+    },
+    {
+      "epoch": 0.723601177955406,
+      "grad_norm": 0.28946155309677124,
+      "learning_rate": 0.0002,
+      "loss": 1.7891,
+      "step": 860
+    },
+    {
+      "epoch": 0.7320151451409339,
+      "grad_norm": 0.3095663785934448,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 870
+    },
+    {
+      "epoch": 0.7404291123264619,
+      "grad_norm": 0.3317491412162781,
+      "learning_rate": 0.0002,
+      "loss": 1.785,
+      "step": 880
+    },
+    {
+      "epoch": 0.7488430795119899,
+      "grad_norm": 0.31324660778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 890
+    },
+    {
+      "epoch": 0.7572570466975179,
+      "grad_norm": 0.3290475606918335,
+      "learning_rate": 0.0002,
+      "loss": 1.8753,
+      "step": 900
+    },
+    {
+      "epoch": 0.7656710138830458,
+      "grad_norm": 0.35690343379974365,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 910
+    },
+    {
+      "epoch": 0.7740849810685738,
+      "grad_norm": 0.39558273553848267,
+      "learning_rate": 0.0002,
+      "loss": 1.826,
+      "step": 920
+    },
+    {
+      "epoch": 0.7824989482541018,
+      "grad_norm": 0.34254348278045654,
+      "learning_rate": 0.0002,
+      "loss": 1.8722,
+      "step": 930
+    },
+    {
+      "epoch": 0.7909129154396298,
+      "grad_norm": 0.3560165464878082,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 940
+    },
+    {
+      "epoch": 0.7993268826251577,
+      "grad_norm": 0.30693164467811584,
+      "learning_rate": 0.0002,
+      "loss": 1.7992,
+      "step": 950
+    },
+    {
+      "epoch": 0.8077408498106857,
+      "grad_norm": 0.3394823372364044,
+      "learning_rate": 0.0002,
+      "loss": 1.8029,
+      "step": 960
+    },
+    {
+      "epoch": 0.8161548169962137,
+      "grad_norm": 0.3741514980792999,
+      "learning_rate": 0.0002,
+      "loss": 1.8105,
+      "step": 970
+    },
+    {
+      "epoch": 0.8245687841817417,
+      "grad_norm": 0.3655228316783905,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 980
+    },
+    {
+      "epoch": 0.8329827513672696,
+      "grad_norm": 0.3586033880710602,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 990
+    },
+    {
+      "epoch": 0.8413967185527976,
+      "grad_norm": 0.3459678888320923,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8498106857383256,
+      "grad_norm": 0.3184349834918976,
+      "learning_rate": 0.0002,
+      "loss": 1.8498,
+      "step": 1010
+    },
+    {
+      "epoch": 0.8582246529238536,
+      "grad_norm": 0.3099786043167114,
+      "learning_rate": 0.0002,
+      "loss": 1.7632,
+      "step": 1020
+    },
+    {
+      "epoch": 0.8666386201093815,
+      "grad_norm": 0.30300915241241455,
+      "learning_rate": 0.0002,
+      "loss": 1.8067,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8750525872949095,
+      "grad_norm": 0.3128705620765686,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8834665544804375,
+      "grad_norm": 0.3336263597011566,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8918805216659655,
+      "grad_norm": 0.3801328241825104,
+      "learning_rate": 0.0002,
+      "loss": 1.8375,
+      "step": 1060
+    },
+    {
+      "epoch": 0.9002944888514934,
+      "grad_norm": 0.3122096359729767,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 1070
+    },
+    {
+      "epoch": 0.9087084560370214,
+      "grad_norm": 0.35990869998931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 1080
+    },
+    {
+      "epoch": 0.9171224232225494,
+      "grad_norm": 0.3321819305419922,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1090
+    },
+    {
+      "epoch": 0.9255363904080774,
+      "grad_norm": 0.4202139377593994,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9339503575936053,
+      "grad_norm": 0.32559722661972046,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 1110
+    },
+    {
+      "epoch": 0.9423643247791333,
+      "grad_norm": 0.3098459839820862,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 1120
+    },
+    {
+      "epoch": 0.9507782919646613,
+      "grad_norm": 0.33917108178138733,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1130
+    },
+    {
+      "epoch": 0.9591922591501894,
+      "grad_norm": 0.4055837094783783,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1140
+    },
+    {
+      "epoch": 0.9676062263357172,
+      "grad_norm": 0.32508623600006104,
+      "learning_rate": 0.0002,
+      "loss": 1.8259,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9760201935212452,
+      "grad_norm": 0.30150601267814636,
+      "learning_rate": 0.0002,
+      "loss": 1.782,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9844341607067733,
+      "grad_norm": 0.3042563199996948,
+      "learning_rate": 0.0002,
+      "loss": 1.8291,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9928481278923013,
+      "grad_norm": 0.33254584670066833,
+      "learning_rate": 0.0002,
+      "loss": 1.7847,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9995793016407236,
+      "eval_loss": 1.8077726364135742,
+      "eval_runtime": 38.4359,
+      "eval_samples_per_second": 13.399,
+      "eval_steps_per_second": 1.691,
+      "step": 1188
+    },
+    {
+      "epoch": 1.0012620950778293,
+      "grad_norm": 0.35073035955429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 1190
+    },
+    {
+      "epoch": 1.0096760622633572,
+      "grad_norm": 0.3217269778251648,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1200
+    },
+    {
+      "epoch": 1.018090029448885,
+      "grad_norm": 0.3635033369064331,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1210
+    },
+    {
+      "epoch": 1.0265039966344132,
+      "grad_norm": 0.32468414306640625,
+      "learning_rate": 0.0002,
+      "loss": 1.6949,
+      "step": 1220
+    },
+    {
+      "epoch": 1.034917963819941,
+      "grad_norm": 0.3307163417339325,
+      "learning_rate": 0.0002,
+      "loss": 1.711,
+      "step": 1230
+    },
+    {
+      "epoch": 1.0433319310054692,
+      "grad_norm": 0.34381359815597534,
+      "learning_rate": 0.0002,
+      "loss": 1.7881,
+      "step": 1240
+    },
+    {
+      "epoch": 1.051745898190997,
+      "grad_norm": 0.35874804854393005,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 1250
+    },
+    {
+      "epoch": 1.060159865376525,
+      "grad_norm": 0.3615919351577759,
+      "learning_rate": 0.0002,
+      "loss": 1.7314,
+      "step": 1260
+    },
+    {
+      "epoch": 1.068573832562053,
+      "grad_norm": 0.32835808396339417,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1270
+    },
+    {
+      "epoch": 1.076987799747581,
+      "grad_norm": 0.3876388370990753,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0854017669331089,
+      "grad_norm": 0.39895930886268616,
+      "learning_rate": 0.0002,
+      "loss": 1.7442,
+      "step": 1290
+    },
+    {
+      "epoch": 1.093815734118637,
+      "grad_norm": 0.39081698656082153,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 1300
+    },
+    {
+      "epoch": 1.1022297013041649,
+      "grad_norm": 0.39974215626716614,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1310
+    },
+    {
+      "epoch": 1.110643668489693,
+      "grad_norm": 0.3887332081794739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1320
+    },
+    {
+      "epoch": 1.1190576356752209,
+      "grad_norm": 0.36216408014297485,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 1330
+    },
+    {
+      "epoch": 1.1274716028607488,
+      "grad_norm": 0.36979028582572937,
+      "learning_rate": 0.0002,
+      "loss": 1.762,
+      "step": 1340
+    },
+    {
+      "epoch": 1.1358855700462769,
+      "grad_norm": 0.34052133560180664,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 1350
+    },
+    {
+      "epoch": 1.1442995372318048,
+      "grad_norm": 0.3467716574668884,
+      "learning_rate": 0.0002,
+      "loss": 1.7513,
+      "step": 1360
+    },
+    {
+      "epoch": 1.1527135044173327,
+      "grad_norm": 0.35528799891471863,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1370
+    },
+    {
+      "epoch": 1.1611274716028608,
+      "grad_norm": 0.36282262206077576,
+      "learning_rate": 0.0002,
+      "loss": 1.794,
+      "step": 1380
+    },
+    {
+      "epoch": 1.1695414387883887,
+      "grad_norm": 0.37355899810791016,
+      "learning_rate": 0.0002,
+      "loss": 1.7731,
+      "step": 1390
+    },
+    {
+      "epoch": 1.1779554059739168,
+      "grad_norm": 0.37292736768722534,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1400
+    },
+    {
+      "epoch": 1.1863693731594447,
+      "grad_norm": 0.5892812013626099,
+      "learning_rate": 0.0002,
+      "loss": 1.6916,
+      "step": 1410
+    },
+    {
+      "epoch": 1.1947833403449726,
+      "grad_norm": 0.3712292015552521,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 1420
+    },
+    {
+      "epoch": 1.2031973075305007,
+      "grad_norm": 0.3349577486515045,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1430
+    },
+    {
+      "epoch": 1.2116112747160286,
+      "grad_norm": 0.32591062784194946,
+      "learning_rate": 0.0002,
+      "loss": 1.7412,
+      "step": 1440
+    },
+    {
+      "epoch": 1.2200252419015567,
+      "grad_norm": 0.3840635418891907,
+      "learning_rate": 0.0002,
+      "loss": 1.7406,
+      "step": 1450
+    },
+    {
+      "epoch": 1.2284392090870846,
+      "grad_norm": 0.37238365411758423,
+      "learning_rate": 0.0002,
+      "loss": 1.7276,
+      "step": 1460
+    },
+    {
+      "epoch": 1.2368531762726125,
+      "grad_norm": 0.3731217682361603,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 1470
+    },
+    {
+      "epoch": 1.2452671434581406,
+      "grad_norm": 0.3318967819213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7255,
+      "step": 1480
+    },
+    {
+      "epoch": 1.2536811106436685,
+      "grad_norm": 0.3784034848213196,
+      "learning_rate": 0.0002,
+      "loss": 1.7463,
+      "step": 1490
+    },
+    {
+      "epoch": 1.2620950778291964,
+      "grad_norm": 0.3541383147239685,
+      "learning_rate": 0.0002,
+      "loss": 1.6862,
+      "step": 1500
+    },
+    {
+      "epoch": 1.2705090450147245,
+      "grad_norm": 0.35312485694885254,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1510
+    },
+    {
+      "epoch": 1.2789230122002524,
+      "grad_norm": 0.35272929072380066,
+      "learning_rate": 0.0002,
+      "loss": 1.7029,
+      "step": 1520
+    },
+    {
+      "epoch": 1.2873369793857803,
+      "grad_norm": 0.40988272428512573,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1530
+    },
+    {
+      "epoch": 1.2957509465713084,
+      "grad_norm": 0.3543946146965027,
+      "learning_rate": 0.0002,
+      "loss": 1.6912,
+      "step": 1540
+    },
+    {
+      "epoch": 1.3041649137568363,
+      "grad_norm": 0.35639145970344543,
+      "learning_rate": 0.0002,
+      "loss": 1.6757,
+      "step": 1550
+    },
+    {
+      "epoch": 1.3125788809423642,
+      "grad_norm": 0.3290826678276062,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 1560
+    },
+    {
+      "epoch": 1.3209928481278923,
+      "grad_norm": 0.39264336228370667,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 1570
+    },
+    {
+      "epoch": 1.3294068153134202,
+      "grad_norm": 0.5390415191650391,
+      "learning_rate": 0.0002,
+      "loss": 1.6804,
+      "step": 1580
+    },
+    {
+      "epoch": 1.3378207824989483,
+      "grad_norm": 0.5188116431236267,
+      "learning_rate": 0.0002,
+      "loss": 1.708,
+      "step": 1590
+    },
+    {
+      "epoch": 1.3462347496844762,
+      "grad_norm": 0.37445148825645447,
+      "learning_rate": 0.0002,
+      "loss": 1.6763,
+      "step": 1600
+    },
+    {
+      "epoch": 1.3546487168700043,
+      "grad_norm": 0.3296085298061371,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1610
+    },
+    {
+      "epoch": 1.3630626840555322,
+      "grad_norm": 0.39879581332206726,
+      "learning_rate": 0.0002,
+      "loss": 1.8107,
+      "step": 1620
+    },
+    {
+      "epoch": 1.37147665124106,
+      "grad_norm": 0.36092764139175415,
+      "learning_rate": 0.0002,
+      "loss": 1.6744,
+      "step": 1630
+    },
+    {
+      "epoch": 1.3798906184265882,
+      "grad_norm": 0.37011823058128357,
+      "learning_rate": 0.0002,
+      "loss": 1.7144,
+      "step": 1640
+    },
+    {
+      "epoch": 1.3883045856121161,
+      "grad_norm": 0.40863534808158875,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1650
+    },
+    {
+      "epoch": 1.396718552797644,
+      "grad_norm": 0.337001770734787,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 1660
+    },
+    {
+      "epoch": 1.4051325199831721,
+      "grad_norm": 0.35596707463264465,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1670
+    },
+    {
+      "epoch": 1.4135464871687,
+      "grad_norm": 0.3857671916484833,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 1680
+    },
+    {
+      "epoch": 1.421960454354228,
+      "grad_norm": 0.419502317905426,
+      "learning_rate": 0.0002,
+      "loss": 1.7015,
+      "step": 1690
+    },
+    {
+      "epoch": 1.430374421539756,
+      "grad_norm": 0.35459452867507935,
+      "learning_rate": 0.0002,
+      "loss": 1.7261,
+      "step": 1700
+    },
+    {
+      "epoch": 1.438788388725284,
+      "grad_norm": 0.37246978282928467,
+      "learning_rate": 0.0002,
+      "loss": 1.7361,
+      "step": 1710
+    },
+    {
+      "epoch": 1.4472023559108118,
+      "grad_norm": 0.33091893792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 1720
+    },
+    {
+      "epoch": 1.45561632309634,
+      "grad_norm": 0.37029674649238586,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1730
+    },
+    {
+      "epoch": 1.4640302902818678,
+      "grad_norm": 0.374025821685791,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1740
+    },
+    {
+      "epoch": 1.472444257467396,
+      "grad_norm": 0.3416315019130707,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1750
+    },
+    {
+      "epoch": 1.4808582246529238,
+      "grad_norm": 0.36502841114997864,
+      "learning_rate": 0.0002,
+      "loss": 1.7093,
+      "step": 1760
+    },
+    {
+      "epoch": 1.489272191838452,
+      "grad_norm": 0.35458803176879883,
+      "learning_rate": 0.0002,
+      "loss": 1.6597,
+      "step": 1770
+    },
+    {
+      "epoch": 1.4976861590239798,
+      "grad_norm": 0.4462839663028717,
+      "learning_rate": 0.0002,
+      "loss": 1.675,
+      "step": 1780
+    },
+    {
+      "epoch": 1.5061001262095077,
+      "grad_norm": 0.34836092591285706,
+      "learning_rate": 0.0002,
+      "loss": 1.7267,
+      "step": 1790
+    },
+    {
+      "epoch": 1.5145140933950358,
+      "grad_norm": 0.3445749282836914,
+      "learning_rate": 0.0002,
+      "loss": 1.7295,
+      "step": 1800
+    },
+    {
+      "epoch": 1.5229280605805637,
+      "grad_norm": 0.36012160778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1810
+    },
+    {
+      "epoch": 1.5313420277660916,
+      "grad_norm": 0.4052616059780121,
+      "learning_rate": 0.0002,
+      "loss": 1.6594,
+      "step": 1820
+    },
+    {
+      "epoch": 1.5397559949516197,
+      "grad_norm": 0.3966905474662781,
+      "learning_rate": 0.0002,
+      "loss": 1.72,
+      "step": 1830
+    },
+    {
+      "epoch": 1.5481699621371476,
+      "grad_norm": 0.35028719902038574,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1840
+    },
+    {
+      "epoch": 1.5565839293226755,
+      "grad_norm": 0.3936742842197418,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1850
+    },
+    {
+      "epoch": 1.5649978965082036,
+      "grad_norm": 0.34473296999931335,
+      "learning_rate": 0.0002,
+      "loss": 1.7579,
+      "step": 1860
+    },
+    {
+      "epoch": 1.5734118636937318,
+      "grad_norm": 0.4328365623950958,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1870
+    },
+    {
+      "epoch": 1.5818258308792594,
+      "grad_norm": 0.3566315472126007,
+      "learning_rate": 0.0002,
+      "loss": 1.7098,
+      "step": 1880
+    },
+    {
+      "epoch": 1.5902397980647875,
+      "grad_norm": 0.3301256597042084,
+      "learning_rate": 0.0002,
+      "loss": 1.6095,
+      "step": 1890
+    },
+    {
+      "epoch": 1.5986537652503157,
+      "grad_norm": 0.3743041455745697,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 1900
+    },
+    {
+      "epoch": 1.6070677324358436,
+      "grad_norm": 0.3735344707965851,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 1910
+    },
+    {
+      "epoch": 1.6154816996213714,
+      "grad_norm": 0.42191144824028015,
+      "learning_rate": 0.0002,
+      "loss": 1.7445,
+      "step": 1920
+    },
+    {
+      "epoch": 1.6238956668068996,
+      "grad_norm": 0.3787207305431366,
+      "learning_rate": 0.0002,
+      "loss": 1.6978,
+      "step": 1930
+    },
+    {
+      "epoch": 1.6323096339924275,
+      "grad_norm": 0.35647350549697876,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 1940
+    },
+    {
+      "epoch": 1.6407236011779553,
+      "grad_norm": 0.39791446924209595,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 1950
+    },
+    {
+      "epoch": 1.6491375683634835,
+      "grad_norm": 0.37341275811195374,
+      "learning_rate": 0.0002,
+      "loss": 1.7293,
+      "step": 1960
+    },
+    {
+      "epoch": 1.6575515355490114,
+      "grad_norm": 0.3722686469554901,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1970
+    },
+    {
+      "epoch": 1.6659655027345392,
+      "grad_norm": 0.37467387318611145,
+      "learning_rate": 0.0002,
+      "loss": 1.6383,
+      "step": 1980
+    },
+    {
+      "epoch": 1.6743794699200674,
+      "grad_norm": 0.37109461426734924,
+      "learning_rate": 0.0002,
+      "loss": 1.7439,
+      "step": 1990
+    },
+    {
+      "epoch": 1.6827934371055953,
+      "grad_norm": 0.4008837044239044,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 2000
+    },
+    {
+      "epoch": 1.6912074042911232,
+      "grad_norm": 0.3316999673843384,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2010
+    },
+    {
+      "epoch": 1.6996213714766513,
+      "grad_norm": 0.3683805465698242,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 2020
+    },
+    {
+      "epoch": 1.7080353386621794,
+      "grad_norm": 0.4163658320903778,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 2030
+    },
+    {
+      "epoch": 1.716449305847707,
+      "grad_norm": 0.4245431125164032,
+      "learning_rate": 0.0002,
+      "loss": 1.741,
+      "step": 2040
+    },
+    {
+      "epoch": 1.7248632730332352,
+      "grad_norm": 0.36732038855552673,
+      "learning_rate": 0.0002,
+      "loss": 1.7184,
+      "step": 2050
+    },
+    {
+      "epoch": 1.7332772402187633,
+      "grad_norm": 0.34981656074523926,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 2060
+    },
+    {
+      "epoch": 1.7416912074042912,
+      "grad_norm": 0.38588812947273254,
+      "learning_rate": 0.0002,
+      "loss": 1.7545,
+      "step": 2070
+    },
+    {
+      "epoch": 1.750105174589819,
+      "grad_norm": 0.39914557337760925,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 2080
+    },
+    {
+      "epoch": 1.7585191417753472,
+      "grad_norm": 0.36068692803382874,
+      "learning_rate": 0.0002,
+      "loss": 1.7049,
+      "step": 2090
+    },
+    {
+      "epoch": 1.766933108960875,
+      "grad_norm": 0.3983287215232849,
+      "learning_rate": 0.0002,
+      "loss": 1.7537,
+      "step": 2100
+    },
+    {
+      "epoch": 1.775347076146403,
+      "grad_norm": 0.45008400082588196,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2110
+    },
+    {
+      "epoch": 1.783761043331931,
+      "grad_norm": 0.3618052303791046,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 2120
+    },
+    {
+      "epoch": 1.792175010517459,
+      "grad_norm": 0.38745400309562683,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 2130
+    },
+    {
+      "epoch": 1.8005889777029869,
+      "grad_norm": 0.3413826525211334,
+      "learning_rate": 0.0002,
+      "loss": 1.7387,
+      "step": 2140
+    },
+    {
+      "epoch": 1.809002944888515,
+      "grad_norm": 0.35983747243881226,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 2150
+    },
+    {
+      "epoch": 1.8174169120740429,
+      "grad_norm": 0.40926849842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 2160
+    },
+    {
+      "epoch": 1.8258308792595708,
+      "grad_norm": 0.3543093800544739,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2170
+    },
+    {
+      "epoch": 1.8342448464450989,
+      "grad_norm": 0.42690935730934143,
+      "learning_rate": 0.0002,
+      "loss": 1.7812,
+      "step": 2180
+    },
+    {
+      "epoch": 1.842658813630627,
+      "grad_norm": 0.40282756090164185,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 2190
+    },
+    {
+      "epoch": 1.8510727808161547,
+      "grad_norm": 0.36568400263786316,
+      "learning_rate": 0.0002,
+      "loss": 1.7411,
+      "step": 2200
+    },
+    {
+      "epoch": 1.8594867480016828,
+      "grad_norm": 0.43159013986587524,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 2210
+    },
+    {
+      "epoch": 1.867900715187211,
+      "grad_norm": 0.3554118573665619,
+      "learning_rate": 0.0002,
+      "loss": 1.7298,
+      "step": 2220
+    },
+    {
+      "epoch": 1.8763146823727388,
+      "grad_norm": 0.43349072337150574,
+      "learning_rate": 0.0002,
+      "loss": 1.7157,
+      "step": 2230
+    },
+    {
+      "epoch": 1.8847286495582667,
+      "grad_norm": 0.36486536264419556,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 2240
+    },
+    {
+      "epoch": 1.8931426167437948,
+      "grad_norm": 0.39260047674179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6901,
+      "step": 2250
+    },
+    {
+      "epoch": 1.9015565839293227,
+      "grad_norm": 0.3741776943206787,
+      "learning_rate": 0.0002,
+      "loss": 1.6691,
+      "step": 2260
+    },
+    {
+      "epoch": 1.9099705511148506,
+      "grad_norm": 0.3961946964263916,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2270
+    },
+    {
+      "epoch": 1.9183845183003787,
+      "grad_norm": 0.3659731149673462,
+      "learning_rate": 0.0002,
+      "loss": 1.737,
+      "step": 2280
+    },
+    {
+      "epoch": 1.9267984854859066,
+      "grad_norm": 0.34744107723236084,
+      "learning_rate": 0.0002,
+      "loss": 1.7342,
+      "step": 2290
+    },
+    {
+      "epoch": 1.9352124526714345,
+      "grad_norm": 0.3607442378997803,
+      "learning_rate": 0.0002,
+      "loss": 1.7162,
+      "step": 2300
+    },
+    {
+      "epoch": 1.9436264198569626,
+      "grad_norm": 0.331464558839798,
+      "learning_rate": 0.0002,
+      "loss": 1.6673,
+      "step": 2310
+    },
+    {
+      "epoch": 1.9520403870424905,
+      "grad_norm": 0.3904414474964142,
+      "learning_rate": 0.0002,
+      "loss": 1.7101,
+      "step": 2320
+    },
+    {
+      "epoch": 1.9604543542280184,
+      "grad_norm": 0.37584832310676575,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 2330
+    },
+    {
+      "epoch": 1.9688683214135465,
+      "grad_norm": 0.3698684275150299,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 2340
+    },
+    {
+      "epoch": 1.9772822885990746,
+      "grad_norm": 0.40571412444114685,
+      "learning_rate": 0.0002,
+      "loss": 1.7764,
+      "step": 2350
+    },
+    {
+      "epoch": 1.9856962557846023,
+      "grad_norm": 0.40059587359428406,
+      "learning_rate": 0.0002,
+      "loss": 1.744,
+      "step": 2360
+    },
+    {
+      "epoch": 1.9941102229701304,
+      "grad_norm": 0.4168248474597931,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2370
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8055059909820557,
+      "eval_runtime": 38.422,
+      "eval_samples_per_second": 13.404,
+      "eval_steps_per_second": 1.692,
+      "step": 2377
+    },
+    {
+      "epoch": 2.0025241901556585,
+      "grad_norm": 0.35205352306365967,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 2380
+    },
+    {
+      "epoch": 2.010938157341186,
+      "grad_norm": 0.3979377746582031,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 2390
+    },
+    {
+      "epoch": 2.0193521245267143,
+      "grad_norm": 0.396491676568985,
+      "learning_rate": 0.0002,
+      "loss": 1.6421,
+      "step": 2400
+    },
+    {
+      "epoch": 2.0277660917122424,
+      "grad_norm": 0.44712209701538086,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2410
+    },
+    {
+      "epoch": 2.03618005889777,
+      "grad_norm": 0.4454420208930969,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 2420
+    },
+    {
+      "epoch": 2.044594026083298,
+      "grad_norm": 0.4170038402080536,
+      "learning_rate": 0.0002,
+      "loss": 1.6635,
+      "step": 2430
+    },
+    {
+      "epoch": 2.0530079932688263,
+      "grad_norm": 0.4309595227241516,
+      "learning_rate": 0.0002,
+      "loss": 1.6512,
+      "step": 2440
+    },
+    {
+      "epoch": 2.0614219604543544,
+      "grad_norm": 0.4241602122783661,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 2450
+    },
+    {
+      "epoch": 2.069835927639882,
+      "grad_norm": 0.4370540678501129,
+      "learning_rate": 0.0002,
+      "loss": 1.6162,
+      "step": 2460
+    },
+    {
+      "epoch": 2.0782498948254102,
+      "grad_norm": 0.43985554575920105,
+      "learning_rate": 0.0002,
+      "loss": 1.6354,
+      "step": 2470
+    },
+    {
+      "epoch": 2.0866638620109383,
+      "grad_norm": 0.4158105254173279,
+      "learning_rate": 0.0002,
+      "loss": 1.6954,
+      "step": 2480
+    },
+    {
+      "epoch": 2.095077829196466,
+      "grad_norm": 0.441549152135849,
+      "learning_rate": 0.0002,
+      "loss": 1.6114,
+      "step": 2490
+    },
+    {
+      "epoch": 2.103491796381994,
+      "grad_norm": 0.385718435049057,
+      "learning_rate": 0.0002,
+      "loss": 1.5485,
+      "step": 2500
+    },
+    {
+      "epoch": 2.1119057635675222,
+      "grad_norm": 0.43146514892578125,
+      "learning_rate": 0.0002,
+      "loss": 1.5894,
+      "step": 2510
+    },
+    {
+      "epoch": 2.12031973075305,
+      "grad_norm": 0.41663315892219543,
+      "learning_rate": 0.0002,
+      "loss": 1.6414,
+      "step": 2520
+    },
+    {
+      "epoch": 2.128733697938578,
+      "grad_norm": 0.4410698115825653,
+      "learning_rate": 0.0002,
+      "loss": 1.6527,
+      "step": 2530
+    },
+    {
+      "epoch": 2.137147665124106,
+      "grad_norm": 0.4472278952598572,
+      "learning_rate": 0.0002,
+      "loss": 1.6124,
+      "step": 2540
+    },
+    {
+      "epoch": 2.145561632309634,
+      "grad_norm": 0.3879167437553406,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2550
+    },
+    {
+      "epoch": 2.153975599495162,
+      "grad_norm": 0.4212203025817871,
+      "learning_rate": 0.0002,
+      "loss": 1.6682,
+      "step": 2560
+    },
+    {
+      "epoch": 2.16238956668069,
+      "grad_norm": 0.42841723561286926,
+      "learning_rate": 0.0002,
+      "loss": 1.6036,
+      "step": 2570
+    },
+    {
+      "epoch": 2.1708035338662177,
+      "grad_norm": 0.39272481203079224,
+      "learning_rate": 0.0002,
+      "loss": 1.5962,
+      "step": 2580
+    },
+    {
+      "epoch": 2.179217501051746,
+      "grad_norm": 0.4075261354446411,
+      "learning_rate": 0.0002,
+      "loss": 1.681,
+      "step": 2590
+    },
+    {
+      "epoch": 2.187631468237274,
+      "grad_norm": 0.5358437895774841,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 2600
+    },
+    {
+      "epoch": 2.1960454354228016,
+      "grad_norm": 0.4738350212574005,
+      "learning_rate": 0.0002,
+      "loss": 1.6423,
+      "step": 2610
+    },
+    {
+      "epoch": 2.2044594026083297,
+      "grad_norm": 0.446789026260376,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 2620
+    },
+    {
+      "epoch": 2.212873369793858,
+      "grad_norm": 0.4615374505519867,
+      "learning_rate": 0.0002,
+      "loss": 1.6246,
+      "step": 2630
+    },
+    {
+      "epoch": 2.221287336979386,
+      "grad_norm": 0.46901994943618774,
+      "learning_rate": 0.0002,
+      "loss": 1.6205,
+      "step": 2640
+    },
+    {
+      "epoch": 2.2297013041649136,
+      "grad_norm": 0.46267789602279663,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 2.2381152713504417,
+      "grad_norm": 0.4383080005645752,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2660
+    },
+    {
+      "epoch": 2.24652923853597,
+      "grad_norm": 0.4070609509944916,
+      "learning_rate": 0.0002,
+      "loss": 1.5745,
+      "step": 2670
+    },
+    {
+      "epoch": 2.2549432057214975,
+      "grad_norm": 0.4572339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.6125,
+      "step": 2680
+    },
+    {
+      "epoch": 2.2633571729070256,
+      "grad_norm": 0.393265038728714,
+      "learning_rate": 0.0002,
+      "loss": 1.5671,
+      "step": 2690
+    },
+    {
+      "epoch": 2.2717711400925538,
+      "grad_norm": 0.46144717931747437,
+      "learning_rate": 0.0002,
+      "loss": 1.6239,
+      "step": 2700
+    },
+    {
+      "epoch": 2.2801851072780814,
+      "grad_norm": 0.45077767968177795,
+      "learning_rate": 0.0002,
+      "loss": 1.5992,
+      "step": 2710
+    },
+    {
+      "epoch": 2.2885990744636096,
+      "grad_norm": 0.5697639584541321,
+      "learning_rate": 0.0002,
+      "loss": 1.6261,
+      "step": 2720
+    },
+    {
+      "epoch": 2.2970130416491377,
+      "grad_norm": 0.4855510890483856,
+      "learning_rate": 0.0002,
+      "loss": 1.6192,
+      "step": 2730
+    },
+    {
+      "epoch": 2.3054270088346653,
+      "grad_norm": 0.4440622627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 2740
+    },
+    {
+      "epoch": 2.3138409760201935,
+      "grad_norm": 0.3904096782207489,
+      "learning_rate": 0.0002,
+      "loss": 1.6496,
+      "step": 2750
+    },
+    {
+      "epoch": 2.3222549432057216,
+      "grad_norm": 0.5225510597229004,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 2760
+    },
+    {
+      "epoch": 2.3306689103912497,
+      "grad_norm": 0.44866397976875305,
+      "learning_rate": 0.0002,
+      "loss": 1.6082,
+      "step": 2770
+    },
+    {
+      "epoch": 2.3390828775767774,
+      "grad_norm": 0.5167056322097778,
+      "learning_rate": 0.0002,
+      "loss": 1.6087,
+      "step": 2780
+    },
+    {
+      "epoch": 2.3474968447623055,
+      "grad_norm": 0.45913267135620117,
+      "learning_rate": 0.0002,
+      "loss": 1.6136,
+      "step": 2790
+    },
+    {
+      "epoch": 2.3559108119478336,
+      "grad_norm": 0.45787590742111206,
+      "learning_rate": 0.0002,
+      "loss": 1.6564,
+      "step": 2800
+    },
+    {
+      "epoch": 2.3643247791333613,
+      "grad_norm": 0.4633352756500244,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2810
+    },
+    {
+      "epoch": 2.3727387463188894,
+      "grad_norm": 0.46390071511268616,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 2820
+    },
+    {
+      "epoch": 2.3811527135044175,
+      "grad_norm": 0.4261005222797394,
+      "learning_rate": 0.0002,
+      "loss": 1.6039,
+      "step": 2830
+    },
+    {
+      "epoch": 2.389566680689945,
+      "grad_norm": 0.4283634424209595,
+      "learning_rate": 0.0002,
+      "loss": 1.6364,
+      "step": 2840
+    },
+    {
+      "epoch": 2.3979806478754733,
+      "grad_norm": 0.4955291450023651,
+      "learning_rate": 0.0002,
+      "loss": 1.6382,
+      "step": 2850
+    },
+    {
+      "epoch": 2.4063946150610014,
+      "grad_norm": 0.4740189015865326,
+      "learning_rate": 0.0002,
+      "loss": 1.6173,
+      "step": 2860
+    },
+    {
+      "epoch": 2.414808582246529,
+      "grad_norm": 0.4222276508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2870
+    },
+    {
+      "epoch": 2.423222549432057,
+      "grad_norm": 0.4982149004936218,
+      "learning_rate": 0.0002,
+      "loss": 1.5602,
+      "step": 2880
+    },
+    {
+      "epoch": 2.4316365166175853,
+      "grad_norm": 0.5217409133911133,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 2890
+    },
+    {
+      "epoch": 2.4400504838031134,
+      "grad_norm": 0.4555884897708893,
+      "learning_rate": 0.0002,
+      "loss": 1.5804,
+      "step": 2900
+    },
+    {
+      "epoch": 2.448464450988641,
+      "grad_norm": 0.43178579211235046,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2910
+    },
+    {
+      "epoch": 2.456878418174169,
+      "grad_norm": 0.4788478910923004,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2920
+    },
+    {
+      "epoch": 2.465292385359697,
+      "grad_norm": 0.43689873814582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 2930
+    },
+    {
+      "epoch": 2.473706352545225,
+      "grad_norm": 0.5115197896957397,
+      "learning_rate": 0.0002,
+      "loss": 1.6196,
+      "step": 2940
+    },
+    {
+      "epoch": 2.482120319730753,
+      "grad_norm": 0.5290159583091736,
+      "learning_rate": 0.0002,
+      "loss": 1.689,
+      "step": 2950
+    },
+    {
+      "epoch": 2.490534286916281,
+      "grad_norm": 0.46042463183403015,
+      "learning_rate": 0.0002,
+      "loss": 1.6499,
+      "step": 2960
+    },
+    {
+      "epoch": 2.498948254101809,
+      "grad_norm": 0.4359915852546692,
+      "learning_rate": 0.0002,
+      "loss": 1.6664,
+      "step": 2970
+    },
+    {
+      "epoch": 2.507362221287337,
+      "grad_norm": 0.46352964639663696,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 2980
+    },
+    {
+      "epoch": 2.515776188472865,
+      "grad_norm": 0.5324268341064453,
+      "learning_rate": 0.0002,
+      "loss": 1.6501,
+      "step": 2990
+    },
+    {
+      "epoch": 2.5241901556583928,
+      "grad_norm": 0.5929607152938843,
+      "learning_rate": 0.0002,
+      "loss": 1.6115,
+      "step": 3000
+    },
+    {
+      "epoch": 2.532604122843921,
+      "grad_norm": 0.4811333417892456,
+      "learning_rate": 0.0002,
+      "loss": 1.6772,
+      "step": 3010
+    },
+    {
+      "epoch": 2.541018090029449,
+      "grad_norm": 0.4662701487541199,
+      "learning_rate": 0.0002,
+      "loss": 1.7023,
+      "step": 3020
+    },
+    {
+      "epoch": 2.549432057214977,
+      "grad_norm": 0.4582270681858063,
+      "learning_rate": 0.0002,
+      "loss": 1.5426,
+      "step": 3030
+    },
+    {
+      "epoch": 2.557846024400505,
+      "grad_norm": 0.4679982662200928,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 3040
+    },
+    {
+      "epoch": 2.566259991586033,
+      "grad_norm": 0.4380294680595398,
+      "learning_rate": 0.0002,
+      "loss": 1.5442,
+      "step": 3050
+    },
+    {
+      "epoch": 2.5746739587715606,
+      "grad_norm": 0.44295763969421387,
+      "learning_rate": 0.0002,
+      "loss": 1.6055,
+      "step": 3060
+    },
+    {
+      "epoch": 2.5830879259570887,
+      "grad_norm": 0.5131027698516846,
+      "learning_rate": 0.0002,
+      "loss": 1.5775,
+      "step": 3070
+    },
+    {
+      "epoch": 2.591501893142617,
+      "grad_norm": 0.47567516565322876,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 3080
+    },
+    {
+      "epoch": 2.599915860328145,
+      "grad_norm": 0.49002596735954285,
+      "learning_rate": 0.0002,
+      "loss": 1.5671,
+      "step": 3090
+    },
+    {
+      "epoch": 2.6083298275136726,
+      "grad_norm": 0.44856327772140503,
+      "learning_rate": 0.0002,
+      "loss": 1.5445,
+      "step": 3100
+    },
+    {
+      "epoch": 2.6167437946992007,
+      "grad_norm": 0.4480142593383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5797,
+      "step": 3110
+    },
+    {
+      "epoch": 2.6251577618847284,
+      "grad_norm": 0.4317494034767151,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 3120
+    },
+    {
+      "epoch": 2.6335717290702565,
+      "grad_norm": 0.42580848932266235,
+      "learning_rate": 0.0002,
+      "loss": 1.6321,
+      "step": 3130
+    },
+    {
+      "epoch": 2.6419856962557846,
+      "grad_norm": 0.4516814947128296,
+      "learning_rate": 0.0002,
+      "loss": 1.6483,
+      "step": 3140
+    },
+    {
+      "epoch": 2.6503996634413127,
+      "grad_norm": 0.4438435733318329,
+      "learning_rate": 0.0002,
+      "loss": 1.695,
+      "step": 3150
+    },
+    {
+      "epoch": 2.6588136306268404,
+      "grad_norm": 0.4385356307029724,
+      "learning_rate": 0.0002,
+      "loss": 1.6938,
+      "step": 3160
+    },
+    {
+      "epoch": 2.6672275978123685,
+      "grad_norm": 0.5064112544059753,
+      "learning_rate": 0.0002,
+      "loss": 1.6139,
+      "step": 3170
+    },
+    {
+      "epoch": 2.6756415649978966,
+      "grad_norm": 0.49163177609443665,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 3180
+    },
+    {
+      "epoch": 2.6840555321834243,
+      "grad_norm": 0.49339258670806885,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 3190
+    },
+    {
+      "epoch": 2.6924694993689524,
+      "grad_norm": 0.440950870513916,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 3200
+    },
+    {
+      "epoch": 2.7008834665544805,
+      "grad_norm": 0.4283970594406128,
+      "learning_rate": 0.0002,
+      "loss": 1.6305,
+      "step": 3210
+    },
+    {
+      "epoch": 2.7092974337400086,
+      "grad_norm": 0.43875712156295776,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 3220
+    },
+    {
+      "epoch": 2.7177114009255363,
+      "grad_norm": 0.49332964420318604,
+      "learning_rate": 0.0002,
+      "loss": 1.6129,
+      "step": 3230
+    },
+    {
+      "epoch": 2.7261253681110644,
+      "grad_norm": 0.5225692391395569,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 3240
+    },
+    {
+      "epoch": 2.734539335296592,
+      "grad_norm": 0.4856489300727844,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 3250
+    },
+    {
+      "epoch": 2.74295330248212,
+      "grad_norm": 0.46918296813964844,
+      "learning_rate": 0.0002,
+      "loss": 1.6463,
+      "step": 3260
+    },
+    {
+      "epoch": 2.7513672696676483,
+      "grad_norm": 0.4802931249141693,
+      "learning_rate": 0.0002,
+      "loss": 1.6819,
+      "step": 3270
+    },
+    {
+      "epoch": 2.7597812368531764,
+      "grad_norm": 0.4485355615615845,
+      "learning_rate": 0.0002,
+      "loss": 1.6246,
+      "step": 3280
+    },
+    {
+      "epoch": 2.768195204038704,
+      "grad_norm": 0.43944594264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 3290
+    },
+    {
+      "epoch": 2.7766091712242322,
+      "grad_norm": 0.46847742795944214,
+      "learning_rate": 0.0002,
+      "loss": 1.6501,
+      "step": 3300
+    },
+    {
+      "epoch": 2.7850231384097603,
+      "grad_norm": 0.4816027879714966,
+      "learning_rate": 0.0002,
+      "loss": 1.5969,
+      "step": 3310
+    },
+    {
+      "epoch": 2.793437105595288,
+      "grad_norm": 0.453960120677948,
+      "learning_rate": 0.0002,
+      "loss": 1.6293,
+      "step": 3320
+    },
+    {
+      "epoch": 2.801851072780816,
+      "grad_norm": 0.4816017150878906,
+      "learning_rate": 0.0002,
+      "loss": 1.6429,
+      "step": 3330
+    },
+    {
+      "epoch": 2.8102650399663442,
+      "grad_norm": 0.4461034834384918,
+      "learning_rate": 0.0002,
+      "loss": 1.6683,
+      "step": 3340
+    },
+    {
+      "epoch": 2.8186790071518724,
+      "grad_norm": 0.48821821808815,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 3350
+    },
+    {
+      "epoch": 2.8270929743374,
+      "grad_norm": 0.4574853777885437,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 3360
+    },
+    {
+      "epoch": 2.835506941522928,
+      "grad_norm": 0.42062026262283325,
+      "learning_rate": 0.0002,
+      "loss": 1.6651,
+      "step": 3370
+    },
+    {
+      "epoch": 2.843920908708456,
+      "grad_norm": 0.4499834477901459,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 3380
+    },
+    {
+      "epoch": 2.852334875893984,
+      "grad_norm": 0.4780360758304596,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 3390
+    },
+    {
+      "epoch": 2.860748843079512,
+      "grad_norm": 0.45422887802124023,
+      "learning_rate": 0.0002,
+      "loss": 1.5882,
+      "step": 3400
+    },
+    {
+      "epoch": 2.86916281026504,
+      "grad_norm": 0.4590015709400177,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 3410
+    },
+    {
+      "epoch": 2.877576777450568,
+      "grad_norm": 0.45689624547958374,
+      "learning_rate": 0.0002,
+      "loss": 1.6746,
+      "step": 3420
+    },
+    {
+      "epoch": 2.885990744636096,
+      "grad_norm": 0.46953922510147095,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3430
+    },
+    {
+      "epoch": 2.8944047118216236,
+      "grad_norm": 0.4791966378688812,
+      "learning_rate": 0.0002,
+      "loss": 1.6015,
+      "step": 3440
+    },
+    {
+      "epoch": 2.9028186790071517,
+      "grad_norm": 0.4842296242713928,
+      "learning_rate": 0.0002,
+      "loss": 1.694,
+      "step": 3450
+    },
+    {
+      "epoch": 2.91123264619268,
+      "grad_norm": 0.47219768166542053,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3460
+    },
+    {
+      "epoch": 2.919646613378208,
+      "grad_norm": 0.4622127115726471,
+      "learning_rate": 0.0002,
+      "loss": 1.6486,
+      "step": 3470
+    },
+    {
+      "epoch": 2.9280605805637356,
+      "grad_norm": 0.46832820773124695,
+      "learning_rate": 0.0002,
+      "loss": 1.6485,
+      "step": 3480
+    },
+    {
+      "epoch": 2.9364745477492638,
+      "grad_norm": 0.44582483172416687,
+      "learning_rate": 0.0002,
+      "loss": 1.6366,
+      "step": 3490
+    },
+    {
+      "epoch": 2.944888514934792,
+      "grad_norm": 0.4987219274044037,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 3500
+    },
+    {
+      "epoch": 2.9533024821203195,
+      "grad_norm": 0.43750956654548645,
+      "learning_rate": 0.0002,
+      "loss": 1.5991,
+      "step": 3510
+    },
+    {
+      "epoch": 2.9617164493058477,
+      "grad_norm": 0.49962925910949707,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 3520
+    },
+    {
+      "epoch": 2.9701304164913758,
+      "grad_norm": 0.5189590454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 3530
+    },
+    {
+      "epoch": 2.978544383676904,
+      "grad_norm": 0.391317754983902,
+      "learning_rate": 0.0002,
+      "loss": 1.6688,
+      "step": 3540
+    },
+    {
+      "epoch": 2.9869583508624316,
+      "grad_norm": 0.44934695959091187,
+      "learning_rate": 0.0002,
+      "loss": 1.5884,
+      "step": 3550
+    },
+    {
+      "epoch": 2.9953723180479597,
+      "grad_norm": 0.4740142226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.5688,
+      "step": 3560
+    },
+    {
+      "epoch": 2.9995793016407237,
+      "eval_loss": 1.8266887664794922,
+      "eval_runtime": 37.9445,
+      "eval_samples_per_second": 13.572,
+      "eval_steps_per_second": 1.713,
+      "step": 3565
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 9504,
+  "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": 1.6500335652215194e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/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-0.8-num-5623-sd-42/checkpoint-3565/training_args.bin b/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-0.8-num-5623-sd-42/checkpoint-3565/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f1502d478cfbb1424f707352d007b740bde5e373
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-3565/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df2b79d3acefeedef5a0229881de39ec68ef9b40046a60d7976a49f7e6b3b936
+size 5560
diff --git a/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-0.8-num-5623-sd-42/checkpoint-4754/README.md b/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-0.8-num-5623-sd-42/checkpoint-4754/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-4754/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_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-0.8-num-5623-sd-42/checkpoint-4754/adapter_config.json b/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-0.8-num-5623-sd-42/checkpoint-4754/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-4754/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_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-0.8-num-5623-sd-42/checkpoint-4754/adapter_model.safetensors b/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-0.8-num-5623-sd-42/checkpoint-4754/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..bcc66ea61bc15269026188d40da0c63fac1d2464
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-4754/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34bb60dd6a14b909b3a4b1b2e39bc03628c58c8097adcf5996e5b1e86852650a
+size 109069176
diff --git a/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-0.8-num-5623-sd-42/checkpoint-4754/optimizer.pt b/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-0.8-num-5623-sd-42/checkpoint-4754/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..913c4afa58476db9fb005acc4db228143df926fa
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-4754/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b0b5b8677319130cb0db8505e3bf2423e75a6e4db8c1ed002e2f856367d3c04
+size 55532666
diff --git a/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-0.8-num-5623-sd-42/checkpoint-4754/rng_state.pth b/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-0.8-num-5623-sd-42/checkpoint-4754/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..86a5ff8ddb03f16119c5416749514cd1bc4a9c5d
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-4754/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fae11d62103c7b8451a989acb28df2f459fd59d22bb21b6a1b17d66ae0d502f6
+size 14244
diff --git a/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-0.8-num-5623-sd-42/checkpoint-4754/scheduler.pt b/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-0.8-num-5623-sd-42/checkpoint-4754/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..4d54a6cd5fade505ebe9ce4e9ee0babc5788f107
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-4754/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efe349c3c5c33e28a25a0a294d49d09b87b69bc75373770eaa85bbc11de7321c
+size 1064
diff --git a/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-0.8-num-5623-sd-42/checkpoint-4754/special_tokens_map.json b/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-0.8-num-5623-sd-42/checkpoint-4754/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-4754/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_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-0.8-num-5623-sd-42/checkpoint-4754/tokenizer.json b/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-0.8-num-5623-sd-42/checkpoint-4754/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-4754/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_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-0.8-num-5623-sd-42/checkpoint-4754/tokenizer_config.json b/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-0.8-num-5623-sd-42/checkpoint-4754/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-4754/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_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-0.8-num-5623-sd-42/checkpoint-4754/trainer_state.json b/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-0.8-num-5623-sd-42/checkpoint-4754/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..2736d78056fd97b7353eb82cf68e7122741f214a
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-4754/trainer_state.json
@@ -0,0 +1,3390 @@
+{
+  "best_metric": 1.8055059909820557,
+  "best_model_checkpoint": "outputs-001/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-0.8-num-5623-sd-42/checkpoint-2377",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 4754,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008413967185527976,
+      "grad_norm": 0.5458821654319763,
+      "learning_rate": 0.0002,
+      "loss": 2.56,
+      "step": 10
+    },
+    {
+      "epoch": 0.016827934371055953,
+      "grad_norm": 0.7293308973312378,
+      "learning_rate": 0.0002,
+      "loss": 2.3235,
+      "step": 20
+    },
+    {
+      "epoch": 0.02524190155658393,
+      "grad_norm": 0.47792306542396545,
+      "learning_rate": 0.0002,
+      "loss": 2.0815,
+      "step": 30
+    },
+    {
+      "epoch": 0.033655868742111905,
+      "grad_norm": 0.5944402813911438,
+      "learning_rate": 0.0002,
+      "loss": 1.9718,
+      "step": 40
+    },
+    {
+      "epoch": 0.04206983592763988,
+      "grad_norm": 0.5415359735488892,
+      "learning_rate": 0.0002,
+      "loss": 1.8848,
+      "step": 50
+    },
+    {
+      "epoch": 0.05048380311316786,
+      "grad_norm": 0.535713791847229,
+      "learning_rate": 0.0002,
+      "loss": 1.8953,
+      "step": 60
+    },
+    {
+      "epoch": 0.058897770298695834,
+      "grad_norm": 0.5184146761894226,
+      "learning_rate": 0.0002,
+      "loss": 1.937,
+      "step": 70
+    },
+    {
+      "epoch": 0.06731173748422381,
+      "grad_norm": 0.458926796913147,
+      "learning_rate": 0.0002,
+      "loss": 1.8396,
+      "step": 80
+    },
+    {
+      "epoch": 0.07572570466975179,
+      "grad_norm": 0.4780142307281494,
+      "learning_rate": 0.0002,
+      "loss": 1.8677,
+      "step": 90
+    },
+    {
+      "epoch": 0.08413967185527976,
+      "grad_norm": 0.79965740442276,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 100
+    },
+    {
+      "epoch": 0.09255363904080774,
+      "grad_norm": 0.4498862028121948,
+      "learning_rate": 0.0002,
+      "loss": 1.9081,
+      "step": 110
+    },
+    {
+      "epoch": 0.10096760622633572,
+      "grad_norm": 0.39338430762290955,
+      "learning_rate": 0.0002,
+      "loss": 1.8503,
+      "step": 120
+    },
+    {
+      "epoch": 0.10938157341186369,
+      "grad_norm": 0.9588953852653503,
+      "learning_rate": 0.0002,
+      "loss": 1.8637,
+      "step": 130
+    },
+    {
+      "epoch": 0.11779554059739167,
+      "grad_norm": 0.41675639152526855,
+      "learning_rate": 0.0002,
+      "loss": 1.8676,
+      "step": 140
+    },
+    {
+      "epoch": 0.12620950778291964,
+      "grad_norm": 0.44519832730293274,
+      "learning_rate": 0.0002,
+      "loss": 1.8904,
+      "step": 150
+    },
+    {
+      "epoch": 0.13462347496844762,
+      "grad_norm": 0.4176260530948639,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 160
+    },
+    {
+      "epoch": 0.1430374421539756,
+      "grad_norm": 0.35840365290641785,
+      "learning_rate": 0.0002,
+      "loss": 1.8398,
+      "step": 170
+    },
+    {
+      "epoch": 0.15145140933950357,
+      "grad_norm": 0.3794495463371277,
+      "learning_rate": 0.0002,
+      "loss": 1.8666,
+      "step": 180
+    },
+    {
+      "epoch": 0.15986537652503155,
+      "grad_norm": 0.4563522934913635,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 190
+    },
+    {
+      "epoch": 0.16827934371055953,
+      "grad_norm": 0.37057486176490784,
+      "learning_rate": 0.0002,
+      "loss": 1.8893,
+      "step": 200
+    },
+    {
+      "epoch": 0.1766933108960875,
+      "grad_norm": 0.44081518054008484,
+      "learning_rate": 0.0002,
+      "loss": 1.7995,
+      "step": 210
+    },
+    {
+      "epoch": 0.18510727808161548,
+      "grad_norm": 0.46078577637672424,
+      "learning_rate": 0.0002,
+      "loss": 1.9048,
+      "step": 220
+    },
+    {
+      "epoch": 0.19352124526714345,
+      "grad_norm": 0.36132094264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.8403,
+      "step": 230
+    },
+    {
+      "epoch": 0.20193521245267143,
+      "grad_norm": 0.3747289180755615,
+      "learning_rate": 0.0002,
+      "loss": 1.8827,
+      "step": 240
+    },
+    {
+      "epoch": 0.2103491796381994,
+      "grad_norm": 0.3540179133415222,
+      "learning_rate": 0.0002,
+      "loss": 1.8382,
+      "step": 250
+    },
+    {
+      "epoch": 0.21876314682372738,
+      "grad_norm": 0.3461375832557678,
+      "learning_rate": 0.0002,
+      "loss": 1.8196,
+      "step": 260
+    },
+    {
+      "epoch": 0.22717711400925536,
+      "grad_norm": 0.3436960279941559,
+      "learning_rate": 0.0002,
+      "loss": 1.8509,
+      "step": 270
+    },
+    {
+      "epoch": 0.23559108119478334,
+      "grad_norm": 0.35403719544410706,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 280
+    },
+    {
+      "epoch": 0.2440050483803113,
+      "grad_norm": 0.37142616510391235,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 290
+    },
+    {
+      "epoch": 0.2524190155658393,
+      "grad_norm": 0.3307955861091614,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 300
+    },
+    {
+      "epoch": 0.2608329827513673,
+      "grad_norm": 0.32855314016342163,
+      "learning_rate": 0.0002,
+      "loss": 1.817,
+      "step": 310
+    },
+    {
+      "epoch": 0.26924694993689524,
+      "grad_norm": 0.3299003839492798,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 320
+    },
+    {
+      "epoch": 0.27766091712242325,
+      "grad_norm": 0.44311287999153137,
+      "learning_rate": 0.0002,
+      "loss": 1.8129,
+      "step": 330
+    },
+    {
+      "epoch": 0.2860748843079512,
+      "grad_norm": 0.32989758253097534,
+      "learning_rate": 0.0002,
+      "loss": 1.8232,
+      "step": 340
+    },
+    {
+      "epoch": 0.2944888514934792,
+      "grad_norm": 0.34400200843811035,
+      "learning_rate": 0.0002,
+      "loss": 1.7716,
+      "step": 350
+    },
+    {
+      "epoch": 0.30290281867900715,
+      "grad_norm": 0.36286211013793945,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 360
+    },
+    {
+      "epoch": 0.31131678586453515,
+      "grad_norm": 0.406827837228775,
+      "learning_rate": 0.0002,
+      "loss": 1.8025,
+      "step": 370
+    },
+    {
+      "epoch": 0.3197307530500631,
+      "grad_norm": 0.36299195885658264,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 380
+    },
+    {
+      "epoch": 0.3281447202355911,
+      "grad_norm": 0.3477257192134857,
+      "learning_rate": 0.0002,
+      "loss": 1.837,
+      "step": 390
+    },
+    {
+      "epoch": 0.33655868742111905,
+      "grad_norm": 0.3730369210243225,
+      "learning_rate": 0.0002,
+      "loss": 1.7767,
+      "step": 400
+    },
+    {
+      "epoch": 0.34497265460664706,
+      "grad_norm": 0.4644559919834137,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 410
+    },
+    {
+      "epoch": 0.353386621792175,
+      "grad_norm": 0.406576544046402,
+      "learning_rate": 0.0002,
+      "loss": 1.7538,
+      "step": 420
+    },
+    {
+      "epoch": 0.361800588977703,
+      "grad_norm": 0.3612699508666992,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 430
+    },
+    {
+      "epoch": 0.37021455616323096,
+      "grad_norm": 0.3243742287158966,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 440
+    },
+    {
+      "epoch": 0.37862852334875896,
+      "grad_norm": 0.36671221256256104,
+      "learning_rate": 0.0002,
+      "loss": 1.8851,
+      "step": 450
+    },
+    {
+      "epoch": 0.3870424905342869,
+      "grad_norm": 0.3565002381801605,
+      "learning_rate": 0.0002,
+      "loss": 1.8853,
+      "step": 460
+    },
+    {
+      "epoch": 0.3954564577198149,
+      "grad_norm": 0.34630221128463745,
+      "learning_rate": 0.0002,
+      "loss": 1.8923,
+      "step": 470
+    },
+    {
+      "epoch": 0.40387042490534286,
+      "grad_norm": 0.3353537321090698,
+      "learning_rate": 0.0002,
+      "loss": 1.8234,
+      "step": 480
+    },
+    {
+      "epoch": 0.41228439209087087,
+      "grad_norm": 0.4015921950340271,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 490
+    },
+    {
+      "epoch": 0.4206983592763988,
+      "grad_norm": 0.5489419102668762,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 500
+    },
+    {
+      "epoch": 0.4291123264619268,
+      "grad_norm": 0.4193589985370636,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 510
+    },
+    {
+      "epoch": 0.43752629364745477,
+      "grad_norm": 0.3418922424316406,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 520
+    },
+    {
+      "epoch": 0.44594026083298277,
+      "grad_norm": 0.32668185234069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 530
+    },
+    {
+      "epoch": 0.4543542280185107,
+      "grad_norm": 0.3094325661659241,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 540
+    },
+    {
+      "epoch": 0.4627681952040387,
+      "grad_norm": 0.3743017315864563,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 550
+    },
+    {
+      "epoch": 0.47118216238956667,
+      "grad_norm": 0.3295630216598511,
+      "learning_rate": 0.0002,
+      "loss": 1.8451,
+      "step": 560
+    },
+    {
+      "epoch": 0.4795961295750947,
+      "grad_norm": 1.6124513149261475,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 570
+    },
+    {
+      "epoch": 0.4880100967606226,
+      "grad_norm": 0.3245585858821869,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 580
+    },
+    {
+      "epoch": 0.49642406394615063,
+      "grad_norm": 0.3332934081554413,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 590
+    },
+    {
+      "epoch": 0.5048380311316786,
+      "grad_norm": 0.3836138844490051,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 600
+    },
+    {
+      "epoch": 0.5132519983172066,
+      "grad_norm": 0.32953888177871704,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 610
+    },
+    {
+      "epoch": 0.5216659655027346,
+      "grad_norm": 0.36291512846946716,
+      "learning_rate": 0.0002,
+      "loss": 1.7729,
+      "step": 620
+    },
+    {
+      "epoch": 0.5300799326882625,
+      "grad_norm": 0.3237783908843994,
+      "learning_rate": 0.0002,
+      "loss": 1.7758,
+      "step": 630
+    },
+    {
+      "epoch": 0.5384938998737905,
+      "grad_norm": 0.38882696628570557,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 640
+    },
+    {
+      "epoch": 0.5469078670593185,
+      "grad_norm": 0.37821972370147705,
+      "learning_rate": 0.0002,
+      "loss": 1.8624,
+      "step": 650
+    },
+    {
+      "epoch": 0.5553218342448465,
+      "grad_norm": 0.3556285500526428,
+      "learning_rate": 0.0002,
+      "loss": 1.8075,
+      "step": 660
+    },
+    {
+      "epoch": 0.5637358014303744,
+      "grad_norm": 0.347499281167984,
+      "learning_rate": 0.0002,
+      "loss": 1.778,
+      "step": 670
+    },
+    {
+      "epoch": 0.5721497686159024,
+      "grad_norm": 0.3176489472389221,
+      "learning_rate": 0.0002,
+      "loss": 1.8066,
+      "step": 680
+    },
+    {
+      "epoch": 0.5805637358014304,
+      "grad_norm": 0.30220088362693787,
+      "learning_rate": 0.0002,
+      "loss": 1.7257,
+      "step": 690
+    },
+    {
+      "epoch": 0.5889777029869584,
+      "grad_norm": 0.3711601793766022,
+      "learning_rate": 0.0002,
+      "loss": 1.8415,
+      "step": 700
+    },
+    {
+      "epoch": 0.5973916701724863,
+      "grad_norm": 0.3311759829521179,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 710
+    },
+    {
+      "epoch": 0.6058056373580143,
+      "grad_norm": 0.34824270009994507,
+      "learning_rate": 0.0002,
+      "loss": 1.7712,
+      "step": 720
+    },
+    {
+      "epoch": 0.6142196045435423,
+      "grad_norm": 0.29668381810188293,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 730
+    },
+    {
+      "epoch": 0.6226335717290703,
+      "grad_norm": 0.36087489128112793,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 740
+    },
+    {
+      "epoch": 0.6310475389145982,
+      "grad_norm": 0.31590089201927185,
+      "learning_rate": 0.0002,
+      "loss": 1.7956,
+      "step": 750
+    },
+    {
+      "epoch": 0.6394615061001262,
+      "grad_norm": 0.37632957100868225,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 760
+    },
+    {
+      "epoch": 0.6478754732856542,
+      "grad_norm": 0.3360748589038849,
+      "learning_rate": 0.0002,
+      "loss": 1.8499,
+      "step": 770
+    },
+    {
+      "epoch": 0.6562894404711822,
+      "grad_norm": 0.3420640528202057,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 780
+    },
+    {
+      "epoch": 0.6647034076567101,
+      "grad_norm": 0.5734959244728088,
+      "learning_rate": 0.0002,
+      "loss": 1.8353,
+      "step": 790
+    },
+    {
+      "epoch": 0.6731173748422381,
+      "grad_norm": 0.36440837383270264,
+      "learning_rate": 0.0002,
+      "loss": 1.7746,
+      "step": 800
+    },
+    {
+      "epoch": 0.6815313420277661,
+      "grad_norm": 0.3179708421230316,
+      "learning_rate": 0.0002,
+      "loss": 1.7532,
+      "step": 810
+    },
+    {
+      "epoch": 0.6899453092132941,
+      "grad_norm": 0.34122881293296814,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 820
+    },
+    {
+      "epoch": 0.698359276398822,
+      "grad_norm": 0.31886112689971924,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 830
+    },
+    {
+      "epoch": 0.70677324358435,
+      "grad_norm": 0.31782326102256775,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 840
+    },
+    {
+      "epoch": 0.715187210769878,
+      "grad_norm": 0.36052989959716797,
+      "learning_rate": 0.0002,
+      "loss": 1.7588,
+      "step": 850
+    },
+    {
+      "epoch": 0.723601177955406,
+      "grad_norm": 0.28946155309677124,
+      "learning_rate": 0.0002,
+      "loss": 1.7891,
+      "step": 860
+    },
+    {
+      "epoch": 0.7320151451409339,
+      "grad_norm": 0.3095663785934448,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 870
+    },
+    {
+      "epoch": 0.7404291123264619,
+      "grad_norm": 0.3317491412162781,
+      "learning_rate": 0.0002,
+      "loss": 1.785,
+      "step": 880
+    },
+    {
+      "epoch": 0.7488430795119899,
+      "grad_norm": 0.31324660778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 890
+    },
+    {
+      "epoch": 0.7572570466975179,
+      "grad_norm": 0.3290475606918335,
+      "learning_rate": 0.0002,
+      "loss": 1.8753,
+      "step": 900
+    },
+    {
+      "epoch": 0.7656710138830458,
+      "grad_norm": 0.35690343379974365,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 910
+    },
+    {
+      "epoch": 0.7740849810685738,
+      "grad_norm": 0.39558273553848267,
+      "learning_rate": 0.0002,
+      "loss": 1.826,
+      "step": 920
+    },
+    {
+      "epoch": 0.7824989482541018,
+      "grad_norm": 0.34254348278045654,
+      "learning_rate": 0.0002,
+      "loss": 1.8722,
+      "step": 930
+    },
+    {
+      "epoch": 0.7909129154396298,
+      "grad_norm": 0.3560165464878082,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 940
+    },
+    {
+      "epoch": 0.7993268826251577,
+      "grad_norm": 0.30693164467811584,
+      "learning_rate": 0.0002,
+      "loss": 1.7992,
+      "step": 950
+    },
+    {
+      "epoch": 0.8077408498106857,
+      "grad_norm": 0.3394823372364044,
+      "learning_rate": 0.0002,
+      "loss": 1.8029,
+      "step": 960
+    },
+    {
+      "epoch": 0.8161548169962137,
+      "grad_norm": 0.3741514980792999,
+      "learning_rate": 0.0002,
+      "loss": 1.8105,
+      "step": 970
+    },
+    {
+      "epoch": 0.8245687841817417,
+      "grad_norm": 0.3655228316783905,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 980
+    },
+    {
+      "epoch": 0.8329827513672696,
+      "grad_norm": 0.3586033880710602,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 990
+    },
+    {
+      "epoch": 0.8413967185527976,
+      "grad_norm": 0.3459678888320923,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8498106857383256,
+      "grad_norm": 0.3184349834918976,
+      "learning_rate": 0.0002,
+      "loss": 1.8498,
+      "step": 1010
+    },
+    {
+      "epoch": 0.8582246529238536,
+      "grad_norm": 0.3099786043167114,
+      "learning_rate": 0.0002,
+      "loss": 1.7632,
+      "step": 1020
+    },
+    {
+      "epoch": 0.8666386201093815,
+      "grad_norm": 0.30300915241241455,
+      "learning_rate": 0.0002,
+      "loss": 1.8067,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8750525872949095,
+      "grad_norm": 0.3128705620765686,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8834665544804375,
+      "grad_norm": 0.3336263597011566,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8918805216659655,
+      "grad_norm": 0.3801328241825104,
+      "learning_rate": 0.0002,
+      "loss": 1.8375,
+      "step": 1060
+    },
+    {
+      "epoch": 0.9002944888514934,
+      "grad_norm": 0.3122096359729767,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 1070
+    },
+    {
+      "epoch": 0.9087084560370214,
+      "grad_norm": 0.35990869998931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 1080
+    },
+    {
+      "epoch": 0.9171224232225494,
+      "grad_norm": 0.3321819305419922,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1090
+    },
+    {
+      "epoch": 0.9255363904080774,
+      "grad_norm": 0.4202139377593994,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9339503575936053,
+      "grad_norm": 0.32559722661972046,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 1110
+    },
+    {
+      "epoch": 0.9423643247791333,
+      "grad_norm": 0.3098459839820862,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 1120
+    },
+    {
+      "epoch": 0.9507782919646613,
+      "grad_norm": 0.33917108178138733,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1130
+    },
+    {
+      "epoch": 0.9591922591501894,
+      "grad_norm": 0.4055837094783783,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1140
+    },
+    {
+      "epoch": 0.9676062263357172,
+      "grad_norm": 0.32508623600006104,
+      "learning_rate": 0.0002,
+      "loss": 1.8259,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9760201935212452,
+      "grad_norm": 0.30150601267814636,
+      "learning_rate": 0.0002,
+      "loss": 1.782,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9844341607067733,
+      "grad_norm": 0.3042563199996948,
+      "learning_rate": 0.0002,
+      "loss": 1.8291,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9928481278923013,
+      "grad_norm": 0.33254584670066833,
+      "learning_rate": 0.0002,
+      "loss": 1.7847,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9995793016407236,
+      "eval_loss": 1.8077726364135742,
+      "eval_runtime": 38.4359,
+      "eval_samples_per_second": 13.399,
+      "eval_steps_per_second": 1.691,
+      "step": 1188
+    },
+    {
+      "epoch": 1.0012620950778293,
+      "grad_norm": 0.35073035955429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 1190
+    },
+    {
+      "epoch": 1.0096760622633572,
+      "grad_norm": 0.3217269778251648,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1200
+    },
+    {
+      "epoch": 1.018090029448885,
+      "grad_norm": 0.3635033369064331,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1210
+    },
+    {
+      "epoch": 1.0265039966344132,
+      "grad_norm": 0.32468414306640625,
+      "learning_rate": 0.0002,
+      "loss": 1.6949,
+      "step": 1220
+    },
+    {
+      "epoch": 1.034917963819941,
+      "grad_norm": 0.3307163417339325,
+      "learning_rate": 0.0002,
+      "loss": 1.711,
+      "step": 1230
+    },
+    {
+      "epoch": 1.0433319310054692,
+      "grad_norm": 0.34381359815597534,
+      "learning_rate": 0.0002,
+      "loss": 1.7881,
+      "step": 1240
+    },
+    {
+      "epoch": 1.051745898190997,
+      "grad_norm": 0.35874804854393005,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 1250
+    },
+    {
+      "epoch": 1.060159865376525,
+      "grad_norm": 0.3615919351577759,
+      "learning_rate": 0.0002,
+      "loss": 1.7314,
+      "step": 1260
+    },
+    {
+      "epoch": 1.068573832562053,
+      "grad_norm": 0.32835808396339417,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1270
+    },
+    {
+      "epoch": 1.076987799747581,
+      "grad_norm": 0.3876388370990753,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0854017669331089,
+      "grad_norm": 0.39895930886268616,
+      "learning_rate": 0.0002,
+      "loss": 1.7442,
+      "step": 1290
+    },
+    {
+      "epoch": 1.093815734118637,
+      "grad_norm": 0.39081698656082153,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 1300
+    },
+    {
+      "epoch": 1.1022297013041649,
+      "grad_norm": 0.39974215626716614,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1310
+    },
+    {
+      "epoch": 1.110643668489693,
+      "grad_norm": 0.3887332081794739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1320
+    },
+    {
+      "epoch": 1.1190576356752209,
+      "grad_norm": 0.36216408014297485,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 1330
+    },
+    {
+      "epoch": 1.1274716028607488,
+      "grad_norm": 0.36979028582572937,
+      "learning_rate": 0.0002,
+      "loss": 1.762,
+      "step": 1340
+    },
+    {
+      "epoch": 1.1358855700462769,
+      "grad_norm": 0.34052133560180664,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 1350
+    },
+    {
+      "epoch": 1.1442995372318048,
+      "grad_norm": 0.3467716574668884,
+      "learning_rate": 0.0002,
+      "loss": 1.7513,
+      "step": 1360
+    },
+    {
+      "epoch": 1.1527135044173327,
+      "grad_norm": 0.35528799891471863,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1370
+    },
+    {
+      "epoch": 1.1611274716028608,
+      "grad_norm": 0.36282262206077576,
+      "learning_rate": 0.0002,
+      "loss": 1.794,
+      "step": 1380
+    },
+    {
+      "epoch": 1.1695414387883887,
+      "grad_norm": 0.37355899810791016,
+      "learning_rate": 0.0002,
+      "loss": 1.7731,
+      "step": 1390
+    },
+    {
+      "epoch": 1.1779554059739168,
+      "grad_norm": 0.37292736768722534,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1400
+    },
+    {
+      "epoch": 1.1863693731594447,
+      "grad_norm": 0.5892812013626099,
+      "learning_rate": 0.0002,
+      "loss": 1.6916,
+      "step": 1410
+    },
+    {
+      "epoch": 1.1947833403449726,
+      "grad_norm": 0.3712292015552521,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 1420
+    },
+    {
+      "epoch": 1.2031973075305007,
+      "grad_norm": 0.3349577486515045,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1430
+    },
+    {
+      "epoch": 1.2116112747160286,
+      "grad_norm": 0.32591062784194946,
+      "learning_rate": 0.0002,
+      "loss": 1.7412,
+      "step": 1440
+    },
+    {
+      "epoch": 1.2200252419015567,
+      "grad_norm": 0.3840635418891907,
+      "learning_rate": 0.0002,
+      "loss": 1.7406,
+      "step": 1450
+    },
+    {
+      "epoch": 1.2284392090870846,
+      "grad_norm": 0.37238365411758423,
+      "learning_rate": 0.0002,
+      "loss": 1.7276,
+      "step": 1460
+    },
+    {
+      "epoch": 1.2368531762726125,
+      "grad_norm": 0.3731217682361603,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 1470
+    },
+    {
+      "epoch": 1.2452671434581406,
+      "grad_norm": 0.3318967819213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7255,
+      "step": 1480
+    },
+    {
+      "epoch": 1.2536811106436685,
+      "grad_norm": 0.3784034848213196,
+      "learning_rate": 0.0002,
+      "loss": 1.7463,
+      "step": 1490
+    },
+    {
+      "epoch": 1.2620950778291964,
+      "grad_norm": 0.3541383147239685,
+      "learning_rate": 0.0002,
+      "loss": 1.6862,
+      "step": 1500
+    },
+    {
+      "epoch": 1.2705090450147245,
+      "grad_norm": 0.35312485694885254,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1510
+    },
+    {
+      "epoch": 1.2789230122002524,
+      "grad_norm": 0.35272929072380066,
+      "learning_rate": 0.0002,
+      "loss": 1.7029,
+      "step": 1520
+    },
+    {
+      "epoch": 1.2873369793857803,
+      "grad_norm": 0.40988272428512573,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1530
+    },
+    {
+      "epoch": 1.2957509465713084,
+      "grad_norm": 0.3543946146965027,
+      "learning_rate": 0.0002,
+      "loss": 1.6912,
+      "step": 1540
+    },
+    {
+      "epoch": 1.3041649137568363,
+      "grad_norm": 0.35639145970344543,
+      "learning_rate": 0.0002,
+      "loss": 1.6757,
+      "step": 1550
+    },
+    {
+      "epoch": 1.3125788809423642,
+      "grad_norm": 0.3290826678276062,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 1560
+    },
+    {
+      "epoch": 1.3209928481278923,
+      "grad_norm": 0.39264336228370667,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 1570
+    },
+    {
+      "epoch": 1.3294068153134202,
+      "grad_norm": 0.5390415191650391,
+      "learning_rate": 0.0002,
+      "loss": 1.6804,
+      "step": 1580
+    },
+    {
+      "epoch": 1.3378207824989483,
+      "grad_norm": 0.5188116431236267,
+      "learning_rate": 0.0002,
+      "loss": 1.708,
+      "step": 1590
+    },
+    {
+      "epoch": 1.3462347496844762,
+      "grad_norm": 0.37445148825645447,
+      "learning_rate": 0.0002,
+      "loss": 1.6763,
+      "step": 1600
+    },
+    {
+      "epoch": 1.3546487168700043,
+      "grad_norm": 0.3296085298061371,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1610
+    },
+    {
+      "epoch": 1.3630626840555322,
+      "grad_norm": 0.39879581332206726,
+      "learning_rate": 0.0002,
+      "loss": 1.8107,
+      "step": 1620
+    },
+    {
+      "epoch": 1.37147665124106,
+      "grad_norm": 0.36092764139175415,
+      "learning_rate": 0.0002,
+      "loss": 1.6744,
+      "step": 1630
+    },
+    {
+      "epoch": 1.3798906184265882,
+      "grad_norm": 0.37011823058128357,
+      "learning_rate": 0.0002,
+      "loss": 1.7144,
+      "step": 1640
+    },
+    {
+      "epoch": 1.3883045856121161,
+      "grad_norm": 0.40863534808158875,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1650
+    },
+    {
+      "epoch": 1.396718552797644,
+      "grad_norm": 0.337001770734787,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 1660
+    },
+    {
+      "epoch": 1.4051325199831721,
+      "grad_norm": 0.35596707463264465,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1670
+    },
+    {
+      "epoch": 1.4135464871687,
+      "grad_norm": 0.3857671916484833,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 1680
+    },
+    {
+      "epoch": 1.421960454354228,
+      "grad_norm": 0.419502317905426,
+      "learning_rate": 0.0002,
+      "loss": 1.7015,
+      "step": 1690
+    },
+    {
+      "epoch": 1.430374421539756,
+      "grad_norm": 0.35459452867507935,
+      "learning_rate": 0.0002,
+      "loss": 1.7261,
+      "step": 1700
+    },
+    {
+      "epoch": 1.438788388725284,
+      "grad_norm": 0.37246978282928467,
+      "learning_rate": 0.0002,
+      "loss": 1.7361,
+      "step": 1710
+    },
+    {
+      "epoch": 1.4472023559108118,
+      "grad_norm": 0.33091893792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 1720
+    },
+    {
+      "epoch": 1.45561632309634,
+      "grad_norm": 0.37029674649238586,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1730
+    },
+    {
+      "epoch": 1.4640302902818678,
+      "grad_norm": 0.374025821685791,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1740
+    },
+    {
+      "epoch": 1.472444257467396,
+      "grad_norm": 0.3416315019130707,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1750
+    },
+    {
+      "epoch": 1.4808582246529238,
+      "grad_norm": 0.36502841114997864,
+      "learning_rate": 0.0002,
+      "loss": 1.7093,
+      "step": 1760
+    },
+    {
+      "epoch": 1.489272191838452,
+      "grad_norm": 0.35458803176879883,
+      "learning_rate": 0.0002,
+      "loss": 1.6597,
+      "step": 1770
+    },
+    {
+      "epoch": 1.4976861590239798,
+      "grad_norm": 0.4462839663028717,
+      "learning_rate": 0.0002,
+      "loss": 1.675,
+      "step": 1780
+    },
+    {
+      "epoch": 1.5061001262095077,
+      "grad_norm": 0.34836092591285706,
+      "learning_rate": 0.0002,
+      "loss": 1.7267,
+      "step": 1790
+    },
+    {
+      "epoch": 1.5145140933950358,
+      "grad_norm": 0.3445749282836914,
+      "learning_rate": 0.0002,
+      "loss": 1.7295,
+      "step": 1800
+    },
+    {
+      "epoch": 1.5229280605805637,
+      "grad_norm": 0.36012160778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1810
+    },
+    {
+      "epoch": 1.5313420277660916,
+      "grad_norm": 0.4052616059780121,
+      "learning_rate": 0.0002,
+      "loss": 1.6594,
+      "step": 1820
+    },
+    {
+      "epoch": 1.5397559949516197,
+      "grad_norm": 0.3966905474662781,
+      "learning_rate": 0.0002,
+      "loss": 1.72,
+      "step": 1830
+    },
+    {
+      "epoch": 1.5481699621371476,
+      "grad_norm": 0.35028719902038574,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1840
+    },
+    {
+      "epoch": 1.5565839293226755,
+      "grad_norm": 0.3936742842197418,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1850
+    },
+    {
+      "epoch": 1.5649978965082036,
+      "grad_norm": 0.34473296999931335,
+      "learning_rate": 0.0002,
+      "loss": 1.7579,
+      "step": 1860
+    },
+    {
+      "epoch": 1.5734118636937318,
+      "grad_norm": 0.4328365623950958,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1870
+    },
+    {
+      "epoch": 1.5818258308792594,
+      "grad_norm": 0.3566315472126007,
+      "learning_rate": 0.0002,
+      "loss": 1.7098,
+      "step": 1880
+    },
+    {
+      "epoch": 1.5902397980647875,
+      "grad_norm": 0.3301256597042084,
+      "learning_rate": 0.0002,
+      "loss": 1.6095,
+      "step": 1890
+    },
+    {
+      "epoch": 1.5986537652503157,
+      "grad_norm": 0.3743041455745697,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 1900
+    },
+    {
+      "epoch": 1.6070677324358436,
+      "grad_norm": 0.3735344707965851,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 1910
+    },
+    {
+      "epoch": 1.6154816996213714,
+      "grad_norm": 0.42191144824028015,
+      "learning_rate": 0.0002,
+      "loss": 1.7445,
+      "step": 1920
+    },
+    {
+      "epoch": 1.6238956668068996,
+      "grad_norm": 0.3787207305431366,
+      "learning_rate": 0.0002,
+      "loss": 1.6978,
+      "step": 1930
+    },
+    {
+      "epoch": 1.6323096339924275,
+      "grad_norm": 0.35647350549697876,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 1940
+    },
+    {
+      "epoch": 1.6407236011779553,
+      "grad_norm": 0.39791446924209595,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 1950
+    },
+    {
+      "epoch": 1.6491375683634835,
+      "grad_norm": 0.37341275811195374,
+      "learning_rate": 0.0002,
+      "loss": 1.7293,
+      "step": 1960
+    },
+    {
+      "epoch": 1.6575515355490114,
+      "grad_norm": 0.3722686469554901,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1970
+    },
+    {
+      "epoch": 1.6659655027345392,
+      "grad_norm": 0.37467387318611145,
+      "learning_rate": 0.0002,
+      "loss": 1.6383,
+      "step": 1980
+    },
+    {
+      "epoch": 1.6743794699200674,
+      "grad_norm": 0.37109461426734924,
+      "learning_rate": 0.0002,
+      "loss": 1.7439,
+      "step": 1990
+    },
+    {
+      "epoch": 1.6827934371055953,
+      "grad_norm": 0.4008837044239044,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 2000
+    },
+    {
+      "epoch": 1.6912074042911232,
+      "grad_norm": 0.3316999673843384,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2010
+    },
+    {
+      "epoch": 1.6996213714766513,
+      "grad_norm": 0.3683805465698242,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 2020
+    },
+    {
+      "epoch": 1.7080353386621794,
+      "grad_norm": 0.4163658320903778,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 2030
+    },
+    {
+      "epoch": 1.716449305847707,
+      "grad_norm": 0.4245431125164032,
+      "learning_rate": 0.0002,
+      "loss": 1.741,
+      "step": 2040
+    },
+    {
+      "epoch": 1.7248632730332352,
+      "grad_norm": 0.36732038855552673,
+      "learning_rate": 0.0002,
+      "loss": 1.7184,
+      "step": 2050
+    },
+    {
+      "epoch": 1.7332772402187633,
+      "grad_norm": 0.34981656074523926,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 2060
+    },
+    {
+      "epoch": 1.7416912074042912,
+      "grad_norm": 0.38588812947273254,
+      "learning_rate": 0.0002,
+      "loss": 1.7545,
+      "step": 2070
+    },
+    {
+      "epoch": 1.750105174589819,
+      "grad_norm": 0.39914557337760925,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 2080
+    },
+    {
+      "epoch": 1.7585191417753472,
+      "grad_norm": 0.36068692803382874,
+      "learning_rate": 0.0002,
+      "loss": 1.7049,
+      "step": 2090
+    },
+    {
+      "epoch": 1.766933108960875,
+      "grad_norm": 0.3983287215232849,
+      "learning_rate": 0.0002,
+      "loss": 1.7537,
+      "step": 2100
+    },
+    {
+      "epoch": 1.775347076146403,
+      "grad_norm": 0.45008400082588196,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2110
+    },
+    {
+      "epoch": 1.783761043331931,
+      "grad_norm": 0.3618052303791046,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 2120
+    },
+    {
+      "epoch": 1.792175010517459,
+      "grad_norm": 0.38745400309562683,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 2130
+    },
+    {
+      "epoch": 1.8005889777029869,
+      "grad_norm": 0.3413826525211334,
+      "learning_rate": 0.0002,
+      "loss": 1.7387,
+      "step": 2140
+    },
+    {
+      "epoch": 1.809002944888515,
+      "grad_norm": 0.35983747243881226,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 2150
+    },
+    {
+      "epoch": 1.8174169120740429,
+      "grad_norm": 0.40926849842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 2160
+    },
+    {
+      "epoch": 1.8258308792595708,
+      "grad_norm": 0.3543093800544739,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2170
+    },
+    {
+      "epoch": 1.8342448464450989,
+      "grad_norm": 0.42690935730934143,
+      "learning_rate": 0.0002,
+      "loss": 1.7812,
+      "step": 2180
+    },
+    {
+      "epoch": 1.842658813630627,
+      "grad_norm": 0.40282756090164185,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 2190
+    },
+    {
+      "epoch": 1.8510727808161547,
+      "grad_norm": 0.36568400263786316,
+      "learning_rate": 0.0002,
+      "loss": 1.7411,
+      "step": 2200
+    },
+    {
+      "epoch": 1.8594867480016828,
+      "grad_norm": 0.43159013986587524,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 2210
+    },
+    {
+      "epoch": 1.867900715187211,
+      "grad_norm": 0.3554118573665619,
+      "learning_rate": 0.0002,
+      "loss": 1.7298,
+      "step": 2220
+    },
+    {
+      "epoch": 1.8763146823727388,
+      "grad_norm": 0.43349072337150574,
+      "learning_rate": 0.0002,
+      "loss": 1.7157,
+      "step": 2230
+    },
+    {
+      "epoch": 1.8847286495582667,
+      "grad_norm": 0.36486536264419556,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 2240
+    },
+    {
+      "epoch": 1.8931426167437948,
+      "grad_norm": 0.39260047674179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6901,
+      "step": 2250
+    },
+    {
+      "epoch": 1.9015565839293227,
+      "grad_norm": 0.3741776943206787,
+      "learning_rate": 0.0002,
+      "loss": 1.6691,
+      "step": 2260
+    },
+    {
+      "epoch": 1.9099705511148506,
+      "grad_norm": 0.3961946964263916,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2270
+    },
+    {
+      "epoch": 1.9183845183003787,
+      "grad_norm": 0.3659731149673462,
+      "learning_rate": 0.0002,
+      "loss": 1.737,
+      "step": 2280
+    },
+    {
+      "epoch": 1.9267984854859066,
+      "grad_norm": 0.34744107723236084,
+      "learning_rate": 0.0002,
+      "loss": 1.7342,
+      "step": 2290
+    },
+    {
+      "epoch": 1.9352124526714345,
+      "grad_norm": 0.3607442378997803,
+      "learning_rate": 0.0002,
+      "loss": 1.7162,
+      "step": 2300
+    },
+    {
+      "epoch": 1.9436264198569626,
+      "grad_norm": 0.331464558839798,
+      "learning_rate": 0.0002,
+      "loss": 1.6673,
+      "step": 2310
+    },
+    {
+      "epoch": 1.9520403870424905,
+      "grad_norm": 0.3904414474964142,
+      "learning_rate": 0.0002,
+      "loss": 1.7101,
+      "step": 2320
+    },
+    {
+      "epoch": 1.9604543542280184,
+      "grad_norm": 0.37584832310676575,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 2330
+    },
+    {
+      "epoch": 1.9688683214135465,
+      "grad_norm": 0.3698684275150299,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 2340
+    },
+    {
+      "epoch": 1.9772822885990746,
+      "grad_norm": 0.40571412444114685,
+      "learning_rate": 0.0002,
+      "loss": 1.7764,
+      "step": 2350
+    },
+    {
+      "epoch": 1.9856962557846023,
+      "grad_norm": 0.40059587359428406,
+      "learning_rate": 0.0002,
+      "loss": 1.744,
+      "step": 2360
+    },
+    {
+      "epoch": 1.9941102229701304,
+      "grad_norm": 0.4168248474597931,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2370
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8055059909820557,
+      "eval_runtime": 38.422,
+      "eval_samples_per_second": 13.404,
+      "eval_steps_per_second": 1.692,
+      "step": 2377
+    },
+    {
+      "epoch": 2.0025241901556585,
+      "grad_norm": 0.35205352306365967,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 2380
+    },
+    {
+      "epoch": 2.010938157341186,
+      "grad_norm": 0.3979377746582031,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 2390
+    },
+    {
+      "epoch": 2.0193521245267143,
+      "grad_norm": 0.396491676568985,
+      "learning_rate": 0.0002,
+      "loss": 1.6421,
+      "step": 2400
+    },
+    {
+      "epoch": 2.0277660917122424,
+      "grad_norm": 0.44712209701538086,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2410
+    },
+    {
+      "epoch": 2.03618005889777,
+      "grad_norm": 0.4454420208930969,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 2420
+    },
+    {
+      "epoch": 2.044594026083298,
+      "grad_norm": 0.4170038402080536,
+      "learning_rate": 0.0002,
+      "loss": 1.6635,
+      "step": 2430
+    },
+    {
+      "epoch": 2.0530079932688263,
+      "grad_norm": 0.4309595227241516,
+      "learning_rate": 0.0002,
+      "loss": 1.6512,
+      "step": 2440
+    },
+    {
+      "epoch": 2.0614219604543544,
+      "grad_norm": 0.4241602122783661,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 2450
+    },
+    {
+      "epoch": 2.069835927639882,
+      "grad_norm": 0.4370540678501129,
+      "learning_rate": 0.0002,
+      "loss": 1.6162,
+      "step": 2460
+    },
+    {
+      "epoch": 2.0782498948254102,
+      "grad_norm": 0.43985554575920105,
+      "learning_rate": 0.0002,
+      "loss": 1.6354,
+      "step": 2470
+    },
+    {
+      "epoch": 2.0866638620109383,
+      "grad_norm": 0.4158105254173279,
+      "learning_rate": 0.0002,
+      "loss": 1.6954,
+      "step": 2480
+    },
+    {
+      "epoch": 2.095077829196466,
+      "grad_norm": 0.441549152135849,
+      "learning_rate": 0.0002,
+      "loss": 1.6114,
+      "step": 2490
+    },
+    {
+      "epoch": 2.103491796381994,
+      "grad_norm": 0.385718435049057,
+      "learning_rate": 0.0002,
+      "loss": 1.5485,
+      "step": 2500
+    },
+    {
+      "epoch": 2.1119057635675222,
+      "grad_norm": 0.43146514892578125,
+      "learning_rate": 0.0002,
+      "loss": 1.5894,
+      "step": 2510
+    },
+    {
+      "epoch": 2.12031973075305,
+      "grad_norm": 0.41663315892219543,
+      "learning_rate": 0.0002,
+      "loss": 1.6414,
+      "step": 2520
+    },
+    {
+      "epoch": 2.128733697938578,
+      "grad_norm": 0.4410698115825653,
+      "learning_rate": 0.0002,
+      "loss": 1.6527,
+      "step": 2530
+    },
+    {
+      "epoch": 2.137147665124106,
+      "grad_norm": 0.4472278952598572,
+      "learning_rate": 0.0002,
+      "loss": 1.6124,
+      "step": 2540
+    },
+    {
+      "epoch": 2.145561632309634,
+      "grad_norm": 0.3879167437553406,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2550
+    },
+    {
+      "epoch": 2.153975599495162,
+      "grad_norm": 0.4212203025817871,
+      "learning_rate": 0.0002,
+      "loss": 1.6682,
+      "step": 2560
+    },
+    {
+      "epoch": 2.16238956668069,
+      "grad_norm": 0.42841723561286926,
+      "learning_rate": 0.0002,
+      "loss": 1.6036,
+      "step": 2570
+    },
+    {
+      "epoch": 2.1708035338662177,
+      "grad_norm": 0.39272481203079224,
+      "learning_rate": 0.0002,
+      "loss": 1.5962,
+      "step": 2580
+    },
+    {
+      "epoch": 2.179217501051746,
+      "grad_norm": 0.4075261354446411,
+      "learning_rate": 0.0002,
+      "loss": 1.681,
+      "step": 2590
+    },
+    {
+      "epoch": 2.187631468237274,
+      "grad_norm": 0.5358437895774841,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 2600
+    },
+    {
+      "epoch": 2.1960454354228016,
+      "grad_norm": 0.4738350212574005,
+      "learning_rate": 0.0002,
+      "loss": 1.6423,
+      "step": 2610
+    },
+    {
+      "epoch": 2.2044594026083297,
+      "grad_norm": 0.446789026260376,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 2620
+    },
+    {
+      "epoch": 2.212873369793858,
+      "grad_norm": 0.4615374505519867,
+      "learning_rate": 0.0002,
+      "loss": 1.6246,
+      "step": 2630
+    },
+    {
+      "epoch": 2.221287336979386,
+      "grad_norm": 0.46901994943618774,
+      "learning_rate": 0.0002,
+      "loss": 1.6205,
+      "step": 2640
+    },
+    {
+      "epoch": 2.2297013041649136,
+      "grad_norm": 0.46267789602279663,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 2.2381152713504417,
+      "grad_norm": 0.4383080005645752,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2660
+    },
+    {
+      "epoch": 2.24652923853597,
+      "grad_norm": 0.4070609509944916,
+      "learning_rate": 0.0002,
+      "loss": 1.5745,
+      "step": 2670
+    },
+    {
+      "epoch": 2.2549432057214975,
+      "grad_norm": 0.4572339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.6125,
+      "step": 2680
+    },
+    {
+      "epoch": 2.2633571729070256,
+      "grad_norm": 0.393265038728714,
+      "learning_rate": 0.0002,
+      "loss": 1.5671,
+      "step": 2690
+    },
+    {
+      "epoch": 2.2717711400925538,
+      "grad_norm": 0.46144717931747437,
+      "learning_rate": 0.0002,
+      "loss": 1.6239,
+      "step": 2700
+    },
+    {
+      "epoch": 2.2801851072780814,
+      "grad_norm": 0.45077767968177795,
+      "learning_rate": 0.0002,
+      "loss": 1.5992,
+      "step": 2710
+    },
+    {
+      "epoch": 2.2885990744636096,
+      "grad_norm": 0.5697639584541321,
+      "learning_rate": 0.0002,
+      "loss": 1.6261,
+      "step": 2720
+    },
+    {
+      "epoch": 2.2970130416491377,
+      "grad_norm": 0.4855510890483856,
+      "learning_rate": 0.0002,
+      "loss": 1.6192,
+      "step": 2730
+    },
+    {
+      "epoch": 2.3054270088346653,
+      "grad_norm": 0.4440622627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 2740
+    },
+    {
+      "epoch": 2.3138409760201935,
+      "grad_norm": 0.3904096782207489,
+      "learning_rate": 0.0002,
+      "loss": 1.6496,
+      "step": 2750
+    },
+    {
+      "epoch": 2.3222549432057216,
+      "grad_norm": 0.5225510597229004,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 2760
+    },
+    {
+      "epoch": 2.3306689103912497,
+      "grad_norm": 0.44866397976875305,
+      "learning_rate": 0.0002,
+      "loss": 1.6082,
+      "step": 2770
+    },
+    {
+      "epoch": 2.3390828775767774,
+      "grad_norm": 0.5167056322097778,
+      "learning_rate": 0.0002,
+      "loss": 1.6087,
+      "step": 2780
+    },
+    {
+      "epoch": 2.3474968447623055,
+      "grad_norm": 0.45913267135620117,
+      "learning_rate": 0.0002,
+      "loss": 1.6136,
+      "step": 2790
+    },
+    {
+      "epoch": 2.3559108119478336,
+      "grad_norm": 0.45787590742111206,
+      "learning_rate": 0.0002,
+      "loss": 1.6564,
+      "step": 2800
+    },
+    {
+      "epoch": 2.3643247791333613,
+      "grad_norm": 0.4633352756500244,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2810
+    },
+    {
+      "epoch": 2.3727387463188894,
+      "grad_norm": 0.46390071511268616,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 2820
+    },
+    {
+      "epoch": 2.3811527135044175,
+      "grad_norm": 0.4261005222797394,
+      "learning_rate": 0.0002,
+      "loss": 1.6039,
+      "step": 2830
+    },
+    {
+      "epoch": 2.389566680689945,
+      "grad_norm": 0.4283634424209595,
+      "learning_rate": 0.0002,
+      "loss": 1.6364,
+      "step": 2840
+    },
+    {
+      "epoch": 2.3979806478754733,
+      "grad_norm": 0.4955291450023651,
+      "learning_rate": 0.0002,
+      "loss": 1.6382,
+      "step": 2850
+    },
+    {
+      "epoch": 2.4063946150610014,
+      "grad_norm": 0.4740189015865326,
+      "learning_rate": 0.0002,
+      "loss": 1.6173,
+      "step": 2860
+    },
+    {
+      "epoch": 2.414808582246529,
+      "grad_norm": 0.4222276508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2870
+    },
+    {
+      "epoch": 2.423222549432057,
+      "grad_norm": 0.4982149004936218,
+      "learning_rate": 0.0002,
+      "loss": 1.5602,
+      "step": 2880
+    },
+    {
+      "epoch": 2.4316365166175853,
+      "grad_norm": 0.5217409133911133,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 2890
+    },
+    {
+      "epoch": 2.4400504838031134,
+      "grad_norm": 0.4555884897708893,
+      "learning_rate": 0.0002,
+      "loss": 1.5804,
+      "step": 2900
+    },
+    {
+      "epoch": 2.448464450988641,
+      "grad_norm": 0.43178579211235046,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2910
+    },
+    {
+      "epoch": 2.456878418174169,
+      "grad_norm": 0.4788478910923004,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2920
+    },
+    {
+      "epoch": 2.465292385359697,
+      "grad_norm": 0.43689873814582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 2930
+    },
+    {
+      "epoch": 2.473706352545225,
+      "grad_norm": 0.5115197896957397,
+      "learning_rate": 0.0002,
+      "loss": 1.6196,
+      "step": 2940
+    },
+    {
+      "epoch": 2.482120319730753,
+      "grad_norm": 0.5290159583091736,
+      "learning_rate": 0.0002,
+      "loss": 1.689,
+      "step": 2950
+    },
+    {
+      "epoch": 2.490534286916281,
+      "grad_norm": 0.46042463183403015,
+      "learning_rate": 0.0002,
+      "loss": 1.6499,
+      "step": 2960
+    },
+    {
+      "epoch": 2.498948254101809,
+      "grad_norm": 0.4359915852546692,
+      "learning_rate": 0.0002,
+      "loss": 1.6664,
+      "step": 2970
+    },
+    {
+      "epoch": 2.507362221287337,
+      "grad_norm": 0.46352964639663696,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 2980
+    },
+    {
+      "epoch": 2.515776188472865,
+      "grad_norm": 0.5324268341064453,
+      "learning_rate": 0.0002,
+      "loss": 1.6501,
+      "step": 2990
+    },
+    {
+      "epoch": 2.5241901556583928,
+      "grad_norm": 0.5929607152938843,
+      "learning_rate": 0.0002,
+      "loss": 1.6115,
+      "step": 3000
+    },
+    {
+      "epoch": 2.532604122843921,
+      "grad_norm": 0.4811333417892456,
+      "learning_rate": 0.0002,
+      "loss": 1.6772,
+      "step": 3010
+    },
+    {
+      "epoch": 2.541018090029449,
+      "grad_norm": 0.4662701487541199,
+      "learning_rate": 0.0002,
+      "loss": 1.7023,
+      "step": 3020
+    },
+    {
+      "epoch": 2.549432057214977,
+      "grad_norm": 0.4582270681858063,
+      "learning_rate": 0.0002,
+      "loss": 1.5426,
+      "step": 3030
+    },
+    {
+      "epoch": 2.557846024400505,
+      "grad_norm": 0.4679982662200928,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 3040
+    },
+    {
+      "epoch": 2.566259991586033,
+      "grad_norm": 0.4380294680595398,
+      "learning_rate": 0.0002,
+      "loss": 1.5442,
+      "step": 3050
+    },
+    {
+      "epoch": 2.5746739587715606,
+      "grad_norm": 0.44295763969421387,
+      "learning_rate": 0.0002,
+      "loss": 1.6055,
+      "step": 3060
+    },
+    {
+      "epoch": 2.5830879259570887,
+      "grad_norm": 0.5131027698516846,
+      "learning_rate": 0.0002,
+      "loss": 1.5775,
+      "step": 3070
+    },
+    {
+      "epoch": 2.591501893142617,
+      "grad_norm": 0.47567516565322876,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 3080
+    },
+    {
+      "epoch": 2.599915860328145,
+      "grad_norm": 0.49002596735954285,
+      "learning_rate": 0.0002,
+      "loss": 1.5671,
+      "step": 3090
+    },
+    {
+      "epoch": 2.6083298275136726,
+      "grad_norm": 0.44856327772140503,
+      "learning_rate": 0.0002,
+      "loss": 1.5445,
+      "step": 3100
+    },
+    {
+      "epoch": 2.6167437946992007,
+      "grad_norm": 0.4480142593383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5797,
+      "step": 3110
+    },
+    {
+      "epoch": 2.6251577618847284,
+      "grad_norm": 0.4317494034767151,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 3120
+    },
+    {
+      "epoch": 2.6335717290702565,
+      "grad_norm": 0.42580848932266235,
+      "learning_rate": 0.0002,
+      "loss": 1.6321,
+      "step": 3130
+    },
+    {
+      "epoch": 2.6419856962557846,
+      "grad_norm": 0.4516814947128296,
+      "learning_rate": 0.0002,
+      "loss": 1.6483,
+      "step": 3140
+    },
+    {
+      "epoch": 2.6503996634413127,
+      "grad_norm": 0.4438435733318329,
+      "learning_rate": 0.0002,
+      "loss": 1.695,
+      "step": 3150
+    },
+    {
+      "epoch": 2.6588136306268404,
+      "grad_norm": 0.4385356307029724,
+      "learning_rate": 0.0002,
+      "loss": 1.6938,
+      "step": 3160
+    },
+    {
+      "epoch": 2.6672275978123685,
+      "grad_norm": 0.5064112544059753,
+      "learning_rate": 0.0002,
+      "loss": 1.6139,
+      "step": 3170
+    },
+    {
+      "epoch": 2.6756415649978966,
+      "grad_norm": 0.49163177609443665,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 3180
+    },
+    {
+      "epoch": 2.6840555321834243,
+      "grad_norm": 0.49339258670806885,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 3190
+    },
+    {
+      "epoch": 2.6924694993689524,
+      "grad_norm": 0.440950870513916,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 3200
+    },
+    {
+      "epoch": 2.7008834665544805,
+      "grad_norm": 0.4283970594406128,
+      "learning_rate": 0.0002,
+      "loss": 1.6305,
+      "step": 3210
+    },
+    {
+      "epoch": 2.7092974337400086,
+      "grad_norm": 0.43875712156295776,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 3220
+    },
+    {
+      "epoch": 2.7177114009255363,
+      "grad_norm": 0.49332964420318604,
+      "learning_rate": 0.0002,
+      "loss": 1.6129,
+      "step": 3230
+    },
+    {
+      "epoch": 2.7261253681110644,
+      "grad_norm": 0.5225692391395569,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 3240
+    },
+    {
+      "epoch": 2.734539335296592,
+      "grad_norm": 0.4856489300727844,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 3250
+    },
+    {
+      "epoch": 2.74295330248212,
+      "grad_norm": 0.46918296813964844,
+      "learning_rate": 0.0002,
+      "loss": 1.6463,
+      "step": 3260
+    },
+    {
+      "epoch": 2.7513672696676483,
+      "grad_norm": 0.4802931249141693,
+      "learning_rate": 0.0002,
+      "loss": 1.6819,
+      "step": 3270
+    },
+    {
+      "epoch": 2.7597812368531764,
+      "grad_norm": 0.4485355615615845,
+      "learning_rate": 0.0002,
+      "loss": 1.6246,
+      "step": 3280
+    },
+    {
+      "epoch": 2.768195204038704,
+      "grad_norm": 0.43944594264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 3290
+    },
+    {
+      "epoch": 2.7766091712242322,
+      "grad_norm": 0.46847742795944214,
+      "learning_rate": 0.0002,
+      "loss": 1.6501,
+      "step": 3300
+    },
+    {
+      "epoch": 2.7850231384097603,
+      "grad_norm": 0.4816027879714966,
+      "learning_rate": 0.0002,
+      "loss": 1.5969,
+      "step": 3310
+    },
+    {
+      "epoch": 2.793437105595288,
+      "grad_norm": 0.453960120677948,
+      "learning_rate": 0.0002,
+      "loss": 1.6293,
+      "step": 3320
+    },
+    {
+      "epoch": 2.801851072780816,
+      "grad_norm": 0.4816017150878906,
+      "learning_rate": 0.0002,
+      "loss": 1.6429,
+      "step": 3330
+    },
+    {
+      "epoch": 2.8102650399663442,
+      "grad_norm": 0.4461034834384918,
+      "learning_rate": 0.0002,
+      "loss": 1.6683,
+      "step": 3340
+    },
+    {
+      "epoch": 2.8186790071518724,
+      "grad_norm": 0.48821821808815,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 3350
+    },
+    {
+      "epoch": 2.8270929743374,
+      "grad_norm": 0.4574853777885437,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 3360
+    },
+    {
+      "epoch": 2.835506941522928,
+      "grad_norm": 0.42062026262283325,
+      "learning_rate": 0.0002,
+      "loss": 1.6651,
+      "step": 3370
+    },
+    {
+      "epoch": 2.843920908708456,
+      "grad_norm": 0.4499834477901459,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 3380
+    },
+    {
+      "epoch": 2.852334875893984,
+      "grad_norm": 0.4780360758304596,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 3390
+    },
+    {
+      "epoch": 2.860748843079512,
+      "grad_norm": 0.45422887802124023,
+      "learning_rate": 0.0002,
+      "loss": 1.5882,
+      "step": 3400
+    },
+    {
+      "epoch": 2.86916281026504,
+      "grad_norm": 0.4590015709400177,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 3410
+    },
+    {
+      "epoch": 2.877576777450568,
+      "grad_norm": 0.45689624547958374,
+      "learning_rate": 0.0002,
+      "loss": 1.6746,
+      "step": 3420
+    },
+    {
+      "epoch": 2.885990744636096,
+      "grad_norm": 0.46953922510147095,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3430
+    },
+    {
+      "epoch": 2.8944047118216236,
+      "grad_norm": 0.4791966378688812,
+      "learning_rate": 0.0002,
+      "loss": 1.6015,
+      "step": 3440
+    },
+    {
+      "epoch": 2.9028186790071517,
+      "grad_norm": 0.4842296242713928,
+      "learning_rate": 0.0002,
+      "loss": 1.694,
+      "step": 3450
+    },
+    {
+      "epoch": 2.91123264619268,
+      "grad_norm": 0.47219768166542053,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3460
+    },
+    {
+      "epoch": 2.919646613378208,
+      "grad_norm": 0.4622127115726471,
+      "learning_rate": 0.0002,
+      "loss": 1.6486,
+      "step": 3470
+    },
+    {
+      "epoch": 2.9280605805637356,
+      "grad_norm": 0.46832820773124695,
+      "learning_rate": 0.0002,
+      "loss": 1.6485,
+      "step": 3480
+    },
+    {
+      "epoch": 2.9364745477492638,
+      "grad_norm": 0.44582483172416687,
+      "learning_rate": 0.0002,
+      "loss": 1.6366,
+      "step": 3490
+    },
+    {
+      "epoch": 2.944888514934792,
+      "grad_norm": 0.4987219274044037,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 3500
+    },
+    {
+      "epoch": 2.9533024821203195,
+      "grad_norm": 0.43750956654548645,
+      "learning_rate": 0.0002,
+      "loss": 1.5991,
+      "step": 3510
+    },
+    {
+      "epoch": 2.9617164493058477,
+      "grad_norm": 0.49962925910949707,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 3520
+    },
+    {
+      "epoch": 2.9701304164913758,
+      "grad_norm": 0.5189590454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 3530
+    },
+    {
+      "epoch": 2.978544383676904,
+      "grad_norm": 0.391317754983902,
+      "learning_rate": 0.0002,
+      "loss": 1.6688,
+      "step": 3540
+    },
+    {
+      "epoch": 2.9869583508624316,
+      "grad_norm": 0.44934695959091187,
+      "learning_rate": 0.0002,
+      "loss": 1.5884,
+      "step": 3550
+    },
+    {
+      "epoch": 2.9953723180479597,
+      "grad_norm": 0.4740142226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.5688,
+      "step": 3560
+    },
+    {
+      "epoch": 2.9995793016407237,
+      "eval_loss": 1.8266887664794922,
+      "eval_runtime": 37.9445,
+      "eval_samples_per_second": 13.572,
+      "eval_steps_per_second": 1.713,
+      "step": 3565
+    },
+    {
+      "epoch": 3.003786285233488,
+      "grad_norm": 0.4523724615573883,
+      "learning_rate": 0.0002,
+      "loss": 1.5939,
+      "step": 3570
+    },
+    {
+      "epoch": 3.0122002524190155,
+      "grad_norm": 0.5261380076408386,
+      "learning_rate": 0.0002,
+      "loss": 1.526,
+      "step": 3580
+    },
+    {
+      "epoch": 3.0206142196045436,
+      "grad_norm": 0.48664888739585876,
+      "learning_rate": 0.0002,
+      "loss": 1.4946,
+      "step": 3590
+    },
+    {
+      "epoch": 3.0290281867900717,
+      "grad_norm": 0.5070882439613342,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 3600
+    },
+    {
+      "epoch": 3.0374421539755994,
+      "grad_norm": 0.5816011428833008,
+      "learning_rate": 0.0002,
+      "loss": 1.5316,
+      "step": 3610
+    },
+    {
+      "epoch": 3.0458561211611275,
+      "grad_norm": 0.6610211730003357,
+      "learning_rate": 0.0002,
+      "loss": 1.5682,
+      "step": 3620
+    },
+    {
+      "epoch": 3.0542700883466556,
+      "grad_norm": 0.5257703065872192,
+      "learning_rate": 0.0002,
+      "loss": 1.5699,
+      "step": 3630
+    },
+    {
+      "epoch": 3.0626840555321833,
+      "grad_norm": 0.5574390888214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4438,
+      "step": 3640
+    },
+    {
+      "epoch": 3.0710980227177114,
+      "grad_norm": 0.5682297348976135,
+      "learning_rate": 0.0002,
+      "loss": 1.547,
+      "step": 3650
+    },
+    {
+      "epoch": 3.0795119899032395,
+      "grad_norm": 0.5798383355140686,
+      "learning_rate": 0.0002,
+      "loss": 1.5743,
+      "step": 3660
+    },
+    {
+      "epoch": 3.087925957088767,
+      "grad_norm": 0.5458289980888367,
+      "learning_rate": 0.0002,
+      "loss": 1.4339,
+      "step": 3670
+    },
+    {
+      "epoch": 3.0963399242742953,
+      "grad_norm": 0.5599102973937988,
+      "learning_rate": 0.0002,
+      "loss": 1.46,
+      "step": 3680
+    },
+    {
+      "epoch": 3.1047538914598234,
+      "grad_norm": 0.5023021697998047,
+      "learning_rate": 0.0002,
+      "loss": 1.4589,
+      "step": 3690
+    },
+    {
+      "epoch": 3.113167858645351,
+      "grad_norm": 0.5448206067085266,
+      "learning_rate": 0.0002,
+      "loss": 1.5114,
+      "step": 3700
+    },
+    {
+      "epoch": 3.121581825830879,
+      "grad_norm": 0.5760458707809448,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 3710
+    },
+    {
+      "epoch": 3.1299957930164073,
+      "grad_norm": 0.6018968224525452,
+      "learning_rate": 0.0002,
+      "loss": 1.4789,
+      "step": 3720
+    },
+    {
+      "epoch": 3.1384097602019354,
+      "grad_norm": 0.5767101049423218,
+      "learning_rate": 0.0002,
+      "loss": 1.5518,
+      "step": 3730
+    },
+    {
+      "epoch": 3.146823727387463,
+      "grad_norm": 0.5333963632583618,
+      "learning_rate": 0.0002,
+      "loss": 1.5032,
+      "step": 3740
+    },
+    {
+      "epoch": 3.155237694572991,
+      "grad_norm": 0.5918396711349487,
+      "learning_rate": 0.0002,
+      "loss": 1.4812,
+      "step": 3750
+    },
+    {
+      "epoch": 3.1636516617585193,
+      "grad_norm": 0.5931203365325928,
+      "learning_rate": 0.0002,
+      "loss": 1.4618,
+      "step": 3760
+    },
+    {
+      "epoch": 3.172065628944047,
+      "grad_norm": 0.6562168598175049,
+      "learning_rate": 0.0002,
+      "loss": 1.5592,
+      "step": 3770
+    },
+    {
+      "epoch": 3.180479596129575,
+      "grad_norm": 0.5820156335830688,
+      "learning_rate": 0.0002,
+      "loss": 1.4932,
+      "step": 3780
+    },
+    {
+      "epoch": 3.188893563315103,
+      "grad_norm": 0.5784737467765808,
+      "learning_rate": 0.0002,
+      "loss": 1.4523,
+      "step": 3790
+    },
+    {
+      "epoch": 3.197307530500631,
+      "grad_norm": 0.5506529808044434,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 3800
+    },
+    {
+      "epoch": 3.205721497686159,
+      "grad_norm": 0.6101595163345337,
+      "learning_rate": 0.0002,
+      "loss": 1.4819,
+      "step": 3810
+    },
+    {
+      "epoch": 3.214135464871687,
+      "grad_norm": 0.5597806572914124,
+      "learning_rate": 0.0002,
+      "loss": 1.5185,
+      "step": 3820
+    },
+    {
+      "epoch": 3.222549432057215,
+      "grad_norm": 0.5641011595726013,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 3830
+    },
+    {
+      "epoch": 3.230963399242743,
+      "grad_norm": 0.5892080068588257,
+      "learning_rate": 0.0002,
+      "loss": 1.4702,
+      "step": 3840
+    },
+    {
+      "epoch": 3.239377366428271,
+      "grad_norm": 0.6034760475158691,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 3850
+    },
+    {
+      "epoch": 3.247791333613799,
+      "grad_norm": 0.5112439393997192,
+      "learning_rate": 0.0002,
+      "loss": 1.5499,
+      "step": 3860
+    },
+    {
+      "epoch": 3.256205300799327,
+      "grad_norm": 0.56565922498703,
+      "learning_rate": 0.0002,
+      "loss": 1.5132,
+      "step": 3870
+    },
+    {
+      "epoch": 3.264619267984855,
+      "grad_norm": 0.6155247092247009,
+      "learning_rate": 0.0002,
+      "loss": 1.4892,
+      "step": 3880
+    },
+    {
+      "epoch": 3.273033235170383,
+      "grad_norm": 0.6064623594284058,
+      "learning_rate": 0.0002,
+      "loss": 1.5118,
+      "step": 3890
+    },
+    {
+      "epoch": 3.2814472023559107,
+      "grad_norm": 0.6313768029212952,
+      "learning_rate": 0.0002,
+      "loss": 1.5236,
+      "step": 3900
+    },
+    {
+      "epoch": 3.289861169541439,
+      "grad_norm": 0.5903939008712769,
+      "learning_rate": 0.0002,
+      "loss": 1.5551,
+      "step": 3910
+    },
+    {
+      "epoch": 3.298275136726967,
+      "grad_norm": 0.5770667195320129,
+      "learning_rate": 0.0002,
+      "loss": 1.5703,
+      "step": 3920
+    },
+    {
+      "epoch": 3.3066891039124946,
+      "grad_norm": 0.5785196423530579,
+      "learning_rate": 0.0002,
+      "loss": 1.5159,
+      "step": 3930
+    },
+    {
+      "epoch": 3.3151030710980227,
+      "grad_norm": 0.6468310356140137,
+      "learning_rate": 0.0002,
+      "loss": 1.5277,
+      "step": 3940
+    },
+    {
+      "epoch": 3.323517038283551,
+      "grad_norm": 0.6200279593467712,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 3950
+    },
+    {
+      "epoch": 3.3319310054690785,
+      "grad_norm": 0.5779302716255188,
+      "learning_rate": 0.0002,
+      "loss": 1.5264,
+      "step": 3960
+    },
+    {
+      "epoch": 3.3403449726546066,
+      "grad_norm": 0.5463796854019165,
+      "learning_rate": 0.0002,
+      "loss": 1.4861,
+      "step": 3970
+    },
+    {
+      "epoch": 3.3487589398401347,
+      "grad_norm": 0.6117855906486511,
+      "learning_rate": 0.0002,
+      "loss": 1.541,
+      "step": 3980
+    },
+    {
+      "epoch": 3.357172907025663,
+      "grad_norm": 0.5554766058921814,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 3990
+    },
+    {
+      "epoch": 3.3655868742111905,
+      "grad_norm": 0.6012870073318481,
+      "learning_rate": 0.0002,
+      "loss": 1.5004,
+      "step": 4000
+    },
+    {
+      "epoch": 3.3740008413967186,
+      "grad_norm": 0.5443974137306213,
+      "learning_rate": 0.0002,
+      "loss": 1.473,
+      "step": 4010
+    },
+    {
+      "epoch": 3.3824148085822463,
+      "grad_norm": 0.6636057496070862,
+      "learning_rate": 0.0002,
+      "loss": 1.5139,
+      "step": 4020
+    },
+    {
+      "epoch": 3.3908287757677744,
+      "grad_norm": 0.5801246166229248,
+      "learning_rate": 0.0002,
+      "loss": 1.5141,
+      "step": 4030
+    },
+    {
+      "epoch": 3.3992427429533025,
+      "grad_norm": 0.5668839812278748,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4040
+    },
+    {
+      "epoch": 3.4076567101388306,
+      "grad_norm": 0.7763481736183167,
+      "learning_rate": 0.0002,
+      "loss": 1.523,
+      "step": 4050
+    },
+    {
+      "epoch": 3.4160706773243583,
+      "grad_norm": 0.6675992608070374,
+      "learning_rate": 0.0002,
+      "loss": 1.4932,
+      "step": 4060
+    },
+    {
+      "epoch": 3.4244846445098864,
+      "grad_norm": 0.6290077567100525,
+      "learning_rate": 0.0002,
+      "loss": 1.4959,
+      "step": 4070
+    },
+    {
+      "epoch": 3.4328986116954145,
+      "grad_norm": 0.6040239930152893,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 4080
+    },
+    {
+      "epoch": 3.441312578880942,
+      "grad_norm": 0.6237877607345581,
+      "learning_rate": 0.0002,
+      "loss": 1.5711,
+      "step": 4090
+    },
+    {
+      "epoch": 3.4497265460664703,
+      "grad_norm": 0.5343508124351501,
+      "learning_rate": 0.0002,
+      "loss": 1.4961,
+      "step": 4100
+    },
+    {
+      "epoch": 3.4581405132519984,
+      "grad_norm": 0.6817412972450256,
+      "learning_rate": 0.0002,
+      "loss": 1.5123,
+      "step": 4110
+    },
+    {
+      "epoch": 3.466554480437526,
+      "grad_norm": 0.7115170359611511,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 4120
+    },
+    {
+      "epoch": 3.4749684476230542,
+      "grad_norm": 0.6127332448959351,
+      "learning_rate": 0.0002,
+      "loss": 1.5275,
+      "step": 4130
+    },
+    {
+      "epoch": 3.4833824148085824,
+      "grad_norm": 0.5745994448661804,
+      "learning_rate": 0.0002,
+      "loss": 1.557,
+      "step": 4140
+    },
+    {
+      "epoch": 3.49179638199411,
+      "grad_norm": 0.6248795390129089,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4150
+    },
+    {
+      "epoch": 3.500210349179638,
+      "grad_norm": 0.5821124911308289,
+      "learning_rate": 0.0002,
+      "loss": 1.4885,
+      "step": 4160
+    },
+    {
+      "epoch": 3.5086243163651663,
+      "grad_norm": 0.561416506767273,
+      "learning_rate": 0.0002,
+      "loss": 1.4937,
+      "step": 4170
+    },
+    {
+      "epoch": 3.5170382835506944,
+      "grad_norm": 0.5848962664604187,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 4180
+    },
+    {
+      "epoch": 3.525452250736222,
+      "grad_norm": 0.5335569977760315,
+      "learning_rate": 0.0002,
+      "loss": 1.5892,
+      "step": 4190
+    },
+    {
+      "epoch": 3.53386621792175,
+      "grad_norm": 0.547964870929718,
+      "learning_rate": 0.0002,
+      "loss": 1.5152,
+      "step": 4200
+    },
+    {
+      "epoch": 3.542280185107278,
+      "grad_norm": 0.6157727241516113,
+      "learning_rate": 0.0002,
+      "loss": 1.4887,
+      "step": 4210
+    },
+    {
+      "epoch": 3.550694152292806,
+      "grad_norm": 0.6163121461868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5484,
+      "step": 4220
+    },
+    {
+      "epoch": 3.559108119478334,
+      "grad_norm": 0.5844616293907166,
+      "learning_rate": 0.0002,
+      "loss": 1.5833,
+      "step": 4230
+    },
+    {
+      "epoch": 3.567522086663862,
+      "grad_norm": 0.7104926109313965,
+      "learning_rate": 0.0002,
+      "loss": 1.5305,
+      "step": 4240
+    },
+    {
+      "epoch": 3.57593605384939,
+      "grad_norm": 0.5055213570594788,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4250
+    },
+    {
+      "epoch": 3.584350021034918,
+      "grad_norm": 0.611676812171936,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 4260
+    },
+    {
+      "epoch": 3.592763988220446,
+      "grad_norm": 0.6326440572738647,
+      "learning_rate": 0.0002,
+      "loss": 1.5048,
+      "step": 4270
+    },
+    {
+      "epoch": 3.6011779554059737,
+      "grad_norm": 0.6290925741195679,
+      "learning_rate": 0.0002,
+      "loss": 1.5122,
+      "step": 4280
+    },
+    {
+      "epoch": 3.609591922591502,
+      "grad_norm": 0.5691978931427002,
+      "learning_rate": 0.0002,
+      "loss": 1.5654,
+      "step": 4290
+    },
+    {
+      "epoch": 3.61800588977703,
+      "grad_norm": 0.6071329116821289,
+      "learning_rate": 0.0002,
+      "loss": 1.4854,
+      "step": 4300
+    },
+    {
+      "epoch": 3.626419856962558,
+      "grad_norm": 0.606573224067688,
+      "learning_rate": 0.0002,
+      "loss": 1.5336,
+      "step": 4310
+    },
+    {
+      "epoch": 3.6348338241480858,
+      "grad_norm": 0.5515419244766235,
+      "learning_rate": 0.0002,
+      "loss": 1.6437,
+      "step": 4320
+    },
+    {
+      "epoch": 3.643247791333614,
+      "grad_norm": 0.5964660048484802,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 4330
+    },
+    {
+      "epoch": 3.6516617585191415,
+      "grad_norm": 0.5774146914482117,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 4340
+    },
+    {
+      "epoch": 3.6600757257046697,
+      "grad_norm": 0.5732731223106384,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 4350
+    },
+    {
+      "epoch": 3.6684896928901978,
+      "grad_norm": 0.7354163527488708,
+      "learning_rate": 0.0002,
+      "loss": 1.5682,
+      "step": 4360
+    },
+    {
+      "epoch": 3.676903660075726,
+      "grad_norm": 0.6220902800559998,
+      "learning_rate": 0.0002,
+      "loss": 1.5225,
+      "step": 4370
+    },
+    {
+      "epoch": 3.6853176272612536,
+      "grad_norm": 0.6053991317749023,
+      "learning_rate": 0.0002,
+      "loss": 1.4838,
+      "step": 4380
+    },
+    {
+      "epoch": 3.6937315944467817,
+      "grad_norm": 0.67010897397995,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4390
+    },
+    {
+      "epoch": 3.70214556163231,
+      "grad_norm": 0.6139186024665833,
+      "learning_rate": 0.0002,
+      "loss": 1.5381,
+      "step": 4400
+    },
+    {
+      "epoch": 3.7105595288178375,
+      "grad_norm": 0.5433071851730347,
+      "learning_rate": 0.0002,
+      "loss": 1.5088,
+      "step": 4410
+    },
+    {
+      "epoch": 3.7189734960033656,
+      "grad_norm": 0.5453870296478271,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 4420
+    },
+    {
+      "epoch": 3.7273874631888937,
+      "grad_norm": 0.6401727199554443,
+      "learning_rate": 0.0002,
+      "loss": 1.4549,
+      "step": 4430
+    },
+    {
+      "epoch": 3.735801430374422,
+      "grad_norm": 0.6049367189407349,
+      "learning_rate": 0.0002,
+      "loss": 1.503,
+      "step": 4440
+    },
+    {
+      "epoch": 3.7442153975599495,
+      "grad_norm": 0.5740529298782349,
+      "learning_rate": 0.0002,
+      "loss": 1.5268,
+      "step": 4450
+    },
+    {
+      "epoch": 3.7526293647454776,
+      "grad_norm": 0.6521880626678467,
+      "learning_rate": 0.0002,
+      "loss": 1.5183,
+      "step": 4460
+    },
+    {
+      "epoch": 3.7610433319310053,
+      "grad_norm": 0.7096368074417114,
+      "learning_rate": 0.0002,
+      "loss": 1.5741,
+      "step": 4470
+    },
+    {
+      "epoch": 3.7694572991165334,
+      "grad_norm": 0.5886474251747131,
+      "learning_rate": 0.0002,
+      "loss": 1.5786,
+      "step": 4480
+    },
+    {
+      "epoch": 3.7778712663020615,
+      "grad_norm": 0.5821043252944946,
+      "learning_rate": 0.0002,
+      "loss": 1.5887,
+      "step": 4490
+    },
+    {
+      "epoch": 3.7862852334875896,
+      "grad_norm": 0.628892183303833,
+      "learning_rate": 0.0002,
+      "loss": 1.5777,
+      "step": 4500
+    },
+    {
+      "epoch": 3.7946992006731173,
+      "grad_norm": 0.5962669849395752,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 4510
+    },
+    {
+      "epoch": 3.8031131678586454,
+      "grad_norm": 0.6635549068450928,
+      "learning_rate": 0.0002,
+      "loss": 1.5267,
+      "step": 4520
+    },
+    {
+      "epoch": 3.811527135044173,
+      "grad_norm": 0.6010760068893433,
+      "learning_rate": 0.0002,
+      "loss": 1.5058,
+      "step": 4530
+    },
+    {
+      "epoch": 3.819941102229701,
+      "grad_norm": 0.6322658658027649,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 4540
+    },
+    {
+      "epoch": 3.8283550694152293,
+      "grad_norm": 0.5893137454986572,
+      "learning_rate": 0.0002,
+      "loss": 1.5029,
+      "step": 4550
+    },
+    {
+      "epoch": 3.8367690366007574,
+      "grad_norm": 0.7829602360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.5435,
+      "step": 4560
+    },
+    {
+      "epoch": 3.845183003786285,
+      "grad_norm": 0.6190396547317505,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 4570
+    },
+    {
+      "epoch": 3.853596970971813,
+      "grad_norm": 0.6662813425064087,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 4580
+    },
+    {
+      "epoch": 3.8620109381573413,
+      "grad_norm": 0.5809855461120605,
+      "learning_rate": 0.0002,
+      "loss": 1.5065,
+      "step": 4590
+    },
+    {
+      "epoch": 3.870424905342869,
+      "grad_norm": 0.5779069662094116,
+      "learning_rate": 0.0002,
+      "loss": 1.5041,
+      "step": 4600
+    },
+    {
+      "epoch": 3.878838872528397,
+      "grad_norm": 0.5603038668632507,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 4610
+    },
+    {
+      "epoch": 3.887252839713925,
+      "grad_norm": 0.6274181008338928,
+      "learning_rate": 0.0002,
+      "loss": 1.5372,
+      "step": 4620
+    },
+    {
+      "epoch": 3.8956668068994533,
+      "grad_norm": 0.6810959577560425,
+      "learning_rate": 0.0002,
+      "loss": 1.4996,
+      "step": 4630
+    },
+    {
+      "epoch": 3.904080774084981,
+      "grad_norm": 0.5647315979003906,
+      "learning_rate": 0.0002,
+      "loss": 1.4956,
+      "step": 4640
+    },
+    {
+      "epoch": 3.912494741270509,
+      "grad_norm": 0.6830295324325562,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 4650
+    },
+    {
+      "epoch": 3.920908708456037,
+      "grad_norm": 0.652565598487854,
+      "learning_rate": 0.0002,
+      "loss": 1.535,
+      "step": 4660
+    },
+    {
+      "epoch": 3.929322675641565,
+      "grad_norm": 0.5806284546852112,
+      "learning_rate": 0.0002,
+      "loss": 1.4772,
+      "step": 4670
+    },
+    {
+      "epoch": 3.937736642827093,
+      "grad_norm": 0.6825073957443237,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 4680
+    },
+    {
+      "epoch": 3.946150610012621,
+      "grad_norm": 0.6149451732635498,
+      "learning_rate": 0.0002,
+      "loss": 1.5516,
+      "step": 4690
+    },
+    {
+      "epoch": 3.954564577198149,
+      "grad_norm": 0.6152557134628296,
+      "learning_rate": 0.0002,
+      "loss": 1.5608,
+      "step": 4700
+    },
+    {
+      "epoch": 3.962978544383677,
+      "grad_norm": 0.6239011883735657,
+      "learning_rate": 0.0002,
+      "loss": 1.4897,
+      "step": 4710
+    },
+    {
+      "epoch": 3.971392511569205,
+      "grad_norm": 0.6485443115234375,
+      "learning_rate": 0.0002,
+      "loss": 1.538,
+      "step": 4720
+    },
+    {
+      "epoch": 3.9798064787547327,
+      "grad_norm": 0.6449228525161743,
+      "learning_rate": 0.0002,
+      "loss": 1.5226,
+      "step": 4730
+    },
+    {
+      "epoch": 3.988220445940261,
+      "grad_norm": 0.6526407599449158,
+      "learning_rate": 0.0002,
+      "loss": 1.5087,
+      "step": 4740
+    },
+    {
+      "epoch": 3.996634413125789,
+      "grad_norm": 0.6277706027030945,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4750
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.871641755104065,
+      "eval_runtime": 37.9637,
+      "eval_samples_per_second": 13.566,
+      "eval_steps_per_second": 1.712,
+      "step": 4754
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 9504,
+  "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": 2.2000447536286925e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/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-0.8-num-5623-sd-42/checkpoint-4754/training_args.bin b/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-0.8-num-5623-sd-42/checkpoint-4754/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f1502d478cfbb1424f707352d007b740bde5e373
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-4754/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df2b79d3acefeedef5a0229881de39ec68ef9b40046a60d7976a49f7e6b3b936
+size 5560
diff --git a/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-0.8-num-5623-sd-42/checkpoint-5942/README.md b/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-0.8-num-5623-sd-42/checkpoint-5942/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-5942/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_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-0.8-num-5623-sd-42/checkpoint-5942/adapter_config.json b/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-0.8-num-5623-sd-42/checkpoint-5942/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-5942/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_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-0.8-num-5623-sd-42/checkpoint-5942/adapter_model.safetensors b/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-0.8-num-5623-sd-42/checkpoint-5942/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2703957f8d94a723eab1fe0b0ddbf31cb25ed6ec
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-5942/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6018665777bb34b616cd34fa662867cfff157e84dc6d90c38e4e02be569b37a5
+size 109069176
diff --git a/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-0.8-num-5623-sd-42/checkpoint-5942/optimizer.pt b/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-0.8-num-5623-sd-42/checkpoint-5942/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..04de002f07da9be69e1c8ba2e5f8e331e02f98b4
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-5942/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0441013b17afbf1cf92a2b1afa32eb7d2412ef831ed1b830a1dcee519ff7a238
+size 55532666
diff --git a/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-0.8-num-5623-sd-42/checkpoint-5942/rng_state.pth b/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-0.8-num-5623-sd-42/checkpoint-5942/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..55f0cc2def4211824a3a8e06f82b1a470ad9dafe
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-5942/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a36eb29aabc5fe6416332a206ea0c4dbeb34353f5b7eaf5a21c684c91e17bf40
+size 14244
diff --git a/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-0.8-num-5623-sd-42/checkpoint-5942/scheduler.pt b/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-0.8-num-5623-sd-42/checkpoint-5942/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..55843a93ad559b4be7fa0c6d8fa39bfe1db40228
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-5942/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a95a8659297b5b75e90aa56fc1ddf4df2316d927d2ed84dbcf2a6a29fc7ad9c1
+size 1064
diff --git a/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-0.8-num-5623-sd-42/checkpoint-5942/special_tokens_map.json b/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-0.8-num-5623-sd-42/checkpoint-5942/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-5942/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_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-0.8-num-5623-sd-42/checkpoint-5942/tokenizer.json b/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-0.8-num-5623-sd-42/checkpoint-5942/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-5942/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_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-0.8-num-5623-sd-42/checkpoint-5942/tokenizer_config.json b/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-0.8-num-5623-sd-42/checkpoint-5942/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-5942/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_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-0.8-num-5623-sd-42/checkpoint-5942/trainer_state.json b/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-0.8-num-5623-sd-42/checkpoint-5942/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..77ea3a9b7f5d93e668e7c59a41c648dae57eda5f
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-5942/trainer_state.json
@@ -0,0 +1,4231 @@
+{
+  "best_metric": 1.8055059909820557,
+  "best_model_checkpoint": "outputs-001/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-0.8-num-5623-sd-42/checkpoint-2377",
+  "epoch": 4.999579301640724,
+  "eval_steps": 10,
+  "global_step": 5942,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008413967185527976,
+      "grad_norm": 0.5458821654319763,
+      "learning_rate": 0.0002,
+      "loss": 2.56,
+      "step": 10
+    },
+    {
+      "epoch": 0.016827934371055953,
+      "grad_norm": 0.7293308973312378,
+      "learning_rate": 0.0002,
+      "loss": 2.3235,
+      "step": 20
+    },
+    {
+      "epoch": 0.02524190155658393,
+      "grad_norm": 0.47792306542396545,
+      "learning_rate": 0.0002,
+      "loss": 2.0815,
+      "step": 30
+    },
+    {
+      "epoch": 0.033655868742111905,
+      "grad_norm": 0.5944402813911438,
+      "learning_rate": 0.0002,
+      "loss": 1.9718,
+      "step": 40
+    },
+    {
+      "epoch": 0.04206983592763988,
+      "grad_norm": 0.5415359735488892,
+      "learning_rate": 0.0002,
+      "loss": 1.8848,
+      "step": 50
+    },
+    {
+      "epoch": 0.05048380311316786,
+      "grad_norm": 0.535713791847229,
+      "learning_rate": 0.0002,
+      "loss": 1.8953,
+      "step": 60
+    },
+    {
+      "epoch": 0.058897770298695834,
+      "grad_norm": 0.5184146761894226,
+      "learning_rate": 0.0002,
+      "loss": 1.937,
+      "step": 70
+    },
+    {
+      "epoch": 0.06731173748422381,
+      "grad_norm": 0.458926796913147,
+      "learning_rate": 0.0002,
+      "loss": 1.8396,
+      "step": 80
+    },
+    {
+      "epoch": 0.07572570466975179,
+      "grad_norm": 0.4780142307281494,
+      "learning_rate": 0.0002,
+      "loss": 1.8677,
+      "step": 90
+    },
+    {
+      "epoch": 0.08413967185527976,
+      "grad_norm": 0.79965740442276,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 100
+    },
+    {
+      "epoch": 0.09255363904080774,
+      "grad_norm": 0.4498862028121948,
+      "learning_rate": 0.0002,
+      "loss": 1.9081,
+      "step": 110
+    },
+    {
+      "epoch": 0.10096760622633572,
+      "grad_norm": 0.39338430762290955,
+      "learning_rate": 0.0002,
+      "loss": 1.8503,
+      "step": 120
+    },
+    {
+      "epoch": 0.10938157341186369,
+      "grad_norm": 0.9588953852653503,
+      "learning_rate": 0.0002,
+      "loss": 1.8637,
+      "step": 130
+    },
+    {
+      "epoch": 0.11779554059739167,
+      "grad_norm": 0.41675639152526855,
+      "learning_rate": 0.0002,
+      "loss": 1.8676,
+      "step": 140
+    },
+    {
+      "epoch": 0.12620950778291964,
+      "grad_norm": 0.44519832730293274,
+      "learning_rate": 0.0002,
+      "loss": 1.8904,
+      "step": 150
+    },
+    {
+      "epoch": 0.13462347496844762,
+      "grad_norm": 0.4176260530948639,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 160
+    },
+    {
+      "epoch": 0.1430374421539756,
+      "grad_norm": 0.35840365290641785,
+      "learning_rate": 0.0002,
+      "loss": 1.8398,
+      "step": 170
+    },
+    {
+      "epoch": 0.15145140933950357,
+      "grad_norm": 0.3794495463371277,
+      "learning_rate": 0.0002,
+      "loss": 1.8666,
+      "step": 180
+    },
+    {
+      "epoch": 0.15986537652503155,
+      "grad_norm": 0.4563522934913635,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 190
+    },
+    {
+      "epoch": 0.16827934371055953,
+      "grad_norm": 0.37057486176490784,
+      "learning_rate": 0.0002,
+      "loss": 1.8893,
+      "step": 200
+    },
+    {
+      "epoch": 0.1766933108960875,
+      "grad_norm": 0.44081518054008484,
+      "learning_rate": 0.0002,
+      "loss": 1.7995,
+      "step": 210
+    },
+    {
+      "epoch": 0.18510727808161548,
+      "grad_norm": 0.46078577637672424,
+      "learning_rate": 0.0002,
+      "loss": 1.9048,
+      "step": 220
+    },
+    {
+      "epoch": 0.19352124526714345,
+      "grad_norm": 0.36132094264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.8403,
+      "step": 230
+    },
+    {
+      "epoch": 0.20193521245267143,
+      "grad_norm": 0.3747289180755615,
+      "learning_rate": 0.0002,
+      "loss": 1.8827,
+      "step": 240
+    },
+    {
+      "epoch": 0.2103491796381994,
+      "grad_norm": 0.3540179133415222,
+      "learning_rate": 0.0002,
+      "loss": 1.8382,
+      "step": 250
+    },
+    {
+      "epoch": 0.21876314682372738,
+      "grad_norm": 0.3461375832557678,
+      "learning_rate": 0.0002,
+      "loss": 1.8196,
+      "step": 260
+    },
+    {
+      "epoch": 0.22717711400925536,
+      "grad_norm": 0.3436960279941559,
+      "learning_rate": 0.0002,
+      "loss": 1.8509,
+      "step": 270
+    },
+    {
+      "epoch": 0.23559108119478334,
+      "grad_norm": 0.35403719544410706,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 280
+    },
+    {
+      "epoch": 0.2440050483803113,
+      "grad_norm": 0.37142616510391235,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 290
+    },
+    {
+      "epoch": 0.2524190155658393,
+      "grad_norm": 0.3307955861091614,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 300
+    },
+    {
+      "epoch": 0.2608329827513673,
+      "grad_norm": 0.32855314016342163,
+      "learning_rate": 0.0002,
+      "loss": 1.817,
+      "step": 310
+    },
+    {
+      "epoch": 0.26924694993689524,
+      "grad_norm": 0.3299003839492798,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 320
+    },
+    {
+      "epoch": 0.27766091712242325,
+      "grad_norm": 0.44311287999153137,
+      "learning_rate": 0.0002,
+      "loss": 1.8129,
+      "step": 330
+    },
+    {
+      "epoch": 0.2860748843079512,
+      "grad_norm": 0.32989758253097534,
+      "learning_rate": 0.0002,
+      "loss": 1.8232,
+      "step": 340
+    },
+    {
+      "epoch": 0.2944888514934792,
+      "grad_norm": 0.34400200843811035,
+      "learning_rate": 0.0002,
+      "loss": 1.7716,
+      "step": 350
+    },
+    {
+      "epoch": 0.30290281867900715,
+      "grad_norm": 0.36286211013793945,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 360
+    },
+    {
+      "epoch": 0.31131678586453515,
+      "grad_norm": 0.406827837228775,
+      "learning_rate": 0.0002,
+      "loss": 1.8025,
+      "step": 370
+    },
+    {
+      "epoch": 0.3197307530500631,
+      "grad_norm": 0.36299195885658264,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 380
+    },
+    {
+      "epoch": 0.3281447202355911,
+      "grad_norm": 0.3477257192134857,
+      "learning_rate": 0.0002,
+      "loss": 1.837,
+      "step": 390
+    },
+    {
+      "epoch": 0.33655868742111905,
+      "grad_norm": 0.3730369210243225,
+      "learning_rate": 0.0002,
+      "loss": 1.7767,
+      "step": 400
+    },
+    {
+      "epoch": 0.34497265460664706,
+      "grad_norm": 0.4644559919834137,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 410
+    },
+    {
+      "epoch": 0.353386621792175,
+      "grad_norm": 0.406576544046402,
+      "learning_rate": 0.0002,
+      "loss": 1.7538,
+      "step": 420
+    },
+    {
+      "epoch": 0.361800588977703,
+      "grad_norm": 0.3612699508666992,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 430
+    },
+    {
+      "epoch": 0.37021455616323096,
+      "grad_norm": 0.3243742287158966,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 440
+    },
+    {
+      "epoch": 0.37862852334875896,
+      "grad_norm": 0.36671221256256104,
+      "learning_rate": 0.0002,
+      "loss": 1.8851,
+      "step": 450
+    },
+    {
+      "epoch": 0.3870424905342869,
+      "grad_norm": 0.3565002381801605,
+      "learning_rate": 0.0002,
+      "loss": 1.8853,
+      "step": 460
+    },
+    {
+      "epoch": 0.3954564577198149,
+      "grad_norm": 0.34630221128463745,
+      "learning_rate": 0.0002,
+      "loss": 1.8923,
+      "step": 470
+    },
+    {
+      "epoch": 0.40387042490534286,
+      "grad_norm": 0.3353537321090698,
+      "learning_rate": 0.0002,
+      "loss": 1.8234,
+      "step": 480
+    },
+    {
+      "epoch": 0.41228439209087087,
+      "grad_norm": 0.4015921950340271,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 490
+    },
+    {
+      "epoch": 0.4206983592763988,
+      "grad_norm": 0.5489419102668762,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 500
+    },
+    {
+      "epoch": 0.4291123264619268,
+      "grad_norm": 0.4193589985370636,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 510
+    },
+    {
+      "epoch": 0.43752629364745477,
+      "grad_norm": 0.3418922424316406,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 520
+    },
+    {
+      "epoch": 0.44594026083298277,
+      "grad_norm": 0.32668185234069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 530
+    },
+    {
+      "epoch": 0.4543542280185107,
+      "grad_norm": 0.3094325661659241,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 540
+    },
+    {
+      "epoch": 0.4627681952040387,
+      "grad_norm": 0.3743017315864563,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 550
+    },
+    {
+      "epoch": 0.47118216238956667,
+      "grad_norm": 0.3295630216598511,
+      "learning_rate": 0.0002,
+      "loss": 1.8451,
+      "step": 560
+    },
+    {
+      "epoch": 0.4795961295750947,
+      "grad_norm": 1.6124513149261475,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 570
+    },
+    {
+      "epoch": 0.4880100967606226,
+      "grad_norm": 0.3245585858821869,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 580
+    },
+    {
+      "epoch": 0.49642406394615063,
+      "grad_norm": 0.3332934081554413,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 590
+    },
+    {
+      "epoch": 0.5048380311316786,
+      "grad_norm": 0.3836138844490051,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 600
+    },
+    {
+      "epoch": 0.5132519983172066,
+      "grad_norm": 0.32953888177871704,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 610
+    },
+    {
+      "epoch": 0.5216659655027346,
+      "grad_norm": 0.36291512846946716,
+      "learning_rate": 0.0002,
+      "loss": 1.7729,
+      "step": 620
+    },
+    {
+      "epoch": 0.5300799326882625,
+      "grad_norm": 0.3237783908843994,
+      "learning_rate": 0.0002,
+      "loss": 1.7758,
+      "step": 630
+    },
+    {
+      "epoch": 0.5384938998737905,
+      "grad_norm": 0.38882696628570557,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 640
+    },
+    {
+      "epoch": 0.5469078670593185,
+      "grad_norm": 0.37821972370147705,
+      "learning_rate": 0.0002,
+      "loss": 1.8624,
+      "step": 650
+    },
+    {
+      "epoch": 0.5553218342448465,
+      "grad_norm": 0.3556285500526428,
+      "learning_rate": 0.0002,
+      "loss": 1.8075,
+      "step": 660
+    },
+    {
+      "epoch": 0.5637358014303744,
+      "grad_norm": 0.347499281167984,
+      "learning_rate": 0.0002,
+      "loss": 1.778,
+      "step": 670
+    },
+    {
+      "epoch": 0.5721497686159024,
+      "grad_norm": 0.3176489472389221,
+      "learning_rate": 0.0002,
+      "loss": 1.8066,
+      "step": 680
+    },
+    {
+      "epoch": 0.5805637358014304,
+      "grad_norm": 0.30220088362693787,
+      "learning_rate": 0.0002,
+      "loss": 1.7257,
+      "step": 690
+    },
+    {
+      "epoch": 0.5889777029869584,
+      "grad_norm": 0.3711601793766022,
+      "learning_rate": 0.0002,
+      "loss": 1.8415,
+      "step": 700
+    },
+    {
+      "epoch": 0.5973916701724863,
+      "grad_norm": 0.3311759829521179,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 710
+    },
+    {
+      "epoch": 0.6058056373580143,
+      "grad_norm": 0.34824270009994507,
+      "learning_rate": 0.0002,
+      "loss": 1.7712,
+      "step": 720
+    },
+    {
+      "epoch": 0.6142196045435423,
+      "grad_norm": 0.29668381810188293,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 730
+    },
+    {
+      "epoch": 0.6226335717290703,
+      "grad_norm": 0.36087489128112793,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 740
+    },
+    {
+      "epoch": 0.6310475389145982,
+      "grad_norm": 0.31590089201927185,
+      "learning_rate": 0.0002,
+      "loss": 1.7956,
+      "step": 750
+    },
+    {
+      "epoch": 0.6394615061001262,
+      "grad_norm": 0.37632957100868225,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 760
+    },
+    {
+      "epoch": 0.6478754732856542,
+      "grad_norm": 0.3360748589038849,
+      "learning_rate": 0.0002,
+      "loss": 1.8499,
+      "step": 770
+    },
+    {
+      "epoch": 0.6562894404711822,
+      "grad_norm": 0.3420640528202057,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 780
+    },
+    {
+      "epoch": 0.6647034076567101,
+      "grad_norm": 0.5734959244728088,
+      "learning_rate": 0.0002,
+      "loss": 1.8353,
+      "step": 790
+    },
+    {
+      "epoch": 0.6731173748422381,
+      "grad_norm": 0.36440837383270264,
+      "learning_rate": 0.0002,
+      "loss": 1.7746,
+      "step": 800
+    },
+    {
+      "epoch": 0.6815313420277661,
+      "grad_norm": 0.3179708421230316,
+      "learning_rate": 0.0002,
+      "loss": 1.7532,
+      "step": 810
+    },
+    {
+      "epoch": 0.6899453092132941,
+      "grad_norm": 0.34122881293296814,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 820
+    },
+    {
+      "epoch": 0.698359276398822,
+      "grad_norm": 0.31886112689971924,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 830
+    },
+    {
+      "epoch": 0.70677324358435,
+      "grad_norm": 0.31782326102256775,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 840
+    },
+    {
+      "epoch": 0.715187210769878,
+      "grad_norm": 0.36052989959716797,
+      "learning_rate": 0.0002,
+      "loss": 1.7588,
+      "step": 850
+    },
+    {
+      "epoch": 0.723601177955406,
+      "grad_norm": 0.28946155309677124,
+      "learning_rate": 0.0002,
+      "loss": 1.7891,
+      "step": 860
+    },
+    {
+      "epoch": 0.7320151451409339,
+      "grad_norm": 0.3095663785934448,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 870
+    },
+    {
+      "epoch": 0.7404291123264619,
+      "grad_norm": 0.3317491412162781,
+      "learning_rate": 0.0002,
+      "loss": 1.785,
+      "step": 880
+    },
+    {
+      "epoch": 0.7488430795119899,
+      "grad_norm": 0.31324660778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 890
+    },
+    {
+      "epoch": 0.7572570466975179,
+      "grad_norm": 0.3290475606918335,
+      "learning_rate": 0.0002,
+      "loss": 1.8753,
+      "step": 900
+    },
+    {
+      "epoch": 0.7656710138830458,
+      "grad_norm": 0.35690343379974365,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 910
+    },
+    {
+      "epoch": 0.7740849810685738,
+      "grad_norm": 0.39558273553848267,
+      "learning_rate": 0.0002,
+      "loss": 1.826,
+      "step": 920
+    },
+    {
+      "epoch": 0.7824989482541018,
+      "grad_norm": 0.34254348278045654,
+      "learning_rate": 0.0002,
+      "loss": 1.8722,
+      "step": 930
+    },
+    {
+      "epoch": 0.7909129154396298,
+      "grad_norm": 0.3560165464878082,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 940
+    },
+    {
+      "epoch": 0.7993268826251577,
+      "grad_norm": 0.30693164467811584,
+      "learning_rate": 0.0002,
+      "loss": 1.7992,
+      "step": 950
+    },
+    {
+      "epoch": 0.8077408498106857,
+      "grad_norm": 0.3394823372364044,
+      "learning_rate": 0.0002,
+      "loss": 1.8029,
+      "step": 960
+    },
+    {
+      "epoch": 0.8161548169962137,
+      "grad_norm": 0.3741514980792999,
+      "learning_rate": 0.0002,
+      "loss": 1.8105,
+      "step": 970
+    },
+    {
+      "epoch": 0.8245687841817417,
+      "grad_norm": 0.3655228316783905,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 980
+    },
+    {
+      "epoch": 0.8329827513672696,
+      "grad_norm": 0.3586033880710602,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 990
+    },
+    {
+      "epoch": 0.8413967185527976,
+      "grad_norm": 0.3459678888320923,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8498106857383256,
+      "grad_norm": 0.3184349834918976,
+      "learning_rate": 0.0002,
+      "loss": 1.8498,
+      "step": 1010
+    },
+    {
+      "epoch": 0.8582246529238536,
+      "grad_norm": 0.3099786043167114,
+      "learning_rate": 0.0002,
+      "loss": 1.7632,
+      "step": 1020
+    },
+    {
+      "epoch": 0.8666386201093815,
+      "grad_norm": 0.30300915241241455,
+      "learning_rate": 0.0002,
+      "loss": 1.8067,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8750525872949095,
+      "grad_norm": 0.3128705620765686,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8834665544804375,
+      "grad_norm": 0.3336263597011566,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8918805216659655,
+      "grad_norm": 0.3801328241825104,
+      "learning_rate": 0.0002,
+      "loss": 1.8375,
+      "step": 1060
+    },
+    {
+      "epoch": 0.9002944888514934,
+      "grad_norm": 0.3122096359729767,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 1070
+    },
+    {
+      "epoch": 0.9087084560370214,
+      "grad_norm": 0.35990869998931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 1080
+    },
+    {
+      "epoch": 0.9171224232225494,
+      "grad_norm": 0.3321819305419922,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1090
+    },
+    {
+      "epoch": 0.9255363904080774,
+      "grad_norm": 0.4202139377593994,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9339503575936053,
+      "grad_norm": 0.32559722661972046,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 1110
+    },
+    {
+      "epoch": 0.9423643247791333,
+      "grad_norm": 0.3098459839820862,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 1120
+    },
+    {
+      "epoch": 0.9507782919646613,
+      "grad_norm": 0.33917108178138733,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1130
+    },
+    {
+      "epoch": 0.9591922591501894,
+      "grad_norm": 0.4055837094783783,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1140
+    },
+    {
+      "epoch": 0.9676062263357172,
+      "grad_norm": 0.32508623600006104,
+      "learning_rate": 0.0002,
+      "loss": 1.8259,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9760201935212452,
+      "grad_norm": 0.30150601267814636,
+      "learning_rate": 0.0002,
+      "loss": 1.782,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9844341607067733,
+      "grad_norm": 0.3042563199996948,
+      "learning_rate": 0.0002,
+      "loss": 1.8291,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9928481278923013,
+      "grad_norm": 0.33254584670066833,
+      "learning_rate": 0.0002,
+      "loss": 1.7847,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9995793016407236,
+      "eval_loss": 1.8077726364135742,
+      "eval_runtime": 38.4359,
+      "eval_samples_per_second": 13.399,
+      "eval_steps_per_second": 1.691,
+      "step": 1188
+    },
+    {
+      "epoch": 1.0012620950778293,
+      "grad_norm": 0.35073035955429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 1190
+    },
+    {
+      "epoch": 1.0096760622633572,
+      "grad_norm": 0.3217269778251648,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1200
+    },
+    {
+      "epoch": 1.018090029448885,
+      "grad_norm": 0.3635033369064331,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1210
+    },
+    {
+      "epoch": 1.0265039966344132,
+      "grad_norm": 0.32468414306640625,
+      "learning_rate": 0.0002,
+      "loss": 1.6949,
+      "step": 1220
+    },
+    {
+      "epoch": 1.034917963819941,
+      "grad_norm": 0.3307163417339325,
+      "learning_rate": 0.0002,
+      "loss": 1.711,
+      "step": 1230
+    },
+    {
+      "epoch": 1.0433319310054692,
+      "grad_norm": 0.34381359815597534,
+      "learning_rate": 0.0002,
+      "loss": 1.7881,
+      "step": 1240
+    },
+    {
+      "epoch": 1.051745898190997,
+      "grad_norm": 0.35874804854393005,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 1250
+    },
+    {
+      "epoch": 1.060159865376525,
+      "grad_norm": 0.3615919351577759,
+      "learning_rate": 0.0002,
+      "loss": 1.7314,
+      "step": 1260
+    },
+    {
+      "epoch": 1.068573832562053,
+      "grad_norm": 0.32835808396339417,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1270
+    },
+    {
+      "epoch": 1.076987799747581,
+      "grad_norm": 0.3876388370990753,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0854017669331089,
+      "grad_norm": 0.39895930886268616,
+      "learning_rate": 0.0002,
+      "loss": 1.7442,
+      "step": 1290
+    },
+    {
+      "epoch": 1.093815734118637,
+      "grad_norm": 0.39081698656082153,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 1300
+    },
+    {
+      "epoch": 1.1022297013041649,
+      "grad_norm": 0.39974215626716614,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1310
+    },
+    {
+      "epoch": 1.110643668489693,
+      "grad_norm": 0.3887332081794739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1320
+    },
+    {
+      "epoch": 1.1190576356752209,
+      "grad_norm": 0.36216408014297485,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 1330
+    },
+    {
+      "epoch": 1.1274716028607488,
+      "grad_norm": 0.36979028582572937,
+      "learning_rate": 0.0002,
+      "loss": 1.762,
+      "step": 1340
+    },
+    {
+      "epoch": 1.1358855700462769,
+      "grad_norm": 0.34052133560180664,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 1350
+    },
+    {
+      "epoch": 1.1442995372318048,
+      "grad_norm": 0.3467716574668884,
+      "learning_rate": 0.0002,
+      "loss": 1.7513,
+      "step": 1360
+    },
+    {
+      "epoch": 1.1527135044173327,
+      "grad_norm": 0.35528799891471863,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1370
+    },
+    {
+      "epoch": 1.1611274716028608,
+      "grad_norm": 0.36282262206077576,
+      "learning_rate": 0.0002,
+      "loss": 1.794,
+      "step": 1380
+    },
+    {
+      "epoch": 1.1695414387883887,
+      "grad_norm": 0.37355899810791016,
+      "learning_rate": 0.0002,
+      "loss": 1.7731,
+      "step": 1390
+    },
+    {
+      "epoch": 1.1779554059739168,
+      "grad_norm": 0.37292736768722534,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1400
+    },
+    {
+      "epoch": 1.1863693731594447,
+      "grad_norm": 0.5892812013626099,
+      "learning_rate": 0.0002,
+      "loss": 1.6916,
+      "step": 1410
+    },
+    {
+      "epoch": 1.1947833403449726,
+      "grad_norm": 0.3712292015552521,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 1420
+    },
+    {
+      "epoch": 1.2031973075305007,
+      "grad_norm": 0.3349577486515045,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1430
+    },
+    {
+      "epoch": 1.2116112747160286,
+      "grad_norm": 0.32591062784194946,
+      "learning_rate": 0.0002,
+      "loss": 1.7412,
+      "step": 1440
+    },
+    {
+      "epoch": 1.2200252419015567,
+      "grad_norm": 0.3840635418891907,
+      "learning_rate": 0.0002,
+      "loss": 1.7406,
+      "step": 1450
+    },
+    {
+      "epoch": 1.2284392090870846,
+      "grad_norm": 0.37238365411758423,
+      "learning_rate": 0.0002,
+      "loss": 1.7276,
+      "step": 1460
+    },
+    {
+      "epoch": 1.2368531762726125,
+      "grad_norm": 0.3731217682361603,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 1470
+    },
+    {
+      "epoch": 1.2452671434581406,
+      "grad_norm": 0.3318967819213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7255,
+      "step": 1480
+    },
+    {
+      "epoch": 1.2536811106436685,
+      "grad_norm": 0.3784034848213196,
+      "learning_rate": 0.0002,
+      "loss": 1.7463,
+      "step": 1490
+    },
+    {
+      "epoch": 1.2620950778291964,
+      "grad_norm": 0.3541383147239685,
+      "learning_rate": 0.0002,
+      "loss": 1.6862,
+      "step": 1500
+    },
+    {
+      "epoch": 1.2705090450147245,
+      "grad_norm": 0.35312485694885254,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1510
+    },
+    {
+      "epoch": 1.2789230122002524,
+      "grad_norm": 0.35272929072380066,
+      "learning_rate": 0.0002,
+      "loss": 1.7029,
+      "step": 1520
+    },
+    {
+      "epoch": 1.2873369793857803,
+      "grad_norm": 0.40988272428512573,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1530
+    },
+    {
+      "epoch": 1.2957509465713084,
+      "grad_norm": 0.3543946146965027,
+      "learning_rate": 0.0002,
+      "loss": 1.6912,
+      "step": 1540
+    },
+    {
+      "epoch": 1.3041649137568363,
+      "grad_norm": 0.35639145970344543,
+      "learning_rate": 0.0002,
+      "loss": 1.6757,
+      "step": 1550
+    },
+    {
+      "epoch": 1.3125788809423642,
+      "grad_norm": 0.3290826678276062,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 1560
+    },
+    {
+      "epoch": 1.3209928481278923,
+      "grad_norm": 0.39264336228370667,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 1570
+    },
+    {
+      "epoch": 1.3294068153134202,
+      "grad_norm": 0.5390415191650391,
+      "learning_rate": 0.0002,
+      "loss": 1.6804,
+      "step": 1580
+    },
+    {
+      "epoch": 1.3378207824989483,
+      "grad_norm": 0.5188116431236267,
+      "learning_rate": 0.0002,
+      "loss": 1.708,
+      "step": 1590
+    },
+    {
+      "epoch": 1.3462347496844762,
+      "grad_norm": 0.37445148825645447,
+      "learning_rate": 0.0002,
+      "loss": 1.6763,
+      "step": 1600
+    },
+    {
+      "epoch": 1.3546487168700043,
+      "grad_norm": 0.3296085298061371,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1610
+    },
+    {
+      "epoch": 1.3630626840555322,
+      "grad_norm": 0.39879581332206726,
+      "learning_rate": 0.0002,
+      "loss": 1.8107,
+      "step": 1620
+    },
+    {
+      "epoch": 1.37147665124106,
+      "grad_norm": 0.36092764139175415,
+      "learning_rate": 0.0002,
+      "loss": 1.6744,
+      "step": 1630
+    },
+    {
+      "epoch": 1.3798906184265882,
+      "grad_norm": 0.37011823058128357,
+      "learning_rate": 0.0002,
+      "loss": 1.7144,
+      "step": 1640
+    },
+    {
+      "epoch": 1.3883045856121161,
+      "grad_norm": 0.40863534808158875,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1650
+    },
+    {
+      "epoch": 1.396718552797644,
+      "grad_norm": 0.337001770734787,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 1660
+    },
+    {
+      "epoch": 1.4051325199831721,
+      "grad_norm": 0.35596707463264465,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1670
+    },
+    {
+      "epoch": 1.4135464871687,
+      "grad_norm": 0.3857671916484833,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 1680
+    },
+    {
+      "epoch": 1.421960454354228,
+      "grad_norm": 0.419502317905426,
+      "learning_rate": 0.0002,
+      "loss": 1.7015,
+      "step": 1690
+    },
+    {
+      "epoch": 1.430374421539756,
+      "grad_norm": 0.35459452867507935,
+      "learning_rate": 0.0002,
+      "loss": 1.7261,
+      "step": 1700
+    },
+    {
+      "epoch": 1.438788388725284,
+      "grad_norm": 0.37246978282928467,
+      "learning_rate": 0.0002,
+      "loss": 1.7361,
+      "step": 1710
+    },
+    {
+      "epoch": 1.4472023559108118,
+      "grad_norm": 0.33091893792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 1720
+    },
+    {
+      "epoch": 1.45561632309634,
+      "grad_norm": 0.37029674649238586,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1730
+    },
+    {
+      "epoch": 1.4640302902818678,
+      "grad_norm": 0.374025821685791,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1740
+    },
+    {
+      "epoch": 1.472444257467396,
+      "grad_norm": 0.3416315019130707,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1750
+    },
+    {
+      "epoch": 1.4808582246529238,
+      "grad_norm": 0.36502841114997864,
+      "learning_rate": 0.0002,
+      "loss": 1.7093,
+      "step": 1760
+    },
+    {
+      "epoch": 1.489272191838452,
+      "grad_norm": 0.35458803176879883,
+      "learning_rate": 0.0002,
+      "loss": 1.6597,
+      "step": 1770
+    },
+    {
+      "epoch": 1.4976861590239798,
+      "grad_norm": 0.4462839663028717,
+      "learning_rate": 0.0002,
+      "loss": 1.675,
+      "step": 1780
+    },
+    {
+      "epoch": 1.5061001262095077,
+      "grad_norm": 0.34836092591285706,
+      "learning_rate": 0.0002,
+      "loss": 1.7267,
+      "step": 1790
+    },
+    {
+      "epoch": 1.5145140933950358,
+      "grad_norm": 0.3445749282836914,
+      "learning_rate": 0.0002,
+      "loss": 1.7295,
+      "step": 1800
+    },
+    {
+      "epoch": 1.5229280605805637,
+      "grad_norm": 0.36012160778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1810
+    },
+    {
+      "epoch": 1.5313420277660916,
+      "grad_norm": 0.4052616059780121,
+      "learning_rate": 0.0002,
+      "loss": 1.6594,
+      "step": 1820
+    },
+    {
+      "epoch": 1.5397559949516197,
+      "grad_norm": 0.3966905474662781,
+      "learning_rate": 0.0002,
+      "loss": 1.72,
+      "step": 1830
+    },
+    {
+      "epoch": 1.5481699621371476,
+      "grad_norm": 0.35028719902038574,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1840
+    },
+    {
+      "epoch": 1.5565839293226755,
+      "grad_norm": 0.3936742842197418,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1850
+    },
+    {
+      "epoch": 1.5649978965082036,
+      "grad_norm": 0.34473296999931335,
+      "learning_rate": 0.0002,
+      "loss": 1.7579,
+      "step": 1860
+    },
+    {
+      "epoch": 1.5734118636937318,
+      "grad_norm": 0.4328365623950958,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1870
+    },
+    {
+      "epoch": 1.5818258308792594,
+      "grad_norm": 0.3566315472126007,
+      "learning_rate": 0.0002,
+      "loss": 1.7098,
+      "step": 1880
+    },
+    {
+      "epoch": 1.5902397980647875,
+      "grad_norm": 0.3301256597042084,
+      "learning_rate": 0.0002,
+      "loss": 1.6095,
+      "step": 1890
+    },
+    {
+      "epoch": 1.5986537652503157,
+      "grad_norm": 0.3743041455745697,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 1900
+    },
+    {
+      "epoch": 1.6070677324358436,
+      "grad_norm": 0.3735344707965851,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 1910
+    },
+    {
+      "epoch": 1.6154816996213714,
+      "grad_norm": 0.42191144824028015,
+      "learning_rate": 0.0002,
+      "loss": 1.7445,
+      "step": 1920
+    },
+    {
+      "epoch": 1.6238956668068996,
+      "grad_norm": 0.3787207305431366,
+      "learning_rate": 0.0002,
+      "loss": 1.6978,
+      "step": 1930
+    },
+    {
+      "epoch": 1.6323096339924275,
+      "grad_norm": 0.35647350549697876,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 1940
+    },
+    {
+      "epoch": 1.6407236011779553,
+      "grad_norm": 0.39791446924209595,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 1950
+    },
+    {
+      "epoch": 1.6491375683634835,
+      "grad_norm": 0.37341275811195374,
+      "learning_rate": 0.0002,
+      "loss": 1.7293,
+      "step": 1960
+    },
+    {
+      "epoch": 1.6575515355490114,
+      "grad_norm": 0.3722686469554901,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1970
+    },
+    {
+      "epoch": 1.6659655027345392,
+      "grad_norm": 0.37467387318611145,
+      "learning_rate": 0.0002,
+      "loss": 1.6383,
+      "step": 1980
+    },
+    {
+      "epoch": 1.6743794699200674,
+      "grad_norm": 0.37109461426734924,
+      "learning_rate": 0.0002,
+      "loss": 1.7439,
+      "step": 1990
+    },
+    {
+      "epoch": 1.6827934371055953,
+      "grad_norm": 0.4008837044239044,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 2000
+    },
+    {
+      "epoch": 1.6912074042911232,
+      "grad_norm": 0.3316999673843384,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2010
+    },
+    {
+      "epoch": 1.6996213714766513,
+      "grad_norm": 0.3683805465698242,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 2020
+    },
+    {
+      "epoch": 1.7080353386621794,
+      "grad_norm": 0.4163658320903778,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 2030
+    },
+    {
+      "epoch": 1.716449305847707,
+      "grad_norm": 0.4245431125164032,
+      "learning_rate": 0.0002,
+      "loss": 1.741,
+      "step": 2040
+    },
+    {
+      "epoch": 1.7248632730332352,
+      "grad_norm": 0.36732038855552673,
+      "learning_rate": 0.0002,
+      "loss": 1.7184,
+      "step": 2050
+    },
+    {
+      "epoch": 1.7332772402187633,
+      "grad_norm": 0.34981656074523926,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 2060
+    },
+    {
+      "epoch": 1.7416912074042912,
+      "grad_norm": 0.38588812947273254,
+      "learning_rate": 0.0002,
+      "loss": 1.7545,
+      "step": 2070
+    },
+    {
+      "epoch": 1.750105174589819,
+      "grad_norm": 0.39914557337760925,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 2080
+    },
+    {
+      "epoch": 1.7585191417753472,
+      "grad_norm": 0.36068692803382874,
+      "learning_rate": 0.0002,
+      "loss": 1.7049,
+      "step": 2090
+    },
+    {
+      "epoch": 1.766933108960875,
+      "grad_norm": 0.3983287215232849,
+      "learning_rate": 0.0002,
+      "loss": 1.7537,
+      "step": 2100
+    },
+    {
+      "epoch": 1.775347076146403,
+      "grad_norm": 0.45008400082588196,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2110
+    },
+    {
+      "epoch": 1.783761043331931,
+      "grad_norm": 0.3618052303791046,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 2120
+    },
+    {
+      "epoch": 1.792175010517459,
+      "grad_norm": 0.38745400309562683,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 2130
+    },
+    {
+      "epoch": 1.8005889777029869,
+      "grad_norm": 0.3413826525211334,
+      "learning_rate": 0.0002,
+      "loss": 1.7387,
+      "step": 2140
+    },
+    {
+      "epoch": 1.809002944888515,
+      "grad_norm": 0.35983747243881226,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 2150
+    },
+    {
+      "epoch": 1.8174169120740429,
+      "grad_norm": 0.40926849842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 2160
+    },
+    {
+      "epoch": 1.8258308792595708,
+      "grad_norm": 0.3543093800544739,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2170
+    },
+    {
+      "epoch": 1.8342448464450989,
+      "grad_norm": 0.42690935730934143,
+      "learning_rate": 0.0002,
+      "loss": 1.7812,
+      "step": 2180
+    },
+    {
+      "epoch": 1.842658813630627,
+      "grad_norm": 0.40282756090164185,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 2190
+    },
+    {
+      "epoch": 1.8510727808161547,
+      "grad_norm": 0.36568400263786316,
+      "learning_rate": 0.0002,
+      "loss": 1.7411,
+      "step": 2200
+    },
+    {
+      "epoch": 1.8594867480016828,
+      "grad_norm": 0.43159013986587524,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 2210
+    },
+    {
+      "epoch": 1.867900715187211,
+      "grad_norm": 0.3554118573665619,
+      "learning_rate": 0.0002,
+      "loss": 1.7298,
+      "step": 2220
+    },
+    {
+      "epoch": 1.8763146823727388,
+      "grad_norm": 0.43349072337150574,
+      "learning_rate": 0.0002,
+      "loss": 1.7157,
+      "step": 2230
+    },
+    {
+      "epoch": 1.8847286495582667,
+      "grad_norm": 0.36486536264419556,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 2240
+    },
+    {
+      "epoch": 1.8931426167437948,
+      "grad_norm": 0.39260047674179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6901,
+      "step": 2250
+    },
+    {
+      "epoch": 1.9015565839293227,
+      "grad_norm": 0.3741776943206787,
+      "learning_rate": 0.0002,
+      "loss": 1.6691,
+      "step": 2260
+    },
+    {
+      "epoch": 1.9099705511148506,
+      "grad_norm": 0.3961946964263916,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2270
+    },
+    {
+      "epoch": 1.9183845183003787,
+      "grad_norm": 0.3659731149673462,
+      "learning_rate": 0.0002,
+      "loss": 1.737,
+      "step": 2280
+    },
+    {
+      "epoch": 1.9267984854859066,
+      "grad_norm": 0.34744107723236084,
+      "learning_rate": 0.0002,
+      "loss": 1.7342,
+      "step": 2290
+    },
+    {
+      "epoch": 1.9352124526714345,
+      "grad_norm": 0.3607442378997803,
+      "learning_rate": 0.0002,
+      "loss": 1.7162,
+      "step": 2300
+    },
+    {
+      "epoch": 1.9436264198569626,
+      "grad_norm": 0.331464558839798,
+      "learning_rate": 0.0002,
+      "loss": 1.6673,
+      "step": 2310
+    },
+    {
+      "epoch": 1.9520403870424905,
+      "grad_norm": 0.3904414474964142,
+      "learning_rate": 0.0002,
+      "loss": 1.7101,
+      "step": 2320
+    },
+    {
+      "epoch": 1.9604543542280184,
+      "grad_norm": 0.37584832310676575,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 2330
+    },
+    {
+      "epoch": 1.9688683214135465,
+      "grad_norm": 0.3698684275150299,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 2340
+    },
+    {
+      "epoch": 1.9772822885990746,
+      "grad_norm": 0.40571412444114685,
+      "learning_rate": 0.0002,
+      "loss": 1.7764,
+      "step": 2350
+    },
+    {
+      "epoch": 1.9856962557846023,
+      "grad_norm": 0.40059587359428406,
+      "learning_rate": 0.0002,
+      "loss": 1.744,
+      "step": 2360
+    },
+    {
+      "epoch": 1.9941102229701304,
+      "grad_norm": 0.4168248474597931,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2370
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8055059909820557,
+      "eval_runtime": 38.422,
+      "eval_samples_per_second": 13.404,
+      "eval_steps_per_second": 1.692,
+      "step": 2377
+    },
+    {
+      "epoch": 2.0025241901556585,
+      "grad_norm": 0.35205352306365967,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 2380
+    },
+    {
+      "epoch": 2.010938157341186,
+      "grad_norm": 0.3979377746582031,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 2390
+    },
+    {
+      "epoch": 2.0193521245267143,
+      "grad_norm": 0.396491676568985,
+      "learning_rate": 0.0002,
+      "loss": 1.6421,
+      "step": 2400
+    },
+    {
+      "epoch": 2.0277660917122424,
+      "grad_norm": 0.44712209701538086,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2410
+    },
+    {
+      "epoch": 2.03618005889777,
+      "grad_norm": 0.4454420208930969,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 2420
+    },
+    {
+      "epoch": 2.044594026083298,
+      "grad_norm": 0.4170038402080536,
+      "learning_rate": 0.0002,
+      "loss": 1.6635,
+      "step": 2430
+    },
+    {
+      "epoch": 2.0530079932688263,
+      "grad_norm": 0.4309595227241516,
+      "learning_rate": 0.0002,
+      "loss": 1.6512,
+      "step": 2440
+    },
+    {
+      "epoch": 2.0614219604543544,
+      "grad_norm": 0.4241602122783661,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 2450
+    },
+    {
+      "epoch": 2.069835927639882,
+      "grad_norm": 0.4370540678501129,
+      "learning_rate": 0.0002,
+      "loss": 1.6162,
+      "step": 2460
+    },
+    {
+      "epoch": 2.0782498948254102,
+      "grad_norm": 0.43985554575920105,
+      "learning_rate": 0.0002,
+      "loss": 1.6354,
+      "step": 2470
+    },
+    {
+      "epoch": 2.0866638620109383,
+      "grad_norm": 0.4158105254173279,
+      "learning_rate": 0.0002,
+      "loss": 1.6954,
+      "step": 2480
+    },
+    {
+      "epoch": 2.095077829196466,
+      "grad_norm": 0.441549152135849,
+      "learning_rate": 0.0002,
+      "loss": 1.6114,
+      "step": 2490
+    },
+    {
+      "epoch": 2.103491796381994,
+      "grad_norm": 0.385718435049057,
+      "learning_rate": 0.0002,
+      "loss": 1.5485,
+      "step": 2500
+    },
+    {
+      "epoch": 2.1119057635675222,
+      "grad_norm": 0.43146514892578125,
+      "learning_rate": 0.0002,
+      "loss": 1.5894,
+      "step": 2510
+    },
+    {
+      "epoch": 2.12031973075305,
+      "grad_norm": 0.41663315892219543,
+      "learning_rate": 0.0002,
+      "loss": 1.6414,
+      "step": 2520
+    },
+    {
+      "epoch": 2.128733697938578,
+      "grad_norm": 0.4410698115825653,
+      "learning_rate": 0.0002,
+      "loss": 1.6527,
+      "step": 2530
+    },
+    {
+      "epoch": 2.137147665124106,
+      "grad_norm": 0.4472278952598572,
+      "learning_rate": 0.0002,
+      "loss": 1.6124,
+      "step": 2540
+    },
+    {
+      "epoch": 2.145561632309634,
+      "grad_norm": 0.3879167437553406,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2550
+    },
+    {
+      "epoch": 2.153975599495162,
+      "grad_norm": 0.4212203025817871,
+      "learning_rate": 0.0002,
+      "loss": 1.6682,
+      "step": 2560
+    },
+    {
+      "epoch": 2.16238956668069,
+      "grad_norm": 0.42841723561286926,
+      "learning_rate": 0.0002,
+      "loss": 1.6036,
+      "step": 2570
+    },
+    {
+      "epoch": 2.1708035338662177,
+      "grad_norm": 0.39272481203079224,
+      "learning_rate": 0.0002,
+      "loss": 1.5962,
+      "step": 2580
+    },
+    {
+      "epoch": 2.179217501051746,
+      "grad_norm": 0.4075261354446411,
+      "learning_rate": 0.0002,
+      "loss": 1.681,
+      "step": 2590
+    },
+    {
+      "epoch": 2.187631468237274,
+      "grad_norm": 0.5358437895774841,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 2600
+    },
+    {
+      "epoch": 2.1960454354228016,
+      "grad_norm": 0.4738350212574005,
+      "learning_rate": 0.0002,
+      "loss": 1.6423,
+      "step": 2610
+    },
+    {
+      "epoch": 2.2044594026083297,
+      "grad_norm": 0.446789026260376,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 2620
+    },
+    {
+      "epoch": 2.212873369793858,
+      "grad_norm": 0.4615374505519867,
+      "learning_rate": 0.0002,
+      "loss": 1.6246,
+      "step": 2630
+    },
+    {
+      "epoch": 2.221287336979386,
+      "grad_norm": 0.46901994943618774,
+      "learning_rate": 0.0002,
+      "loss": 1.6205,
+      "step": 2640
+    },
+    {
+      "epoch": 2.2297013041649136,
+      "grad_norm": 0.46267789602279663,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 2.2381152713504417,
+      "grad_norm": 0.4383080005645752,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2660
+    },
+    {
+      "epoch": 2.24652923853597,
+      "grad_norm": 0.4070609509944916,
+      "learning_rate": 0.0002,
+      "loss": 1.5745,
+      "step": 2670
+    },
+    {
+      "epoch": 2.2549432057214975,
+      "grad_norm": 0.4572339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.6125,
+      "step": 2680
+    },
+    {
+      "epoch": 2.2633571729070256,
+      "grad_norm": 0.393265038728714,
+      "learning_rate": 0.0002,
+      "loss": 1.5671,
+      "step": 2690
+    },
+    {
+      "epoch": 2.2717711400925538,
+      "grad_norm": 0.46144717931747437,
+      "learning_rate": 0.0002,
+      "loss": 1.6239,
+      "step": 2700
+    },
+    {
+      "epoch": 2.2801851072780814,
+      "grad_norm": 0.45077767968177795,
+      "learning_rate": 0.0002,
+      "loss": 1.5992,
+      "step": 2710
+    },
+    {
+      "epoch": 2.2885990744636096,
+      "grad_norm": 0.5697639584541321,
+      "learning_rate": 0.0002,
+      "loss": 1.6261,
+      "step": 2720
+    },
+    {
+      "epoch": 2.2970130416491377,
+      "grad_norm": 0.4855510890483856,
+      "learning_rate": 0.0002,
+      "loss": 1.6192,
+      "step": 2730
+    },
+    {
+      "epoch": 2.3054270088346653,
+      "grad_norm": 0.4440622627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 2740
+    },
+    {
+      "epoch": 2.3138409760201935,
+      "grad_norm": 0.3904096782207489,
+      "learning_rate": 0.0002,
+      "loss": 1.6496,
+      "step": 2750
+    },
+    {
+      "epoch": 2.3222549432057216,
+      "grad_norm": 0.5225510597229004,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 2760
+    },
+    {
+      "epoch": 2.3306689103912497,
+      "grad_norm": 0.44866397976875305,
+      "learning_rate": 0.0002,
+      "loss": 1.6082,
+      "step": 2770
+    },
+    {
+      "epoch": 2.3390828775767774,
+      "grad_norm": 0.5167056322097778,
+      "learning_rate": 0.0002,
+      "loss": 1.6087,
+      "step": 2780
+    },
+    {
+      "epoch": 2.3474968447623055,
+      "grad_norm": 0.45913267135620117,
+      "learning_rate": 0.0002,
+      "loss": 1.6136,
+      "step": 2790
+    },
+    {
+      "epoch": 2.3559108119478336,
+      "grad_norm": 0.45787590742111206,
+      "learning_rate": 0.0002,
+      "loss": 1.6564,
+      "step": 2800
+    },
+    {
+      "epoch": 2.3643247791333613,
+      "grad_norm": 0.4633352756500244,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2810
+    },
+    {
+      "epoch": 2.3727387463188894,
+      "grad_norm": 0.46390071511268616,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 2820
+    },
+    {
+      "epoch": 2.3811527135044175,
+      "grad_norm": 0.4261005222797394,
+      "learning_rate": 0.0002,
+      "loss": 1.6039,
+      "step": 2830
+    },
+    {
+      "epoch": 2.389566680689945,
+      "grad_norm": 0.4283634424209595,
+      "learning_rate": 0.0002,
+      "loss": 1.6364,
+      "step": 2840
+    },
+    {
+      "epoch": 2.3979806478754733,
+      "grad_norm": 0.4955291450023651,
+      "learning_rate": 0.0002,
+      "loss": 1.6382,
+      "step": 2850
+    },
+    {
+      "epoch": 2.4063946150610014,
+      "grad_norm": 0.4740189015865326,
+      "learning_rate": 0.0002,
+      "loss": 1.6173,
+      "step": 2860
+    },
+    {
+      "epoch": 2.414808582246529,
+      "grad_norm": 0.4222276508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2870
+    },
+    {
+      "epoch": 2.423222549432057,
+      "grad_norm": 0.4982149004936218,
+      "learning_rate": 0.0002,
+      "loss": 1.5602,
+      "step": 2880
+    },
+    {
+      "epoch": 2.4316365166175853,
+      "grad_norm": 0.5217409133911133,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 2890
+    },
+    {
+      "epoch": 2.4400504838031134,
+      "grad_norm": 0.4555884897708893,
+      "learning_rate": 0.0002,
+      "loss": 1.5804,
+      "step": 2900
+    },
+    {
+      "epoch": 2.448464450988641,
+      "grad_norm": 0.43178579211235046,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2910
+    },
+    {
+      "epoch": 2.456878418174169,
+      "grad_norm": 0.4788478910923004,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2920
+    },
+    {
+      "epoch": 2.465292385359697,
+      "grad_norm": 0.43689873814582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 2930
+    },
+    {
+      "epoch": 2.473706352545225,
+      "grad_norm": 0.5115197896957397,
+      "learning_rate": 0.0002,
+      "loss": 1.6196,
+      "step": 2940
+    },
+    {
+      "epoch": 2.482120319730753,
+      "grad_norm": 0.5290159583091736,
+      "learning_rate": 0.0002,
+      "loss": 1.689,
+      "step": 2950
+    },
+    {
+      "epoch": 2.490534286916281,
+      "grad_norm": 0.46042463183403015,
+      "learning_rate": 0.0002,
+      "loss": 1.6499,
+      "step": 2960
+    },
+    {
+      "epoch": 2.498948254101809,
+      "grad_norm": 0.4359915852546692,
+      "learning_rate": 0.0002,
+      "loss": 1.6664,
+      "step": 2970
+    },
+    {
+      "epoch": 2.507362221287337,
+      "grad_norm": 0.46352964639663696,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 2980
+    },
+    {
+      "epoch": 2.515776188472865,
+      "grad_norm": 0.5324268341064453,
+      "learning_rate": 0.0002,
+      "loss": 1.6501,
+      "step": 2990
+    },
+    {
+      "epoch": 2.5241901556583928,
+      "grad_norm": 0.5929607152938843,
+      "learning_rate": 0.0002,
+      "loss": 1.6115,
+      "step": 3000
+    },
+    {
+      "epoch": 2.532604122843921,
+      "grad_norm": 0.4811333417892456,
+      "learning_rate": 0.0002,
+      "loss": 1.6772,
+      "step": 3010
+    },
+    {
+      "epoch": 2.541018090029449,
+      "grad_norm": 0.4662701487541199,
+      "learning_rate": 0.0002,
+      "loss": 1.7023,
+      "step": 3020
+    },
+    {
+      "epoch": 2.549432057214977,
+      "grad_norm": 0.4582270681858063,
+      "learning_rate": 0.0002,
+      "loss": 1.5426,
+      "step": 3030
+    },
+    {
+      "epoch": 2.557846024400505,
+      "grad_norm": 0.4679982662200928,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 3040
+    },
+    {
+      "epoch": 2.566259991586033,
+      "grad_norm": 0.4380294680595398,
+      "learning_rate": 0.0002,
+      "loss": 1.5442,
+      "step": 3050
+    },
+    {
+      "epoch": 2.5746739587715606,
+      "grad_norm": 0.44295763969421387,
+      "learning_rate": 0.0002,
+      "loss": 1.6055,
+      "step": 3060
+    },
+    {
+      "epoch": 2.5830879259570887,
+      "grad_norm": 0.5131027698516846,
+      "learning_rate": 0.0002,
+      "loss": 1.5775,
+      "step": 3070
+    },
+    {
+      "epoch": 2.591501893142617,
+      "grad_norm": 0.47567516565322876,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 3080
+    },
+    {
+      "epoch": 2.599915860328145,
+      "grad_norm": 0.49002596735954285,
+      "learning_rate": 0.0002,
+      "loss": 1.5671,
+      "step": 3090
+    },
+    {
+      "epoch": 2.6083298275136726,
+      "grad_norm": 0.44856327772140503,
+      "learning_rate": 0.0002,
+      "loss": 1.5445,
+      "step": 3100
+    },
+    {
+      "epoch": 2.6167437946992007,
+      "grad_norm": 0.4480142593383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5797,
+      "step": 3110
+    },
+    {
+      "epoch": 2.6251577618847284,
+      "grad_norm": 0.4317494034767151,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 3120
+    },
+    {
+      "epoch": 2.6335717290702565,
+      "grad_norm": 0.42580848932266235,
+      "learning_rate": 0.0002,
+      "loss": 1.6321,
+      "step": 3130
+    },
+    {
+      "epoch": 2.6419856962557846,
+      "grad_norm": 0.4516814947128296,
+      "learning_rate": 0.0002,
+      "loss": 1.6483,
+      "step": 3140
+    },
+    {
+      "epoch": 2.6503996634413127,
+      "grad_norm": 0.4438435733318329,
+      "learning_rate": 0.0002,
+      "loss": 1.695,
+      "step": 3150
+    },
+    {
+      "epoch": 2.6588136306268404,
+      "grad_norm": 0.4385356307029724,
+      "learning_rate": 0.0002,
+      "loss": 1.6938,
+      "step": 3160
+    },
+    {
+      "epoch": 2.6672275978123685,
+      "grad_norm": 0.5064112544059753,
+      "learning_rate": 0.0002,
+      "loss": 1.6139,
+      "step": 3170
+    },
+    {
+      "epoch": 2.6756415649978966,
+      "grad_norm": 0.49163177609443665,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 3180
+    },
+    {
+      "epoch": 2.6840555321834243,
+      "grad_norm": 0.49339258670806885,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 3190
+    },
+    {
+      "epoch": 2.6924694993689524,
+      "grad_norm": 0.440950870513916,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 3200
+    },
+    {
+      "epoch": 2.7008834665544805,
+      "grad_norm": 0.4283970594406128,
+      "learning_rate": 0.0002,
+      "loss": 1.6305,
+      "step": 3210
+    },
+    {
+      "epoch": 2.7092974337400086,
+      "grad_norm": 0.43875712156295776,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 3220
+    },
+    {
+      "epoch": 2.7177114009255363,
+      "grad_norm": 0.49332964420318604,
+      "learning_rate": 0.0002,
+      "loss": 1.6129,
+      "step": 3230
+    },
+    {
+      "epoch": 2.7261253681110644,
+      "grad_norm": 0.5225692391395569,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 3240
+    },
+    {
+      "epoch": 2.734539335296592,
+      "grad_norm": 0.4856489300727844,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 3250
+    },
+    {
+      "epoch": 2.74295330248212,
+      "grad_norm": 0.46918296813964844,
+      "learning_rate": 0.0002,
+      "loss": 1.6463,
+      "step": 3260
+    },
+    {
+      "epoch": 2.7513672696676483,
+      "grad_norm": 0.4802931249141693,
+      "learning_rate": 0.0002,
+      "loss": 1.6819,
+      "step": 3270
+    },
+    {
+      "epoch": 2.7597812368531764,
+      "grad_norm": 0.4485355615615845,
+      "learning_rate": 0.0002,
+      "loss": 1.6246,
+      "step": 3280
+    },
+    {
+      "epoch": 2.768195204038704,
+      "grad_norm": 0.43944594264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 3290
+    },
+    {
+      "epoch": 2.7766091712242322,
+      "grad_norm": 0.46847742795944214,
+      "learning_rate": 0.0002,
+      "loss": 1.6501,
+      "step": 3300
+    },
+    {
+      "epoch": 2.7850231384097603,
+      "grad_norm": 0.4816027879714966,
+      "learning_rate": 0.0002,
+      "loss": 1.5969,
+      "step": 3310
+    },
+    {
+      "epoch": 2.793437105595288,
+      "grad_norm": 0.453960120677948,
+      "learning_rate": 0.0002,
+      "loss": 1.6293,
+      "step": 3320
+    },
+    {
+      "epoch": 2.801851072780816,
+      "grad_norm": 0.4816017150878906,
+      "learning_rate": 0.0002,
+      "loss": 1.6429,
+      "step": 3330
+    },
+    {
+      "epoch": 2.8102650399663442,
+      "grad_norm": 0.4461034834384918,
+      "learning_rate": 0.0002,
+      "loss": 1.6683,
+      "step": 3340
+    },
+    {
+      "epoch": 2.8186790071518724,
+      "grad_norm": 0.48821821808815,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 3350
+    },
+    {
+      "epoch": 2.8270929743374,
+      "grad_norm": 0.4574853777885437,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 3360
+    },
+    {
+      "epoch": 2.835506941522928,
+      "grad_norm": 0.42062026262283325,
+      "learning_rate": 0.0002,
+      "loss": 1.6651,
+      "step": 3370
+    },
+    {
+      "epoch": 2.843920908708456,
+      "grad_norm": 0.4499834477901459,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 3380
+    },
+    {
+      "epoch": 2.852334875893984,
+      "grad_norm": 0.4780360758304596,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 3390
+    },
+    {
+      "epoch": 2.860748843079512,
+      "grad_norm": 0.45422887802124023,
+      "learning_rate": 0.0002,
+      "loss": 1.5882,
+      "step": 3400
+    },
+    {
+      "epoch": 2.86916281026504,
+      "grad_norm": 0.4590015709400177,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 3410
+    },
+    {
+      "epoch": 2.877576777450568,
+      "grad_norm": 0.45689624547958374,
+      "learning_rate": 0.0002,
+      "loss": 1.6746,
+      "step": 3420
+    },
+    {
+      "epoch": 2.885990744636096,
+      "grad_norm": 0.46953922510147095,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3430
+    },
+    {
+      "epoch": 2.8944047118216236,
+      "grad_norm": 0.4791966378688812,
+      "learning_rate": 0.0002,
+      "loss": 1.6015,
+      "step": 3440
+    },
+    {
+      "epoch": 2.9028186790071517,
+      "grad_norm": 0.4842296242713928,
+      "learning_rate": 0.0002,
+      "loss": 1.694,
+      "step": 3450
+    },
+    {
+      "epoch": 2.91123264619268,
+      "grad_norm": 0.47219768166542053,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3460
+    },
+    {
+      "epoch": 2.919646613378208,
+      "grad_norm": 0.4622127115726471,
+      "learning_rate": 0.0002,
+      "loss": 1.6486,
+      "step": 3470
+    },
+    {
+      "epoch": 2.9280605805637356,
+      "grad_norm": 0.46832820773124695,
+      "learning_rate": 0.0002,
+      "loss": 1.6485,
+      "step": 3480
+    },
+    {
+      "epoch": 2.9364745477492638,
+      "grad_norm": 0.44582483172416687,
+      "learning_rate": 0.0002,
+      "loss": 1.6366,
+      "step": 3490
+    },
+    {
+      "epoch": 2.944888514934792,
+      "grad_norm": 0.4987219274044037,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 3500
+    },
+    {
+      "epoch": 2.9533024821203195,
+      "grad_norm": 0.43750956654548645,
+      "learning_rate": 0.0002,
+      "loss": 1.5991,
+      "step": 3510
+    },
+    {
+      "epoch": 2.9617164493058477,
+      "grad_norm": 0.49962925910949707,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 3520
+    },
+    {
+      "epoch": 2.9701304164913758,
+      "grad_norm": 0.5189590454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 3530
+    },
+    {
+      "epoch": 2.978544383676904,
+      "grad_norm": 0.391317754983902,
+      "learning_rate": 0.0002,
+      "loss": 1.6688,
+      "step": 3540
+    },
+    {
+      "epoch": 2.9869583508624316,
+      "grad_norm": 0.44934695959091187,
+      "learning_rate": 0.0002,
+      "loss": 1.5884,
+      "step": 3550
+    },
+    {
+      "epoch": 2.9953723180479597,
+      "grad_norm": 0.4740142226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.5688,
+      "step": 3560
+    },
+    {
+      "epoch": 2.9995793016407237,
+      "eval_loss": 1.8266887664794922,
+      "eval_runtime": 37.9445,
+      "eval_samples_per_second": 13.572,
+      "eval_steps_per_second": 1.713,
+      "step": 3565
+    },
+    {
+      "epoch": 3.003786285233488,
+      "grad_norm": 0.4523724615573883,
+      "learning_rate": 0.0002,
+      "loss": 1.5939,
+      "step": 3570
+    },
+    {
+      "epoch": 3.0122002524190155,
+      "grad_norm": 0.5261380076408386,
+      "learning_rate": 0.0002,
+      "loss": 1.526,
+      "step": 3580
+    },
+    {
+      "epoch": 3.0206142196045436,
+      "grad_norm": 0.48664888739585876,
+      "learning_rate": 0.0002,
+      "loss": 1.4946,
+      "step": 3590
+    },
+    {
+      "epoch": 3.0290281867900717,
+      "grad_norm": 0.5070882439613342,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 3600
+    },
+    {
+      "epoch": 3.0374421539755994,
+      "grad_norm": 0.5816011428833008,
+      "learning_rate": 0.0002,
+      "loss": 1.5316,
+      "step": 3610
+    },
+    {
+      "epoch": 3.0458561211611275,
+      "grad_norm": 0.6610211730003357,
+      "learning_rate": 0.0002,
+      "loss": 1.5682,
+      "step": 3620
+    },
+    {
+      "epoch": 3.0542700883466556,
+      "grad_norm": 0.5257703065872192,
+      "learning_rate": 0.0002,
+      "loss": 1.5699,
+      "step": 3630
+    },
+    {
+      "epoch": 3.0626840555321833,
+      "grad_norm": 0.5574390888214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4438,
+      "step": 3640
+    },
+    {
+      "epoch": 3.0710980227177114,
+      "grad_norm": 0.5682297348976135,
+      "learning_rate": 0.0002,
+      "loss": 1.547,
+      "step": 3650
+    },
+    {
+      "epoch": 3.0795119899032395,
+      "grad_norm": 0.5798383355140686,
+      "learning_rate": 0.0002,
+      "loss": 1.5743,
+      "step": 3660
+    },
+    {
+      "epoch": 3.087925957088767,
+      "grad_norm": 0.5458289980888367,
+      "learning_rate": 0.0002,
+      "loss": 1.4339,
+      "step": 3670
+    },
+    {
+      "epoch": 3.0963399242742953,
+      "grad_norm": 0.5599102973937988,
+      "learning_rate": 0.0002,
+      "loss": 1.46,
+      "step": 3680
+    },
+    {
+      "epoch": 3.1047538914598234,
+      "grad_norm": 0.5023021697998047,
+      "learning_rate": 0.0002,
+      "loss": 1.4589,
+      "step": 3690
+    },
+    {
+      "epoch": 3.113167858645351,
+      "grad_norm": 0.5448206067085266,
+      "learning_rate": 0.0002,
+      "loss": 1.5114,
+      "step": 3700
+    },
+    {
+      "epoch": 3.121581825830879,
+      "grad_norm": 0.5760458707809448,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 3710
+    },
+    {
+      "epoch": 3.1299957930164073,
+      "grad_norm": 0.6018968224525452,
+      "learning_rate": 0.0002,
+      "loss": 1.4789,
+      "step": 3720
+    },
+    {
+      "epoch": 3.1384097602019354,
+      "grad_norm": 0.5767101049423218,
+      "learning_rate": 0.0002,
+      "loss": 1.5518,
+      "step": 3730
+    },
+    {
+      "epoch": 3.146823727387463,
+      "grad_norm": 0.5333963632583618,
+      "learning_rate": 0.0002,
+      "loss": 1.5032,
+      "step": 3740
+    },
+    {
+      "epoch": 3.155237694572991,
+      "grad_norm": 0.5918396711349487,
+      "learning_rate": 0.0002,
+      "loss": 1.4812,
+      "step": 3750
+    },
+    {
+      "epoch": 3.1636516617585193,
+      "grad_norm": 0.5931203365325928,
+      "learning_rate": 0.0002,
+      "loss": 1.4618,
+      "step": 3760
+    },
+    {
+      "epoch": 3.172065628944047,
+      "grad_norm": 0.6562168598175049,
+      "learning_rate": 0.0002,
+      "loss": 1.5592,
+      "step": 3770
+    },
+    {
+      "epoch": 3.180479596129575,
+      "grad_norm": 0.5820156335830688,
+      "learning_rate": 0.0002,
+      "loss": 1.4932,
+      "step": 3780
+    },
+    {
+      "epoch": 3.188893563315103,
+      "grad_norm": 0.5784737467765808,
+      "learning_rate": 0.0002,
+      "loss": 1.4523,
+      "step": 3790
+    },
+    {
+      "epoch": 3.197307530500631,
+      "grad_norm": 0.5506529808044434,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 3800
+    },
+    {
+      "epoch": 3.205721497686159,
+      "grad_norm": 0.6101595163345337,
+      "learning_rate": 0.0002,
+      "loss": 1.4819,
+      "step": 3810
+    },
+    {
+      "epoch": 3.214135464871687,
+      "grad_norm": 0.5597806572914124,
+      "learning_rate": 0.0002,
+      "loss": 1.5185,
+      "step": 3820
+    },
+    {
+      "epoch": 3.222549432057215,
+      "grad_norm": 0.5641011595726013,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 3830
+    },
+    {
+      "epoch": 3.230963399242743,
+      "grad_norm": 0.5892080068588257,
+      "learning_rate": 0.0002,
+      "loss": 1.4702,
+      "step": 3840
+    },
+    {
+      "epoch": 3.239377366428271,
+      "grad_norm": 0.6034760475158691,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 3850
+    },
+    {
+      "epoch": 3.247791333613799,
+      "grad_norm": 0.5112439393997192,
+      "learning_rate": 0.0002,
+      "loss": 1.5499,
+      "step": 3860
+    },
+    {
+      "epoch": 3.256205300799327,
+      "grad_norm": 0.56565922498703,
+      "learning_rate": 0.0002,
+      "loss": 1.5132,
+      "step": 3870
+    },
+    {
+      "epoch": 3.264619267984855,
+      "grad_norm": 0.6155247092247009,
+      "learning_rate": 0.0002,
+      "loss": 1.4892,
+      "step": 3880
+    },
+    {
+      "epoch": 3.273033235170383,
+      "grad_norm": 0.6064623594284058,
+      "learning_rate": 0.0002,
+      "loss": 1.5118,
+      "step": 3890
+    },
+    {
+      "epoch": 3.2814472023559107,
+      "grad_norm": 0.6313768029212952,
+      "learning_rate": 0.0002,
+      "loss": 1.5236,
+      "step": 3900
+    },
+    {
+      "epoch": 3.289861169541439,
+      "grad_norm": 0.5903939008712769,
+      "learning_rate": 0.0002,
+      "loss": 1.5551,
+      "step": 3910
+    },
+    {
+      "epoch": 3.298275136726967,
+      "grad_norm": 0.5770667195320129,
+      "learning_rate": 0.0002,
+      "loss": 1.5703,
+      "step": 3920
+    },
+    {
+      "epoch": 3.3066891039124946,
+      "grad_norm": 0.5785196423530579,
+      "learning_rate": 0.0002,
+      "loss": 1.5159,
+      "step": 3930
+    },
+    {
+      "epoch": 3.3151030710980227,
+      "grad_norm": 0.6468310356140137,
+      "learning_rate": 0.0002,
+      "loss": 1.5277,
+      "step": 3940
+    },
+    {
+      "epoch": 3.323517038283551,
+      "grad_norm": 0.6200279593467712,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 3950
+    },
+    {
+      "epoch": 3.3319310054690785,
+      "grad_norm": 0.5779302716255188,
+      "learning_rate": 0.0002,
+      "loss": 1.5264,
+      "step": 3960
+    },
+    {
+      "epoch": 3.3403449726546066,
+      "grad_norm": 0.5463796854019165,
+      "learning_rate": 0.0002,
+      "loss": 1.4861,
+      "step": 3970
+    },
+    {
+      "epoch": 3.3487589398401347,
+      "grad_norm": 0.6117855906486511,
+      "learning_rate": 0.0002,
+      "loss": 1.541,
+      "step": 3980
+    },
+    {
+      "epoch": 3.357172907025663,
+      "grad_norm": 0.5554766058921814,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 3990
+    },
+    {
+      "epoch": 3.3655868742111905,
+      "grad_norm": 0.6012870073318481,
+      "learning_rate": 0.0002,
+      "loss": 1.5004,
+      "step": 4000
+    },
+    {
+      "epoch": 3.3740008413967186,
+      "grad_norm": 0.5443974137306213,
+      "learning_rate": 0.0002,
+      "loss": 1.473,
+      "step": 4010
+    },
+    {
+      "epoch": 3.3824148085822463,
+      "grad_norm": 0.6636057496070862,
+      "learning_rate": 0.0002,
+      "loss": 1.5139,
+      "step": 4020
+    },
+    {
+      "epoch": 3.3908287757677744,
+      "grad_norm": 0.5801246166229248,
+      "learning_rate": 0.0002,
+      "loss": 1.5141,
+      "step": 4030
+    },
+    {
+      "epoch": 3.3992427429533025,
+      "grad_norm": 0.5668839812278748,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4040
+    },
+    {
+      "epoch": 3.4076567101388306,
+      "grad_norm": 0.7763481736183167,
+      "learning_rate": 0.0002,
+      "loss": 1.523,
+      "step": 4050
+    },
+    {
+      "epoch": 3.4160706773243583,
+      "grad_norm": 0.6675992608070374,
+      "learning_rate": 0.0002,
+      "loss": 1.4932,
+      "step": 4060
+    },
+    {
+      "epoch": 3.4244846445098864,
+      "grad_norm": 0.6290077567100525,
+      "learning_rate": 0.0002,
+      "loss": 1.4959,
+      "step": 4070
+    },
+    {
+      "epoch": 3.4328986116954145,
+      "grad_norm": 0.6040239930152893,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 4080
+    },
+    {
+      "epoch": 3.441312578880942,
+      "grad_norm": 0.6237877607345581,
+      "learning_rate": 0.0002,
+      "loss": 1.5711,
+      "step": 4090
+    },
+    {
+      "epoch": 3.4497265460664703,
+      "grad_norm": 0.5343508124351501,
+      "learning_rate": 0.0002,
+      "loss": 1.4961,
+      "step": 4100
+    },
+    {
+      "epoch": 3.4581405132519984,
+      "grad_norm": 0.6817412972450256,
+      "learning_rate": 0.0002,
+      "loss": 1.5123,
+      "step": 4110
+    },
+    {
+      "epoch": 3.466554480437526,
+      "grad_norm": 0.7115170359611511,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 4120
+    },
+    {
+      "epoch": 3.4749684476230542,
+      "grad_norm": 0.6127332448959351,
+      "learning_rate": 0.0002,
+      "loss": 1.5275,
+      "step": 4130
+    },
+    {
+      "epoch": 3.4833824148085824,
+      "grad_norm": 0.5745994448661804,
+      "learning_rate": 0.0002,
+      "loss": 1.557,
+      "step": 4140
+    },
+    {
+      "epoch": 3.49179638199411,
+      "grad_norm": 0.6248795390129089,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4150
+    },
+    {
+      "epoch": 3.500210349179638,
+      "grad_norm": 0.5821124911308289,
+      "learning_rate": 0.0002,
+      "loss": 1.4885,
+      "step": 4160
+    },
+    {
+      "epoch": 3.5086243163651663,
+      "grad_norm": 0.561416506767273,
+      "learning_rate": 0.0002,
+      "loss": 1.4937,
+      "step": 4170
+    },
+    {
+      "epoch": 3.5170382835506944,
+      "grad_norm": 0.5848962664604187,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 4180
+    },
+    {
+      "epoch": 3.525452250736222,
+      "grad_norm": 0.5335569977760315,
+      "learning_rate": 0.0002,
+      "loss": 1.5892,
+      "step": 4190
+    },
+    {
+      "epoch": 3.53386621792175,
+      "grad_norm": 0.547964870929718,
+      "learning_rate": 0.0002,
+      "loss": 1.5152,
+      "step": 4200
+    },
+    {
+      "epoch": 3.542280185107278,
+      "grad_norm": 0.6157727241516113,
+      "learning_rate": 0.0002,
+      "loss": 1.4887,
+      "step": 4210
+    },
+    {
+      "epoch": 3.550694152292806,
+      "grad_norm": 0.6163121461868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5484,
+      "step": 4220
+    },
+    {
+      "epoch": 3.559108119478334,
+      "grad_norm": 0.5844616293907166,
+      "learning_rate": 0.0002,
+      "loss": 1.5833,
+      "step": 4230
+    },
+    {
+      "epoch": 3.567522086663862,
+      "grad_norm": 0.7104926109313965,
+      "learning_rate": 0.0002,
+      "loss": 1.5305,
+      "step": 4240
+    },
+    {
+      "epoch": 3.57593605384939,
+      "grad_norm": 0.5055213570594788,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4250
+    },
+    {
+      "epoch": 3.584350021034918,
+      "grad_norm": 0.611676812171936,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 4260
+    },
+    {
+      "epoch": 3.592763988220446,
+      "grad_norm": 0.6326440572738647,
+      "learning_rate": 0.0002,
+      "loss": 1.5048,
+      "step": 4270
+    },
+    {
+      "epoch": 3.6011779554059737,
+      "grad_norm": 0.6290925741195679,
+      "learning_rate": 0.0002,
+      "loss": 1.5122,
+      "step": 4280
+    },
+    {
+      "epoch": 3.609591922591502,
+      "grad_norm": 0.5691978931427002,
+      "learning_rate": 0.0002,
+      "loss": 1.5654,
+      "step": 4290
+    },
+    {
+      "epoch": 3.61800588977703,
+      "grad_norm": 0.6071329116821289,
+      "learning_rate": 0.0002,
+      "loss": 1.4854,
+      "step": 4300
+    },
+    {
+      "epoch": 3.626419856962558,
+      "grad_norm": 0.606573224067688,
+      "learning_rate": 0.0002,
+      "loss": 1.5336,
+      "step": 4310
+    },
+    {
+      "epoch": 3.6348338241480858,
+      "grad_norm": 0.5515419244766235,
+      "learning_rate": 0.0002,
+      "loss": 1.6437,
+      "step": 4320
+    },
+    {
+      "epoch": 3.643247791333614,
+      "grad_norm": 0.5964660048484802,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 4330
+    },
+    {
+      "epoch": 3.6516617585191415,
+      "grad_norm": 0.5774146914482117,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 4340
+    },
+    {
+      "epoch": 3.6600757257046697,
+      "grad_norm": 0.5732731223106384,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 4350
+    },
+    {
+      "epoch": 3.6684896928901978,
+      "grad_norm": 0.7354163527488708,
+      "learning_rate": 0.0002,
+      "loss": 1.5682,
+      "step": 4360
+    },
+    {
+      "epoch": 3.676903660075726,
+      "grad_norm": 0.6220902800559998,
+      "learning_rate": 0.0002,
+      "loss": 1.5225,
+      "step": 4370
+    },
+    {
+      "epoch": 3.6853176272612536,
+      "grad_norm": 0.6053991317749023,
+      "learning_rate": 0.0002,
+      "loss": 1.4838,
+      "step": 4380
+    },
+    {
+      "epoch": 3.6937315944467817,
+      "grad_norm": 0.67010897397995,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4390
+    },
+    {
+      "epoch": 3.70214556163231,
+      "grad_norm": 0.6139186024665833,
+      "learning_rate": 0.0002,
+      "loss": 1.5381,
+      "step": 4400
+    },
+    {
+      "epoch": 3.7105595288178375,
+      "grad_norm": 0.5433071851730347,
+      "learning_rate": 0.0002,
+      "loss": 1.5088,
+      "step": 4410
+    },
+    {
+      "epoch": 3.7189734960033656,
+      "grad_norm": 0.5453870296478271,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 4420
+    },
+    {
+      "epoch": 3.7273874631888937,
+      "grad_norm": 0.6401727199554443,
+      "learning_rate": 0.0002,
+      "loss": 1.4549,
+      "step": 4430
+    },
+    {
+      "epoch": 3.735801430374422,
+      "grad_norm": 0.6049367189407349,
+      "learning_rate": 0.0002,
+      "loss": 1.503,
+      "step": 4440
+    },
+    {
+      "epoch": 3.7442153975599495,
+      "grad_norm": 0.5740529298782349,
+      "learning_rate": 0.0002,
+      "loss": 1.5268,
+      "step": 4450
+    },
+    {
+      "epoch": 3.7526293647454776,
+      "grad_norm": 0.6521880626678467,
+      "learning_rate": 0.0002,
+      "loss": 1.5183,
+      "step": 4460
+    },
+    {
+      "epoch": 3.7610433319310053,
+      "grad_norm": 0.7096368074417114,
+      "learning_rate": 0.0002,
+      "loss": 1.5741,
+      "step": 4470
+    },
+    {
+      "epoch": 3.7694572991165334,
+      "grad_norm": 0.5886474251747131,
+      "learning_rate": 0.0002,
+      "loss": 1.5786,
+      "step": 4480
+    },
+    {
+      "epoch": 3.7778712663020615,
+      "grad_norm": 0.5821043252944946,
+      "learning_rate": 0.0002,
+      "loss": 1.5887,
+      "step": 4490
+    },
+    {
+      "epoch": 3.7862852334875896,
+      "grad_norm": 0.628892183303833,
+      "learning_rate": 0.0002,
+      "loss": 1.5777,
+      "step": 4500
+    },
+    {
+      "epoch": 3.7946992006731173,
+      "grad_norm": 0.5962669849395752,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 4510
+    },
+    {
+      "epoch": 3.8031131678586454,
+      "grad_norm": 0.6635549068450928,
+      "learning_rate": 0.0002,
+      "loss": 1.5267,
+      "step": 4520
+    },
+    {
+      "epoch": 3.811527135044173,
+      "grad_norm": 0.6010760068893433,
+      "learning_rate": 0.0002,
+      "loss": 1.5058,
+      "step": 4530
+    },
+    {
+      "epoch": 3.819941102229701,
+      "grad_norm": 0.6322658658027649,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 4540
+    },
+    {
+      "epoch": 3.8283550694152293,
+      "grad_norm": 0.5893137454986572,
+      "learning_rate": 0.0002,
+      "loss": 1.5029,
+      "step": 4550
+    },
+    {
+      "epoch": 3.8367690366007574,
+      "grad_norm": 0.7829602360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.5435,
+      "step": 4560
+    },
+    {
+      "epoch": 3.845183003786285,
+      "grad_norm": 0.6190396547317505,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 4570
+    },
+    {
+      "epoch": 3.853596970971813,
+      "grad_norm": 0.6662813425064087,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 4580
+    },
+    {
+      "epoch": 3.8620109381573413,
+      "grad_norm": 0.5809855461120605,
+      "learning_rate": 0.0002,
+      "loss": 1.5065,
+      "step": 4590
+    },
+    {
+      "epoch": 3.870424905342869,
+      "grad_norm": 0.5779069662094116,
+      "learning_rate": 0.0002,
+      "loss": 1.5041,
+      "step": 4600
+    },
+    {
+      "epoch": 3.878838872528397,
+      "grad_norm": 0.5603038668632507,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 4610
+    },
+    {
+      "epoch": 3.887252839713925,
+      "grad_norm": 0.6274181008338928,
+      "learning_rate": 0.0002,
+      "loss": 1.5372,
+      "step": 4620
+    },
+    {
+      "epoch": 3.8956668068994533,
+      "grad_norm": 0.6810959577560425,
+      "learning_rate": 0.0002,
+      "loss": 1.4996,
+      "step": 4630
+    },
+    {
+      "epoch": 3.904080774084981,
+      "grad_norm": 0.5647315979003906,
+      "learning_rate": 0.0002,
+      "loss": 1.4956,
+      "step": 4640
+    },
+    {
+      "epoch": 3.912494741270509,
+      "grad_norm": 0.6830295324325562,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 4650
+    },
+    {
+      "epoch": 3.920908708456037,
+      "grad_norm": 0.652565598487854,
+      "learning_rate": 0.0002,
+      "loss": 1.535,
+      "step": 4660
+    },
+    {
+      "epoch": 3.929322675641565,
+      "grad_norm": 0.5806284546852112,
+      "learning_rate": 0.0002,
+      "loss": 1.4772,
+      "step": 4670
+    },
+    {
+      "epoch": 3.937736642827093,
+      "grad_norm": 0.6825073957443237,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 4680
+    },
+    {
+      "epoch": 3.946150610012621,
+      "grad_norm": 0.6149451732635498,
+      "learning_rate": 0.0002,
+      "loss": 1.5516,
+      "step": 4690
+    },
+    {
+      "epoch": 3.954564577198149,
+      "grad_norm": 0.6152557134628296,
+      "learning_rate": 0.0002,
+      "loss": 1.5608,
+      "step": 4700
+    },
+    {
+      "epoch": 3.962978544383677,
+      "grad_norm": 0.6239011883735657,
+      "learning_rate": 0.0002,
+      "loss": 1.4897,
+      "step": 4710
+    },
+    {
+      "epoch": 3.971392511569205,
+      "grad_norm": 0.6485443115234375,
+      "learning_rate": 0.0002,
+      "loss": 1.538,
+      "step": 4720
+    },
+    {
+      "epoch": 3.9798064787547327,
+      "grad_norm": 0.6449228525161743,
+      "learning_rate": 0.0002,
+      "loss": 1.5226,
+      "step": 4730
+    },
+    {
+      "epoch": 3.988220445940261,
+      "grad_norm": 0.6526407599449158,
+      "learning_rate": 0.0002,
+      "loss": 1.5087,
+      "step": 4740
+    },
+    {
+      "epoch": 3.996634413125789,
+      "grad_norm": 0.6277706027030945,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4750
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.871641755104065,
+      "eval_runtime": 37.9637,
+      "eval_samples_per_second": 13.566,
+      "eval_steps_per_second": 1.712,
+      "step": 4754
+    },
+    {
+      "epoch": 4.005048380311317,
+      "grad_norm": 0.6994837522506714,
+      "learning_rate": 0.0002,
+      "loss": 1.4744,
+      "step": 4760
+    },
+    {
+      "epoch": 4.013462347496845,
+      "grad_norm": 0.8728373050689697,
+      "learning_rate": 0.0002,
+      "loss": 1.4433,
+      "step": 4770
+    },
+    {
+      "epoch": 4.021876314682372,
+      "grad_norm": 0.688679575920105,
+      "learning_rate": 0.0002,
+      "loss": 1.3329,
+      "step": 4780
+    },
+    {
+      "epoch": 4.0302902818679005,
+      "grad_norm": 0.6313387155532837,
+      "learning_rate": 0.0002,
+      "loss": 1.3999,
+      "step": 4790
+    },
+    {
+      "epoch": 4.038704249053429,
+      "grad_norm": 0.6577984690666199,
+      "learning_rate": 0.0002,
+      "loss": 1.3346,
+      "step": 4800
+    },
+    {
+      "epoch": 4.047118216238957,
+      "grad_norm": 0.7938185930252075,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 4810
+    },
+    {
+      "epoch": 4.055532183424485,
+      "grad_norm": 0.760399580001831,
+      "learning_rate": 0.0002,
+      "loss": 1.3716,
+      "step": 4820
+    },
+    {
+      "epoch": 4.063946150610013,
+      "grad_norm": 0.7329602241516113,
+      "learning_rate": 0.0002,
+      "loss": 1.4321,
+      "step": 4830
+    },
+    {
+      "epoch": 4.07236011779554,
+      "grad_norm": 0.7778576016426086,
+      "learning_rate": 0.0002,
+      "loss": 1.4133,
+      "step": 4840
+    },
+    {
+      "epoch": 4.080774084981068,
+      "grad_norm": 0.8235865235328674,
+      "learning_rate": 0.0002,
+      "loss": 1.4372,
+      "step": 4850
+    },
+    {
+      "epoch": 4.089188052166596,
+      "grad_norm": 0.7743754386901855,
+      "learning_rate": 0.0002,
+      "loss": 1.3719,
+      "step": 4860
+    },
+    {
+      "epoch": 4.0976020193521245,
+      "grad_norm": 0.8145367503166199,
+      "learning_rate": 0.0002,
+      "loss": 1.3787,
+      "step": 4870
+    },
+    {
+      "epoch": 4.106015986537653,
+      "grad_norm": 0.8517307639122009,
+      "learning_rate": 0.0002,
+      "loss": 1.356,
+      "step": 4880
+    },
+    {
+      "epoch": 4.114429953723181,
+      "grad_norm": 0.8208953142166138,
+      "learning_rate": 0.0002,
+      "loss": 1.4191,
+      "step": 4890
+    },
+    {
+      "epoch": 4.122843920908709,
+      "grad_norm": 0.8437790870666504,
+      "learning_rate": 0.0002,
+      "loss": 1.3189,
+      "step": 4900
+    },
+    {
+      "epoch": 4.131257888094236,
+      "grad_norm": 0.716672420501709,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 4910
+    },
+    {
+      "epoch": 4.139671855279764,
+      "grad_norm": 0.7656235098838806,
+      "learning_rate": 0.0002,
+      "loss": 1.4392,
+      "step": 4920
+    },
+    {
+      "epoch": 4.148085822465292,
+      "grad_norm": 0.7209306955337524,
+      "learning_rate": 0.0002,
+      "loss": 1.3408,
+      "step": 4930
+    },
+    {
+      "epoch": 4.1564997896508205,
+      "grad_norm": 0.7731267809867859,
+      "learning_rate": 0.0002,
+      "loss": 1.3639,
+      "step": 4940
+    },
+    {
+      "epoch": 4.164913756836349,
+      "grad_norm": 0.7477553486824036,
+      "learning_rate": 0.0002,
+      "loss": 1.4151,
+      "step": 4950
+    },
+    {
+      "epoch": 4.173327724021877,
+      "grad_norm": 0.7372981309890747,
+      "learning_rate": 0.0002,
+      "loss": 1.3485,
+      "step": 4960
+    },
+    {
+      "epoch": 4.181741691207404,
+      "grad_norm": 0.6582154035568237,
+      "learning_rate": 0.0002,
+      "loss": 1.3901,
+      "step": 4970
+    },
+    {
+      "epoch": 4.190155658392932,
+      "grad_norm": 0.7003206610679626,
+      "learning_rate": 0.0002,
+      "loss": 1.3343,
+      "step": 4980
+    },
+    {
+      "epoch": 4.19856962557846,
+      "grad_norm": 0.735223650932312,
+      "learning_rate": 0.0002,
+      "loss": 1.4098,
+      "step": 4990
+    },
+    {
+      "epoch": 4.206983592763988,
+      "grad_norm": 0.7832302451133728,
+      "learning_rate": 0.0002,
+      "loss": 1.3564,
+      "step": 5000
+    },
+    {
+      "epoch": 4.215397559949516,
+      "grad_norm": 0.8819546103477478,
+      "learning_rate": 0.0002,
+      "loss": 1.3622,
+      "step": 5010
+    },
+    {
+      "epoch": 4.2238115271350445,
+      "grad_norm": 0.9325336813926697,
+      "learning_rate": 0.0002,
+      "loss": 1.4438,
+      "step": 5020
+    },
+    {
+      "epoch": 4.232225494320572,
+      "grad_norm": 0.7007517218589783,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 5030
+    },
+    {
+      "epoch": 4.2406394615061,
+      "grad_norm": 0.7118321061134338,
+      "learning_rate": 0.0002,
+      "loss": 1.3683,
+      "step": 5040
+    },
+    {
+      "epoch": 4.249053428691628,
+      "grad_norm": 0.6578946709632874,
+      "learning_rate": 0.0002,
+      "loss": 1.2365,
+      "step": 5050
+    },
+    {
+      "epoch": 4.257467395877156,
+      "grad_norm": 0.9438983798027039,
+      "learning_rate": 0.0002,
+      "loss": 1.3696,
+      "step": 5060
+    },
+    {
+      "epoch": 4.265881363062684,
+      "grad_norm": 0.703037679195404,
+      "learning_rate": 0.0002,
+      "loss": 1.3868,
+      "step": 5070
+    },
+    {
+      "epoch": 4.274295330248212,
+      "grad_norm": 0.7286025285720825,
+      "learning_rate": 0.0002,
+      "loss": 1.3687,
+      "step": 5080
+    },
+    {
+      "epoch": 4.28270929743374,
+      "grad_norm": 0.750689685344696,
+      "learning_rate": 0.0002,
+      "loss": 1.3605,
+      "step": 5090
+    },
+    {
+      "epoch": 4.291123264619268,
+      "grad_norm": 0.869753360748291,
+      "learning_rate": 0.0002,
+      "loss": 1.5089,
+      "step": 5100
+    },
+    {
+      "epoch": 4.299537231804796,
+      "grad_norm": 0.8712980151176453,
+      "learning_rate": 0.0002,
+      "loss": 1.4128,
+      "step": 5110
+    },
+    {
+      "epoch": 4.307951198990324,
+      "grad_norm": 0.690263569355011,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 5120
+    },
+    {
+      "epoch": 4.316365166175852,
+      "grad_norm": 0.7114760279655457,
+      "learning_rate": 0.0002,
+      "loss": 1.4088,
+      "step": 5130
+    },
+    {
+      "epoch": 4.32477913336138,
+      "grad_norm": 0.7588112354278564,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 5140
+    },
+    {
+      "epoch": 4.333193100546908,
+      "grad_norm": 0.7556202411651611,
+      "learning_rate": 0.0002,
+      "loss": 1.4408,
+      "step": 5150
+    },
+    {
+      "epoch": 4.341607067732435,
+      "grad_norm": 0.8357610702514648,
+      "learning_rate": 0.0002,
+      "loss": 1.4203,
+      "step": 5160
+    },
+    {
+      "epoch": 4.3500210349179635,
+      "grad_norm": 0.8054035902023315,
+      "learning_rate": 0.0002,
+      "loss": 1.3348,
+      "step": 5170
+    },
+    {
+      "epoch": 4.358435002103492,
+      "grad_norm": 0.7637107968330383,
+      "learning_rate": 0.0002,
+      "loss": 1.3109,
+      "step": 5180
+    },
+    {
+      "epoch": 4.36684896928902,
+      "grad_norm": 0.757481038570404,
+      "learning_rate": 0.0002,
+      "loss": 1.3744,
+      "step": 5190
+    },
+    {
+      "epoch": 4.375262936474548,
+      "grad_norm": 0.7185863852500916,
+      "learning_rate": 0.0002,
+      "loss": 1.3622,
+      "step": 5200
+    },
+    {
+      "epoch": 4.383676903660076,
+      "grad_norm": 0.7326455116271973,
+      "learning_rate": 0.0002,
+      "loss": 1.3896,
+      "step": 5210
+    },
+    {
+      "epoch": 4.392090870845603,
+      "grad_norm": 0.7980523109436035,
+      "learning_rate": 0.0002,
+      "loss": 1.4098,
+      "step": 5220
+    },
+    {
+      "epoch": 4.400504838031131,
+      "grad_norm": 0.8526999354362488,
+      "learning_rate": 0.0002,
+      "loss": 1.3783,
+      "step": 5230
+    },
+    {
+      "epoch": 4.4089188052166595,
+      "grad_norm": 0.7012337446212769,
+      "learning_rate": 0.0002,
+      "loss": 1.4022,
+      "step": 5240
+    },
+    {
+      "epoch": 4.417332772402188,
+      "grad_norm": 0.8217827677726746,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 5250
+    },
+    {
+      "epoch": 4.425746739587716,
+      "grad_norm": 0.7141005396842957,
+      "learning_rate": 0.0002,
+      "loss": 1.3482,
+      "step": 5260
+    },
+    {
+      "epoch": 4.434160706773244,
+      "grad_norm": 0.7094302177429199,
+      "learning_rate": 0.0002,
+      "loss": 1.3699,
+      "step": 5270
+    },
+    {
+      "epoch": 4.442574673958772,
+      "grad_norm": 0.7234613299369812,
+      "learning_rate": 0.0002,
+      "loss": 1.3527,
+      "step": 5280
+    },
+    {
+      "epoch": 4.450988641144299,
+      "grad_norm": 0.7530457973480225,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 5290
+    },
+    {
+      "epoch": 4.459402608329827,
+      "grad_norm": 0.7300912141799927,
+      "learning_rate": 0.0002,
+      "loss": 1.3944,
+      "step": 5300
+    },
+    {
+      "epoch": 4.467816575515355,
+      "grad_norm": 0.825443685054779,
+      "learning_rate": 0.0002,
+      "loss": 1.3844,
+      "step": 5310
+    },
+    {
+      "epoch": 4.4762305427008835,
+      "grad_norm": 0.7559658885002136,
+      "learning_rate": 0.0002,
+      "loss": 1.3648,
+      "step": 5320
+    },
+    {
+      "epoch": 4.484644509886412,
+      "grad_norm": 0.8817561268806458,
+      "learning_rate": 0.0002,
+      "loss": 1.4364,
+      "step": 5330
+    },
+    {
+      "epoch": 4.49305847707194,
+      "grad_norm": 0.8203575611114502,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 5340
+    },
+    {
+      "epoch": 4.501472444257468,
+      "grad_norm": 0.7677690982818604,
+      "learning_rate": 0.0002,
+      "loss": 1.3996,
+      "step": 5350
+    },
+    {
+      "epoch": 4.509886411442995,
+      "grad_norm": 0.657085120677948,
+      "learning_rate": 0.0002,
+      "loss": 1.4142,
+      "step": 5360
+    },
+    {
+      "epoch": 4.518300378628523,
+      "grad_norm": 0.7939504384994507,
+      "learning_rate": 0.0002,
+      "loss": 1.3722,
+      "step": 5370
+    },
+    {
+      "epoch": 4.526714345814051,
+      "grad_norm": 0.6971889138221741,
+      "learning_rate": 0.0002,
+      "loss": 1.4361,
+      "step": 5380
+    },
+    {
+      "epoch": 4.535128312999579,
+      "grad_norm": 0.6984175443649292,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 5390
+    },
+    {
+      "epoch": 4.5435422801851075,
+      "grad_norm": 0.8504858613014221,
+      "learning_rate": 0.0002,
+      "loss": 1.341,
+      "step": 5400
+    },
+    {
+      "epoch": 4.551956247370635,
+      "grad_norm": 0.9134073853492737,
+      "learning_rate": 0.0002,
+      "loss": 1.4026,
+      "step": 5410
+    },
+    {
+      "epoch": 4.560370214556163,
+      "grad_norm": 0.7765598893165588,
+      "learning_rate": 0.0002,
+      "loss": 1.4375,
+      "step": 5420
+    },
+    {
+      "epoch": 4.568784181741691,
+      "grad_norm": 0.6991009712219238,
+      "learning_rate": 0.0002,
+      "loss": 1.4832,
+      "step": 5430
+    },
+    {
+      "epoch": 4.577198148927219,
+      "grad_norm": 0.8393039107322693,
+      "learning_rate": 0.0002,
+      "loss": 1.4021,
+      "step": 5440
+    },
+    {
+      "epoch": 4.585612116112747,
+      "grad_norm": 0.7685918211936951,
+      "learning_rate": 0.0002,
+      "loss": 1.3976,
+      "step": 5450
+    },
+    {
+      "epoch": 4.594026083298275,
+      "grad_norm": 0.7135679721832275,
+      "learning_rate": 0.0002,
+      "loss": 1.3883,
+      "step": 5460
+    },
+    {
+      "epoch": 4.6024400504838034,
+      "grad_norm": 0.6728870868682861,
+      "learning_rate": 0.0002,
+      "loss": 1.4083,
+      "step": 5470
+    },
+    {
+      "epoch": 4.610854017669331,
+      "grad_norm": 0.7139479517936707,
+      "learning_rate": 0.0002,
+      "loss": 1.3698,
+      "step": 5480
+    },
+    {
+      "epoch": 4.619267984854859,
+      "grad_norm": 0.8476598858833313,
+      "learning_rate": 0.0002,
+      "loss": 1.3498,
+      "step": 5490
+    },
+    {
+      "epoch": 4.627681952040387,
+      "grad_norm": 0.8034361004829407,
+      "learning_rate": 0.0002,
+      "loss": 1.3389,
+      "step": 5500
+    },
+    {
+      "epoch": 4.636095919225915,
+      "grad_norm": 0.7452183961868286,
+      "learning_rate": 0.0002,
+      "loss": 1.4179,
+      "step": 5510
+    },
+    {
+      "epoch": 4.644509886411443,
+      "grad_norm": 0.8394148945808411,
+      "learning_rate": 0.0002,
+      "loss": 1.4031,
+      "step": 5520
+    },
+    {
+      "epoch": 4.652923853596971,
+      "grad_norm": 0.7480153441429138,
+      "learning_rate": 0.0002,
+      "loss": 1.4561,
+      "step": 5530
+    },
+    {
+      "epoch": 4.661337820782499,
+      "grad_norm": 0.7781714797019958,
+      "learning_rate": 0.0002,
+      "loss": 1.378,
+      "step": 5540
+    },
+    {
+      "epoch": 4.669751787968027,
+      "grad_norm": 1.0058213472366333,
+      "learning_rate": 0.0002,
+      "loss": 1.3924,
+      "step": 5550
+    },
+    {
+      "epoch": 4.678165755153555,
+      "grad_norm": 0.7403179407119751,
+      "learning_rate": 0.0002,
+      "loss": 1.4198,
+      "step": 5560
+    },
+    {
+      "epoch": 4.686579722339083,
+      "grad_norm": 0.7270476818084717,
+      "learning_rate": 0.0002,
+      "loss": 1.4328,
+      "step": 5570
+    },
+    {
+      "epoch": 4.694993689524611,
+      "grad_norm": 0.760877788066864,
+      "learning_rate": 0.0002,
+      "loss": 1.378,
+      "step": 5580
+    },
+    {
+      "epoch": 4.703407656710139,
+      "grad_norm": 0.8097004890441895,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 5590
+    },
+    {
+      "epoch": 4.711821623895667,
+      "grad_norm": 0.9096523523330688,
+      "learning_rate": 0.0002,
+      "loss": 1.3661,
+      "step": 5600
+    },
+    {
+      "epoch": 4.720235591081195,
+      "grad_norm": 0.7262444496154785,
+      "learning_rate": 0.0002,
+      "loss": 1.4012,
+      "step": 5610
+    },
+    {
+      "epoch": 4.7286495582667225,
+      "grad_norm": 0.8207762837409973,
+      "learning_rate": 0.0002,
+      "loss": 1.422,
+      "step": 5620
+    },
+    {
+      "epoch": 4.737063525452251,
+      "grad_norm": 0.8089601993560791,
+      "learning_rate": 0.0002,
+      "loss": 1.4017,
+      "step": 5630
+    },
+    {
+      "epoch": 4.745477492637779,
+      "grad_norm": 0.7609543800354004,
+      "learning_rate": 0.0002,
+      "loss": 1.3675,
+      "step": 5640
+    },
+    {
+      "epoch": 4.753891459823307,
+      "grad_norm": 0.7273501753807068,
+      "learning_rate": 0.0002,
+      "loss": 1.4085,
+      "step": 5650
+    },
+    {
+      "epoch": 4.762305427008835,
+      "grad_norm": 0.7800219058990479,
+      "learning_rate": 0.0002,
+      "loss": 1.3849,
+      "step": 5660
+    },
+    {
+      "epoch": 4.770719394194362,
+      "grad_norm": 0.8558377623558044,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 5670
+    },
+    {
+      "epoch": 4.77913336137989,
+      "grad_norm": 0.7131547927856445,
+      "learning_rate": 0.0002,
+      "loss": 1.3831,
+      "step": 5680
+    },
+    {
+      "epoch": 4.787547328565418,
+      "grad_norm": 0.7651025056838989,
+      "learning_rate": 0.0002,
+      "loss": 1.407,
+      "step": 5690
+    },
+    {
+      "epoch": 4.7959612957509465,
+      "grad_norm": 0.8129976391792297,
+      "learning_rate": 0.0002,
+      "loss": 1.3882,
+      "step": 5700
+    },
+    {
+      "epoch": 4.804375262936475,
+      "grad_norm": 0.8019895553588867,
+      "learning_rate": 0.0002,
+      "loss": 1.4347,
+      "step": 5710
+    },
+    {
+      "epoch": 4.812789230122003,
+      "grad_norm": 0.7692018151283264,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 5720
+    },
+    {
+      "epoch": 4.821203197307531,
+      "grad_norm": 0.6893943548202515,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 5730
+    },
+    {
+      "epoch": 4.829617164493058,
+      "grad_norm": 0.6881810426712036,
+      "learning_rate": 0.0002,
+      "loss": 1.4453,
+      "step": 5740
+    },
+    {
+      "epoch": 4.838031131678586,
+      "grad_norm": 0.7838267683982849,
+      "learning_rate": 0.0002,
+      "loss": 1.4775,
+      "step": 5750
+    },
+    {
+      "epoch": 4.846445098864114,
+      "grad_norm": 0.727799117565155,
+      "learning_rate": 0.0002,
+      "loss": 1.3857,
+      "step": 5760
+    },
+    {
+      "epoch": 4.8548590660496425,
+      "grad_norm": 0.7458277344703674,
+      "learning_rate": 0.0002,
+      "loss": 1.4685,
+      "step": 5770
+    },
+    {
+      "epoch": 4.863273033235171,
+      "grad_norm": 0.903802216053009,
+      "learning_rate": 0.0002,
+      "loss": 1.4426,
+      "step": 5780
+    },
+    {
+      "epoch": 4.871687000420699,
+      "grad_norm": 0.7983472347259521,
+      "learning_rate": 0.0002,
+      "loss": 1.451,
+      "step": 5790
+    },
+    {
+      "epoch": 4.880100967606227,
+      "grad_norm": 0.6894361972808838,
+      "learning_rate": 0.0002,
+      "loss": 1.4534,
+      "step": 5800
+    },
+    {
+      "epoch": 4.888514934791754,
+      "grad_norm": 0.7499409317970276,
+      "learning_rate": 0.0002,
+      "loss": 1.4486,
+      "step": 5810
+    },
+    {
+      "epoch": 4.896928901977282,
+      "grad_norm": 0.7362820506095886,
+      "learning_rate": 0.0002,
+      "loss": 1.4253,
+      "step": 5820
+    },
+    {
+      "epoch": 4.90534286916281,
+      "grad_norm": 0.8341619968414307,
+      "learning_rate": 0.0002,
+      "loss": 1.3763,
+      "step": 5830
+    },
+    {
+      "epoch": 4.913756836348338,
+      "grad_norm": 0.9604470133781433,
+      "learning_rate": 0.0002,
+      "loss": 1.3748,
+      "step": 5840
+    },
+    {
+      "epoch": 4.9221708035338665,
+      "grad_norm": 0.8916844129562378,
+      "learning_rate": 0.0002,
+      "loss": 1.3658,
+      "step": 5850
+    },
+    {
+      "epoch": 4.930584770719394,
+      "grad_norm": 0.8519647121429443,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 5860
+    },
+    {
+      "epoch": 4.938998737904922,
+      "grad_norm": 0.7946906089782715,
+      "learning_rate": 0.0002,
+      "loss": 1.424,
+      "step": 5870
+    },
+    {
+      "epoch": 4.94741270509045,
+      "grad_norm": 0.7843789458274841,
+      "learning_rate": 0.0002,
+      "loss": 1.4071,
+      "step": 5880
+    },
+    {
+      "epoch": 4.955826672275978,
+      "grad_norm": 0.707618772983551,
+      "learning_rate": 0.0002,
+      "loss": 1.4021,
+      "step": 5890
+    },
+    {
+      "epoch": 4.964240639461506,
+      "grad_norm": 0.7704206109046936,
+      "learning_rate": 0.0002,
+      "loss": 1.502,
+      "step": 5900
+    },
+    {
+      "epoch": 4.972654606647034,
+      "grad_norm": 0.7160256505012512,
+      "learning_rate": 0.0002,
+      "loss": 1.4456,
+      "step": 5910
+    },
+    {
+      "epoch": 4.981068573832562,
+      "grad_norm": 0.7020420432090759,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 5920
+    },
+    {
+      "epoch": 4.98948254101809,
+      "grad_norm": 0.7576286792755127,
+      "learning_rate": 0.0002,
+      "loss": 1.4037,
+      "step": 5930
+    },
+    {
+      "epoch": 4.997896508203618,
+      "grad_norm": 0.8573036789894104,
+      "learning_rate": 0.0002,
+      "loss": 1.414,
+      "step": 5940
+    },
+    {
+      "epoch": 4.999579301640724,
+      "eval_loss": 1.9353811740875244,
+      "eval_runtime": 37.9208,
+      "eval_samples_per_second": 13.581,
+      "eval_steps_per_second": 1.714,
+      "step": 5942
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 9504,
+  "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": 2.7500559420358656e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/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-0.8-num-5623-sd-42/checkpoint-5942/training_args.bin b/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-0.8-num-5623-sd-42/checkpoint-5942/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f1502d478cfbb1424f707352d007b740bde5e373
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-5942/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df2b79d3acefeedef5a0229881de39ec68ef9b40046a60d7976a49f7e6b3b936
+size 5560
diff --git a/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-0.8-num-5623-sd-42/checkpoint-7131/README.md b/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-0.8-num-5623-sd-42/checkpoint-7131/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-7131/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_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-0.8-num-5623-sd-42/checkpoint-7131/adapter_config.json b/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-0.8-num-5623-sd-42/checkpoint-7131/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-7131/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_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-0.8-num-5623-sd-42/checkpoint-7131/adapter_model.safetensors b/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-0.8-num-5623-sd-42/checkpoint-7131/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ddc54a30306eb450d8a0bab8b59976c957d1e6d6
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-7131/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b550c6fb37ebab4c946fa28d7dd8ef81f3e06eb6addb32229bae94c960e2946c
+size 109069176
diff --git a/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-0.8-num-5623-sd-42/checkpoint-7131/optimizer.pt b/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-0.8-num-5623-sd-42/checkpoint-7131/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1156b68bf74a1e6a384134df8aae6de37a85baa9
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-7131/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c76177c28be06b4676d4cfc29c9d51e132c88d59f9f1f18de61ef1d0ad426488
+size 55532666
diff --git a/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-0.8-num-5623-sd-42/checkpoint-7131/rng_state.pth b/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-0.8-num-5623-sd-42/checkpoint-7131/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..49c6aa9c6911c6e0207230c9932fdff6528b8862
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-7131/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a097e50fb002ecd808587da0d513b7600feddbfe54140d1afd35f54d81ae974
+size 14244
diff --git a/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-0.8-num-5623-sd-42/checkpoint-7131/scheduler.pt b/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-0.8-num-5623-sd-42/checkpoint-7131/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..35f4beb5ec1490e09e1f60d088fa916c080a994f
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-7131/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14292654b4b970cbe29282170061263fd878014175c090c19c21eb65c6808883
+size 1064
diff --git a/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-0.8-num-5623-sd-42/checkpoint-7131/special_tokens_map.json b/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-0.8-num-5623-sd-42/checkpoint-7131/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-7131/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_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-0.8-num-5623-sd-42/checkpoint-7131/tokenizer.json b/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-0.8-num-5623-sd-42/checkpoint-7131/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-7131/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_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-0.8-num-5623-sd-42/checkpoint-7131/tokenizer_config.json b/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-0.8-num-5623-sd-42/checkpoint-7131/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-7131/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_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-0.8-num-5623-sd-42/checkpoint-7131/trainer_state.json b/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-0.8-num-5623-sd-42/checkpoint-7131/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..7499bebc8e24b0b6efe3636f4964f8d0c1e4c43c
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-7131/trainer_state.json
@@ -0,0 +1,5072 @@
+{
+  "best_metric": 1.8055059909820557,
+  "best_model_checkpoint": "outputs-001/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-0.8-num-5623-sd-42/checkpoint-2377",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 7131,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008413967185527976,
+      "grad_norm": 0.5458821654319763,
+      "learning_rate": 0.0002,
+      "loss": 2.56,
+      "step": 10
+    },
+    {
+      "epoch": 0.016827934371055953,
+      "grad_norm": 0.7293308973312378,
+      "learning_rate": 0.0002,
+      "loss": 2.3235,
+      "step": 20
+    },
+    {
+      "epoch": 0.02524190155658393,
+      "grad_norm": 0.47792306542396545,
+      "learning_rate": 0.0002,
+      "loss": 2.0815,
+      "step": 30
+    },
+    {
+      "epoch": 0.033655868742111905,
+      "grad_norm": 0.5944402813911438,
+      "learning_rate": 0.0002,
+      "loss": 1.9718,
+      "step": 40
+    },
+    {
+      "epoch": 0.04206983592763988,
+      "grad_norm": 0.5415359735488892,
+      "learning_rate": 0.0002,
+      "loss": 1.8848,
+      "step": 50
+    },
+    {
+      "epoch": 0.05048380311316786,
+      "grad_norm": 0.535713791847229,
+      "learning_rate": 0.0002,
+      "loss": 1.8953,
+      "step": 60
+    },
+    {
+      "epoch": 0.058897770298695834,
+      "grad_norm": 0.5184146761894226,
+      "learning_rate": 0.0002,
+      "loss": 1.937,
+      "step": 70
+    },
+    {
+      "epoch": 0.06731173748422381,
+      "grad_norm": 0.458926796913147,
+      "learning_rate": 0.0002,
+      "loss": 1.8396,
+      "step": 80
+    },
+    {
+      "epoch": 0.07572570466975179,
+      "grad_norm": 0.4780142307281494,
+      "learning_rate": 0.0002,
+      "loss": 1.8677,
+      "step": 90
+    },
+    {
+      "epoch": 0.08413967185527976,
+      "grad_norm": 0.79965740442276,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 100
+    },
+    {
+      "epoch": 0.09255363904080774,
+      "grad_norm": 0.4498862028121948,
+      "learning_rate": 0.0002,
+      "loss": 1.9081,
+      "step": 110
+    },
+    {
+      "epoch": 0.10096760622633572,
+      "grad_norm": 0.39338430762290955,
+      "learning_rate": 0.0002,
+      "loss": 1.8503,
+      "step": 120
+    },
+    {
+      "epoch": 0.10938157341186369,
+      "grad_norm": 0.9588953852653503,
+      "learning_rate": 0.0002,
+      "loss": 1.8637,
+      "step": 130
+    },
+    {
+      "epoch": 0.11779554059739167,
+      "grad_norm": 0.41675639152526855,
+      "learning_rate": 0.0002,
+      "loss": 1.8676,
+      "step": 140
+    },
+    {
+      "epoch": 0.12620950778291964,
+      "grad_norm": 0.44519832730293274,
+      "learning_rate": 0.0002,
+      "loss": 1.8904,
+      "step": 150
+    },
+    {
+      "epoch": 0.13462347496844762,
+      "grad_norm": 0.4176260530948639,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 160
+    },
+    {
+      "epoch": 0.1430374421539756,
+      "grad_norm": 0.35840365290641785,
+      "learning_rate": 0.0002,
+      "loss": 1.8398,
+      "step": 170
+    },
+    {
+      "epoch": 0.15145140933950357,
+      "grad_norm": 0.3794495463371277,
+      "learning_rate": 0.0002,
+      "loss": 1.8666,
+      "step": 180
+    },
+    {
+      "epoch": 0.15986537652503155,
+      "grad_norm": 0.4563522934913635,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 190
+    },
+    {
+      "epoch": 0.16827934371055953,
+      "grad_norm": 0.37057486176490784,
+      "learning_rate": 0.0002,
+      "loss": 1.8893,
+      "step": 200
+    },
+    {
+      "epoch": 0.1766933108960875,
+      "grad_norm": 0.44081518054008484,
+      "learning_rate": 0.0002,
+      "loss": 1.7995,
+      "step": 210
+    },
+    {
+      "epoch": 0.18510727808161548,
+      "grad_norm": 0.46078577637672424,
+      "learning_rate": 0.0002,
+      "loss": 1.9048,
+      "step": 220
+    },
+    {
+      "epoch": 0.19352124526714345,
+      "grad_norm": 0.36132094264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.8403,
+      "step": 230
+    },
+    {
+      "epoch": 0.20193521245267143,
+      "grad_norm": 0.3747289180755615,
+      "learning_rate": 0.0002,
+      "loss": 1.8827,
+      "step": 240
+    },
+    {
+      "epoch": 0.2103491796381994,
+      "grad_norm": 0.3540179133415222,
+      "learning_rate": 0.0002,
+      "loss": 1.8382,
+      "step": 250
+    },
+    {
+      "epoch": 0.21876314682372738,
+      "grad_norm": 0.3461375832557678,
+      "learning_rate": 0.0002,
+      "loss": 1.8196,
+      "step": 260
+    },
+    {
+      "epoch": 0.22717711400925536,
+      "grad_norm": 0.3436960279941559,
+      "learning_rate": 0.0002,
+      "loss": 1.8509,
+      "step": 270
+    },
+    {
+      "epoch": 0.23559108119478334,
+      "grad_norm": 0.35403719544410706,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 280
+    },
+    {
+      "epoch": 0.2440050483803113,
+      "grad_norm": 0.37142616510391235,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 290
+    },
+    {
+      "epoch": 0.2524190155658393,
+      "grad_norm": 0.3307955861091614,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 300
+    },
+    {
+      "epoch": 0.2608329827513673,
+      "grad_norm": 0.32855314016342163,
+      "learning_rate": 0.0002,
+      "loss": 1.817,
+      "step": 310
+    },
+    {
+      "epoch": 0.26924694993689524,
+      "grad_norm": 0.3299003839492798,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 320
+    },
+    {
+      "epoch": 0.27766091712242325,
+      "grad_norm": 0.44311287999153137,
+      "learning_rate": 0.0002,
+      "loss": 1.8129,
+      "step": 330
+    },
+    {
+      "epoch": 0.2860748843079512,
+      "grad_norm": 0.32989758253097534,
+      "learning_rate": 0.0002,
+      "loss": 1.8232,
+      "step": 340
+    },
+    {
+      "epoch": 0.2944888514934792,
+      "grad_norm": 0.34400200843811035,
+      "learning_rate": 0.0002,
+      "loss": 1.7716,
+      "step": 350
+    },
+    {
+      "epoch": 0.30290281867900715,
+      "grad_norm": 0.36286211013793945,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 360
+    },
+    {
+      "epoch": 0.31131678586453515,
+      "grad_norm": 0.406827837228775,
+      "learning_rate": 0.0002,
+      "loss": 1.8025,
+      "step": 370
+    },
+    {
+      "epoch": 0.3197307530500631,
+      "grad_norm": 0.36299195885658264,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 380
+    },
+    {
+      "epoch": 0.3281447202355911,
+      "grad_norm": 0.3477257192134857,
+      "learning_rate": 0.0002,
+      "loss": 1.837,
+      "step": 390
+    },
+    {
+      "epoch": 0.33655868742111905,
+      "grad_norm": 0.3730369210243225,
+      "learning_rate": 0.0002,
+      "loss": 1.7767,
+      "step": 400
+    },
+    {
+      "epoch": 0.34497265460664706,
+      "grad_norm": 0.4644559919834137,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 410
+    },
+    {
+      "epoch": 0.353386621792175,
+      "grad_norm": 0.406576544046402,
+      "learning_rate": 0.0002,
+      "loss": 1.7538,
+      "step": 420
+    },
+    {
+      "epoch": 0.361800588977703,
+      "grad_norm": 0.3612699508666992,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 430
+    },
+    {
+      "epoch": 0.37021455616323096,
+      "grad_norm": 0.3243742287158966,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 440
+    },
+    {
+      "epoch": 0.37862852334875896,
+      "grad_norm": 0.36671221256256104,
+      "learning_rate": 0.0002,
+      "loss": 1.8851,
+      "step": 450
+    },
+    {
+      "epoch": 0.3870424905342869,
+      "grad_norm": 0.3565002381801605,
+      "learning_rate": 0.0002,
+      "loss": 1.8853,
+      "step": 460
+    },
+    {
+      "epoch": 0.3954564577198149,
+      "grad_norm": 0.34630221128463745,
+      "learning_rate": 0.0002,
+      "loss": 1.8923,
+      "step": 470
+    },
+    {
+      "epoch": 0.40387042490534286,
+      "grad_norm": 0.3353537321090698,
+      "learning_rate": 0.0002,
+      "loss": 1.8234,
+      "step": 480
+    },
+    {
+      "epoch": 0.41228439209087087,
+      "grad_norm": 0.4015921950340271,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 490
+    },
+    {
+      "epoch": 0.4206983592763988,
+      "grad_norm": 0.5489419102668762,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 500
+    },
+    {
+      "epoch": 0.4291123264619268,
+      "grad_norm": 0.4193589985370636,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 510
+    },
+    {
+      "epoch": 0.43752629364745477,
+      "grad_norm": 0.3418922424316406,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 520
+    },
+    {
+      "epoch": 0.44594026083298277,
+      "grad_norm": 0.32668185234069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 530
+    },
+    {
+      "epoch": 0.4543542280185107,
+      "grad_norm": 0.3094325661659241,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 540
+    },
+    {
+      "epoch": 0.4627681952040387,
+      "grad_norm": 0.3743017315864563,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 550
+    },
+    {
+      "epoch": 0.47118216238956667,
+      "grad_norm": 0.3295630216598511,
+      "learning_rate": 0.0002,
+      "loss": 1.8451,
+      "step": 560
+    },
+    {
+      "epoch": 0.4795961295750947,
+      "grad_norm": 1.6124513149261475,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 570
+    },
+    {
+      "epoch": 0.4880100967606226,
+      "grad_norm": 0.3245585858821869,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 580
+    },
+    {
+      "epoch": 0.49642406394615063,
+      "grad_norm": 0.3332934081554413,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 590
+    },
+    {
+      "epoch": 0.5048380311316786,
+      "grad_norm": 0.3836138844490051,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 600
+    },
+    {
+      "epoch": 0.5132519983172066,
+      "grad_norm": 0.32953888177871704,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 610
+    },
+    {
+      "epoch": 0.5216659655027346,
+      "grad_norm": 0.36291512846946716,
+      "learning_rate": 0.0002,
+      "loss": 1.7729,
+      "step": 620
+    },
+    {
+      "epoch": 0.5300799326882625,
+      "grad_norm": 0.3237783908843994,
+      "learning_rate": 0.0002,
+      "loss": 1.7758,
+      "step": 630
+    },
+    {
+      "epoch": 0.5384938998737905,
+      "grad_norm": 0.38882696628570557,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 640
+    },
+    {
+      "epoch": 0.5469078670593185,
+      "grad_norm": 0.37821972370147705,
+      "learning_rate": 0.0002,
+      "loss": 1.8624,
+      "step": 650
+    },
+    {
+      "epoch": 0.5553218342448465,
+      "grad_norm": 0.3556285500526428,
+      "learning_rate": 0.0002,
+      "loss": 1.8075,
+      "step": 660
+    },
+    {
+      "epoch": 0.5637358014303744,
+      "grad_norm": 0.347499281167984,
+      "learning_rate": 0.0002,
+      "loss": 1.778,
+      "step": 670
+    },
+    {
+      "epoch": 0.5721497686159024,
+      "grad_norm": 0.3176489472389221,
+      "learning_rate": 0.0002,
+      "loss": 1.8066,
+      "step": 680
+    },
+    {
+      "epoch": 0.5805637358014304,
+      "grad_norm": 0.30220088362693787,
+      "learning_rate": 0.0002,
+      "loss": 1.7257,
+      "step": 690
+    },
+    {
+      "epoch": 0.5889777029869584,
+      "grad_norm": 0.3711601793766022,
+      "learning_rate": 0.0002,
+      "loss": 1.8415,
+      "step": 700
+    },
+    {
+      "epoch": 0.5973916701724863,
+      "grad_norm": 0.3311759829521179,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 710
+    },
+    {
+      "epoch": 0.6058056373580143,
+      "grad_norm": 0.34824270009994507,
+      "learning_rate": 0.0002,
+      "loss": 1.7712,
+      "step": 720
+    },
+    {
+      "epoch": 0.6142196045435423,
+      "grad_norm": 0.29668381810188293,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 730
+    },
+    {
+      "epoch": 0.6226335717290703,
+      "grad_norm": 0.36087489128112793,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 740
+    },
+    {
+      "epoch": 0.6310475389145982,
+      "grad_norm": 0.31590089201927185,
+      "learning_rate": 0.0002,
+      "loss": 1.7956,
+      "step": 750
+    },
+    {
+      "epoch": 0.6394615061001262,
+      "grad_norm": 0.37632957100868225,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 760
+    },
+    {
+      "epoch": 0.6478754732856542,
+      "grad_norm": 0.3360748589038849,
+      "learning_rate": 0.0002,
+      "loss": 1.8499,
+      "step": 770
+    },
+    {
+      "epoch": 0.6562894404711822,
+      "grad_norm": 0.3420640528202057,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 780
+    },
+    {
+      "epoch": 0.6647034076567101,
+      "grad_norm": 0.5734959244728088,
+      "learning_rate": 0.0002,
+      "loss": 1.8353,
+      "step": 790
+    },
+    {
+      "epoch": 0.6731173748422381,
+      "grad_norm": 0.36440837383270264,
+      "learning_rate": 0.0002,
+      "loss": 1.7746,
+      "step": 800
+    },
+    {
+      "epoch": 0.6815313420277661,
+      "grad_norm": 0.3179708421230316,
+      "learning_rate": 0.0002,
+      "loss": 1.7532,
+      "step": 810
+    },
+    {
+      "epoch": 0.6899453092132941,
+      "grad_norm": 0.34122881293296814,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 820
+    },
+    {
+      "epoch": 0.698359276398822,
+      "grad_norm": 0.31886112689971924,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 830
+    },
+    {
+      "epoch": 0.70677324358435,
+      "grad_norm": 0.31782326102256775,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 840
+    },
+    {
+      "epoch": 0.715187210769878,
+      "grad_norm": 0.36052989959716797,
+      "learning_rate": 0.0002,
+      "loss": 1.7588,
+      "step": 850
+    },
+    {
+      "epoch": 0.723601177955406,
+      "grad_norm": 0.28946155309677124,
+      "learning_rate": 0.0002,
+      "loss": 1.7891,
+      "step": 860
+    },
+    {
+      "epoch": 0.7320151451409339,
+      "grad_norm": 0.3095663785934448,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 870
+    },
+    {
+      "epoch": 0.7404291123264619,
+      "grad_norm": 0.3317491412162781,
+      "learning_rate": 0.0002,
+      "loss": 1.785,
+      "step": 880
+    },
+    {
+      "epoch": 0.7488430795119899,
+      "grad_norm": 0.31324660778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 890
+    },
+    {
+      "epoch": 0.7572570466975179,
+      "grad_norm": 0.3290475606918335,
+      "learning_rate": 0.0002,
+      "loss": 1.8753,
+      "step": 900
+    },
+    {
+      "epoch": 0.7656710138830458,
+      "grad_norm": 0.35690343379974365,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 910
+    },
+    {
+      "epoch": 0.7740849810685738,
+      "grad_norm": 0.39558273553848267,
+      "learning_rate": 0.0002,
+      "loss": 1.826,
+      "step": 920
+    },
+    {
+      "epoch": 0.7824989482541018,
+      "grad_norm": 0.34254348278045654,
+      "learning_rate": 0.0002,
+      "loss": 1.8722,
+      "step": 930
+    },
+    {
+      "epoch": 0.7909129154396298,
+      "grad_norm": 0.3560165464878082,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 940
+    },
+    {
+      "epoch": 0.7993268826251577,
+      "grad_norm": 0.30693164467811584,
+      "learning_rate": 0.0002,
+      "loss": 1.7992,
+      "step": 950
+    },
+    {
+      "epoch": 0.8077408498106857,
+      "grad_norm": 0.3394823372364044,
+      "learning_rate": 0.0002,
+      "loss": 1.8029,
+      "step": 960
+    },
+    {
+      "epoch": 0.8161548169962137,
+      "grad_norm": 0.3741514980792999,
+      "learning_rate": 0.0002,
+      "loss": 1.8105,
+      "step": 970
+    },
+    {
+      "epoch": 0.8245687841817417,
+      "grad_norm": 0.3655228316783905,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 980
+    },
+    {
+      "epoch": 0.8329827513672696,
+      "grad_norm": 0.3586033880710602,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 990
+    },
+    {
+      "epoch": 0.8413967185527976,
+      "grad_norm": 0.3459678888320923,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8498106857383256,
+      "grad_norm": 0.3184349834918976,
+      "learning_rate": 0.0002,
+      "loss": 1.8498,
+      "step": 1010
+    },
+    {
+      "epoch": 0.8582246529238536,
+      "grad_norm": 0.3099786043167114,
+      "learning_rate": 0.0002,
+      "loss": 1.7632,
+      "step": 1020
+    },
+    {
+      "epoch": 0.8666386201093815,
+      "grad_norm": 0.30300915241241455,
+      "learning_rate": 0.0002,
+      "loss": 1.8067,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8750525872949095,
+      "grad_norm": 0.3128705620765686,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8834665544804375,
+      "grad_norm": 0.3336263597011566,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8918805216659655,
+      "grad_norm": 0.3801328241825104,
+      "learning_rate": 0.0002,
+      "loss": 1.8375,
+      "step": 1060
+    },
+    {
+      "epoch": 0.9002944888514934,
+      "grad_norm": 0.3122096359729767,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 1070
+    },
+    {
+      "epoch": 0.9087084560370214,
+      "grad_norm": 0.35990869998931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 1080
+    },
+    {
+      "epoch": 0.9171224232225494,
+      "grad_norm": 0.3321819305419922,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1090
+    },
+    {
+      "epoch": 0.9255363904080774,
+      "grad_norm": 0.4202139377593994,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9339503575936053,
+      "grad_norm": 0.32559722661972046,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 1110
+    },
+    {
+      "epoch": 0.9423643247791333,
+      "grad_norm": 0.3098459839820862,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 1120
+    },
+    {
+      "epoch": 0.9507782919646613,
+      "grad_norm": 0.33917108178138733,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1130
+    },
+    {
+      "epoch": 0.9591922591501894,
+      "grad_norm": 0.4055837094783783,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1140
+    },
+    {
+      "epoch": 0.9676062263357172,
+      "grad_norm": 0.32508623600006104,
+      "learning_rate": 0.0002,
+      "loss": 1.8259,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9760201935212452,
+      "grad_norm": 0.30150601267814636,
+      "learning_rate": 0.0002,
+      "loss": 1.782,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9844341607067733,
+      "grad_norm": 0.3042563199996948,
+      "learning_rate": 0.0002,
+      "loss": 1.8291,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9928481278923013,
+      "grad_norm": 0.33254584670066833,
+      "learning_rate": 0.0002,
+      "loss": 1.7847,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9995793016407236,
+      "eval_loss": 1.8077726364135742,
+      "eval_runtime": 38.4359,
+      "eval_samples_per_second": 13.399,
+      "eval_steps_per_second": 1.691,
+      "step": 1188
+    },
+    {
+      "epoch": 1.0012620950778293,
+      "grad_norm": 0.35073035955429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 1190
+    },
+    {
+      "epoch": 1.0096760622633572,
+      "grad_norm": 0.3217269778251648,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1200
+    },
+    {
+      "epoch": 1.018090029448885,
+      "grad_norm": 0.3635033369064331,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1210
+    },
+    {
+      "epoch": 1.0265039966344132,
+      "grad_norm": 0.32468414306640625,
+      "learning_rate": 0.0002,
+      "loss": 1.6949,
+      "step": 1220
+    },
+    {
+      "epoch": 1.034917963819941,
+      "grad_norm": 0.3307163417339325,
+      "learning_rate": 0.0002,
+      "loss": 1.711,
+      "step": 1230
+    },
+    {
+      "epoch": 1.0433319310054692,
+      "grad_norm": 0.34381359815597534,
+      "learning_rate": 0.0002,
+      "loss": 1.7881,
+      "step": 1240
+    },
+    {
+      "epoch": 1.051745898190997,
+      "grad_norm": 0.35874804854393005,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 1250
+    },
+    {
+      "epoch": 1.060159865376525,
+      "grad_norm": 0.3615919351577759,
+      "learning_rate": 0.0002,
+      "loss": 1.7314,
+      "step": 1260
+    },
+    {
+      "epoch": 1.068573832562053,
+      "grad_norm": 0.32835808396339417,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1270
+    },
+    {
+      "epoch": 1.076987799747581,
+      "grad_norm": 0.3876388370990753,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0854017669331089,
+      "grad_norm": 0.39895930886268616,
+      "learning_rate": 0.0002,
+      "loss": 1.7442,
+      "step": 1290
+    },
+    {
+      "epoch": 1.093815734118637,
+      "grad_norm": 0.39081698656082153,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 1300
+    },
+    {
+      "epoch": 1.1022297013041649,
+      "grad_norm": 0.39974215626716614,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1310
+    },
+    {
+      "epoch": 1.110643668489693,
+      "grad_norm": 0.3887332081794739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1320
+    },
+    {
+      "epoch": 1.1190576356752209,
+      "grad_norm": 0.36216408014297485,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 1330
+    },
+    {
+      "epoch": 1.1274716028607488,
+      "grad_norm": 0.36979028582572937,
+      "learning_rate": 0.0002,
+      "loss": 1.762,
+      "step": 1340
+    },
+    {
+      "epoch": 1.1358855700462769,
+      "grad_norm": 0.34052133560180664,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 1350
+    },
+    {
+      "epoch": 1.1442995372318048,
+      "grad_norm": 0.3467716574668884,
+      "learning_rate": 0.0002,
+      "loss": 1.7513,
+      "step": 1360
+    },
+    {
+      "epoch": 1.1527135044173327,
+      "grad_norm": 0.35528799891471863,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1370
+    },
+    {
+      "epoch": 1.1611274716028608,
+      "grad_norm": 0.36282262206077576,
+      "learning_rate": 0.0002,
+      "loss": 1.794,
+      "step": 1380
+    },
+    {
+      "epoch": 1.1695414387883887,
+      "grad_norm": 0.37355899810791016,
+      "learning_rate": 0.0002,
+      "loss": 1.7731,
+      "step": 1390
+    },
+    {
+      "epoch": 1.1779554059739168,
+      "grad_norm": 0.37292736768722534,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1400
+    },
+    {
+      "epoch": 1.1863693731594447,
+      "grad_norm": 0.5892812013626099,
+      "learning_rate": 0.0002,
+      "loss": 1.6916,
+      "step": 1410
+    },
+    {
+      "epoch": 1.1947833403449726,
+      "grad_norm": 0.3712292015552521,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 1420
+    },
+    {
+      "epoch": 1.2031973075305007,
+      "grad_norm": 0.3349577486515045,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1430
+    },
+    {
+      "epoch": 1.2116112747160286,
+      "grad_norm": 0.32591062784194946,
+      "learning_rate": 0.0002,
+      "loss": 1.7412,
+      "step": 1440
+    },
+    {
+      "epoch": 1.2200252419015567,
+      "grad_norm": 0.3840635418891907,
+      "learning_rate": 0.0002,
+      "loss": 1.7406,
+      "step": 1450
+    },
+    {
+      "epoch": 1.2284392090870846,
+      "grad_norm": 0.37238365411758423,
+      "learning_rate": 0.0002,
+      "loss": 1.7276,
+      "step": 1460
+    },
+    {
+      "epoch": 1.2368531762726125,
+      "grad_norm": 0.3731217682361603,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 1470
+    },
+    {
+      "epoch": 1.2452671434581406,
+      "grad_norm": 0.3318967819213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7255,
+      "step": 1480
+    },
+    {
+      "epoch": 1.2536811106436685,
+      "grad_norm": 0.3784034848213196,
+      "learning_rate": 0.0002,
+      "loss": 1.7463,
+      "step": 1490
+    },
+    {
+      "epoch": 1.2620950778291964,
+      "grad_norm": 0.3541383147239685,
+      "learning_rate": 0.0002,
+      "loss": 1.6862,
+      "step": 1500
+    },
+    {
+      "epoch": 1.2705090450147245,
+      "grad_norm": 0.35312485694885254,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1510
+    },
+    {
+      "epoch": 1.2789230122002524,
+      "grad_norm": 0.35272929072380066,
+      "learning_rate": 0.0002,
+      "loss": 1.7029,
+      "step": 1520
+    },
+    {
+      "epoch": 1.2873369793857803,
+      "grad_norm": 0.40988272428512573,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1530
+    },
+    {
+      "epoch": 1.2957509465713084,
+      "grad_norm": 0.3543946146965027,
+      "learning_rate": 0.0002,
+      "loss": 1.6912,
+      "step": 1540
+    },
+    {
+      "epoch": 1.3041649137568363,
+      "grad_norm": 0.35639145970344543,
+      "learning_rate": 0.0002,
+      "loss": 1.6757,
+      "step": 1550
+    },
+    {
+      "epoch": 1.3125788809423642,
+      "grad_norm": 0.3290826678276062,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 1560
+    },
+    {
+      "epoch": 1.3209928481278923,
+      "grad_norm": 0.39264336228370667,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 1570
+    },
+    {
+      "epoch": 1.3294068153134202,
+      "grad_norm": 0.5390415191650391,
+      "learning_rate": 0.0002,
+      "loss": 1.6804,
+      "step": 1580
+    },
+    {
+      "epoch": 1.3378207824989483,
+      "grad_norm": 0.5188116431236267,
+      "learning_rate": 0.0002,
+      "loss": 1.708,
+      "step": 1590
+    },
+    {
+      "epoch": 1.3462347496844762,
+      "grad_norm": 0.37445148825645447,
+      "learning_rate": 0.0002,
+      "loss": 1.6763,
+      "step": 1600
+    },
+    {
+      "epoch": 1.3546487168700043,
+      "grad_norm": 0.3296085298061371,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1610
+    },
+    {
+      "epoch": 1.3630626840555322,
+      "grad_norm": 0.39879581332206726,
+      "learning_rate": 0.0002,
+      "loss": 1.8107,
+      "step": 1620
+    },
+    {
+      "epoch": 1.37147665124106,
+      "grad_norm": 0.36092764139175415,
+      "learning_rate": 0.0002,
+      "loss": 1.6744,
+      "step": 1630
+    },
+    {
+      "epoch": 1.3798906184265882,
+      "grad_norm": 0.37011823058128357,
+      "learning_rate": 0.0002,
+      "loss": 1.7144,
+      "step": 1640
+    },
+    {
+      "epoch": 1.3883045856121161,
+      "grad_norm": 0.40863534808158875,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1650
+    },
+    {
+      "epoch": 1.396718552797644,
+      "grad_norm": 0.337001770734787,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 1660
+    },
+    {
+      "epoch": 1.4051325199831721,
+      "grad_norm": 0.35596707463264465,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1670
+    },
+    {
+      "epoch": 1.4135464871687,
+      "grad_norm": 0.3857671916484833,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 1680
+    },
+    {
+      "epoch": 1.421960454354228,
+      "grad_norm": 0.419502317905426,
+      "learning_rate": 0.0002,
+      "loss": 1.7015,
+      "step": 1690
+    },
+    {
+      "epoch": 1.430374421539756,
+      "grad_norm": 0.35459452867507935,
+      "learning_rate": 0.0002,
+      "loss": 1.7261,
+      "step": 1700
+    },
+    {
+      "epoch": 1.438788388725284,
+      "grad_norm": 0.37246978282928467,
+      "learning_rate": 0.0002,
+      "loss": 1.7361,
+      "step": 1710
+    },
+    {
+      "epoch": 1.4472023559108118,
+      "grad_norm": 0.33091893792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 1720
+    },
+    {
+      "epoch": 1.45561632309634,
+      "grad_norm": 0.37029674649238586,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1730
+    },
+    {
+      "epoch": 1.4640302902818678,
+      "grad_norm": 0.374025821685791,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1740
+    },
+    {
+      "epoch": 1.472444257467396,
+      "grad_norm": 0.3416315019130707,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1750
+    },
+    {
+      "epoch": 1.4808582246529238,
+      "grad_norm": 0.36502841114997864,
+      "learning_rate": 0.0002,
+      "loss": 1.7093,
+      "step": 1760
+    },
+    {
+      "epoch": 1.489272191838452,
+      "grad_norm": 0.35458803176879883,
+      "learning_rate": 0.0002,
+      "loss": 1.6597,
+      "step": 1770
+    },
+    {
+      "epoch": 1.4976861590239798,
+      "grad_norm": 0.4462839663028717,
+      "learning_rate": 0.0002,
+      "loss": 1.675,
+      "step": 1780
+    },
+    {
+      "epoch": 1.5061001262095077,
+      "grad_norm": 0.34836092591285706,
+      "learning_rate": 0.0002,
+      "loss": 1.7267,
+      "step": 1790
+    },
+    {
+      "epoch": 1.5145140933950358,
+      "grad_norm": 0.3445749282836914,
+      "learning_rate": 0.0002,
+      "loss": 1.7295,
+      "step": 1800
+    },
+    {
+      "epoch": 1.5229280605805637,
+      "grad_norm": 0.36012160778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1810
+    },
+    {
+      "epoch": 1.5313420277660916,
+      "grad_norm": 0.4052616059780121,
+      "learning_rate": 0.0002,
+      "loss": 1.6594,
+      "step": 1820
+    },
+    {
+      "epoch": 1.5397559949516197,
+      "grad_norm": 0.3966905474662781,
+      "learning_rate": 0.0002,
+      "loss": 1.72,
+      "step": 1830
+    },
+    {
+      "epoch": 1.5481699621371476,
+      "grad_norm": 0.35028719902038574,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1840
+    },
+    {
+      "epoch": 1.5565839293226755,
+      "grad_norm": 0.3936742842197418,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1850
+    },
+    {
+      "epoch": 1.5649978965082036,
+      "grad_norm": 0.34473296999931335,
+      "learning_rate": 0.0002,
+      "loss": 1.7579,
+      "step": 1860
+    },
+    {
+      "epoch": 1.5734118636937318,
+      "grad_norm": 0.4328365623950958,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1870
+    },
+    {
+      "epoch": 1.5818258308792594,
+      "grad_norm": 0.3566315472126007,
+      "learning_rate": 0.0002,
+      "loss": 1.7098,
+      "step": 1880
+    },
+    {
+      "epoch": 1.5902397980647875,
+      "grad_norm": 0.3301256597042084,
+      "learning_rate": 0.0002,
+      "loss": 1.6095,
+      "step": 1890
+    },
+    {
+      "epoch": 1.5986537652503157,
+      "grad_norm": 0.3743041455745697,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 1900
+    },
+    {
+      "epoch": 1.6070677324358436,
+      "grad_norm": 0.3735344707965851,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 1910
+    },
+    {
+      "epoch": 1.6154816996213714,
+      "grad_norm": 0.42191144824028015,
+      "learning_rate": 0.0002,
+      "loss": 1.7445,
+      "step": 1920
+    },
+    {
+      "epoch": 1.6238956668068996,
+      "grad_norm": 0.3787207305431366,
+      "learning_rate": 0.0002,
+      "loss": 1.6978,
+      "step": 1930
+    },
+    {
+      "epoch": 1.6323096339924275,
+      "grad_norm": 0.35647350549697876,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 1940
+    },
+    {
+      "epoch": 1.6407236011779553,
+      "grad_norm": 0.39791446924209595,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 1950
+    },
+    {
+      "epoch": 1.6491375683634835,
+      "grad_norm": 0.37341275811195374,
+      "learning_rate": 0.0002,
+      "loss": 1.7293,
+      "step": 1960
+    },
+    {
+      "epoch": 1.6575515355490114,
+      "grad_norm": 0.3722686469554901,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1970
+    },
+    {
+      "epoch": 1.6659655027345392,
+      "grad_norm": 0.37467387318611145,
+      "learning_rate": 0.0002,
+      "loss": 1.6383,
+      "step": 1980
+    },
+    {
+      "epoch": 1.6743794699200674,
+      "grad_norm": 0.37109461426734924,
+      "learning_rate": 0.0002,
+      "loss": 1.7439,
+      "step": 1990
+    },
+    {
+      "epoch": 1.6827934371055953,
+      "grad_norm": 0.4008837044239044,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 2000
+    },
+    {
+      "epoch": 1.6912074042911232,
+      "grad_norm": 0.3316999673843384,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2010
+    },
+    {
+      "epoch": 1.6996213714766513,
+      "grad_norm": 0.3683805465698242,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 2020
+    },
+    {
+      "epoch": 1.7080353386621794,
+      "grad_norm": 0.4163658320903778,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 2030
+    },
+    {
+      "epoch": 1.716449305847707,
+      "grad_norm": 0.4245431125164032,
+      "learning_rate": 0.0002,
+      "loss": 1.741,
+      "step": 2040
+    },
+    {
+      "epoch": 1.7248632730332352,
+      "grad_norm": 0.36732038855552673,
+      "learning_rate": 0.0002,
+      "loss": 1.7184,
+      "step": 2050
+    },
+    {
+      "epoch": 1.7332772402187633,
+      "grad_norm": 0.34981656074523926,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 2060
+    },
+    {
+      "epoch": 1.7416912074042912,
+      "grad_norm": 0.38588812947273254,
+      "learning_rate": 0.0002,
+      "loss": 1.7545,
+      "step": 2070
+    },
+    {
+      "epoch": 1.750105174589819,
+      "grad_norm": 0.39914557337760925,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 2080
+    },
+    {
+      "epoch": 1.7585191417753472,
+      "grad_norm": 0.36068692803382874,
+      "learning_rate": 0.0002,
+      "loss": 1.7049,
+      "step": 2090
+    },
+    {
+      "epoch": 1.766933108960875,
+      "grad_norm": 0.3983287215232849,
+      "learning_rate": 0.0002,
+      "loss": 1.7537,
+      "step": 2100
+    },
+    {
+      "epoch": 1.775347076146403,
+      "grad_norm": 0.45008400082588196,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2110
+    },
+    {
+      "epoch": 1.783761043331931,
+      "grad_norm": 0.3618052303791046,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 2120
+    },
+    {
+      "epoch": 1.792175010517459,
+      "grad_norm": 0.38745400309562683,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 2130
+    },
+    {
+      "epoch": 1.8005889777029869,
+      "grad_norm": 0.3413826525211334,
+      "learning_rate": 0.0002,
+      "loss": 1.7387,
+      "step": 2140
+    },
+    {
+      "epoch": 1.809002944888515,
+      "grad_norm": 0.35983747243881226,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 2150
+    },
+    {
+      "epoch": 1.8174169120740429,
+      "grad_norm": 0.40926849842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 2160
+    },
+    {
+      "epoch": 1.8258308792595708,
+      "grad_norm": 0.3543093800544739,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2170
+    },
+    {
+      "epoch": 1.8342448464450989,
+      "grad_norm": 0.42690935730934143,
+      "learning_rate": 0.0002,
+      "loss": 1.7812,
+      "step": 2180
+    },
+    {
+      "epoch": 1.842658813630627,
+      "grad_norm": 0.40282756090164185,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 2190
+    },
+    {
+      "epoch": 1.8510727808161547,
+      "grad_norm": 0.36568400263786316,
+      "learning_rate": 0.0002,
+      "loss": 1.7411,
+      "step": 2200
+    },
+    {
+      "epoch": 1.8594867480016828,
+      "grad_norm": 0.43159013986587524,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 2210
+    },
+    {
+      "epoch": 1.867900715187211,
+      "grad_norm": 0.3554118573665619,
+      "learning_rate": 0.0002,
+      "loss": 1.7298,
+      "step": 2220
+    },
+    {
+      "epoch": 1.8763146823727388,
+      "grad_norm": 0.43349072337150574,
+      "learning_rate": 0.0002,
+      "loss": 1.7157,
+      "step": 2230
+    },
+    {
+      "epoch": 1.8847286495582667,
+      "grad_norm": 0.36486536264419556,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 2240
+    },
+    {
+      "epoch": 1.8931426167437948,
+      "grad_norm": 0.39260047674179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6901,
+      "step": 2250
+    },
+    {
+      "epoch": 1.9015565839293227,
+      "grad_norm": 0.3741776943206787,
+      "learning_rate": 0.0002,
+      "loss": 1.6691,
+      "step": 2260
+    },
+    {
+      "epoch": 1.9099705511148506,
+      "grad_norm": 0.3961946964263916,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2270
+    },
+    {
+      "epoch": 1.9183845183003787,
+      "grad_norm": 0.3659731149673462,
+      "learning_rate": 0.0002,
+      "loss": 1.737,
+      "step": 2280
+    },
+    {
+      "epoch": 1.9267984854859066,
+      "grad_norm": 0.34744107723236084,
+      "learning_rate": 0.0002,
+      "loss": 1.7342,
+      "step": 2290
+    },
+    {
+      "epoch": 1.9352124526714345,
+      "grad_norm": 0.3607442378997803,
+      "learning_rate": 0.0002,
+      "loss": 1.7162,
+      "step": 2300
+    },
+    {
+      "epoch": 1.9436264198569626,
+      "grad_norm": 0.331464558839798,
+      "learning_rate": 0.0002,
+      "loss": 1.6673,
+      "step": 2310
+    },
+    {
+      "epoch": 1.9520403870424905,
+      "grad_norm": 0.3904414474964142,
+      "learning_rate": 0.0002,
+      "loss": 1.7101,
+      "step": 2320
+    },
+    {
+      "epoch": 1.9604543542280184,
+      "grad_norm": 0.37584832310676575,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 2330
+    },
+    {
+      "epoch": 1.9688683214135465,
+      "grad_norm": 0.3698684275150299,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 2340
+    },
+    {
+      "epoch": 1.9772822885990746,
+      "grad_norm": 0.40571412444114685,
+      "learning_rate": 0.0002,
+      "loss": 1.7764,
+      "step": 2350
+    },
+    {
+      "epoch": 1.9856962557846023,
+      "grad_norm": 0.40059587359428406,
+      "learning_rate": 0.0002,
+      "loss": 1.744,
+      "step": 2360
+    },
+    {
+      "epoch": 1.9941102229701304,
+      "grad_norm": 0.4168248474597931,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2370
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8055059909820557,
+      "eval_runtime": 38.422,
+      "eval_samples_per_second": 13.404,
+      "eval_steps_per_second": 1.692,
+      "step": 2377
+    },
+    {
+      "epoch": 2.0025241901556585,
+      "grad_norm": 0.35205352306365967,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 2380
+    },
+    {
+      "epoch": 2.010938157341186,
+      "grad_norm": 0.3979377746582031,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 2390
+    },
+    {
+      "epoch": 2.0193521245267143,
+      "grad_norm": 0.396491676568985,
+      "learning_rate": 0.0002,
+      "loss": 1.6421,
+      "step": 2400
+    },
+    {
+      "epoch": 2.0277660917122424,
+      "grad_norm": 0.44712209701538086,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2410
+    },
+    {
+      "epoch": 2.03618005889777,
+      "grad_norm": 0.4454420208930969,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 2420
+    },
+    {
+      "epoch": 2.044594026083298,
+      "grad_norm": 0.4170038402080536,
+      "learning_rate": 0.0002,
+      "loss": 1.6635,
+      "step": 2430
+    },
+    {
+      "epoch": 2.0530079932688263,
+      "grad_norm": 0.4309595227241516,
+      "learning_rate": 0.0002,
+      "loss": 1.6512,
+      "step": 2440
+    },
+    {
+      "epoch": 2.0614219604543544,
+      "grad_norm": 0.4241602122783661,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 2450
+    },
+    {
+      "epoch": 2.069835927639882,
+      "grad_norm": 0.4370540678501129,
+      "learning_rate": 0.0002,
+      "loss": 1.6162,
+      "step": 2460
+    },
+    {
+      "epoch": 2.0782498948254102,
+      "grad_norm": 0.43985554575920105,
+      "learning_rate": 0.0002,
+      "loss": 1.6354,
+      "step": 2470
+    },
+    {
+      "epoch": 2.0866638620109383,
+      "grad_norm": 0.4158105254173279,
+      "learning_rate": 0.0002,
+      "loss": 1.6954,
+      "step": 2480
+    },
+    {
+      "epoch": 2.095077829196466,
+      "grad_norm": 0.441549152135849,
+      "learning_rate": 0.0002,
+      "loss": 1.6114,
+      "step": 2490
+    },
+    {
+      "epoch": 2.103491796381994,
+      "grad_norm": 0.385718435049057,
+      "learning_rate": 0.0002,
+      "loss": 1.5485,
+      "step": 2500
+    },
+    {
+      "epoch": 2.1119057635675222,
+      "grad_norm": 0.43146514892578125,
+      "learning_rate": 0.0002,
+      "loss": 1.5894,
+      "step": 2510
+    },
+    {
+      "epoch": 2.12031973075305,
+      "grad_norm": 0.41663315892219543,
+      "learning_rate": 0.0002,
+      "loss": 1.6414,
+      "step": 2520
+    },
+    {
+      "epoch": 2.128733697938578,
+      "grad_norm": 0.4410698115825653,
+      "learning_rate": 0.0002,
+      "loss": 1.6527,
+      "step": 2530
+    },
+    {
+      "epoch": 2.137147665124106,
+      "grad_norm": 0.4472278952598572,
+      "learning_rate": 0.0002,
+      "loss": 1.6124,
+      "step": 2540
+    },
+    {
+      "epoch": 2.145561632309634,
+      "grad_norm": 0.3879167437553406,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2550
+    },
+    {
+      "epoch": 2.153975599495162,
+      "grad_norm": 0.4212203025817871,
+      "learning_rate": 0.0002,
+      "loss": 1.6682,
+      "step": 2560
+    },
+    {
+      "epoch": 2.16238956668069,
+      "grad_norm": 0.42841723561286926,
+      "learning_rate": 0.0002,
+      "loss": 1.6036,
+      "step": 2570
+    },
+    {
+      "epoch": 2.1708035338662177,
+      "grad_norm": 0.39272481203079224,
+      "learning_rate": 0.0002,
+      "loss": 1.5962,
+      "step": 2580
+    },
+    {
+      "epoch": 2.179217501051746,
+      "grad_norm": 0.4075261354446411,
+      "learning_rate": 0.0002,
+      "loss": 1.681,
+      "step": 2590
+    },
+    {
+      "epoch": 2.187631468237274,
+      "grad_norm": 0.5358437895774841,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 2600
+    },
+    {
+      "epoch": 2.1960454354228016,
+      "grad_norm": 0.4738350212574005,
+      "learning_rate": 0.0002,
+      "loss": 1.6423,
+      "step": 2610
+    },
+    {
+      "epoch": 2.2044594026083297,
+      "grad_norm": 0.446789026260376,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 2620
+    },
+    {
+      "epoch": 2.212873369793858,
+      "grad_norm": 0.4615374505519867,
+      "learning_rate": 0.0002,
+      "loss": 1.6246,
+      "step": 2630
+    },
+    {
+      "epoch": 2.221287336979386,
+      "grad_norm": 0.46901994943618774,
+      "learning_rate": 0.0002,
+      "loss": 1.6205,
+      "step": 2640
+    },
+    {
+      "epoch": 2.2297013041649136,
+      "grad_norm": 0.46267789602279663,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 2.2381152713504417,
+      "grad_norm": 0.4383080005645752,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2660
+    },
+    {
+      "epoch": 2.24652923853597,
+      "grad_norm": 0.4070609509944916,
+      "learning_rate": 0.0002,
+      "loss": 1.5745,
+      "step": 2670
+    },
+    {
+      "epoch": 2.2549432057214975,
+      "grad_norm": 0.4572339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.6125,
+      "step": 2680
+    },
+    {
+      "epoch": 2.2633571729070256,
+      "grad_norm": 0.393265038728714,
+      "learning_rate": 0.0002,
+      "loss": 1.5671,
+      "step": 2690
+    },
+    {
+      "epoch": 2.2717711400925538,
+      "grad_norm": 0.46144717931747437,
+      "learning_rate": 0.0002,
+      "loss": 1.6239,
+      "step": 2700
+    },
+    {
+      "epoch": 2.2801851072780814,
+      "grad_norm": 0.45077767968177795,
+      "learning_rate": 0.0002,
+      "loss": 1.5992,
+      "step": 2710
+    },
+    {
+      "epoch": 2.2885990744636096,
+      "grad_norm": 0.5697639584541321,
+      "learning_rate": 0.0002,
+      "loss": 1.6261,
+      "step": 2720
+    },
+    {
+      "epoch": 2.2970130416491377,
+      "grad_norm": 0.4855510890483856,
+      "learning_rate": 0.0002,
+      "loss": 1.6192,
+      "step": 2730
+    },
+    {
+      "epoch": 2.3054270088346653,
+      "grad_norm": 0.4440622627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 2740
+    },
+    {
+      "epoch": 2.3138409760201935,
+      "grad_norm": 0.3904096782207489,
+      "learning_rate": 0.0002,
+      "loss": 1.6496,
+      "step": 2750
+    },
+    {
+      "epoch": 2.3222549432057216,
+      "grad_norm": 0.5225510597229004,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 2760
+    },
+    {
+      "epoch": 2.3306689103912497,
+      "grad_norm": 0.44866397976875305,
+      "learning_rate": 0.0002,
+      "loss": 1.6082,
+      "step": 2770
+    },
+    {
+      "epoch": 2.3390828775767774,
+      "grad_norm": 0.5167056322097778,
+      "learning_rate": 0.0002,
+      "loss": 1.6087,
+      "step": 2780
+    },
+    {
+      "epoch": 2.3474968447623055,
+      "grad_norm": 0.45913267135620117,
+      "learning_rate": 0.0002,
+      "loss": 1.6136,
+      "step": 2790
+    },
+    {
+      "epoch": 2.3559108119478336,
+      "grad_norm": 0.45787590742111206,
+      "learning_rate": 0.0002,
+      "loss": 1.6564,
+      "step": 2800
+    },
+    {
+      "epoch": 2.3643247791333613,
+      "grad_norm": 0.4633352756500244,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2810
+    },
+    {
+      "epoch": 2.3727387463188894,
+      "grad_norm": 0.46390071511268616,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 2820
+    },
+    {
+      "epoch": 2.3811527135044175,
+      "grad_norm": 0.4261005222797394,
+      "learning_rate": 0.0002,
+      "loss": 1.6039,
+      "step": 2830
+    },
+    {
+      "epoch": 2.389566680689945,
+      "grad_norm": 0.4283634424209595,
+      "learning_rate": 0.0002,
+      "loss": 1.6364,
+      "step": 2840
+    },
+    {
+      "epoch": 2.3979806478754733,
+      "grad_norm": 0.4955291450023651,
+      "learning_rate": 0.0002,
+      "loss": 1.6382,
+      "step": 2850
+    },
+    {
+      "epoch": 2.4063946150610014,
+      "grad_norm": 0.4740189015865326,
+      "learning_rate": 0.0002,
+      "loss": 1.6173,
+      "step": 2860
+    },
+    {
+      "epoch": 2.414808582246529,
+      "grad_norm": 0.4222276508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2870
+    },
+    {
+      "epoch": 2.423222549432057,
+      "grad_norm": 0.4982149004936218,
+      "learning_rate": 0.0002,
+      "loss": 1.5602,
+      "step": 2880
+    },
+    {
+      "epoch": 2.4316365166175853,
+      "grad_norm": 0.5217409133911133,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 2890
+    },
+    {
+      "epoch": 2.4400504838031134,
+      "grad_norm": 0.4555884897708893,
+      "learning_rate": 0.0002,
+      "loss": 1.5804,
+      "step": 2900
+    },
+    {
+      "epoch": 2.448464450988641,
+      "grad_norm": 0.43178579211235046,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2910
+    },
+    {
+      "epoch": 2.456878418174169,
+      "grad_norm": 0.4788478910923004,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2920
+    },
+    {
+      "epoch": 2.465292385359697,
+      "grad_norm": 0.43689873814582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 2930
+    },
+    {
+      "epoch": 2.473706352545225,
+      "grad_norm": 0.5115197896957397,
+      "learning_rate": 0.0002,
+      "loss": 1.6196,
+      "step": 2940
+    },
+    {
+      "epoch": 2.482120319730753,
+      "grad_norm": 0.5290159583091736,
+      "learning_rate": 0.0002,
+      "loss": 1.689,
+      "step": 2950
+    },
+    {
+      "epoch": 2.490534286916281,
+      "grad_norm": 0.46042463183403015,
+      "learning_rate": 0.0002,
+      "loss": 1.6499,
+      "step": 2960
+    },
+    {
+      "epoch": 2.498948254101809,
+      "grad_norm": 0.4359915852546692,
+      "learning_rate": 0.0002,
+      "loss": 1.6664,
+      "step": 2970
+    },
+    {
+      "epoch": 2.507362221287337,
+      "grad_norm": 0.46352964639663696,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 2980
+    },
+    {
+      "epoch": 2.515776188472865,
+      "grad_norm": 0.5324268341064453,
+      "learning_rate": 0.0002,
+      "loss": 1.6501,
+      "step": 2990
+    },
+    {
+      "epoch": 2.5241901556583928,
+      "grad_norm": 0.5929607152938843,
+      "learning_rate": 0.0002,
+      "loss": 1.6115,
+      "step": 3000
+    },
+    {
+      "epoch": 2.532604122843921,
+      "grad_norm": 0.4811333417892456,
+      "learning_rate": 0.0002,
+      "loss": 1.6772,
+      "step": 3010
+    },
+    {
+      "epoch": 2.541018090029449,
+      "grad_norm": 0.4662701487541199,
+      "learning_rate": 0.0002,
+      "loss": 1.7023,
+      "step": 3020
+    },
+    {
+      "epoch": 2.549432057214977,
+      "grad_norm": 0.4582270681858063,
+      "learning_rate": 0.0002,
+      "loss": 1.5426,
+      "step": 3030
+    },
+    {
+      "epoch": 2.557846024400505,
+      "grad_norm": 0.4679982662200928,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 3040
+    },
+    {
+      "epoch": 2.566259991586033,
+      "grad_norm": 0.4380294680595398,
+      "learning_rate": 0.0002,
+      "loss": 1.5442,
+      "step": 3050
+    },
+    {
+      "epoch": 2.5746739587715606,
+      "grad_norm": 0.44295763969421387,
+      "learning_rate": 0.0002,
+      "loss": 1.6055,
+      "step": 3060
+    },
+    {
+      "epoch": 2.5830879259570887,
+      "grad_norm": 0.5131027698516846,
+      "learning_rate": 0.0002,
+      "loss": 1.5775,
+      "step": 3070
+    },
+    {
+      "epoch": 2.591501893142617,
+      "grad_norm": 0.47567516565322876,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 3080
+    },
+    {
+      "epoch": 2.599915860328145,
+      "grad_norm": 0.49002596735954285,
+      "learning_rate": 0.0002,
+      "loss": 1.5671,
+      "step": 3090
+    },
+    {
+      "epoch": 2.6083298275136726,
+      "grad_norm": 0.44856327772140503,
+      "learning_rate": 0.0002,
+      "loss": 1.5445,
+      "step": 3100
+    },
+    {
+      "epoch": 2.6167437946992007,
+      "grad_norm": 0.4480142593383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5797,
+      "step": 3110
+    },
+    {
+      "epoch": 2.6251577618847284,
+      "grad_norm": 0.4317494034767151,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 3120
+    },
+    {
+      "epoch": 2.6335717290702565,
+      "grad_norm": 0.42580848932266235,
+      "learning_rate": 0.0002,
+      "loss": 1.6321,
+      "step": 3130
+    },
+    {
+      "epoch": 2.6419856962557846,
+      "grad_norm": 0.4516814947128296,
+      "learning_rate": 0.0002,
+      "loss": 1.6483,
+      "step": 3140
+    },
+    {
+      "epoch": 2.6503996634413127,
+      "grad_norm": 0.4438435733318329,
+      "learning_rate": 0.0002,
+      "loss": 1.695,
+      "step": 3150
+    },
+    {
+      "epoch": 2.6588136306268404,
+      "grad_norm": 0.4385356307029724,
+      "learning_rate": 0.0002,
+      "loss": 1.6938,
+      "step": 3160
+    },
+    {
+      "epoch": 2.6672275978123685,
+      "grad_norm": 0.5064112544059753,
+      "learning_rate": 0.0002,
+      "loss": 1.6139,
+      "step": 3170
+    },
+    {
+      "epoch": 2.6756415649978966,
+      "grad_norm": 0.49163177609443665,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 3180
+    },
+    {
+      "epoch": 2.6840555321834243,
+      "grad_norm": 0.49339258670806885,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 3190
+    },
+    {
+      "epoch": 2.6924694993689524,
+      "grad_norm": 0.440950870513916,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 3200
+    },
+    {
+      "epoch": 2.7008834665544805,
+      "grad_norm": 0.4283970594406128,
+      "learning_rate": 0.0002,
+      "loss": 1.6305,
+      "step": 3210
+    },
+    {
+      "epoch": 2.7092974337400086,
+      "grad_norm": 0.43875712156295776,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 3220
+    },
+    {
+      "epoch": 2.7177114009255363,
+      "grad_norm": 0.49332964420318604,
+      "learning_rate": 0.0002,
+      "loss": 1.6129,
+      "step": 3230
+    },
+    {
+      "epoch": 2.7261253681110644,
+      "grad_norm": 0.5225692391395569,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 3240
+    },
+    {
+      "epoch": 2.734539335296592,
+      "grad_norm": 0.4856489300727844,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 3250
+    },
+    {
+      "epoch": 2.74295330248212,
+      "grad_norm": 0.46918296813964844,
+      "learning_rate": 0.0002,
+      "loss": 1.6463,
+      "step": 3260
+    },
+    {
+      "epoch": 2.7513672696676483,
+      "grad_norm": 0.4802931249141693,
+      "learning_rate": 0.0002,
+      "loss": 1.6819,
+      "step": 3270
+    },
+    {
+      "epoch": 2.7597812368531764,
+      "grad_norm": 0.4485355615615845,
+      "learning_rate": 0.0002,
+      "loss": 1.6246,
+      "step": 3280
+    },
+    {
+      "epoch": 2.768195204038704,
+      "grad_norm": 0.43944594264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 3290
+    },
+    {
+      "epoch": 2.7766091712242322,
+      "grad_norm": 0.46847742795944214,
+      "learning_rate": 0.0002,
+      "loss": 1.6501,
+      "step": 3300
+    },
+    {
+      "epoch": 2.7850231384097603,
+      "grad_norm": 0.4816027879714966,
+      "learning_rate": 0.0002,
+      "loss": 1.5969,
+      "step": 3310
+    },
+    {
+      "epoch": 2.793437105595288,
+      "grad_norm": 0.453960120677948,
+      "learning_rate": 0.0002,
+      "loss": 1.6293,
+      "step": 3320
+    },
+    {
+      "epoch": 2.801851072780816,
+      "grad_norm": 0.4816017150878906,
+      "learning_rate": 0.0002,
+      "loss": 1.6429,
+      "step": 3330
+    },
+    {
+      "epoch": 2.8102650399663442,
+      "grad_norm": 0.4461034834384918,
+      "learning_rate": 0.0002,
+      "loss": 1.6683,
+      "step": 3340
+    },
+    {
+      "epoch": 2.8186790071518724,
+      "grad_norm": 0.48821821808815,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 3350
+    },
+    {
+      "epoch": 2.8270929743374,
+      "grad_norm": 0.4574853777885437,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 3360
+    },
+    {
+      "epoch": 2.835506941522928,
+      "grad_norm": 0.42062026262283325,
+      "learning_rate": 0.0002,
+      "loss": 1.6651,
+      "step": 3370
+    },
+    {
+      "epoch": 2.843920908708456,
+      "grad_norm": 0.4499834477901459,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 3380
+    },
+    {
+      "epoch": 2.852334875893984,
+      "grad_norm": 0.4780360758304596,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 3390
+    },
+    {
+      "epoch": 2.860748843079512,
+      "grad_norm": 0.45422887802124023,
+      "learning_rate": 0.0002,
+      "loss": 1.5882,
+      "step": 3400
+    },
+    {
+      "epoch": 2.86916281026504,
+      "grad_norm": 0.4590015709400177,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 3410
+    },
+    {
+      "epoch": 2.877576777450568,
+      "grad_norm": 0.45689624547958374,
+      "learning_rate": 0.0002,
+      "loss": 1.6746,
+      "step": 3420
+    },
+    {
+      "epoch": 2.885990744636096,
+      "grad_norm": 0.46953922510147095,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3430
+    },
+    {
+      "epoch": 2.8944047118216236,
+      "grad_norm": 0.4791966378688812,
+      "learning_rate": 0.0002,
+      "loss": 1.6015,
+      "step": 3440
+    },
+    {
+      "epoch": 2.9028186790071517,
+      "grad_norm": 0.4842296242713928,
+      "learning_rate": 0.0002,
+      "loss": 1.694,
+      "step": 3450
+    },
+    {
+      "epoch": 2.91123264619268,
+      "grad_norm": 0.47219768166542053,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3460
+    },
+    {
+      "epoch": 2.919646613378208,
+      "grad_norm": 0.4622127115726471,
+      "learning_rate": 0.0002,
+      "loss": 1.6486,
+      "step": 3470
+    },
+    {
+      "epoch": 2.9280605805637356,
+      "grad_norm": 0.46832820773124695,
+      "learning_rate": 0.0002,
+      "loss": 1.6485,
+      "step": 3480
+    },
+    {
+      "epoch": 2.9364745477492638,
+      "grad_norm": 0.44582483172416687,
+      "learning_rate": 0.0002,
+      "loss": 1.6366,
+      "step": 3490
+    },
+    {
+      "epoch": 2.944888514934792,
+      "grad_norm": 0.4987219274044037,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 3500
+    },
+    {
+      "epoch": 2.9533024821203195,
+      "grad_norm": 0.43750956654548645,
+      "learning_rate": 0.0002,
+      "loss": 1.5991,
+      "step": 3510
+    },
+    {
+      "epoch": 2.9617164493058477,
+      "grad_norm": 0.49962925910949707,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 3520
+    },
+    {
+      "epoch": 2.9701304164913758,
+      "grad_norm": 0.5189590454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 3530
+    },
+    {
+      "epoch": 2.978544383676904,
+      "grad_norm": 0.391317754983902,
+      "learning_rate": 0.0002,
+      "loss": 1.6688,
+      "step": 3540
+    },
+    {
+      "epoch": 2.9869583508624316,
+      "grad_norm": 0.44934695959091187,
+      "learning_rate": 0.0002,
+      "loss": 1.5884,
+      "step": 3550
+    },
+    {
+      "epoch": 2.9953723180479597,
+      "grad_norm": 0.4740142226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.5688,
+      "step": 3560
+    },
+    {
+      "epoch": 2.9995793016407237,
+      "eval_loss": 1.8266887664794922,
+      "eval_runtime": 37.9445,
+      "eval_samples_per_second": 13.572,
+      "eval_steps_per_second": 1.713,
+      "step": 3565
+    },
+    {
+      "epoch": 3.003786285233488,
+      "grad_norm": 0.4523724615573883,
+      "learning_rate": 0.0002,
+      "loss": 1.5939,
+      "step": 3570
+    },
+    {
+      "epoch": 3.0122002524190155,
+      "grad_norm": 0.5261380076408386,
+      "learning_rate": 0.0002,
+      "loss": 1.526,
+      "step": 3580
+    },
+    {
+      "epoch": 3.0206142196045436,
+      "grad_norm": 0.48664888739585876,
+      "learning_rate": 0.0002,
+      "loss": 1.4946,
+      "step": 3590
+    },
+    {
+      "epoch": 3.0290281867900717,
+      "grad_norm": 0.5070882439613342,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 3600
+    },
+    {
+      "epoch": 3.0374421539755994,
+      "grad_norm": 0.5816011428833008,
+      "learning_rate": 0.0002,
+      "loss": 1.5316,
+      "step": 3610
+    },
+    {
+      "epoch": 3.0458561211611275,
+      "grad_norm": 0.6610211730003357,
+      "learning_rate": 0.0002,
+      "loss": 1.5682,
+      "step": 3620
+    },
+    {
+      "epoch": 3.0542700883466556,
+      "grad_norm": 0.5257703065872192,
+      "learning_rate": 0.0002,
+      "loss": 1.5699,
+      "step": 3630
+    },
+    {
+      "epoch": 3.0626840555321833,
+      "grad_norm": 0.5574390888214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4438,
+      "step": 3640
+    },
+    {
+      "epoch": 3.0710980227177114,
+      "grad_norm": 0.5682297348976135,
+      "learning_rate": 0.0002,
+      "loss": 1.547,
+      "step": 3650
+    },
+    {
+      "epoch": 3.0795119899032395,
+      "grad_norm": 0.5798383355140686,
+      "learning_rate": 0.0002,
+      "loss": 1.5743,
+      "step": 3660
+    },
+    {
+      "epoch": 3.087925957088767,
+      "grad_norm": 0.5458289980888367,
+      "learning_rate": 0.0002,
+      "loss": 1.4339,
+      "step": 3670
+    },
+    {
+      "epoch": 3.0963399242742953,
+      "grad_norm": 0.5599102973937988,
+      "learning_rate": 0.0002,
+      "loss": 1.46,
+      "step": 3680
+    },
+    {
+      "epoch": 3.1047538914598234,
+      "grad_norm": 0.5023021697998047,
+      "learning_rate": 0.0002,
+      "loss": 1.4589,
+      "step": 3690
+    },
+    {
+      "epoch": 3.113167858645351,
+      "grad_norm": 0.5448206067085266,
+      "learning_rate": 0.0002,
+      "loss": 1.5114,
+      "step": 3700
+    },
+    {
+      "epoch": 3.121581825830879,
+      "grad_norm": 0.5760458707809448,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 3710
+    },
+    {
+      "epoch": 3.1299957930164073,
+      "grad_norm": 0.6018968224525452,
+      "learning_rate": 0.0002,
+      "loss": 1.4789,
+      "step": 3720
+    },
+    {
+      "epoch": 3.1384097602019354,
+      "grad_norm": 0.5767101049423218,
+      "learning_rate": 0.0002,
+      "loss": 1.5518,
+      "step": 3730
+    },
+    {
+      "epoch": 3.146823727387463,
+      "grad_norm": 0.5333963632583618,
+      "learning_rate": 0.0002,
+      "loss": 1.5032,
+      "step": 3740
+    },
+    {
+      "epoch": 3.155237694572991,
+      "grad_norm": 0.5918396711349487,
+      "learning_rate": 0.0002,
+      "loss": 1.4812,
+      "step": 3750
+    },
+    {
+      "epoch": 3.1636516617585193,
+      "grad_norm": 0.5931203365325928,
+      "learning_rate": 0.0002,
+      "loss": 1.4618,
+      "step": 3760
+    },
+    {
+      "epoch": 3.172065628944047,
+      "grad_norm": 0.6562168598175049,
+      "learning_rate": 0.0002,
+      "loss": 1.5592,
+      "step": 3770
+    },
+    {
+      "epoch": 3.180479596129575,
+      "grad_norm": 0.5820156335830688,
+      "learning_rate": 0.0002,
+      "loss": 1.4932,
+      "step": 3780
+    },
+    {
+      "epoch": 3.188893563315103,
+      "grad_norm": 0.5784737467765808,
+      "learning_rate": 0.0002,
+      "loss": 1.4523,
+      "step": 3790
+    },
+    {
+      "epoch": 3.197307530500631,
+      "grad_norm": 0.5506529808044434,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 3800
+    },
+    {
+      "epoch": 3.205721497686159,
+      "grad_norm": 0.6101595163345337,
+      "learning_rate": 0.0002,
+      "loss": 1.4819,
+      "step": 3810
+    },
+    {
+      "epoch": 3.214135464871687,
+      "grad_norm": 0.5597806572914124,
+      "learning_rate": 0.0002,
+      "loss": 1.5185,
+      "step": 3820
+    },
+    {
+      "epoch": 3.222549432057215,
+      "grad_norm": 0.5641011595726013,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 3830
+    },
+    {
+      "epoch": 3.230963399242743,
+      "grad_norm": 0.5892080068588257,
+      "learning_rate": 0.0002,
+      "loss": 1.4702,
+      "step": 3840
+    },
+    {
+      "epoch": 3.239377366428271,
+      "grad_norm": 0.6034760475158691,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 3850
+    },
+    {
+      "epoch": 3.247791333613799,
+      "grad_norm": 0.5112439393997192,
+      "learning_rate": 0.0002,
+      "loss": 1.5499,
+      "step": 3860
+    },
+    {
+      "epoch": 3.256205300799327,
+      "grad_norm": 0.56565922498703,
+      "learning_rate": 0.0002,
+      "loss": 1.5132,
+      "step": 3870
+    },
+    {
+      "epoch": 3.264619267984855,
+      "grad_norm": 0.6155247092247009,
+      "learning_rate": 0.0002,
+      "loss": 1.4892,
+      "step": 3880
+    },
+    {
+      "epoch": 3.273033235170383,
+      "grad_norm": 0.6064623594284058,
+      "learning_rate": 0.0002,
+      "loss": 1.5118,
+      "step": 3890
+    },
+    {
+      "epoch": 3.2814472023559107,
+      "grad_norm": 0.6313768029212952,
+      "learning_rate": 0.0002,
+      "loss": 1.5236,
+      "step": 3900
+    },
+    {
+      "epoch": 3.289861169541439,
+      "grad_norm": 0.5903939008712769,
+      "learning_rate": 0.0002,
+      "loss": 1.5551,
+      "step": 3910
+    },
+    {
+      "epoch": 3.298275136726967,
+      "grad_norm": 0.5770667195320129,
+      "learning_rate": 0.0002,
+      "loss": 1.5703,
+      "step": 3920
+    },
+    {
+      "epoch": 3.3066891039124946,
+      "grad_norm": 0.5785196423530579,
+      "learning_rate": 0.0002,
+      "loss": 1.5159,
+      "step": 3930
+    },
+    {
+      "epoch": 3.3151030710980227,
+      "grad_norm": 0.6468310356140137,
+      "learning_rate": 0.0002,
+      "loss": 1.5277,
+      "step": 3940
+    },
+    {
+      "epoch": 3.323517038283551,
+      "grad_norm": 0.6200279593467712,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 3950
+    },
+    {
+      "epoch": 3.3319310054690785,
+      "grad_norm": 0.5779302716255188,
+      "learning_rate": 0.0002,
+      "loss": 1.5264,
+      "step": 3960
+    },
+    {
+      "epoch": 3.3403449726546066,
+      "grad_norm": 0.5463796854019165,
+      "learning_rate": 0.0002,
+      "loss": 1.4861,
+      "step": 3970
+    },
+    {
+      "epoch": 3.3487589398401347,
+      "grad_norm": 0.6117855906486511,
+      "learning_rate": 0.0002,
+      "loss": 1.541,
+      "step": 3980
+    },
+    {
+      "epoch": 3.357172907025663,
+      "grad_norm": 0.5554766058921814,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 3990
+    },
+    {
+      "epoch": 3.3655868742111905,
+      "grad_norm": 0.6012870073318481,
+      "learning_rate": 0.0002,
+      "loss": 1.5004,
+      "step": 4000
+    },
+    {
+      "epoch": 3.3740008413967186,
+      "grad_norm": 0.5443974137306213,
+      "learning_rate": 0.0002,
+      "loss": 1.473,
+      "step": 4010
+    },
+    {
+      "epoch": 3.3824148085822463,
+      "grad_norm": 0.6636057496070862,
+      "learning_rate": 0.0002,
+      "loss": 1.5139,
+      "step": 4020
+    },
+    {
+      "epoch": 3.3908287757677744,
+      "grad_norm": 0.5801246166229248,
+      "learning_rate": 0.0002,
+      "loss": 1.5141,
+      "step": 4030
+    },
+    {
+      "epoch": 3.3992427429533025,
+      "grad_norm": 0.5668839812278748,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4040
+    },
+    {
+      "epoch": 3.4076567101388306,
+      "grad_norm": 0.7763481736183167,
+      "learning_rate": 0.0002,
+      "loss": 1.523,
+      "step": 4050
+    },
+    {
+      "epoch": 3.4160706773243583,
+      "grad_norm": 0.6675992608070374,
+      "learning_rate": 0.0002,
+      "loss": 1.4932,
+      "step": 4060
+    },
+    {
+      "epoch": 3.4244846445098864,
+      "grad_norm": 0.6290077567100525,
+      "learning_rate": 0.0002,
+      "loss": 1.4959,
+      "step": 4070
+    },
+    {
+      "epoch": 3.4328986116954145,
+      "grad_norm": 0.6040239930152893,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 4080
+    },
+    {
+      "epoch": 3.441312578880942,
+      "grad_norm": 0.6237877607345581,
+      "learning_rate": 0.0002,
+      "loss": 1.5711,
+      "step": 4090
+    },
+    {
+      "epoch": 3.4497265460664703,
+      "grad_norm": 0.5343508124351501,
+      "learning_rate": 0.0002,
+      "loss": 1.4961,
+      "step": 4100
+    },
+    {
+      "epoch": 3.4581405132519984,
+      "grad_norm": 0.6817412972450256,
+      "learning_rate": 0.0002,
+      "loss": 1.5123,
+      "step": 4110
+    },
+    {
+      "epoch": 3.466554480437526,
+      "grad_norm": 0.7115170359611511,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 4120
+    },
+    {
+      "epoch": 3.4749684476230542,
+      "grad_norm": 0.6127332448959351,
+      "learning_rate": 0.0002,
+      "loss": 1.5275,
+      "step": 4130
+    },
+    {
+      "epoch": 3.4833824148085824,
+      "grad_norm": 0.5745994448661804,
+      "learning_rate": 0.0002,
+      "loss": 1.557,
+      "step": 4140
+    },
+    {
+      "epoch": 3.49179638199411,
+      "grad_norm": 0.6248795390129089,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4150
+    },
+    {
+      "epoch": 3.500210349179638,
+      "grad_norm": 0.5821124911308289,
+      "learning_rate": 0.0002,
+      "loss": 1.4885,
+      "step": 4160
+    },
+    {
+      "epoch": 3.5086243163651663,
+      "grad_norm": 0.561416506767273,
+      "learning_rate": 0.0002,
+      "loss": 1.4937,
+      "step": 4170
+    },
+    {
+      "epoch": 3.5170382835506944,
+      "grad_norm": 0.5848962664604187,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 4180
+    },
+    {
+      "epoch": 3.525452250736222,
+      "grad_norm": 0.5335569977760315,
+      "learning_rate": 0.0002,
+      "loss": 1.5892,
+      "step": 4190
+    },
+    {
+      "epoch": 3.53386621792175,
+      "grad_norm": 0.547964870929718,
+      "learning_rate": 0.0002,
+      "loss": 1.5152,
+      "step": 4200
+    },
+    {
+      "epoch": 3.542280185107278,
+      "grad_norm": 0.6157727241516113,
+      "learning_rate": 0.0002,
+      "loss": 1.4887,
+      "step": 4210
+    },
+    {
+      "epoch": 3.550694152292806,
+      "grad_norm": 0.6163121461868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5484,
+      "step": 4220
+    },
+    {
+      "epoch": 3.559108119478334,
+      "grad_norm": 0.5844616293907166,
+      "learning_rate": 0.0002,
+      "loss": 1.5833,
+      "step": 4230
+    },
+    {
+      "epoch": 3.567522086663862,
+      "grad_norm": 0.7104926109313965,
+      "learning_rate": 0.0002,
+      "loss": 1.5305,
+      "step": 4240
+    },
+    {
+      "epoch": 3.57593605384939,
+      "grad_norm": 0.5055213570594788,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4250
+    },
+    {
+      "epoch": 3.584350021034918,
+      "grad_norm": 0.611676812171936,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 4260
+    },
+    {
+      "epoch": 3.592763988220446,
+      "grad_norm": 0.6326440572738647,
+      "learning_rate": 0.0002,
+      "loss": 1.5048,
+      "step": 4270
+    },
+    {
+      "epoch": 3.6011779554059737,
+      "grad_norm": 0.6290925741195679,
+      "learning_rate": 0.0002,
+      "loss": 1.5122,
+      "step": 4280
+    },
+    {
+      "epoch": 3.609591922591502,
+      "grad_norm": 0.5691978931427002,
+      "learning_rate": 0.0002,
+      "loss": 1.5654,
+      "step": 4290
+    },
+    {
+      "epoch": 3.61800588977703,
+      "grad_norm": 0.6071329116821289,
+      "learning_rate": 0.0002,
+      "loss": 1.4854,
+      "step": 4300
+    },
+    {
+      "epoch": 3.626419856962558,
+      "grad_norm": 0.606573224067688,
+      "learning_rate": 0.0002,
+      "loss": 1.5336,
+      "step": 4310
+    },
+    {
+      "epoch": 3.6348338241480858,
+      "grad_norm": 0.5515419244766235,
+      "learning_rate": 0.0002,
+      "loss": 1.6437,
+      "step": 4320
+    },
+    {
+      "epoch": 3.643247791333614,
+      "grad_norm": 0.5964660048484802,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 4330
+    },
+    {
+      "epoch": 3.6516617585191415,
+      "grad_norm": 0.5774146914482117,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 4340
+    },
+    {
+      "epoch": 3.6600757257046697,
+      "grad_norm": 0.5732731223106384,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 4350
+    },
+    {
+      "epoch": 3.6684896928901978,
+      "grad_norm": 0.7354163527488708,
+      "learning_rate": 0.0002,
+      "loss": 1.5682,
+      "step": 4360
+    },
+    {
+      "epoch": 3.676903660075726,
+      "grad_norm": 0.6220902800559998,
+      "learning_rate": 0.0002,
+      "loss": 1.5225,
+      "step": 4370
+    },
+    {
+      "epoch": 3.6853176272612536,
+      "grad_norm": 0.6053991317749023,
+      "learning_rate": 0.0002,
+      "loss": 1.4838,
+      "step": 4380
+    },
+    {
+      "epoch": 3.6937315944467817,
+      "grad_norm": 0.67010897397995,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4390
+    },
+    {
+      "epoch": 3.70214556163231,
+      "grad_norm": 0.6139186024665833,
+      "learning_rate": 0.0002,
+      "loss": 1.5381,
+      "step": 4400
+    },
+    {
+      "epoch": 3.7105595288178375,
+      "grad_norm": 0.5433071851730347,
+      "learning_rate": 0.0002,
+      "loss": 1.5088,
+      "step": 4410
+    },
+    {
+      "epoch": 3.7189734960033656,
+      "grad_norm": 0.5453870296478271,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 4420
+    },
+    {
+      "epoch": 3.7273874631888937,
+      "grad_norm": 0.6401727199554443,
+      "learning_rate": 0.0002,
+      "loss": 1.4549,
+      "step": 4430
+    },
+    {
+      "epoch": 3.735801430374422,
+      "grad_norm": 0.6049367189407349,
+      "learning_rate": 0.0002,
+      "loss": 1.503,
+      "step": 4440
+    },
+    {
+      "epoch": 3.7442153975599495,
+      "grad_norm": 0.5740529298782349,
+      "learning_rate": 0.0002,
+      "loss": 1.5268,
+      "step": 4450
+    },
+    {
+      "epoch": 3.7526293647454776,
+      "grad_norm": 0.6521880626678467,
+      "learning_rate": 0.0002,
+      "loss": 1.5183,
+      "step": 4460
+    },
+    {
+      "epoch": 3.7610433319310053,
+      "grad_norm": 0.7096368074417114,
+      "learning_rate": 0.0002,
+      "loss": 1.5741,
+      "step": 4470
+    },
+    {
+      "epoch": 3.7694572991165334,
+      "grad_norm": 0.5886474251747131,
+      "learning_rate": 0.0002,
+      "loss": 1.5786,
+      "step": 4480
+    },
+    {
+      "epoch": 3.7778712663020615,
+      "grad_norm": 0.5821043252944946,
+      "learning_rate": 0.0002,
+      "loss": 1.5887,
+      "step": 4490
+    },
+    {
+      "epoch": 3.7862852334875896,
+      "grad_norm": 0.628892183303833,
+      "learning_rate": 0.0002,
+      "loss": 1.5777,
+      "step": 4500
+    },
+    {
+      "epoch": 3.7946992006731173,
+      "grad_norm": 0.5962669849395752,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 4510
+    },
+    {
+      "epoch": 3.8031131678586454,
+      "grad_norm": 0.6635549068450928,
+      "learning_rate": 0.0002,
+      "loss": 1.5267,
+      "step": 4520
+    },
+    {
+      "epoch": 3.811527135044173,
+      "grad_norm": 0.6010760068893433,
+      "learning_rate": 0.0002,
+      "loss": 1.5058,
+      "step": 4530
+    },
+    {
+      "epoch": 3.819941102229701,
+      "grad_norm": 0.6322658658027649,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 4540
+    },
+    {
+      "epoch": 3.8283550694152293,
+      "grad_norm": 0.5893137454986572,
+      "learning_rate": 0.0002,
+      "loss": 1.5029,
+      "step": 4550
+    },
+    {
+      "epoch": 3.8367690366007574,
+      "grad_norm": 0.7829602360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.5435,
+      "step": 4560
+    },
+    {
+      "epoch": 3.845183003786285,
+      "grad_norm": 0.6190396547317505,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 4570
+    },
+    {
+      "epoch": 3.853596970971813,
+      "grad_norm": 0.6662813425064087,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 4580
+    },
+    {
+      "epoch": 3.8620109381573413,
+      "grad_norm": 0.5809855461120605,
+      "learning_rate": 0.0002,
+      "loss": 1.5065,
+      "step": 4590
+    },
+    {
+      "epoch": 3.870424905342869,
+      "grad_norm": 0.5779069662094116,
+      "learning_rate": 0.0002,
+      "loss": 1.5041,
+      "step": 4600
+    },
+    {
+      "epoch": 3.878838872528397,
+      "grad_norm": 0.5603038668632507,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 4610
+    },
+    {
+      "epoch": 3.887252839713925,
+      "grad_norm": 0.6274181008338928,
+      "learning_rate": 0.0002,
+      "loss": 1.5372,
+      "step": 4620
+    },
+    {
+      "epoch": 3.8956668068994533,
+      "grad_norm": 0.6810959577560425,
+      "learning_rate": 0.0002,
+      "loss": 1.4996,
+      "step": 4630
+    },
+    {
+      "epoch": 3.904080774084981,
+      "grad_norm": 0.5647315979003906,
+      "learning_rate": 0.0002,
+      "loss": 1.4956,
+      "step": 4640
+    },
+    {
+      "epoch": 3.912494741270509,
+      "grad_norm": 0.6830295324325562,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 4650
+    },
+    {
+      "epoch": 3.920908708456037,
+      "grad_norm": 0.652565598487854,
+      "learning_rate": 0.0002,
+      "loss": 1.535,
+      "step": 4660
+    },
+    {
+      "epoch": 3.929322675641565,
+      "grad_norm": 0.5806284546852112,
+      "learning_rate": 0.0002,
+      "loss": 1.4772,
+      "step": 4670
+    },
+    {
+      "epoch": 3.937736642827093,
+      "grad_norm": 0.6825073957443237,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 4680
+    },
+    {
+      "epoch": 3.946150610012621,
+      "grad_norm": 0.6149451732635498,
+      "learning_rate": 0.0002,
+      "loss": 1.5516,
+      "step": 4690
+    },
+    {
+      "epoch": 3.954564577198149,
+      "grad_norm": 0.6152557134628296,
+      "learning_rate": 0.0002,
+      "loss": 1.5608,
+      "step": 4700
+    },
+    {
+      "epoch": 3.962978544383677,
+      "grad_norm": 0.6239011883735657,
+      "learning_rate": 0.0002,
+      "loss": 1.4897,
+      "step": 4710
+    },
+    {
+      "epoch": 3.971392511569205,
+      "grad_norm": 0.6485443115234375,
+      "learning_rate": 0.0002,
+      "loss": 1.538,
+      "step": 4720
+    },
+    {
+      "epoch": 3.9798064787547327,
+      "grad_norm": 0.6449228525161743,
+      "learning_rate": 0.0002,
+      "loss": 1.5226,
+      "step": 4730
+    },
+    {
+      "epoch": 3.988220445940261,
+      "grad_norm": 0.6526407599449158,
+      "learning_rate": 0.0002,
+      "loss": 1.5087,
+      "step": 4740
+    },
+    {
+      "epoch": 3.996634413125789,
+      "grad_norm": 0.6277706027030945,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4750
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.871641755104065,
+      "eval_runtime": 37.9637,
+      "eval_samples_per_second": 13.566,
+      "eval_steps_per_second": 1.712,
+      "step": 4754
+    },
+    {
+      "epoch": 4.005048380311317,
+      "grad_norm": 0.6994837522506714,
+      "learning_rate": 0.0002,
+      "loss": 1.4744,
+      "step": 4760
+    },
+    {
+      "epoch": 4.013462347496845,
+      "grad_norm": 0.8728373050689697,
+      "learning_rate": 0.0002,
+      "loss": 1.4433,
+      "step": 4770
+    },
+    {
+      "epoch": 4.021876314682372,
+      "grad_norm": 0.688679575920105,
+      "learning_rate": 0.0002,
+      "loss": 1.3329,
+      "step": 4780
+    },
+    {
+      "epoch": 4.0302902818679005,
+      "grad_norm": 0.6313387155532837,
+      "learning_rate": 0.0002,
+      "loss": 1.3999,
+      "step": 4790
+    },
+    {
+      "epoch": 4.038704249053429,
+      "grad_norm": 0.6577984690666199,
+      "learning_rate": 0.0002,
+      "loss": 1.3346,
+      "step": 4800
+    },
+    {
+      "epoch": 4.047118216238957,
+      "grad_norm": 0.7938185930252075,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 4810
+    },
+    {
+      "epoch": 4.055532183424485,
+      "grad_norm": 0.760399580001831,
+      "learning_rate": 0.0002,
+      "loss": 1.3716,
+      "step": 4820
+    },
+    {
+      "epoch": 4.063946150610013,
+      "grad_norm": 0.7329602241516113,
+      "learning_rate": 0.0002,
+      "loss": 1.4321,
+      "step": 4830
+    },
+    {
+      "epoch": 4.07236011779554,
+      "grad_norm": 0.7778576016426086,
+      "learning_rate": 0.0002,
+      "loss": 1.4133,
+      "step": 4840
+    },
+    {
+      "epoch": 4.080774084981068,
+      "grad_norm": 0.8235865235328674,
+      "learning_rate": 0.0002,
+      "loss": 1.4372,
+      "step": 4850
+    },
+    {
+      "epoch": 4.089188052166596,
+      "grad_norm": 0.7743754386901855,
+      "learning_rate": 0.0002,
+      "loss": 1.3719,
+      "step": 4860
+    },
+    {
+      "epoch": 4.0976020193521245,
+      "grad_norm": 0.8145367503166199,
+      "learning_rate": 0.0002,
+      "loss": 1.3787,
+      "step": 4870
+    },
+    {
+      "epoch": 4.106015986537653,
+      "grad_norm": 0.8517307639122009,
+      "learning_rate": 0.0002,
+      "loss": 1.356,
+      "step": 4880
+    },
+    {
+      "epoch": 4.114429953723181,
+      "grad_norm": 0.8208953142166138,
+      "learning_rate": 0.0002,
+      "loss": 1.4191,
+      "step": 4890
+    },
+    {
+      "epoch": 4.122843920908709,
+      "grad_norm": 0.8437790870666504,
+      "learning_rate": 0.0002,
+      "loss": 1.3189,
+      "step": 4900
+    },
+    {
+      "epoch": 4.131257888094236,
+      "grad_norm": 0.716672420501709,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 4910
+    },
+    {
+      "epoch": 4.139671855279764,
+      "grad_norm": 0.7656235098838806,
+      "learning_rate": 0.0002,
+      "loss": 1.4392,
+      "step": 4920
+    },
+    {
+      "epoch": 4.148085822465292,
+      "grad_norm": 0.7209306955337524,
+      "learning_rate": 0.0002,
+      "loss": 1.3408,
+      "step": 4930
+    },
+    {
+      "epoch": 4.1564997896508205,
+      "grad_norm": 0.7731267809867859,
+      "learning_rate": 0.0002,
+      "loss": 1.3639,
+      "step": 4940
+    },
+    {
+      "epoch": 4.164913756836349,
+      "grad_norm": 0.7477553486824036,
+      "learning_rate": 0.0002,
+      "loss": 1.4151,
+      "step": 4950
+    },
+    {
+      "epoch": 4.173327724021877,
+      "grad_norm": 0.7372981309890747,
+      "learning_rate": 0.0002,
+      "loss": 1.3485,
+      "step": 4960
+    },
+    {
+      "epoch": 4.181741691207404,
+      "grad_norm": 0.6582154035568237,
+      "learning_rate": 0.0002,
+      "loss": 1.3901,
+      "step": 4970
+    },
+    {
+      "epoch": 4.190155658392932,
+      "grad_norm": 0.7003206610679626,
+      "learning_rate": 0.0002,
+      "loss": 1.3343,
+      "step": 4980
+    },
+    {
+      "epoch": 4.19856962557846,
+      "grad_norm": 0.735223650932312,
+      "learning_rate": 0.0002,
+      "loss": 1.4098,
+      "step": 4990
+    },
+    {
+      "epoch": 4.206983592763988,
+      "grad_norm": 0.7832302451133728,
+      "learning_rate": 0.0002,
+      "loss": 1.3564,
+      "step": 5000
+    },
+    {
+      "epoch": 4.215397559949516,
+      "grad_norm": 0.8819546103477478,
+      "learning_rate": 0.0002,
+      "loss": 1.3622,
+      "step": 5010
+    },
+    {
+      "epoch": 4.2238115271350445,
+      "grad_norm": 0.9325336813926697,
+      "learning_rate": 0.0002,
+      "loss": 1.4438,
+      "step": 5020
+    },
+    {
+      "epoch": 4.232225494320572,
+      "grad_norm": 0.7007517218589783,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 5030
+    },
+    {
+      "epoch": 4.2406394615061,
+      "grad_norm": 0.7118321061134338,
+      "learning_rate": 0.0002,
+      "loss": 1.3683,
+      "step": 5040
+    },
+    {
+      "epoch": 4.249053428691628,
+      "grad_norm": 0.6578946709632874,
+      "learning_rate": 0.0002,
+      "loss": 1.2365,
+      "step": 5050
+    },
+    {
+      "epoch": 4.257467395877156,
+      "grad_norm": 0.9438983798027039,
+      "learning_rate": 0.0002,
+      "loss": 1.3696,
+      "step": 5060
+    },
+    {
+      "epoch": 4.265881363062684,
+      "grad_norm": 0.703037679195404,
+      "learning_rate": 0.0002,
+      "loss": 1.3868,
+      "step": 5070
+    },
+    {
+      "epoch": 4.274295330248212,
+      "grad_norm": 0.7286025285720825,
+      "learning_rate": 0.0002,
+      "loss": 1.3687,
+      "step": 5080
+    },
+    {
+      "epoch": 4.28270929743374,
+      "grad_norm": 0.750689685344696,
+      "learning_rate": 0.0002,
+      "loss": 1.3605,
+      "step": 5090
+    },
+    {
+      "epoch": 4.291123264619268,
+      "grad_norm": 0.869753360748291,
+      "learning_rate": 0.0002,
+      "loss": 1.5089,
+      "step": 5100
+    },
+    {
+      "epoch": 4.299537231804796,
+      "grad_norm": 0.8712980151176453,
+      "learning_rate": 0.0002,
+      "loss": 1.4128,
+      "step": 5110
+    },
+    {
+      "epoch": 4.307951198990324,
+      "grad_norm": 0.690263569355011,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 5120
+    },
+    {
+      "epoch": 4.316365166175852,
+      "grad_norm": 0.7114760279655457,
+      "learning_rate": 0.0002,
+      "loss": 1.4088,
+      "step": 5130
+    },
+    {
+      "epoch": 4.32477913336138,
+      "grad_norm": 0.7588112354278564,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 5140
+    },
+    {
+      "epoch": 4.333193100546908,
+      "grad_norm": 0.7556202411651611,
+      "learning_rate": 0.0002,
+      "loss": 1.4408,
+      "step": 5150
+    },
+    {
+      "epoch": 4.341607067732435,
+      "grad_norm": 0.8357610702514648,
+      "learning_rate": 0.0002,
+      "loss": 1.4203,
+      "step": 5160
+    },
+    {
+      "epoch": 4.3500210349179635,
+      "grad_norm": 0.8054035902023315,
+      "learning_rate": 0.0002,
+      "loss": 1.3348,
+      "step": 5170
+    },
+    {
+      "epoch": 4.358435002103492,
+      "grad_norm": 0.7637107968330383,
+      "learning_rate": 0.0002,
+      "loss": 1.3109,
+      "step": 5180
+    },
+    {
+      "epoch": 4.36684896928902,
+      "grad_norm": 0.757481038570404,
+      "learning_rate": 0.0002,
+      "loss": 1.3744,
+      "step": 5190
+    },
+    {
+      "epoch": 4.375262936474548,
+      "grad_norm": 0.7185863852500916,
+      "learning_rate": 0.0002,
+      "loss": 1.3622,
+      "step": 5200
+    },
+    {
+      "epoch": 4.383676903660076,
+      "grad_norm": 0.7326455116271973,
+      "learning_rate": 0.0002,
+      "loss": 1.3896,
+      "step": 5210
+    },
+    {
+      "epoch": 4.392090870845603,
+      "grad_norm": 0.7980523109436035,
+      "learning_rate": 0.0002,
+      "loss": 1.4098,
+      "step": 5220
+    },
+    {
+      "epoch": 4.400504838031131,
+      "grad_norm": 0.8526999354362488,
+      "learning_rate": 0.0002,
+      "loss": 1.3783,
+      "step": 5230
+    },
+    {
+      "epoch": 4.4089188052166595,
+      "grad_norm": 0.7012337446212769,
+      "learning_rate": 0.0002,
+      "loss": 1.4022,
+      "step": 5240
+    },
+    {
+      "epoch": 4.417332772402188,
+      "grad_norm": 0.8217827677726746,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 5250
+    },
+    {
+      "epoch": 4.425746739587716,
+      "grad_norm": 0.7141005396842957,
+      "learning_rate": 0.0002,
+      "loss": 1.3482,
+      "step": 5260
+    },
+    {
+      "epoch": 4.434160706773244,
+      "grad_norm": 0.7094302177429199,
+      "learning_rate": 0.0002,
+      "loss": 1.3699,
+      "step": 5270
+    },
+    {
+      "epoch": 4.442574673958772,
+      "grad_norm": 0.7234613299369812,
+      "learning_rate": 0.0002,
+      "loss": 1.3527,
+      "step": 5280
+    },
+    {
+      "epoch": 4.450988641144299,
+      "grad_norm": 0.7530457973480225,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 5290
+    },
+    {
+      "epoch": 4.459402608329827,
+      "grad_norm": 0.7300912141799927,
+      "learning_rate": 0.0002,
+      "loss": 1.3944,
+      "step": 5300
+    },
+    {
+      "epoch": 4.467816575515355,
+      "grad_norm": 0.825443685054779,
+      "learning_rate": 0.0002,
+      "loss": 1.3844,
+      "step": 5310
+    },
+    {
+      "epoch": 4.4762305427008835,
+      "grad_norm": 0.7559658885002136,
+      "learning_rate": 0.0002,
+      "loss": 1.3648,
+      "step": 5320
+    },
+    {
+      "epoch": 4.484644509886412,
+      "grad_norm": 0.8817561268806458,
+      "learning_rate": 0.0002,
+      "loss": 1.4364,
+      "step": 5330
+    },
+    {
+      "epoch": 4.49305847707194,
+      "grad_norm": 0.8203575611114502,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 5340
+    },
+    {
+      "epoch": 4.501472444257468,
+      "grad_norm": 0.7677690982818604,
+      "learning_rate": 0.0002,
+      "loss": 1.3996,
+      "step": 5350
+    },
+    {
+      "epoch": 4.509886411442995,
+      "grad_norm": 0.657085120677948,
+      "learning_rate": 0.0002,
+      "loss": 1.4142,
+      "step": 5360
+    },
+    {
+      "epoch": 4.518300378628523,
+      "grad_norm": 0.7939504384994507,
+      "learning_rate": 0.0002,
+      "loss": 1.3722,
+      "step": 5370
+    },
+    {
+      "epoch": 4.526714345814051,
+      "grad_norm": 0.6971889138221741,
+      "learning_rate": 0.0002,
+      "loss": 1.4361,
+      "step": 5380
+    },
+    {
+      "epoch": 4.535128312999579,
+      "grad_norm": 0.6984175443649292,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 5390
+    },
+    {
+      "epoch": 4.5435422801851075,
+      "grad_norm": 0.8504858613014221,
+      "learning_rate": 0.0002,
+      "loss": 1.341,
+      "step": 5400
+    },
+    {
+      "epoch": 4.551956247370635,
+      "grad_norm": 0.9134073853492737,
+      "learning_rate": 0.0002,
+      "loss": 1.4026,
+      "step": 5410
+    },
+    {
+      "epoch": 4.560370214556163,
+      "grad_norm": 0.7765598893165588,
+      "learning_rate": 0.0002,
+      "loss": 1.4375,
+      "step": 5420
+    },
+    {
+      "epoch": 4.568784181741691,
+      "grad_norm": 0.6991009712219238,
+      "learning_rate": 0.0002,
+      "loss": 1.4832,
+      "step": 5430
+    },
+    {
+      "epoch": 4.577198148927219,
+      "grad_norm": 0.8393039107322693,
+      "learning_rate": 0.0002,
+      "loss": 1.4021,
+      "step": 5440
+    },
+    {
+      "epoch": 4.585612116112747,
+      "grad_norm": 0.7685918211936951,
+      "learning_rate": 0.0002,
+      "loss": 1.3976,
+      "step": 5450
+    },
+    {
+      "epoch": 4.594026083298275,
+      "grad_norm": 0.7135679721832275,
+      "learning_rate": 0.0002,
+      "loss": 1.3883,
+      "step": 5460
+    },
+    {
+      "epoch": 4.6024400504838034,
+      "grad_norm": 0.6728870868682861,
+      "learning_rate": 0.0002,
+      "loss": 1.4083,
+      "step": 5470
+    },
+    {
+      "epoch": 4.610854017669331,
+      "grad_norm": 0.7139479517936707,
+      "learning_rate": 0.0002,
+      "loss": 1.3698,
+      "step": 5480
+    },
+    {
+      "epoch": 4.619267984854859,
+      "grad_norm": 0.8476598858833313,
+      "learning_rate": 0.0002,
+      "loss": 1.3498,
+      "step": 5490
+    },
+    {
+      "epoch": 4.627681952040387,
+      "grad_norm": 0.8034361004829407,
+      "learning_rate": 0.0002,
+      "loss": 1.3389,
+      "step": 5500
+    },
+    {
+      "epoch": 4.636095919225915,
+      "grad_norm": 0.7452183961868286,
+      "learning_rate": 0.0002,
+      "loss": 1.4179,
+      "step": 5510
+    },
+    {
+      "epoch": 4.644509886411443,
+      "grad_norm": 0.8394148945808411,
+      "learning_rate": 0.0002,
+      "loss": 1.4031,
+      "step": 5520
+    },
+    {
+      "epoch": 4.652923853596971,
+      "grad_norm": 0.7480153441429138,
+      "learning_rate": 0.0002,
+      "loss": 1.4561,
+      "step": 5530
+    },
+    {
+      "epoch": 4.661337820782499,
+      "grad_norm": 0.7781714797019958,
+      "learning_rate": 0.0002,
+      "loss": 1.378,
+      "step": 5540
+    },
+    {
+      "epoch": 4.669751787968027,
+      "grad_norm": 1.0058213472366333,
+      "learning_rate": 0.0002,
+      "loss": 1.3924,
+      "step": 5550
+    },
+    {
+      "epoch": 4.678165755153555,
+      "grad_norm": 0.7403179407119751,
+      "learning_rate": 0.0002,
+      "loss": 1.4198,
+      "step": 5560
+    },
+    {
+      "epoch": 4.686579722339083,
+      "grad_norm": 0.7270476818084717,
+      "learning_rate": 0.0002,
+      "loss": 1.4328,
+      "step": 5570
+    },
+    {
+      "epoch": 4.694993689524611,
+      "grad_norm": 0.760877788066864,
+      "learning_rate": 0.0002,
+      "loss": 1.378,
+      "step": 5580
+    },
+    {
+      "epoch": 4.703407656710139,
+      "grad_norm": 0.8097004890441895,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 5590
+    },
+    {
+      "epoch": 4.711821623895667,
+      "grad_norm": 0.9096523523330688,
+      "learning_rate": 0.0002,
+      "loss": 1.3661,
+      "step": 5600
+    },
+    {
+      "epoch": 4.720235591081195,
+      "grad_norm": 0.7262444496154785,
+      "learning_rate": 0.0002,
+      "loss": 1.4012,
+      "step": 5610
+    },
+    {
+      "epoch": 4.7286495582667225,
+      "grad_norm": 0.8207762837409973,
+      "learning_rate": 0.0002,
+      "loss": 1.422,
+      "step": 5620
+    },
+    {
+      "epoch": 4.737063525452251,
+      "grad_norm": 0.8089601993560791,
+      "learning_rate": 0.0002,
+      "loss": 1.4017,
+      "step": 5630
+    },
+    {
+      "epoch": 4.745477492637779,
+      "grad_norm": 0.7609543800354004,
+      "learning_rate": 0.0002,
+      "loss": 1.3675,
+      "step": 5640
+    },
+    {
+      "epoch": 4.753891459823307,
+      "grad_norm": 0.7273501753807068,
+      "learning_rate": 0.0002,
+      "loss": 1.4085,
+      "step": 5650
+    },
+    {
+      "epoch": 4.762305427008835,
+      "grad_norm": 0.7800219058990479,
+      "learning_rate": 0.0002,
+      "loss": 1.3849,
+      "step": 5660
+    },
+    {
+      "epoch": 4.770719394194362,
+      "grad_norm": 0.8558377623558044,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 5670
+    },
+    {
+      "epoch": 4.77913336137989,
+      "grad_norm": 0.7131547927856445,
+      "learning_rate": 0.0002,
+      "loss": 1.3831,
+      "step": 5680
+    },
+    {
+      "epoch": 4.787547328565418,
+      "grad_norm": 0.7651025056838989,
+      "learning_rate": 0.0002,
+      "loss": 1.407,
+      "step": 5690
+    },
+    {
+      "epoch": 4.7959612957509465,
+      "grad_norm": 0.8129976391792297,
+      "learning_rate": 0.0002,
+      "loss": 1.3882,
+      "step": 5700
+    },
+    {
+      "epoch": 4.804375262936475,
+      "grad_norm": 0.8019895553588867,
+      "learning_rate": 0.0002,
+      "loss": 1.4347,
+      "step": 5710
+    },
+    {
+      "epoch": 4.812789230122003,
+      "grad_norm": 0.7692018151283264,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 5720
+    },
+    {
+      "epoch": 4.821203197307531,
+      "grad_norm": 0.6893943548202515,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 5730
+    },
+    {
+      "epoch": 4.829617164493058,
+      "grad_norm": 0.6881810426712036,
+      "learning_rate": 0.0002,
+      "loss": 1.4453,
+      "step": 5740
+    },
+    {
+      "epoch": 4.838031131678586,
+      "grad_norm": 0.7838267683982849,
+      "learning_rate": 0.0002,
+      "loss": 1.4775,
+      "step": 5750
+    },
+    {
+      "epoch": 4.846445098864114,
+      "grad_norm": 0.727799117565155,
+      "learning_rate": 0.0002,
+      "loss": 1.3857,
+      "step": 5760
+    },
+    {
+      "epoch": 4.8548590660496425,
+      "grad_norm": 0.7458277344703674,
+      "learning_rate": 0.0002,
+      "loss": 1.4685,
+      "step": 5770
+    },
+    {
+      "epoch": 4.863273033235171,
+      "grad_norm": 0.903802216053009,
+      "learning_rate": 0.0002,
+      "loss": 1.4426,
+      "step": 5780
+    },
+    {
+      "epoch": 4.871687000420699,
+      "grad_norm": 0.7983472347259521,
+      "learning_rate": 0.0002,
+      "loss": 1.451,
+      "step": 5790
+    },
+    {
+      "epoch": 4.880100967606227,
+      "grad_norm": 0.6894361972808838,
+      "learning_rate": 0.0002,
+      "loss": 1.4534,
+      "step": 5800
+    },
+    {
+      "epoch": 4.888514934791754,
+      "grad_norm": 0.7499409317970276,
+      "learning_rate": 0.0002,
+      "loss": 1.4486,
+      "step": 5810
+    },
+    {
+      "epoch": 4.896928901977282,
+      "grad_norm": 0.7362820506095886,
+      "learning_rate": 0.0002,
+      "loss": 1.4253,
+      "step": 5820
+    },
+    {
+      "epoch": 4.90534286916281,
+      "grad_norm": 0.8341619968414307,
+      "learning_rate": 0.0002,
+      "loss": 1.3763,
+      "step": 5830
+    },
+    {
+      "epoch": 4.913756836348338,
+      "grad_norm": 0.9604470133781433,
+      "learning_rate": 0.0002,
+      "loss": 1.3748,
+      "step": 5840
+    },
+    {
+      "epoch": 4.9221708035338665,
+      "grad_norm": 0.8916844129562378,
+      "learning_rate": 0.0002,
+      "loss": 1.3658,
+      "step": 5850
+    },
+    {
+      "epoch": 4.930584770719394,
+      "grad_norm": 0.8519647121429443,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 5860
+    },
+    {
+      "epoch": 4.938998737904922,
+      "grad_norm": 0.7946906089782715,
+      "learning_rate": 0.0002,
+      "loss": 1.424,
+      "step": 5870
+    },
+    {
+      "epoch": 4.94741270509045,
+      "grad_norm": 0.7843789458274841,
+      "learning_rate": 0.0002,
+      "loss": 1.4071,
+      "step": 5880
+    },
+    {
+      "epoch": 4.955826672275978,
+      "grad_norm": 0.707618772983551,
+      "learning_rate": 0.0002,
+      "loss": 1.4021,
+      "step": 5890
+    },
+    {
+      "epoch": 4.964240639461506,
+      "grad_norm": 0.7704206109046936,
+      "learning_rate": 0.0002,
+      "loss": 1.502,
+      "step": 5900
+    },
+    {
+      "epoch": 4.972654606647034,
+      "grad_norm": 0.7160256505012512,
+      "learning_rate": 0.0002,
+      "loss": 1.4456,
+      "step": 5910
+    },
+    {
+      "epoch": 4.981068573832562,
+      "grad_norm": 0.7020420432090759,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 5920
+    },
+    {
+      "epoch": 4.98948254101809,
+      "grad_norm": 0.7576286792755127,
+      "learning_rate": 0.0002,
+      "loss": 1.4037,
+      "step": 5930
+    },
+    {
+      "epoch": 4.997896508203618,
+      "grad_norm": 0.8573036789894104,
+      "learning_rate": 0.0002,
+      "loss": 1.414,
+      "step": 5940
+    },
+    {
+      "epoch": 4.999579301640724,
+      "eval_loss": 1.9353811740875244,
+      "eval_runtime": 37.9208,
+      "eval_samples_per_second": 13.581,
+      "eval_steps_per_second": 1.714,
+      "step": 5942
+    },
+    {
+      "epoch": 5.006310475389146,
+      "grad_norm": 0.8204267621040344,
+      "learning_rate": 0.0002,
+      "loss": 1.2418,
+      "step": 5950
+    },
+    {
+      "epoch": 5.014724442574674,
+      "grad_norm": 0.976840615272522,
+      "learning_rate": 0.0002,
+      "loss": 1.235,
+      "step": 5960
+    },
+    {
+      "epoch": 5.023138409760202,
+      "grad_norm": 0.8765613436698914,
+      "learning_rate": 0.0002,
+      "loss": 1.2134,
+      "step": 5970
+    },
+    {
+      "epoch": 5.03155237694573,
+      "grad_norm": 1.1793042421340942,
+      "learning_rate": 0.0002,
+      "loss": 1.2748,
+      "step": 5980
+    },
+    {
+      "epoch": 5.039966344131258,
+      "grad_norm": 0.971062958240509,
+      "learning_rate": 0.0002,
+      "loss": 1.2412,
+      "step": 5990
+    },
+    {
+      "epoch": 5.0483803113167856,
+      "grad_norm": 0.8649757504463196,
+      "learning_rate": 0.0002,
+      "loss": 1.1819,
+      "step": 6000
+    },
+    {
+      "epoch": 5.056794278502314,
+      "grad_norm": 0.9563034176826477,
+      "learning_rate": 0.0002,
+      "loss": 1.1654,
+      "step": 6010
+    },
+    {
+      "epoch": 5.065208245687842,
+      "grad_norm": 1.0093994140625,
+      "learning_rate": 0.0002,
+      "loss": 1.2238,
+      "step": 6020
+    },
+    {
+      "epoch": 5.07362221287337,
+      "grad_norm": 1.004213571548462,
+      "learning_rate": 0.0002,
+      "loss": 1.2519,
+      "step": 6030
+    },
+    {
+      "epoch": 5.082036180058898,
+      "grad_norm": 0.8307787179946899,
+      "learning_rate": 0.0002,
+      "loss": 1.2379,
+      "step": 6040
+    },
+    {
+      "epoch": 5.090450147244426,
+      "grad_norm": 0.9117848873138428,
+      "learning_rate": 0.0002,
+      "loss": 1.2282,
+      "step": 6050
+    },
+    {
+      "epoch": 5.098864114429953,
+      "grad_norm": 1.0269840955734253,
+      "learning_rate": 0.0002,
+      "loss": 1.2582,
+      "step": 6060
+    },
+    {
+      "epoch": 5.1072780816154815,
+      "grad_norm": 0.9079542756080627,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 6070
+    },
+    {
+      "epoch": 5.11569204880101,
+      "grad_norm": 0.885702908039093,
+      "learning_rate": 0.0002,
+      "loss": 1.215,
+      "step": 6080
+    },
+    {
+      "epoch": 5.124106015986538,
+      "grad_norm": 0.9976128339767456,
+      "learning_rate": 0.0002,
+      "loss": 1.2406,
+      "step": 6090
+    },
+    {
+      "epoch": 5.132519983172066,
+      "grad_norm": 0.8472117185592651,
+      "learning_rate": 0.0002,
+      "loss": 1.3082,
+      "step": 6100
+    },
+    {
+      "epoch": 5.140933950357594,
+      "grad_norm": 1.0385161638259888,
+      "learning_rate": 0.0002,
+      "loss": 1.226,
+      "step": 6110
+    },
+    {
+      "epoch": 5.149347917543121,
+      "grad_norm": 0.8948383927345276,
+      "learning_rate": 0.0002,
+      "loss": 1.213,
+      "step": 6120
+    },
+    {
+      "epoch": 5.157761884728649,
+      "grad_norm": 1.2613716125488281,
+      "learning_rate": 0.0002,
+      "loss": 1.2213,
+      "step": 6130
+    },
+    {
+      "epoch": 5.166175851914177,
+      "grad_norm": 0.9933410286903381,
+      "learning_rate": 0.0002,
+      "loss": 1.2632,
+      "step": 6140
+    },
+    {
+      "epoch": 5.1745898190997055,
+      "grad_norm": 0.9673663973808289,
+      "learning_rate": 0.0002,
+      "loss": 1.1715,
+      "step": 6150
+    },
+    {
+      "epoch": 5.183003786285234,
+      "grad_norm": 0.9969648122787476,
+      "learning_rate": 0.0002,
+      "loss": 1.2947,
+      "step": 6160
+    },
+    {
+      "epoch": 5.191417753470762,
+      "grad_norm": 1.2163258790969849,
+      "learning_rate": 0.0002,
+      "loss": 1.2416,
+      "step": 6170
+    },
+    {
+      "epoch": 5.19983172065629,
+      "grad_norm": 0.9163419604301453,
+      "learning_rate": 0.0002,
+      "loss": 1.2221,
+      "step": 6180
+    },
+    {
+      "epoch": 5.208245687841817,
+      "grad_norm": 0.9225585460662842,
+      "learning_rate": 0.0002,
+      "loss": 1.2624,
+      "step": 6190
+    },
+    {
+      "epoch": 5.216659655027345,
+      "grad_norm": 0.9205296635627747,
+      "learning_rate": 0.0002,
+      "loss": 1.2932,
+      "step": 6200
+    },
+    {
+      "epoch": 5.225073622212873,
+      "grad_norm": 1.0655443668365479,
+      "learning_rate": 0.0002,
+      "loss": 1.1825,
+      "step": 6210
+    },
+    {
+      "epoch": 5.233487589398401,
+      "grad_norm": 1.0854865312576294,
+      "learning_rate": 0.0002,
+      "loss": 1.2613,
+      "step": 6220
+    },
+    {
+      "epoch": 5.2419015565839295,
+      "grad_norm": 0.8489186763763428,
+      "learning_rate": 0.0002,
+      "loss": 1.3045,
+      "step": 6230
+    },
+    {
+      "epoch": 5.250315523769458,
+      "grad_norm": 0.910391628742218,
+      "learning_rate": 0.0002,
+      "loss": 1.2708,
+      "step": 6240
+    },
+    {
+      "epoch": 5.258729490954985,
+      "grad_norm": 0.925507128238678,
+      "learning_rate": 0.0002,
+      "loss": 1.1914,
+      "step": 6250
+    },
+    {
+      "epoch": 5.267143458140513,
+      "grad_norm": 1.1069735288619995,
+      "learning_rate": 0.0002,
+      "loss": 1.3368,
+      "step": 6260
+    },
+    {
+      "epoch": 5.275557425326041,
+      "grad_norm": 0.9705119132995605,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 6270
+    },
+    {
+      "epoch": 5.283971392511569,
+      "grad_norm": 0.9752426147460938,
+      "learning_rate": 0.0002,
+      "loss": 1.2602,
+      "step": 6280
+    },
+    {
+      "epoch": 5.292385359697097,
+      "grad_norm": 1.021359920501709,
+      "learning_rate": 0.0002,
+      "loss": 1.2043,
+      "step": 6290
+    },
+    {
+      "epoch": 5.3007993268826255,
+      "grad_norm": 1.148606300354004,
+      "learning_rate": 0.0002,
+      "loss": 1.2848,
+      "step": 6300
+    },
+    {
+      "epoch": 5.309213294068153,
+      "grad_norm": 0.8909247517585754,
+      "learning_rate": 0.0002,
+      "loss": 1.2201,
+      "step": 6310
+    },
+    {
+      "epoch": 5.317627261253681,
+      "grad_norm": 0.9879156351089478,
+      "learning_rate": 0.0002,
+      "loss": 1.2376,
+      "step": 6320
+    },
+    {
+      "epoch": 5.326041228439209,
+      "grad_norm": 0.9473357200622559,
+      "learning_rate": 0.0002,
+      "loss": 1.2638,
+      "step": 6330
+    },
+    {
+      "epoch": 5.334455195624737,
+      "grad_norm": 1.1422028541564941,
+      "learning_rate": 0.0002,
+      "loss": 1.232,
+      "step": 6340
+    },
+    {
+      "epoch": 5.342869162810265,
+      "grad_norm": 0.9942235350608826,
+      "learning_rate": 0.0002,
+      "loss": 1.263,
+      "step": 6350
+    },
+    {
+      "epoch": 5.351283129995793,
+      "grad_norm": 0.9535723924636841,
+      "learning_rate": 0.0002,
+      "loss": 1.3032,
+      "step": 6360
+    },
+    {
+      "epoch": 5.359697097181321,
+      "grad_norm": 0.9020892381668091,
+      "learning_rate": 0.0002,
+      "loss": 1.2908,
+      "step": 6370
+    },
+    {
+      "epoch": 5.368111064366849,
+      "grad_norm": 1.0626472234725952,
+      "learning_rate": 0.0002,
+      "loss": 1.2023,
+      "step": 6380
+    },
+    {
+      "epoch": 5.376525031552377,
+      "grad_norm": 1.1395848989486694,
+      "learning_rate": 0.0002,
+      "loss": 1.2555,
+      "step": 6390
+    },
+    {
+      "epoch": 5.384938998737905,
+      "grad_norm": 0.9274451732635498,
+      "learning_rate": 0.0002,
+      "loss": 1.2839,
+      "step": 6400
+    },
+    {
+      "epoch": 5.393352965923433,
+      "grad_norm": 0.8108699917793274,
+      "learning_rate": 0.0002,
+      "loss": 1.2819,
+      "step": 6410
+    },
+    {
+      "epoch": 5.401766933108961,
+      "grad_norm": 1.1805564165115356,
+      "learning_rate": 0.0002,
+      "loss": 1.2589,
+      "step": 6420
+    },
+    {
+      "epoch": 5.410180900294489,
+      "grad_norm": 0.8321298360824585,
+      "learning_rate": 0.0002,
+      "loss": 1.3549,
+      "step": 6430
+    },
+    {
+      "epoch": 5.418594867480017,
+      "grad_norm": 0.8981925249099731,
+      "learning_rate": 0.0002,
+      "loss": 1.2925,
+      "step": 6440
+    },
+    {
+      "epoch": 5.4270088346655445,
+      "grad_norm": 1.0730986595153809,
+      "learning_rate": 0.0002,
+      "loss": 1.258,
+      "step": 6450
+    },
+    {
+      "epoch": 5.435422801851073,
+      "grad_norm": 1.0584609508514404,
+      "learning_rate": 0.0002,
+      "loss": 1.26,
+      "step": 6460
+    },
+    {
+      "epoch": 5.443836769036601,
+      "grad_norm": 1.0792299509048462,
+      "learning_rate": 0.0002,
+      "loss": 1.2847,
+      "step": 6470
+    },
+    {
+      "epoch": 5.452250736222129,
+      "grad_norm": 0.9101872444152832,
+      "learning_rate": 0.0002,
+      "loss": 1.2035,
+      "step": 6480
+    },
+    {
+      "epoch": 5.460664703407657,
+      "grad_norm": 0.9910100698471069,
+      "learning_rate": 0.0002,
+      "loss": 1.2574,
+      "step": 6490
+    },
+    {
+      "epoch": 5.469078670593185,
+      "grad_norm": 1.041412353515625,
+      "learning_rate": 0.0002,
+      "loss": 1.3098,
+      "step": 6500
+    },
+    {
+      "epoch": 5.477492637778712,
+      "grad_norm": 1.0091687440872192,
+      "learning_rate": 0.0002,
+      "loss": 1.2812,
+      "step": 6510
+    },
+    {
+      "epoch": 5.48590660496424,
+      "grad_norm": 0.8755383491516113,
+      "learning_rate": 0.0002,
+      "loss": 1.2523,
+      "step": 6520
+    },
+    {
+      "epoch": 5.4943205721497685,
+      "grad_norm": 0.980212390422821,
+      "learning_rate": 0.0002,
+      "loss": 1.3042,
+      "step": 6530
+    },
+    {
+      "epoch": 5.502734539335297,
+      "grad_norm": 0.9356869459152222,
+      "learning_rate": 0.0002,
+      "loss": 1.2873,
+      "step": 6540
+    },
+    {
+      "epoch": 5.511148506520825,
+      "grad_norm": 0.9008095264434814,
+      "learning_rate": 0.0002,
+      "loss": 1.2254,
+      "step": 6550
+    },
+    {
+      "epoch": 5.519562473706353,
+      "grad_norm": 0.8908938765525818,
+      "learning_rate": 0.0002,
+      "loss": 1.2818,
+      "step": 6560
+    },
+    {
+      "epoch": 5.52797644089188,
+      "grad_norm": 1.1423932313919067,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 6570
+    },
+    {
+      "epoch": 5.536390408077408,
+      "grad_norm": 1.0508161783218384,
+      "learning_rate": 0.0002,
+      "loss": 1.3039,
+      "step": 6580
+    },
+    {
+      "epoch": 5.544804375262936,
+      "grad_norm": 0.8357517719268799,
+      "learning_rate": 0.0002,
+      "loss": 1.2446,
+      "step": 6590
+    },
+    {
+      "epoch": 5.5532183424484645,
+      "grad_norm": 0.9892540574073792,
+      "learning_rate": 0.0002,
+      "loss": 1.3037,
+      "step": 6600
+    },
+    {
+      "epoch": 5.561632309633993,
+      "grad_norm": 1.0048326253890991,
+      "learning_rate": 0.0002,
+      "loss": 1.3028,
+      "step": 6610
+    },
+    {
+      "epoch": 5.570046276819521,
+      "grad_norm": 0.9801995158195496,
+      "learning_rate": 0.0002,
+      "loss": 1.2152,
+      "step": 6620
+    },
+    {
+      "epoch": 5.578460244005049,
+      "grad_norm": 0.9899214506149292,
+      "learning_rate": 0.0002,
+      "loss": 1.2606,
+      "step": 6630
+    },
+    {
+      "epoch": 5.586874211190576,
+      "grad_norm": 1.1911814212799072,
+      "learning_rate": 0.0002,
+      "loss": 1.2043,
+      "step": 6640
+    },
+    {
+      "epoch": 5.595288178376104,
+      "grad_norm": 1.0368894338607788,
+      "learning_rate": 0.0002,
+      "loss": 1.3458,
+      "step": 6650
+    },
+    {
+      "epoch": 5.603702145561632,
+      "grad_norm": 1.1248382329940796,
+      "learning_rate": 0.0002,
+      "loss": 1.2595,
+      "step": 6660
+    },
+    {
+      "epoch": 5.61211611274716,
+      "grad_norm": 0.9765539765357971,
+      "learning_rate": 0.0002,
+      "loss": 1.2548,
+      "step": 6670
+    },
+    {
+      "epoch": 5.6205300799326885,
+      "grad_norm": 0.9810206890106201,
+      "learning_rate": 0.0002,
+      "loss": 1.3451,
+      "step": 6680
+    },
+    {
+      "epoch": 5.628944047118217,
+      "grad_norm": 1.100386619567871,
+      "learning_rate": 0.0002,
+      "loss": 1.2952,
+      "step": 6690
+    },
+    {
+      "epoch": 5.637358014303744,
+      "grad_norm": 0.8824519515037537,
+      "learning_rate": 0.0002,
+      "loss": 1.2467,
+      "step": 6700
+    },
+    {
+      "epoch": 5.645771981489272,
+      "grad_norm": 1.0864064693450928,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 6710
+    },
+    {
+      "epoch": 5.6541859486748,
+      "grad_norm": 1.1614511013031006,
+      "learning_rate": 0.0002,
+      "loss": 1.2479,
+      "step": 6720
+    },
+    {
+      "epoch": 5.662599915860328,
+      "grad_norm": 1.0762972831726074,
+      "learning_rate": 0.0002,
+      "loss": 1.2753,
+      "step": 6730
+    },
+    {
+      "epoch": 5.671013883045856,
+      "grad_norm": 0.9408974647521973,
+      "learning_rate": 0.0002,
+      "loss": 1.2741,
+      "step": 6740
+    },
+    {
+      "epoch": 5.679427850231384,
+      "grad_norm": 0.8906030058860779,
+      "learning_rate": 0.0002,
+      "loss": 1.2431,
+      "step": 6750
+    },
+    {
+      "epoch": 5.687841817416912,
+      "grad_norm": 0.9527303576469421,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 6760
+    },
+    {
+      "epoch": 5.69625578460244,
+      "grad_norm": 0.9471196532249451,
+      "learning_rate": 0.0002,
+      "loss": 1.322,
+      "step": 6770
+    },
+    {
+      "epoch": 5.704669751787968,
+      "grad_norm": 0.9186838865280151,
+      "learning_rate": 0.0002,
+      "loss": 1.2514,
+      "step": 6780
+    },
+    {
+      "epoch": 5.713083718973496,
+      "grad_norm": 0.9225441813468933,
+      "learning_rate": 0.0002,
+      "loss": 1.2347,
+      "step": 6790
+    },
+    {
+      "epoch": 5.721497686159024,
+      "grad_norm": 0.9712982773780823,
+      "learning_rate": 0.0002,
+      "loss": 1.1849,
+      "step": 6800
+    },
+    {
+      "epoch": 5.729911653344552,
+      "grad_norm": 1.0743170976638794,
+      "learning_rate": 0.0002,
+      "loss": 1.2431,
+      "step": 6810
+    },
+    {
+      "epoch": 5.73832562053008,
+      "grad_norm": 1.2738113403320312,
+      "learning_rate": 0.0002,
+      "loss": 1.2136,
+      "step": 6820
+    },
+    {
+      "epoch": 5.7467395877156076,
+      "grad_norm": 0.9386790990829468,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 6830
+    },
+    {
+      "epoch": 5.755153554901136,
+      "grad_norm": 1.0817769765853882,
+      "learning_rate": 0.0002,
+      "loss": 1.285,
+      "step": 6840
+    },
+    {
+      "epoch": 5.763567522086664,
+      "grad_norm": 1.1040263175964355,
+      "learning_rate": 0.0002,
+      "loss": 1.2247,
+      "step": 6850
+    },
+    {
+      "epoch": 5.771981489272192,
+      "grad_norm": 1.0656492710113525,
+      "learning_rate": 0.0002,
+      "loss": 1.2507,
+      "step": 6860
+    },
+    {
+      "epoch": 5.78039545645772,
+      "grad_norm": 0.9550157189369202,
+      "learning_rate": 0.0002,
+      "loss": 1.2999,
+      "step": 6870
+    },
+    {
+      "epoch": 5.788809423643248,
+      "grad_norm": 1.0130870342254639,
+      "learning_rate": 0.0002,
+      "loss": 1.3201,
+      "step": 6880
+    },
+    {
+      "epoch": 5.797223390828776,
+      "grad_norm": 1.0675787925720215,
+      "learning_rate": 0.0002,
+      "loss": 1.3392,
+      "step": 6890
+    },
+    {
+      "epoch": 5.8056373580143035,
+      "grad_norm": 0.9537774920463562,
+      "learning_rate": 0.0002,
+      "loss": 1.2949,
+      "step": 6900
+    },
+    {
+      "epoch": 5.814051325199832,
+      "grad_norm": 0.9640319347381592,
+      "learning_rate": 0.0002,
+      "loss": 1.2658,
+      "step": 6910
+    },
+    {
+      "epoch": 5.82246529238536,
+      "grad_norm": 0.8917992115020752,
+      "learning_rate": 0.0002,
+      "loss": 1.2199,
+      "step": 6920
+    },
+    {
+      "epoch": 5.830879259570888,
+      "grad_norm": 0.9881822466850281,
+      "learning_rate": 0.0002,
+      "loss": 1.373,
+      "step": 6930
+    },
+    {
+      "epoch": 5.839293226756416,
+      "grad_norm": 0.9136882424354553,
+      "learning_rate": 0.0002,
+      "loss": 1.323,
+      "step": 6940
+    },
+    {
+      "epoch": 5.847707193941943,
+      "grad_norm": 0.9086098074913025,
+      "learning_rate": 0.0002,
+      "loss": 1.3159,
+      "step": 6950
+    },
+    {
+      "epoch": 5.856121161127471,
+      "grad_norm": 0.9443018436431885,
+      "learning_rate": 0.0002,
+      "loss": 1.2624,
+      "step": 6960
+    },
+    {
+      "epoch": 5.864535128312999,
+      "grad_norm": 0.9915381669998169,
+      "learning_rate": 0.0002,
+      "loss": 1.3224,
+      "step": 6970
+    },
+    {
+      "epoch": 5.8729490954985275,
+      "grad_norm": 0.8939146995544434,
+      "learning_rate": 0.0002,
+      "loss": 1.337,
+      "step": 6980
+    },
+    {
+      "epoch": 5.881363062684056,
+      "grad_norm": 1.3672245740890503,
+      "learning_rate": 0.0002,
+      "loss": 1.2611,
+      "step": 6990
+    },
+    {
+      "epoch": 5.889777029869584,
+      "grad_norm": 1.0116257667541504,
+      "learning_rate": 0.0002,
+      "loss": 1.3012,
+      "step": 7000
+    },
+    {
+      "epoch": 5.898190997055112,
+      "grad_norm": 1.1561565399169922,
+      "learning_rate": 0.0002,
+      "loss": 1.3128,
+      "step": 7010
+    },
+    {
+      "epoch": 5.906604964240639,
+      "grad_norm": 0.9900678992271423,
+      "learning_rate": 0.0002,
+      "loss": 1.2301,
+      "step": 7020
+    },
+    {
+      "epoch": 5.915018931426167,
+      "grad_norm": 0.9297345876693726,
+      "learning_rate": 0.0002,
+      "loss": 1.2845,
+      "step": 7030
+    },
+    {
+      "epoch": 5.923432898611695,
+      "grad_norm": 0.9357825517654419,
+      "learning_rate": 0.0002,
+      "loss": 1.2317,
+      "step": 7040
+    },
+    {
+      "epoch": 5.931846865797223,
+      "grad_norm": 1.049317717552185,
+      "learning_rate": 0.0002,
+      "loss": 1.2303,
+      "step": 7050
+    },
+    {
+      "epoch": 5.9402608329827515,
+      "grad_norm": 0.950633704662323,
+      "learning_rate": 0.0002,
+      "loss": 1.3243,
+      "step": 7060
+    },
+    {
+      "epoch": 5.94867480016828,
+      "grad_norm": 0.854581892490387,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 7070
+    },
+    {
+      "epoch": 5.957088767353808,
+      "grad_norm": 0.9097039699554443,
+      "learning_rate": 0.0002,
+      "loss": 1.3252,
+      "step": 7080
+    },
+    {
+      "epoch": 5.965502734539335,
+      "grad_norm": 0.9072173237800598,
+      "learning_rate": 0.0002,
+      "loss": 1.291,
+      "step": 7090
+    },
+    {
+      "epoch": 5.973916701724863,
+      "grad_norm": 1.0470727682113647,
+      "learning_rate": 0.0002,
+      "loss": 1.2724,
+      "step": 7100
+    },
+    {
+      "epoch": 5.982330668910391,
+      "grad_norm": 1.2628462314605713,
+      "learning_rate": 0.0002,
+      "loss": 1.3324,
+      "step": 7110
+    },
+    {
+      "epoch": 5.990744636095919,
+      "grad_norm": 1.055279016494751,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 7120
+    },
+    {
+      "epoch": 5.9991586032814475,
+      "grad_norm": 0.966194212436676,
+      "learning_rate": 0.0002,
+      "loss": 1.3234,
+      "step": 7130
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.0427448749542236,
+      "eval_runtime": 37.8426,
+      "eval_samples_per_second": 13.609,
+      "eval_steps_per_second": 1.718,
+      "step": 7131
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 9504,
+  "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": 3.300067130443039e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/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-0.8-num-5623-sd-42/checkpoint-7131/training_args.bin b/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-0.8-num-5623-sd-42/checkpoint-7131/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f1502d478cfbb1424f707352d007b740bde5e373
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-7131/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df2b79d3acefeedef5a0229881de39ec68ef9b40046a60d7976a49f7e6b3b936
+size 5560
diff --git a/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-0.8-num-5623-sd-42/checkpoint-8319/README.md b/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-0.8-num-5623-sd-42/checkpoint-8319/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-8319/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_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-0.8-num-5623-sd-42/checkpoint-8319/adapter_config.json b/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-0.8-num-5623-sd-42/checkpoint-8319/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-8319/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_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-0.8-num-5623-sd-42/checkpoint-8319/adapter_model.safetensors b/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-0.8-num-5623-sd-42/checkpoint-8319/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5322f8a46798c9f9f585b13e82028241b9621341
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-8319/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aae23ae6783d269b6970fa7b8d2f0a1f2fab5480b7822808680363eda0810bb6
+size 109069176
diff --git a/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-0.8-num-5623-sd-42/checkpoint-8319/optimizer.pt b/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-0.8-num-5623-sd-42/checkpoint-8319/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0c06afd6b3a5a404e779911f90d39ed7473457b7
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-8319/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35efff070d4b5549eccb52fcd5c5c01015d0c90e53e56776b01fc0bfa96e3464
+size 55532666
diff --git a/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-0.8-num-5623-sd-42/checkpoint-8319/rng_state.pth b/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-0.8-num-5623-sd-42/checkpoint-8319/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f4761a05d0fc7027566474f86993a848fb9c150f
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-8319/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d29b17e02e08ea002a69cf031f4a7f63c9e6cdb9a81b89ef25bcc4d24814d5dd
+size 14244
diff --git a/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-0.8-num-5623-sd-42/checkpoint-8319/scheduler.pt b/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-0.8-num-5623-sd-42/checkpoint-8319/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f451e9b8f166df97b2436cbcc45efc9c1bf9ec87
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-8319/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d62b9f612cf8891fea8d0ca32ca5dd7d386a32839809ea28d01ed4efd8643f3f
+size 1064
diff --git a/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-0.8-num-5623-sd-42/checkpoint-8319/special_tokens_map.json b/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-0.8-num-5623-sd-42/checkpoint-8319/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-8319/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_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-0.8-num-5623-sd-42/checkpoint-8319/tokenizer.json b/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-0.8-num-5623-sd-42/checkpoint-8319/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-8319/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_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-0.8-num-5623-sd-42/checkpoint-8319/tokenizer_config.json b/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-0.8-num-5623-sd-42/checkpoint-8319/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-8319/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_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-0.8-num-5623-sd-42/checkpoint-8319/trainer_state.json b/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-0.8-num-5623-sd-42/checkpoint-8319/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..d0728e792f0d1d3c4dba5107b680cc3830e2d4ea
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-8319/trainer_state.json
@@ -0,0 +1,5906 @@
+{
+  "best_metric": 1.8055059909820557,
+  "best_model_checkpoint": "outputs-001/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-0.8-num-5623-sd-42/checkpoint-2377",
+  "epoch": 6.999579301640724,
+  "eval_steps": 10,
+  "global_step": 8319,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008413967185527976,
+      "grad_norm": 0.5458821654319763,
+      "learning_rate": 0.0002,
+      "loss": 2.56,
+      "step": 10
+    },
+    {
+      "epoch": 0.016827934371055953,
+      "grad_norm": 0.7293308973312378,
+      "learning_rate": 0.0002,
+      "loss": 2.3235,
+      "step": 20
+    },
+    {
+      "epoch": 0.02524190155658393,
+      "grad_norm": 0.47792306542396545,
+      "learning_rate": 0.0002,
+      "loss": 2.0815,
+      "step": 30
+    },
+    {
+      "epoch": 0.033655868742111905,
+      "grad_norm": 0.5944402813911438,
+      "learning_rate": 0.0002,
+      "loss": 1.9718,
+      "step": 40
+    },
+    {
+      "epoch": 0.04206983592763988,
+      "grad_norm": 0.5415359735488892,
+      "learning_rate": 0.0002,
+      "loss": 1.8848,
+      "step": 50
+    },
+    {
+      "epoch": 0.05048380311316786,
+      "grad_norm": 0.535713791847229,
+      "learning_rate": 0.0002,
+      "loss": 1.8953,
+      "step": 60
+    },
+    {
+      "epoch": 0.058897770298695834,
+      "grad_norm": 0.5184146761894226,
+      "learning_rate": 0.0002,
+      "loss": 1.937,
+      "step": 70
+    },
+    {
+      "epoch": 0.06731173748422381,
+      "grad_norm": 0.458926796913147,
+      "learning_rate": 0.0002,
+      "loss": 1.8396,
+      "step": 80
+    },
+    {
+      "epoch": 0.07572570466975179,
+      "grad_norm": 0.4780142307281494,
+      "learning_rate": 0.0002,
+      "loss": 1.8677,
+      "step": 90
+    },
+    {
+      "epoch": 0.08413967185527976,
+      "grad_norm": 0.79965740442276,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 100
+    },
+    {
+      "epoch": 0.09255363904080774,
+      "grad_norm": 0.4498862028121948,
+      "learning_rate": 0.0002,
+      "loss": 1.9081,
+      "step": 110
+    },
+    {
+      "epoch": 0.10096760622633572,
+      "grad_norm": 0.39338430762290955,
+      "learning_rate": 0.0002,
+      "loss": 1.8503,
+      "step": 120
+    },
+    {
+      "epoch": 0.10938157341186369,
+      "grad_norm": 0.9588953852653503,
+      "learning_rate": 0.0002,
+      "loss": 1.8637,
+      "step": 130
+    },
+    {
+      "epoch": 0.11779554059739167,
+      "grad_norm": 0.41675639152526855,
+      "learning_rate": 0.0002,
+      "loss": 1.8676,
+      "step": 140
+    },
+    {
+      "epoch": 0.12620950778291964,
+      "grad_norm": 0.44519832730293274,
+      "learning_rate": 0.0002,
+      "loss": 1.8904,
+      "step": 150
+    },
+    {
+      "epoch": 0.13462347496844762,
+      "grad_norm": 0.4176260530948639,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 160
+    },
+    {
+      "epoch": 0.1430374421539756,
+      "grad_norm": 0.35840365290641785,
+      "learning_rate": 0.0002,
+      "loss": 1.8398,
+      "step": 170
+    },
+    {
+      "epoch": 0.15145140933950357,
+      "grad_norm": 0.3794495463371277,
+      "learning_rate": 0.0002,
+      "loss": 1.8666,
+      "step": 180
+    },
+    {
+      "epoch": 0.15986537652503155,
+      "grad_norm": 0.4563522934913635,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 190
+    },
+    {
+      "epoch": 0.16827934371055953,
+      "grad_norm": 0.37057486176490784,
+      "learning_rate": 0.0002,
+      "loss": 1.8893,
+      "step": 200
+    },
+    {
+      "epoch": 0.1766933108960875,
+      "grad_norm": 0.44081518054008484,
+      "learning_rate": 0.0002,
+      "loss": 1.7995,
+      "step": 210
+    },
+    {
+      "epoch": 0.18510727808161548,
+      "grad_norm": 0.46078577637672424,
+      "learning_rate": 0.0002,
+      "loss": 1.9048,
+      "step": 220
+    },
+    {
+      "epoch": 0.19352124526714345,
+      "grad_norm": 0.36132094264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.8403,
+      "step": 230
+    },
+    {
+      "epoch": 0.20193521245267143,
+      "grad_norm": 0.3747289180755615,
+      "learning_rate": 0.0002,
+      "loss": 1.8827,
+      "step": 240
+    },
+    {
+      "epoch": 0.2103491796381994,
+      "grad_norm": 0.3540179133415222,
+      "learning_rate": 0.0002,
+      "loss": 1.8382,
+      "step": 250
+    },
+    {
+      "epoch": 0.21876314682372738,
+      "grad_norm": 0.3461375832557678,
+      "learning_rate": 0.0002,
+      "loss": 1.8196,
+      "step": 260
+    },
+    {
+      "epoch": 0.22717711400925536,
+      "grad_norm": 0.3436960279941559,
+      "learning_rate": 0.0002,
+      "loss": 1.8509,
+      "step": 270
+    },
+    {
+      "epoch": 0.23559108119478334,
+      "grad_norm": 0.35403719544410706,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 280
+    },
+    {
+      "epoch": 0.2440050483803113,
+      "grad_norm": 0.37142616510391235,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 290
+    },
+    {
+      "epoch": 0.2524190155658393,
+      "grad_norm": 0.3307955861091614,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 300
+    },
+    {
+      "epoch": 0.2608329827513673,
+      "grad_norm": 0.32855314016342163,
+      "learning_rate": 0.0002,
+      "loss": 1.817,
+      "step": 310
+    },
+    {
+      "epoch": 0.26924694993689524,
+      "grad_norm": 0.3299003839492798,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 320
+    },
+    {
+      "epoch": 0.27766091712242325,
+      "grad_norm": 0.44311287999153137,
+      "learning_rate": 0.0002,
+      "loss": 1.8129,
+      "step": 330
+    },
+    {
+      "epoch": 0.2860748843079512,
+      "grad_norm": 0.32989758253097534,
+      "learning_rate": 0.0002,
+      "loss": 1.8232,
+      "step": 340
+    },
+    {
+      "epoch": 0.2944888514934792,
+      "grad_norm": 0.34400200843811035,
+      "learning_rate": 0.0002,
+      "loss": 1.7716,
+      "step": 350
+    },
+    {
+      "epoch": 0.30290281867900715,
+      "grad_norm": 0.36286211013793945,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 360
+    },
+    {
+      "epoch": 0.31131678586453515,
+      "grad_norm": 0.406827837228775,
+      "learning_rate": 0.0002,
+      "loss": 1.8025,
+      "step": 370
+    },
+    {
+      "epoch": 0.3197307530500631,
+      "grad_norm": 0.36299195885658264,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 380
+    },
+    {
+      "epoch": 0.3281447202355911,
+      "grad_norm": 0.3477257192134857,
+      "learning_rate": 0.0002,
+      "loss": 1.837,
+      "step": 390
+    },
+    {
+      "epoch": 0.33655868742111905,
+      "grad_norm": 0.3730369210243225,
+      "learning_rate": 0.0002,
+      "loss": 1.7767,
+      "step": 400
+    },
+    {
+      "epoch": 0.34497265460664706,
+      "grad_norm": 0.4644559919834137,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 410
+    },
+    {
+      "epoch": 0.353386621792175,
+      "grad_norm": 0.406576544046402,
+      "learning_rate": 0.0002,
+      "loss": 1.7538,
+      "step": 420
+    },
+    {
+      "epoch": 0.361800588977703,
+      "grad_norm": 0.3612699508666992,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 430
+    },
+    {
+      "epoch": 0.37021455616323096,
+      "grad_norm": 0.3243742287158966,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 440
+    },
+    {
+      "epoch": 0.37862852334875896,
+      "grad_norm": 0.36671221256256104,
+      "learning_rate": 0.0002,
+      "loss": 1.8851,
+      "step": 450
+    },
+    {
+      "epoch": 0.3870424905342869,
+      "grad_norm": 0.3565002381801605,
+      "learning_rate": 0.0002,
+      "loss": 1.8853,
+      "step": 460
+    },
+    {
+      "epoch": 0.3954564577198149,
+      "grad_norm": 0.34630221128463745,
+      "learning_rate": 0.0002,
+      "loss": 1.8923,
+      "step": 470
+    },
+    {
+      "epoch": 0.40387042490534286,
+      "grad_norm": 0.3353537321090698,
+      "learning_rate": 0.0002,
+      "loss": 1.8234,
+      "step": 480
+    },
+    {
+      "epoch": 0.41228439209087087,
+      "grad_norm": 0.4015921950340271,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 490
+    },
+    {
+      "epoch": 0.4206983592763988,
+      "grad_norm": 0.5489419102668762,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 500
+    },
+    {
+      "epoch": 0.4291123264619268,
+      "grad_norm": 0.4193589985370636,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 510
+    },
+    {
+      "epoch": 0.43752629364745477,
+      "grad_norm": 0.3418922424316406,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 520
+    },
+    {
+      "epoch": 0.44594026083298277,
+      "grad_norm": 0.32668185234069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 530
+    },
+    {
+      "epoch": 0.4543542280185107,
+      "grad_norm": 0.3094325661659241,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 540
+    },
+    {
+      "epoch": 0.4627681952040387,
+      "grad_norm": 0.3743017315864563,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 550
+    },
+    {
+      "epoch": 0.47118216238956667,
+      "grad_norm": 0.3295630216598511,
+      "learning_rate": 0.0002,
+      "loss": 1.8451,
+      "step": 560
+    },
+    {
+      "epoch": 0.4795961295750947,
+      "grad_norm": 1.6124513149261475,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 570
+    },
+    {
+      "epoch": 0.4880100967606226,
+      "grad_norm": 0.3245585858821869,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 580
+    },
+    {
+      "epoch": 0.49642406394615063,
+      "grad_norm": 0.3332934081554413,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 590
+    },
+    {
+      "epoch": 0.5048380311316786,
+      "grad_norm": 0.3836138844490051,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 600
+    },
+    {
+      "epoch": 0.5132519983172066,
+      "grad_norm": 0.32953888177871704,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 610
+    },
+    {
+      "epoch": 0.5216659655027346,
+      "grad_norm": 0.36291512846946716,
+      "learning_rate": 0.0002,
+      "loss": 1.7729,
+      "step": 620
+    },
+    {
+      "epoch": 0.5300799326882625,
+      "grad_norm": 0.3237783908843994,
+      "learning_rate": 0.0002,
+      "loss": 1.7758,
+      "step": 630
+    },
+    {
+      "epoch": 0.5384938998737905,
+      "grad_norm": 0.38882696628570557,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 640
+    },
+    {
+      "epoch": 0.5469078670593185,
+      "grad_norm": 0.37821972370147705,
+      "learning_rate": 0.0002,
+      "loss": 1.8624,
+      "step": 650
+    },
+    {
+      "epoch": 0.5553218342448465,
+      "grad_norm": 0.3556285500526428,
+      "learning_rate": 0.0002,
+      "loss": 1.8075,
+      "step": 660
+    },
+    {
+      "epoch": 0.5637358014303744,
+      "grad_norm": 0.347499281167984,
+      "learning_rate": 0.0002,
+      "loss": 1.778,
+      "step": 670
+    },
+    {
+      "epoch": 0.5721497686159024,
+      "grad_norm": 0.3176489472389221,
+      "learning_rate": 0.0002,
+      "loss": 1.8066,
+      "step": 680
+    },
+    {
+      "epoch": 0.5805637358014304,
+      "grad_norm": 0.30220088362693787,
+      "learning_rate": 0.0002,
+      "loss": 1.7257,
+      "step": 690
+    },
+    {
+      "epoch": 0.5889777029869584,
+      "grad_norm": 0.3711601793766022,
+      "learning_rate": 0.0002,
+      "loss": 1.8415,
+      "step": 700
+    },
+    {
+      "epoch": 0.5973916701724863,
+      "grad_norm": 0.3311759829521179,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 710
+    },
+    {
+      "epoch": 0.6058056373580143,
+      "grad_norm": 0.34824270009994507,
+      "learning_rate": 0.0002,
+      "loss": 1.7712,
+      "step": 720
+    },
+    {
+      "epoch": 0.6142196045435423,
+      "grad_norm": 0.29668381810188293,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 730
+    },
+    {
+      "epoch": 0.6226335717290703,
+      "grad_norm": 0.36087489128112793,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 740
+    },
+    {
+      "epoch": 0.6310475389145982,
+      "grad_norm": 0.31590089201927185,
+      "learning_rate": 0.0002,
+      "loss": 1.7956,
+      "step": 750
+    },
+    {
+      "epoch": 0.6394615061001262,
+      "grad_norm": 0.37632957100868225,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 760
+    },
+    {
+      "epoch": 0.6478754732856542,
+      "grad_norm": 0.3360748589038849,
+      "learning_rate": 0.0002,
+      "loss": 1.8499,
+      "step": 770
+    },
+    {
+      "epoch": 0.6562894404711822,
+      "grad_norm": 0.3420640528202057,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 780
+    },
+    {
+      "epoch": 0.6647034076567101,
+      "grad_norm": 0.5734959244728088,
+      "learning_rate": 0.0002,
+      "loss": 1.8353,
+      "step": 790
+    },
+    {
+      "epoch": 0.6731173748422381,
+      "grad_norm": 0.36440837383270264,
+      "learning_rate": 0.0002,
+      "loss": 1.7746,
+      "step": 800
+    },
+    {
+      "epoch": 0.6815313420277661,
+      "grad_norm": 0.3179708421230316,
+      "learning_rate": 0.0002,
+      "loss": 1.7532,
+      "step": 810
+    },
+    {
+      "epoch": 0.6899453092132941,
+      "grad_norm": 0.34122881293296814,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 820
+    },
+    {
+      "epoch": 0.698359276398822,
+      "grad_norm": 0.31886112689971924,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 830
+    },
+    {
+      "epoch": 0.70677324358435,
+      "grad_norm": 0.31782326102256775,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 840
+    },
+    {
+      "epoch": 0.715187210769878,
+      "grad_norm": 0.36052989959716797,
+      "learning_rate": 0.0002,
+      "loss": 1.7588,
+      "step": 850
+    },
+    {
+      "epoch": 0.723601177955406,
+      "grad_norm": 0.28946155309677124,
+      "learning_rate": 0.0002,
+      "loss": 1.7891,
+      "step": 860
+    },
+    {
+      "epoch": 0.7320151451409339,
+      "grad_norm": 0.3095663785934448,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 870
+    },
+    {
+      "epoch": 0.7404291123264619,
+      "grad_norm": 0.3317491412162781,
+      "learning_rate": 0.0002,
+      "loss": 1.785,
+      "step": 880
+    },
+    {
+      "epoch": 0.7488430795119899,
+      "grad_norm": 0.31324660778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 890
+    },
+    {
+      "epoch": 0.7572570466975179,
+      "grad_norm": 0.3290475606918335,
+      "learning_rate": 0.0002,
+      "loss": 1.8753,
+      "step": 900
+    },
+    {
+      "epoch": 0.7656710138830458,
+      "grad_norm": 0.35690343379974365,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 910
+    },
+    {
+      "epoch": 0.7740849810685738,
+      "grad_norm": 0.39558273553848267,
+      "learning_rate": 0.0002,
+      "loss": 1.826,
+      "step": 920
+    },
+    {
+      "epoch": 0.7824989482541018,
+      "grad_norm": 0.34254348278045654,
+      "learning_rate": 0.0002,
+      "loss": 1.8722,
+      "step": 930
+    },
+    {
+      "epoch": 0.7909129154396298,
+      "grad_norm": 0.3560165464878082,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 940
+    },
+    {
+      "epoch": 0.7993268826251577,
+      "grad_norm": 0.30693164467811584,
+      "learning_rate": 0.0002,
+      "loss": 1.7992,
+      "step": 950
+    },
+    {
+      "epoch": 0.8077408498106857,
+      "grad_norm": 0.3394823372364044,
+      "learning_rate": 0.0002,
+      "loss": 1.8029,
+      "step": 960
+    },
+    {
+      "epoch": 0.8161548169962137,
+      "grad_norm": 0.3741514980792999,
+      "learning_rate": 0.0002,
+      "loss": 1.8105,
+      "step": 970
+    },
+    {
+      "epoch": 0.8245687841817417,
+      "grad_norm": 0.3655228316783905,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 980
+    },
+    {
+      "epoch": 0.8329827513672696,
+      "grad_norm": 0.3586033880710602,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 990
+    },
+    {
+      "epoch": 0.8413967185527976,
+      "grad_norm": 0.3459678888320923,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8498106857383256,
+      "grad_norm": 0.3184349834918976,
+      "learning_rate": 0.0002,
+      "loss": 1.8498,
+      "step": 1010
+    },
+    {
+      "epoch": 0.8582246529238536,
+      "grad_norm": 0.3099786043167114,
+      "learning_rate": 0.0002,
+      "loss": 1.7632,
+      "step": 1020
+    },
+    {
+      "epoch": 0.8666386201093815,
+      "grad_norm": 0.30300915241241455,
+      "learning_rate": 0.0002,
+      "loss": 1.8067,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8750525872949095,
+      "grad_norm": 0.3128705620765686,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8834665544804375,
+      "grad_norm": 0.3336263597011566,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8918805216659655,
+      "grad_norm": 0.3801328241825104,
+      "learning_rate": 0.0002,
+      "loss": 1.8375,
+      "step": 1060
+    },
+    {
+      "epoch": 0.9002944888514934,
+      "grad_norm": 0.3122096359729767,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 1070
+    },
+    {
+      "epoch": 0.9087084560370214,
+      "grad_norm": 0.35990869998931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 1080
+    },
+    {
+      "epoch": 0.9171224232225494,
+      "grad_norm": 0.3321819305419922,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1090
+    },
+    {
+      "epoch": 0.9255363904080774,
+      "grad_norm": 0.4202139377593994,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9339503575936053,
+      "grad_norm": 0.32559722661972046,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 1110
+    },
+    {
+      "epoch": 0.9423643247791333,
+      "grad_norm": 0.3098459839820862,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 1120
+    },
+    {
+      "epoch": 0.9507782919646613,
+      "grad_norm": 0.33917108178138733,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1130
+    },
+    {
+      "epoch": 0.9591922591501894,
+      "grad_norm": 0.4055837094783783,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1140
+    },
+    {
+      "epoch": 0.9676062263357172,
+      "grad_norm": 0.32508623600006104,
+      "learning_rate": 0.0002,
+      "loss": 1.8259,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9760201935212452,
+      "grad_norm": 0.30150601267814636,
+      "learning_rate": 0.0002,
+      "loss": 1.782,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9844341607067733,
+      "grad_norm": 0.3042563199996948,
+      "learning_rate": 0.0002,
+      "loss": 1.8291,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9928481278923013,
+      "grad_norm": 0.33254584670066833,
+      "learning_rate": 0.0002,
+      "loss": 1.7847,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9995793016407236,
+      "eval_loss": 1.8077726364135742,
+      "eval_runtime": 38.4359,
+      "eval_samples_per_second": 13.399,
+      "eval_steps_per_second": 1.691,
+      "step": 1188
+    },
+    {
+      "epoch": 1.0012620950778293,
+      "grad_norm": 0.35073035955429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 1190
+    },
+    {
+      "epoch": 1.0096760622633572,
+      "grad_norm": 0.3217269778251648,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1200
+    },
+    {
+      "epoch": 1.018090029448885,
+      "grad_norm": 0.3635033369064331,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1210
+    },
+    {
+      "epoch": 1.0265039966344132,
+      "grad_norm": 0.32468414306640625,
+      "learning_rate": 0.0002,
+      "loss": 1.6949,
+      "step": 1220
+    },
+    {
+      "epoch": 1.034917963819941,
+      "grad_norm": 0.3307163417339325,
+      "learning_rate": 0.0002,
+      "loss": 1.711,
+      "step": 1230
+    },
+    {
+      "epoch": 1.0433319310054692,
+      "grad_norm": 0.34381359815597534,
+      "learning_rate": 0.0002,
+      "loss": 1.7881,
+      "step": 1240
+    },
+    {
+      "epoch": 1.051745898190997,
+      "grad_norm": 0.35874804854393005,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 1250
+    },
+    {
+      "epoch": 1.060159865376525,
+      "grad_norm": 0.3615919351577759,
+      "learning_rate": 0.0002,
+      "loss": 1.7314,
+      "step": 1260
+    },
+    {
+      "epoch": 1.068573832562053,
+      "grad_norm": 0.32835808396339417,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1270
+    },
+    {
+      "epoch": 1.076987799747581,
+      "grad_norm": 0.3876388370990753,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0854017669331089,
+      "grad_norm": 0.39895930886268616,
+      "learning_rate": 0.0002,
+      "loss": 1.7442,
+      "step": 1290
+    },
+    {
+      "epoch": 1.093815734118637,
+      "grad_norm": 0.39081698656082153,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 1300
+    },
+    {
+      "epoch": 1.1022297013041649,
+      "grad_norm": 0.39974215626716614,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1310
+    },
+    {
+      "epoch": 1.110643668489693,
+      "grad_norm": 0.3887332081794739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1320
+    },
+    {
+      "epoch": 1.1190576356752209,
+      "grad_norm": 0.36216408014297485,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 1330
+    },
+    {
+      "epoch": 1.1274716028607488,
+      "grad_norm": 0.36979028582572937,
+      "learning_rate": 0.0002,
+      "loss": 1.762,
+      "step": 1340
+    },
+    {
+      "epoch": 1.1358855700462769,
+      "grad_norm": 0.34052133560180664,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 1350
+    },
+    {
+      "epoch": 1.1442995372318048,
+      "grad_norm": 0.3467716574668884,
+      "learning_rate": 0.0002,
+      "loss": 1.7513,
+      "step": 1360
+    },
+    {
+      "epoch": 1.1527135044173327,
+      "grad_norm": 0.35528799891471863,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1370
+    },
+    {
+      "epoch": 1.1611274716028608,
+      "grad_norm": 0.36282262206077576,
+      "learning_rate": 0.0002,
+      "loss": 1.794,
+      "step": 1380
+    },
+    {
+      "epoch": 1.1695414387883887,
+      "grad_norm": 0.37355899810791016,
+      "learning_rate": 0.0002,
+      "loss": 1.7731,
+      "step": 1390
+    },
+    {
+      "epoch": 1.1779554059739168,
+      "grad_norm": 0.37292736768722534,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1400
+    },
+    {
+      "epoch": 1.1863693731594447,
+      "grad_norm": 0.5892812013626099,
+      "learning_rate": 0.0002,
+      "loss": 1.6916,
+      "step": 1410
+    },
+    {
+      "epoch": 1.1947833403449726,
+      "grad_norm": 0.3712292015552521,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 1420
+    },
+    {
+      "epoch": 1.2031973075305007,
+      "grad_norm": 0.3349577486515045,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1430
+    },
+    {
+      "epoch": 1.2116112747160286,
+      "grad_norm": 0.32591062784194946,
+      "learning_rate": 0.0002,
+      "loss": 1.7412,
+      "step": 1440
+    },
+    {
+      "epoch": 1.2200252419015567,
+      "grad_norm": 0.3840635418891907,
+      "learning_rate": 0.0002,
+      "loss": 1.7406,
+      "step": 1450
+    },
+    {
+      "epoch": 1.2284392090870846,
+      "grad_norm": 0.37238365411758423,
+      "learning_rate": 0.0002,
+      "loss": 1.7276,
+      "step": 1460
+    },
+    {
+      "epoch": 1.2368531762726125,
+      "grad_norm": 0.3731217682361603,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 1470
+    },
+    {
+      "epoch": 1.2452671434581406,
+      "grad_norm": 0.3318967819213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7255,
+      "step": 1480
+    },
+    {
+      "epoch": 1.2536811106436685,
+      "grad_norm": 0.3784034848213196,
+      "learning_rate": 0.0002,
+      "loss": 1.7463,
+      "step": 1490
+    },
+    {
+      "epoch": 1.2620950778291964,
+      "grad_norm": 0.3541383147239685,
+      "learning_rate": 0.0002,
+      "loss": 1.6862,
+      "step": 1500
+    },
+    {
+      "epoch": 1.2705090450147245,
+      "grad_norm": 0.35312485694885254,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1510
+    },
+    {
+      "epoch": 1.2789230122002524,
+      "grad_norm": 0.35272929072380066,
+      "learning_rate": 0.0002,
+      "loss": 1.7029,
+      "step": 1520
+    },
+    {
+      "epoch": 1.2873369793857803,
+      "grad_norm": 0.40988272428512573,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1530
+    },
+    {
+      "epoch": 1.2957509465713084,
+      "grad_norm": 0.3543946146965027,
+      "learning_rate": 0.0002,
+      "loss": 1.6912,
+      "step": 1540
+    },
+    {
+      "epoch": 1.3041649137568363,
+      "grad_norm": 0.35639145970344543,
+      "learning_rate": 0.0002,
+      "loss": 1.6757,
+      "step": 1550
+    },
+    {
+      "epoch": 1.3125788809423642,
+      "grad_norm": 0.3290826678276062,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 1560
+    },
+    {
+      "epoch": 1.3209928481278923,
+      "grad_norm": 0.39264336228370667,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 1570
+    },
+    {
+      "epoch": 1.3294068153134202,
+      "grad_norm": 0.5390415191650391,
+      "learning_rate": 0.0002,
+      "loss": 1.6804,
+      "step": 1580
+    },
+    {
+      "epoch": 1.3378207824989483,
+      "grad_norm": 0.5188116431236267,
+      "learning_rate": 0.0002,
+      "loss": 1.708,
+      "step": 1590
+    },
+    {
+      "epoch": 1.3462347496844762,
+      "grad_norm": 0.37445148825645447,
+      "learning_rate": 0.0002,
+      "loss": 1.6763,
+      "step": 1600
+    },
+    {
+      "epoch": 1.3546487168700043,
+      "grad_norm": 0.3296085298061371,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1610
+    },
+    {
+      "epoch": 1.3630626840555322,
+      "grad_norm": 0.39879581332206726,
+      "learning_rate": 0.0002,
+      "loss": 1.8107,
+      "step": 1620
+    },
+    {
+      "epoch": 1.37147665124106,
+      "grad_norm": 0.36092764139175415,
+      "learning_rate": 0.0002,
+      "loss": 1.6744,
+      "step": 1630
+    },
+    {
+      "epoch": 1.3798906184265882,
+      "grad_norm": 0.37011823058128357,
+      "learning_rate": 0.0002,
+      "loss": 1.7144,
+      "step": 1640
+    },
+    {
+      "epoch": 1.3883045856121161,
+      "grad_norm": 0.40863534808158875,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1650
+    },
+    {
+      "epoch": 1.396718552797644,
+      "grad_norm": 0.337001770734787,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 1660
+    },
+    {
+      "epoch": 1.4051325199831721,
+      "grad_norm": 0.35596707463264465,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1670
+    },
+    {
+      "epoch": 1.4135464871687,
+      "grad_norm": 0.3857671916484833,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 1680
+    },
+    {
+      "epoch": 1.421960454354228,
+      "grad_norm": 0.419502317905426,
+      "learning_rate": 0.0002,
+      "loss": 1.7015,
+      "step": 1690
+    },
+    {
+      "epoch": 1.430374421539756,
+      "grad_norm": 0.35459452867507935,
+      "learning_rate": 0.0002,
+      "loss": 1.7261,
+      "step": 1700
+    },
+    {
+      "epoch": 1.438788388725284,
+      "grad_norm": 0.37246978282928467,
+      "learning_rate": 0.0002,
+      "loss": 1.7361,
+      "step": 1710
+    },
+    {
+      "epoch": 1.4472023559108118,
+      "grad_norm": 0.33091893792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 1720
+    },
+    {
+      "epoch": 1.45561632309634,
+      "grad_norm": 0.37029674649238586,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1730
+    },
+    {
+      "epoch": 1.4640302902818678,
+      "grad_norm": 0.374025821685791,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1740
+    },
+    {
+      "epoch": 1.472444257467396,
+      "grad_norm": 0.3416315019130707,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1750
+    },
+    {
+      "epoch": 1.4808582246529238,
+      "grad_norm": 0.36502841114997864,
+      "learning_rate": 0.0002,
+      "loss": 1.7093,
+      "step": 1760
+    },
+    {
+      "epoch": 1.489272191838452,
+      "grad_norm": 0.35458803176879883,
+      "learning_rate": 0.0002,
+      "loss": 1.6597,
+      "step": 1770
+    },
+    {
+      "epoch": 1.4976861590239798,
+      "grad_norm": 0.4462839663028717,
+      "learning_rate": 0.0002,
+      "loss": 1.675,
+      "step": 1780
+    },
+    {
+      "epoch": 1.5061001262095077,
+      "grad_norm": 0.34836092591285706,
+      "learning_rate": 0.0002,
+      "loss": 1.7267,
+      "step": 1790
+    },
+    {
+      "epoch": 1.5145140933950358,
+      "grad_norm": 0.3445749282836914,
+      "learning_rate": 0.0002,
+      "loss": 1.7295,
+      "step": 1800
+    },
+    {
+      "epoch": 1.5229280605805637,
+      "grad_norm": 0.36012160778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1810
+    },
+    {
+      "epoch": 1.5313420277660916,
+      "grad_norm": 0.4052616059780121,
+      "learning_rate": 0.0002,
+      "loss": 1.6594,
+      "step": 1820
+    },
+    {
+      "epoch": 1.5397559949516197,
+      "grad_norm": 0.3966905474662781,
+      "learning_rate": 0.0002,
+      "loss": 1.72,
+      "step": 1830
+    },
+    {
+      "epoch": 1.5481699621371476,
+      "grad_norm": 0.35028719902038574,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1840
+    },
+    {
+      "epoch": 1.5565839293226755,
+      "grad_norm": 0.3936742842197418,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1850
+    },
+    {
+      "epoch": 1.5649978965082036,
+      "grad_norm": 0.34473296999931335,
+      "learning_rate": 0.0002,
+      "loss": 1.7579,
+      "step": 1860
+    },
+    {
+      "epoch": 1.5734118636937318,
+      "grad_norm": 0.4328365623950958,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1870
+    },
+    {
+      "epoch": 1.5818258308792594,
+      "grad_norm": 0.3566315472126007,
+      "learning_rate": 0.0002,
+      "loss": 1.7098,
+      "step": 1880
+    },
+    {
+      "epoch": 1.5902397980647875,
+      "grad_norm": 0.3301256597042084,
+      "learning_rate": 0.0002,
+      "loss": 1.6095,
+      "step": 1890
+    },
+    {
+      "epoch": 1.5986537652503157,
+      "grad_norm": 0.3743041455745697,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 1900
+    },
+    {
+      "epoch": 1.6070677324358436,
+      "grad_norm": 0.3735344707965851,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 1910
+    },
+    {
+      "epoch": 1.6154816996213714,
+      "grad_norm": 0.42191144824028015,
+      "learning_rate": 0.0002,
+      "loss": 1.7445,
+      "step": 1920
+    },
+    {
+      "epoch": 1.6238956668068996,
+      "grad_norm": 0.3787207305431366,
+      "learning_rate": 0.0002,
+      "loss": 1.6978,
+      "step": 1930
+    },
+    {
+      "epoch": 1.6323096339924275,
+      "grad_norm": 0.35647350549697876,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 1940
+    },
+    {
+      "epoch": 1.6407236011779553,
+      "grad_norm": 0.39791446924209595,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 1950
+    },
+    {
+      "epoch": 1.6491375683634835,
+      "grad_norm": 0.37341275811195374,
+      "learning_rate": 0.0002,
+      "loss": 1.7293,
+      "step": 1960
+    },
+    {
+      "epoch": 1.6575515355490114,
+      "grad_norm": 0.3722686469554901,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1970
+    },
+    {
+      "epoch": 1.6659655027345392,
+      "grad_norm": 0.37467387318611145,
+      "learning_rate": 0.0002,
+      "loss": 1.6383,
+      "step": 1980
+    },
+    {
+      "epoch": 1.6743794699200674,
+      "grad_norm": 0.37109461426734924,
+      "learning_rate": 0.0002,
+      "loss": 1.7439,
+      "step": 1990
+    },
+    {
+      "epoch": 1.6827934371055953,
+      "grad_norm": 0.4008837044239044,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 2000
+    },
+    {
+      "epoch": 1.6912074042911232,
+      "grad_norm": 0.3316999673843384,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2010
+    },
+    {
+      "epoch": 1.6996213714766513,
+      "grad_norm": 0.3683805465698242,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 2020
+    },
+    {
+      "epoch": 1.7080353386621794,
+      "grad_norm": 0.4163658320903778,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 2030
+    },
+    {
+      "epoch": 1.716449305847707,
+      "grad_norm": 0.4245431125164032,
+      "learning_rate": 0.0002,
+      "loss": 1.741,
+      "step": 2040
+    },
+    {
+      "epoch": 1.7248632730332352,
+      "grad_norm": 0.36732038855552673,
+      "learning_rate": 0.0002,
+      "loss": 1.7184,
+      "step": 2050
+    },
+    {
+      "epoch": 1.7332772402187633,
+      "grad_norm": 0.34981656074523926,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 2060
+    },
+    {
+      "epoch": 1.7416912074042912,
+      "grad_norm": 0.38588812947273254,
+      "learning_rate": 0.0002,
+      "loss": 1.7545,
+      "step": 2070
+    },
+    {
+      "epoch": 1.750105174589819,
+      "grad_norm": 0.39914557337760925,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 2080
+    },
+    {
+      "epoch": 1.7585191417753472,
+      "grad_norm": 0.36068692803382874,
+      "learning_rate": 0.0002,
+      "loss": 1.7049,
+      "step": 2090
+    },
+    {
+      "epoch": 1.766933108960875,
+      "grad_norm": 0.3983287215232849,
+      "learning_rate": 0.0002,
+      "loss": 1.7537,
+      "step": 2100
+    },
+    {
+      "epoch": 1.775347076146403,
+      "grad_norm": 0.45008400082588196,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2110
+    },
+    {
+      "epoch": 1.783761043331931,
+      "grad_norm": 0.3618052303791046,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 2120
+    },
+    {
+      "epoch": 1.792175010517459,
+      "grad_norm": 0.38745400309562683,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 2130
+    },
+    {
+      "epoch": 1.8005889777029869,
+      "grad_norm": 0.3413826525211334,
+      "learning_rate": 0.0002,
+      "loss": 1.7387,
+      "step": 2140
+    },
+    {
+      "epoch": 1.809002944888515,
+      "grad_norm": 0.35983747243881226,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 2150
+    },
+    {
+      "epoch": 1.8174169120740429,
+      "grad_norm": 0.40926849842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 2160
+    },
+    {
+      "epoch": 1.8258308792595708,
+      "grad_norm": 0.3543093800544739,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2170
+    },
+    {
+      "epoch": 1.8342448464450989,
+      "grad_norm": 0.42690935730934143,
+      "learning_rate": 0.0002,
+      "loss": 1.7812,
+      "step": 2180
+    },
+    {
+      "epoch": 1.842658813630627,
+      "grad_norm": 0.40282756090164185,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 2190
+    },
+    {
+      "epoch": 1.8510727808161547,
+      "grad_norm": 0.36568400263786316,
+      "learning_rate": 0.0002,
+      "loss": 1.7411,
+      "step": 2200
+    },
+    {
+      "epoch": 1.8594867480016828,
+      "grad_norm": 0.43159013986587524,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 2210
+    },
+    {
+      "epoch": 1.867900715187211,
+      "grad_norm": 0.3554118573665619,
+      "learning_rate": 0.0002,
+      "loss": 1.7298,
+      "step": 2220
+    },
+    {
+      "epoch": 1.8763146823727388,
+      "grad_norm": 0.43349072337150574,
+      "learning_rate": 0.0002,
+      "loss": 1.7157,
+      "step": 2230
+    },
+    {
+      "epoch": 1.8847286495582667,
+      "grad_norm": 0.36486536264419556,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 2240
+    },
+    {
+      "epoch": 1.8931426167437948,
+      "grad_norm": 0.39260047674179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6901,
+      "step": 2250
+    },
+    {
+      "epoch": 1.9015565839293227,
+      "grad_norm": 0.3741776943206787,
+      "learning_rate": 0.0002,
+      "loss": 1.6691,
+      "step": 2260
+    },
+    {
+      "epoch": 1.9099705511148506,
+      "grad_norm": 0.3961946964263916,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2270
+    },
+    {
+      "epoch": 1.9183845183003787,
+      "grad_norm": 0.3659731149673462,
+      "learning_rate": 0.0002,
+      "loss": 1.737,
+      "step": 2280
+    },
+    {
+      "epoch": 1.9267984854859066,
+      "grad_norm": 0.34744107723236084,
+      "learning_rate": 0.0002,
+      "loss": 1.7342,
+      "step": 2290
+    },
+    {
+      "epoch": 1.9352124526714345,
+      "grad_norm": 0.3607442378997803,
+      "learning_rate": 0.0002,
+      "loss": 1.7162,
+      "step": 2300
+    },
+    {
+      "epoch": 1.9436264198569626,
+      "grad_norm": 0.331464558839798,
+      "learning_rate": 0.0002,
+      "loss": 1.6673,
+      "step": 2310
+    },
+    {
+      "epoch": 1.9520403870424905,
+      "grad_norm": 0.3904414474964142,
+      "learning_rate": 0.0002,
+      "loss": 1.7101,
+      "step": 2320
+    },
+    {
+      "epoch": 1.9604543542280184,
+      "grad_norm": 0.37584832310676575,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 2330
+    },
+    {
+      "epoch": 1.9688683214135465,
+      "grad_norm": 0.3698684275150299,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 2340
+    },
+    {
+      "epoch": 1.9772822885990746,
+      "grad_norm": 0.40571412444114685,
+      "learning_rate": 0.0002,
+      "loss": 1.7764,
+      "step": 2350
+    },
+    {
+      "epoch": 1.9856962557846023,
+      "grad_norm": 0.40059587359428406,
+      "learning_rate": 0.0002,
+      "loss": 1.744,
+      "step": 2360
+    },
+    {
+      "epoch": 1.9941102229701304,
+      "grad_norm": 0.4168248474597931,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2370
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8055059909820557,
+      "eval_runtime": 38.422,
+      "eval_samples_per_second": 13.404,
+      "eval_steps_per_second": 1.692,
+      "step": 2377
+    },
+    {
+      "epoch": 2.0025241901556585,
+      "grad_norm": 0.35205352306365967,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 2380
+    },
+    {
+      "epoch": 2.010938157341186,
+      "grad_norm": 0.3979377746582031,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 2390
+    },
+    {
+      "epoch": 2.0193521245267143,
+      "grad_norm": 0.396491676568985,
+      "learning_rate": 0.0002,
+      "loss": 1.6421,
+      "step": 2400
+    },
+    {
+      "epoch": 2.0277660917122424,
+      "grad_norm": 0.44712209701538086,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2410
+    },
+    {
+      "epoch": 2.03618005889777,
+      "grad_norm": 0.4454420208930969,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 2420
+    },
+    {
+      "epoch": 2.044594026083298,
+      "grad_norm": 0.4170038402080536,
+      "learning_rate": 0.0002,
+      "loss": 1.6635,
+      "step": 2430
+    },
+    {
+      "epoch": 2.0530079932688263,
+      "grad_norm": 0.4309595227241516,
+      "learning_rate": 0.0002,
+      "loss": 1.6512,
+      "step": 2440
+    },
+    {
+      "epoch": 2.0614219604543544,
+      "grad_norm": 0.4241602122783661,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 2450
+    },
+    {
+      "epoch": 2.069835927639882,
+      "grad_norm": 0.4370540678501129,
+      "learning_rate": 0.0002,
+      "loss": 1.6162,
+      "step": 2460
+    },
+    {
+      "epoch": 2.0782498948254102,
+      "grad_norm": 0.43985554575920105,
+      "learning_rate": 0.0002,
+      "loss": 1.6354,
+      "step": 2470
+    },
+    {
+      "epoch": 2.0866638620109383,
+      "grad_norm": 0.4158105254173279,
+      "learning_rate": 0.0002,
+      "loss": 1.6954,
+      "step": 2480
+    },
+    {
+      "epoch": 2.095077829196466,
+      "grad_norm": 0.441549152135849,
+      "learning_rate": 0.0002,
+      "loss": 1.6114,
+      "step": 2490
+    },
+    {
+      "epoch": 2.103491796381994,
+      "grad_norm": 0.385718435049057,
+      "learning_rate": 0.0002,
+      "loss": 1.5485,
+      "step": 2500
+    },
+    {
+      "epoch": 2.1119057635675222,
+      "grad_norm": 0.43146514892578125,
+      "learning_rate": 0.0002,
+      "loss": 1.5894,
+      "step": 2510
+    },
+    {
+      "epoch": 2.12031973075305,
+      "grad_norm": 0.41663315892219543,
+      "learning_rate": 0.0002,
+      "loss": 1.6414,
+      "step": 2520
+    },
+    {
+      "epoch": 2.128733697938578,
+      "grad_norm": 0.4410698115825653,
+      "learning_rate": 0.0002,
+      "loss": 1.6527,
+      "step": 2530
+    },
+    {
+      "epoch": 2.137147665124106,
+      "grad_norm": 0.4472278952598572,
+      "learning_rate": 0.0002,
+      "loss": 1.6124,
+      "step": 2540
+    },
+    {
+      "epoch": 2.145561632309634,
+      "grad_norm": 0.3879167437553406,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2550
+    },
+    {
+      "epoch": 2.153975599495162,
+      "grad_norm": 0.4212203025817871,
+      "learning_rate": 0.0002,
+      "loss": 1.6682,
+      "step": 2560
+    },
+    {
+      "epoch": 2.16238956668069,
+      "grad_norm": 0.42841723561286926,
+      "learning_rate": 0.0002,
+      "loss": 1.6036,
+      "step": 2570
+    },
+    {
+      "epoch": 2.1708035338662177,
+      "grad_norm": 0.39272481203079224,
+      "learning_rate": 0.0002,
+      "loss": 1.5962,
+      "step": 2580
+    },
+    {
+      "epoch": 2.179217501051746,
+      "grad_norm": 0.4075261354446411,
+      "learning_rate": 0.0002,
+      "loss": 1.681,
+      "step": 2590
+    },
+    {
+      "epoch": 2.187631468237274,
+      "grad_norm": 0.5358437895774841,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 2600
+    },
+    {
+      "epoch": 2.1960454354228016,
+      "grad_norm": 0.4738350212574005,
+      "learning_rate": 0.0002,
+      "loss": 1.6423,
+      "step": 2610
+    },
+    {
+      "epoch": 2.2044594026083297,
+      "grad_norm": 0.446789026260376,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 2620
+    },
+    {
+      "epoch": 2.212873369793858,
+      "grad_norm": 0.4615374505519867,
+      "learning_rate": 0.0002,
+      "loss": 1.6246,
+      "step": 2630
+    },
+    {
+      "epoch": 2.221287336979386,
+      "grad_norm": 0.46901994943618774,
+      "learning_rate": 0.0002,
+      "loss": 1.6205,
+      "step": 2640
+    },
+    {
+      "epoch": 2.2297013041649136,
+      "grad_norm": 0.46267789602279663,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 2.2381152713504417,
+      "grad_norm": 0.4383080005645752,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2660
+    },
+    {
+      "epoch": 2.24652923853597,
+      "grad_norm": 0.4070609509944916,
+      "learning_rate": 0.0002,
+      "loss": 1.5745,
+      "step": 2670
+    },
+    {
+      "epoch": 2.2549432057214975,
+      "grad_norm": 0.4572339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.6125,
+      "step": 2680
+    },
+    {
+      "epoch": 2.2633571729070256,
+      "grad_norm": 0.393265038728714,
+      "learning_rate": 0.0002,
+      "loss": 1.5671,
+      "step": 2690
+    },
+    {
+      "epoch": 2.2717711400925538,
+      "grad_norm": 0.46144717931747437,
+      "learning_rate": 0.0002,
+      "loss": 1.6239,
+      "step": 2700
+    },
+    {
+      "epoch": 2.2801851072780814,
+      "grad_norm": 0.45077767968177795,
+      "learning_rate": 0.0002,
+      "loss": 1.5992,
+      "step": 2710
+    },
+    {
+      "epoch": 2.2885990744636096,
+      "grad_norm": 0.5697639584541321,
+      "learning_rate": 0.0002,
+      "loss": 1.6261,
+      "step": 2720
+    },
+    {
+      "epoch": 2.2970130416491377,
+      "grad_norm": 0.4855510890483856,
+      "learning_rate": 0.0002,
+      "loss": 1.6192,
+      "step": 2730
+    },
+    {
+      "epoch": 2.3054270088346653,
+      "grad_norm": 0.4440622627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 2740
+    },
+    {
+      "epoch": 2.3138409760201935,
+      "grad_norm": 0.3904096782207489,
+      "learning_rate": 0.0002,
+      "loss": 1.6496,
+      "step": 2750
+    },
+    {
+      "epoch": 2.3222549432057216,
+      "grad_norm": 0.5225510597229004,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 2760
+    },
+    {
+      "epoch": 2.3306689103912497,
+      "grad_norm": 0.44866397976875305,
+      "learning_rate": 0.0002,
+      "loss": 1.6082,
+      "step": 2770
+    },
+    {
+      "epoch": 2.3390828775767774,
+      "grad_norm": 0.5167056322097778,
+      "learning_rate": 0.0002,
+      "loss": 1.6087,
+      "step": 2780
+    },
+    {
+      "epoch": 2.3474968447623055,
+      "grad_norm": 0.45913267135620117,
+      "learning_rate": 0.0002,
+      "loss": 1.6136,
+      "step": 2790
+    },
+    {
+      "epoch": 2.3559108119478336,
+      "grad_norm": 0.45787590742111206,
+      "learning_rate": 0.0002,
+      "loss": 1.6564,
+      "step": 2800
+    },
+    {
+      "epoch": 2.3643247791333613,
+      "grad_norm": 0.4633352756500244,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2810
+    },
+    {
+      "epoch": 2.3727387463188894,
+      "grad_norm": 0.46390071511268616,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 2820
+    },
+    {
+      "epoch": 2.3811527135044175,
+      "grad_norm": 0.4261005222797394,
+      "learning_rate": 0.0002,
+      "loss": 1.6039,
+      "step": 2830
+    },
+    {
+      "epoch": 2.389566680689945,
+      "grad_norm": 0.4283634424209595,
+      "learning_rate": 0.0002,
+      "loss": 1.6364,
+      "step": 2840
+    },
+    {
+      "epoch": 2.3979806478754733,
+      "grad_norm": 0.4955291450023651,
+      "learning_rate": 0.0002,
+      "loss": 1.6382,
+      "step": 2850
+    },
+    {
+      "epoch": 2.4063946150610014,
+      "grad_norm": 0.4740189015865326,
+      "learning_rate": 0.0002,
+      "loss": 1.6173,
+      "step": 2860
+    },
+    {
+      "epoch": 2.414808582246529,
+      "grad_norm": 0.4222276508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2870
+    },
+    {
+      "epoch": 2.423222549432057,
+      "grad_norm": 0.4982149004936218,
+      "learning_rate": 0.0002,
+      "loss": 1.5602,
+      "step": 2880
+    },
+    {
+      "epoch": 2.4316365166175853,
+      "grad_norm": 0.5217409133911133,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 2890
+    },
+    {
+      "epoch": 2.4400504838031134,
+      "grad_norm": 0.4555884897708893,
+      "learning_rate": 0.0002,
+      "loss": 1.5804,
+      "step": 2900
+    },
+    {
+      "epoch": 2.448464450988641,
+      "grad_norm": 0.43178579211235046,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2910
+    },
+    {
+      "epoch": 2.456878418174169,
+      "grad_norm": 0.4788478910923004,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2920
+    },
+    {
+      "epoch": 2.465292385359697,
+      "grad_norm": 0.43689873814582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 2930
+    },
+    {
+      "epoch": 2.473706352545225,
+      "grad_norm": 0.5115197896957397,
+      "learning_rate": 0.0002,
+      "loss": 1.6196,
+      "step": 2940
+    },
+    {
+      "epoch": 2.482120319730753,
+      "grad_norm": 0.5290159583091736,
+      "learning_rate": 0.0002,
+      "loss": 1.689,
+      "step": 2950
+    },
+    {
+      "epoch": 2.490534286916281,
+      "grad_norm": 0.46042463183403015,
+      "learning_rate": 0.0002,
+      "loss": 1.6499,
+      "step": 2960
+    },
+    {
+      "epoch": 2.498948254101809,
+      "grad_norm": 0.4359915852546692,
+      "learning_rate": 0.0002,
+      "loss": 1.6664,
+      "step": 2970
+    },
+    {
+      "epoch": 2.507362221287337,
+      "grad_norm": 0.46352964639663696,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 2980
+    },
+    {
+      "epoch": 2.515776188472865,
+      "grad_norm": 0.5324268341064453,
+      "learning_rate": 0.0002,
+      "loss": 1.6501,
+      "step": 2990
+    },
+    {
+      "epoch": 2.5241901556583928,
+      "grad_norm": 0.5929607152938843,
+      "learning_rate": 0.0002,
+      "loss": 1.6115,
+      "step": 3000
+    },
+    {
+      "epoch": 2.532604122843921,
+      "grad_norm": 0.4811333417892456,
+      "learning_rate": 0.0002,
+      "loss": 1.6772,
+      "step": 3010
+    },
+    {
+      "epoch": 2.541018090029449,
+      "grad_norm": 0.4662701487541199,
+      "learning_rate": 0.0002,
+      "loss": 1.7023,
+      "step": 3020
+    },
+    {
+      "epoch": 2.549432057214977,
+      "grad_norm": 0.4582270681858063,
+      "learning_rate": 0.0002,
+      "loss": 1.5426,
+      "step": 3030
+    },
+    {
+      "epoch": 2.557846024400505,
+      "grad_norm": 0.4679982662200928,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 3040
+    },
+    {
+      "epoch": 2.566259991586033,
+      "grad_norm": 0.4380294680595398,
+      "learning_rate": 0.0002,
+      "loss": 1.5442,
+      "step": 3050
+    },
+    {
+      "epoch": 2.5746739587715606,
+      "grad_norm": 0.44295763969421387,
+      "learning_rate": 0.0002,
+      "loss": 1.6055,
+      "step": 3060
+    },
+    {
+      "epoch": 2.5830879259570887,
+      "grad_norm": 0.5131027698516846,
+      "learning_rate": 0.0002,
+      "loss": 1.5775,
+      "step": 3070
+    },
+    {
+      "epoch": 2.591501893142617,
+      "grad_norm": 0.47567516565322876,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 3080
+    },
+    {
+      "epoch": 2.599915860328145,
+      "grad_norm": 0.49002596735954285,
+      "learning_rate": 0.0002,
+      "loss": 1.5671,
+      "step": 3090
+    },
+    {
+      "epoch": 2.6083298275136726,
+      "grad_norm": 0.44856327772140503,
+      "learning_rate": 0.0002,
+      "loss": 1.5445,
+      "step": 3100
+    },
+    {
+      "epoch": 2.6167437946992007,
+      "grad_norm": 0.4480142593383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5797,
+      "step": 3110
+    },
+    {
+      "epoch": 2.6251577618847284,
+      "grad_norm": 0.4317494034767151,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 3120
+    },
+    {
+      "epoch": 2.6335717290702565,
+      "grad_norm": 0.42580848932266235,
+      "learning_rate": 0.0002,
+      "loss": 1.6321,
+      "step": 3130
+    },
+    {
+      "epoch": 2.6419856962557846,
+      "grad_norm": 0.4516814947128296,
+      "learning_rate": 0.0002,
+      "loss": 1.6483,
+      "step": 3140
+    },
+    {
+      "epoch": 2.6503996634413127,
+      "grad_norm": 0.4438435733318329,
+      "learning_rate": 0.0002,
+      "loss": 1.695,
+      "step": 3150
+    },
+    {
+      "epoch": 2.6588136306268404,
+      "grad_norm": 0.4385356307029724,
+      "learning_rate": 0.0002,
+      "loss": 1.6938,
+      "step": 3160
+    },
+    {
+      "epoch": 2.6672275978123685,
+      "grad_norm": 0.5064112544059753,
+      "learning_rate": 0.0002,
+      "loss": 1.6139,
+      "step": 3170
+    },
+    {
+      "epoch": 2.6756415649978966,
+      "grad_norm": 0.49163177609443665,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 3180
+    },
+    {
+      "epoch": 2.6840555321834243,
+      "grad_norm": 0.49339258670806885,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 3190
+    },
+    {
+      "epoch": 2.6924694993689524,
+      "grad_norm": 0.440950870513916,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 3200
+    },
+    {
+      "epoch": 2.7008834665544805,
+      "grad_norm": 0.4283970594406128,
+      "learning_rate": 0.0002,
+      "loss": 1.6305,
+      "step": 3210
+    },
+    {
+      "epoch": 2.7092974337400086,
+      "grad_norm": 0.43875712156295776,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 3220
+    },
+    {
+      "epoch": 2.7177114009255363,
+      "grad_norm": 0.49332964420318604,
+      "learning_rate": 0.0002,
+      "loss": 1.6129,
+      "step": 3230
+    },
+    {
+      "epoch": 2.7261253681110644,
+      "grad_norm": 0.5225692391395569,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 3240
+    },
+    {
+      "epoch": 2.734539335296592,
+      "grad_norm": 0.4856489300727844,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 3250
+    },
+    {
+      "epoch": 2.74295330248212,
+      "grad_norm": 0.46918296813964844,
+      "learning_rate": 0.0002,
+      "loss": 1.6463,
+      "step": 3260
+    },
+    {
+      "epoch": 2.7513672696676483,
+      "grad_norm": 0.4802931249141693,
+      "learning_rate": 0.0002,
+      "loss": 1.6819,
+      "step": 3270
+    },
+    {
+      "epoch": 2.7597812368531764,
+      "grad_norm": 0.4485355615615845,
+      "learning_rate": 0.0002,
+      "loss": 1.6246,
+      "step": 3280
+    },
+    {
+      "epoch": 2.768195204038704,
+      "grad_norm": 0.43944594264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 3290
+    },
+    {
+      "epoch": 2.7766091712242322,
+      "grad_norm": 0.46847742795944214,
+      "learning_rate": 0.0002,
+      "loss": 1.6501,
+      "step": 3300
+    },
+    {
+      "epoch": 2.7850231384097603,
+      "grad_norm": 0.4816027879714966,
+      "learning_rate": 0.0002,
+      "loss": 1.5969,
+      "step": 3310
+    },
+    {
+      "epoch": 2.793437105595288,
+      "grad_norm": 0.453960120677948,
+      "learning_rate": 0.0002,
+      "loss": 1.6293,
+      "step": 3320
+    },
+    {
+      "epoch": 2.801851072780816,
+      "grad_norm": 0.4816017150878906,
+      "learning_rate": 0.0002,
+      "loss": 1.6429,
+      "step": 3330
+    },
+    {
+      "epoch": 2.8102650399663442,
+      "grad_norm": 0.4461034834384918,
+      "learning_rate": 0.0002,
+      "loss": 1.6683,
+      "step": 3340
+    },
+    {
+      "epoch": 2.8186790071518724,
+      "grad_norm": 0.48821821808815,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 3350
+    },
+    {
+      "epoch": 2.8270929743374,
+      "grad_norm": 0.4574853777885437,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 3360
+    },
+    {
+      "epoch": 2.835506941522928,
+      "grad_norm": 0.42062026262283325,
+      "learning_rate": 0.0002,
+      "loss": 1.6651,
+      "step": 3370
+    },
+    {
+      "epoch": 2.843920908708456,
+      "grad_norm": 0.4499834477901459,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 3380
+    },
+    {
+      "epoch": 2.852334875893984,
+      "grad_norm": 0.4780360758304596,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 3390
+    },
+    {
+      "epoch": 2.860748843079512,
+      "grad_norm": 0.45422887802124023,
+      "learning_rate": 0.0002,
+      "loss": 1.5882,
+      "step": 3400
+    },
+    {
+      "epoch": 2.86916281026504,
+      "grad_norm": 0.4590015709400177,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 3410
+    },
+    {
+      "epoch": 2.877576777450568,
+      "grad_norm": 0.45689624547958374,
+      "learning_rate": 0.0002,
+      "loss": 1.6746,
+      "step": 3420
+    },
+    {
+      "epoch": 2.885990744636096,
+      "grad_norm": 0.46953922510147095,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3430
+    },
+    {
+      "epoch": 2.8944047118216236,
+      "grad_norm": 0.4791966378688812,
+      "learning_rate": 0.0002,
+      "loss": 1.6015,
+      "step": 3440
+    },
+    {
+      "epoch": 2.9028186790071517,
+      "grad_norm": 0.4842296242713928,
+      "learning_rate": 0.0002,
+      "loss": 1.694,
+      "step": 3450
+    },
+    {
+      "epoch": 2.91123264619268,
+      "grad_norm": 0.47219768166542053,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3460
+    },
+    {
+      "epoch": 2.919646613378208,
+      "grad_norm": 0.4622127115726471,
+      "learning_rate": 0.0002,
+      "loss": 1.6486,
+      "step": 3470
+    },
+    {
+      "epoch": 2.9280605805637356,
+      "grad_norm": 0.46832820773124695,
+      "learning_rate": 0.0002,
+      "loss": 1.6485,
+      "step": 3480
+    },
+    {
+      "epoch": 2.9364745477492638,
+      "grad_norm": 0.44582483172416687,
+      "learning_rate": 0.0002,
+      "loss": 1.6366,
+      "step": 3490
+    },
+    {
+      "epoch": 2.944888514934792,
+      "grad_norm": 0.4987219274044037,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 3500
+    },
+    {
+      "epoch": 2.9533024821203195,
+      "grad_norm": 0.43750956654548645,
+      "learning_rate": 0.0002,
+      "loss": 1.5991,
+      "step": 3510
+    },
+    {
+      "epoch": 2.9617164493058477,
+      "grad_norm": 0.49962925910949707,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 3520
+    },
+    {
+      "epoch": 2.9701304164913758,
+      "grad_norm": 0.5189590454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 3530
+    },
+    {
+      "epoch": 2.978544383676904,
+      "grad_norm": 0.391317754983902,
+      "learning_rate": 0.0002,
+      "loss": 1.6688,
+      "step": 3540
+    },
+    {
+      "epoch": 2.9869583508624316,
+      "grad_norm": 0.44934695959091187,
+      "learning_rate": 0.0002,
+      "loss": 1.5884,
+      "step": 3550
+    },
+    {
+      "epoch": 2.9953723180479597,
+      "grad_norm": 0.4740142226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.5688,
+      "step": 3560
+    },
+    {
+      "epoch": 2.9995793016407237,
+      "eval_loss": 1.8266887664794922,
+      "eval_runtime": 37.9445,
+      "eval_samples_per_second": 13.572,
+      "eval_steps_per_second": 1.713,
+      "step": 3565
+    },
+    {
+      "epoch": 3.003786285233488,
+      "grad_norm": 0.4523724615573883,
+      "learning_rate": 0.0002,
+      "loss": 1.5939,
+      "step": 3570
+    },
+    {
+      "epoch": 3.0122002524190155,
+      "grad_norm": 0.5261380076408386,
+      "learning_rate": 0.0002,
+      "loss": 1.526,
+      "step": 3580
+    },
+    {
+      "epoch": 3.0206142196045436,
+      "grad_norm": 0.48664888739585876,
+      "learning_rate": 0.0002,
+      "loss": 1.4946,
+      "step": 3590
+    },
+    {
+      "epoch": 3.0290281867900717,
+      "grad_norm": 0.5070882439613342,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 3600
+    },
+    {
+      "epoch": 3.0374421539755994,
+      "grad_norm": 0.5816011428833008,
+      "learning_rate": 0.0002,
+      "loss": 1.5316,
+      "step": 3610
+    },
+    {
+      "epoch": 3.0458561211611275,
+      "grad_norm": 0.6610211730003357,
+      "learning_rate": 0.0002,
+      "loss": 1.5682,
+      "step": 3620
+    },
+    {
+      "epoch": 3.0542700883466556,
+      "grad_norm": 0.5257703065872192,
+      "learning_rate": 0.0002,
+      "loss": 1.5699,
+      "step": 3630
+    },
+    {
+      "epoch": 3.0626840555321833,
+      "grad_norm": 0.5574390888214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4438,
+      "step": 3640
+    },
+    {
+      "epoch": 3.0710980227177114,
+      "grad_norm": 0.5682297348976135,
+      "learning_rate": 0.0002,
+      "loss": 1.547,
+      "step": 3650
+    },
+    {
+      "epoch": 3.0795119899032395,
+      "grad_norm": 0.5798383355140686,
+      "learning_rate": 0.0002,
+      "loss": 1.5743,
+      "step": 3660
+    },
+    {
+      "epoch": 3.087925957088767,
+      "grad_norm": 0.5458289980888367,
+      "learning_rate": 0.0002,
+      "loss": 1.4339,
+      "step": 3670
+    },
+    {
+      "epoch": 3.0963399242742953,
+      "grad_norm": 0.5599102973937988,
+      "learning_rate": 0.0002,
+      "loss": 1.46,
+      "step": 3680
+    },
+    {
+      "epoch": 3.1047538914598234,
+      "grad_norm": 0.5023021697998047,
+      "learning_rate": 0.0002,
+      "loss": 1.4589,
+      "step": 3690
+    },
+    {
+      "epoch": 3.113167858645351,
+      "grad_norm": 0.5448206067085266,
+      "learning_rate": 0.0002,
+      "loss": 1.5114,
+      "step": 3700
+    },
+    {
+      "epoch": 3.121581825830879,
+      "grad_norm": 0.5760458707809448,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 3710
+    },
+    {
+      "epoch": 3.1299957930164073,
+      "grad_norm": 0.6018968224525452,
+      "learning_rate": 0.0002,
+      "loss": 1.4789,
+      "step": 3720
+    },
+    {
+      "epoch": 3.1384097602019354,
+      "grad_norm": 0.5767101049423218,
+      "learning_rate": 0.0002,
+      "loss": 1.5518,
+      "step": 3730
+    },
+    {
+      "epoch": 3.146823727387463,
+      "grad_norm": 0.5333963632583618,
+      "learning_rate": 0.0002,
+      "loss": 1.5032,
+      "step": 3740
+    },
+    {
+      "epoch": 3.155237694572991,
+      "grad_norm": 0.5918396711349487,
+      "learning_rate": 0.0002,
+      "loss": 1.4812,
+      "step": 3750
+    },
+    {
+      "epoch": 3.1636516617585193,
+      "grad_norm": 0.5931203365325928,
+      "learning_rate": 0.0002,
+      "loss": 1.4618,
+      "step": 3760
+    },
+    {
+      "epoch": 3.172065628944047,
+      "grad_norm": 0.6562168598175049,
+      "learning_rate": 0.0002,
+      "loss": 1.5592,
+      "step": 3770
+    },
+    {
+      "epoch": 3.180479596129575,
+      "grad_norm": 0.5820156335830688,
+      "learning_rate": 0.0002,
+      "loss": 1.4932,
+      "step": 3780
+    },
+    {
+      "epoch": 3.188893563315103,
+      "grad_norm": 0.5784737467765808,
+      "learning_rate": 0.0002,
+      "loss": 1.4523,
+      "step": 3790
+    },
+    {
+      "epoch": 3.197307530500631,
+      "grad_norm": 0.5506529808044434,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 3800
+    },
+    {
+      "epoch": 3.205721497686159,
+      "grad_norm": 0.6101595163345337,
+      "learning_rate": 0.0002,
+      "loss": 1.4819,
+      "step": 3810
+    },
+    {
+      "epoch": 3.214135464871687,
+      "grad_norm": 0.5597806572914124,
+      "learning_rate": 0.0002,
+      "loss": 1.5185,
+      "step": 3820
+    },
+    {
+      "epoch": 3.222549432057215,
+      "grad_norm": 0.5641011595726013,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 3830
+    },
+    {
+      "epoch": 3.230963399242743,
+      "grad_norm": 0.5892080068588257,
+      "learning_rate": 0.0002,
+      "loss": 1.4702,
+      "step": 3840
+    },
+    {
+      "epoch": 3.239377366428271,
+      "grad_norm": 0.6034760475158691,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 3850
+    },
+    {
+      "epoch": 3.247791333613799,
+      "grad_norm": 0.5112439393997192,
+      "learning_rate": 0.0002,
+      "loss": 1.5499,
+      "step": 3860
+    },
+    {
+      "epoch": 3.256205300799327,
+      "grad_norm": 0.56565922498703,
+      "learning_rate": 0.0002,
+      "loss": 1.5132,
+      "step": 3870
+    },
+    {
+      "epoch": 3.264619267984855,
+      "grad_norm": 0.6155247092247009,
+      "learning_rate": 0.0002,
+      "loss": 1.4892,
+      "step": 3880
+    },
+    {
+      "epoch": 3.273033235170383,
+      "grad_norm": 0.6064623594284058,
+      "learning_rate": 0.0002,
+      "loss": 1.5118,
+      "step": 3890
+    },
+    {
+      "epoch": 3.2814472023559107,
+      "grad_norm": 0.6313768029212952,
+      "learning_rate": 0.0002,
+      "loss": 1.5236,
+      "step": 3900
+    },
+    {
+      "epoch": 3.289861169541439,
+      "grad_norm": 0.5903939008712769,
+      "learning_rate": 0.0002,
+      "loss": 1.5551,
+      "step": 3910
+    },
+    {
+      "epoch": 3.298275136726967,
+      "grad_norm": 0.5770667195320129,
+      "learning_rate": 0.0002,
+      "loss": 1.5703,
+      "step": 3920
+    },
+    {
+      "epoch": 3.3066891039124946,
+      "grad_norm": 0.5785196423530579,
+      "learning_rate": 0.0002,
+      "loss": 1.5159,
+      "step": 3930
+    },
+    {
+      "epoch": 3.3151030710980227,
+      "grad_norm": 0.6468310356140137,
+      "learning_rate": 0.0002,
+      "loss": 1.5277,
+      "step": 3940
+    },
+    {
+      "epoch": 3.323517038283551,
+      "grad_norm": 0.6200279593467712,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 3950
+    },
+    {
+      "epoch": 3.3319310054690785,
+      "grad_norm": 0.5779302716255188,
+      "learning_rate": 0.0002,
+      "loss": 1.5264,
+      "step": 3960
+    },
+    {
+      "epoch": 3.3403449726546066,
+      "grad_norm": 0.5463796854019165,
+      "learning_rate": 0.0002,
+      "loss": 1.4861,
+      "step": 3970
+    },
+    {
+      "epoch": 3.3487589398401347,
+      "grad_norm": 0.6117855906486511,
+      "learning_rate": 0.0002,
+      "loss": 1.541,
+      "step": 3980
+    },
+    {
+      "epoch": 3.357172907025663,
+      "grad_norm": 0.5554766058921814,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 3990
+    },
+    {
+      "epoch": 3.3655868742111905,
+      "grad_norm": 0.6012870073318481,
+      "learning_rate": 0.0002,
+      "loss": 1.5004,
+      "step": 4000
+    },
+    {
+      "epoch": 3.3740008413967186,
+      "grad_norm": 0.5443974137306213,
+      "learning_rate": 0.0002,
+      "loss": 1.473,
+      "step": 4010
+    },
+    {
+      "epoch": 3.3824148085822463,
+      "grad_norm": 0.6636057496070862,
+      "learning_rate": 0.0002,
+      "loss": 1.5139,
+      "step": 4020
+    },
+    {
+      "epoch": 3.3908287757677744,
+      "grad_norm": 0.5801246166229248,
+      "learning_rate": 0.0002,
+      "loss": 1.5141,
+      "step": 4030
+    },
+    {
+      "epoch": 3.3992427429533025,
+      "grad_norm": 0.5668839812278748,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4040
+    },
+    {
+      "epoch": 3.4076567101388306,
+      "grad_norm": 0.7763481736183167,
+      "learning_rate": 0.0002,
+      "loss": 1.523,
+      "step": 4050
+    },
+    {
+      "epoch": 3.4160706773243583,
+      "grad_norm": 0.6675992608070374,
+      "learning_rate": 0.0002,
+      "loss": 1.4932,
+      "step": 4060
+    },
+    {
+      "epoch": 3.4244846445098864,
+      "grad_norm": 0.6290077567100525,
+      "learning_rate": 0.0002,
+      "loss": 1.4959,
+      "step": 4070
+    },
+    {
+      "epoch": 3.4328986116954145,
+      "grad_norm": 0.6040239930152893,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 4080
+    },
+    {
+      "epoch": 3.441312578880942,
+      "grad_norm": 0.6237877607345581,
+      "learning_rate": 0.0002,
+      "loss": 1.5711,
+      "step": 4090
+    },
+    {
+      "epoch": 3.4497265460664703,
+      "grad_norm": 0.5343508124351501,
+      "learning_rate": 0.0002,
+      "loss": 1.4961,
+      "step": 4100
+    },
+    {
+      "epoch": 3.4581405132519984,
+      "grad_norm": 0.6817412972450256,
+      "learning_rate": 0.0002,
+      "loss": 1.5123,
+      "step": 4110
+    },
+    {
+      "epoch": 3.466554480437526,
+      "grad_norm": 0.7115170359611511,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 4120
+    },
+    {
+      "epoch": 3.4749684476230542,
+      "grad_norm": 0.6127332448959351,
+      "learning_rate": 0.0002,
+      "loss": 1.5275,
+      "step": 4130
+    },
+    {
+      "epoch": 3.4833824148085824,
+      "grad_norm": 0.5745994448661804,
+      "learning_rate": 0.0002,
+      "loss": 1.557,
+      "step": 4140
+    },
+    {
+      "epoch": 3.49179638199411,
+      "grad_norm": 0.6248795390129089,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4150
+    },
+    {
+      "epoch": 3.500210349179638,
+      "grad_norm": 0.5821124911308289,
+      "learning_rate": 0.0002,
+      "loss": 1.4885,
+      "step": 4160
+    },
+    {
+      "epoch": 3.5086243163651663,
+      "grad_norm": 0.561416506767273,
+      "learning_rate": 0.0002,
+      "loss": 1.4937,
+      "step": 4170
+    },
+    {
+      "epoch": 3.5170382835506944,
+      "grad_norm": 0.5848962664604187,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 4180
+    },
+    {
+      "epoch": 3.525452250736222,
+      "grad_norm": 0.5335569977760315,
+      "learning_rate": 0.0002,
+      "loss": 1.5892,
+      "step": 4190
+    },
+    {
+      "epoch": 3.53386621792175,
+      "grad_norm": 0.547964870929718,
+      "learning_rate": 0.0002,
+      "loss": 1.5152,
+      "step": 4200
+    },
+    {
+      "epoch": 3.542280185107278,
+      "grad_norm": 0.6157727241516113,
+      "learning_rate": 0.0002,
+      "loss": 1.4887,
+      "step": 4210
+    },
+    {
+      "epoch": 3.550694152292806,
+      "grad_norm": 0.6163121461868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5484,
+      "step": 4220
+    },
+    {
+      "epoch": 3.559108119478334,
+      "grad_norm": 0.5844616293907166,
+      "learning_rate": 0.0002,
+      "loss": 1.5833,
+      "step": 4230
+    },
+    {
+      "epoch": 3.567522086663862,
+      "grad_norm": 0.7104926109313965,
+      "learning_rate": 0.0002,
+      "loss": 1.5305,
+      "step": 4240
+    },
+    {
+      "epoch": 3.57593605384939,
+      "grad_norm": 0.5055213570594788,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4250
+    },
+    {
+      "epoch": 3.584350021034918,
+      "grad_norm": 0.611676812171936,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 4260
+    },
+    {
+      "epoch": 3.592763988220446,
+      "grad_norm": 0.6326440572738647,
+      "learning_rate": 0.0002,
+      "loss": 1.5048,
+      "step": 4270
+    },
+    {
+      "epoch": 3.6011779554059737,
+      "grad_norm": 0.6290925741195679,
+      "learning_rate": 0.0002,
+      "loss": 1.5122,
+      "step": 4280
+    },
+    {
+      "epoch": 3.609591922591502,
+      "grad_norm": 0.5691978931427002,
+      "learning_rate": 0.0002,
+      "loss": 1.5654,
+      "step": 4290
+    },
+    {
+      "epoch": 3.61800588977703,
+      "grad_norm": 0.6071329116821289,
+      "learning_rate": 0.0002,
+      "loss": 1.4854,
+      "step": 4300
+    },
+    {
+      "epoch": 3.626419856962558,
+      "grad_norm": 0.606573224067688,
+      "learning_rate": 0.0002,
+      "loss": 1.5336,
+      "step": 4310
+    },
+    {
+      "epoch": 3.6348338241480858,
+      "grad_norm": 0.5515419244766235,
+      "learning_rate": 0.0002,
+      "loss": 1.6437,
+      "step": 4320
+    },
+    {
+      "epoch": 3.643247791333614,
+      "grad_norm": 0.5964660048484802,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 4330
+    },
+    {
+      "epoch": 3.6516617585191415,
+      "grad_norm": 0.5774146914482117,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 4340
+    },
+    {
+      "epoch": 3.6600757257046697,
+      "grad_norm": 0.5732731223106384,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 4350
+    },
+    {
+      "epoch": 3.6684896928901978,
+      "grad_norm": 0.7354163527488708,
+      "learning_rate": 0.0002,
+      "loss": 1.5682,
+      "step": 4360
+    },
+    {
+      "epoch": 3.676903660075726,
+      "grad_norm": 0.6220902800559998,
+      "learning_rate": 0.0002,
+      "loss": 1.5225,
+      "step": 4370
+    },
+    {
+      "epoch": 3.6853176272612536,
+      "grad_norm": 0.6053991317749023,
+      "learning_rate": 0.0002,
+      "loss": 1.4838,
+      "step": 4380
+    },
+    {
+      "epoch": 3.6937315944467817,
+      "grad_norm": 0.67010897397995,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4390
+    },
+    {
+      "epoch": 3.70214556163231,
+      "grad_norm": 0.6139186024665833,
+      "learning_rate": 0.0002,
+      "loss": 1.5381,
+      "step": 4400
+    },
+    {
+      "epoch": 3.7105595288178375,
+      "grad_norm": 0.5433071851730347,
+      "learning_rate": 0.0002,
+      "loss": 1.5088,
+      "step": 4410
+    },
+    {
+      "epoch": 3.7189734960033656,
+      "grad_norm": 0.5453870296478271,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 4420
+    },
+    {
+      "epoch": 3.7273874631888937,
+      "grad_norm": 0.6401727199554443,
+      "learning_rate": 0.0002,
+      "loss": 1.4549,
+      "step": 4430
+    },
+    {
+      "epoch": 3.735801430374422,
+      "grad_norm": 0.6049367189407349,
+      "learning_rate": 0.0002,
+      "loss": 1.503,
+      "step": 4440
+    },
+    {
+      "epoch": 3.7442153975599495,
+      "grad_norm": 0.5740529298782349,
+      "learning_rate": 0.0002,
+      "loss": 1.5268,
+      "step": 4450
+    },
+    {
+      "epoch": 3.7526293647454776,
+      "grad_norm": 0.6521880626678467,
+      "learning_rate": 0.0002,
+      "loss": 1.5183,
+      "step": 4460
+    },
+    {
+      "epoch": 3.7610433319310053,
+      "grad_norm": 0.7096368074417114,
+      "learning_rate": 0.0002,
+      "loss": 1.5741,
+      "step": 4470
+    },
+    {
+      "epoch": 3.7694572991165334,
+      "grad_norm": 0.5886474251747131,
+      "learning_rate": 0.0002,
+      "loss": 1.5786,
+      "step": 4480
+    },
+    {
+      "epoch": 3.7778712663020615,
+      "grad_norm": 0.5821043252944946,
+      "learning_rate": 0.0002,
+      "loss": 1.5887,
+      "step": 4490
+    },
+    {
+      "epoch": 3.7862852334875896,
+      "grad_norm": 0.628892183303833,
+      "learning_rate": 0.0002,
+      "loss": 1.5777,
+      "step": 4500
+    },
+    {
+      "epoch": 3.7946992006731173,
+      "grad_norm": 0.5962669849395752,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 4510
+    },
+    {
+      "epoch": 3.8031131678586454,
+      "grad_norm": 0.6635549068450928,
+      "learning_rate": 0.0002,
+      "loss": 1.5267,
+      "step": 4520
+    },
+    {
+      "epoch": 3.811527135044173,
+      "grad_norm": 0.6010760068893433,
+      "learning_rate": 0.0002,
+      "loss": 1.5058,
+      "step": 4530
+    },
+    {
+      "epoch": 3.819941102229701,
+      "grad_norm": 0.6322658658027649,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 4540
+    },
+    {
+      "epoch": 3.8283550694152293,
+      "grad_norm": 0.5893137454986572,
+      "learning_rate": 0.0002,
+      "loss": 1.5029,
+      "step": 4550
+    },
+    {
+      "epoch": 3.8367690366007574,
+      "grad_norm": 0.7829602360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.5435,
+      "step": 4560
+    },
+    {
+      "epoch": 3.845183003786285,
+      "grad_norm": 0.6190396547317505,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 4570
+    },
+    {
+      "epoch": 3.853596970971813,
+      "grad_norm": 0.6662813425064087,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 4580
+    },
+    {
+      "epoch": 3.8620109381573413,
+      "grad_norm": 0.5809855461120605,
+      "learning_rate": 0.0002,
+      "loss": 1.5065,
+      "step": 4590
+    },
+    {
+      "epoch": 3.870424905342869,
+      "grad_norm": 0.5779069662094116,
+      "learning_rate": 0.0002,
+      "loss": 1.5041,
+      "step": 4600
+    },
+    {
+      "epoch": 3.878838872528397,
+      "grad_norm": 0.5603038668632507,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 4610
+    },
+    {
+      "epoch": 3.887252839713925,
+      "grad_norm": 0.6274181008338928,
+      "learning_rate": 0.0002,
+      "loss": 1.5372,
+      "step": 4620
+    },
+    {
+      "epoch": 3.8956668068994533,
+      "grad_norm": 0.6810959577560425,
+      "learning_rate": 0.0002,
+      "loss": 1.4996,
+      "step": 4630
+    },
+    {
+      "epoch": 3.904080774084981,
+      "grad_norm": 0.5647315979003906,
+      "learning_rate": 0.0002,
+      "loss": 1.4956,
+      "step": 4640
+    },
+    {
+      "epoch": 3.912494741270509,
+      "grad_norm": 0.6830295324325562,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 4650
+    },
+    {
+      "epoch": 3.920908708456037,
+      "grad_norm": 0.652565598487854,
+      "learning_rate": 0.0002,
+      "loss": 1.535,
+      "step": 4660
+    },
+    {
+      "epoch": 3.929322675641565,
+      "grad_norm": 0.5806284546852112,
+      "learning_rate": 0.0002,
+      "loss": 1.4772,
+      "step": 4670
+    },
+    {
+      "epoch": 3.937736642827093,
+      "grad_norm": 0.6825073957443237,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 4680
+    },
+    {
+      "epoch": 3.946150610012621,
+      "grad_norm": 0.6149451732635498,
+      "learning_rate": 0.0002,
+      "loss": 1.5516,
+      "step": 4690
+    },
+    {
+      "epoch": 3.954564577198149,
+      "grad_norm": 0.6152557134628296,
+      "learning_rate": 0.0002,
+      "loss": 1.5608,
+      "step": 4700
+    },
+    {
+      "epoch": 3.962978544383677,
+      "grad_norm": 0.6239011883735657,
+      "learning_rate": 0.0002,
+      "loss": 1.4897,
+      "step": 4710
+    },
+    {
+      "epoch": 3.971392511569205,
+      "grad_norm": 0.6485443115234375,
+      "learning_rate": 0.0002,
+      "loss": 1.538,
+      "step": 4720
+    },
+    {
+      "epoch": 3.9798064787547327,
+      "grad_norm": 0.6449228525161743,
+      "learning_rate": 0.0002,
+      "loss": 1.5226,
+      "step": 4730
+    },
+    {
+      "epoch": 3.988220445940261,
+      "grad_norm": 0.6526407599449158,
+      "learning_rate": 0.0002,
+      "loss": 1.5087,
+      "step": 4740
+    },
+    {
+      "epoch": 3.996634413125789,
+      "grad_norm": 0.6277706027030945,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4750
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.871641755104065,
+      "eval_runtime": 37.9637,
+      "eval_samples_per_second": 13.566,
+      "eval_steps_per_second": 1.712,
+      "step": 4754
+    },
+    {
+      "epoch": 4.005048380311317,
+      "grad_norm": 0.6994837522506714,
+      "learning_rate": 0.0002,
+      "loss": 1.4744,
+      "step": 4760
+    },
+    {
+      "epoch": 4.013462347496845,
+      "grad_norm": 0.8728373050689697,
+      "learning_rate": 0.0002,
+      "loss": 1.4433,
+      "step": 4770
+    },
+    {
+      "epoch": 4.021876314682372,
+      "grad_norm": 0.688679575920105,
+      "learning_rate": 0.0002,
+      "loss": 1.3329,
+      "step": 4780
+    },
+    {
+      "epoch": 4.0302902818679005,
+      "grad_norm": 0.6313387155532837,
+      "learning_rate": 0.0002,
+      "loss": 1.3999,
+      "step": 4790
+    },
+    {
+      "epoch": 4.038704249053429,
+      "grad_norm": 0.6577984690666199,
+      "learning_rate": 0.0002,
+      "loss": 1.3346,
+      "step": 4800
+    },
+    {
+      "epoch": 4.047118216238957,
+      "grad_norm": 0.7938185930252075,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 4810
+    },
+    {
+      "epoch": 4.055532183424485,
+      "grad_norm": 0.760399580001831,
+      "learning_rate": 0.0002,
+      "loss": 1.3716,
+      "step": 4820
+    },
+    {
+      "epoch": 4.063946150610013,
+      "grad_norm": 0.7329602241516113,
+      "learning_rate": 0.0002,
+      "loss": 1.4321,
+      "step": 4830
+    },
+    {
+      "epoch": 4.07236011779554,
+      "grad_norm": 0.7778576016426086,
+      "learning_rate": 0.0002,
+      "loss": 1.4133,
+      "step": 4840
+    },
+    {
+      "epoch": 4.080774084981068,
+      "grad_norm": 0.8235865235328674,
+      "learning_rate": 0.0002,
+      "loss": 1.4372,
+      "step": 4850
+    },
+    {
+      "epoch": 4.089188052166596,
+      "grad_norm": 0.7743754386901855,
+      "learning_rate": 0.0002,
+      "loss": 1.3719,
+      "step": 4860
+    },
+    {
+      "epoch": 4.0976020193521245,
+      "grad_norm": 0.8145367503166199,
+      "learning_rate": 0.0002,
+      "loss": 1.3787,
+      "step": 4870
+    },
+    {
+      "epoch": 4.106015986537653,
+      "grad_norm": 0.8517307639122009,
+      "learning_rate": 0.0002,
+      "loss": 1.356,
+      "step": 4880
+    },
+    {
+      "epoch": 4.114429953723181,
+      "grad_norm": 0.8208953142166138,
+      "learning_rate": 0.0002,
+      "loss": 1.4191,
+      "step": 4890
+    },
+    {
+      "epoch": 4.122843920908709,
+      "grad_norm": 0.8437790870666504,
+      "learning_rate": 0.0002,
+      "loss": 1.3189,
+      "step": 4900
+    },
+    {
+      "epoch": 4.131257888094236,
+      "grad_norm": 0.716672420501709,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 4910
+    },
+    {
+      "epoch": 4.139671855279764,
+      "grad_norm": 0.7656235098838806,
+      "learning_rate": 0.0002,
+      "loss": 1.4392,
+      "step": 4920
+    },
+    {
+      "epoch": 4.148085822465292,
+      "grad_norm": 0.7209306955337524,
+      "learning_rate": 0.0002,
+      "loss": 1.3408,
+      "step": 4930
+    },
+    {
+      "epoch": 4.1564997896508205,
+      "grad_norm": 0.7731267809867859,
+      "learning_rate": 0.0002,
+      "loss": 1.3639,
+      "step": 4940
+    },
+    {
+      "epoch": 4.164913756836349,
+      "grad_norm": 0.7477553486824036,
+      "learning_rate": 0.0002,
+      "loss": 1.4151,
+      "step": 4950
+    },
+    {
+      "epoch": 4.173327724021877,
+      "grad_norm": 0.7372981309890747,
+      "learning_rate": 0.0002,
+      "loss": 1.3485,
+      "step": 4960
+    },
+    {
+      "epoch": 4.181741691207404,
+      "grad_norm": 0.6582154035568237,
+      "learning_rate": 0.0002,
+      "loss": 1.3901,
+      "step": 4970
+    },
+    {
+      "epoch": 4.190155658392932,
+      "grad_norm": 0.7003206610679626,
+      "learning_rate": 0.0002,
+      "loss": 1.3343,
+      "step": 4980
+    },
+    {
+      "epoch": 4.19856962557846,
+      "grad_norm": 0.735223650932312,
+      "learning_rate": 0.0002,
+      "loss": 1.4098,
+      "step": 4990
+    },
+    {
+      "epoch": 4.206983592763988,
+      "grad_norm": 0.7832302451133728,
+      "learning_rate": 0.0002,
+      "loss": 1.3564,
+      "step": 5000
+    },
+    {
+      "epoch": 4.215397559949516,
+      "grad_norm": 0.8819546103477478,
+      "learning_rate": 0.0002,
+      "loss": 1.3622,
+      "step": 5010
+    },
+    {
+      "epoch": 4.2238115271350445,
+      "grad_norm": 0.9325336813926697,
+      "learning_rate": 0.0002,
+      "loss": 1.4438,
+      "step": 5020
+    },
+    {
+      "epoch": 4.232225494320572,
+      "grad_norm": 0.7007517218589783,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 5030
+    },
+    {
+      "epoch": 4.2406394615061,
+      "grad_norm": 0.7118321061134338,
+      "learning_rate": 0.0002,
+      "loss": 1.3683,
+      "step": 5040
+    },
+    {
+      "epoch": 4.249053428691628,
+      "grad_norm": 0.6578946709632874,
+      "learning_rate": 0.0002,
+      "loss": 1.2365,
+      "step": 5050
+    },
+    {
+      "epoch": 4.257467395877156,
+      "grad_norm": 0.9438983798027039,
+      "learning_rate": 0.0002,
+      "loss": 1.3696,
+      "step": 5060
+    },
+    {
+      "epoch": 4.265881363062684,
+      "grad_norm": 0.703037679195404,
+      "learning_rate": 0.0002,
+      "loss": 1.3868,
+      "step": 5070
+    },
+    {
+      "epoch": 4.274295330248212,
+      "grad_norm": 0.7286025285720825,
+      "learning_rate": 0.0002,
+      "loss": 1.3687,
+      "step": 5080
+    },
+    {
+      "epoch": 4.28270929743374,
+      "grad_norm": 0.750689685344696,
+      "learning_rate": 0.0002,
+      "loss": 1.3605,
+      "step": 5090
+    },
+    {
+      "epoch": 4.291123264619268,
+      "grad_norm": 0.869753360748291,
+      "learning_rate": 0.0002,
+      "loss": 1.5089,
+      "step": 5100
+    },
+    {
+      "epoch": 4.299537231804796,
+      "grad_norm": 0.8712980151176453,
+      "learning_rate": 0.0002,
+      "loss": 1.4128,
+      "step": 5110
+    },
+    {
+      "epoch": 4.307951198990324,
+      "grad_norm": 0.690263569355011,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 5120
+    },
+    {
+      "epoch": 4.316365166175852,
+      "grad_norm": 0.7114760279655457,
+      "learning_rate": 0.0002,
+      "loss": 1.4088,
+      "step": 5130
+    },
+    {
+      "epoch": 4.32477913336138,
+      "grad_norm": 0.7588112354278564,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 5140
+    },
+    {
+      "epoch": 4.333193100546908,
+      "grad_norm": 0.7556202411651611,
+      "learning_rate": 0.0002,
+      "loss": 1.4408,
+      "step": 5150
+    },
+    {
+      "epoch": 4.341607067732435,
+      "grad_norm": 0.8357610702514648,
+      "learning_rate": 0.0002,
+      "loss": 1.4203,
+      "step": 5160
+    },
+    {
+      "epoch": 4.3500210349179635,
+      "grad_norm": 0.8054035902023315,
+      "learning_rate": 0.0002,
+      "loss": 1.3348,
+      "step": 5170
+    },
+    {
+      "epoch": 4.358435002103492,
+      "grad_norm": 0.7637107968330383,
+      "learning_rate": 0.0002,
+      "loss": 1.3109,
+      "step": 5180
+    },
+    {
+      "epoch": 4.36684896928902,
+      "grad_norm": 0.757481038570404,
+      "learning_rate": 0.0002,
+      "loss": 1.3744,
+      "step": 5190
+    },
+    {
+      "epoch": 4.375262936474548,
+      "grad_norm": 0.7185863852500916,
+      "learning_rate": 0.0002,
+      "loss": 1.3622,
+      "step": 5200
+    },
+    {
+      "epoch": 4.383676903660076,
+      "grad_norm": 0.7326455116271973,
+      "learning_rate": 0.0002,
+      "loss": 1.3896,
+      "step": 5210
+    },
+    {
+      "epoch": 4.392090870845603,
+      "grad_norm": 0.7980523109436035,
+      "learning_rate": 0.0002,
+      "loss": 1.4098,
+      "step": 5220
+    },
+    {
+      "epoch": 4.400504838031131,
+      "grad_norm": 0.8526999354362488,
+      "learning_rate": 0.0002,
+      "loss": 1.3783,
+      "step": 5230
+    },
+    {
+      "epoch": 4.4089188052166595,
+      "grad_norm": 0.7012337446212769,
+      "learning_rate": 0.0002,
+      "loss": 1.4022,
+      "step": 5240
+    },
+    {
+      "epoch": 4.417332772402188,
+      "grad_norm": 0.8217827677726746,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 5250
+    },
+    {
+      "epoch": 4.425746739587716,
+      "grad_norm": 0.7141005396842957,
+      "learning_rate": 0.0002,
+      "loss": 1.3482,
+      "step": 5260
+    },
+    {
+      "epoch": 4.434160706773244,
+      "grad_norm": 0.7094302177429199,
+      "learning_rate": 0.0002,
+      "loss": 1.3699,
+      "step": 5270
+    },
+    {
+      "epoch": 4.442574673958772,
+      "grad_norm": 0.7234613299369812,
+      "learning_rate": 0.0002,
+      "loss": 1.3527,
+      "step": 5280
+    },
+    {
+      "epoch": 4.450988641144299,
+      "grad_norm": 0.7530457973480225,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 5290
+    },
+    {
+      "epoch": 4.459402608329827,
+      "grad_norm": 0.7300912141799927,
+      "learning_rate": 0.0002,
+      "loss": 1.3944,
+      "step": 5300
+    },
+    {
+      "epoch": 4.467816575515355,
+      "grad_norm": 0.825443685054779,
+      "learning_rate": 0.0002,
+      "loss": 1.3844,
+      "step": 5310
+    },
+    {
+      "epoch": 4.4762305427008835,
+      "grad_norm": 0.7559658885002136,
+      "learning_rate": 0.0002,
+      "loss": 1.3648,
+      "step": 5320
+    },
+    {
+      "epoch": 4.484644509886412,
+      "grad_norm": 0.8817561268806458,
+      "learning_rate": 0.0002,
+      "loss": 1.4364,
+      "step": 5330
+    },
+    {
+      "epoch": 4.49305847707194,
+      "grad_norm": 0.8203575611114502,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 5340
+    },
+    {
+      "epoch": 4.501472444257468,
+      "grad_norm": 0.7677690982818604,
+      "learning_rate": 0.0002,
+      "loss": 1.3996,
+      "step": 5350
+    },
+    {
+      "epoch": 4.509886411442995,
+      "grad_norm": 0.657085120677948,
+      "learning_rate": 0.0002,
+      "loss": 1.4142,
+      "step": 5360
+    },
+    {
+      "epoch": 4.518300378628523,
+      "grad_norm": 0.7939504384994507,
+      "learning_rate": 0.0002,
+      "loss": 1.3722,
+      "step": 5370
+    },
+    {
+      "epoch": 4.526714345814051,
+      "grad_norm": 0.6971889138221741,
+      "learning_rate": 0.0002,
+      "loss": 1.4361,
+      "step": 5380
+    },
+    {
+      "epoch": 4.535128312999579,
+      "grad_norm": 0.6984175443649292,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 5390
+    },
+    {
+      "epoch": 4.5435422801851075,
+      "grad_norm": 0.8504858613014221,
+      "learning_rate": 0.0002,
+      "loss": 1.341,
+      "step": 5400
+    },
+    {
+      "epoch": 4.551956247370635,
+      "grad_norm": 0.9134073853492737,
+      "learning_rate": 0.0002,
+      "loss": 1.4026,
+      "step": 5410
+    },
+    {
+      "epoch": 4.560370214556163,
+      "grad_norm": 0.7765598893165588,
+      "learning_rate": 0.0002,
+      "loss": 1.4375,
+      "step": 5420
+    },
+    {
+      "epoch": 4.568784181741691,
+      "grad_norm": 0.6991009712219238,
+      "learning_rate": 0.0002,
+      "loss": 1.4832,
+      "step": 5430
+    },
+    {
+      "epoch": 4.577198148927219,
+      "grad_norm": 0.8393039107322693,
+      "learning_rate": 0.0002,
+      "loss": 1.4021,
+      "step": 5440
+    },
+    {
+      "epoch": 4.585612116112747,
+      "grad_norm": 0.7685918211936951,
+      "learning_rate": 0.0002,
+      "loss": 1.3976,
+      "step": 5450
+    },
+    {
+      "epoch": 4.594026083298275,
+      "grad_norm": 0.7135679721832275,
+      "learning_rate": 0.0002,
+      "loss": 1.3883,
+      "step": 5460
+    },
+    {
+      "epoch": 4.6024400504838034,
+      "grad_norm": 0.6728870868682861,
+      "learning_rate": 0.0002,
+      "loss": 1.4083,
+      "step": 5470
+    },
+    {
+      "epoch": 4.610854017669331,
+      "grad_norm": 0.7139479517936707,
+      "learning_rate": 0.0002,
+      "loss": 1.3698,
+      "step": 5480
+    },
+    {
+      "epoch": 4.619267984854859,
+      "grad_norm": 0.8476598858833313,
+      "learning_rate": 0.0002,
+      "loss": 1.3498,
+      "step": 5490
+    },
+    {
+      "epoch": 4.627681952040387,
+      "grad_norm": 0.8034361004829407,
+      "learning_rate": 0.0002,
+      "loss": 1.3389,
+      "step": 5500
+    },
+    {
+      "epoch": 4.636095919225915,
+      "grad_norm": 0.7452183961868286,
+      "learning_rate": 0.0002,
+      "loss": 1.4179,
+      "step": 5510
+    },
+    {
+      "epoch": 4.644509886411443,
+      "grad_norm": 0.8394148945808411,
+      "learning_rate": 0.0002,
+      "loss": 1.4031,
+      "step": 5520
+    },
+    {
+      "epoch": 4.652923853596971,
+      "grad_norm": 0.7480153441429138,
+      "learning_rate": 0.0002,
+      "loss": 1.4561,
+      "step": 5530
+    },
+    {
+      "epoch": 4.661337820782499,
+      "grad_norm": 0.7781714797019958,
+      "learning_rate": 0.0002,
+      "loss": 1.378,
+      "step": 5540
+    },
+    {
+      "epoch": 4.669751787968027,
+      "grad_norm": 1.0058213472366333,
+      "learning_rate": 0.0002,
+      "loss": 1.3924,
+      "step": 5550
+    },
+    {
+      "epoch": 4.678165755153555,
+      "grad_norm": 0.7403179407119751,
+      "learning_rate": 0.0002,
+      "loss": 1.4198,
+      "step": 5560
+    },
+    {
+      "epoch": 4.686579722339083,
+      "grad_norm": 0.7270476818084717,
+      "learning_rate": 0.0002,
+      "loss": 1.4328,
+      "step": 5570
+    },
+    {
+      "epoch": 4.694993689524611,
+      "grad_norm": 0.760877788066864,
+      "learning_rate": 0.0002,
+      "loss": 1.378,
+      "step": 5580
+    },
+    {
+      "epoch": 4.703407656710139,
+      "grad_norm": 0.8097004890441895,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 5590
+    },
+    {
+      "epoch": 4.711821623895667,
+      "grad_norm": 0.9096523523330688,
+      "learning_rate": 0.0002,
+      "loss": 1.3661,
+      "step": 5600
+    },
+    {
+      "epoch": 4.720235591081195,
+      "grad_norm": 0.7262444496154785,
+      "learning_rate": 0.0002,
+      "loss": 1.4012,
+      "step": 5610
+    },
+    {
+      "epoch": 4.7286495582667225,
+      "grad_norm": 0.8207762837409973,
+      "learning_rate": 0.0002,
+      "loss": 1.422,
+      "step": 5620
+    },
+    {
+      "epoch": 4.737063525452251,
+      "grad_norm": 0.8089601993560791,
+      "learning_rate": 0.0002,
+      "loss": 1.4017,
+      "step": 5630
+    },
+    {
+      "epoch": 4.745477492637779,
+      "grad_norm": 0.7609543800354004,
+      "learning_rate": 0.0002,
+      "loss": 1.3675,
+      "step": 5640
+    },
+    {
+      "epoch": 4.753891459823307,
+      "grad_norm": 0.7273501753807068,
+      "learning_rate": 0.0002,
+      "loss": 1.4085,
+      "step": 5650
+    },
+    {
+      "epoch": 4.762305427008835,
+      "grad_norm": 0.7800219058990479,
+      "learning_rate": 0.0002,
+      "loss": 1.3849,
+      "step": 5660
+    },
+    {
+      "epoch": 4.770719394194362,
+      "grad_norm": 0.8558377623558044,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 5670
+    },
+    {
+      "epoch": 4.77913336137989,
+      "grad_norm": 0.7131547927856445,
+      "learning_rate": 0.0002,
+      "loss": 1.3831,
+      "step": 5680
+    },
+    {
+      "epoch": 4.787547328565418,
+      "grad_norm": 0.7651025056838989,
+      "learning_rate": 0.0002,
+      "loss": 1.407,
+      "step": 5690
+    },
+    {
+      "epoch": 4.7959612957509465,
+      "grad_norm": 0.8129976391792297,
+      "learning_rate": 0.0002,
+      "loss": 1.3882,
+      "step": 5700
+    },
+    {
+      "epoch": 4.804375262936475,
+      "grad_norm": 0.8019895553588867,
+      "learning_rate": 0.0002,
+      "loss": 1.4347,
+      "step": 5710
+    },
+    {
+      "epoch": 4.812789230122003,
+      "grad_norm": 0.7692018151283264,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 5720
+    },
+    {
+      "epoch": 4.821203197307531,
+      "grad_norm": 0.6893943548202515,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 5730
+    },
+    {
+      "epoch": 4.829617164493058,
+      "grad_norm": 0.6881810426712036,
+      "learning_rate": 0.0002,
+      "loss": 1.4453,
+      "step": 5740
+    },
+    {
+      "epoch": 4.838031131678586,
+      "grad_norm": 0.7838267683982849,
+      "learning_rate": 0.0002,
+      "loss": 1.4775,
+      "step": 5750
+    },
+    {
+      "epoch": 4.846445098864114,
+      "grad_norm": 0.727799117565155,
+      "learning_rate": 0.0002,
+      "loss": 1.3857,
+      "step": 5760
+    },
+    {
+      "epoch": 4.8548590660496425,
+      "grad_norm": 0.7458277344703674,
+      "learning_rate": 0.0002,
+      "loss": 1.4685,
+      "step": 5770
+    },
+    {
+      "epoch": 4.863273033235171,
+      "grad_norm": 0.903802216053009,
+      "learning_rate": 0.0002,
+      "loss": 1.4426,
+      "step": 5780
+    },
+    {
+      "epoch": 4.871687000420699,
+      "grad_norm": 0.7983472347259521,
+      "learning_rate": 0.0002,
+      "loss": 1.451,
+      "step": 5790
+    },
+    {
+      "epoch": 4.880100967606227,
+      "grad_norm": 0.6894361972808838,
+      "learning_rate": 0.0002,
+      "loss": 1.4534,
+      "step": 5800
+    },
+    {
+      "epoch": 4.888514934791754,
+      "grad_norm": 0.7499409317970276,
+      "learning_rate": 0.0002,
+      "loss": 1.4486,
+      "step": 5810
+    },
+    {
+      "epoch": 4.896928901977282,
+      "grad_norm": 0.7362820506095886,
+      "learning_rate": 0.0002,
+      "loss": 1.4253,
+      "step": 5820
+    },
+    {
+      "epoch": 4.90534286916281,
+      "grad_norm": 0.8341619968414307,
+      "learning_rate": 0.0002,
+      "loss": 1.3763,
+      "step": 5830
+    },
+    {
+      "epoch": 4.913756836348338,
+      "grad_norm": 0.9604470133781433,
+      "learning_rate": 0.0002,
+      "loss": 1.3748,
+      "step": 5840
+    },
+    {
+      "epoch": 4.9221708035338665,
+      "grad_norm": 0.8916844129562378,
+      "learning_rate": 0.0002,
+      "loss": 1.3658,
+      "step": 5850
+    },
+    {
+      "epoch": 4.930584770719394,
+      "grad_norm": 0.8519647121429443,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 5860
+    },
+    {
+      "epoch": 4.938998737904922,
+      "grad_norm": 0.7946906089782715,
+      "learning_rate": 0.0002,
+      "loss": 1.424,
+      "step": 5870
+    },
+    {
+      "epoch": 4.94741270509045,
+      "grad_norm": 0.7843789458274841,
+      "learning_rate": 0.0002,
+      "loss": 1.4071,
+      "step": 5880
+    },
+    {
+      "epoch": 4.955826672275978,
+      "grad_norm": 0.707618772983551,
+      "learning_rate": 0.0002,
+      "loss": 1.4021,
+      "step": 5890
+    },
+    {
+      "epoch": 4.964240639461506,
+      "grad_norm": 0.7704206109046936,
+      "learning_rate": 0.0002,
+      "loss": 1.502,
+      "step": 5900
+    },
+    {
+      "epoch": 4.972654606647034,
+      "grad_norm": 0.7160256505012512,
+      "learning_rate": 0.0002,
+      "loss": 1.4456,
+      "step": 5910
+    },
+    {
+      "epoch": 4.981068573832562,
+      "grad_norm": 0.7020420432090759,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 5920
+    },
+    {
+      "epoch": 4.98948254101809,
+      "grad_norm": 0.7576286792755127,
+      "learning_rate": 0.0002,
+      "loss": 1.4037,
+      "step": 5930
+    },
+    {
+      "epoch": 4.997896508203618,
+      "grad_norm": 0.8573036789894104,
+      "learning_rate": 0.0002,
+      "loss": 1.414,
+      "step": 5940
+    },
+    {
+      "epoch": 4.999579301640724,
+      "eval_loss": 1.9353811740875244,
+      "eval_runtime": 37.9208,
+      "eval_samples_per_second": 13.581,
+      "eval_steps_per_second": 1.714,
+      "step": 5942
+    },
+    {
+      "epoch": 5.006310475389146,
+      "grad_norm": 0.8204267621040344,
+      "learning_rate": 0.0002,
+      "loss": 1.2418,
+      "step": 5950
+    },
+    {
+      "epoch": 5.014724442574674,
+      "grad_norm": 0.976840615272522,
+      "learning_rate": 0.0002,
+      "loss": 1.235,
+      "step": 5960
+    },
+    {
+      "epoch": 5.023138409760202,
+      "grad_norm": 0.8765613436698914,
+      "learning_rate": 0.0002,
+      "loss": 1.2134,
+      "step": 5970
+    },
+    {
+      "epoch": 5.03155237694573,
+      "grad_norm": 1.1793042421340942,
+      "learning_rate": 0.0002,
+      "loss": 1.2748,
+      "step": 5980
+    },
+    {
+      "epoch": 5.039966344131258,
+      "grad_norm": 0.971062958240509,
+      "learning_rate": 0.0002,
+      "loss": 1.2412,
+      "step": 5990
+    },
+    {
+      "epoch": 5.0483803113167856,
+      "grad_norm": 0.8649757504463196,
+      "learning_rate": 0.0002,
+      "loss": 1.1819,
+      "step": 6000
+    },
+    {
+      "epoch": 5.056794278502314,
+      "grad_norm": 0.9563034176826477,
+      "learning_rate": 0.0002,
+      "loss": 1.1654,
+      "step": 6010
+    },
+    {
+      "epoch": 5.065208245687842,
+      "grad_norm": 1.0093994140625,
+      "learning_rate": 0.0002,
+      "loss": 1.2238,
+      "step": 6020
+    },
+    {
+      "epoch": 5.07362221287337,
+      "grad_norm": 1.004213571548462,
+      "learning_rate": 0.0002,
+      "loss": 1.2519,
+      "step": 6030
+    },
+    {
+      "epoch": 5.082036180058898,
+      "grad_norm": 0.8307787179946899,
+      "learning_rate": 0.0002,
+      "loss": 1.2379,
+      "step": 6040
+    },
+    {
+      "epoch": 5.090450147244426,
+      "grad_norm": 0.9117848873138428,
+      "learning_rate": 0.0002,
+      "loss": 1.2282,
+      "step": 6050
+    },
+    {
+      "epoch": 5.098864114429953,
+      "grad_norm": 1.0269840955734253,
+      "learning_rate": 0.0002,
+      "loss": 1.2582,
+      "step": 6060
+    },
+    {
+      "epoch": 5.1072780816154815,
+      "grad_norm": 0.9079542756080627,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 6070
+    },
+    {
+      "epoch": 5.11569204880101,
+      "grad_norm": 0.885702908039093,
+      "learning_rate": 0.0002,
+      "loss": 1.215,
+      "step": 6080
+    },
+    {
+      "epoch": 5.124106015986538,
+      "grad_norm": 0.9976128339767456,
+      "learning_rate": 0.0002,
+      "loss": 1.2406,
+      "step": 6090
+    },
+    {
+      "epoch": 5.132519983172066,
+      "grad_norm": 0.8472117185592651,
+      "learning_rate": 0.0002,
+      "loss": 1.3082,
+      "step": 6100
+    },
+    {
+      "epoch": 5.140933950357594,
+      "grad_norm": 1.0385161638259888,
+      "learning_rate": 0.0002,
+      "loss": 1.226,
+      "step": 6110
+    },
+    {
+      "epoch": 5.149347917543121,
+      "grad_norm": 0.8948383927345276,
+      "learning_rate": 0.0002,
+      "loss": 1.213,
+      "step": 6120
+    },
+    {
+      "epoch": 5.157761884728649,
+      "grad_norm": 1.2613716125488281,
+      "learning_rate": 0.0002,
+      "loss": 1.2213,
+      "step": 6130
+    },
+    {
+      "epoch": 5.166175851914177,
+      "grad_norm": 0.9933410286903381,
+      "learning_rate": 0.0002,
+      "loss": 1.2632,
+      "step": 6140
+    },
+    {
+      "epoch": 5.1745898190997055,
+      "grad_norm": 0.9673663973808289,
+      "learning_rate": 0.0002,
+      "loss": 1.1715,
+      "step": 6150
+    },
+    {
+      "epoch": 5.183003786285234,
+      "grad_norm": 0.9969648122787476,
+      "learning_rate": 0.0002,
+      "loss": 1.2947,
+      "step": 6160
+    },
+    {
+      "epoch": 5.191417753470762,
+      "grad_norm": 1.2163258790969849,
+      "learning_rate": 0.0002,
+      "loss": 1.2416,
+      "step": 6170
+    },
+    {
+      "epoch": 5.19983172065629,
+      "grad_norm": 0.9163419604301453,
+      "learning_rate": 0.0002,
+      "loss": 1.2221,
+      "step": 6180
+    },
+    {
+      "epoch": 5.208245687841817,
+      "grad_norm": 0.9225585460662842,
+      "learning_rate": 0.0002,
+      "loss": 1.2624,
+      "step": 6190
+    },
+    {
+      "epoch": 5.216659655027345,
+      "grad_norm": 0.9205296635627747,
+      "learning_rate": 0.0002,
+      "loss": 1.2932,
+      "step": 6200
+    },
+    {
+      "epoch": 5.225073622212873,
+      "grad_norm": 1.0655443668365479,
+      "learning_rate": 0.0002,
+      "loss": 1.1825,
+      "step": 6210
+    },
+    {
+      "epoch": 5.233487589398401,
+      "grad_norm": 1.0854865312576294,
+      "learning_rate": 0.0002,
+      "loss": 1.2613,
+      "step": 6220
+    },
+    {
+      "epoch": 5.2419015565839295,
+      "grad_norm": 0.8489186763763428,
+      "learning_rate": 0.0002,
+      "loss": 1.3045,
+      "step": 6230
+    },
+    {
+      "epoch": 5.250315523769458,
+      "grad_norm": 0.910391628742218,
+      "learning_rate": 0.0002,
+      "loss": 1.2708,
+      "step": 6240
+    },
+    {
+      "epoch": 5.258729490954985,
+      "grad_norm": 0.925507128238678,
+      "learning_rate": 0.0002,
+      "loss": 1.1914,
+      "step": 6250
+    },
+    {
+      "epoch": 5.267143458140513,
+      "grad_norm": 1.1069735288619995,
+      "learning_rate": 0.0002,
+      "loss": 1.3368,
+      "step": 6260
+    },
+    {
+      "epoch": 5.275557425326041,
+      "grad_norm": 0.9705119132995605,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 6270
+    },
+    {
+      "epoch": 5.283971392511569,
+      "grad_norm": 0.9752426147460938,
+      "learning_rate": 0.0002,
+      "loss": 1.2602,
+      "step": 6280
+    },
+    {
+      "epoch": 5.292385359697097,
+      "grad_norm": 1.021359920501709,
+      "learning_rate": 0.0002,
+      "loss": 1.2043,
+      "step": 6290
+    },
+    {
+      "epoch": 5.3007993268826255,
+      "grad_norm": 1.148606300354004,
+      "learning_rate": 0.0002,
+      "loss": 1.2848,
+      "step": 6300
+    },
+    {
+      "epoch": 5.309213294068153,
+      "grad_norm": 0.8909247517585754,
+      "learning_rate": 0.0002,
+      "loss": 1.2201,
+      "step": 6310
+    },
+    {
+      "epoch": 5.317627261253681,
+      "grad_norm": 0.9879156351089478,
+      "learning_rate": 0.0002,
+      "loss": 1.2376,
+      "step": 6320
+    },
+    {
+      "epoch": 5.326041228439209,
+      "grad_norm": 0.9473357200622559,
+      "learning_rate": 0.0002,
+      "loss": 1.2638,
+      "step": 6330
+    },
+    {
+      "epoch": 5.334455195624737,
+      "grad_norm": 1.1422028541564941,
+      "learning_rate": 0.0002,
+      "loss": 1.232,
+      "step": 6340
+    },
+    {
+      "epoch": 5.342869162810265,
+      "grad_norm": 0.9942235350608826,
+      "learning_rate": 0.0002,
+      "loss": 1.263,
+      "step": 6350
+    },
+    {
+      "epoch": 5.351283129995793,
+      "grad_norm": 0.9535723924636841,
+      "learning_rate": 0.0002,
+      "loss": 1.3032,
+      "step": 6360
+    },
+    {
+      "epoch": 5.359697097181321,
+      "grad_norm": 0.9020892381668091,
+      "learning_rate": 0.0002,
+      "loss": 1.2908,
+      "step": 6370
+    },
+    {
+      "epoch": 5.368111064366849,
+      "grad_norm": 1.0626472234725952,
+      "learning_rate": 0.0002,
+      "loss": 1.2023,
+      "step": 6380
+    },
+    {
+      "epoch": 5.376525031552377,
+      "grad_norm": 1.1395848989486694,
+      "learning_rate": 0.0002,
+      "loss": 1.2555,
+      "step": 6390
+    },
+    {
+      "epoch": 5.384938998737905,
+      "grad_norm": 0.9274451732635498,
+      "learning_rate": 0.0002,
+      "loss": 1.2839,
+      "step": 6400
+    },
+    {
+      "epoch": 5.393352965923433,
+      "grad_norm": 0.8108699917793274,
+      "learning_rate": 0.0002,
+      "loss": 1.2819,
+      "step": 6410
+    },
+    {
+      "epoch": 5.401766933108961,
+      "grad_norm": 1.1805564165115356,
+      "learning_rate": 0.0002,
+      "loss": 1.2589,
+      "step": 6420
+    },
+    {
+      "epoch": 5.410180900294489,
+      "grad_norm": 0.8321298360824585,
+      "learning_rate": 0.0002,
+      "loss": 1.3549,
+      "step": 6430
+    },
+    {
+      "epoch": 5.418594867480017,
+      "grad_norm": 0.8981925249099731,
+      "learning_rate": 0.0002,
+      "loss": 1.2925,
+      "step": 6440
+    },
+    {
+      "epoch": 5.4270088346655445,
+      "grad_norm": 1.0730986595153809,
+      "learning_rate": 0.0002,
+      "loss": 1.258,
+      "step": 6450
+    },
+    {
+      "epoch": 5.435422801851073,
+      "grad_norm": 1.0584609508514404,
+      "learning_rate": 0.0002,
+      "loss": 1.26,
+      "step": 6460
+    },
+    {
+      "epoch": 5.443836769036601,
+      "grad_norm": 1.0792299509048462,
+      "learning_rate": 0.0002,
+      "loss": 1.2847,
+      "step": 6470
+    },
+    {
+      "epoch": 5.452250736222129,
+      "grad_norm": 0.9101872444152832,
+      "learning_rate": 0.0002,
+      "loss": 1.2035,
+      "step": 6480
+    },
+    {
+      "epoch": 5.460664703407657,
+      "grad_norm": 0.9910100698471069,
+      "learning_rate": 0.0002,
+      "loss": 1.2574,
+      "step": 6490
+    },
+    {
+      "epoch": 5.469078670593185,
+      "grad_norm": 1.041412353515625,
+      "learning_rate": 0.0002,
+      "loss": 1.3098,
+      "step": 6500
+    },
+    {
+      "epoch": 5.477492637778712,
+      "grad_norm": 1.0091687440872192,
+      "learning_rate": 0.0002,
+      "loss": 1.2812,
+      "step": 6510
+    },
+    {
+      "epoch": 5.48590660496424,
+      "grad_norm": 0.8755383491516113,
+      "learning_rate": 0.0002,
+      "loss": 1.2523,
+      "step": 6520
+    },
+    {
+      "epoch": 5.4943205721497685,
+      "grad_norm": 0.980212390422821,
+      "learning_rate": 0.0002,
+      "loss": 1.3042,
+      "step": 6530
+    },
+    {
+      "epoch": 5.502734539335297,
+      "grad_norm": 0.9356869459152222,
+      "learning_rate": 0.0002,
+      "loss": 1.2873,
+      "step": 6540
+    },
+    {
+      "epoch": 5.511148506520825,
+      "grad_norm": 0.9008095264434814,
+      "learning_rate": 0.0002,
+      "loss": 1.2254,
+      "step": 6550
+    },
+    {
+      "epoch": 5.519562473706353,
+      "grad_norm": 0.8908938765525818,
+      "learning_rate": 0.0002,
+      "loss": 1.2818,
+      "step": 6560
+    },
+    {
+      "epoch": 5.52797644089188,
+      "grad_norm": 1.1423932313919067,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 6570
+    },
+    {
+      "epoch": 5.536390408077408,
+      "grad_norm": 1.0508161783218384,
+      "learning_rate": 0.0002,
+      "loss": 1.3039,
+      "step": 6580
+    },
+    {
+      "epoch": 5.544804375262936,
+      "grad_norm": 0.8357517719268799,
+      "learning_rate": 0.0002,
+      "loss": 1.2446,
+      "step": 6590
+    },
+    {
+      "epoch": 5.5532183424484645,
+      "grad_norm": 0.9892540574073792,
+      "learning_rate": 0.0002,
+      "loss": 1.3037,
+      "step": 6600
+    },
+    {
+      "epoch": 5.561632309633993,
+      "grad_norm": 1.0048326253890991,
+      "learning_rate": 0.0002,
+      "loss": 1.3028,
+      "step": 6610
+    },
+    {
+      "epoch": 5.570046276819521,
+      "grad_norm": 0.9801995158195496,
+      "learning_rate": 0.0002,
+      "loss": 1.2152,
+      "step": 6620
+    },
+    {
+      "epoch": 5.578460244005049,
+      "grad_norm": 0.9899214506149292,
+      "learning_rate": 0.0002,
+      "loss": 1.2606,
+      "step": 6630
+    },
+    {
+      "epoch": 5.586874211190576,
+      "grad_norm": 1.1911814212799072,
+      "learning_rate": 0.0002,
+      "loss": 1.2043,
+      "step": 6640
+    },
+    {
+      "epoch": 5.595288178376104,
+      "grad_norm": 1.0368894338607788,
+      "learning_rate": 0.0002,
+      "loss": 1.3458,
+      "step": 6650
+    },
+    {
+      "epoch": 5.603702145561632,
+      "grad_norm": 1.1248382329940796,
+      "learning_rate": 0.0002,
+      "loss": 1.2595,
+      "step": 6660
+    },
+    {
+      "epoch": 5.61211611274716,
+      "grad_norm": 0.9765539765357971,
+      "learning_rate": 0.0002,
+      "loss": 1.2548,
+      "step": 6670
+    },
+    {
+      "epoch": 5.6205300799326885,
+      "grad_norm": 0.9810206890106201,
+      "learning_rate": 0.0002,
+      "loss": 1.3451,
+      "step": 6680
+    },
+    {
+      "epoch": 5.628944047118217,
+      "grad_norm": 1.100386619567871,
+      "learning_rate": 0.0002,
+      "loss": 1.2952,
+      "step": 6690
+    },
+    {
+      "epoch": 5.637358014303744,
+      "grad_norm": 0.8824519515037537,
+      "learning_rate": 0.0002,
+      "loss": 1.2467,
+      "step": 6700
+    },
+    {
+      "epoch": 5.645771981489272,
+      "grad_norm": 1.0864064693450928,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 6710
+    },
+    {
+      "epoch": 5.6541859486748,
+      "grad_norm": 1.1614511013031006,
+      "learning_rate": 0.0002,
+      "loss": 1.2479,
+      "step": 6720
+    },
+    {
+      "epoch": 5.662599915860328,
+      "grad_norm": 1.0762972831726074,
+      "learning_rate": 0.0002,
+      "loss": 1.2753,
+      "step": 6730
+    },
+    {
+      "epoch": 5.671013883045856,
+      "grad_norm": 0.9408974647521973,
+      "learning_rate": 0.0002,
+      "loss": 1.2741,
+      "step": 6740
+    },
+    {
+      "epoch": 5.679427850231384,
+      "grad_norm": 0.8906030058860779,
+      "learning_rate": 0.0002,
+      "loss": 1.2431,
+      "step": 6750
+    },
+    {
+      "epoch": 5.687841817416912,
+      "grad_norm": 0.9527303576469421,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 6760
+    },
+    {
+      "epoch": 5.69625578460244,
+      "grad_norm": 0.9471196532249451,
+      "learning_rate": 0.0002,
+      "loss": 1.322,
+      "step": 6770
+    },
+    {
+      "epoch": 5.704669751787968,
+      "grad_norm": 0.9186838865280151,
+      "learning_rate": 0.0002,
+      "loss": 1.2514,
+      "step": 6780
+    },
+    {
+      "epoch": 5.713083718973496,
+      "grad_norm": 0.9225441813468933,
+      "learning_rate": 0.0002,
+      "loss": 1.2347,
+      "step": 6790
+    },
+    {
+      "epoch": 5.721497686159024,
+      "grad_norm": 0.9712982773780823,
+      "learning_rate": 0.0002,
+      "loss": 1.1849,
+      "step": 6800
+    },
+    {
+      "epoch": 5.729911653344552,
+      "grad_norm": 1.0743170976638794,
+      "learning_rate": 0.0002,
+      "loss": 1.2431,
+      "step": 6810
+    },
+    {
+      "epoch": 5.73832562053008,
+      "grad_norm": 1.2738113403320312,
+      "learning_rate": 0.0002,
+      "loss": 1.2136,
+      "step": 6820
+    },
+    {
+      "epoch": 5.7467395877156076,
+      "grad_norm": 0.9386790990829468,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 6830
+    },
+    {
+      "epoch": 5.755153554901136,
+      "grad_norm": 1.0817769765853882,
+      "learning_rate": 0.0002,
+      "loss": 1.285,
+      "step": 6840
+    },
+    {
+      "epoch": 5.763567522086664,
+      "grad_norm": 1.1040263175964355,
+      "learning_rate": 0.0002,
+      "loss": 1.2247,
+      "step": 6850
+    },
+    {
+      "epoch": 5.771981489272192,
+      "grad_norm": 1.0656492710113525,
+      "learning_rate": 0.0002,
+      "loss": 1.2507,
+      "step": 6860
+    },
+    {
+      "epoch": 5.78039545645772,
+      "grad_norm": 0.9550157189369202,
+      "learning_rate": 0.0002,
+      "loss": 1.2999,
+      "step": 6870
+    },
+    {
+      "epoch": 5.788809423643248,
+      "grad_norm": 1.0130870342254639,
+      "learning_rate": 0.0002,
+      "loss": 1.3201,
+      "step": 6880
+    },
+    {
+      "epoch": 5.797223390828776,
+      "grad_norm": 1.0675787925720215,
+      "learning_rate": 0.0002,
+      "loss": 1.3392,
+      "step": 6890
+    },
+    {
+      "epoch": 5.8056373580143035,
+      "grad_norm": 0.9537774920463562,
+      "learning_rate": 0.0002,
+      "loss": 1.2949,
+      "step": 6900
+    },
+    {
+      "epoch": 5.814051325199832,
+      "grad_norm": 0.9640319347381592,
+      "learning_rate": 0.0002,
+      "loss": 1.2658,
+      "step": 6910
+    },
+    {
+      "epoch": 5.82246529238536,
+      "grad_norm": 0.8917992115020752,
+      "learning_rate": 0.0002,
+      "loss": 1.2199,
+      "step": 6920
+    },
+    {
+      "epoch": 5.830879259570888,
+      "grad_norm": 0.9881822466850281,
+      "learning_rate": 0.0002,
+      "loss": 1.373,
+      "step": 6930
+    },
+    {
+      "epoch": 5.839293226756416,
+      "grad_norm": 0.9136882424354553,
+      "learning_rate": 0.0002,
+      "loss": 1.323,
+      "step": 6940
+    },
+    {
+      "epoch": 5.847707193941943,
+      "grad_norm": 0.9086098074913025,
+      "learning_rate": 0.0002,
+      "loss": 1.3159,
+      "step": 6950
+    },
+    {
+      "epoch": 5.856121161127471,
+      "grad_norm": 0.9443018436431885,
+      "learning_rate": 0.0002,
+      "loss": 1.2624,
+      "step": 6960
+    },
+    {
+      "epoch": 5.864535128312999,
+      "grad_norm": 0.9915381669998169,
+      "learning_rate": 0.0002,
+      "loss": 1.3224,
+      "step": 6970
+    },
+    {
+      "epoch": 5.8729490954985275,
+      "grad_norm": 0.8939146995544434,
+      "learning_rate": 0.0002,
+      "loss": 1.337,
+      "step": 6980
+    },
+    {
+      "epoch": 5.881363062684056,
+      "grad_norm": 1.3672245740890503,
+      "learning_rate": 0.0002,
+      "loss": 1.2611,
+      "step": 6990
+    },
+    {
+      "epoch": 5.889777029869584,
+      "grad_norm": 1.0116257667541504,
+      "learning_rate": 0.0002,
+      "loss": 1.3012,
+      "step": 7000
+    },
+    {
+      "epoch": 5.898190997055112,
+      "grad_norm": 1.1561565399169922,
+      "learning_rate": 0.0002,
+      "loss": 1.3128,
+      "step": 7010
+    },
+    {
+      "epoch": 5.906604964240639,
+      "grad_norm": 0.9900678992271423,
+      "learning_rate": 0.0002,
+      "loss": 1.2301,
+      "step": 7020
+    },
+    {
+      "epoch": 5.915018931426167,
+      "grad_norm": 0.9297345876693726,
+      "learning_rate": 0.0002,
+      "loss": 1.2845,
+      "step": 7030
+    },
+    {
+      "epoch": 5.923432898611695,
+      "grad_norm": 0.9357825517654419,
+      "learning_rate": 0.0002,
+      "loss": 1.2317,
+      "step": 7040
+    },
+    {
+      "epoch": 5.931846865797223,
+      "grad_norm": 1.049317717552185,
+      "learning_rate": 0.0002,
+      "loss": 1.2303,
+      "step": 7050
+    },
+    {
+      "epoch": 5.9402608329827515,
+      "grad_norm": 0.950633704662323,
+      "learning_rate": 0.0002,
+      "loss": 1.3243,
+      "step": 7060
+    },
+    {
+      "epoch": 5.94867480016828,
+      "grad_norm": 0.854581892490387,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 7070
+    },
+    {
+      "epoch": 5.957088767353808,
+      "grad_norm": 0.9097039699554443,
+      "learning_rate": 0.0002,
+      "loss": 1.3252,
+      "step": 7080
+    },
+    {
+      "epoch": 5.965502734539335,
+      "grad_norm": 0.9072173237800598,
+      "learning_rate": 0.0002,
+      "loss": 1.291,
+      "step": 7090
+    },
+    {
+      "epoch": 5.973916701724863,
+      "grad_norm": 1.0470727682113647,
+      "learning_rate": 0.0002,
+      "loss": 1.2724,
+      "step": 7100
+    },
+    {
+      "epoch": 5.982330668910391,
+      "grad_norm": 1.2628462314605713,
+      "learning_rate": 0.0002,
+      "loss": 1.3324,
+      "step": 7110
+    },
+    {
+      "epoch": 5.990744636095919,
+      "grad_norm": 1.055279016494751,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 7120
+    },
+    {
+      "epoch": 5.9991586032814475,
+      "grad_norm": 0.966194212436676,
+      "learning_rate": 0.0002,
+      "loss": 1.3234,
+      "step": 7130
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.0427448749542236,
+      "eval_runtime": 37.8426,
+      "eval_samples_per_second": 13.609,
+      "eval_steps_per_second": 1.718,
+      "step": 7131
+    },
+    {
+      "epoch": 6.007572570466976,
+      "grad_norm": 1.4037928581237793,
+      "learning_rate": 0.0002,
+      "loss": 1.1308,
+      "step": 7140
+    },
+    {
+      "epoch": 6.015986537652503,
+      "grad_norm": 1.1081010103225708,
+      "learning_rate": 0.0002,
+      "loss": 1.047,
+      "step": 7150
+    },
+    {
+      "epoch": 6.024400504838031,
+      "grad_norm": 1.1585499048233032,
+      "learning_rate": 0.0002,
+      "loss": 1.1368,
+      "step": 7160
+    },
+    {
+      "epoch": 6.032814472023559,
+      "grad_norm": 1.0822780132293701,
+      "learning_rate": 0.0002,
+      "loss": 1.0192,
+      "step": 7170
+    },
+    {
+      "epoch": 6.041228439209087,
+      "grad_norm": 0.9662094712257385,
+      "learning_rate": 0.0002,
+      "loss": 1.0755,
+      "step": 7180
+    },
+    {
+      "epoch": 6.049642406394615,
+      "grad_norm": 1.063936710357666,
+      "learning_rate": 0.0002,
+      "loss": 1.1366,
+      "step": 7190
+    },
+    {
+      "epoch": 6.058056373580143,
+      "grad_norm": 1.0349032878875732,
+      "learning_rate": 0.0002,
+      "loss": 1.0121,
+      "step": 7200
+    },
+    {
+      "epoch": 6.066470340765671,
+      "grad_norm": 1.0312575101852417,
+      "learning_rate": 0.0002,
+      "loss": 1.0591,
+      "step": 7210
+    },
+    {
+      "epoch": 6.074884307951199,
+      "grad_norm": 1.1942846775054932,
+      "learning_rate": 0.0002,
+      "loss": 1.1824,
+      "step": 7220
+    },
+    {
+      "epoch": 6.083298275136727,
+      "grad_norm": 1.0816049575805664,
+      "learning_rate": 0.0002,
+      "loss": 1.1034,
+      "step": 7230
+    },
+    {
+      "epoch": 6.091712242322255,
+      "grad_norm": 0.9985513687133789,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 7240
+    },
+    {
+      "epoch": 6.100126209507783,
+      "grad_norm": 1.2573972940444946,
+      "learning_rate": 0.0002,
+      "loss": 1.0367,
+      "step": 7250
+    },
+    {
+      "epoch": 6.108540176693311,
+      "grad_norm": 1.1182395219802856,
+      "learning_rate": 0.0002,
+      "loss": 1.1051,
+      "step": 7260
+    },
+    {
+      "epoch": 6.116954143878839,
+      "grad_norm": 0.9679344296455383,
+      "learning_rate": 0.0002,
+      "loss": 1.1219,
+      "step": 7270
+    },
+    {
+      "epoch": 6.1253681110643665,
+      "grad_norm": 1.0913981199264526,
+      "learning_rate": 0.0002,
+      "loss": 1.1192,
+      "step": 7280
+    },
+    {
+      "epoch": 6.133782078249895,
+      "grad_norm": 1.1291013956069946,
+      "learning_rate": 0.0002,
+      "loss": 1.0411,
+      "step": 7290
+    },
+    {
+      "epoch": 6.142196045435423,
+      "grad_norm": 1.2679595947265625,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 7300
+    },
+    {
+      "epoch": 6.150610012620951,
+      "grad_norm": 1.2350026369094849,
+      "learning_rate": 0.0002,
+      "loss": 1.0875,
+      "step": 7310
+    },
+    {
+      "epoch": 6.159023979806479,
+      "grad_norm": 1.3213104009628296,
+      "learning_rate": 0.0002,
+      "loss": 1.1139,
+      "step": 7320
+    },
+    {
+      "epoch": 6.167437946992007,
+      "grad_norm": 1.1924850940704346,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 7330
+    },
+    {
+      "epoch": 6.175851914177534,
+      "grad_norm": 1.1890000104904175,
+      "learning_rate": 0.0002,
+      "loss": 1.1242,
+      "step": 7340
+    },
+    {
+      "epoch": 6.184265881363062,
+      "grad_norm": 1.3821455240249634,
+      "learning_rate": 0.0002,
+      "loss": 1.1341,
+      "step": 7350
+    },
+    {
+      "epoch": 6.1926798485485905,
+      "grad_norm": 1.1217057704925537,
+      "learning_rate": 0.0002,
+      "loss": 1.0748,
+      "step": 7360
+    },
+    {
+      "epoch": 6.201093815734119,
+      "grad_norm": 1.2441548109054565,
+      "learning_rate": 0.0002,
+      "loss": 1.159,
+      "step": 7370
+    },
+    {
+      "epoch": 6.209507782919647,
+      "grad_norm": 1.0837615728378296,
+      "learning_rate": 0.0002,
+      "loss": 1.1199,
+      "step": 7380
+    },
+    {
+      "epoch": 6.217921750105175,
+      "grad_norm": 1.164304256439209,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 7390
+    },
+    {
+      "epoch": 6.226335717290702,
+      "grad_norm": 1.3129467964172363,
+      "learning_rate": 0.0002,
+      "loss": 1.1325,
+      "step": 7400
+    },
+    {
+      "epoch": 6.23474968447623,
+      "grad_norm": 1.1938153505325317,
+      "learning_rate": 0.0002,
+      "loss": 1.1537,
+      "step": 7410
+    },
+    {
+      "epoch": 6.243163651661758,
+      "grad_norm": 1.4348443746566772,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 7420
+    },
+    {
+      "epoch": 6.2515776188472865,
+      "grad_norm": 1.132301926612854,
+      "learning_rate": 0.0002,
+      "loss": 1.0778,
+      "step": 7430
+    },
+    {
+      "epoch": 6.259991586032815,
+      "grad_norm": 1.136966586112976,
+      "learning_rate": 0.0002,
+      "loss": 1.1148,
+      "step": 7440
+    },
+    {
+      "epoch": 6.268405553218343,
+      "grad_norm": 1.12801194190979,
+      "learning_rate": 0.0002,
+      "loss": 1.096,
+      "step": 7450
+    },
+    {
+      "epoch": 6.276819520403871,
+      "grad_norm": 1.0246902704238892,
+      "learning_rate": 0.0002,
+      "loss": 1.0408,
+      "step": 7460
+    },
+    {
+      "epoch": 6.285233487589398,
+      "grad_norm": 1.1066974401474,
+      "learning_rate": 0.0002,
+      "loss": 1.0389,
+      "step": 7470
+    },
+    {
+      "epoch": 6.293647454774926,
+      "grad_norm": 1.012710690498352,
+      "learning_rate": 0.0002,
+      "loss": 1.1589,
+      "step": 7480
+    },
+    {
+      "epoch": 6.302061421960454,
+      "grad_norm": 1.2227119207382202,
+      "learning_rate": 0.0002,
+      "loss": 1.1049,
+      "step": 7490
+    },
+    {
+      "epoch": 6.310475389145982,
+      "grad_norm": 0.9736923575401306,
+      "learning_rate": 0.0002,
+      "loss": 1.1376,
+      "step": 7500
+    },
+    {
+      "epoch": 6.3188893563315105,
+      "grad_norm": 1.2945268154144287,
+      "learning_rate": 0.0002,
+      "loss": 1.1017,
+      "step": 7510
+    },
+    {
+      "epoch": 6.327303323517039,
+      "grad_norm": 1.1579312086105347,
+      "learning_rate": 0.0002,
+      "loss": 1.0724,
+      "step": 7520
+    },
+    {
+      "epoch": 6.335717290702567,
+      "grad_norm": 1.2404558658599854,
+      "learning_rate": 0.0002,
+      "loss": 1.0899,
+      "step": 7530
+    },
+    {
+      "epoch": 6.344131257888094,
+      "grad_norm": 1.4673258066177368,
+      "learning_rate": 0.0002,
+      "loss": 1.1635,
+      "step": 7540
+    },
+    {
+      "epoch": 6.352545225073622,
+      "grad_norm": 1.2268997430801392,
+      "learning_rate": 0.0002,
+      "loss": 1.128,
+      "step": 7550
+    },
+    {
+      "epoch": 6.36095919225915,
+      "grad_norm": 0.9772747159004211,
+      "learning_rate": 0.0002,
+      "loss": 1.0932,
+      "step": 7560
+    },
+    {
+      "epoch": 6.369373159444678,
+      "grad_norm": 1.0205204486846924,
+      "learning_rate": 0.0002,
+      "loss": 1.1214,
+      "step": 7570
+    },
+    {
+      "epoch": 6.377787126630206,
+      "grad_norm": 1.2227109670639038,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 7580
+    },
+    {
+      "epoch": 6.3862010938157345,
+      "grad_norm": 1.0708507299423218,
+      "learning_rate": 0.0002,
+      "loss": 1.1115,
+      "step": 7590
+    },
+    {
+      "epoch": 6.394615061001262,
+      "grad_norm": 1.1427522897720337,
+      "learning_rate": 0.0002,
+      "loss": 1.1018,
+      "step": 7600
+    },
+    {
+      "epoch": 6.40302902818679,
+      "grad_norm": 1.0706431865692139,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 7610
+    },
+    {
+      "epoch": 6.411442995372318,
+      "grad_norm": 1.1358282566070557,
+      "learning_rate": 0.0002,
+      "loss": 1.0933,
+      "step": 7620
+    },
+    {
+      "epoch": 6.419856962557846,
+      "grad_norm": 1.4011822938919067,
+      "learning_rate": 0.0002,
+      "loss": 1.1075,
+      "step": 7630
+    },
+    {
+      "epoch": 6.428270929743374,
+      "grad_norm": 1.5616450309753418,
+      "learning_rate": 0.0002,
+      "loss": 1.1269,
+      "step": 7640
+    },
+    {
+      "epoch": 6.436684896928902,
+      "grad_norm": 1.1442687511444092,
+      "learning_rate": 0.0002,
+      "loss": 1.0953,
+      "step": 7650
+    },
+    {
+      "epoch": 6.44509886411443,
+      "grad_norm": 1.164803147315979,
+      "learning_rate": 0.0002,
+      "loss": 1.1341,
+      "step": 7660
+    },
+    {
+      "epoch": 6.453512831299958,
+      "grad_norm": 1.3184553384780884,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 7670
+    },
+    {
+      "epoch": 6.461926798485486,
+      "grad_norm": 1.2701894044876099,
+      "learning_rate": 0.0002,
+      "loss": 1.1526,
+      "step": 7680
+    },
+    {
+      "epoch": 6.470340765671014,
+      "grad_norm": 1.1998416185379028,
+      "learning_rate": 0.0002,
+      "loss": 1.2119,
+      "step": 7690
+    },
+    {
+      "epoch": 6.478754732856542,
+      "grad_norm": 1.156459927558899,
+      "learning_rate": 0.0002,
+      "loss": 1.1528,
+      "step": 7700
+    },
+    {
+      "epoch": 6.48716870004207,
+      "grad_norm": 1.0217190980911255,
+      "learning_rate": 0.0002,
+      "loss": 1.2122,
+      "step": 7710
+    },
+    {
+      "epoch": 6.495582667227598,
+      "grad_norm": 1.230372428894043,
+      "learning_rate": 0.0002,
+      "loss": 1.0917,
+      "step": 7720
+    },
+    {
+      "epoch": 6.5039966344131255,
+      "grad_norm": 1.105675220489502,
+      "learning_rate": 0.0002,
+      "loss": 1.119,
+      "step": 7730
+    },
+    {
+      "epoch": 6.512410601598654,
+      "grad_norm": 1.1623669862747192,
+      "learning_rate": 0.0002,
+      "loss": 1.0758,
+      "step": 7740
+    },
+    {
+      "epoch": 6.520824568784182,
+      "grad_norm": 1.2884684801101685,
+      "learning_rate": 0.0002,
+      "loss": 1.1548,
+      "step": 7750
+    },
+    {
+      "epoch": 6.52923853596971,
+      "grad_norm": 1.1785279512405396,
+      "learning_rate": 0.0002,
+      "loss": 1.142,
+      "step": 7760
+    },
+    {
+      "epoch": 6.537652503155238,
+      "grad_norm": 1.0607101917266846,
+      "learning_rate": 0.0002,
+      "loss": 1.1598,
+      "step": 7770
+    },
+    {
+      "epoch": 6.546066470340766,
+      "grad_norm": 1.21990168094635,
+      "learning_rate": 0.0002,
+      "loss": 1.1472,
+      "step": 7780
+    },
+    {
+      "epoch": 6.554480437526293,
+      "grad_norm": 1.1498621702194214,
+      "learning_rate": 0.0002,
+      "loss": 1.1468,
+      "step": 7790
+    },
+    {
+      "epoch": 6.562894404711821,
+      "grad_norm": 1.263929009437561,
+      "learning_rate": 0.0002,
+      "loss": 1.1847,
+      "step": 7800
+    },
+    {
+      "epoch": 6.5713083718973495,
+      "grad_norm": 1.1580625772476196,
+      "learning_rate": 0.0002,
+      "loss": 1.1177,
+      "step": 7810
+    },
+    {
+      "epoch": 6.579722339082878,
+      "grad_norm": 1.4431294202804565,
+      "learning_rate": 0.0002,
+      "loss": 1.1313,
+      "step": 7820
+    },
+    {
+      "epoch": 6.588136306268406,
+      "grad_norm": 1.1309990882873535,
+      "learning_rate": 0.0002,
+      "loss": 1.1944,
+      "step": 7830
+    },
+    {
+      "epoch": 6.596550273453934,
+      "grad_norm": 1.0543386936187744,
+      "learning_rate": 0.0002,
+      "loss": 1.1156,
+      "step": 7840
+    },
+    {
+      "epoch": 6.604964240639461,
+      "grad_norm": 1.2180639505386353,
+      "learning_rate": 0.0002,
+      "loss": 1.0945,
+      "step": 7850
+    },
+    {
+      "epoch": 6.613378207824989,
+      "grad_norm": 1.0631271600723267,
+      "learning_rate": 0.0002,
+      "loss": 1.1318,
+      "step": 7860
+    },
+    {
+      "epoch": 6.621792175010517,
+      "grad_norm": 1.138885498046875,
+      "learning_rate": 0.0002,
+      "loss": 1.1792,
+      "step": 7870
+    },
+    {
+      "epoch": 6.630206142196045,
+      "grad_norm": 1.1117745637893677,
+      "learning_rate": 0.0002,
+      "loss": 1.1805,
+      "step": 7880
+    },
+    {
+      "epoch": 6.6386201093815735,
+      "grad_norm": 1.3734886646270752,
+      "learning_rate": 0.0002,
+      "loss": 1.15,
+      "step": 7890
+    },
+    {
+      "epoch": 6.647034076567102,
+      "grad_norm": 1.236003041267395,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 7900
+    },
+    {
+      "epoch": 6.65544804375263,
+      "grad_norm": 1.2206000089645386,
+      "learning_rate": 0.0002,
+      "loss": 1.1718,
+      "step": 7910
+    },
+    {
+      "epoch": 6.663862010938157,
+      "grad_norm": 1.2842656373977661,
+      "learning_rate": 0.0002,
+      "loss": 1.1637,
+      "step": 7920
+    },
+    {
+      "epoch": 6.672275978123685,
+      "grad_norm": 1.2365005016326904,
+      "learning_rate": 0.0002,
+      "loss": 1.2219,
+      "step": 7930
+    },
+    {
+      "epoch": 6.680689945309213,
+      "grad_norm": 1.256620168685913,
+      "learning_rate": 0.0002,
+      "loss": 1.0827,
+      "step": 7940
+    },
+    {
+      "epoch": 6.689103912494741,
+      "grad_norm": 1.3232917785644531,
+      "learning_rate": 0.0002,
+      "loss": 1.1788,
+      "step": 7950
+    },
+    {
+      "epoch": 6.6975178796802695,
+      "grad_norm": 1.2470088005065918,
+      "learning_rate": 0.0002,
+      "loss": 1.2042,
+      "step": 7960
+    },
+    {
+      "epoch": 6.705931846865798,
+      "grad_norm": 1.0511926412582397,
+      "learning_rate": 0.0002,
+      "loss": 1.0959,
+      "step": 7970
+    },
+    {
+      "epoch": 6.714345814051326,
+      "grad_norm": 1.107310175895691,
+      "learning_rate": 0.0002,
+      "loss": 1.118,
+      "step": 7980
+    },
+    {
+      "epoch": 6.722759781236853,
+      "grad_norm": 1.4069843292236328,
+      "learning_rate": 0.0002,
+      "loss": 1.2109,
+      "step": 7990
+    },
+    {
+      "epoch": 6.731173748422381,
+      "grad_norm": 1.0800836086273193,
+      "learning_rate": 0.0002,
+      "loss": 1.1298,
+      "step": 8000
+    },
+    {
+      "epoch": 6.739587715607909,
+      "grad_norm": 1.1676300764083862,
+      "learning_rate": 0.0002,
+      "loss": 1.1824,
+      "step": 8010
+    },
+    {
+      "epoch": 6.748001682793437,
+      "grad_norm": 1.0579663515090942,
+      "learning_rate": 0.0002,
+      "loss": 1.1253,
+      "step": 8020
+    },
+    {
+      "epoch": 6.756415649978965,
+      "grad_norm": 1.2770029306411743,
+      "learning_rate": 0.0002,
+      "loss": 1.1542,
+      "step": 8030
+    },
+    {
+      "epoch": 6.764829617164493,
+      "grad_norm": 1.0981038808822632,
+      "learning_rate": 0.0002,
+      "loss": 1.1519,
+      "step": 8040
+    },
+    {
+      "epoch": 6.773243584350021,
+      "grad_norm": 1.1194742918014526,
+      "learning_rate": 0.0002,
+      "loss": 1.1422,
+      "step": 8050
+    },
+    {
+      "epoch": 6.781657551535549,
+      "grad_norm": 1.0130012035369873,
+      "learning_rate": 0.0002,
+      "loss": 1.1463,
+      "step": 8060
+    },
+    {
+      "epoch": 6.790071518721077,
+      "grad_norm": 1.2051167488098145,
+      "learning_rate": 0.0002,
+      "loss": 1.2008,
+      "step": 8070
+    },
+    {
+      "epoch": 6.798485485906605,
+      "grad_norm": 1.095689058303833,
+      "learning_rate": 0.0002,
+      "loss": 1.142,
+      "step": 8080
+    },
+    {
+      "epoch": 6.806899453092133,
+      "grad_norm": 1.2275174856185913,
+      "learning_rate": 0.0002,
+      "loss": 1.1352,
+      "step": 8090
+    },
+    {
+      "epoch": 6.815313420277661,
+      "grad_norm": 1.1439805030822754,
+      "learning_rate": 0.0002,
+      "loss": 1.1453,
+      "step": 8100
+    },
+    {
+      "epoch": 6.8237273874631885,
+      "grad_norm": 1.276331901550293,
+      "learning_rate": 0.0002,
+      "loss": 1.1624,
+      "step": 8110
+    },
+    {
+      "epoch": 6.832141354648717,
+      "grad_norm": 1.0450139045715332,
+      "learning_rate": 0.0002,
+      "loss": 1.1686,
+      "step": 8120
+    },
+    {
+      "epoch": 6.840555321834245,
+      "grad_norm": 1.1189453601837158,
+      "learning_rate": 0.0002,
+      "loss": 1.1783,
+      "step": 8130
+    },
+    {
+      "epoch": 6.848969289019773,
+      "grad_norm": 1.194640874862671,
+      "learning_rate": 0.0002,
+      "loss": 1.1093,
+      "step": 8140
+    },
+    {
+      "epoch": 6.857383256205301,
+      "grad_norm": 1.095372200012207,
+      "learning_rate": 0.0002,
+      "loss": 1.1559,
+      "step": 8150
+    },
+    {
+      "epoch": 6.865797223390829,
+      "grad_norm": 1.2416104078292847,
+      "learning_rate": 0.0002,
+      "loss": 1.165,
+      "step": 8160
+    },
+    {
+      "epoch": 6.874211190576357,
+      "grad_norm": 1.2402868270874023,
+      "learning_rate": 0.0002,
+      "loss": 1.2174,
+      "step": 8170
+    },
+    {
+      "epoch": 6.882625157761884,
+      "grad_norm": 1.1317291259765625,
+      "learning_rate": 0.0002,
+      "loss": 1.1306,
+      "step": 8180
+    },
+    {
+      "epoch": 6.8910391249474126,
+      "grad_norm": 1.0581914186477661,
+      "learning_rate": 0.0002,
+      "loss": 1.1944,
+      "step": 8190
+    },
+    {
+      "epoch": 6.899453092132941,
+      "grad_norm": 1.3540890216827393,
+      "learning_rate": 0.0002,
+      "loss": 1.1271,
+      "step": 8200
+    },
+    {
+      "epoch": 6.907867059318469,
+      "grad_norm": 1.213672399520874,
+      "learning_rate": 0.0002,
+      "loss": 1.2119,
+      "step": 8210
+    },
+    {
+      "epoch": 6.916281026503997,
+      "grad_norm": 1.2654485702514648,
+      "learning_rate": 0.0002,
+      "loss": 1.1406,
+      "step": 8220
+    },
+    {
+      "epoch": 6.924694993689524,
+      "grad_norm": 1.203903317451477,
+      "learning_rate": 0.0002,
+      "loss": 1.205,
+      "step": 8230
+    },
+    {
+      "epoch": 6.933108960875052,
+      "grad_norm": 1.1332030296325684,
+      "learning_rate": 0.0002,
+      "loss": 1.1635,
+      "step": 8240
+    },
+    {
+      "epoch": 6.94152292806058,
+      "grad_norm": 1.2699192762374878,
+      "learning_rate": 0.0002,
+      "loss": 1.1148,
+      "step": 8250
+    },
+    {
+      "epoch": 6.9499368952461085,
+      "grad_norm": 1.2728958129882812,
+      "learning_rate": 0.0002,
+      "loss": 1.1831,
+      "step": 8260
+    },
+    {
+      "epoch": 6.958350862431637,
+      "grad_norm": 1.238410472869873,
+      "learning_rate": 0.0002,
+      "loss": 1.1757,
+      "step": 8270
+    },
+    {
+      "epoch": 6.966764829617165,
+      "grad_norm": 1.403863549232483,
+      "learning_rate": 0.0002,
+      "loss": 1.1499,
+      "step": 8280
+    },
+    {
+      "epoch": 6.975178796802693,
+      "grad_norm": 1.1096396446228027,
+      "learning_rate": 0.0002,
+      "loss": 1.1515,
+      "step": 8290
+    },
+    {
+      "epoch": 6.98359276398822,
+      "grad_norm": 1.1043379306793213,
+      "learning_rate": 0.0002,
+      "loss": 1.2049,
+      "step": 8300
+    },
+    {
+      "epoch": 6.992006731173748,
+      "grad_norm": 1.391754388809204,
+      "learning_rate": 0.0002,
+      "loss": 1.1255,
+      "step": 8310
+    },
+    {
+      "epoch": 6.999579301640724,
+      "eval_loss": 2.1421656608581543,
+      "eval_runtime": 37.8262,
+      "eval_samples_per_second": 13.615,
+      "eval_steps_per_second": 1.718,
+      "step": 8319
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 9504,
+  "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": 3.850078318850212e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/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-0.8-num-5623-sd-42/checkpoint-8319/training_args.bin b/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-0.8-num-5623-sd-42/checkpoint-8319/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f1502d478cfbb1424f707352d007b740bde5e373
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-8319/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df2b79d3acefeedef5a0229881de39ec68ef9b40046a60d7976a49f7e6b3b936
+size 5560
diff --git a/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-0.8-num-5623-sd-42/checkpoint-9504/README.md b/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-0.8-num-5623-sd-42/checkpoint-9504/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-9504/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_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-0.8-num-5623-sd-42/checkpoint-9504/adapter_config.json b/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-0.8-num-5623-sd-42/checkpoint-9504/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-9504/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_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-0.8-num-5623-sd-42/checkpoint-9504/adapter_model.safetensors b/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-0.8-num-5623-sd-42/checkpoint-9504/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..877a1445c41083bf3775e855f5f8d04a996d35e4
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-9504/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1550273ea68bc5a4e788e74c1c78bf524ec1b61043bf5f20b357927ff807a309
+size 109069176
diff --git a/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-0.8-num-5623-sd-42/checkpoint-9504/optimizer.pt b/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-0.8-num-5623-sd-42/checkpoint-9504/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0c78b121fcf1c09de960a5ad8805a36487a4ebbd
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-9504/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f50e70ec2342f84423f0873190aa676fd8b4c8f338b92928987c390474e1f228
+size 55532666
diff --git a/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-0.8-num-5623-sd-42/checkpoint-9504/rng_state.pth b/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-0.8-num-5623-sd-42/checkpoint-9504/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..466d6085e3ccf304eda44eb2738464a7b9bafcd5
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-9504/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92900cc70e521535f52ac8e843c2753bcf71d5073db21bdc901368b594c00657
+size 14244
diff --git a/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-0.8-num-5623-sd-42/checkpoint-9504/scheduler.pt b/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-0.8-num-5623-sd-42/checkpoint-9504/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..affbf7ce62b8cef2cd803c35235c37107c3a7913
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-9504/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f92060c6b01adb16ebce9a8e91b1e7112c58c4535a075036e50137831397c75
+size 1064
diff --git a/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-0.8-num-5623-sd-42/checkpoint-9504/special_tokens_map.json b/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-0.8-num-5623-sd-42/checkpoint-9504/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-9504/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_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-0.8-num-5623-sd-42/checkpoint-9504/tokenizer.json b/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-0.8-num-5623-sd-42/checkpoint-9504/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-9504/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_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-0.8-num-5623-sd-42/checkpoint-9504/tokenizer_config.json b/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-0.8-num-5623-sd-42/checkpoint-9504/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-9504/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_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-0.8-num-5623-sd-42/checkpoint-9504/trainer_state.json b/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-0.8-num-5623-sd-42/checkpoint-9504/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..6be0e4e303145eb3bbd12fbdaf5eb56198a31cd4
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-9504/trainer_state.json
@@ -0,0 +1,6747 @@
+{
+  "best_metric": 1.8055059909820557,
+  "best_model_checkpoint": "outputs-001/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-0.8-num-5623-sd-42/checkpoint-2377",
+  "epoch": 7.996634413125789,
+  "eval_steps": 10,
+  "global_step": 9504,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008413967185527976,
+      "grad_norm": 0.5458821654319763,
+      "learning_rate": 0.0002,
+      "loss": 2.56,
+      "step": 10
+    },
+    {
+      "epoch": 0.016827934371055953,
+      "grad_norm": 0.7293308973312378,
+      "learning_rate": 0.0002,
+      "loss": 2.3235,
+      "step": 20
+    },
+    {
+      "epoch": 0.02524190155658393,
+      "grad_norm": 0.47792306542396545,
+      "learning_rate": 0.0002,
+      "loss": 2.0815,
+      "step": 30
+    },
+    {
+      "epoch": 0.033655868742111905,
+      "grad_norm": 0.5944402813911438,
+      "learning_rate": 0.0002,
+      "loss": 1.9718,
+      "step": 40
+    },
+    {
+      "epoch": 0.04206983592763988,
+      "grad_norm": 0.5415359735488892,
+      "learning_rate": 0.0002,
+      "loss": 1.8848,
+      "step": 50
+    },
+    {
+      "epoch": 0.05048380311316786,
+      "grad_norm": 0.535713791847229,
+      "learning_rate": 0.0002,
+      "loss": 1.8953,
+      "step": 60
+    },
+    {
+      "epoch": 0.058897770298695834,
+      "grad_norm": 0.5184146761894226,
+      "learning_rate": 0.0002,
+      "loss": 1.937,
+      "step": 70
+    },
+    {
+      "epoch": 0.06731173748422381,
+      "grad_norm": 0.458926796913147,
+      "learning_rate": 0.0002,
+      "loss": 1.8396,
+      "step": 80
+    },
+    {
+      "epoch": 0.07572570466975179,
+      "grad_norm": 0.4780142307281494,
+      "learning_rate": 0.0002,
+      "loss": 1.8677,
+      "step": 90
+    },
+    {
+      "epoch": 0.08413967185527976,
+      "grad_norm": 0.79965740442276,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 100
+    },
+    {
+      "epoch": 0.09255363904080774,
+      "grad_norm": 0.4498862028121948,
+      "learning_rate": 0.0002,
+      "loss": 1.9081,
+      "step": 110
+    },
+    {
+      "epoch": 0.10096760622633572,
+      "grad_norm": 0.39338430762290955,
+      "learning_rate": 0.0002,
+      "loss": 1.8503,
+      "step": 120
+    },
+    {
+      "epoch": 0.10938157341186369,
+      "grad_norm": 0.9588953852653503,
+      "learning_rate": 0.0002,
+      "loss": 1.8637,
+      "step": 130
+    },
+    {
+      "epoch": 0.11779554059739167,
+      "grad_norm": 0.41675639152526855,
+      "learning_rate": 0.0002,
+      "loss": 1.8676,
+      "step": 140
+    },
+    {
+      "epoch": 0.12620950778291964,
+      "grad_norm": 0.44519832730293274,
+      "learning_rate": 0.0002,
+      "loss": 1.8904,
+      "step": 150
+    },
+    {
+      "epoch": 0.13462347496844762,
+      "grad_norm": 0.4176260530948639,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 160
+    },
+    {
+      "epoch": 0.1430374421539756,
+      "grad_norm": 0.35840365290641785,
+      "learning_rate": 0.0002,
+      "loss": 1.8398,
+      "step": 170
+    },
+    {
+      "epoch": 0.15145140933950357,
+      "grad_norm": 0.3794495463371277,
+      "learning_rate": 0.0002,
+      "loss": 1.8666,
+      "step": 180
+    },
+    {
+      "epoch": 0.15986537652503155,
+      "grad_norm": 0.4563522934913635,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 190
+    },
+    {
+      "epoch": 0.16827934371055953,
+      "grad_norm": 0.37057486176490784,
+      "learning_rate": 0.0002,
+      "loss": 1.8893,
+      "step": 200
+    },
+    {
+      "epoch": 0.1766933108960875,
+      "grad_norm": 0.44081518054008484,
+      "learning_rate": 0.0002,
+      "loss": 1.7995,
+      "step": 210
+    },
+    {
+      "epoch": 0.18510727808161548,
+      "grad_norm": 0.46078577637672424,
+      "learning_rate": 0.0002,
+      "loss": 1.9048,
+      "step": 220
+    },
+    {
+      "epoch": 0.19352124526714345,
+      "grad_norm": 0.36132094264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.8403,
+      "step": 230
+    },
+    {
+      "epoch": 0.20193521245267143,
+      "grad_norm": 0.3747289180755615,
+      "learning_rate": 0.0002,
+      "loss": 1.8827,
+      "step": 240
+    },
+    {
+      "epoch": 0.2103491796381994,
+      "grad_norm": 0.3540179133415222,
+      "learning_rate": 0.0002,
+      "loss": 1.8382,
+      "step": 250
+    },
+    {
+      "epoch": 0.21876314682372738,
+      "grad_norm": 0.3461375832557678,
+      "learning_rate": 0.0002,
+      "loss": 1.8196,
+      "step": 260
+    },
+    {
+      "epoch": 0.22717711400925536,
+      "grad_norm": 0.3436960279941559,
+      "learning_rate": 0.0002,
+      "loss": 1.8509,
+      "step": 270
+    },
+    {
+      "epoch": 0.23559108119478334,
+      "grad_norm": 0.35403719544410706,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 280
+    },
+    {
+      "epoch": 0.2440050483803113,
+      "grad_norm": 0.37142616510391235,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 290
+    },
+    {
+      "epoch": 0.2524190155658393,
+      "grad_norm": 0.3307955861091614,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 300
+    },
+    {
+      "epoch": 0.2608329827513673,
+      "grad_norm": 0.32855314016342163,
+      "learning_rate": 0.0002,
+      "loss": 1.817,
+      "step": 310
+    },
+    {
+      "epoch": 0.26924694993689524,
+      "grad_norm": 0.3299003839492798,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 320
+    },
+    {
+      "epoch": 0.27766091712242325,
+      "grad_norm": 0.44311287999153137,
+      "learning_rate": 0.0002,
+      "loss": 1.8129,
+      "step": 330
+    },
+    {
+      "epoch": 0.2860748843079512,
+      "grad_norm": 0.32989758253097534,
+      "learning_rate": 0.0002,
+      "loss": 1.8232,
+      "step": 340
+    },
+    {
+      "epoch": 0.2944888514934792,
+      "grad_norm": 0.34400200843811035,
+      "learning_rate": 0.0002,
+      "loss": 1.7716,
+      "step": 350
+    },
+    {
+      "epoch": 0.30290281867900715,
+      "grad_norm": 0.36286211013793945,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 360
+    },
+    {
+      "epoch": 0.31131678586453515,
+      "grad_norm": 0.406827837228775,
+      "learning_rate": 0.0002,
+      "loss": 1.8025,
+      "step": 370
+    },
+    {
+      "epoch": 0.3197307530500631,
+      "grad_norm": 0.36299195885658264,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 380
+    },
+    {
+      "epoch": 0.3281447202355911,
+      "grad_norm": 0.3477257192134857,
+      "learning_rate": 0.0002,
+      "loss": 1.837,
+      "step": 390
+    },
+    {
+      "epoch": 0.33655868742111905,
+      "grad_norm": 0.3730369210243225,
+      "learning_rate": 0.0002,
+      "loss": 1.7767,
+      "step": 400
+    },
+    {
+      "epoch": 0.34497265460664706,
+      "grad_norm": 0.4644559919834137,
+      "learning_rate": 0.0002,
+      "loss": 1.7747,
+      "step": 410
+    },
+    {
+      "epoch": 0.353386621792175,
+      "grad_norm": 0.406576544046402,
+      "learning_rate": 0.0002,
+      "loss": 1.7538,
+      "step": 420
+    },
+    {
+      "epoch": 0.361800588977703,
+      "grad_norm": 0.3612699508666992,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 430
+    },
+    {
+      "epoch": 0.37021455616323096,
+      "grad_norm": 0.3243742287158966,
+      "learning_rate": 0.0002,
+      "loss": 1.7473,
+      "step": 440
+    },
+    {
+      "epoch": 0.37862852334875896,
+      "grad_norm": 0.36671221256256104,
+      "learning_rate": 0.0002,
+      "loss": 1.8851,
+      "step": 450
+    },
+    {
+      "epoch": 0.3870424905342869,
+      "grad_norm": 0.3565002381801605,
+      "learning_rate": 0.0002,
+      "loss": 1.8853,
+      "step": 460
+    },
+    {
+      "epoch": 0.3954564577198149,
+      "grad_norm": 0.34630221128463745,
+      "learning_rate": 0.0002,
+      "loss": 1.8923,
+      "step": 470
+    },
+    {
+      "epoch": 0.40387042490534286,
+      "grad_norm": 0.3353537321090698,
+      "learning_rate": 0.0002,
+      "loss": 1.8234,
+      "step": 480
+    },
+    {
+      "epoch": 0.41228439209087087,
+      "grad_norm": 0.4015921950340271,
+      "learning_rate": 0.0002,
+      "loss": 1.7135,
+      "step": 490
+    },
+    {
+      "epoch": 0.4206983592763988,
+      "grad_norm": 0.5489419102668762,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 500
+    },
+    {
+      "epoch": 0.4291123264619268,
+      "grad_norm": 0.4193589985370636,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 510
+    },
+    {
+      "epoch": 0.43752629364745477,
+      "grad_norm": 0.3418922424316406,
+      "learning_rate": 0.0002,
+      "loss": 1.8416,
+      "step": 520
+    },
+    {
+      "epoch": 0.44594026083298277,
+      "grad_norm": 0.32668185234069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 530
+    },
+    {
+      "epoch": 0.4543542280185107,
+      "grad_norm": 0.3094325661659241,
+      "learning_rate": 0.0002,
+      "loss": 1.7501,
+      "step": 540
+    },
+    {
+      "epoch": 0.4627681952040387,
+      "grad_norm": 0.3743017315864563,
+      "learning_rate": 0.0002,
+      "loss": 1.7438,
+      "step": 550
+    },
+    {
+      "epoch": 0.47118216238956667,
+      "grad_norm": 0.3295630216598511,
+      "learning_rate": 0.0002,
+      "loss": 1.8451,
+      "step": 560
+    },
+    {
+      "epoch": 0.4795961295750947,
+      "grad_norm": 1.6124513149261475,
+      "learning_rate": 0.0002,
+      "loss": 1.7529,
+      "step": 570
+    },
+    {
+      "epoch": 0.4880100967606226,
+      "grad_norm": 0.3245585858821869,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 580
+    },
+    {
+      "epoch": 0.49642406394615063,
+      "grad_norm": 0.3332934081554413,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 590
+    },
+    {
+      "epoch": 0.5048380311316786,
+      "grad_norm": 0.3836138844490051,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 600
+    },
+    {
+      "epoch": 0.5132519983172066,
+      "grad_norm": 0.32953888177871704,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 610
+    },
+    {
+      "epoch": 0.5216659655027346,
+      "grad_norm": 0.36291512846946716,
+      "learning_rate": 0.0002,
+      "loss": 1.7729,
+      "step": 620
+    },
+    {
+      "epoch": 0.5300799326882625,
+      "grad_norm": 0.3237783908843994,
+      "learning_rate": 0.0002,
+      "loss": 1.7758,
+      "step": 630
+    },
+    {
+      "epoch": 0.5384938998737905,
+      "grad_norm": 0.38882696628570557,
+      "learning_rate": 0.0002,
+      "loss": 1.8352,
+      "step": 640
+    },
+    {
+      "epoch": 0.5469078670593185,
+      "grad_norm": 0.37821972370147705,
+      "learning_rate": 0.0002,
+      "loss": 1.8624,
+      "step": 650
+    },
+    {
+      "epoch": 0.5553218342448465,
+      "grad_norm": 0.3556285500526428,
+      "learning_rate": 0.0002,
+      "loss": 1.8075,
+      "step": 660
+    },
+    {
+      "epoch": 0.5637358014303744,
+      "grad_norm": 0.347499281167984,
+      "learning_rate": 0.0002,
+      "loss": 1.778,
+      "step": 670
+    },
+    {
+      "epoch": 0.5721497686159024,
+      "grad_norm": 0.3176489472389221,
+      "learning_rate": 0.0002,
+      "loss": 1.8066,
+      "step": 680
+    },
+    {
+      "epoch": 0.5805637358014304,
+      "grad_norm": 0.30220088362693787,
+      "learning_rate": 0.0002,
+      "loss": 1.7257,
+      "step": 690
+    },
+    {
+      "epoch": 0.5889777029869584,
+      "grad_norm": 0.3711601793766022,
+      "learning_rate": 0.0002,
+      "loss": 1.8415,
+      "step": 700
+    },
+    {
+      "epoch": 0.5973916701724863,
+      "grad_norm": 0.3311759829521179,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 710
+    },
+    {
+      "epoch": 0.6058056373580143,
+      "grad_norm": 0.34824270009994507,
+      "learning_rate": 0.0002,
+      "loss": 1.7712,
+      "step": 720
+    },
+    {
+      "epoch": 0.6142196045435423,
+      "grad_norm": 0.29668381810188293,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 730
+    },
+    {
+      "epoch": 0.6226335717290703,
+      "grad_norm": 0.36087489128112793,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 740
+    },
+    {
+      "epoch": 0.6310475389145982,
+      "grad_norm": 0.31590089201927185,
+      "learning_rate": 0.0002,
+      "loss": 1.7956,
+      "step": 750
+    },
+    {
+      "epoch": 0.6394615061001262,
+      "grad_norm": 0.37632957100868225,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 760
+    },
+    {
+      "epoch": 0.6478754732856542,
+      "grad_norm": 0.3360748589038849,
+      "learning_rate": 0.0002,
+      "loss": 1.8499,
+      "step": 770
+    },
+    {
+      "epoch": 0.6562894404711822,
+      "grad_norm": 0.3420640528202057,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 780
+    },
+    {
+      "epoch": 0.6647034076567101,
+      "grad_norm": 0.5734959244728088,
+      "learning_rate": 0.0002,
+      "loss": 1.8353,
+      "step": 790
+    },
+    {
+      "epoch": 0.6731173748422381,
+      "grad_norm": 0.36440837383270264,
+      "learning_rate": 0.0002,
+      "loss": 1.7746,
+      "step": 800
+    },
+    {
+      "epoch": 0.6815313420277661,
+      "grad_norm": 0.3179708421230316,
+      "learning_rate": 0.0002,
+      "loss": 1.7532,
+      "step": 810
+    },
+    {
+      "epoch": 0.6899453092132941,
+      "grad_norm": 0.34122881293296814,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 820
+    },
+    {
+      "epoch": 0.698359276398822,
+      "grad_norm": 0.31886112689971924,
+      "learning_rate": 0.0002,
+      "loss": 1.8167,
+      "step": 830
+    },
+    {
+      "epoch": 0.70677324358435,
+      "grad_norm": 0.31782326102256775,
+      "learning_rate": 0.0002,
+      "loss": 1.7505,
+      "step": 840
+    },
+    {
+      "epoch": 0.715187210769878,
+      "grad_norm": 0.36052989959716797,
+      "learning_rate": 0.0002,
+      "loss": 1.7588,
+      "step": 850
+    },
+    {
+      "epoch": 0.723601177955406,
+      "grad_norm": 0.28946155309677124,
+      "learning_rate": 0.0002,
+      "loss": 1.7891,
+      "step": 860
+    },
+    {
+      "epoch": 0.7320151451409339,
+      "grad_norm": 0.3095663785934448,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 870
+    },
+    {
+      "epoch": 0.7404291123264619,
+      "grad_norm": 0.3317491412162781,
+      "learning_rate": 0.0002,
+      "loss": 1.785,
+      "step": 880
+    },
+    {
+      "epoch": 0.7488430795119899,
+      "grad_norm": 0.31324660778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 890
+    },
+    {
+      "epoch": 0.7572570466975179,
+      "grad_norm": 0.3290475606918335,
+      "learning_rate": 0.0002,
+      "loss": 1.8753,
+      "step": 900
+    },
+    {
+      "epoch": 0.7656710138830458,
+      "grad_norm": 0.35690343379974365,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 910
+    },
+    {
+      "epoch": 0.7740849810685738,
+      "grad_norm": 0.39558273553848267,
+      "learning_rate": 0.0002,
+      "loss": 1.826,
+      "step": 920
+    },
+    {
+      "epoch": 0.7824989482541018,
+      "grad_norm": 0.34254348278045654,
+      "learning_rate": 0.0002,
+      "loss": 1.8722,
+      "step": 930
+    },
+    {
+      "epoch": 0.7909129154396298,
+      "grad_norm": 0.3560165464878082,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 940
+    },
+    {
+      "epoch": 0.7993268826251577,
+      "grad_norm": 0.30693164467811584,
+      "learning_rate": 0.0002,
+      "loss": 1.7992,
+      "step": 950
+    },
+    {
+      "epoch": 0.8077408498106857,
+      "grad_norm": 0.3394823372364044,
+      "learning_rate": 0.0002,
+      "loss": 1.8029,
+      "step": 960
+    },
+    {
+      "epoch": 0.8161548169962137,
+      "grad_norm": 0.3741514980792999,
+      "learning_rate": 0.0002,
+      "loss": 1.8105,
+      "step": 970
+    },
+    {
+      "epoch": 0.8245687841817417,
+      "grad_norm": 0.3655228316783905,
+      "learning_rate": 0.0002,
+      "loss": 1.7849,
+      "step": 980
+    },
+    {
+      "epoch": 0.8329827513672696,
+      "grad_norm": 0.3586033880710602,
+      "learning_rate": 0.0002,
+      "loss": 1.8449,
+      "step": 990
+    },
+    {
+      "epoch": 0.8413967185527976,
+      "grad_norm": 0.3459678888320923,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8498106857383256,
+      "grad_norm": 0.3184349834918976,
+      "learning_rate": 0.0002,
+      "loss": 1.8498,
+      "step": 1010
+    },
+    {
+      "epoch": 0.8582246529238536,
+      "grad_norm": 0.3099786043167114,
+      "learning_rate": 0.0002,
+      "loss": 1.7632,
+      "step": 1020
+    },
+    {
+      "epoch": 0.8666386201093815,
+      "grad_norm": 0.30300915241241455,
+      "learning_rate": 0.0002,
+      "loss": 1.8067,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8750525872949095,
+      "grad_norm": 0.3128705620765686,
+      "learning_rate": 0.0002,
+      "loss": 1.7923,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8834665544804375,
+      "grad_norm": 0.3336263597011566,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8918805216659655,
+      "grad_norm": 0.3801328241825104,
+      "learning_rate": 0.0002,
+      "loss": 1.8375,
+      "step": 1060
+    },
+    {
+      "epoch": 0.9002944888514934,
+      "grad_norm": 0.3122096359729767,
+      "learning_rate": 0.0002,
+      "loss": 1.7757,
+      "step": 1070
+    },
+    {
+      "epoch": 0.9087084560370214,
+      "grad_norm": 0.35990869998931885,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 1080
+    },
+    {
+      "epoch": 0.9171224232225494,
+      "grad_norm": 0.3321819305419922,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1090
+    },
+    {
+      "epoch": 0.9255363904080774,
+      "grad_norm": 0.4202139377593994,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9339503575936053,
+      "grad_norm": 0.32559722661972046,
+      "learning_rate": 0.0002,
+      "loss": 1.8056,
+      "step": 1110
+    },
+    {
+      "epoch": 0.9423643247791333,
+      "grad_norm": 0.3098459839820862,
+      "learning_rate": 0.0002,
+      "loss": 1.812,
+      "step": 1120
+    },
+    {
+      "epoch": 0.9507782919646613,
+      "grad_norm": 0.33917108178138733,
+      "learning_rate": 0.0002,
+      "loss": 1.8252,
+      "step": 1130
+    },
+    {
+      "epoch": 0.9591922591501894,
+      "grad_norm": 0.4055837094783783,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1140
+    },
+    {
+      "epoch": 0.9676062263357172,
+      "grad_norm": 0.32508623600006104,
+      "learning_rate": 0.0002,
+      "loss": 1.8259,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9760201935212452,
+      "grad_norm": 0.30150601267814636,
+      "learning_rate": 0.0002,
+      "loss": 1.782,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9844341607067733,
+      "grad_norm": 0.3042563199996948,
+      "learning_rate": 0.0002,
+      "loss": 1.8291,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9928481278923013,
+      "grad_norm": 0.33254584670066833,
+      "learning_rate": 0.0002,
+      "loss": 1.7847,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9995793016407236,
+      "eval_loss": 1.8077726364135742,
+      "eval_runtime": 38.4359,
+      "eval_samples_per_second": 13.399,
+      "eval_steps_per_second": 1.691,
+      "step": 1188
+    },
+    {
+      "epoch": 1.0012620950778293,
+      "grad_norm": 0.35073035955429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 1190
+    },
+    {
+      "epoch": 1.0096760622633572,
+      "grad_norm": 0.3217269778251648,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1200
+    },
+    {
+      "epoch": 1.018090029448885,
+      "grad_norm": 0.3635033369064331,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1210
+    },
+    {
+      "epoch": 1.0265039966344132,
+      "grad_norm": 0.32468414306640625,
+      "learning_rate": 0.0002,
+      "loss": 1.6949,
+      "step": 1220
+    },
+    {
+      "epoch": 1.034917963819941,
+      "grad_norm": 0.3307163417339325,
+      "learning_rate": 0.0002,
+      "loss": 1.711,
+      "step": 1230
+    },
+    {
+      "epoch": 1.0433319310054692,
+      "grad_norm": 0.34381359815597534,
+      "learning_rate": 0.0002,
+      "loss": 1.7881,
+      "step": 1240
+    },
+    {
+      "epoch": 1.051745898190997,
+      "grad_norm": 0.35874804854393005,
+      "learning_rate": 0.0002,
+      "loss": 1.612,
+      "step": 1250
+    },
+    {
+      "epoch": 1.060159865376525,
+      "grad_norm": 0.3615919351577759,
+      "learning_rate": 0.0002,
+      "loss": 1.7314,
+      "step": 1260
+    },
+    {
+      "epoch": 1.068573832562053,
+      "grad_norm": 0.32835808396339417,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 1270
+    },
+    {
+      "epoch": 1.076987799747581,
+      "grad_norm": 0.3876388370990753,
+      "learning_rate": 0.0002,
+      "loss": 1.7193,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0854017669331089,
+      "grad_norm": 0.39895930886268616,
+      "learning_rate": 0.0002,
+      "loss": 1.7442,
+      "step": 1290
+    },
+    {
+      "epoch": 1.093815734118637,
+      "grad_norm": 0.39081698656082153,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 1300
+    },
+    {
+      "epoch": 1.1022297013041649,
+      "grad_norm": 0.39974215626716614,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1310
+    },
+    {
+      "epoch": 1.110643668489693,
+      "grad_norm": 0.3887332081794739,
+      "learning_rate": 0.0002,
+      "loss": 1.7506,
+      "step": 1320
+    },
+    {
+      "epoch": 1.1190576356752209,
+      "grad_norm": 0.36216408014297485,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 1330
+    },
+    {
+      "epoch": 1.1274716028607488,
+      "grad_norm": 0.36979028582572937,
+      "learning_rate": 0.0002,
+      "loss": 1.762,
+      "step": 1340
+    },
+    {
+      "epoch": 1.1358855700462769,
+      "grad_norm": 0.34052133560180664,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 1350
+    },
+    {
+      "epoch": 1.1442995372318048,
+      "grad_norm": 0.3467716574668884,
+      "learning_rate": 0.0002,
+      "loss": 1.7513,
+      "step": 1360
+    },
+    {
+      "epoch": 1.1527135044173327,
+      "grad_norm": 0.35528799891471863,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1370
+    },
+    {
+      "epoch": 1.1611274716028608,
+      "grad_norm": 0.36282262206077576,
+      "learning_rate": 0.0002,
+      "loss": 1.794,
+      "step": 1380
+    },
+    {
+      "epoch": 1.1695414387883887,
+      "grad_norm": 0.37355899810791016,
+      "learning_rate": 0.0002,
+      "loss": 1.7731,
+      "step": 1390
+    },
+    {
+      "epoch": 1.1779554059739168,
+      "grad_norm": 0.37292736768722534,
+      "learning_rate": 0.0002,
+      "loss": 1.7483,
+      "step": 1400
+    },
+    {
+      "epoch": 1.1863693731594447,
+      "grad_norm": 0.5892812013626099,
+      "learning_rate": 0.0002,
+      "loss": 1.6916,
+      "step": 1410
+    },
+    {
+      "epoch": 1.1947833403449726,
+      "grad_norm": 0.3712292015552521,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 1420
+    },
+    {
+      "epoch": 1.2031973075305007,
+      "grad_norm": 0.3349577486515045,
+      "learning_rate": 0.0002,
+      "loss": 1.7709,
+      "step": 1430
+    },
+    {
+      "epoch": 1.2116112747160286,
+      "grad_norm": 0.32591062784194946,
+      "learning_rate": 0.0002,
+      "loss": 1.7412,
+      "step": 1440
+    },
+    {
+      "epoch": 1.2200252419015567,
+      "grad_norm": 0.3840635418891907,
+      "learning_rate": 0.0002,
+      "loss": 1.7406,
+      "step": 1450
+    },
+    {
+      "epoch": 1.2284392090870846,
+      "grad_norm": 0.37238365411758423,
+      "learning_rate": 0.0002,
+      "loss": 1.7276,
+      "step": 1460
+    },
+    {
+      "epoch": 1.2368531762726125,
+      "grad_norm": 0.3731217682361603,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 1470
+    },
+    {
+      "epoch": 1.2452671434581406,
+      "grad_norm": 0.3318967819213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7255,
+      "step": 1480
+    },
+    {
+      "epoch": 1.2536811106436685,
+      "grad_norm": 0.3784034848213196,
+      "learning_rate": 0.0002,
+      "loss": 1.7463,
+      "step": 1490
+    },
+    {
+      "epoch": 1.2620950778291964,
+      "grad_norm": 0.3541383147239685,
+      "learning_rate": 0.0002,
+      "loss": 1.6862,
+      "step": 1500
+    },
+    {
+      "epoch": 1.2705090450147245,
+      "grad_norm": 0.35312485694885254,
+      "learning_rate": 0.0002,
+      "loss": 1.8394,
+      "step": 1510
+    },
+    {
+      "epoch": 1.2789230122002524,
+      "grad_norm": 0.35272929072380066,
+      "learning_rate": 0.0002,
+      "loss": 1.7029,
+      "step": 1520
+    },
+    {
+      "epoch": 1.2873369793857803,
+      "grad_norm": 0.40988272428512573,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 1530
+    },
+    {
+      "epoch": 1.2957509465713084,
+      "grad_norm": 0.3543946146965027,
+      "learning_rate": 0.0002,
+      "loss": 1.6912,
+      "step": 1540
+    },
+    {
+      "epoch": 1.3041649137568363,
+      "grad_norm": 0.35639145970344543,
+      "learning_rate": 0.0002,
+      "loss": 1.6757,
+      "step": 1550
+    },
+    {
+      "epoch": 1.3125788809423642,
+      "grad_norm": 0.3290826678276062,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 1560
+    },
+    {
+      "epoch": 1.3209928481278923,
+      "grad_norm": 0.39264336228370667,
+      "learning_rate": 0.0002,
+      "loss": 1.7369,
+      "step": 1570
+    },
+    {
+      "epoch": 1.3294068153134202,
+      "grad_norm": 0.5390415191650391,
+      "learning_rate": 0.0002,
+      "loss": 1.6804,
+      "step": 1580
+    },
+    {
+      "epoch": 1.3378207824989483,
+      "grad_norm": 0.5188116431236267,
+      "learning_rate": 0.0002,
+      "loss": 1.708,
+      "step": 1590
+    },
+    {
+      "epoch": 1.3462347496844762,
+      "grad_norm": 0.37445148825645447,
+      "learning_rate": 0.0002,
+      "loss": 1.6763,
+      "step": 1600
+    },
+    {
+      "epoch": 1.3546487168700043,
+      "grad_norm": 0.3296085298061371,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1610
+    },
+    {
+      "epoch": 1.3630626840555322,
+      "grad_norm": 0.39879581332206726,
+      "learning_rate": 0.0002,
+      "loss": 1.8107,
+      "step": 1620
+    },
+    {
+      "epoch": 1.37147665124106,
+      "grad_norm": 0.36092764139175415,
+      "learning_rate": 0.0002,
+      "loss": 1.6744,
+      "step": 1630
+    },
+    {
+      "epoch": 1.3798906184265882,
+      "grad_norm": 0.37011823058128357,
+      "learning_rate": 0.0002,
+      "loss": 1.7144,
+      "step": 1640
+    },
+    {
+      "epoch": 1.3883045856121161,
+      "grad_norm": 0.40863534808158875,
+      "learning_rate": 0.0002,
+      "loss": 1.7396,
+      "step": 1650
+    },
+    {
+      "epoch": 1.396718552797644,
+      "grad_norm": 0.337001770734787,
+      "learning_rate": 0.0002,
+      "loss": 1.7901,
+      "step": 1660
+    },
+    {
+      "epoch": 1.4051325199831721,
+      "grad_norm": 0.35596707463264465,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1670
+    },
+    {
+      "epoch": 1.4135464871687,
+      "grad_norm": 0.3857671916484833,
+      "learning_rate": 0.0002,
+      "loss": 1.7717,
+      "step": 1680
+    },
+    {
+      "epoch": 1.421960454354228,
+      "grad_norm": 0.419502317905426,
+      "learning_rate": 0.0002,
+      "loss": 1.7015,
+      "step": 1690
+    },
+    {
+      "epoch": 1.430374421539756,
+      "grad_norm": 0.35459452867507935,
+      "learning_rate": 0.0002,
+      "loss": 1.7261,
+      "step": 1700
+    },
+    {
+      "epoch": 1.438788388725284,
+      "grad_norm": 0.37246978282928467,
+      "learning_rate": 0.0002,
+      "loss": 1.7361,
+      "step": 1710
+    },
+    {
+      "epoch": 1.4472023559108118,
+      "grad_norm": 0.33091893792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.6762,
+      "step": 1720
+    },
+    {
+      "epoch": 1.45561632309634,
+      "grad_norm": 0.37029674649238586,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 1730
+    },
+    {
+      "epoch": 1.4640302902818678,
+      "grad_norm": 0.374025821685791,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1740
+    },
+    {
+      "epoch": 1.472444257467396,
+      "grad_norm": 0.3416315019130707,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1750
+    },
+    {
+      "epoch": 1.4808582246529238,
+      "grad_norm": 0.36502841114997864,
+      "learning_rate": 0.0002,
+      "loss": 1.7093,
+      "step": 1760
+    },
+    {
+      "epoch": 1.489272191838452,
+      "grad_norm": 0.35458803176879883,
+      "learning_rate": 0.0002,
+      "loss": 1.6597,
+      "step": 1770
+    },
+    {
+      "epoch": 1.4976861590239798,
+      "grad_norm": 0.4462839663028717,
+      "learning_rate": 0.0002,
+      "loss": 1.675,
+      "step": 1780
+    },
+    {
+      "epoch": 1.5061001262095077,
+      "grad_norm": 0.34836092591285706,
+      "learning_rate": 0.0002,
+      "loss": 1.7267,
+      "step": 1790
+    },
+    {
+      "epoch": 1.5145140933950358,
+      "grad_norm": 0.3445749282836914,
+      "learning_rate": 0.0002,
+      "loss": 1.7295,
+      "step": 1800
+    },
+    {
+      "epoch": 1.5229280605805637,
+      "grad_norm": 0.36012160778045654,
+      "learning_rate": 0.0002,
+      "loss": 1.7386,
+      "step": 1810
+    },
+    {
+      "epoch": 1.5313420277660916,
+      "grad_norm": 0.4052616059780121,
+      "learning_rate": 0.0002,
+      "loss": 1.6594,
+      "step": 1820
+    },
+    {
+      "epoch": 1.5397559949516197,
+      "grad_norm": 0.3966905474662781,
+      "learning_rate": 0.0002,
+      "loss": 1.72,
+      "step": 1830
+    },
+    {
+      "epoch": 1.5481699621371476,
+      "grad_norm": 0.35028719902038574,
+      "learning_rate": 0.0002,
+      "loss": 1.7595,
+      "step": 1840
+    },
+    {
+      "epoch": 1.5565839293226755,
+      "grad_norm": 0.3936742842197418,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 1850
+    },
+    {
+      "epoch": 1.5649978965082036,
+      "grad_norm": 0.34473296999931335,
+      "learning_rate": 0.0002,
+      "loss": 1.7579,
+      "step": 1860
+    },
+    {
+      "epoch": 1.5734118636937318,
+      "grad_norm": 0.4328365623950958,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 1870
+    },
+    {
+      "epoch": 1.5818258308792594,
+      "grad_norm": 0.3566315472126007,
+      "learning_rate": 0.0002,
+      "loss": 1.7098,
+      "step": 1880
+    },
+    {
+      "epoch": 1.5902397980647875,
+      "grad_norm": 0.3301256597042084,
+      "learning_rate": 0.0002,
+      "loss": 1.6095,
+      "step": 1890
+    },
+    {
+      "epoch": 1.5986537652503157,
+      "grad_norm": 0.3743041455745697,
+      "learning_rate": 0.0002,
+      "loss": 1.748,
+      "step": 1900
+    },
+    {
+      "epoch": 1.6070677324358436,
+      "grad_norm": 0.3735344707965851,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 1910
+    },
+    {
+      "epoch": 1.6154816996213714,
+      "grad_norm": 0.42191144824028015,
+      "learning_rate": 0.0002,
+      "loss": 1.7445,
+      "step": 1920
+    },
+    {
+      "epoch": 1.6238956668068996,
+      "grad_norm": 0.3787207305431366,
+      "learning_rate": 0.0002,
+      "loss": 1.6978,
+      "step": 1930
+    },
+    {
+      "epoch": 1.6323096339924275,
+      "grad_norm": 0.35647350549697876,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 1940
+    },
+    {
+      "epoch": 1.6407236011779553,
+      "grad_norm": 0.39791446924209595,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 1950
+    },
+    {
+      "epoch": 1.6491375683634835,
+      "grad_norm": 0.37341275811195374,
+      "learning_rate": 0.0002,
+      "loss": 1.7293,
+      "step": 1960
+    },
+    {
+      "epoch": 1.6575515355490114,
+      "grad_norm": 0.3722686469554901,
+      "learning_rate": 0.0002,
+      "loss": 1.6781,
+      "step": 1970
+    },
+    {
+      "epoch": 1.6659655027345392,
+      "grad_norm": 0.37467387318611145,
+      "learning_rate": 0.0002,
+      "loss": 1.6383,
+      "step": 1980
+    },
+    {
+      "epoch": 1.6743794699200674,
+      "grad_norm": 0.37109461426734924,
+      "learning_rate": 0.0002,
+      "loss": 1.7439,
+      "step": 1990
+    },
+    {
+      "epoch": 1.6827934371055953,
+      "grad_norm": 0.4008837044239044,
+      "learning_rate": 0.0002,
+      "loss": 1.7206,
+      "step": 2000
+    },
+    {
+      "epoch": 1.6912074042911232,
+      "grad_norm": 0.3316999673843384,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2010
+    },
+    {
+      "epoch": 1.6996213714766513,
+      "grad_norm": 0.3683805465698242,
+      "learning_rate": 0.0002,
+      "loss": 1.7325,
+      "step": 2020
+    },
+    {
+      "epoch": 1.7080353386621794,
+      "grad_norm": 0.4163658320903778,
+      "learning_rate": 0.0002,
+      "loss": 1.7451,
+      "step": 2030
+    },
+    {
+      "epoch": 1.716449305847707,
+      "grad_norm": 0.4245431125164032,
+      "learning_rate": 0.0002,
+      "loss": 1.741,
+      "step": 2040
+    },
+    {
+      "epoch": 1.7248632730332352,
+      "grad_norm": 0.36732038855552673,
+      "learning_rate": 0.0002,
+      "loss": 1.7184,
+      "step": 2050
+    },
+    {
+      "epoch": 1.7332772402187633,
+      "grad_norm": 0.34981656074523926,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 2060
+    },
+    {
+      "epoch": 1.7416912074042912,
+      "grad_norm": 0.38588812947273254,
+      "learning_rate": 0.0002,
+      "loss": 1.7545,
+      "step": 2070
+    },
+    {
+      "epoch": 1.750105174589819,
+      "grad_norm": 0.39914557337760925,
+      "learning_rate": 0.0002,
+      "loss": 1.7728,
+      "step": 2080
+    },
+    {
+      "epoch": 1.7585191417753472,
+      "grad_norm": 0.36068692803382874,
+      "learning_rate": 0.0002,
+      "loss": 1.7049,
+      "step": 2090
+    },
+    {
+      "epoch": 1.766933108960875,
+      "grad_norm": 0.3983287215232849,
+      "learning_rate": 0.0002,
+      "loss": 1.7537,
+      "step": 2100
+    },
+    {
+      "epoch": 1.775347076146403,
+      "grad_norm": 0.45008400082588196,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2110
+    },
+    {
+      "epoch": 1.783761043331931,
+      "grad_norm": 0.3618052303791046,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 2120
+    },
+    {
+      "epoch": 1.792175010517459,
+      "grad_norm": 0.38745400309562683,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 2130
+    },
+    {
+      "epoch": 1.8005889777029869,
+      "grad_norm": 0.3413826525211334,
+      "learning_rate": 0.0002,
+      "loss": 1.7387,
+      "step": 2140
+    },
+    {
+      "epoch": 1.809002944888515,
+      "grad_norm": 0.35983747243881226,
+      "learning_rate": 0.0002,
+      "loss": 1.7414,
+      "step": 2150
+    },
+    {
+      "epoch": 1.8174169120740429,
+      "grad_norm": 0.40926849842071533,
+      "learning_rate": 0.0002,
+      "loss": 1.7892,
+      "step": 2160
+    },
+    {
+      "epoch": 1.8258308792595708,
+      "grad_norm": 0.3543093800544739,
+      "learning_rate": 0.0002,
+      "loss": 1.6823,
+      "step": 2170
+    },
+    {
+      "epoch": 1.8342448464450989,
+      "grad_norm": 0.42690935730934143,
+      "learning_rate": 0.0002,
+      "loss": 1.7812,
+      "step": 2180
+    },
+    {
+      "epoch": 1.842658813630627,
+      "grad_norm": 0.40282756090164185,
+      "learning_rate": 0.0002,
+      "loss": 1.7471,
+      "step": 2190
+    },
+    {
+      "epoch": 1.8510727808161547,
+      "grad_norm": 0.36568400263786316,
+      "learning_rate": 0.0002,
+      "loss": 1.7411,
+      "step": 2200
+    },
+    {
+      "epoch": 1.8594867480016828,
+      "grad_norm": 0.43159013986587524,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 2210
+    },
+    {
+      "epoch": 1.867900715187211,
+      "grad_norm": 0.3554118573665619,
+      "learning_rate": 0.0002,
+      "loss": 1.7298,
+      "step": 2220
+    },
+    {
+      "epoch": 1.8763146823727388,
+      "grad_norm": 0.43349072337150574,
+      "learning_rate": 0.0002,
+      "loss": 1.7157,
+      "step": 2230
+    },
+    {
+      "epoch": 1.8847286495582667,
+      "grad_norm": 0.36486536264419556,
+      "learning_rate": 0.0002,
+      "loss": 1.7302,
+      "step": 2240
+    },
+    {
+      "epoch": 1.8931426167437948,
+      "grad_norm": 0.39260047674179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6901,
+      "step": 2250
+    },
+    {
+      "epoch": 1.9015565839293227,
+      "grad_norm": 0.3741776943206787,
+      "learning_rate": 0.0002,
+      "loss": 1.6691,
+      "step": 2260
+    },
+    {
+      "epoch": 1.9099705511148506,
+      "grad_norm": 0.3961946964263916,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2270
+    },
+    {
+      "epoch": 1.9183845183003787,
+      "grad_norm": 0.3659731149673462,
+      "learning_rate": 0.0002,
+      "loss": 1.737,
+      "step": 2280
+    },
+    {
+      "epoch": 1.9267984854859066,
+      "grad_norm": 0.34744107723236084,
+      "learning_rate": 0.0002,
+      "loss": 1.7342,
+      "step": 2290
+    },
+    {
+      "epoch": 1.9352124526714345,
+      "grad_norm": 0.3607442378997803,
+      "learning_rate": 0.0002,
+      "loss": 1.7162,
+      "step": 2300
+    },
+    {
+      "epoch": 1.9436264198569626,
+      "grad_norm": 0.331464558839798,
+      "learning_rate": 0.0002,
+      "loss": 1.6673,
+      "step": 2310
+    },
+    {
+      "epoch": 1.9520403870424905,
+      "grad_norm": 0.3904414474964142,
+      "learning_rate": 0.0002,
+      "loss": 1.7101,
+      "step": 2320
+    },
+    {
+      "epoch": 1.9604543542280184,
+      "grad_norm": 0.37584832310676575,
+      "learning_rate": 0.0002,
+      "loss": 1.7327,
+      "step": 2330
+    },
+    {
+      "epoch": 1.9688683214135465,
+      "grad_norm": 0.3698684275150299,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 2340
+    },
+    {
+      "epoch": 1.9772822885990746,
+      "grad_norm": 0.40571412444114685,
+      "learning_rate": 0.0002,
+      "loss": 1.7764,
+      "step": 2350
+    },
+    {
+      "epoch": 1.9856962557846023,
+      "grad_norm": 0.40059587359428406,
+      "learning_rate": 0.0002,
+      "loss": 1.744,
+      "step": 2360
+    },
+    {
+      "epoch": 1.9941102229701304,
+      "grad_norm": 0.4168248474597931,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2370
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8055059909820557,
+      "eval_runtime": 38.422,
+      "eval_samples_per_second": 13.404,
+      "eval_steps_per_second": 1.692,
+      "step": 2377
+    },
+    {
+      "epoch": 2.0025241901556585,
+      "grad_norm": 0.35205352306365967,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 2380
+    },
+    {
+      "epoch": 2.010938157341186,
+      "grad_norm": 0.3979377746582031,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 2390
+    },
+    {
+      "epoch": 2.0193521245267143,
+      "grad_norm": 0.396491676568985,
+      "learning_rate": 0.0002,
+      "loss": 1.6421,
+      "step": 2400
+    },
+    {
+      "epoch": 2.0277660917122424,
+      "grad_norm": 0.44712209701538086,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2410
+    },
+    {
+      "epoch": 2.03618005889777,
+      "grad_norm": 0.4454420208930969,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 2420
+    },
+    {
+      "epoch": 2.044594026083298,
+      "grad_norm": 0.4170038402080536,
+      "learning_rate": 0.0002,
+      "loss": 1.6635,
+      "step": 2430
+    },
+    {
+      "epoch": 2.0530079932688263,
+      "grad_norm": 0.4309595227241516,
+      "learning_rate": 0.0002,
+      "loss": 1.6512,
+      "step": 2440
+    },
+    {
+      "epoch": 2.0614219604543544,
+      "grad_norm": 0.4241602122783661,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 2450
+    },
+    {
+      "epoch": 2.069835927639882,
+      "grad_norm": 0.4370540678501129,
+      "learning_rate": 0.0002,
+      "loss": 1.6162,
+      "step": 2460
+    },
+    {
+      "epoch": 2.0782498948254102,
+      "grad_norm": 0.43985554575920105,
+      "learning_rate": 0.0002,
+      "loss": 1.6354,
+      "step": 2470
+    },
+    {
+      "epoch": 2.0866638620109383,
+      "grad_norm": 0.4158105254173279,
+      "learning_rate": 0.0002,
+      "loss": 1.6954,
+      "step": 2480
+    },
+    {
+      "epoch": 2.095077829196466,
+      "grad_norm": 0.441549152135849,
+      "learning_rate": 0.0002,
+      "loss": 1.6114,
+      "step": 2490
+    },
+    {
+      "epoch": 2.103491796381994,
+      "grad_norm": 0.385718435049057,
+      "learning_rate": 0.0002,
+      "loss": 1.5485,
+      "step": 2500
+    },
+    {
+      "epoch": 2.1119057635675222,
+      "grad_norm": 0.43146514892578125,
+      "learning_rate": 0.0002,
+      "loss": 1.5894,
+      "step": 2510
+    },
+    {
+      "epoch": 2.12031973075305,
+      "grad_norm": 0.41663315892219543,
+      "learning_rate": 0.0002,
+      "loss": 1.6414,
+      "step": 2520
+    },
+    {
+      "epoch": 2.128733697938578,
+      "grad_norm": 0.4410698115825653,
+      "learning_rate": 0.0002,
+      "loss": 1.6527,
+      "step": 2530
+    },
+    {
+      "epoch": 2.137147665124106,
+      "grad_norm": 0.4472278952598572,
+      "learning_rate": 0.0002,
+      "loss": 1.6124,
+      "step": 2540
+    },
+    {
+      "epoch": 2.145561632309634,
+      "grad_norm": 0.3879167437553406,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 2550
+    },
+    {
+      "epoch": 2.153975599495162,
+      "grad_norm": 0.4212203025817871,
+      "learning_rate": 0.0002,
+      "loss": 1.6682,
+      "step": 2560
+    },
+    {
+      "epoch": 2.16238956668069,
+      "grad_norm": 0.42841723561286926,
+      "learning_rate": 0.0002,
+      "loss": 1.6036,
+      "step": 2570
+    },
+    {
+      "epoch": 2.1708035338662177,
+      "grad_norm": 0.39272481203079224,
+      "learning_rate": 0.0002,
+      "loss": 1.5962,
+      "step": 2580
+    },
+    {
+      "epoch": 2.179217501051746,
+      "grad_norm": 0.4075261354446411,
+      "learning_rate": 0.0002,
+      "loss": 1.681,
+      "step": 2590
+    },
+    {
+      "epoch": 2.187631468237274,
+      "grad_norm": 0.5358437895774841,
+      "learning_rate": 0.0002,
+      "loss": 1.6601,
+      "step": 2600
+    },
+    {
+      "epoch": 2.1960454354228016,
+      "grad_norm": 0.4738350212574005,
+      "learning_rate": 0.0002,
+      "loss": 1.6423,
+      "step": 2610
+    },
+    {
+      "epoch": 2.2044594026083297,
+      "grad_norm": 0.446789026260376,
+      "learning_rate": 0.0002,
+      "loss": 1.6386,
+      "step": 2620
+    },
+    {
+      "epoch": 2.212873369793858,
+      "grad_norm": 0.4615374505519867,
+      "learning_rate": 0.0002,
+      "loss": 1.6246,
+      "step": 2630
+    },
+    {
+      "epoch": 2.221287336979386,
+      "grad_norm": 0.46901994943618774,
+      "learning_rate": 0.0002,
+      "loss": 1.6205,
+      "step": 2640
+    },
+    {
+      "epoch": 2.2297013041649136,
+      "grad_norm": 0.46267789602279663,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 2.2381152713504417,
+      "grad_norm": 0.4383080005645752,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2660
+    },
+    {
+      "epoch": 2.24652923853597,
+      "grad_norm": 0.4070609509944916,
+      "learning_rate": 0.0002,
+      "loss": 1.5745,
+      "step": 2670
+    },
+    {
+      "epoch": 2.2549432057214975,
+      "grad_norm": 0.4572339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.6125,
+      "step": 2680
+    },
+    {
+      "epoch": 2.2633571729070256,
+      "grad_norm": 0.393265038728714,
+      "learning_rate": 0.0002,
+      "loss": 1.5671,
+      "step": 2690
+    },
+    {
+      "epoch": 2.2717711400925538,
+      "grad_norm": 0.46144717931747437,
+      "learning_rate": 0.0002,
+      "loss": 1.6239,
+      "step": 2700
+    },
+    {
+      "epoch": 2.2801851072780814,
+      "grad_norm": 0.45077767968177795,
+      "learning_rate": 0.0002,
+      "loss": 1.5992,
+      "step": 2710
+    },
+    {
+      "epoch": 2.2885990744636096,
+      "grad_norm": 0.5697639584541321,
+      "learning_rate": 0.0002,
+      "loss": 1.6261,
+      "step": 2720
+    },
+    {
+      "epoch": 2.2970130416491377,
+      "grad_norm": 0.4855510890483856,
+      "learning_rate": 0.0002,
+      "loss": 1.6192,
+      "step": 2730
+    },
+    {
+      "epoch": 2.3054270088346653,
+      "grad_norm": 0.4440622627735138,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 2740
+    },
+    {
+      "epoch": 2.3138409760201935,
+      "grad_norm": 0.3904096782207489,
+      "learning_rate": 0.0002,
+      "loss": 1.6496,
+      "step": 2750
+    },
+    {
+      "epoch": 2.3222549432057216,
+      "grad_norm": 0.5225510597229004,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 2760
+    },
+    {
+      "epoch": 2.3306689103912497,
+      "grad_norm": 0.44866397976875305,
+      "learning_rate": 0.0002,
+      "loss": 1.6082,
+      "step": 2770
+    },
+    {
+      "epoch": 2.3390828775767774,
+      "grad_norm": 0.5167056322097778,
+      "learning_rate": 0.0002,
+      "loss": 1.6087,
+      "step": 2780
+    },
+    {
+      "epoch": 2.3474968447623055,
+      "grad_norm": 0.45913267135620117,
+      "learning_rate": 0.0002,
+      "loss": 1.6136,
+      "step": 2790
+    },
+    {
+      "epoch": 2.3559108119478336,
+      "grad_norm": 0.45787590742111206,
+      "learning_rate": 0.0002,
+      "loss": 1.6564,
+      "step": 2800
+    },
+    {
+      "epoch": 2.3643247791333613,
+      "grad_norm": 0.4633352756500244,
+      "learning_rate": 0.0002,
+      "loss": 1.6868,
+      "step": 2810
+    },
+    {
+      "epoch": 2.3727387463188894,
+      "grad_norm": 0.46390071511268616,
+      "learning_rate": 0.0002,
+      "loss": 1.6316,
+      "step": 2820
+    },
+    {
+      "epoch": 2.3811527135044175,
+      "grad_norm": 0.4261005222797394,
+      "learning_rate": 0.0002,
+      "loss": 1.6039,
+      "step": 2830
+    },
+    {
+      "epoch": 2.389566680689945,
+      "grad_norm": 0.4283634424209595,
+      "learning_rate": 0.0002,
+      "loss": 1.6364,
+      "step": 2840
+    },
+    {
+      "epoch": 2.3979806478754733,
+      "grad_norm": 0.4955291450023651,
+      "learning_rate": 0.0002,
+      "loss": 1.6382,
+      "step": 2850
+    },
+    {
+      "epoch": 2.4063946150610014,
+      "grad_norm": 0.4740189015865326,
+      "learning_rate": 0.0002,
+      "loss": 1.6173,
+      "step": 2860
+    },
+    {
+      "epoch": 2.414808582246529,
+      "grad_norm": 0.4222276508808136,
+      "learning_rate": 0.0002,
+      "loss": 1.6403,
+      "step": 2870
+    },
+    {
+      "epoch": 2.423222549432057,
+      "grad_norm": 0.4982149004936218,
+      "learning_rate": 0.0002,
+      "loss": 1.5602,
+      "step": 2880
+    },
+    {
+      "epoch": 2.4316365166175853,
+      "grad_norm": 0.5217409133911133,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 2890
+    },
+    {
+      "epoch": 2.4400504838031134,
+      "grad_norm": 0.4555884897708893,
+      "learning_rate": 0.0002,
+      "loss": 1.5804,
+      "step": 2900
+    },
+    {
+      "epoch": 2.448464450988641,
+      "grad_norm": 0.43178579211235046,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2910
+    },
+    {
+      "epoch": 2.456878418174169,
+      "grad_norm": 0.4788478910923004,
+      "learning_rate": 0.0002,
+      "loss": 1.6824,
+      "step": 2920
+    },
+    {
+      "epoch": 2.465292385359697,
+      "grad_norm": 0.43689873814582825,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 2930
+    },
+    {
+      "epoch": 2.473706352545225,
+      "grad_norm": 0.5115197896957397,
+      "learning_rate": 0.0002,
+      "loss": 1.6196,
+      "step": 2940
+    },
+    {
+      "epoch": 2.482120319730753,
+      "grad_norm": 0.5290159583091736,
+      "learning_rate": 0.0002,
+      "loss": 1.689,
+      "step": 2950
+    },
+    {
+      "epoch": 2.490534286916281,
+      "grad_norm": 0.46042463183403015,
+      "learning_rate": 0.0002,
+      "loss": 1.6499,
+      "step": 2960
+    },
+    {
+      "epoch": 2.498948254101809,
+      "grad_norm": 0.4359915852546692,
+      "learning_rate": 0.0002,
+      "loss": 1.6664,
+      "step": 2970
+    },
+    {
+      "epoch": 2.507362221287337,
+      "grad_norm": 0.46352964639663696,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 2980
+    },
+    {
+      "epoch": 2.515776188472865,
+      "grad_norm": 0.5324268341064453,
+      "learning_rate": 0.0002,
+      "loss": 1.6501,
+      "step": 2990
+    },
+    {
+      "epoch": 2.5241901556583928,
+      "grad_norm": 0.5929607152938843,
+      "learning_rate": 0.0002,
+      "loss": 1.6115,
+      "step": 3000
+    },
+    {
+      "epoch": 2.532604122843921,
+      "grad_norm": 0.4811333417892456,
+      "learning_rate": 0.0002,
+      "loss": 1.6772,
+      "step": 3010
+    },
+    {
+      "epoch": 2.541018090029449,
+      "grad_norm": 0.4662701487541199,
+      "learning_rate": 0.0002,
+      "loss": 1.7023,
+      "step": 3020
+    },
+    {
+      "epoch": 2.549432057214977,
+      "grad_norm": 0.4582270681858063,
+      "learning_rate": 0.0002,
+      "loss": 1.5426,
+      "step": 3030
+    },
+    {
+      "epoch": 2.557846024400505,
+      "grad_norm": 0.4679982662200928,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 3040
+    },
+    {
+      "epoch": 2.566259991586033,
+      "grad_norm": 0.4380294680595398,
+      "learning_rate": 0.0002,
+      "loss": 1.5442,
+      "step": 3050
+    },
+    {
+      "epoch": 2.5746739587715606,
+      "grad_norm": 0.44295763969421387,
+      "learning_rate": 0.0002,
+      "loss": 1.6055,
+      "step": 3060
+    },
+    {
+      "epoch": 2.5830879259570887,
+      "grad_norm": 0.5131027698516846,
+      "learning_rate": 0.0002,
+      "loss": 1.5775,
+      "step": 3070
+    },
+    {
+      "epoch": 2.591501893142617,
+      "grad_norm": 0.47567516565322876,
+      "learning_rate": 0.0002,
+      "loss": 1.546,
+      "step": 3080
+    },
+    {
+      "epoch": 2.599915860328145,
+      "grad_norm": 0.49002596735954285,
+      "learning_rate": 0.0002,
+      "loss": 1.5671,
+      "step": 3090
+    },
+    {
+      "epoch": 2.6083298275136726,
+      "grad_norm": 0.44856327772140503,
+      "learning_rate": 0.0002,
+      "loss": 1.5445,
+      "step": 3100
+    },
+    {
+      "epoch": 2.6167437946992007,
+      "grad_norm": 0.4480142593383789,
+      "learning_rate": 0.0002,
+      "loss": 1.5797,
+      "step": 3110
+    },
+    {
+      "epoch": 2.6251577618847284,
+      "grad_norm": 0.4317494034767151,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 3120
+    },
+    {
+      "epoch": 2.6335717290702565,
+      "grad_norm": 0.42580848932266235,
+      "learning_rate": 0.0002,
+      "loss": 1.6321,
+      "step": 3130
+    },
+    {
+      "epoch": 2.6419856962557846,
+      "grad_norm": 0.4516814947128296,
+      "learning_rate": 0.0002,
+      "loss": 1.6483,
+      "step": 3140
+    },
+    {
+      "epoch": 2.6503996634413127,
+      "grad_norm": 0.4438435733318329,
+      "learning_rate": 0.0002,
+      "loss": 1.695,
+      "step": 3150
+    },
+    {
+      "epoch": 2.6588136306268404,
+      "grad_norm": 0.4385356307029724,
+      "learning_rate": 0.0002,
+      "loss": 1.6938,
+      "step": 3160
+    },
+    {
+      "epoch": 2.6672275978123685,
+      "grad_norm": 0.5064112544059753,
+      "learning_rate": 0.0002,
+      "loss": 1.6139,
+      "step": 3170
+    },
+    {
+      "epoch": 2.6756415649978966,
+      "grad_norm": 0.49163177609443665,
+      "learning_rate": 0.0002,
+      "loss": 1.7189,
+      "step": 3180
+    },
+    {
+      "epoch": 2.6840555321834243,
+      "grad_norm": 0.49339258670806885,
+      "learning_rate": 0.0002,
+      "loss": 1.7323,
+      "step": 3190
+    },
+    {
+      "epoch": 2.6924694993689524,
+      "grad_norm": 0.440950870513916,
+      "learning_rate": 0.0002,
+      "loss": 1.6508,
+      "step": 3200
+    },
+    {
+      "epoch": 2.7008834665544805,
+      "grad_norm": 0.4283970594406128,
+      "learning_rate": 0.0002,
+      "loss": 1.6305,
+      "step": 3210
+    },
+    {
+      "epoch": 2.7092974337400086,
+      "grad_norm": 0.43875712156295776,
+      "learning_rate": 0.0002,
+      "loss": 1.5935,
+      "step": 3220
+    },
+    {
+      "epoch": 2.7177114009255363,
+      "grad_norm": 0.49332964420318604,
+      "learning_rate": 0.0002,
+      "loss": 1.6129,
+      "step": 3230
+    },
+    {
+      "epoch": 2.7261253681110644,
+      "grad_norm": 0.5225692391395569,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 3240
+    },
+    {
+      "epoch": 2.734539335296592,
+      "grad_norm": 0.4856489300727844,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 3250
+    },
+    {
+      "epoch": 2.74295330248212,
+      "grad_norm": 0.46918296813964844,
+      "learning_rate": 0.0002,
+      "loss": 1.6463,
+      "step": 3260
+    },
+    {
+      "epoch": 2.7513672696676483,
+      "grad_norm": 0.4802931249141693,
+      "learning_rate": 0.0002,
+      "loss": 1.6819,
+      "step": 3270
+    },
+    {
+      "epoch": 2.7597812368531764,
+      "grad_norm": 0.4485355615615845,
+      "learning_rate": 0.0002,
+      "loss": 1.6246,
+      "step": 3280
+    },
+    {
+      "epoch": 2.768195204038704,
+      "grad_norm": 0.43944594264030457,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 3290
+    },
+    {
+      "epoch": 2.7766091712242322,
+      "grad_norm": 0.46847742795944214,
+      "learning_rate": 0.0002,
+      "loss": 1.6501,
+      "step": 3300
+    },
+    {
+      "epoch": 2.7850231384097603,
+      "grad_norm": 0.4816027879714966,
+      "learning_rate": 0.0002,
+      "loss": 1.5969,
+      "step": 3310
+    },
+    {
+      "epoch": 2.793437105595288,
+      "grad_norm": 0.453960120677948,
+      "learning_rate": 0.0002,
+      "loss": 1.6293,
+      "step": 3320
+    },
+    {
+      "epoch": 2.801851072780816,
+      "grad_norm": 0.4816017150878906,
+      "learning_rate": 0.0002,
+      "loss": 1.6429,
+      "step": 3330
+    },
+    {
+      "epoch": 2.8102650399663442,
+      "grad_norm": 0.4461034834384918,
+      "learning_rate": 0.0002,
+      "loss": 1.6683,
+      "step": 3340
+    },
+    {
+      "epoch": 2.8186790071518724,
+      "grad_norm": 0.48821821808815,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 3350
+    },
+    {
+      "epoch": 2.8270929743374,
+      "grad_norm": 0.4574853777885437,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 3360
+    },
+    {
+      "epoch": 2.835506941522928,
+      "grad_norm": 0.42062026262283325,
+      "learning_rate": 0.0002,
+      "loss": 1.6651,
+      "step": 3370
+    },
+    {
+      "epoch": 2.843920908708456,
+      "grad_norm": 0.4499834477901459,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 3380
+    },
+    {
+      "epoch": 2.852334875893984,
+      "grad_norm": 0.4780360758304596,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 3390
+    },
+    {
+      "epoch": 2.860748843079512,
+      "grad_norm": 0.45422887802124023,
+      "learning_rate": 0.0002,
+      "loss": 1.5882,
+      "step": 3400
+    },
+    {
+      "epoch": 2.86916281026504,
+      "grad_norm": 0.4590015709400177,
+      "learning_rate": 0.0002,
+      "loss": 1.6028,
+      "step": 3410
+    },
+    {
+      "epoch": 2.877576777450568,
+      "grad_norm": 0.45689624547958374,
+      "learning_rate": 0.0002,
+      "loss": 1.6746,
+      "step": 3420
+    },
+    {
+      "epoch": 2.885990744636096,
+      "grad_norm": 0.46953922510147095,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3430
+    },
+    {
+      "epoch": 2.8944047118216236,
+      "grad_norm": 0.4791966378688812,
+      "learning_rate": 0.0002,
+      "loss": 1.6015,
+      "step": 3440
+    },
+    {
+      "epoch": 2.9028186790071517,
+      "grad_norm": 0.4842296242713928,
+      "learning_rate": 0.0002,
+      "loss": 1.694,
+      "step": 3450
+    },
+    {
+      "epoch": 2.91123264619268,
+      "grad_norm": 0.47219768166542053,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3460
+    },
+    {
+      "epoch": 2.919646613378208,
+      "grad_norm": 0.4622127115726471,
+      "learning_rate": 0.0002,
+      "loss": 1.6486,
+      "step": 3470
+    },
+    {
+      "epoch": 2.9280605805637356,
+      "grad_norm": 0.46832820773124695,
+      "learning_rate": 0.0002,
+      "loss": 1.6485,
+      "step": 3480
+    },
+    {
+      "epoch": 2.9364745477492638,
+      "grad_norm": 0.44582483172416687,
+      "learning_rate": 0.0002,
+      "loss": 1.6366,
+      "step": 3490
+    },
+    {
+      "epoch": 2.944888514934792,
+      "grad_norm": 0.4987219274044037,
+      "learning_rate": 0.0002,
+      "loss": 1.6859,
+      "step": 3500
+    },
+    {
+      "epoch": 2.9533024821203195,
+      "grad_norm": 0.43750956654548645,
+      "learning_rate": 0.0002,
+      "loss": 1.5991,
+      "step": 3510
+    },
+    {
+      "epoch": 2.9617164493058477,
+      "grad_norm": 0.49962925910949707,
+      "learning_rate": 0.0002,
+      "loss": 1.6236,
+      "step": 3520
+    },
+    {
+      "epoch": 2.9701304164913758,
+      "grad_norm": 0.5189590454101562,
+      "learning_rate": 0.0002,
+      "loss": 1.5859,
+      "step": 3530
+    },
+    {
+      "epoch": 2.978544383676904,
+      "grad_norm": 0.391317754983902,
+      "learning_rate": 0.0002,
+      "loss": 1.6688,
+      "step": 3540
+    },
+    {
+      "epoch": 2.9869583508624316,
+      "grad_norm": 0.44934695959091187,
+      "learning_rate": 0.0002,
+      "loss": 1.5884,
+      "step": 3550
+    },
+    {
+      "epoch": 2.9953723180479597,
+      "grad_norm": 0.4740142226219177,
+      "learning_rate": 0.0002,
+      "loss": 1.5688,
+      "step": 3560
+    },
+    {
+      "epoch": 2.9995793016407237,
+      "eval_loss": 1.8266887664794922,
+      "eval_runtime": 37.9445,
+      "eval_samples_per_second": 13.572,
+      "eval_steps_per_second": 1.713,
+      "step": 3565
+    },
+    {
+      "epoch": 3.003786285233488,
+      "grad_norm": 0.4523724615573883,
+      "learning_rate": 0.0002,
+      "loss": 1.5939,
+      "step": 3570
+    },
+    {
+      "epoch": 3.0122002524190155,
+      "grad_norm": 0.5261380076408386,
+      "learning_rate": 0.0002,
+      "loss": 1.526,
+      "step": 3580
+    },
+    {
+      "epoch": 3.0206142196045436,
+      "grad_norm": 0.48664888739585876,
+      "learning_rate": 0.0002,
+      "loss": 1.4946,
+      "step": 3590
+    },
+    {
+      "epoch": 3.0290281867900717,
+      "grad_norm": 0.5070882439613342,
+      "learning_rate": 0.0002,
+      "loss": 1.5193,
+      "step": 3600
+    },
+    {
+      "epoch": 3.0374421539755994,
+      "grad_norm": 0.5816011428833008,
+      "learning_rate": 0.0002,
+      "loss": 1.5316,
+      "step": 3610
+    },
+    {
+      "epoch": 3.0458561211611275,
+      "grad_norm": 0.6610211730003357,
+      "learning_rate": 0.0002,
+      "loss": 1.5682,
+      "step": 3620
+    },
+    {
+      "epoch": 3.0542700883466556,
+      "grad_norm": 0.5257703065872192,
+      "learning_rate": 0.0002,
+      "loss": 1.5699,
+      "step": 3630
+    },
+    {
+      "epoch": 3.0626840555321833,
+      "grad_norm": 0.5574390888214111,
+      "learning_rate": 0.0002,
+      "loss": 1.4438,
+      "step": 3640
+    },
+    {
+      "epoch": 3.0710980227177114,
+      "grad_norm": 0.5682297348976135,
+      "learning_rate": 0.0002,
+      "loss": 1.547,
+      "step": 3650
+    },
+    {
+      "epoch": 3.0795119899032395,
+      "grad_norm": 0.5798383355140686,
+      "learning_rate": 0.0002,
+      "loss": 1.5743,
+      "step": 3660
+    },
+    {
+      "epoch": 3.087925957088767,
+      "grad_norm": 0.5458289980888367,
+      "learning_rate": 0.0002,
+      "loss": 1.4339,
+      "step": 3670
+    },
+    {
+      "epoch": 3.0963399242742953,
+      "grad_norm": 0.5599102973937988,
+      "learning_rate": 0.0002,
+      "loss": 1.46,
+      "step": 3680
+    },
+    {
+      "epoch": 3.1047538914598234,
+      "grad_norm": 0.5023021697998047,
+      "learning_rate": 0.0002,
+      "loss": 1.4589,
+      "step": 3690
+    },
+    {
+      "epoch": 3.113167858645351,
+      "grad_norm": 0.5448206067085266,
+      "learning_rate": 0.0002,
+      "loss": 1.5114,
+      "step": 3700
+    },
+    {
+      "epoch": 3.121581825830879,
+      "grad_norm": 0.5760458707809448,
+      "learning_rate": 0.0002,
+      "loss": 1.4692,
+      "step": 3710
+    },
+    {
+      "epoch": 3.1299957930164073,
+      "grad_norm": 0.6018968224525452,
+      "learning_rate": 0.0002,
+      "loss": 1.4789,
+      "step": 3720
+    },
+    {
+      "epoch": 3.1384097602019354,
+      "grad_norm": 0.5767101049423218,
+      "learning_rate": 0.0002,
+      "loss": 1.5518,
+      "step": 3730
+    },
+    {
+      "epoch": 3.146823727387463,
+      "grad_norm": 0.5333963632583618,
+      "learning_rate": 0.0002,
+      "loss": 1.5032,
+      "step": 3740
+    },
+    {
+      "epoch": 3.155237694572991,
+      "grad_norm": 0.5918396711349487,
+      "learning_rate": 0.0002,
+      "loss": 1.4812,
+      "step": 3750
+    },
+    {
+      "epoch": 3.1636516617585193,
+      "grad_norm": 0.5931203365325928,
+      "learning_rate": 0.0002,
+      "loss": 1.4618,
+      "step": 3760
+    },
+    {
+      "epoch": 3.172065628944047,
+      "grad_norm": 0.6562168598175049,
+      "learning_rate": 0.0002,
+      "loss": 1.5592,
+      "step": 3770
+    },
+    {
+      "epoch": 3.180479596129575,
+      "grad_norm": 0.5820156335830688,
+      "learning_rate": 0.0002,
+      "loss": 1.4932,
+      "step": 3780
+    },
+    {
+      "epoch": 3.188893563315103,
+      "grad_norm": 0.5784737467765808,
+      "learning_rate": 0.0002,
+      "loss": 1.4523,
+      "step": 3790
+    },
+    {
+      "epoch": 3.197307530500631,
+      "grad_norm": 0.5506529808044434,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 3800
+    },
+    {
+      "epoch": 3.205721497686159,
+      "grad_norm": 0.6101595163345337,
+      "learning_rate": 0.0002,
+      "loss": 1.4819,
+      "step": 3810
+    },
+    {
+      "epoch": 3.214135464871687,
+      "grad_norm": 0.5597806572914124,
+      "learning_rate": 0.0002,
+      "loss": 1.5185,
+      "step": 3820
+    },
+    {
+      "epoch": 3.222549432057215,
+      "grad_norm": 0.5641011595726013,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 3830
+    },
+    {
+      "epoch": 3.230963399242743,
+      "grad_norm": 0.5892080068588257,
+      "learning_rate": 0.0002,
+      "loss": 1.4702,
+      "step": 3840
+    },
+    {
+      "epoch": 3.239377366428271,
+      "grad_norm": 0.6034760475158691,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 3850
+    },
+    {
+      "epoch": 3.247791333613799,
+      "grad_norm": 0.5112439393997192,
+      "learning_rate": 0.0002,
+      "loss": 1.5499,
+      "step": 3860
+    },
+    {
+      "epoch": 3.256205300799327,
+      "grad_norm": 0.56565922498703,
+      "learning_rate": 0.0002,
+      "loss": 1.5132,
+      "step": 3870
+    },
+    {
+      "epoch": 3.264619267984855,
+      "grad_norm": 0.6155247092247009,
+      "learning_rate": 0.0002,
+      "loss": 1.4892,
+      "step": 3880
+    },
+    {
+      "epoch": 3.273033235170383,
+      "grad_norm": 0.6064623594284058,
+      "learning_rate": 0.0002,
+      "loss": 1.5118,
+      "step": 3890
+    },
+    {
+      "epoch": 3.2814472023559107,
+      "grad_norm": 0.6313768029212952,
+      "learning_rate": 0.0002,
+      "loss": 1.5236,
+      "step": 3900
+    },
+    {
+      "epoch": 3.289861169541439,
+      "grad_norm": 0.5903939008712769,
+      "learning_rate": 0.0002,
+      "loss": 1.5551,
+      "step": 3910
+    },
+    {
+      "epoch": 3.298275136726967,
+      "grad_norm": 0.5770667195320129,
+      "learning_rate": 0.0002,
+      "loss": 1.5703,
+      "step": 3920
+    },
+    {
+      "epoch": 3.3066891039124946,
+      "grad_norm": 0.5785196423530579,
+      "learning_rate": 0.0002,
+      "loss": 1.5159,
+      "step": 3930
+    },
+    {
+      "epoch": 3.3151030710980227,
+      "grad_norm": 0.6468310356140137,
+      "learning_rate": 0.0002,
+      "loss": 1.5277,
+      "step": 3940
+    },
+    {
+      "epoch": 3.323517038283551,
+      "grad_norm": 0.6200279593467712,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 3950
+    },
+    {
+      "epoch": 3.3319310054690785,
+      "grad_norm": 0.5779302716255188,
+      "learning_rate": 0.0002,
+      "loss": 1.5264,
+      "step": 3960
+    },
+    {
+      "epoch": 3.3403449726546066,
+      "grad_norm": 0.5463796854019165,
+      "learning_rate": 0.0002,
+      "loss": 1.4861,
+      "step": 3970
+    },
+    {
+      "epoch": 3.3487589398401347,
+      "grad_norm": 0.6117855906486511,
+      "learning_rate": 0.0002,
+      "loss": 1.541,
+      "step": 3980
+    },
+    {
+      "epoch": 3.357172907025663,
+      "grad_norm": 0.5554766058921814,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 3990
+    },
+    {
+      "epoch": 3.3655868742111905,
+      "grad_norm": 0.6012870073318481,
+      "learning_rate": 0.0002,
+      "loss": 1.5004,
+      "step": 4000
+    },
+    {
+      "epoch": 3.3740008413967186,
+      "grad_norm": 0.5443974137306213,
+      "learning_rate": 0.0002,
+      "loss": 1.473,
+      "step": 4010
+    },
+    {
+      "epoch": 3.3824148085822463,
+      "grad_norm": 0.6636057496070862,
+      "learning_rate": 0.0002,
+      "loss": 1.5139,
+      "step": 4020
+    },
+    {
+      "epoch": 3.3908287757677744,
+      "grad_norm": 0.5801246166229248,
+      "learning_rate": 0.0002,
+      "loss": 1.5141,
+      "step": 4030
+    },
+    {
+      "epoch": 3.3992427429533025,
+      "grad_norm": 0.5668839812278748,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4040
+    },
+    {
+      "epoch": 3.4076567101388306,
+      "grad_norm": 0.7763481736183167,
+      "learning_rate": 0.0002,
+      "loss": 1.523,
+      "step": 4050
+    },
+    {
+      "epoch": 3.4160706773243583,
+      "grad_norm": 0.6675992608070374,
+      "learning_rate": 0.0002,
+      "loss": 1.4932,
+      "step": 4060
+    },
+    {
+      "epoch": 3.4244846445098864,
+      "grad_norm": 0.6290077567100525,
+      "learning_rate": 0.0002,
+      "loss": 1.4959,
+      "step": 4070
+    },
+    {
+      "epoch": 3.4328986116954145,
+      "grad_norm": 0.6040239930152893,
+      "learning_rate": 0.0002,
+      "loss": 1.5766,
+      "step": 4080
+    },
+    {
+      "epoch": 3.441312578880942,
+      "grad_norm": 0.6237877607345581,
+      "learning_rate": 0.0002,
+      "loss": 1.5711,
+      "step": 4090
+    },
+    {
+      "epoch": 3.4497265460664703,
+      "grad_norm": 0.5343508124351501,
+      "learning_rate": 0.0002,
+      "loss": 1.4961,
+      "step": 4100
+    },
+    {
+      "epoch": 3.4581405132519984,
+      "grad_norm": 0.6817412972450256,
+      "learning_rate": 0.0002,
+      "loss": 1.5123,
+      "step": 4110
+    },
+    {
+      "epoch": 3.466554480437526,
+      "grad_norm": 0.7115170359611511,
+      "learning_rate": 0.0002,
+      "loss": 1.5377,
+      "step": 4120
+    },
+    {
+      "epoch": 3.4749684476230542,
+      "grad_norm": 0.6127332448959351,
+      "learning_rate": 0.0002,
+      "loss": 1.5275,
+      "step": 4130
+    },
+    {
+      "epoch": 3.4833824148085824,
+      "grad_norm": 0.5745994448661804,
+      "learning_rate": 0.0002,
+      "loss": 1.557,
+      "step": 4140
+    },
+    {
+      "epoch": 3.49179638199411,
+      "grad_norm": 0.6248795390129089,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4150
+    },
+    {
+      "epoch": 3.500210349179638,
+      "grad_norm": 0.5821124911308289,
+      "learning_rate": 0.0002,
+      "loss": 1.4885,
+      "step": 4160
+    },
+    {
+      "epoch": 3.5086243163651663,
+      "grad_norm": 0.561416506767273,
+      "learning_rate": 0.0002,
+      "loss": 1.4937,
+      "step": 4170
+    },
+    {
+      "epoch": 3.5170382835506944,
+      "grad_norm": 0.5848962664604187,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 4180
+    },
+    {
+      "epoch": 3.525452250736222,
+      "grad_norm": 0.5335569977760315,
+      "learning_rate": 0.0002,
+      "loss": 1.5892,
+      "step": 4190
+    },
+    {
+      "epoch": 3.53386621792175,
+      "grad_norm": 0.547964870929718,
+      "learning_rate": 0.0002,
+      "loss": 1.5152,
+      "step": 4200
+    },
+    {
+      "epoch": 3.542280185107278,
+      "grad_norm": 0.6157727241516113,
+      "learning_rate": 0.0002,
+      "loss": 1.4887,
+      "step": 4210
+    },
+    {
+      "epoch": 3.550694152292806,
+      "grad_norm": 0.6163121461868286,
+      "learning_rate": 0.0002,
+      "loss": 1.5484,
+      "step": 4220
+    },
+    {
+      "epoch": 3.559108119478334,
+      "grad_norm": 0.5844616293907166,
+      "learning_rate": 0.0002,
+      "loss": 1.5833,
+      "step": 4230
+    },
+    {
+      "epoch": 3.567522086663862,
+      "grad_norm": 0.7104926109313965,
+      "learning_rate": 0.0002,
+      "loss": 1.5305,
+      "step": 4240
+    },
+    {
+      "epoch": 3.57593605384939,
+      "grad_norm": 0.5055213570594788,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4250
+    },
+    {
+      "epoch": 3.584350021034918,
+      "grad_norm": 0.611676812171936,
+      "learning_rate": 0.0002,
+      "loss": 1.482,
+      "step": 4260
+    },
+    {
+      "epoch": 3.592763988220446,
+      "grad_norm": 0.6326440572738647,
+      "learning_rate": 0.0002,
+      "loss": 1.5048,
+      "step": 4270
+    },
+    {
+      "epoch": 3.6011779554059737,
+      "grad_norm": 0.6290925741195679,
+      "learning_rate": 0.0002,
+      "loss": 1.5122,
+      "step": 4280
+    },
+    {
+      "epoch": 3.609591922591502,
+      "grad_norm": 0.5691978931427002,
+      "learning_rate": 0.0002,
+      "loss": 1.5654,
+      "step": 4290
+    },
+    {
+      "epoch": 3.61800588977703,
+      "grad_norm": 0.6071329116821289,
+      "learning_rate": 0.0002,
+      "loss": 1.4854,
+      "step": 4300
+    },
+    {
+      "epoch": 3.626419856962558,
+      "grad_norm": 0.606573224067688,
+      "learning_rate": 0.0002,
+      "loss": 1.5336,
+      "step": 4310
+    },
+    {
+      "epoch": 3.6348338241480858,
+      "grad_norm": 0.5515419244766235,
+      "learning_rate": 0.0002,
+      "loss": 1.6437,
+      "step": 4320
+    },
+    {
+      "epoch": 3.643247791333614,
+      "grad_norm": 0.5964660048484802,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 4330
+    },
+    {
+      "epoch": 3.6516617585191415,
+      "grad_norm": 0.5774146914482117,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 4340
+    },
+    {
+      "epoch": 3.6600757257046697,
+      "grad_norm": 0.5732731223106384,
+      "learning_rate": 0.0002,
+      "loss": 1.5566,
+      "step": 4350
+    },
+    {
+      "epoch": 3.6684896928901978,
+      "grad_norm": 0.7354163527488708,
+      "learning_rate": 0.0002,
+      "loss": 1.5682,
+      "step": 4360
+    },
+    {
+      "epoch": 3.676903660075726,
+      "grad_norm": 0.6220902800559998,
+      "learning_rate": 0.0002,
+      "loss": 1.5225,
+      "step": 4370
+    },
+    {
+      "epoch": 3.6853176272612536,
+      "grad_norm": 0.6053991317749023,
+      "learning_rate": 0.0002,
+      "loss": 1.4838,
+      "step": 4380
+    },
+    {
+      "epoch": 3.6937315944467817,
+      "grad_norm": 0.67010897397995,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4390
+    },
+    {
+      "epoch": 3.70214556163231,
+      "grad_norm": 0.6139186024665833,
+      "learning_rate": 0.0002,
+      "loss": 1.5381,
+      "step": 4400
+    },
+    {
+      "epoch": 3.7105595288178375,
+      "grad_norm": 0.5433071851730347,
+      "learning_rate": 0.0002,
+      "loss": 1.5088,
+      "step": 4410
+    },
+    {
+      "epoch": 3.7189734960033656,
+      "grad_norm": 0.5453870296478271,
+      "learning_rate": 0.0002,
+      "loss": 1.5337,
+      "step": 4420
+    },
+    {
+      "epoch": 3.7273874631888937,
+      "grad_norm": 0.6401727199554443,
+      "learning_rate": 0.0002,
+      "loss": 1.4549,
+      "step": 4430
+    },
+    {
+      "epoch": 3.735801430374422,
+      "grad_norm": 0.6049367189407349,
+      "learning_rate": 0.0002,
+      "loss": 1.503,
+      "step": 4440
+    },
+    {
+      "epoch": 3.7442153975599495,
+      "grad_norm": 0.5740529298782349,
+      "learning_rate": 0.0002,
+      "loss": 1.5268,
+      "step": 4450
+    },
+    {
+      "epoch": 3.7526293647454776,
+      "grad_norm": 0.6521880626678467,
+      "learning_rate": 0.0002,
+      "loss": 1.5183,
+      "step": 4460
+    },
+    {
+      "epoch": 3.7610433319310053,
+      "grad_norm": 0.7096368074417114,
+      "learning_rate": 0.0002,
+      "loss": 1.5741,
+      "step": 4470
+    },
+    {
+      "epoch": 3.7694572991165334,
+      "grad_norm": 0.5886474251747131,
+      "learning_rate": 0.0002,
+      "loss": 1.5786,
+      "step": 4480
+    },
+    {
+      "epoch": 3.7778712663020615,
+      "grad_norm": 0.5821043252944946,
+      "learning_rate": 0.0002,
+      "loss": 1.5887,
+      "step": 4490
+    },
+    {
+      "epoch": 3.7862852334875896,
+      "grad_norm": 0.628892183303833,
+      "learning_rate": 0.0002,
+      "loss": 1.5777,
+      "step": 4500
+    },
+    {
+      "epoch": 3.7946992006731173,
+      "grad_norm": 0.5962669849395752,
+      "learning_rate": 0.0002,
+      "loss": 1.4708,
+      "step": 4510
+    },
+    {
+      "epoch": 3.8031131678586454,
+      "grad_norm": 0.6635549068450928,
+      "learning_rate": 0.0002,
+      "loss": 1.5267,
+      "step": 4520
+    },
+    {
+      "epoch": 3.811527135044173,
+      "grad_norm": 0.6010760068893433,
+      "learning_rate": 0.0002,
+      "loss": 1.5058,
+      "step": 4530
+    },
+    {
+      "epoch": 3.819941102229701,
+      "grad_norm": 0.6322658658027649,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 4540
+    },
+    {
+      "epoch": 3.8283550694152293,
+      "grad_norm": 0.5893137454986572,
+      "learning_rate": 0.0002,
+      "loss": 1.5029,
+      "step": 4550
+    },
+    {
+      "epoch": 3.8367690366007574,
+      "grad_norm": 0.7829602360725403,
+      "learning_rate": 0.0002,
+      "loss": 1.5435,
+      "step": 4560
+    },
+    {
+      "epoch": 3.845183003786285,
+      "grad_norm": 0.6190396547317505,
+      "learning_rate": 0.0002,
+      "loss": 1.5453,
+      "step": 4570
+    },
+    {
+      "epoch": 3.853596970971813,
+      "grad_norm": 0.6662813425064087,
+      "learning_rate": 0.0002,
+      "loss": 1.5292,
+      "step": 4580
+    },
+    {
+      "epoch": 3.8620109381573413,
+      "grad_norm": 0.5809855461120605,
+      "learning_rate": 0.0002,
+      "loss": 1.5065,
+      "step": 4590
+    },
+    {
+      "epoch": 3.870424905342869,
+      "grad_norm": 0.5779069662094116,
+      "learning_rate": 0.0002,
+      "loss": 1.5041,
+      "step": 4600
+    },
+    {
+      "epoch": 3.878838872528397,
+      "grad_norm": 0.5603038668632507,
+      "learning_rate": 0.0002,
+      "loss": 1.498,
+      "step": 4610
+    },
+    {
+      "epoch": 3.887252839713925,
+      "grad_norm": 0.6274181008338928,
+      "learning_rate": 0.0002,
+      "loss": 1.5372,
+      "step": 4620
+    },
+    {
+      "epoch": 3.8956668068994533,
+      "grad_norm": 0.6810959577560425,
+      "learning_rate": 0.0002,
+      "loss": 1.4996,
+      "step": 4630
+    },
+    {
+      "epoch": 3.904080774084981,
+      "grad_norm": 0.5647315979003906,
+      "learning_rate": 0.0002,
+      "loss": 1.4956,
+      "step": 4640
+    },
+    {
+      "epoch": 3.912494741270509,
+      "grad_norm": 0.6830295324325562,
+      "learning_rate": 0.0002,
+      "loss": 1.5424,
+      "step": 4650
+    },
+    {
+      "epoch": 3.920908708456037,
+      "grad_norm": 0.652565598487854,
+      "learning_rate": 0.0002,
+      "loss": 1.535,
+      "step": 4660
+    },
+    {
+      "epoch": 3.929322675641565,
+      "grad_norm": 0.5806284546852112,
+      "learning_rate": 0.0002,
+      "loss": 1.4772,
+      "step": 4670
+    },
+    {
+      "epoch": 3.937736642827093,
+      "grad_norm": 0.6825073957443237,
+      "learning_rate": 0.0002,
+      "loss": 1.5812,
+      "step": 4680
+    },
+    {
+      "epoch": 3.946150610012621,
+      "grad_norm": 0.6149451732635498,
+      "learning_rate": 0.0002,
+      "loss": 1.5516,
+      "step": 4690
+    },
+    {
+      "epoch": 3.954564577198149,
+      "grad_norm": 0.6152557134628296,
+      "learning_rate": 0.0002,
+      "loss": 1.5608,
+      "step": 4700
+    },
+    {
+      "epoch": 3.962978544383677,
+      "grad_norm": 0.6239011883735657,
+      "learning_rate": 0.0002,
+      "loss": 1.4897,
+      "step": 4710
+    },
+    {
+      "epoch": 3.971392511569205,
+      "grad_norm": 0.6485443115234375,
+      "learning_rate": 0.0002,
+      "loss": 1.538,
+      "step": 4720
+    },
+    {
+      "epoch": 3.9798064787547327,
+      "grad_norm": 0.6449228525161743,
+      "learning_rate": 0.0002,
+      "loss": 1.5226,
+      "step": 4730
+    },
+    {
+      "epoch": 3.988220445940261,
+      "grad_norm": 0.6526407599449158,
+      "learning_rate": 0.0002,
+      "loss": 1.5087,
+      "step": 4740
+    },
+    {
+      "epoch": 3.996634413125789,
+      "grad_norm": 0.6277706027030945,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4750
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.871641755104065,
+      "eval_runtime": 37.9637,
+      "eval_samples_per_second": 13.566,
+      "eval_steps_per_second": 1.712,
+      "step": 4754
+    },
+    {
+      "epoch": 4.005048380311317,
+      "grad_norm": 0.6994837522506714,
+      "learning_rate": 0.0002,
+      "loss": 1.4744,
+      "step": 4760
+    },
+    {
+      "epoch": 4.013462347496845,
+      "grad_norm": 0.8728373050689697,
+      "learning_rate": 0.0002,
+      "loss": 1.4433,
+      "step": 4770
+    },
+    {
+      "epoch": 4.021876314682372,
+      "grad_norm": 0.688679575920105,
+      "learning_rate": 0.0002,
+      "loss": 1.3329,
+      "step": 4780
+    },
+    {
+      "epoch": 4.0302902818679005,
+      "grad_norm": 0.6313387155532837,
+      "learning_rate": 0.0002,
+      "loss": 1.3999,
+      "step": 4790
+    },
+    {
+      "epoch": 4.038704249053429,
+      "grad_norm": 0.6577984690666199,
+      "learning_rate": 0.0002,
+      "loss": 1.3346,
+      "step": 4800
+    },
+    {
+      "epoch": 4.047118216238957,
+      "grad_norm": 0.7938185930252075,
+      "learning_rate": 0.0002,
+      "loss": 1.3403,
+      "step": 4810
+    },
+    {
+      "epoch": 4.055532183424485,
+      "grad_norm": 0.760399580001831,
+      "learning_rate": 0.0002,
+      "loss": 1.3716,
+      "step": 4820
+    },
+    {
+      "epoch": 4.063946150610013,
+      "grad_norm": 0.7329602241516113,
+      "learning_rate": 0.0002,
+      "loss": 1.4321,
+      "step": 4830
+    },
+    {
+      "epoch": 4.07236011779554,
+      "grad_norm": 0.7778576016426086,
+      "learning_rate": 0.0002,
+      "loss": 1.4133,
+      "step": 4840
+    },
+    {
+      "epoch": 4.080774084981068,
+      "grad_norm": 0.8235865235328674,
+      "learning_rate": 0.0002,
+      "loss": 1.4372,
+      "step": 4850
+    },
+    {
+      "epoch": 4.089188052166596,
+      "grad_norm": 0.7743754386901855,
+      "learning_rate": 0.0002,
+      "loss": 1.3719,
+      "step": 4860
+    },
+    {
+      "epoch": 4.0976020193521245,
+      "grad_norm": 0.8145367503166199,
+      "learning_rate": 0.0002,
+      "loss": 1.3787,
+      "step": 4870
+    },
+    {
+      "epoch": 4.106015986537653,
+      "grad_norm": 0.8517307639122009,
+      "learning_rate": 0.0002,
+      "loss": 1.356,
+      "step": 4880
+    },
+    {
+      "epoch": 4.114429953723181,
+      "grad_norm": 0.8208953142166138,
+      "learning_rate": 0.0002,
+      "loss": 1.4191,
+      "step": 4890
+    },
+    {
+      "epoch": 4.122843920908709,
+      "grad_norm": 0.8437790870666504,
+      "learning_rate": 0.0002,
+      "loss": 1.3189,
+      "step": 4900
+    },
+    {
+      "epoch": 4.131257888094236,
+      "grad_norm": 0.716672420501709,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 4910
+    },
+    {
+      "epoch": 4.139671855279764,
+      "grad_norm": 0.7656235098838806,
+      "learning_rate": 0.0002,
+      "loss": 1.4392,
+      "step": 4920
+    },
+    {
+      "epoch": 4.148085822465292,
+      "grad_norm": 0.7209306955337524,
+      "learning_rate": 0.0002,
+      "loss": 1.3408,
+      "step": 4930
+    },
+    {
+      "epoch": 4.1564997896508205,
+      "grad_norm": 0.7731267809867859,
+      "learning_rate": 0.0002,
+      "loss": 1.3639,
+      "step": 4940
+    },
+    {
+      "epoch": 4.164913756836349,
+      "grad_norm": 0.7477553486824036,
+      "learning_rate": 0.0002,
+      "loss": 1.4151,
+      "step": 4950
+    },
+    {
+      "epoch": 4.173327724021877,
+      "grad_norm": 0.7372981309890747,
+      "learning_rate": 0.0002,
+      "loss": 1.3485,
+      "step": 4960
+    },
+    {
+      "epoch": 4.181741691207404,
+      "grad_norm": 0.6582154035568237,
+      "learning_rate": 0.0002,
+      "loss": 1.3901,
+      "step": 4970
+    },
+    {
+      "epoch": 4.190155658392932,
+      "grad_norm": 0.7003206610679626,
+      "learning_rate": 0.0002,
+      "loss": 1.3343,
+      "step": 4980
+    },
+    {
+      "epoch": 4.19856962557846,
+      "grad_norm": 0.735223650932312,
+      "learning_rate": 0.0002,
+      "loss": 1.4098,
+      "step": 4990
+    },
+    {
+      "epoch": 4.206983592763988,
+      "grad_norm": 0.7832302451133728,
+      "learning_rate": 0.0002,
+      "loss": 1.3564,
+      "step": 5000
+    },
+    {
+      "epoch": 4.215397559949516,
+      "grad_norm": 0.8819546103477478,
+      "learning_rate": 0.0002,
+      "loss": 1.3622,
+      "step": 5010
+    },
+    {
+      "epoch": 4.2238115271350445,
+      "grad_norm": 0.9325336813926697,
+      "learning_rate": 0.0002,
+      "loss": 1.4438,
+      "step": 5020
+    },
+    {
+      "epoch": 4.232225494320572,
+      "grad_norm": 0.7007517218589783,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 5030
+    },
+    {
+      "epoch": 4.2406394615061,
+      "grad_norm": 0.7118321061134338,
+      "learning_rate": 0.0002,
+      "loss": 1.3683,
+      "step": 5040
+    },
+    {
+      "epoch": 4.249053428691628,
+      "grad_norm": 0.6578946709632874,
+      "learning_rate": 0.0002,
+      "loss": 1.2365,
+      "step": 5050
+    },
+    {
+      "epoch": 4.257467395877156,
+      "grad_norm": 0.9438983798027039,
+      "learning_rate": 0.0002,
+      "loss": 1.3696,
+      "step": 5060
+    },
+    {
+      "epoch": 4.265881363062684,
+      "grad_norm": 0.703037679195404,
+      "learning_rate": 0.0002,
+      "loss": 1.3868,
+      "step": 5070
+    },
+    {
+      "epoch": 4.274295330248212,
+      "grad_norm": 0.7286025285720825,
+      "learning_rate": 0.0002,
+      "loss": 1.3687,
+      "step": 5080
+    },
+    {
+      "epoch": 4.28270929743374,
+      "grad_norm": 0.750689685344696,
+      "learning_rate": 0.0002,
+      "loss": 1.3605,
+      "step": 5090
+    },
+    {
+      "epoch": 4.291123264619268,
+      "grad_norm": 0.869753360748291,
+      "learning_rate": 0.0002,
+      "loss": 1.5089,
+      "step": 5100
+    },
+    {
+      "epoch": 4.299537231804796,
+      "grad_norm": 0.8712980151176453,
+      "learning_rate": 0.0002,
+      "loss": 1.4128,
+      "step": 5110
+    },
+    {
+      "epoch": 4.307951198990324,
+      "grad_norm": 0.690263569355011,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 5120
+    },
+    {
+      "epoch": 4.316365166175852,
+      "grad_norm": 0.7114760279655457,
+      "learning_rate": 0.0002,
+      "loss": 1.4088,
+      "step": 5130
+    },
+    {
+      "epoch": 4.32477913336138,
+      "grad_norm": 0.7588112354278564,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 5140
+    },
+    {
+      "epoch": 4.333193100546908,
+      "grad_norm": 0.7556202411651611,
+      "learning_rate": 0.0002,
+      "loss": 1.4408,
+      "step": 5150
+    },
+    {
+      "epoch": 4.341607067732435,
+      "grad_norm": 0.8357610702514648,
+      "learning_rate": 0.0002,
+      "loss": 1.4203,
+      "step": 5160
+    },
+    {
+      "epoch": 4.3500210349179635,
+      "grad_norm": 0.8054035902023315,
+      "learning_rate": 0.0002,
+      "loss": 1.3348,
+      "step": 5170
+    },
+    {
+      "epoch": 4.358435002103492,
+      "grad_norm": 0.7637107968330383,
+      "learning_rate": 0.0002,
+      "loss": 1.3109,
+      "step": 5180
+    },
+    {
+      "epoch": 4.36684896928902,
+      "grad_norm": 0.757481038570404,
+      "learning_rate": 0.0002,
+      "loss": 1.3744,
+      "step": 5190
+    },
+    {
+      "epoch": 4.375262936474548,
+      "grad_norm": 0.7185863852500916,
+      "learning_rate": 0.0002,
+      "loss": 1.3622,
+      "step": 5200
+    },
+    {
+      "epoch": 4.383676903660076,
+      "grad_norm": 0.7326455116271973,
+      "learning_rate": 0.0002,
+      "loss": 1.3896,
+      "step": 5210
+    },
+    {
+      "epoch": 4.392090870845603,
+      "grad_norm": 0.7980523109436035,
+      "learning_rate": 0.0002,
+      "loss": 1.4098,
+      "step": 5220
+    },
+    {
+      "epoch": 4.400504838031131,
+      "grad_norm": 0.8526999354362488,
+      "learning_rate": 0.0002,
+      "loss": 1.3783,
+      "step": 5230
+    },
+    {
+      "epoch": 4.4089188052166595,
+      "grad_norm": 0.7012337446212769,
+      "learning_rate": 0.0002,
+      "loss": 1.4022,
+      "step": 5240
+    },
+    {
+      "epoch": 4.417332772402188,
+      "grad_norm": 0.8217827677726746,
+      "learning_rate": 0.0002,
+      "loss": 1.3552,
+      "step": 5250
+    },
+    {
+      "epoch": 4.425746739587716,
+      "grad_norm": 0.7141005396842957,
+      "learning_rate": 0.0002,
+      "loss": 1.3482,
+      "step": 5260
+    },
+    {
+      "epoch": 4.434160706773244,
+      "grad_norm": 0.7094302177429199,
+      "learning_rate": 0.0002,
+      "loss": 1.3699,
+      "step": 5270
+    },
+    {
+      "epoch": 4.442574673958772,
+      "grad_norm": 0.7234613299369812,
+      "learning_rate": 0.0002,
+      "loss": 1.3527,
+      "step": 5280
+    },
+    {
+      "epoch": 4.450988641144299,
+      "grad_norm": 0.7530457973480225,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 5290
+    },
+    {
+      "epoch": 4.459402608329827,
+      "grad_norm": 0.7300912141799927,
+      "learning_rate": 0.0002,
+      "loss": 1.3944,
+      "step": 5300
+    },
+    {
+      "epoch": 4.467816575515355,
+      "grad_norm": 0.825443685054779,
+      "learning_rate": 0.0002,
+      "loss": 1.3844,
+      "step": 5310
+    },
+    {
+      "epoch": 4.4762305427008835,
+      "grad_norm": 0.7559658885002136,
+      "learning_rate": 0.0002,
+      "loss": 1.3648,
+      "step": 5320
+    },
+    {
+      "epoch": 4.484644509886412,
+      "grad_norm": 0.8817561268806458,
+      "learning_rate": 0.0002,
+      "loss": 1.4364,
+      "step": 5330
+    },
+    {
+      "epoch": 4.49305847707194,
+      "grad_norm": 0.8203575611114502,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 5340
+    },
+    {
+      "epoch": 4.501472444257468,
+      "grad_norm": 0.7677690982818604,
+      "learning_rate": 0.0002,
+      "loss": 1.3996,
+      "step": 5350
+    },
+    {
+      "epoch": 4.509886411442995,
+      "grad_norm": 0.657085120677948,
+      "learning_rate": 0.0002,
+      "loss": 1.4142,
+      "step": 5360
+    },
+    {
+      "epoch": 4.518300378628523,
+      "grad_norm": 0.7939504384994507,
+      "learning_rate": 0.0002,
+      "loss": 1.3722,
+      "step": 5370
+    },
+    {
+      "epoch": 4.526714345814051,
+      "grad_norm": 0.6971889138221741,
+      "learning_rate": 0.0002,
+      "loss": 1.4361,
+      "step": 5380
+    },
+    {
+      "epoch": 4.535128312999579,
+      "grad_norm": 0.6984175443649292,
+      "learning_rate": 0.0002,
+      "loss": 1.3637,
+      "step": 5390
+    },
+    {
+      "epoch": 4.5435422801851075,
+      "grad_norm": 0.8504858613014221,
+      "learning_rate": 0.0002,
+      "loss": 1.341,
+      "step": 5400
+    },
+    {
+      "epoch": 4.551956247370635,
+      "grad_norm": 0.9134073853492737,
+      "learning_rate": 0.0002,
+      "loss": 1.4026,
+      "step": 5410
+    },
+    {
+      "epoch": 4.560370214556163,
+      "grad_norm": 0.7765598893165588,
+      "learning_rate": 0.0002,
+      "loss": 1.4375,
+      "step": 5420
+    },
+    {
+      "epoch": 4.568784181741691,
+      "grad_norm": 0.6991009712219238,
+      "learning_rate": 0.0002,
+      "loss": 1.4832,
+      "step": 5430
+    },
+    {
+      "epoch": 4.577198148927219,
+      "grad_norm": 0.8393039107322693,
+      "learning_rate": 0.0002,
+      "loss": 1.4021,
+      "step": 5440
+    },
+    {
+      "epoch": 4.585612116112747,
+      "grad_norm": 0.7685918211936951,
+      "learning_rate": 0.0002,
+      "loss": 1.3976,
+      "step": 5450
+    },
+    {
+      "epoch": 4.594026083298275,
+      "grad_norm": 0.7135679721832275,
+      "learning_rate": 0.0002,
+      "loss": 1.3883,
+      "step": 5460
+    },
+    {
+      "epoch": 4.6024400504838034,
+      "grad_norm": 0.6728870868682861,
+      "learning_rate": 0.0002,
+      "loss": 1.4083,
+      "step": 5470
+    },
+    {
+      "epoch": 4.610854017669331,
+      "grad_norm": 0.7139479517936707,
+      "learning_rate": 0.0002,
+      "loss": 1.3698,
+      "step": 5480
+    },
+    {
+      "epoch": 4.619267984854859,
+      "grad_norm": 0.8476598858833313,
+      "learning_rate": 0.0002,
+      "loss": 1.3498,
+      "step": 5490
+    },
+    {
+      "epoch": 4.627681952040387,
+      "grad_norm": 0.8034361004829407,
+      "learning_rate": 0.0002,
+      "loss": 1.3389,
+      "step": 5500
+    },
+    {
+      "epoch": 4.636095919225915,
+      "grad_norm": 0.7452183961868286,
+      "learning_rate": 0.0002,
+      "loss": 1.4179,
+      "step": 5510
+    },
+    {
+      "epoch": 4.644509886411443,
+      "grad_norm": 0.8394148945808411,
+      "learning_rate": 0.0002,
+      "loss": 1.4031,
+      "step": 5520
+    },
+    {
+      "epoch": 4.652923853596971,
+      "grad_norm": 0.7480153441429138,
+      "learning_rate": 0.0002,
+      "loss": 1.4561,
+      "step": 5530
+    },
+    {
+      "epoch": 4.661337820782499,
+      "grad_norm": 0.7781714797019958,
+      "learning_rate": 0.0002,
+      "loss": 1.378,
+      "step": 5540
+    },
+    {
+      "epoch": 4.669751787968027,
+      "grad_norm": 1.0058213472366333,
+      "learning_rate": 0.0002,
+      "loss": 1.3924,
+      "step": 5550
+    },
+    {
+      "epoch": 4.678165755153555,
+      "grad_norm": 0.7403179407119751,
+      "learning_rate": 0.0002,
+      "loss": 1.4198,
+      "step": 5560
+    },
+    {
+      "epoch": 4.686579722339083,
+      "grad_norm": 0.7270476818084717,
+      "learning_rate": 0.0002,
+      "loss": 1.4328,
+      "step": 5570
+    },
+    {
+      "epoch": 4.694993689524611,
+      "grad_norm": 0.760877788066864,
+      "learning_rate": 0.0002,
+      "loss": 1.378,
+      "step": 5580
+    },
+    {
+      "epoch": 4.703407656710139,
+      "grad_norm": 0.8097004890441895,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 5590
+    },
+    {
+      "epoch": 4.711821623895667,
+      "grad_norm": 0.9096523523330688,
+      "learning_rate": 0.0002,
+      "loss": 1.3661,
+      "step": 5600
+    },
+    {
+      "epoch": 4.720235591081195,
+      "grad_norm": 0.7262444496154785,
+      "learning_rate": 0.0002,
+      "loss": 1.4012,
+      "step": 5610
+    },
+    {
+      "epoch": 4.7286495582667225,
+      "grad_norm": 0.8207762837409973,
+      "learning_rate": 0.0002,
+      "loss": 1.422,
+      "step": 5620
+    },
+    {
+      "epoch": 4.737063525452251,
+      "grad_norm": 0.8089601993560791,
+      "learning_rate": 0.0002,
+      "loss": 1.4017,
+      "step": 5630
+    },
+    {
+      "epoch": 4.745477492637779,
+      "grad_norm": 0.7609543800354004,
+      "learning_rate": 0.0002,
+      "loss": 1.3675,
+      "step": 5640
+    },
+    {
+      "epoch": 4.753891459823307,
+      "grad_norm": 0.7273501753807068,
+      "learning_rate": 0.0002,
+      "loss": 1.4085,
+      "step": 5650
+    },
+    {
+      "epoch": 4.762305427008835,
+      "grad_norm": 0.7800219058990479,
+      "learning_rate": 0.0002,
+      "loss": 1.3849,
+      "step": 5660
+    },
+    {
+      "epoch": 4.770719394194362,
+      "grad_norm": 0.8558377623558044,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 5670
+    },
+    {
+      "epoch": 4.77913336137989,
+      "grad_norm": 0.7131547927856445,
+      "learning_rate": 0.0002,
+      "loss": 1.3831,
+      "step": 5680
+    },
+    {
+      "epoch": 4.787547328565418,
+      "grad_norm": 0.7651025056838989,
+      "learning_rate": 0.0002,
+      "loss": 1.407,
+      "step": 5690
+    },
+    {
+      "epoch": 4.7959612957509465,
+      "grad_norm": 0.8129976391792297,
+      "learning_rate": 0.0002,
+      "loss": 1.3882,
+      "step": 5700
+    },
+    {
+      "epoch": 4.804375262936475,
+      "grad_norm": 0.8019895553588867,
+      "learning_rate": 0.0002,
+      "loss": 1.4347,
+      "step": 5710
+    },
+    {
+      "epoch": 4.812789230122003,
+      "grad_norm": 0.7692018151283264,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 5720
+    },
+    {
+      "epoch": 4.821203197307531,
+      "grad_norm": 0.6893943548202515,
+      "learning_rate": 0.0002,
+      "loss": 1.419,
+      "step": 5730
+    },
+    {
+      "epoch": 4.829617164493058,
+      "grad_norm": 0.6881810426712036,
+      "learning_rate": 0.0002,
+      "loss": 1.4453,
+      "step": 5740
+    },
+    {
+      "epoch": 4.838031131678586,
+      "grad_norm": 0.7838267683982849,
+      "learning_rate": 0.0002,
+      "loss": 1.4775,
+      "step": 5750
+    },
+    {
+      "epoch": 4.846445098864114,
+      "grad_norm": 0.727799117565155,
+      "learning_rate": 0.0002,
+      "loss": 1.3857,
+      "step": 5760
+    },
+    {
+      "epoch": 4.8548590660496425,
+      "grad_norm": 0.7458277344703674,
+      "learning_rate": 0.0002,
+      "loss": 1.4685,
+      "step": 5770
+    },
+    {
+      "epoch": 4.863273033235171,
+      "grad_norm": 0.903802216053009,
+      "learning_rate": 0.0002,
+      "loss": 1.4426,
+      "step": 5780
+    },
+    {
+      "epoch": 4.871687000420699,
+      "grad_norm": 0.7983472347259521,
+      "learning_rate": 0.0002,
+      "loss": 1.451,
+      "step": 5790
+    },
+    {
+      "epoch": 4.880100967606227,
+      "grad_norm": 0.6894361972808838,
+      "learning_rate": 0.0002,
+      "loss": 1.4534,
+      "step": 5800
+    },
+    {
+      "epoch": 4.888514934791754,
+      "grad_norm": 0.7499409317970276,
+      "learning_rate": 0.0002,
+      "loss": 1.4486,
+      "step": 5810
+    },
+    {
+      "epoch": 4.896928901977282,
+      "grad_norm": 0.7362820506095886,
+      "learning_rate": 0.0002,
+      "loss": 1.4253,
+      "step": 5820
+    },
+    {
+      "epoch": 4.90534286916281,
+      "grad_norm": 0.8341619968414307,
+      "learning_rate": 0.0002,
+      "loss": 1.3763,
+      "step": 5830
+    },
+    {
+      "epoch": 4.913756836348338,
+      "grad_norm": 0.9604470133781433,
+      "learning_rate": 0.0002,
+      "loss": 1.3748,
+      "step": 5840
+    },
+    {
+      "epoch": 4.9221708035338665,
+      "grad_norm": 0.8916844129562378,
+      "learning_rate": 0.0002,
+      "loss": 1.3658,
+      "step": 5850
+    },
+    {
+      "epoch": 4.930584770719394,
+      "grad_norm": 0.8519647121429443,
+      "learning_rate": 0.0002,
+      "loss": 1.363,
+      "step": 5860
+    },
+    {
+      "epoch": 4.938998737904922,
+      "grad_norm": 0.7946906089782715,
+      "learning_rate": 0.0002,
+      "loss": 1.424,
+      "step": 5870
+    },
+    {
+      "epoch": 4.94741270509045,
+      "grad_norm": 0.7843789458274841,
+      "learning_rate": 0.0002,
+      "loss": 1.4071,
+      "step": 5880
+    },
+    {
+      "epoch": 4.955826672275978,
+      "grad_norm": 0.707618772983551,
+      "learning_rate": 0.0002,
+      "loss": 1.4021,
+      "step": 5890
+    },
+    {
+      "epoch": 4.964240639461506,
+      "grad_norm": 0.7704206109046936,
+      "learning_rate": 0.0002,
+      "loss": 1.502,
+      "step": 5900
+    },
+    {
+      "epoch": 4.972654606647034,
+      "grad_norm": 0.7160256505012512,
+      "learning_rate": 0.0002,
+      "loss": 1.4456,
+      "step": 5910
+    },
+    {
+      "epoch": 4.981068573832562,
+      "grad_norm": 0.7020420432090759,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 5920
+    },
+    {
+      "epoch": 4.98948254101809,
+      "grad_norm": 0.7576286792755127,
+      "learning_rate": 0.0002,
+      "loss": 1.4037,
+      "step": 5930
+    },
+    {
+      "epoch": 4.997896508203618,
+      "grad_norm": 0.8573036789894104,
+      "learning_rate": 0.0002,
+      "loss": 1.414,
+      "step": 5940
+    },
+    {
+      "epoch": 4.999579301640724,
+      "eval_loss": 1.9353811740875244,
+      "eval_runtime": 37.9208,
+      "eval_samples_per_second": 13.581,
+      "eval_steps_per_second": 1.714,
+      "step": 5942
+    },
+    {
+      "epoch": 5.006310475389146,
+      "grad_norm": 0.8204267621040344,
+      "learning_rate": 0.0002,
+      "loss": 1.2418,
+      "step": 5950
+    },
+    {
+      "epoch": 5.014724442574674,
+      "grad_norm": 0.976840615272522,
+      "learning_rate": 0.0002,
+      "loss": 1.235,
+      "step": 5960
+    },
+    {
+      "epoch": 5.023138409760202,
+      "grad_norm": 0.8765613436698914,
+      "learning_rate": 0.0002,
+      "loss": 1.2134,
+      "step": 5970
+    },
+    {
+      "epoch": 5.03155237694573,
+      "grad_norm": 1.1793042421340942,
+      "learning_rate": 0.0002,
+      "loss": 1.2748,
+      "step": 5980
+    },
+    {
+      "epoch": 5.039966344131258,
+      "grad_norm": 0.971062958240509,
+      "learning_rate": 0.0002,
+      "loss": 1.2412,
+      "step": 5990
+    },
+    {
+      "epoch": 5.0483803113167856,
+      "grad_norm": 0.8649757504463196,
+      "learning_rate": 0.0002,
+      "loss": 1.1819,
+      "step": 6000
+    },
+    {
+      "epoch": 5.056794278502314,
+      "grad_norm": 0.9563034176826477,
+      "learning_rate": 0.0002,
+      "loss": 1.1654,
+      "step": 6010
+    },
+    {
+      "epoch": 5.065208245687842,
+      "grad_norm": 1.0093994140625,
+      "learning_rate": 0.0002,
+      "loss": 1.2238,
+      "step": 6020
+    },
+    {
+      "epoch": 5.07362221287337,
+      "grad_norm": 1.004213571548462,
+      "learning_rate": 0.0002,
+      "loss": 1.2519,
+      "step": 6030
+    },
+    {
+      "epoch": 5.082036180058898,
+      "grad_norm": 0.8307787179946899,
+      "learning_rate": 0.0002,
+      "loss": 1.2379,
+      "step": 6040
+    },
+    {
+      "epoch": 5.090450147244426,
+      "grad_norm": 0.9117848873138428,
+      "learning_rate": 0.0002,
+      "loss": 1.2282,
+      "step": 6050
+    },
+    {
+      "epoch": 5.098864114429953,
+      "grad_norm": 1.0269840955734253,
+      "learning_rate": 0.0002,
+      "loss": 1.2582,
+      "step": 6060
+    },
+    {
+      "epoch": 5.1072780816154815,
+      "grad_norm": 0.9079542756080627,
+      "learning_rate": 0.0002,
+      "loss": 1.1836,
+      "step": 6070
+    },
+    {
+      "epoch": 5.11569204880101,
+      "grad_norm": 0.885702908039093,
+      "learning_rate": 0.0002,
+      "loss": 1.215,
+      "step": 6080
+    },
+    {
+      "epoch": 5.124106015986538,
+      "grad_norm": 0.9976128339767456,
+      "learning_rate": 0.0002,
+      "loss": 1.2406,
+      "step": 6090
+    },
+    {
+      "epoch": 5.132519983172066,
+      "grad_norm": 0.8472117185592651,
+      "learning_rate": 0.0002,
+      "loss": 1.3082,
+      "step": 6100
+    },
+    {
+      "epoch": 5.140933950357594,
+      "grad_norm": 1.0385161638259888,
+      "learning_rate": 0.0002,
+      "loss": 1.226,
+      "step": 6110
+    },
+    {
+      "epoch": 5.149347917543121,
+      "grad_norm": 0.8948383927345276,
+      "learning_rate": 0.0002,
+      "loss": 1.213,
+      "step": 6120
+    },
+    {
+      "epoch": 5.157761884728649,
+      "grad_norm": 1.2613716125488281,
+      "learning_rate": 0.0002,
+      "loss": 1.2213,
+      "step": 6130
+    },
+    {
+      "epoch": 5.166175851914177,
+      "grad_norm": 0.9933410286903381,
+      "learning_rate": 0.0002,
+      "loss": 1.2632,
+      "step": 6140
+    },
+    {
+      "epoch": 5.1745898190997055,
+      "grad_norm": 0.9673663973808289,
+      "learning_rate": 0.0002,
+      "loss": 1.1715,
+      "step": 6150
+    },
+    {
+      "epoch": 5.183003786285234,
+      "grad_norm": 0.9969648122787476,
+      "learning_rate": 0.0002,
+      "loss": 1.2947,
+      "step": 6160
+    },
+    {
+      "epoch": 5.191417753470762,
+      "grad_norm": 1.2163258790969849,
+      "learning_rate": 0.0002,
+      "loss": 1.2416,
+      "step": 6170
+    },
+    {
+      "epoch": 5.19983172065629,
+      "grad_norm": 0.9163419604301453,
+      "learning_rate": 0.0002,
+      "loss": 1.2221,
+      "step": 6180
+    },
+    {
+      "epoch": 5.208245687841817,
+      "grad_norm": 0.9225585460662842,
+      "learning_rate": 0.0002,
+      "loss": 1.2624,
+      "step": 6190
+    },
+    {
+      "epoch": 5.216659655027345,
+      "grad_norm": 0.9205296635627747,
+      "learning_rate": 0.0002,
+      "loss": 1.2932,
+      "step": 6200
+    },
+    {
+      "epoch": 5.225073622212873,
+      "grad_norm": 1.0655443668365479,
+      "learning_rate": 0.0002,
+      "loss": 1.1825,
+      "step": 6210
+    },
+    {
+      "epoch": 5.233487589398401,
+      "grad_norm": 1.0854865312576294,
+      "learning_rate": 0.0002,
+      "loss": 1.2613,
+      "step": 6220
+    },
+    {
+      "epoch": 5.2419015565839295,
+      "grad_norm": 0.8489186763763428,
+      "learning_rate": 0.0002,
+      "loss": 1.3045,
+      "step": 6230
+    },
+    {
+      "epoch": 5.250315523769458,
+      "grad_norm": 0.910391628742218,
+      "learning_rate": 0.0002,
+      "loss": 1.2708,
+      "step": 6240
+    },
+    {
+      "epoch": 5.258729490954985,
+      "grad_norm": 0.925507128238678,
+      "learning_rate": 0.0002,
+      "loss": 1.1914,
+      "step": 6250
+    },
+    {
+      "epoch": 5.267143458140513,
+      "grad_norm": 1.1069735288619995,
+      "learning_rate": 0.0002,
+      "loss": 1.3368,
+      "step": 6260
+    },
+    {
+      "epoch": 5.275557425326041,
+      "grad_norm": 0.9705119132995605,
+      "learning_rate": 0.0002,
+      "loss": 1.2505,
+      "step": 6270
+    },
+    {
+      "epoch": 5.283971392511569,
+      "grad_norm": 0.9752426147460938,
+      "learning_rate": 0.0002,
+      "loss": 1.2602,
+      "step": 6280
+    },
+    {
+      "epoch": 5.292385359697097,
+      "grad_norm": 1.021359920501709,
+      "learning_rate": 0.0002,
+      "loss": 1.2043,
+      "step": 6290
+    },
+    {
+      "epoch": 5.3007993268826255,
+      "grad_norm": 1.148606300354004,
+      "learning_rate": 0.0002,
+      "loss": 1.2848,
+      "step": 6300
+    },
+    {
+      "epoch": 5.309213294068153,
+      "grad_norm": 0.8909247517585754,
+      "learning_rate": 0.0002,
+      "loss": 1.2201,
+      "step": 6310
+    },
+    {
+      "epoch": 5.317627261253681,
+      "grad_norm": 0.9879156351089478,
+      "learning_rate": 0.0002,
+      "loss": 1.2376,
+      "step": 6320
+    },
+    {
+      "epoch": 5.326041228439209,
+      "grad_norm": 0.9473357200622559,
+      "learning_rate": 0.0002,
+      "loss": 1.2638,
+      "step": 6330
+    },
+    {
+      "epoch": 5.334455195624737,
+      "grad_norm": 1.1422028541564941,
+      "learning_rate": 0.0002,
+      "loss": 1.232,
+      "step": 6340
+    },
+    {
+      "epoch": 5.342869162810265,
+      "grad_norm": 0.9942235350608826,
+      "learning_rate": 0.0002,
+      "loss": 1.263,
+      "step": 6350
+    },
+    {
+      "epoch": 5.351283129995793,
+      "grad_norm": 0.9535723924636841,
+      "learning_rate": 0.0002,
+      "loss": 1.3032,
+      "step": 6360
+    },
+    {
+      "epoch": 5.359697097181321,
+      "grad_norm": 0.9020892381668091,
+      "learning_rate": 0.0002,
+      "loss": 1.2908,
+      "step": 6370
+    },
+    {
+      "epoch": 5.368111064366849,
+      "grad_norm": 1.0626472234725952,
+      "learning_rate": 0.0002,
+      "loss": 1.2023,
+      "step": 6380
+    },
+    {
+      "epoch": 5.376525031552377,
+      "grad_norm": 1.1395848989486694,
+      "learning_rate": 0.0002,
+      "loss": 1.2555,
+      "step": 6390
+    },
+    {
+      "epoch": 5.384938998737905,
+      "grad_norm": 0.9274451732635498,
+      "learning_rate": 0.0002,
+      "loss": 1.2839,
+      "step": 6400
+    },
+    {
+      "epoch": 5.393352965923433,
+      "grad_norm": 0.8108699917793274,
+      "learning_rate": 0.0002,
+      "loss": 1.2819,
+      "step": 6410
+    },
+    {
+      "epoch": 5.401766933108961,
+      "grad_norm": 1.1805564165115356,
+      "learning_rate": 0.0002,
+      "loss": 1.2589,
+      "step": 6420
+    },
+    {
+      "epoch": 5.410180900294489,
+      "grad_norm": 0.8321298360824585,
+      "learning_rate": 0.0002,
+      "loss": 1.3549,
+      "step": 6430
+    },
+    {
+      "epoch": 5.418594867480017,
+      "grad_norm": 0.8981925249099731,
+      "learning_rate": 0.0002,
+      "loss": 1.2925,
+      "step": 6440
+    },
+    {
+      "epoch": 5.4270088346655445,
+      "grad_norm": 1.0730986595153809,
+      "learning_rate": 0.0002,
+      "loss": 1.258,
+      "step": 6450
+    },
+    {
+      "epoch": 5.435422801851073,
+      "grad_norm": 1.0584609508514404,
+      "learning_rate": 0.0002,
+      "loss": 1.26,
+      "step": 6460
+    },
+    {
+      "epoch": 5.443836769036601,
+      "grad_norm": 1.0792299509048462,
+      "learning_rate": 0.0002,
+      "loss": 1.2847,
+      "step": 6470
+    },
+    {
+      "epoch": 5.452250736222129,
+      "grad_norm": 0.9101872444152832,
+      "learning_rate": 0.0002,
+      "loss": 1.2035,
+      "step": 6480
+    },
+    {
+      "epoch": 5.460664703407657,
+      "grad_norm": 0.9910100698471069,
+      "learning_rate": 0.0002,
+      "loss": 1.2574,
+      "step": 6490
+    },
+    {
+      "epoch": 5.469078670593185,
+      "grad_norm": 1.041412353515625,
+      "learning_rate": 0.0002,
+      "loss": 1.3098,
+      "step": 6500
+    },
+    {
+      "epoch": 5.477492637778712,
+      "grad_norm": 1.0091687440872192,
+      "learning_rate": 0.0002,
+      "loss": 1.2812,
+      "step": 6510
+    },
+    {
+      "epoch": 5.48590660496424,
+      "grad_norm": 0.8755383491516113,
+      "learning_rate": 0.0002,
+      "loss": 1.2523,
+      "step": 6520
+    },
+    {
+      "epoch": 5.4943205721497685,
+      "grad_norm": 0.980212390422821,
+      "learning_rate": 0.0002,
+      "loss": 1.3042,
+      "step": 6530
+    },
+    {
+      "epoch": 5.502734539335297,
+      "grad_norm": 0.9356869459152222,
+      "learning_rate": 0.0002,
+      "loss": 1.2873,
+      "step": 6540
+    },
+    {
+      "epoch": 5.511148506520825,
+      "grad_norm": 0.9008095264434814,
+      "learning_rate": 0.0002,
+      "loss": 1.2254,
+      "step": 6550
+    },
+    {
+      "epoch": 5.519562473706353,
+      "grad_norm": 0.8908938765525818,
+      "learning_rate": 0.0002,
+      "loss": 1.2818,
+      "step": 6560
+    },
+    {
+      "epoch": 5.52797644089188,
+      "grad_norm": 1.1423932313919067,
+      "learning_rate": 0.0002,
+      "loss": 1.2212,
+      "step": 6570
+    },
+    {
+      "epoch": 5.536390408077408,
+      "grad_norm": 1.0508161783218384,
+      "learning_rate": 0.0002,
+      "loss": 1.3039,
+      "step": 6580
+    },
+    {
+      "epoch": 5.544804375262936,
+      "grad_norm": 0.8357517719268799,
+      "learning_rate": 0.0002,
+      "loss": 1.2446,
+      "step": 6590
+    },
+    {
+      "epoch": 5.5532183424484645,
+      "grad_norm": 0.9892540574073792,
+      "learning_rate": 0.0002,
+      "loss": 1.3037,
+      "step": 6600
+    },
+    {
+      "epoch": 5.561632309633993,
+      "grad_norm": 1.0048326253890991,
+      "learning_rate": 0.0002,
+      "loss": 1.3028,
+      "step": 6610
+    },
+    {
+      "epoch": 5.570046276819521,
+      "grad_norm": 0.9801995158195496,
+      "learning_rate": 0.0002,
+      "loss": 1.2152,
+      "step": 6620
+    },
+    {
+      "epoch": 5.578460244005049,
+      "grad_norm": 0.9899214506149292,
+      "learning_rate": 0.0002,
+      "loss": 1.2606,
+      "step": 6630
+    },
+    {
+      "epoch": 5.586874211190576,
+      "grad_norm": 1.1911814212799072,
+      "learning_rate": 0.0002,
+      "loss": 1.2043,
+      "step": 6640
+    },
+    {
+      "epoch": 5.595288178376104,
+      "grad_norm": 1.0368894338607788,
+      "learning_rate": 0.0002,
+      "loss": 1.3458,
+      "step": 6650
+    },
+    {
+      "epoch": 5.603702145561632,
+      "grad_norm": 1.1248382329940796,
+      "learning_rate": 0.0002,
+      "loss": 1.2595,
+      "step": 6660
+    },
+    {
+      "epoch": 5.61211611274716,
+      "grad_norm": 0.9765539765357971,
+      "learning_rate": 0.0002,
+      "loss": 1.2548,
+      "step": 6670
+    },
+    {
+      "epoch": 5.6205300799326885,
+      "grad_norm": 0.9810206890106201,
+      "learning_rate": 0.0002,
+      "loss": 1.3451,
+      "step": 6680
+    },
+    {
+      "epoch": 5.628944047118217,
+      "grad_norm": 1.100386619567871,
+      "learning_rate": 0.0002,
+      "loss": 1.2952,
+      "step": 6690
+    },
+    {
+      "epoch": 5.637358014303744,
+      "grad_norm": 0.8824519515037537,
+      "learning_rate": 0.0002,
+      "loss": 1.2467,
+      "step": 6700
+    },
+    {
+      "epoch": 5.645771981489272,
+      "grad_norm": 1.0864064693450928,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 6710
+    },
+    {
+      "epoch": 5.6541859486748,
+      "grad_norm": 1.1614511013031006,
+      "learning_rate": 0.0002,
+      "loss": 1.2479,
+      "step": 6720
+    },
+    {
+      "epoch": 5.662599915860328,
+      "grad_norm": 1.0762972831726074,
+      "learning_rate": 0.0002,
+      "loss": 1.2753,
+      "step": 6730
+    },
+    {
+      "epoch": 5.671013883045856,
+      "grad_norm": 0.9408974647521973,
+      "learning_rate": 0.0002,
+      "loss": 1.2741,
+      "step": 6740
+    },
+    {
+      "epoch": 5.679427850231384,
+      "grad_norm": 0.8906030058860779,
+      "learning_rate": 0.0002,
+      "loss": 1.2431,
+      "step": 6750
+    },
+    {
+      "epoch": 5.687841817416912,
+      "grad_norm": 0.9527303576469421,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 6760
+    },
+    {
+      "epoch": 5.69625578460244,
+      "grad_norm": 0.9471196532249451,
+      "learning_rate": 0.0002,
+      "loss": 1.322,
+      "step": 6770
+    },
+    {
+      "epoch": 5.704669751787968,
+      "grad_norm": 0.9186838865280151,
+      "learning_rate": 0.0002,
+      "loss": 1.2514,
+      "step": 6780
+    },
+    {
+      "epoch": 5.713083718973496,
+      "grad_norm": 0.9225441813468933,
+      "learning_rate": 0.0002,
+      "loss": 1.2347,
+      "step": 6790
+    },
+    {
+      "epoch": 5.721497686159024,
+      "grad_norm": 0.9712982773780823,
+      "learning_rate": 0.0002,
+      "loss": 1.1849,
+      "step": 6800
+    },
+    {
+      "epoch": 5.729911653344552,
+      "grad_norm": 1.0743170976638794,
+      "learning_rate": 0.0002,
+      "loss": 1.2431,
+      "step": 6810
+    },
+    {
+      "epoch": 5.73832562053008,
+      "grad_norm": 1.2738113403320312,
+      "learning_rate": 0.0002,
+      "loss": 1.2136,
+      "step": 6820
+    },
+    {
+      "epoch": 5.7467395877156076,
+      "grad_norm": 0.9386790990829468,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 6830
+    },
+    {
+      "epoch": 5.755153554901136,
+      "grad_norm": 1.0817769765853882,
+      "learning_rate": 0.0002,
+      "loss": 1.285,
+      "step": 6840
+    },
+    {
+      "epoch": 5.763567522086664,
+      "grad_norm": 1.1040263175964355,
+      "learning_rate": 0.0002,
+      "loss": 1.2247,
+      "step": 6850
+    },
+    {
+      "epoch": 5.771981489272192,
+      "grad_norm": 1.0656492710113525,
+      "learning_rate": 0.0002,
+      "loss": 1.2507,
+      "step": 6860
+    },
+    {
+      "epoch": 5.78039545645772,
+      "grad_norm": 0.9550157189369202,
+      "learning_rate": 0.0002,
+      "loss": 1.2999,
+      "step": 6870
+    },
+    {
+      "epoch": 5.788809423643248,
+      "grad_norm": 1.0130870342254639,
+      "learning_rate": 0.0002,
+      "loss": 1.3201,
+      "step": 6880
+    },
+    {
+      "epoch": 5.797223390828776,
+      "grad_norm": 1.0675787925720215,
+      "learning_rate": 0.0002,
+      "loss": 1.3392,
+      "step": 6890
+    },
+    {
+      "epoch": 5.8056373580143035,
+      "grad_norm": 0.9537774920463562,
+      "learning_rate": 0.0002,
+      "loss": 1.2949,
+      "step": 6900
+    },
+    {
+      "epoch": 5.814051325199832,
+      "grad_norm": 0.9640319347381592,
+      "learning_rate": 0.0002,
+      "loss": 1.2658,
+      "step": 6910
+    },
+    {
+      "epoch": 5.82246529238536,
+      "grad_norm": 0.8917992115020752,
+      "learning_rate": 0.0002,
+      "loss": 1.2199,
+      "step": 6920
+    },
+    {
+      "epoch": 5.830879259570888,
+      "grad_norm": 0.9881822466850281,
+      "learning_rate": 0.0002,
+      "loss": 1.373,
+      "step": 6930
+    },
+    {
+      "epoch": 5.839293226756416,
+      "grad_norm": 0.9136882424354553,
+      "learning_rate": 0.0002,
+      "loss": 1.323,
+      "step": 6940
+    },
+    {
+      "epoch": 5.847707193941943,
+      "grad_norm": 0.9086098074913025,
+      "learning_rate": 0.0002,
+      "loss": 1.3159,
+      "step": 6950
+    },
+    {
+      "epoch": 5.856121161127471,
+      "grad_norm": 0.9443018436431885,
+      "learning_rate": 0.0002,
+      "loss": 1.2624,
+      "step": 6960
+    },
+    {
+      "epoch": 5.864535128312999,
+      "grad_norm": 0.9915381669998169,
+      "learning_rate": 0.0002,
+      "loss": 1.3224,
+      "step": 6970
+    },
+    {
+      "epoch": 5.8729490954985275,
+      "grad_norm": 0.8939146995544434,
+      "learning_rate": 0.0002,
+      "loss": 1.337,
+      "step": 6980
+    },
+    {
+      "epoch": 5.881363062684056,
+      "grad_norm": 1.3672245740890503,
+      "learning_rate": 0.0002,
+      "loss": 1.2611,
+      "step": 6990
+    },
+    {
+      "epoch": 5.889777029869584,
+      "grad_norm": 1.0116257667541504,
+      "learning_rate": 0.0002,
+      "loss": 1.3012,
+      "step": 7000
+    },
+    {
+      "epoch": 5.898190997055112,
+      "grad_norm": 1.1561565399169922,
+      "learning_rate": 0.0002,
+      "loss": 1.3128,
+      "step": 7010
+    },
+    {
+      "epoch": 5.906604964240639,
+      "grad_norm": 0.9900678992271423,
+      "learning_rate": 0.0002,
+      "loss": 1.2301,
+      "step": 7020
+    },
+    {
+      "epoch": 5.915018931426167,
+      "grad_norm": 0.9297345876693726,
+      "learning_rate": 0.0002,
+      "loss": 1.2845,
+      "step": 7030
+    },
+    {
+      "epoch": 5.923432898611695,
+      "grad_norm": 0.9357825517654419,
+      "learning_rate": 0.0002,
+      "loss": 1.2317,
+      "step": 7040
+    },
+    {
+      "epoch": 5.931846865797223,
+      "grad_norm": 1.049317717552185,
+      "learning_rate": 0.0002,
+      "loss": 1.2303,
+      "step": 7050
+    },
+    {
+      "epoch": 5.9402608329827515,
+      "grad_norm": 0.950633704662323,
+      "learning_rate": 0.0002,
+      "loss": 1.3243,
+      "step": 7060
+    },
+    {
+      "epoch": 5.94867480016828,
+      "grad_norm": 0.854581892490387,
+      "learning_rate": 0.0002,
+      "loss": 1.2758,
+      "step": 7070
+    },
+    {
+      "epoch": 5.957088767353808,
+      "grad_norm": 0.9097039699554443,
+      "learning_rate": 0.0002,
+      "loss": 1.3252,
+      "step": 7080
+    },
+    {
+      "epoch": 5.965502734539335,
+      "grad_norm": 0.9072173237800598,
+      "learning_rate": 0.0002,
+      "loss": 1.291,
+      "step": 7090
+    },
+    {
+      "epoch": 5.973916701724863,
+      "grad_norm": 1.0470727682113647,
+      "learning_rate": 0.0002,
+      "loss": 1.2724,
+      "step": 7100
+    },
+    {
+      "epoch": 5.982330668910391,
+      "grad_norm": 1.2628462314605713,
+      "learning_rate": 0.0002,
+      "loss": 1.3324,
+      "step": 7110
+    },
+    {
+      "epoch": 5.990744636095919,
+      "grad_norm": 1.055279016494751,
+      "learning_rate": 0.0002,
+      "loss": 1.2701,
+      "step": 7120
+    },
+    {
+      "epoch": 5.9991586032814475,
+      "grad_norm": 0.966194212436676,
+      "learning_rate": 0.0002,
+      "loss": 1.3234,
+      "step": 7130
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.0427448749542236,
+      "eval_runtime": 37.8426,
+      "eval_samples_per_second": 13.609,
+      "eval_steps_per_second": 1.718,
+      "step": 7131
+    },
+    {
+      "epoch": 6.007572570466976,
+      "grad_norm": 1.4037928581237793,
+      "learning_rate": 0.0002,
+      "loss": 1.1308,
+      "step": 7140
+    },
+    {
+      "epoch": 6.015986537652503,
+      "grad_norm": 1.1081010103225708,
+      "learning_rate": 0.0002,
+      "loss": 1.047,
+      "step": 7150
+    },
+    {
+      "epoch": 6.024400504838031,
+      "grad_norm": 1.1585499048233032,
+      "learning_rate": 0.0002,
+      "loss": 1.1368,
+      "step": 7160
+    },
+    {
+      "epoch": 6.032814472023559,
+      "grad_norm": 1.0822780132293701,
+      "learning_rate": 0.0002,
+      "loss": 1.0192,
+      "step": 7170
+    },
+    {
+      "epoch": 6.041228439209087,
+      "grad_norm": 0.9662094712257385,
+      "learning_rate": 0.0002,
+      "loss": 1.0755,
+      "step": 7180
+    },
+    {
+      "epoch": 6.049642406394615,
+      "grad_norm": 1.063936710357666,
+      "learning_rate": 0.0002,
+      "loss": 1.1366,
+      "step": 7190
+    },
+    {
+      "epoch": 6.058056373580143,
+      "grad_norm": 1.0349032878875732,
+      "learning_rate": 0.0002,
+      "loss": 1.0121,
+      "step": 7200
+    },
+    {
+      "epoch": 6.066470340765671,
+      "grad_norm": 1.0312575101852417,
+      "learning_rate": 0.0002,
+      "loss": 1.0591,
+      "step": 7210
+    },
+    {
+      "epoch": 6.074884307951199,
+      "grad_norm": 1.1942846775054932,
+      "learning_rate": 0.0002,
+      "loss": 1.1824,
+      "step": 7220
+    },
+    {
+      "epoch": 6.083298275136727,
+      "grad_norm": 1.0816049575805664,
+      "learning_rate": 0.0002,
+      "loss": 1.1034,
+      "step": 7230
+    },
+    {
+      "epoch": 6.091712242322255,
+      "grad_norm": 0.9985513687133789,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 7240
+    },
+    {
+      "epoch": 6.100126209507783,
+      "grad_norm": 1.2573972940444946,
+      "learning_rate": 0.0002,
+      "loss": 1.0367,
+      "step": 7250
+    },
+    {
+      "epoch": 6.108540176693311,
+      "grad_norm": 1.1182395219802856,
+      "learning_rate": 0.0002,
+      "loss": 1.1051,
+      "step": 7260
+    },
+    {
+      "epoch": 6.116954143878839,
+      "grad_norm": 0.9679344296455383,
+      "learning_rate": 0.0002,
+      "loss": 1.1219,
+      "step": 7270
+    },
+    {
+      "epoch": 6.1253681110643665,
+      "grad_norm": 1.0913981199264526,
+      "learning_rate": 0.0002,
+      "loss": 1.1192,
+      "step": 7280
+    },
+    {
+      "epoch": 6.133782078249895,
+      "grad_norm": 1.1291013956069946,
+      "learning_rate": 0.0002,
+      "loss": 1.0411,
+      "step": 7290
+    },
+    {
+      "epoch": 6.142196045435423,
+      "grad_norm": 1.2679595947265625,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 7300
+    },
+    {
+      "epoch": 6.150610012620951,
+      "grad_norm": 1.2350026369094849,
+      "learning_rate": 0.0002,
+      "loss": 1.0875,
+      "step": 7310
+    },
+    {
+      "epoch": 6.159023979806479,
+      "grad_norm": 1.3213104009628296,
+      "learning_rate": 0.0002,
+      "loss": 1.1139,
+      "step": 7320
+    },
+    {
+      "epoch": 6.167437946992007,
+      "grad_norm": 1.1924850940704346,
+      "learning_rate": 0.0002,
+      "loss": 1.1167,
+      "step": 7330
+    },
+    {
+      "epoch": 6.175851914177534,
+      "grad_norm": 1.1890000104904175,
+      "learning_rate": 0.0002,
+      "loss": 1.1242,
+      "step": 7340
+    },
+    {
+      "epoch": 6.184265881363062,
+      "grad_norm": 1.3821455240249634,
+      "learning_rate": 0.0002,
+      "loss": 1.1341,
+      "step": 7350
+    },
+    {
+      "epoch": 6.1926798485485905,
+      "grad_norm": 1.1217057704925537,
+      "learning_rate": 0.0002,
+      "loss": 1.0748,
+      "step": 7360
+    },
+    {
+      "epoch": 6.201093815734119,
+      "grad_norm": 1.2441548109054565,
+      "learning_rate": 0.0002,
+      "loss": 1.159,
+      "step": 7370
+    },
+    {
+      "epoch": 6.209507782919647,
+      "grad_norm": 1.0837615728378296,
+      "learning_rate": 0.0002,
+      "loss": 1.1199,
+      "step": 7380
+    },
+    {
+      "epoch": 6.217921750105175,
+      "grad_norm": 1.164304256439209,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 7390
+    },
+    {
+      "epoch": 6.226335717290702,
+      "grad_norm": 1.3129467964172363,
+      "learning_rate": 0.0002,
+      "loss": 1.1325,
+      "step": 7400
+    },
+    {
+      "epoch": 6.23474968447623,
+      "grad_norm": 1.1938153505325317,
+      "learning_rate": 0.0002,
+      "loss": 1.1537,
+      "step": 7410
+    },
+    {
+      "epoch": 6.243163651661758,
+      "grad_norm": 1.4348443746566772,
+      "learning_rate": 0.0002,
+      "loss": 1.1238,
+      "step": 7420
+    },
+    {
+      "epoch": 6.2515776188472865,
+      "grad_norm": 1.132301926612854,
+      "learning_rate": 0.0002,
+      "loss": 1.0778,
+      "step": 7430
+    },
+    {
+      "epoch": 6.259991586032815,
+      "grad_norm": 1.136966586112976,
+      "learning_rate": 0.0002,
+      "loss": 1.1148,
+      "step": 7440
+    },
+    {
+      "epoch": 6.268405553218343,
+      "grad_norm": 1.12801194190979,
+      "learning_rate": 0.0002,
+      "loss": 1.096,
+      "step": 7450
+    },
+    {
+      "epoch": 6.276819520403871,
+      "grad_norm": 1.0246902704238892,
+      "learning_rate": 0.0002,
+      "loss": 1.0408,
+      "step": 7460
+    },
+    {
+      "epoch": 6.285233487589398,
+      "grad_norm": 1.1066974401474,
+      "learning_rate": 0.0002,
+      "loss": 1.0389,
+      "step": 7470
+    },
+    {
+      "epoch": 6.293647454774926,
+      "grad_norm": 1.012710690498352,
+      "learning_rate": 0.0002,
+      "loss": 1.1589,
+      "step": 7480
+    },
+    {
+      "epoch": 6.302061421960454,
+      "grad_norm": 1.2227119207382202,
+      "learning_rate": 0.0002,
+      "loss": 1.1049,
+      "step": 7490
+    },
+    {
+      "epoch": 6.310475389145982,
+      "grad_norm": 0.9736923575401306,
+      "learning_rate": 0.0002,
+      "loss": 1.1376,
+      "step": 7500
+    },
+    {
+      "epoch": 6.3188893563315105,
+      "grad_norm": 1.2945268154144287,
+      "learning_rate": 0.0002,
+      "loss": 1.1017,
+      "step": 7510
+    },
+    {
+      "epoch": 6.327303323517039,
+      "grad_norm": 1.1579312086105347,
+      "learning_rate": 0.0002,
+      "loss": 1.0724,
+      "step": 7520
+    },
+    {
+      "epoch": 6.335717290702567,
+      "grad_norm": 1.2404558658599854,
+      "learning_rate": 0.0002,
+      "loss": 1.0899,
+      "step": 7530
+    },
+    {
+      "epoch": 6.344131257888094,
+      "grad_norm": 1.4673258066177368,
+      "learning_rate": 0.0002,
+      "loss": 1.1635,
+      "step": 7540
+    },
+    {
+      "epoch": 6.352545225073622,
+      "grad_norm": 1.2268997430801392,
+      "learning_rate": 0.0002,
+      "loss": 1.128,
+      "step": 7550
+    },
+    {
+      "epoch": 6.36095919225915,
+      "grad_norm": 0.9772747159004211,
+      "learning_rate": 0.0002,
+      "loss": 1.0932,
+      "step": 7560
+    },
+    {
+      "epoch": 6.369373159444678,
+      "grad_norm": 1.0205204486846924,
+      "learning_rate": 0.0002,
+      "loss": 1.1214,
+      "step": 7570
+    },
+    {
+      "epoch": 6.377787126630206,
+      "grad_norm": 1.2227109670639038,
+      "learning_rate": 0.0002,
+      "loss": 1.1095,
+      "step": 7580
+    },
+    {
+      "epoch": 6.3862010938157345,
+      "grad_norm": 1.0708507299423218,
+      "learning_rate": 0.0002,
+      "loss": 1.1115,
+      "step": 7590
+    },
+    {
+      "epoch": 6.394615061001262,
+      "grad_norm": 1.1427522897720337,
+      "learning_rate": 0.0002,
+      "loss": 1.1018,
+      "step": 7600
+    },
+    {
+      "epoch": 6.40302902818679,
+      "grad_norm": 1.0706431865692139,
+      "learning_rate": 0.0002,
+      "loss": 1.1079,
+      "step": 7610
+    },
+    {
+      "epoch": 6.411442995372318,
+      "grad_norm": 1.1358282566070557,
+      "learning_rate": 0.0002,
+      "loss": 1.0933,
+      "step": 7620
+    },
+    {
+      "epoch": 6.419856962557846,
+      "grad_norm": 1.4011822938919067,
+      "learning_rate": 0.0002,
+      "loss": 1.1075,
+      "step": 7630
+    },
+    {
+      "epoch": 6.428270929743374,
+      "grad_norm": 1.5616450309753418,
+      "learning_rate": 0.0002,
+      "loss": 1.1269,
+      "step": 7640
+    },
+    {
+      "epoch": 6.436684896928902,
+      "grad_norm": 1.1442687511444092,
+      "learning_rate": 0.0002,
+      "loss": 1.0953,
+      "step": 7650
+    },
+    {
+      "epoch": 6.44509886411443,
+      "grad_norm": 1.164803147315979,
+      "learning_rate": 0.0002,
+      "loss": 1.1341,
+      "step": 7660
+    },
+    {
+      "epoch": 6.453512831299958,
+      "grad_norm": 1.3184553384780884,
+      "learning_rate": 0.0002,
+      "loss": 1.14,
+      "step": 7670
+    },
+    {
+      "epoch": 6.461926798485486,
+      "grad_norm": 1.2701894044876099,
+      "learning_rate": 0.0002,
+      "loss": 1.1526,
+      "step": 7680
+    },
+    {
+      "epoch": 6.470340765671014,
+      "grad_norm": 1.1998416185379028,
+      "learning_rate": 0.0002,
+      "loss": 1.2119,
+      "step": 7690
+    },
+    {
+      "epoch": 6.478754732856542,
+      "grad_norm": 1.156459927558899,
+      "learning_rate": 0.0002,
+      "loss": 1.1528,
+      "step": 7700
+    },
+    {
+      "epoch": 6.48716870004207,
+      "grad_norm": 1.0217190980911255,
+      "learning_rate": 0.0002,
+      "loss": 1.2122,
+      "step": 7710
+    },
+    {
+      "epoch": 6.495582667227598,
+      "grad_norm": 1.230372428894043,
+      "learning_rate": 0.0002,
+      "loss": 1.0917,
+      "step": 7720
+    },
+    {
+      "epoch": 6.5039966344131255,
+      "grad_norm": 1.105675220489502,
+      "learning_rate": 0.0002,
+      "loss": 1.119,
+      "step": 7730
+    },
+    {
+      "epoch": 6.512410601598654,
+      "grad_norm": 1.1623669862747192,
+      "learning_rate": 0.0002,
+      "loss": 1.0758,
+      "step": 7740
+    },
+    {
+      "epoch": 6.520824568784182,
+      "grad_norm": 1.2884684801101685,
+      "learning_rate": 0.0002,
+      "loss": 1.1548,
+      "step": 7750
+    },
+    {
+      "epoch": 6.52923853596971,
+      "grad_norm": 1.1785279512405396,
+      "learning_rate": 0.0002,
+      "loss": 1.142,
+      "step": 7760
+    },
+    {
+      "epoch": 6.537652503155238,
+      "grad_norm": 1.0607101917266846,
+      "learning_rate": 0.0002,
+      "loss": 1.1598,
+      "step": 7770
+    },
+    {
+      "epoch": 6.546066470340766,
+      "grad_norm": 1.21990168094635,
+      "learning_rate": 0.0002,
+      "loss": 1.1472,
+      "step": 7780
+    },
+    {
+      "epoch": 6.554480437526293,
+      "grad_norm": 1.1498621702194214,
+      "learning_rate": 0.0002,
+      "loss": 1.1468,
+      "step": 7790
+    },
+    {
+      "epoch": 6.562894404711821,
+      "grad_norm": 1.263929009437561,
+      "learning_rate": 0.0002,
+      "loss": 1.1847,
+      "step": 7800
+    },
+    {
+      "epoch": 6.5713083718973495,
+      "grad_norm": 1.1580625772476196,
+      "learning_rate": 0.0002,
+      "loss": 1.1177,
+      "step": 7810
+    },
+    {
+      "epoch": 6.579722339082878,
+      "grad_norm": 1.4431294202804565,
+      "learning_rate": 0.0002,
+      "loss": 1.1313,
+      "step": 7820
+    },
+    {
+      "epoch": 6.588136306268406,
+      "grad_norm": 1.1309990882873535,
+      "learning_rate": 0.0002,
+      "loss": 1.1944,
+      "step": 7830
+    },
+    {
+      "epoch": 6.596550273453934,
+      "grad_norm": 1.0543386936187744,
+      "learning_rate": 0.0002,
+      "loss": 1.1156,
+      "step": 7840
+    },
+    {
+      "epoch": 6.604964240639461,
+      "grad_norm": 1.2180639505386353,
+      "learning_rate": 0.0002,
+      "loss": 1.0945,
+      "step": 7850
+    },
+    {
+      "epoch": 6.613378207824989,
+      "grad_norm": 1.0631271600723267,
+      "learning_rate": 0.0002,
+      "loss": 1.1318,
+      "step": 7860
+    },
+    {
+      "epoch": 6.621792175010517,
+      "grad_norm": 1.138885498046875,
+      "learning_rate": 0.0002,
+      "loss": 1.1792,
+      "step": 7870
+    },
+    {
+      "epoch": 6.630206142196045,
+      "grad_norm": 1.1117745637893677,
+      "learning_rate": 0.0002,
+      "loss": 1.1805,
+      "step": 7880
+    },
+    {
+      "epoch": 6.6386201093815735,
+      "grad_norm": 1.3734886646270752,
+      "learning_rate": 0.0002,
+      "loss": 1.15,
+      "step": 7890
+    },
+    {
+      "epoch": 6.647034076567102,
+      "grad_norm": 1.236003041267395,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 7900
+    },
+    {
+      "epoch": 6.65544804375263,
+      "grad_norm": 1.2206000089645386,
+      "learning_rate": 0.0002,
+      "loss": 1.1718,
+      "step": 7910
+    },
+    {
+      "epoch": 6.663862010938157,
+      "grad_norm": 1.2842656373977661,
+      "learning_rate": 0.0002,
+      "loss": 1.1637,
+      "step": 7920
+    },
+    {
+      "epoch": 6.672275978123685,
+      "grad_norm": 1.2365005016326904,
+      "learning_rate": 0.0002,
+      "loss": 1.2219,
+      "step": 7930
+    },
+    {
+      "epoch": 6.680689945309213,
+      "grad_norm": 1.256620168685913,
+      "learning_rate": 0.0002,
+      "loss": 1.0827,
+      "step": 7940
+    },
+    {
+      "epoch": 6.689103912494741,
+      "grad_norm": 1.3232917785644531,
+      "learning_rate": 0.0002,
+      "loss": 1.1788,
+      "step": 7950
+    },
+    {
+      "epoch": 6.6975178796802695,
+      "grad_norm": 1.2470088005065918,
+      "learning_rate": 0.0002,
+      "loss": 1.2042,
+      "step": 7960
+    },
+    {
+      "epoch": 6.705931846865798,
+      "grad_norm": 1.0511926412582397,
+      "learning_rate": 0.0002,
+      "loss": 1.0959,
+      "step": 7970
+    },
+    {
+      "epoch": 6.714345814051326,
+      "grad_norm": 1.107310175895691,
+      "learning_rate": 0.0002,
+      "loss": 1.118,
+      "step": 7980
+    },
+    {
+      "epoch": 6.722759781236853,
+      "grad_norm": 1.4069843292236328,
+      "learning_rate": 0.0002,
+      "loss": 1.2109,
+      "step": 7990
+    },
+    {
+      "epoch": 6.731173748422381,
+      "grad_norm": 1.0800836086273193,
+      "learning_rate": 0.0002,
+      "loss": 1.1298,
+      "step": 8000
+    },
+    {
+      "epoch": 6.739587715607909,
+      "grad_norm": 1.1676300764083862,
+      "learning_rate": 0.0002,
+      "loss": 1.1824,
+      "step": 8010
+    },
+    {
+      "epoch": 6.748001682793437,
+      "grad_norm": 1.0579663515090942,
+      "learning_rate": 0.0002,
+      "loss": 1.1253,
+      "step": 8020
+    },
+    {
+      "epoch": 6.756415649978965,
+      "grad_norm": 1.2770029306411743,
+      "learning_rate": 0.0002,
+      "loss": 1.1542,
+      "step": 8030
+    },
+    {
+      "epoch": 6.764829617164493,
+      "grad_norm": 1.0981038808822632,
+      "learning_rate": 0.0002,
+      "loss": 1.1519,
+      "step": 8040
+    },
+    {
+      "epoch": 6.773243584350021,
+      "grad_norm": 1.1194742918014526,
+      "learning_rate": 0.0002,
+      "loss": 1.1422,
+      "step": 8050
+    },
+    {
+      "epoch": 6.781657551535549,
+      "grad_norm": 1.0130012035369873,
+      "learning_rate": 0.0002,
+      "loss": 1.1463,
+      "step": 8060
+    },
+    {
+      "epoch": 6.790071518721077,
+      "grad_norm": 1.2051167488098145,
+      "learning_rate": 0.0002,
+      "loss": 1.2008,
+      "step": 8070
+    },
+    {
+      "epoch": 6.798485485906605,
+      "grad_norm": 1.095689058303833,
+      "learning_rate": 0.0002,
+      "loss": 1.142,
+      "step": 8080
+    },
+    {
+      "epoch": 6.806899453092133,
+      "grad_norm": 1.2275174856185913,
+      "learning_rate": 0.0002,
+      "loss": 1.1352,
+      "step": 8090
+    },
+    {
+      "epoch": 6.815313420277661,
+      "grad_norm": 1.1439805030822754,
+      "learning_rate": 0.0002,
+      "loss": 1.1453,
+      "step": 8100
+    },
+    {
+      "epoch": 6.8237273874631885,
+      "grad_norm": 1.276331901550293,
+      "learning_rate": 0.0002,
+      "loss": 1.1624,
+      "step": 8110
+    },
+    {
+      "epoch": 6.832141354648717,
+      "grad_norm": 1.0450139045715332,
+      "learning_rate": 0.0002,
+      "loss": 1.1686,
+      "step": 8120
+    },
+    {
+      "epoch": 6.840555321834245,
+      "grad_norm": 1.1189453601837158,
+      "learning_rate": 0.0002,
+      "loss": 1.1783,
+      "step": 8130
+    },
+    {
+      "epoch": 6.848969289019773,
+      "grad_norm": 1.194640874862671,
+      "learning_rate": 0.0002,
+      "loss": 1.1093,
+      "step": 8140
+    },
+    {
+      "epoch": 6.857383256205301,
+      "grad_norm": 1.095372200012207,
+      "learning_rate": 0.0002,
+      "loss": 1.1559,
+      "step": 8150
+    },
+    {
+      "epoch": 6.865797223390829,
+      "grad_norm": 1.2416104078292847,
+      "learning_rate": 0.0002,
+      "loss": 1.165,
+      "step": 8160
+    },
+    {
+      "epoch": 6.874211190576357,
+      "grad_norm": 1.2402868270874023,
+      "learning_rate": 0.0002,
+      "loss": 1.2174,
+      "step": 8170
+    },
+    {
+      "epoch": 6.882625157761884,
+      "grad_norm": 1.1317291259765625,
+      "learning_rate": 0.0002,
+      "loss": 1.1306,
+      "step": 8180
+    },
+    {
+      "epoch": 6.8910391249474126,
+      "grad_norm": 1.0581914186477661,
+      "learning_rate": 0.0002,
+      "loss": 1.1944,
+      "step": 8190
+    },
+    {
+      "epoch": 6.899453092132941,
+      "grad_norm": 1.3540890216827393,
+      "learning_rate": 0.0002,
+      "loss": 1.1271,
+      "step": 8200
+    },
+    {
+      "epoch": 6.907867059318469,
+      "grad_norm": 1.213672399520874,
+      "learning_rate": 0.0002,
+      "loss": 1.2119,
+      "step": 8210
+    },
+    {
+      "epoch": 6.916281026503997,
+      "grad_norm": 1.2654485702514648,
+      "learning_rate": 0.0002,
+      "loss": 1.1406,
+      "step": 8220
+    },
+    {
+      "epoch": 6.924694993689524,
+      "grad_norm": 1.203903317451477,
+      "learning_rate": 0.0002,
+      "loss": 1.205,
+      "step": 8230
+    },
+    {
+      "epoch": 6.933108960875052,
+      "grad_norm": 1.1332030296325684,
+      "learning_rate": 0.0002,
+      "loss": 1.1635,
+      "step": 8240
+    },
+    {
+      "epoch": 6.94152292806058,
+      "grad_norm": 1.2699192762374878,
+      "learning_rate": 0.0002,
+      "loss": 1.1148,
+      "step": 8250
+    },
+    {
+      "epoch": 6.9499368952461085,
+      "grad_norm": 1.2728958129882812,
+      "learning_rate": 0.0002,
+      "loss": 1.1831,
+      "step": 8260
+    },
+    {
+      "epoch": 6.958350862431637,
+      "grad_norm": 1.238410472869873,
+      "learning_rate": 0.0002,
+      "loss": 1.1757,
+      "step": 8270
+    },
+    {
+      "epoch": 6.966764829617165,
+      "grad_norm": 1.403863549232483,
+      "learning_rate": 0.0002,
+      "loss": 1.1499,
+      "step": 8280
+    },
+    {
+      "epoch": 6.975178796802693,
+      "grad_norm": 1.1096396446228027,
+      "learning_rate": 0.0002,
+      "loss": 1.1515,
+      "step": 8290
+    },
+    {
+      "epoch": 6.98359276398822,
+      "grad_norm": 1.1043379306793213,
+      "learning_rate": 0.0002,
+      "loss": 1.2049,
+      "step": 8300
+    },
+    {
+      "epoch": 6.992006731173748,
+      "grad_norm": 1.391754388809204,
+      "learning_rate": 0.0002,
+      "loss": 1.1255,
+      "step": 8310
+    },
+    {
+      "epoch": 6.999579301640724,
+      "eval_loss": 2.1421656608581543,
+      "eval_runtime": 37.8262,
+      "eval_samples_per_second": 13.615,
+      "eval_steps_per_second": 1.718,
+      "step": 8319
+    },
+    {
+      "epoch": 7.000420698359276,
+      "grad_norm": 1.1739230155944824,
+      "learning_rate": 0.0002,
+      "loss": 1.1107,
+      "step": 8320
+    },
+    {
+      "epoch": 7.008834665544804,
+      "grad_norm": 1.5428645610809326,
+      "learning_rate": 0.0002,
+      "loss": 1.0066,
+      "step": 8330
+    },
+    {
+      "epoch": 7.0172486327303325,
+      "grad_norm": 1.307463526725769,
+      "learning_rate": 0.0002,
+      "loss": 0.9885,
+      "step": 8340
+    },
+    {
+      "epoch": 7.025662599915861,
+      "grad_norm": 1.4964789152145386,
+      "learning_rate": 0.0002,
+      "loss": 0.9098,
+      "step": 8350
+    },
+    {
+      "epoch": 7.034076567101389,
+      "grad_norm": 1.2289477586746216,
+      "learning_rate": 0.0002,
+      "loss": 0.8976,
+      "step": 8360
+    },
+    {
+      "epoch": 7.042490534286916,
+      "grad_norm": 1.325327754020691,
+      "learning_rate": 0.0002,
+      "loss": 0.9254,
+      "step": 8370
+    },
+    {
+      "epoch": 7.050904501472444,
+      "grad_norm": 1.4672988653182983,
+      "learning_rate": 0.0002,
+      "loss": 0.8967,
+      "step": 8380
+    },
+    {
+      "epoch": 7.059318468657972,
+      "grad_norm": 1.4184634685516357,
+      "learning_rate": 0.0002,
+      "loss": 0.8927,
+      "step": 8390
+    },
+    {
+      "epoch": 7.0677324358435,
+      "grad_norm": 1.3103536367416382,
+      "learning_rate": 0.0002,
+      "loss": 0.9129,
+      "step": 8400
+    },
+    {
+      "epoch": 7.076146403029028,
+      "grad_norm": 1.2364518642425537,
+      "learning_rate": 0.0002,
+      "loss": 0.997,
+      "step": 8410
+    },
+    {
+      "epoch": 7.0845603702145565,
+      "grad_norm": 1.3712464570999146,
+      "learning_rate": 0.0002,
+      "loss": 0.8776,
+      "step": 8420
+    },
+    {
+      "epoch": 7.092974337400084,
+      "grad_norm": 1.4655892848968506,
+      "learning_rate": 0.0002,
+      "loss": 0.9685,
+      "step": 8430
+    },
+    {
+      "epoch": 7.101388304585612,
+      "grad_norm": 1.3276227712631226,
+      "learning_rate": 0.0002,
+      "loss": 0.9276,
+      "step": 8440
+    },
+    {
+      "epoch": 7.10980227177114,
+      "grad_norm": 1.1355878114700317,
+      "learning_rate": 0.0002,
+      "loss": 0.9695,
+      "step": 8450
+    },
+    {
+      "epoch": 7.118216238956668,
+      "grad_norm": 1.2767117023468018,
+      "learning_rate": 0.0002,
+      "loss": 0.9673,
+      "step": 8460
+    },
+    {
+      "epoch": 7.126630206142196,
+      "grad_norm": 1.4915258884429932,
+      "learning_rate": 0.0002,
+      "loss": 0.9296,
+      "step": 8470
+    },
+    {
+      "epoch": 7.135044173327724,
+      "grad_norm": 1.355043649673462,
+      "learning_rate": 0.0002,
+      "loss": 0.9469,
+      "step": 8480
+    },
+    {
+      "epoch": 7.143458140513252,
+      "grad_norm": 1.0848617553710938,
+      "learning_rate": 0.0002,
+      "loss": 0.9626,
+      "step": 8490
+    },
+    {
+      "epoch": 7.15187210769878,
+      "grad_norm": 1.5321701765060425,
+      "learning_rate": 0.0002,
+      "loss": 0.92,
+      "step": 8500
+    },
+    {
+      "epoch": 7.160286074884308,
+      "grad_norm": 1.4917421340942383,
+      "learning_rate": 0.0002,
+      "loss": 0.9787,
+      "step": 8510
+    },
+    {
+      "epoch": 7.168700042069836,
+      "grad_norm": 1.4249778985977173,
+      "learning_rate": 0.0002,
+      "loss": 0.9709,
+      "step": 8520
+    },
+    {
+      "epoch": 7.177114009255364,
+      "grad_norm": 1.5257216691970825,
+      "learning_rate": 0.0002,
+      "loss": 0.9023,
+      "step": 8530
+    },
+    {
+      "epoch": 7.185527976440892,
+      "grad_norm": 1.4094327688217163,
+      "learning_rate": 0.0002,
+      "loss": 0.9818,
+      "step": 8540
+    },
+    {
+      "epoch": 7.19394194362642,
+      "grad_norm": 1.5506917238235474,
+      "learning_rate": 0.0002,
+      "loss": 0.9676,
+      "step": 8550
+    },
+    {
+      "epoch": 7.2023559108119475,
+      "grad_norm": 1.336599588394165,
+      "learning_rate": 0.0002,
+      "loss": 1.0494,
+      "step": 8560
+    },
+    {
+      "epoch": 7.210769877997476,
+      "grad_norm": 1.2018364667892456,
+      "learning_rate": 0.0002,
+      "loss": 0.9902,
+      "step": 8570
+    },
+    {
+      "epoch": 7.219183845183004,
+      "grad_norm": 1.198525071144104,
+      "learning_rate": 0.0002,
+      "loss": 0.9329,
+      "step": 8580
+    },
+    {
+      "epoch": 7.227597812368532,
+      "grad_norm": 1.4427133798599243,
+      "learning_rate": 0.0002,
+      "loss": 0.8954,
+      "step": 8590
+    },
+    {
+      "epoch": 7.23601177955406,
+      "grad_norm": 1.3134386539459229,
+      "learning_rate": 0.0002,
+      "loss": 0.9827,
+      "step": 8600
+    },
+    {
+      "epoch": 7.244425746739588,
+      "grad_norm": 1.4141706228256226,
+      "learning_rate": 0.0002,
+      "loss": 1.025,
+      "step": 8610
+    },
+    {
+      "epoch": 7.252839713925115,
+      "grad_norm": 1.4951153993606567,
+      "learning_rate": 0.0002,
+      "loss": 1.023,
+      "step": 8620
+    },
+    {
+      "epoch": 7.261253681110643,
+      "grad_norm": 1.383599042892456,
+      "learning_rate": 0.0002,
+      "loss": 0.9595,
+      "step": 8630
+    },
+    {
+      "epoch": 7.2696676482961715,
+      "grad_norm": 1.2315951585769653,
+      "learning_rate": 0.0002,
+      "loss": 0.9775,
+      "step": 8640
+    },
+    {
+      "epoch": 7.2780816154817,
+      "grad_norm": 1.253337025642395,
+      "learning_rate": 0.0002,
+      "loss": 0.9946,
+      "step": 8650
+    },
+    {
+      "epoch": 7.286495582667228,
+      "grad_norm": 1.2234476804733276,
+      "learning_rate": 0.0002,
+      "loss": 1.0381,
+      "step": 8660
+    },
+    {
+      "epoch": 7.294909549852756,
+      "grad_norm": 1.395650863647461,
+      "learning_rate": 0.0002,
+      "loss": 0.9774,
+      "step": 8670
+    },
+    {
+      "epoch": 7.303323517038283,
+      "grad_norm": 1.2411445379257202,
+      "learning_rate": 0.0002,
+      "loss": 0.9234,
+      "step": 8680
+    },
+    {
+      "epoch": 7.311737484223811,
+      "grad_norm": 1.22808837890625,
+      "learning_rate": 0.0002,
+      "loss": 0.975,
+      "step": 8690
+    },
+    {
+      "epoch": 7.320151451409339,
+      "grad_norm": 1.5197114944458008,
+      "learning_rate": 0.0002,
+      "loss": 0.9808,
+      "step": 8700
+    },
+    {
+      "epoch": 7.328565418594867,
+      "grad_norm": 1.3072681427001953,
+      "learning_rate": 0.0002,
+      "loss": 0.96,
+      "step": 8710
+    },
+    {
+      "epoch": 7.3369793857803955,
+      "grad_norm": 1.3035615682601929,
+      "learning_rate": 0.0002,
+      "loss": 0.9386,
+      "step": 8720
+    },
+    {
+      "epoch": 7.345393352965924,
+      "grad_norm": 1.2765713930130005,
+      "learning_rate": 0.0002,
+      "loss": 0.9666,
+      "step": 8730
+    },
+    {
+      "epoch": 7.353807320151452,
+      "grad_norm": 1.419601321220398,
+      "learning_rate": 0.0002,
+      "loss": 0.9581,
+      "step": 8740
+    },
+    {
+      "epoch": 7.362221287336979,
+      "grad_norm": 1.376158595085144,
+      "learning_rate": 0.0002,
+      "loss": 1.0378,
+      "step": 8750
+    },
+    {
+      "epoch": 7.370635254522507,
+      "grad_norm": 1.3880754709243774,
+      "learning_rate": 0.0002,
+      "loss": 0.9947,
+      "step": 8760
+    },
+    {
+      "epoch": 7.379049221708035,
+      "grad_norm": 1.2978262901306152,
+      "learning_rate": 0.0002,
+      "loss": 1.0512,
+      "step": 8770
+    },
+    {
+      "epoch": 7.387463188893563,
+      "grad_norm": 1.5811840295791626,
+      "learning_rate": 0.0002,
+      "loss": 1.0312,
+      "step": 8780
+    },
+    {
+      "epoch": 7.3958771560790915,
+      "grad_norm": 1.3790863752365112,
+      "learning_rate": 0.0002,
+      "loss": 0.9977,
+      "step": 8790
+    },
+    {
+      "epoch": 7.40429112326462,
+      "grad_norm": 1.475306510925293,
+      "learning_rate": 0.0002,
+      "loss": 1.008,
+      "step": 8800
+    },
+    {
+      "epoch": 7.412705090450148,
+      "grad_norm": 1.1038212776184082,
+      "learning_rate": 0.0002,
+      "loss": 0.9752,
+      "step": 8810
+    },
+    {
+      "epoch": 7.421119057635675,
+      "grad_norm": 1.5204451084136963,
+      "learning_rate": 0.0002,
+      "loss": 1.0048,
+      "step": 8820
+    },
+    {
+      "epoch": 7.429533024821203,
+      "grad_norm": 1.7151343822479248,
+      "learning_rate": 0.0002,
+      "loss": 1.019,
+      "step": 8830
+    },
+    {
+      "epoch": 7.437946992006731,
+      "grad_norm": 1.128046989440918,
+      "learning_rate": 0.0002,
+      "loss": 1.0038,
+      "step": 8840
+    },
+    {
+      "epoch": 7.446360959192259,
+      "grad_norm": 1.5780670642852783,
+      "learning_rate": 0.0002,
+      "loss": 1.0377,
+      "step": 8850
+    },
+    {
+      "epoch": 7.454774926377787,
+      "grad_norm": 1.3571979999542236,
+      "learning_rate": 0.0002,
+      "loss": 1.0584,
+      "step": 8860
+    },
+    {
+      "epoch": 7.4631888935633155,
+      "grad_norm": 1.2764537334442139,
+      "learning_rate": 0.0002,
+      "loss": 1.0141,
+      "step": 8870
+    },
+    {
+      "epoch": 7.471602860748843,
+      "grad_norm": 1.3429038524627686,
+      "learning_rate": 0.0002,
+      "loss": 0.9982,
+      "step": 8880
+    },
+    {
+      "epoch": 7.480016827934371,
+      "grad_norm": 1.3288369178771973,
+      "learning_rate": 0.0002,
+      "loss": 0.9671,
+      "step": 8890
+    },
+    {
+      "epoch": 7.488430795119899,
+      "grad_norm": 1.360141396522522,
+      "learning_rate": 0.0002,
+      "loss": 0.9461,
+      "step": 8900
+    },
+    {
+      "epoch": 7.496844762305427,
+      "grad_norm": 1.31229829788208,
+      "learning_rate": 0.0002,
+      "loss": 1.0278,
+      "step": 8910
+    },
+    {
+      "epoch": 7.505258729490955,
+      "grad_norm": 1.530605435371399,
+      "learning_rate": 0.0002,
+      "loss": 0.9945,
+      "step": 8920
+    },
+    {
+      "epoch": 7.513672696676483,
+      "grad_norm": 1.2880185842514038,
+      "learning_rate": 0.0002,
+      "loss": 1.0442,
+      "step": 8930
+    },
+    {
+      "epoch": 7.5220866638620105,
+      "grad_norm": 1.3219470977783203,
+      "learning_rate": 0.0002,
+      "loss": 0.9859,
+      "step": 8940
+    },
+    {
+      "epoch": 7.530500631047539,
+      "grad_norm": 1.565633773803711,
+      "learning_rate": 0.0002,
+      "loss": 1.0664,
+      "step": 8950
+    },
+    {
+      "epoch": 7.538914598233067,
+      "grad_norm": 1.4392317533493042,
+      "learning_rate": 0.0002,
+      "loss": 1.0089,
+      "step": 8960
+    },
+    {
+      "epoch": 7.547328565418595,
+      "grad_norm": 1.4557991027832031,
+      "learning_rate": 0.0002,
+      "loss": 1.0214,
+      "step": 8970
+    },
+    {
+      "epoch": 7.555742532604123,
+      "grad_norm": 1.3411110639572144,
+      "learning_rate": 0.0002,
+      "loss": 1.0247,
+      "step": 8980
+    },
+    {
+      "epoch": 7.564156499789651,
+      "grad_norm": 1.333378791809082,
+      "learning_rate": 0.0002,
+      "loss": 1.0738,
+      "step": 8990
+    },
+    {
+      "epoch": 7.572570466975179,
+      "grad_norm": 1.4422006607055664,
+      "learning_rate": 0.0002,
+      "loss": 1.0429,
+      "step": 9000
+    },
+    {
+      "epoch": 7.580984434160706,
+      "grad_norm": 1.2519633769989014,
+      "learning_rate": 0.0002,
+      "loss": 1.0401,
+      "step": 9010
+    },
+    {
+      "epoch": 7.589398401346235,
+      "grad_norm": 1.3628246784210205,
+      "learning_rate": 0.0002,
+      "loss": 1.0028,
+      "step": 9020
+    },
+    {
+      "epoch": 7.597812368531763,
+      "grad_norm": 1.35457181930542,
+      "learning_rate": 0.0002,
+      "loss": 0.9883,
+      "step": 9030
+    },
+    {
+      "epoch": 7.606226335717291,
+      "grad_norm": 1.4441956281661987,
+      "learning_rate": 0.0002,
+      "loss": 0.9929,
+      "step": 9040
+    },
+    {
+      "epoch": 7.614640302902819,
+      "grad_norm": 1.3812335729599,
+      "learning_rate": 0.0002,
+      "loss": 0.9987,
+      "step": 9050
+    },
+    {
+      "epoch": 7.623054270088347,
+      "grad_norm": 1.3576860427856445,
+      "learning_rate": 0.0002,
+      "loss": 0.9692,
+      "step": 9060
+    },
+    {
+      "epoch": 7.631468237273874,
+      "grad_norm": 1.350433588027954,
+      "learning_rate": 0.0002,
+      "loss": 1.0259,
+      "step": 9070
+    },
+    {
+      "epoch": 7.639882204459402,
+      "grad_norm": 1.3413814306259155,
+      "learning_rate": 0.0002,
+      "loss": 1.0292,
+      "step": 9080
+    },
+    {
+      "epoch": 7.6482961716449305,
+      "grad_norm": 1.2727786302566528,
+      "learning_rate": 0.0002,
+      "loss": 1.016,
+      "step": 9090
+    },
+    {
+      "epoch": 7.656710138830459,
+      "grad_norm": 1.1601275205612183,
+      "learning_rate": 0.0002,
+      "loss": 1.0046,
+      "step": 9100
+    },
+    {
+      "epoch": 7.665124106015987,
+      "grad_norm": 1.5492266416549683,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 9110
+    },
+    {
+      "epoch": 7.673538073201515,
+      "grad_norm": 1.4239033460617065,
+      "learning_rate": 0.0002,
+      "loss": 1.0174,
+      "step": 9120
+    },
+    {
+      "epoch": 7.681952040387042,
+      "grad_norm": 1.4212028980255127,
+      "learning_rate": 0.0002,
+      "loss": 0.9972,
+      "step": 9130
+    },
+    {
+      "epoch": 7.69036600757257,
+      "grad_norm": 1.116467833518982,
+      "learning_rate": 0.0002,
+      "loss": 1.0802,
+      "step": 9140
+    },
+    {
+      "epoch": 7.698779974758098,
+      "grad_norm": 1.299910545349121,
+      "learning_rate": 0.0002,
+      "loss": 1.0311,
+      "step": 9150
+    },
+    {
+      "epoch": 7.707193941943626,
+      "grad_norm": 1.404690146446228,
+      "learning_rate": 0.0002,
+      "loss": 1.0262,
+      "step": 9160
+    },
+    {
+      "epoch": 7.7156079091291545,
+      "grad_norm": 1.383244276046753,
+      "learning_rate": 0.0002,
+      "loss": 0.9633,
+      "step": 9170
+    },
+    {
+      "epoch": 7.724021876314683,
+      "grad_norm": 1.5001360177993774,
+      "learning_rate": 0.0002,
+      "loss": 1.0563,
+      "step": 9180
+    },
+    {
+      "epoch": 7.732435843500211,
+      "grad_norm": 1.4455186128616333,
+      "learning_rate": 0.0002,
+      "loss": 1.0731,
+      "step": 9190
+    },
+    {
+      "epoch": 7.740849810685738,
+      "grad_norm": 1.294964075088501,
+      "learning_rate": 0.0002,
+      "loss": 1.0667,
+      "step": 9200
+    },
+    {
+      "epoch": 7.749263777871266,
+      "grad_norm": 1.31305730342865,
+      "learning_rate": 0.0002,
+      "loss": 0.9649,
+      "step": 9210
+    },
+    {
+      "epoch": 7.757677745056794,
+      "grad_norm": 1.3849674463272095,
+      "learning_rate": 0.0002,
+      "loss": 0.9883,
+      "step": 9220
+    },
+    {
+      "epoch": 7.766091712242322,
+      "grad_norm": 1.6689352989196777,
+      "learning_rate": 0.0002,
+      "loss": 1.0219,
+      "step": 9230
+    },
+    {
+      "epoch": 7.77450567942785,
+      "grad_norm": 1.416099190711975,
+      "learning_rate": 0.0002,
+      "loss": 1.03,
+      "step": 9240
+    },
+    {
+      "epoch": 7.7829196466133785,
+      "grad_norm": 1.5212045907974243,
+      "learning_rate": 0.0002,
+      "loss": 1.0429,
+      "step": 9250
+    },
+    {
+      "epoch": 7.791333613798907,
+      "grad_norm": 1.3623390197753906,
+      "learning_rate": 0.0002,
+      "loss": 1.0607,
+      "step": 9260
+    },
+    {
+      "epoch": 7.799747580984434,
+      "grad_norm": 1.304148554801941,
+      "learning_rate": 0.0002,
+      "loss": 1.0469,
+      "step": 9270
+    },
+    {
+      "epoch": 7.808161548169962,
+      "grad_norm": 1.3833202123641968,
+      "learning_rate": 0.0002,
+      "loss": 1.0316,
+      "step": 9280
+    },
+    {
+      "epoch": 7.81657551535549,
+      "grad_norm": 1.3440886735916138,
+      "learning_rate": 0.0002,
+      "loss": 1.0122,
+      "step": 9290
+    },
+    {
+      "epoch": 7.824989482541018,
+      "grad_norm": 1.2798155546188354,
+      "learning_rate": 0.0002,
+      "loss": 1.0268,
+      "step": 9300
+    },
+    {
+      "epoch": 7.833403449726546,
+      "grad_norm": 1.3755156993865967,
+      "learning_rate": 0.0002,
+      "loss": 1.0521,
+      "step": 9310
+    },
+    {
+      "epoch": 7.841817416912074,
+      "grad_norm": 1.3145397901535034,
+      "learning_rate": 0.0002,
+      "loss": 1.0571,
+      "step": 9320
+    },
+    {
+      "epoch": 7.850231384097602,
+      "grad_norm": 1.6102794408798218,
+      "learning_rate": 0.0002,
+      "loss": 1.0684,
+      "step": 9330
+    },
+    {
+      "epoch": 7.85864535128313,
+      "grad_norm": 1.3959331512451172,
+      "learning_rate": 0.0002,
+      "loss": 1.1878,
+      "step": 9340
+    },
+    {
+      "epoch": 7.867059318468658,
+      "grad_norm": 1.4965628385543823,
+      "learning_rate": 0.0002,
+      "loss": 1.05,
+      "step": 9350
+    },
+    {
+      "epoch": 7.875473285654186,
+      "grad_norm": 1.194201946258545,
+      "learning_rate": 0.0002,
+      "loss": 1.085,
+      "step": 9360
+    },
+    {
+      "epoch": 7.883887252839714,
+      "grad_norm": 1.4831446409225464,
+      "learning_rate": 0.0002,
+      "loss": 1.0712,
+      "step": 9370
+    },
+    {
+      "epoch": 7.892301220025242,
+      "grad_norm": 1.3473302125930786,
+      "learning_rate": 0.0002,
+      "loss": 1.0568,
+      "step": 9380
+    },
+    {
+      "epoch": 7.9007151872107695,
+      "grad_norm": 1.4373382329940796,
+      "learning_rate": 0.0002,
+      "loss": 1.0172,
+      "step": 9390
+    },
+    {
+      "epoch": 7.909129154396298,
+      "grad_norm": 1.4341524839401245,
+      "learning_rate": 0.0002,
+      "loss": 0.9892,
+      "step": 9400
+    },
+    {
+      "epoch": 7.917543121581826,
+      "grad_norm": 1.3210171461105347,
+      "learning_rate": 0.0002,
+      "loss": 1.0428,
+      "step": 9410
+    },
+    {
+      "epoch": 7.925957088767354,
+      "grad_norm": 1.2708462476730347,
+      "learning_rate": 0.0002,
+      "loss": 1.0543,
+      "step": 9420
+    },
+    {
+      "epoch": 7.934371055952882,
+      "grad_norm": 1.4132758378982544,
+      "learning_rate": 0.0002,
+      "loss": 1.0789,
+      "step": 9430
+    },
+    {
+      "epoch": 7.94278502313841,
+      "grad_norm": 1.5193610191345215,
+      "learning_rate": 0.0002,
+      "loss": 1.095,
+      "step": 9440
+    },
+    {
+      "epoch": 7.951198990323938,
+      "grad_norm": 1.427832841873169,
+      "learning_rate": 0.0002,
+      "loss": 0.967,
+      "step": 9450
+    },
+    {
+      "epoch": 7.959612957509465,
+      "grad_norm": 1.380478024482727,
+      "learning_rate": 0.0002,
+      "loss": 1.0052,
+      "step": 9460
+    },
+    {
+      "epoch": 7.9680269246949935,
+      "grad_norm": 1.3083926439285278,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 9470
+    },
+    {
+      "epoch": 7.976440891880522,
+      "grad_norm": 1.3049120903015137,
+      "learning_rate": 0.0002,
+      "loss": 1.0883,
+      "step": 9480
+    },
+    {
+      "epoch": 7.98485485906605,
+      "grad_norm": 1.42048978805542,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 9490
+    },
+    {
+      "epoch": 7.993268826251578,
+      "grad_norm": 1.2492578029632568,
+      "learning_rate": 0.0002,
+      "loss": 1.094,
+      "step": 9500
+    },
+    {
+      "epoch": 7.996634413125789,
+      "eval_loss": 2.3033790588378906,
+      "eval_runtime": 37.8581,
+      "eval_samples_per_second": 13.603,
+      "eval_steps_per_second": 1.717,
+      "step": 9504
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 9504,
+  "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": 4.3982383968210125e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/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-0.8-num-5623-sd-42/checkpoint-9504/training_args.bin b/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-0.8-num-5623-sd-42/checkpoint-9504/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f1502d478cfbb1424f707352d007b740bde5e373
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/checkpoint-9504/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df2b79d3acefeedef5a0229881de39ec68ef9b40046a60d7976a49f7e6b3b936
+size 5560
diff --git a/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-0.8-num-5623-sd-42/special_tokens_map.json b/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-0.8-num-5623-sd-42/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/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_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-0.8-num-5623-sd-42/tokenizer.json b/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-0.8-num-5623-sd-42/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/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_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-0.8-num-5623-sd-42/tokenizer_config.json b/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-0.8-num-5623-sd-42/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/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_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-0.8-num-5623-sd-42/training_args.bin b/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-0.8-num-5623-sd-42/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f1502d478cfbb1424f707352d007b740bde5e373
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df2b79d3acefeedef5a0229881de39ec68ef9b40046a60d7976a49f7e6b3b936
+size 5560
diff --git a/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-0.8-num-5623-sd-42/training_log.jsonl b/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-0.8-num-5623-sd-42/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..2a80dc982f56f7b26632463941f9b4334525b3c9
--- /dev/null
+++ b/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-0.8-num-5623-sd-42/training_log.jsonl
@@ -0,0 +1,8 @@
+{"epoch": 0.9995793016407236, "step": 1188, "epoch_duration": 1650.0460257530212, "total_accumulated_duration": 1650.0460257530212, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "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": 2.56, "grad_norm": 0.5458821654319763, "learning_rate": 0.0002, "epoch": 0.008413967185527976, "step": 10}, {"loss": 2.3235, "grad_norm": 0.7293308973312378, "learning_rate": 0.0002, "epoch": 0.016827934371055953, "step": 20}, {"loss": 2.0815, "grad_norm": 0.47792306542396545, "learning_rate": 0.0002, "epoch": 0.02524190155658393, "step": 30}, {"loss": 1.9718, "grad_norm": 0.5944402813911438, "learning_rate": 0.0002, "epoch": 0.033655868742111905, "step": 40}, {"loss": 1.8848, "grad_norm": 0.5415359735488892, "learning_rate": 0.0002, "epoch": 0.04206983592763988, "step": 50}, {"loss": 1.8953, "grad_norm": 0.535713791847229, "learning_rate": 0.0002, "epoch": 0.05048380311316786, "step": 60}, {"loss": 1.937, "grad_norm": 0.5184146761894226, "learning_rate": 0.0002, "epoch": 0.058897770298695834, "step": 70}, {"loss": 1.8396, "grad_norm": 0.458926796913147, "learning_rate": 0.0002, "epoch": 0.06731173748422381, "step": 80}, {"loss": 1.8677, "grad_norm": 0.4780142307281494, "learning_rate": 0.0002, "epoch": 0.07572570466975179, "step": 90}, {"loss": 1.8593, "grad_norm": 0.79965740442276, "learning_rate": 0.0002, "epoch": 0.08413967185527976, "step": 100}, {"loss": 1.9081, "grad_norm": 0.4498862028121948, "learning_rate": 0.0002, "epoch": 0.09255363904080774, "step": 110}, {"loss": 1.8503, "grad_norm": 0.39338430762290955, "learning_rate": 0.0002, "epoch": 0.10096760622633572, "step": 120}, {"loss": 1.8637, "grad_norm": 0.9588953852653503, "learning_rate": 0.0002, "epoch": 0.10938157341186369, "step": 130}, {"loss": 1.8676, "grad_norm": 0.41675639152526855, "learning_rate": 0.0002, "epoch": 0.11779554059739167, "step": 140}, {"loss": 1.8904, "grad_norm": 0.44519832730293274, "learning_rate": 0.0002, "epoch": 0.12620950778291964, "step": 150}, {"loss": 1.798, "grad_norm": 0.4176260530948639, "learning_rate": 0.0002, "epoch": 0.13462347496844762, "step": 160}, {"loss": 1.8398, "grad_norm": 0.35840365290641785, "learning_rate": 0.0002, "epoch": 0.1430374421539756, "step": 170}, {"loss": 1.8666, "grad_norm": 0.3794495463371277, "learning_rate": 0.0002, "epoch": 0.15145140933950357, "step": 180}, {"loss": 1.8111, "grad_norm": 0.4563522934913635, "learning_rate": 0.0002, "epoch": 0.15986537652503155, "step": 190}, {"loss": 1.8893, "grad_norm": 0.37057486176490784, "learning_rate": 0.0002, "epoch": 0.16827934371055953, "step": 200}, {"loss": 1.7995, "grad_norm": 0.44081518054008484, "learning_rate": 0.0002, "epoch": 0.1766933108960875, "step": 210}, {"loss": 1.9048, "grad_norm": 0.46078577637672424, "learning_rate": 0.0002, "epoch": 0.18510727808161548, "step": 220}, {"loss": 1.8403, "grad_norm": 0.36132094264030457, "learning_rate": 0.0002, "epoch": 0.19352124526714345, "step": 230}, {"loss": 1.8827, "grad_norm": 0.3747289180755615, "learning_rate": 0.0002, "epoch": 0.20193521245267143, "step": 240}, {"loss": 1.8382, "grad_norm": 0.3540179133415222, "learning_rate": 0.0002, "epoch": 0.2103491796381994, "step": 250}, {"loss": 1.8196, "grad_norm": 0.3461375832557678, "learning_rate": 0.0002, "epoch": 0.21876314682372738, "step": 260}, {"loss": 1.8509, "grad_norm": 0.3436960279941559, "learning_rate": 0.0002, "epoch": 0.22717711400925536, "step": 270}, {"loss": 1.8285, "grad_norm": 0.35403719544410706, "learning_rate": 0.0002, "epoch": 0.23559108119478334, "step": 280}, {"loss": 1.8369, "grad_norm": 0.37142616510391235, "learning_rate": 0.0002, "epoch": 0.2440050483803113, "step": 290}, {"loss": 1.8044, "grad_norm": 0.3307955861091614, "learning_rate": 0.0002, "epoch": 0.2524190155658393, "step": 300}, {"loss": 1.817, "grad_norm": 0.32855314016342163, "learning_rate": 0.0002, "epoch": 0.2608329827513673, "step": 310}, {"loss": 1.7803, "grad_norm": 0.3299003839492798, "learning_rate": 0.0002, "epoch": 0.26924694993689524, "step": 320}, {"loss": 1.8129, "grad_norm": 0.44311287999153137, "learning_rate": 0.0002, "epoch": 0.27766091712242325, "step": 330}, {"loss": 1.8232, "grad_norm": 0.32989758253097534, "learning_rate": 0.0002, "epoch": 0.2860748843079512, "step": 340}, {"loss": 1.7716, "grad_norm": 0.34400200843811035, "learning_rate": 0.0002, "epoch": 0.2944888514934792, "step": 350}, {"loss": 1.7619, "grad_norm": 0.36286211013793945, "learning_rate": 0.0002, "epoch": 0.30290281867900715, "step": 360}, {"loss": 1.8025, "grad_norm": 0.406827837228775, "learning_rate": 0.0002, "epoch": 0.31131678586453515, "step": 370}, {"loss": 1.7515, "grad_norm": 0.36299195885658264, "learning_rate": 0.0002, "epoch": 0.3197307530500631, "step": 380}, {"loss": 1.837, "grad_norm": 0.3477257192134857, "learning_rate": 0.0002, "epoch": 0.3281447202355911, "step": 390}, {"loss": 1.7767, "grad_norm": 0.3730369210243225, "learning_rate": 0.0002, "epoch": 0.33655868742111905, "step": 400}, {"loss": 1.7747, "grad_norm": 0.4644559919834137, "learning_rate": 0.0002, "epoch": 0.34497265460664706, "step": 410}, {"loss": 1.7538, "grad_norm": 0.406576544046402, "learning_rate": 0.0002, "epoch": 0.353386621792175, "step": 420}, {"loss": 1.7501, "grad_norm": 0.3612699508666992, "learning_rate": 0.0002, "epoch": 0.361800588977703, "step": 430}, {"loss": 1.7473, "grad_norm": 0.3243742287158966, "learning_rate": 0.0002, "epoch": 0.37021455616323096, "step": 440}, {"loss": 1.8851, "grad_norm": 0.36671221256256104, "learning_rate": 0.0002, "epoch": 0.37862852334875896, "step": 450}, {"loss": 1.8853, "grad_norm": 0.3565002381801605, "learning_rate": 0.0002, "epoch": 0.3870424905342869, "step": 460}, {"loss": 1.8923, "grad_norm": 0.34630221128463745, "learning_rate": 0.0002, "epoch": 0.3954564577198149, "step": 470}, {"loss": 1.8234, "grad_norm": 0.3353537321090698, "learning_rate": 0.0002, "epoch": 0.40387042490534286, "step": 480}, {"loss": 1.7135, "grad_norm": 0.4015921950340271, "learning_rate": 0.0002, "epoch": 0.41228439209087087, "step": 490}, {"loss": 1.7815, "grad_norm": 0.5489419102668762, "learning_rate": 0.0002, "epoch": 0.4206983592763988, "step": 500}, {"loss": 1.7903, "grad_norm": 0.4193589985370636, "learning_rate": 0.0002, "epoch": 0.4291123264619268, "step": 510}, {"loss": 1.8416, "grad_norm": 0.3418922424316406, "learning_rate": 0.0002, "epoch": 0.43752629364745477, "step": 520}, {"loss": 1.7982, "grad_norm": 0.32668185234069824, "learning_rate": 0.0002, "epoch": 0.44594026083298277, "step": 530}, {"loss": 1.7501, "grad_norm": 0.3094325661659241, "learning_rate": 0.0002, "epoch": 0.4543542280185107, "step": 540}, {"loss": 1.7438, "grad_norm": 0.3743017315864563, "learning_rate": 0.0002, "epoch": 0.4627681952040387, "step": 550}, {"loss": 1.8451, "grad_norm": 0.3295630216598511, "learning_rate": 0.0002, "epoch": 0.47118216238956667, "step": 560}, {"loss": 1.7529, "grad_norm": 1.6124513149261475, "learning_rate": 0.0002, "epoch": 0.4795961295750947, "step": 570}, {"loss": 1.8028, "grad_norm": 0.3245585858821869, "learning_rate": 0.0002, "epoch": 0.4880100967606226, "step": 580}, {"loss": 1.7976, "grad_norm": 0.3332934081554413, "learning_rate": 0.0002, "epoch": 0.49642406394615063, "step": 590}, {"loss": 1.7912, "grad_norm": 0.3836138844490051, "learning_rate": 0.0002, "epoch": 0.5048380311316786, "step": 600}, {"loss": 1.8347, "grad_norm": 0.32953888177871704, "learning_rate": 0.0002, "epoch": 0.5132519983172066, "step": 610}, {"loss": 1.7729, "grad_norm": 0.36291512846946716, "learning_rate": 0.0002, "epoch": 0.5216659655027346, "step": 620}, {"loss": 1.7758, "grad_norm": 0.3237783908843994, "learning_rate": 0.0002, "epoch": 0.5300799326882625, "step": 630}, {"loss": 1.8352, "grad_norm": 0.38882696628570557, "learning_rate": 0.0002, "epoch": 0.5384938998737905, "step": 640}, {"loss": 1.8624, "grad_norm": 0.37821972370147705, "learning_rate": 0.0002, "epoch": 0.5469078670593185, "step": 650}, {"loss": 1.8075, "grad_norm": 0.3556285500526428, "learning_rate": 0.0002, "epoch": 0.5553218342448465, "step": 660}, {"loss": 1.778, "grad_norm": 0.347499281167984, "learning_rate": 0.0002, "epoch": 0.5637358014303744, "step": 670}, {"loss": 1.8066, "grad_norm": 0.3176489472389221, "learning_rate": 0.0002, "epoch": 0.5721497686159024, "step": 680}, {"loss": 1.7257, "grad_norm": 0.30220088362693787, "learning_rate": 0.0002, "epoch": 0.5805637358014304, "step": 690}, {"loss": 1.8415, "grad_norm": 0.3711601793766022, "learning_rate": 0.0002, "epoch": 0.5889777029869584, "step": 700}, {"loss": 1.7906, "grad_norm": 0.3311759829521179, "learning_rate": 0.0002, "epoch": 0.5973916701724863, "step": 710}, {"loss": 1.7712, "grad_norm": 0.34824270009994507, "learning_rate": 0.0002, "epoch": 0.6058056373580143, "step": 720}, {"loss": 1.7954, "grad_norm": 0.29668381810188293, "learning_rate": 0.0002, "epoch": 0.6142196045435423, "step": 730}, {"loss": 1.8321, "grad_norm": 0.36087489128112793, "learning_rate": 0.0002, "epoch": 0.6226335717290703, "step": 740}, {"loss": 1.7956, "grad_norm": 0.31590089201927185, "learning_rate": 0.0002, "epoch": 0.6310475389145982, "step": 750}, {"loss": 1.7343, "grad_norm": 0.37632957100868225, "learning_rate": 0.0002, "epoch": 0.6394615061001262, "step": 760}, {"loss": 1.8499, "grad_norm": 0.3360748589038849, "learning_rate": 0.0002, "epoch": 0.6478754732856542, "step": 770}, {"loss": 1.8076, "grad_norm": 0.3420640528202057, "learning_rate": 0.0002, "epoch": 0.6562894404711822, "step": 780}, {"loss": 1.8353, "grad_norm": 0.5734959244728088, "learning_rate": 0.0002, "epoch": 0.6647034076567101, "step": 790}, {"loss": 1.7746, "grad_norm": 0.36440837383270264, "learning_rate": 0.0002, "epoch": 0.6731173748422381, "step": 800}, {"loss": 1.7532, "grad_norm": 0.3179708421230316, "learning_rate": 0.0002, "epoch": 0.6815313420277661, "step": 810}, {"loss": 1.7815, "grad_norm": 0.34122881293296814, "learning_rate": 0.0002, "epoch": 0.6899453092132941, "step": 820}, {"loss": 1.8167, "grad_norm": 0.31886112689971924, "learning_rate": 0.0002, "epoch": 0.698359276398822, "step": 830}, {"loss": 1.7505, "grad_norm": 0.31782326102256775, "learning_rate": 0.0002, "epoch": 0.70677324358435, "step": 840}, {"loss": 1.7588, "grad_norm": 0.36052989959716797, "learning_rate": 0.0002, "epoch": 0.715187210769878, "step": 850}, {"loss": 1.7891, "grad_norm": 0.28946155309677124, "learning_rate": 0.0002, "epoch": 0.723601177955406, "step": 860}, {"loss": 1.7923, "grad_norm": 0.3095663785934448, "learning_rate": 0.0002, "epoch": 0.7320151451409339, "step": 870}, {"loss": 1.785, "grad_norm": 0.3317491412162781, "learning_rate": 0.0002, "epoch": 0.7404291123264619, "step": 880}, {"loss": 1.7709, "grad_norm": 0.31324660778045654, "learning_rate": 0.0002, "epoch": 0.7488430795119899, "step": 890}, {"loss": 1.8753, "grad_norm": 0.3290475606918335, "learning_rate": 0.0002, "epoch": 0.7572570466975179, "step": 900}, {"loss": 1.7679, "grad_norm": 0.35690343379974365, "learning_rate": 0.0002, "epoch": 0.7656710138830458, "step": 910}, {"loss": 1.826, "grad_norm": 0.39558273553848267, "learning_rate": 0.0002, "epoch": 0.7740849810685738, "step": 920}, {"loss": 1.8722, "grad_norm": 0.34254348278045654, "learning_rate": 0.0002, "epoch": 0.7824989482541018, "step": 930}, {"loss": 1.7603, "grad_norm": 0.3560165464878082, "learning_rate": 0.0002, "epoch": 0.7909129154396298, "step": 940}, {"loss": 1.7992, "grad_norm": 0.30693164467811584, "learning_rate": 0.0002, "epoch": 0.7993268826251577, "step": 950}, {"loss": 1.8029, "grad_norm": 0.3394823372364044, "learning_rate": 0.0002, "epoch": 0.8077408498106857, "step": 960}, {"loss": 1.8105, "grad_norm": 0.3741514980792999, "learning_rate": 0.0002, "epoch": 0.8161548169962137, "step": 970}, {"loss": 1.7849, "grad_norm": 0.3655228316783905, "learning_rate": 0.0002, "epoch": 0.8245687841817417, "step": 980}, {"loss": 1.8449, "grad_norm": 0.3586033880710602, "learning_rate": 0.0002, "epoch": 0.8329827513672696, "step": 990}, {"loss": 1.7033, "grad_norm": 0.3459678888320923, "learning_rate": 0.0002, "epoch": 0.8413967185527976, "step": 1000}, {"loss": 1.8498, "grad_norm": 0.3184349834918976, "learning_rate": 0.0002, "epoch": 0.8498106857383256, "step": 1010}, {"loss": 1.7632, "grad_norm": 0.3099786043167114, "learning_rate": 0.0002, "epoch": 0.8582246529238536, "step": 1020}, {"loss": 1.8067, "grad_norm": 0.30300915241241455, "learning_rate": 0.0002, "epoch": 0.8666386201093815, "step": 1030}, {"loss": 1.7923, "grad_norm": 0.3128705620765686, "learning_rate": 0.0002, "epoch": 0.8750525872949095, "step": 1040}, {"loss": 1.8252, "grad_norm": 0.3336263597011566, "learning_rate": 0.0002, "epoch": 0.8834665544804375, "step": 1050}, {"loss": 1.8375, "grad_norm": 0.3801328241825104, "learning_rate": 0.0002, "epoch": 0.8918805216659655, "step": 1060}, {"loss": 1.7757, "grad_norm": 0.3122096359729767, "learning_rate": 0.0002, "epoch": 0.9002944888514934, "step": 1070}, {"loss": 1.8251, "grad_norm": 0.35990869998931885, "learning_rate": 0.0002, "epoch": 0.9087084560370214, "step": 1080}, {"loss": 1.7343, "grad_norm": 0.3321819305419922, "learning_rate": 0.0002, "epoch": 0.9171224232225494, "step": 1090}, {"loss": 1.7595, "grad_norm": 0.4202139377593994, "learning_rate": 0.0002, "epoch": 0.9255363904080774, "step": 1100}, {"loss": 1.8056, "grad_norm": 0.32559722661972046, "learning_rate": 0.0002, "epoch": 0.9339503575936053, "step": 1110}, {"loss": 1.812, "grad_norm": 0.3098459839820862, "learning_rate": 0.0002, "epoch": 0.9423643247791333, "step": 1120}, {"loss": 1.8252, "grad_norm": 0.33917108178138733, "learning_rate": 0.0002, "epoch": 0.9507782919646613, "step": 1130}, {"loss": 1.7709, "grad_norm": 0.4055837094783783, "learning_rate": 0.0002, "epoch": 0.9591922591501894, "step": 1140}, {"loss": 1.8259, "grad_norm": 0.32508623600006104, "learning_rate": 0.0002, "epoch": 0.9676062263357172, "step": 1150}, {"loss": 1.782, "grad_norm": 0.30150601267814636, "learning_rate": 0.0002, "epoch": 0.9760201935212452, "step": 1160}, {"loss": 1.8291, "grad_norm": 0.3042563199996948, "learning_rate": 0.0002, "epoch": 0.9844341607067733, "step": 1170}, {"loss": 1.7847, "grad_norm": 0.33254584670066833, "learning_rate": 0.0002, "epoch": 0.9928481278923013, "step": 1180}]}
+{"epoch": 2.0, "step": 2377, "epoch_duration": 1658.4773199558258, "total_accumulated_duration": 3308.523345708847, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "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_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-0.8-num-5623-sd-42/checkpoint-1188", "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": 2.56, "grad_norm": 0.5458821654319763, "learning_rate": 0.0002, "epoch": 0.008413967185527976, "step": 10}, {"loss": 2.3235, "grad_norm": 0.7293308973312378, "learning_rate": 0.0002, "epoch": 0.016827934371055953, "step": 20}, {"loss": 2.0815, "grad_norm": 0.47792306542396545, "learning_rate": 0.0002, "epoch": 0.02524190155658393, "step": 30}, {"loss": 1.9718, "grad_norm": 0.5944402813911438, "learning_rate": 0.0002, "epoch": 0.033655868742111905, "step": 40}, {"loss": 1.8848, "grad_norm": 0.5415359735488892, "learning_rate": 0.0002, "epoch": 0.04206983592763988, "step": 50}, {"loss": 1.8953, "grad_norm": 0.535713791847229, "learning_rate": 0.0002, "epoch": 0.05048380311316786, "step": 60}, {"loss": 1.937, "grad_norm": 0.5184146761894226, "learning_rate": 0.0002, "epoch": 0.058897770298695834, "step": 70}, {"loss": 1.8396, "grad_norm": 0.458926796913147, "learning_rate": 0.0002, "epoch": 0.06731173748422381, "step": 80}, {"loss": 1.8677, "grad_norm": 0.4780142307281494, "learning_rate": 0.0002, "epoch": 0.07572570466975179, "step": 90}, {"loss": 1.8593, "grad_norm": 0.79965740442276, "learning_rate": 0.0002, "epoch": 0.08413967185527976, "step": 100}, {"loss": 1.9081, "grad_norm": 0.4498862028121948, "learning_rate": 0.0002, "epoch": 0.09255363904080774, "step": 110}, {"loss": 1.8503, "grad_norm": 0.39338430762290955, "learning_rate": 0.0002, "epoch": 0.10096760622633572, "step": 120}, {"loss": 1.8637, "grad_norm": 0.9588953852653503, "learning_rate": 0.0002, "epoch": 0.10938157341186369, "step": 130}, {"loss": 1.8676, "grad_norm": 0.41675639152526855, "learning_rate": 0.0002, "epoch": 0.11779554059739167, "step": 140}, {"loss": 1.8904, "grad_norm": 0.44519832730293274, "learning_rate": 0.0002, "epoch": 0.12620950778291964, "step": 150}, {"loss": 1.798, "grad_norm": 0.4176260530948639, "learning_rate": 0.0002, "epoch": 0.13462347496844762, "step": 160}, {"loss": 1.8398, "grad_norm": 0.35840365290641785, "learning_rate": 0.0002, "epoch": 0.1430374421539756, "step": 170}, {"loss": 1.8666, "grad_norm": 0.3794495463371277, "learning_rate": 0.0002, "epoch": 0.15145140933950357, "step": 180}, {"loss": 1.8111, "grad_norm": 0.4563522934913635, "learning_rate": 0.0002, "epoch": 0.15986537652503155, "step": 190}, {"loss": 1.8893, "grad_norm": 0.37057486176490784, "learning_rate": 0.0002, "epoch": 0.16827934371055953, "step": 200}, {"loss": 1.7995, "grad_norm": 0.44081518054008484, "learning_rate": 0.0002, "epoch": 0.1766933108960875, "step": 210}, {"loss": 1.9048, "grad_norm": 0.46078577637672424, "learning_rate": 0.0002, "epoch": 0.18510727808161548, "step": 220}, {"loss": 1.8403, "grad_norm": 0.36132094264030457, "learning_rate": 0.0002, "epoch": 0.19352124526714345, "step": 230}, {"loss": 1.8827, "grad_norm": 0.3747289180755615, "learning_rate": 0.0002, "epoch": 0.20193521245267143, "step": 240}, {"loss": 1.8382, "grad_norm": 0.3540179133415222, "learning_rate": 0.0002, "epoch": 0.2103491796381994, "step": 250}, {"loss": 1.8196, "grad_norm": 0.3461375832557678, "learning_rate": 0.0002, "epoch": 0.21876314682372738, "step": 260}, {"loss": 1.8509, "grad_norm": 0.3436960279941559, "learning_rate": 0.0002, "epoch": 0.22717711400925536, "step": 270}, {"loss": 1.8285, "grad_norm": 0.35403719544410706, "learning_rate": 0.0002, "epoch": 0.23559108119478334, "step": 280}, {"loss": 1.8369, "grad_norm": 0.37142616510391235, "learning_rate": 0.0002, "epoch": 0.2440050483803113, "step": 290}, {"loss": 1.8044, "grad_norm": 0.3307955861091614, "learning_rate": 0.0002, "epoch": 0.2524190155658393, "step": 300}, {"loss": 1.817, "grad_norm": 0.32855314016342163, "learning_rate": 0.0002, "epoch": 0.2608329827513673, "step": 310}, {"loss": 1.7803, "grad_norm": 0.3299003839492798, "learning_rate": 0.0002, "epoch": 0.26924694993689524, "step": 320}, {"loss": 1.8129, "grad_norm": 0.44311287999153137, "learning_rate": 0.0002, "epoch": 0.27766091712242325, "step": 330}, {"loss": 1.8232, "grad_norm": 0.32989758253097534, "learning_rate": 0.0002, "epoch": 0.2860748843079512, "step": 340}, {"loss": 1.7716, "grad_norm": 0.34400200843811035, "learning_rate": 0.0002, "epoch": 0.2944888514934792, "step": 350}, {"loss": 1.7619, "grad_norm": 0.36286211013793945, "learning_rate": 0.0002, "epoch": 0.30290281867900715, "step": 360}, {"loss": 1.8025, "grad_norm": 0.406827837228775, "learning_rate": 0.0002, "epoch": 0.31131678586453515, "step": 370}, {"loss": 1.7515, "grad_norm": 0.36299195885658264, "learning_rate": 0.0002, "epoch": 0.3197307530500631, "step": 380}, {"loss": 1.837, "grad_norm": 0.3477257192134857, "learning_rate": 0.0002, "epoch": 0.3281447202355911, "step": 390}, {"loss": 1.7767, "grad_norm": 0.3730369210243225, "learning_rate": 0.0002, "epoch": 0.33655868742111905, "step": 400}, {"loss": 1.7747, "grad_norm": 0.4644559919834137, "learning_rate": 0.0002, "epoch": 0.34497265460664706, "step": 410}, {"loss": 1.7538, "grad_norm": 0.406576544046402, "learning_rate": 0.0002, "epoch": 0.353386621792175, "step": 420}, {"loss": 1.7501, "grad_norm": 0.3612699508666992, "learning_rate": 0.0002, "epoch": 0.361800588977703, "step": 430}, {"loss": 1.7473, "grad_norm": 0.3243742287158966, "learning_rate": 0.0002, "epoch": 0.37021455616323096, "step": 440}, {"loss": 1.8851, "grad_norm": 0.36671221256256104, "learning_rate": 0.0002, "epoch": 0.37862852334875896, "step": 450}, {"loss": 1.8853, "grad_norm": 0.3565002381801605, "learning_rate": 0.0002, "epoch": 0.3870424905342869, "step": 460}, {"loss": 1.8923, "grad_norm": 0.34630221128463745, "learning_rate": 0.0002, "epoch": 0.3954564577198149, "step": 470}, {"loss": 1.8234, "grad_norm": 0.3353537321090698, "learning_rate": 0.0002, "epoch": 0.40387042490534286, "step": 480}, {"loss": 1.7135, "grad_norm": 0.4015921950340271, "learning_rate": 0.0002, "epoch": 0.41228439209087087, "step": 490}, {"loss": 1.7815, "grad_norm": 0.5489419102668762, "learning_rate": 0.0002, "epoch": 0.4206983592763988, "step": 500}, {"loss": 1.7903, "grad_norm": 0.4193589985370636, "learning_rate": 0.0002, "epoch": 0.4291123264619268, "step": 510}, {"loss": 1.8416, "grad_norm": 0.3418922424316406, "learning_rate": 0.0002, "epoch": 0.43752629364745477, "step": 520}, {"loss": 1.7982, "grad_norm": 0.32668185234069824, "learning_rate": 0.0002, "epoch": 0.44594026083298277, "step": 530}, {"loss": 1.7501, "grad_norm": 0.3094325661659241, "learning_rate": 0.0002, "epoch": 0.4543542280185107, "step": 540}, {"loss": 1.7438, "grad_norm": 0.3743017315864563, "learning_rate": 0.0002, "epoch": 0.4627681952040387, "step": 550}, {"loss": 1.8451, "grad_norm": 0.3295630216598511, "learning_rate": 0.0002, "epoch": 0.47118216238956667, "step": 560}, {"loss": 1.7529, "grad_norm": 1.6124513149261475, "learning_rate": 0.0002, "epoch": 0.4795961295750947, "step": 570}, {"loss": 1.8028, "grad_norm": 0.3245585858821869, "learning_rate": 0.0002, "epoch": 0.4880100967606226, "step": 580}, {"loss": 1.7976, "grad_norm": 0.3332934081554413, "learning_rate": 0.0002, "epoch": 0.49642406394615063, "step": 590}, {"loss": 1.7912, "grad_norm": 0.3836138844490051, "learning_rate": 0.0002, "epoch": 0.5048380311316786, "step": 600}, {"loss": 1.8347, "grad_norm": 0.32953888177871704, "learning_rate": 0.0002, "epoch": 0.5132519983172066, "step": 610}, {"loss": 1.7729, "grad_norm": 0.36291512846946716, "learning_rate": 0.0002, "epoch": 0.5216659655027346, "step": 620}, {"loss": 1.7758, "grad_norm": 0.3237783908843994, "learning_rate": 0.0002, "epoch": 0.5300799326882625, "step": 630}, {"loss": 1.8352, "grad_norm": 0.38882696628570557, "learning_rate": 0.0002, "epoch": 0.5384938998737905, "step": 640}, {"loss": 1.8624, "grad_norm": 0.37821972370147705, "learning_rate": 0.0002, "epoch": 0.5469078670593185, "step": 650}, {"loss": 1.8075, "grad_norm": 0.3556285500526428, "learning_rate": 0.0002, "epoch": 0.5553218342448465, "step": 660}, {"loss": 1.778, "grad_norm": 0.347499281167984, "learning_rate": 0.0002, "epoch": 0.5637358014303744, "step": 670}, {"loss": 1.8066, "grad_norm": 0.3176489472389221, "learning_rate": 0.0002, "epoch": 0.5721497686159024, "step": 680}, {"loss": 1.7257, "grad_norm": 0.30220088362693787, "learning_rate": 0.0002, "epoch": 0.5805637358014304, "step": 690}, {"loss": 1.8415, "grad_norm": 0.3711601793766022, "learning_rate": 0.0002, "epoch": 0.5889777029869584, "step": 700}, {"loss": 1.7906, "grad_norm": 0.3311759829521179, "learning_rate": 0.0002, "epoch": 0.5973916701724863, "step": 710}, {"loss": 1.7712, "grad_norm": 0.34824270009994507, "learning_rate": 0.0002, "epoch": 0.6058056373580143, "step": 720}, {"loss": 1.7954, "grad_norm": 0.29668381810188293, "learning_rate": 0.0002, "epoch": 0.6142196045435423, "step": 730}, {"loss": 1.8321, "grad_norm": 0.36087489128112793, "learning_rate": 0.0002, "epoch": 0.6226335717290703, "step": 740}, {"loss": 1.7956, "grad_norm": 0.31590089201927185, "learning_rate": 0.0002, "epoch": 0.6310475389145982, "step": 750}, {"loss": 1.7343, "grad_norm": 0.37632957100868225, "learning_rate": 0.0002, "epoch": 0.6394615061001262, "step": 760}, {"loss": 1.8499, "grad_norm": 0.3360748589038849, "learning_rate": 0.0002, "epoch": 0.6478754732856542, "step": 770}, {"loss": 1.8076, "grad_norm": 0.3420640528202057, "learning_rate": 0.0002, "epoch": 0.6562894404711822, "step": 780}, {"loss": 1.8353, "grad_norm": 0.5734959244728088, "learning_rate": 0.0002, "epoch": 0.6647034076567101, "step": 790}, {"loss": 1.7746, "grad_norm": 0.36440837383270264, "learning_rate": 0.0002, "epoch": 0.6731173748422381, "step": 800}, {"loss": 1.7532, "grad_norm": 0.3179708421230316, "learning_rate": 0.0002, "epoch": 0.6815313420277661, "step": 810}, {"loss": 1.7815, "grad_norm": 0.34122881293296814, "learning_rate": 0.0002, "epoch": 0.6899453092132941, "step": 820}, {"loss": 1.8167, "grad_norm": 0.31886112689971924, "learning_rate": 0.0002, "epoch": 0.698359276398822, "step": 830}, {"loss": 1.7505, "grad_norm": 0.31782326102256775, "learning_rate": 0.0002, "epoch": 0.70677324358435, "step": 840}, {"loss": 1.7588, "grad_norm": 0.36052989959716797, "learning_rate": 0.0002, "epoch": 0.715187210769878, "step": 850}, {"loss": 1.7891, "grad_norm": 0.28946155309677124, "learning_rate": 0.0002, "epoch": 0.723601177955406, "step": 860}, {"loss": 1.7923, "grad_norm": 0.3095663785934448, "learning_rate": 0.0002, "epoch": 0.7320151451409339, "step": 870}, {"loss": 1.785, "grad_norm": 0.3317491412162781, "learning_rate": 0.0002, "epoch": 0.7404291123264619, "step": 880}, {"loss": 1.7709, "grad_norm": 0.31324660778045654, "learning_rate": 0.0002, "epoch": 0.7488430795119899, "step": 890}, {"loss": 1.8753, "grad_norm": 0.3290475606918335, "learning_rate": 0.0002, "epoch": 0.7572570466975179, "step": 900}, {"loss": 1.7679, "grad_norm": 0.35690343379974365, "learning_rate": 0.0002, "epoch": 0.7656710138830458, "step": 910}, {"loss": 1.826, "grad_norm": 0.39558273553848267, "learning_rate": 0.0002, "epoch": 0.7740849810685738, "step": 920}, {"loss": 1.8722, "grad_norm": 0.34254348278045654, "learning_rate": 0.0002, "epoch": 0.7824989482541018, "step": 930}, {"loss": 1.7603, "grad_norm": 0.3560165464878082, "learning_rate": 0.0002, "epoch": 0.7909129154396298, "step": 940}, {"loss": 1.7992, "grad_norm": 0.30693164467811584, "learning_rate": 0.0002, "epoch": 0.7993268826251577, "step": 950}, {"loss": 1.8029, "grad_norm": 0.3394823372364044, "learning_rate": 0.0002, "epoch": 0.8077408498106857, "step": 960}, {"loss": 1.8105, "grad_norm": 0.3741514980792999, "learning_rate": 0.0002, "epoch": 0.8161548169962137, "step": 970}, {"loss": 1.7849, "grad_norm": 0.3655228316783905, "learning_rate": 0.0002, "epoch": 0.8245687841817417, "step": 980}, {"loss": 1.8449, "grad_norm": 0.3586033880710602, "learning_rate": 0.0002, "epoch": 0.8329827513672696, "step": 990}, {"loss": 1.7033, "grad_norm": 0.3459678888320923, "learning_rate": 0.0002, "epoch": 0.8413967185527976, "step": 1000}, {"loss": 1.8498, "grad_norm": 0.3184349834918976, "learning_rate": 0.0002, "epoch": 0.8498106857383256, "step": 1010}, {"loss": 1.7632, "grad_norm": 0.3099786043167114, "learning_rate": 0.0002, "epoch": 0.8582246529238536, "step": 1020}, {"loss": 1.8067, "grad_norm": 0.30300915241241455, "learning_rate": 0.0002, "epoch": 0.8666386201093815, "step": 1030}, {"loss": 1.7923, "grad_norm": 0.3128705620765686, "learning_rate": 0.0002, "epoch": 0.8750525872949095, "step": 1040}, {"loss": 1.8252, "grad_norm": 0.3336263597011566, "learning_rate": 0.0002, "epoch": 0.8834665544804375, "step": 1050}, {"loss": 1.8375, "grad_norm": 0.3801328241825104, "learning_rate": 0.0002, "epoch": 0.8918805216659655, "step": 1060}, {"loss": 1.7757, "grad_norm": 0.3122096359729767, "learning_rate": 0.0002, "epoch": 0.9002944888514934, "step": 1070}, {"loss": 1.8251, "grad_norm": 0.35990869998931885, "learning_rate": 0.0002, "epoch": 0.9087084560370214, "step": 1080}, {"loss": 1.7343, "grad_norm": 0.3321819305419922, "learning_rate": 0.0002, "epoch": 0.9171224232225494, "step": 1090}, {"loss": 1.7595, "grad_norm": 0.4202139377593994, "learning_rate": 0.0002, "epoch": 0.9255363904080774, "step": 1100}, {"loss": 1.8056, "grad_norm": 0.32559722661972046, "learning_rate": 0.0002, "epoch": 0.9339503575936053, "step": 1110}, {"loss": 1.812, "grad_norm": 0.3098459839820862, "learning_rate": 0.0002, "epoch": 0.9423643247791333, "step": 1120}, {"loss": 1.8252, "grad_norm": 0.33917108178138733, "learning_rate": 0.0002, "epoch": 0.9507782919646613, "step": 1130}, {"loss": 1.7709, "grad_norm": 0.4055837094783783, "learning_rate": 0.0002, "epoch": 0.9591922591501894, "step": 1140}, {"loss": 1.8259, "grad_norm": 0.32508623600006104, "learning_rate": 0.0002, "epoch": 0.9676062263357172, "step": 1150}, {"loss": 1.782, "grad_norm": 0.30150601267814636, "learning_rate": 0.0002, "epoch": 0.9760201935212452, "step": 1160}, {"loss": 1.8291, "grad_norm": 0.3042563199996948, "learning_rate": 0.0002, "epoch": 0.9844341607067733, "step": 1170}, {"loss": 1.7847, "grad_norm": 0.33254584670066833, "learning_rate": 0.0002, "epoch": 0.9928481278923013, "step": 1180}, {"eval_loss": 1.8077726364135742, "eval_runtime": 38.4359, "eval_samples_per_second": 13.399, "eval_steps_per_second": 1.691, "epoch": 0.9995793016407236, "step": 1188}, {"loss": 1.7414, "grad_norm": 0.35073035955429077, "learning_rate": 0.0002, "epoch": 1.0012620950778293, "step": 1190}, {"loss": 1.7483, "grad_norm": 0.3217269778251648, "learning_rate": 0.0002, "epoch": 1.0096760622633572, "step": 1200}, {"loss": 1.7517, "grad_norm": 0.3635033369064331, "learning_rate": 0.0002, "epoch": 1.018090029448885, "step": 1210}, {"loss": 1.6949, "grad_norm": 0.32468414306640625, "learning_rate": 0.0002, "epoch": 1.0265039966344132, "step": 1220}, {"loss": 1.711, "grad_norm": 0.3307163417339325, "learning_rate": 0.0002, "epoch": 1.034917963819941, "step": 1230}, {"loss": 1.7881, "grad_norm": 0.34381359815597534, "learning_rate": 0.0002, "epoch": 1.0433319310054692, "step": 1240}, {"loss": 1.612, "grad_norm": 0.35874804854393005, "learning_rate": 0.0002, "epoch": 1.051745898190997, "step": 1250}, {"loss": 1.7314, "grad_norm": 0.3615919351577759, "learning_rate": 0.0002, "epoch": 1.060159865376525, "step": 1260}, {"loss": 1.7517, "grad_norm": 0.32835808396339417, "learning_rate": 0.0002, "epoch": 1.068573832562053, "step": 1270}, {"loss": 1.7193, "grad_norm": 0.3876388370990753, "learning_rate": 0.0002, "epoch": 1.076987799747581, "step": 1280}, {"loss": 1.7442, "grad_norm": 0.39895930886268616, "learning_rate": 0.0002, "epoch": 1.0854017669331089, "step": 1290}, {"loss": 1.6601, "grad_norm": 0.39081698656082153, "learning_rate": 0.0002, "epoch": 1.093815734118637, "step": 1300}, {"loss": 1.7623, "grad_norm": 0.39974215626716614, "learning_rate": 0.0002, "epoch": 1.1022297013041649, "step": 1310}, {"loss": 1.7506, "grad_norm": 0.3887332081794739, "learning_rate": 0.0002, "epoch": 1.110643668489693, "step": 1320}, {"loss": 1.7381, "grad_norm": 0.36216408014297485, "learning_rate": 0.0002, "epoch": 1.1190576356752209, "step": 1330}, {"loss": 1.762, "grad_norm": 0.36979028582572937, "learning_rate": 0.0002, "epoch": 1.1274716028607488, "step": 1340}, {"loss": 1.7515, "grad_norm": 0.34052133560180664, "learning_rate": 0.0002, "epoch": 1.1358855700462769, "step": 1350}, {"loss": 1.7513, "grad_norm": 0.3467716574668884, "learning_rate": 0.0002, "epoch": 1.1442995372318048, "step": 1360}, {"loss": 1.7086, "grad_norm": 0.35528799891471863, "learning_rate": 0.0002, "epoch": 1.1527135044173327, "step": 1370}, {"loss": 1.794, "grad_norm": 0.36282262206077576, "learning_rate": 0.0002, "epoch": 1.1611274716028608, "step": 1380}, {"loss": 1.7731, "grad_norm": 0.37355899810791016, "learning_rate": 0.0002, "epoch": 1.1695414387883887, "step": 1390}, {"loss": 1.7483, "grad_norm": 0.37292736768722534, "learning_rate": 0.0002, "epoch": 1.1779554059739168, "step": 1400}, {"loss": 1.6916, "grad_norm": 0.5892812013626099, "learning_rate": 0.0002, "epoch": 1.1863693731594447, "step": 1410}, {"loss": 1.7302, "grad_norm": 0.3712292015552521, "learning_rate": 0.0002, "epoch": 1.1947833403449726, "step": 1420}, {"loss": 1.7709, "grad_norm": 0.3349577486515045, "learning_rate": 0.0002, "epoch": 1.2031973075305007, "step": 1430}, {"loss": 1.7412, "grad_norm": 0.32591062784194946, "learning_rate": 0.0002, "epoch": 1.2116112747160286, "step": 1440}, {"loss": 1.7406, "grad_norm": 0.3840635418891907, "learning_rate": 0.0002, "epoch": 1.2200252419015567, "step": 1450}, {"loss": 1.7276, "grad_norm": 0.37238365411758423, "learning_rate": 0.0002, "epoch": 1.2284392090870846, "step": 1460}, {"loss": 1.7052, "grad_norm": 0.3731217682361603, "learning_rate": 0.0002, "epoch": 1.2368531762726125, "step": 1470}, {"loss": 1.7255, "grad_norm": 0.3318967819213867, "learning_rate": 0.0002, "epoch": 1.2452671434581406, "step": 1480}, {"loss": 1.7463, "grad_norm": 0.3784034848213196, "learning_rate": 0.0002, "epoch": 1.2536811106436685, "step": 1490}, {"loss": 1.6862, "grad_norm": 0.3541383147239685, "learning_rate": 0.0002, "epoch": 1.2620950778291964, "step": 1500}, {"loss": 1.8394, "grad_norm": 0.35312485694885254, "learning_rate": 0.0002, "epoch": 1.2705090450147245, "step": 1510}, {"loss": 1.7029, "grad_norm": 0.35272929072380066, "learning_rate": 0.0002, "epoch": 1.2789230122002524, "step": 1520}, {"loss": 1.7016, "grad_norm": 0.40988272428512573, "learning_rate": 0.0002, "epoch": 1.2873369793857803, "step": 1530}, {"loss": 1.6912, "grad_norm": 0.3543946146965027, "learning_rate": 0.0002, "epoch": 1.2957509465713084, "step": 1540}, {"loss": 1.6757, "grad_norm": 0.35639145970344543, "learning_rate": 0.0002, "epoch": 1.3041649137568363, "step": 1550}, {"loss": 1.6814, "grad_norm": 0.3290826678276062, "learning_rate": 0.0002, "epoch": 1.3125788809423642, "step": 1560}, {"loss": 1.7369, "grad_norm": 0.39264336228370667, "learning_rate": 0.0002, "epoch": 1.3209928481278923, "step": 1570}, {"loss": 1.6804, "grad_norm": 0.5390415191650391, "learning_rate": 0.0002, "epoch": 1.3294068153134202, "step": 1580}, {"loss": 1.708, "grad_norm": 0.5188116431236267, "learning_rate": 0.0002, "epoch": 1.3378207824989483, "step": 1590}, {"loss": 1.6763, "grad_norm": 0.37445148825645447, "learning_rate": 0.0002, "epoch": 1.3462347496844762, "step": 1600}, {"loss": 1.7386, "grad_norm": 0.3296085298061371, "learning_rate": 0.0002, "epoch": 1.3546487168700043, "step": 1610}, {"loss": 1.8107, "grad_norm": 0.39879581332206726, "learning_rate": 0.0002, "epoch": 1.3630626840555322, "step": 1620}, {"loss": 1.6744, "grad_norm": 0.36092764139175415, "learning_rate": 0.0002, "epoch": 1.37147665124106, "step": 1630}, {"loss": 1.7144, "grad_norm": 0.37011823058128357, "learning_rate": 0.0002, "epoch": 1.3798906184265882, "step": 1640}, {"loss": 1.7396, "grad_norm": 0.40863534808158875, "learning_rate": 0.0002, "epoch": 1.3883045856121161, "step": 1650}, {"loss": 1.7901, "grad_norm": 0.337001770734787, "learning_rate": 0.0002, "epoch": 1.396718552797644, "step": 1660}, {"loss": 1.7044, "grad_norm": 0.35596707463264465, "learning_rate": 0.0002, "epoch": 1.4051325199831721, "step": 1670}, {"loss": 1.7717, "grad_norm": 0.3857671916484833, "learning_rate": 0.0002, "epoch": 1.4135464871687, "step": 1680}, {"loss": 1.7015, "grad_norm": 0.419502317905426, "learning_rate": 0.0002, "epoch": 1.421960454354228, "step": 1690}, {"loss": 1.7261, "grad_norm": 0.35459452867507935, "learning_rate": 0.0002, "epoch": 1.430374421539756, "step": 1700}, {"loss": 1.7361, "grad_norm": 0.37246978282928467, "learning_rate": 0.0002, "epoch": 1.438788388725284, "step": 1710}, {"loss": 1.6762, "grad_norm": 0.33091893792152405, "learning_rate": 0.0002, "epoch": 1.4472023559108118, "step": 1720}, {"loss": 1.7044, "grad_norm": 0.37029674649238586, "learning_rate": 0.0002, "epoch": 1.45561632309634, "step": 1730}, {"loss": 1.7117, "grad_norm": 0.374025821685791, "learning_rate": 0.0002, "epoch": 1.4640302902818678, "step": 1740}, {"loss": 1.7549, "grad_norm": 0.3416315019130707, "learning_rate": 0.0002, "epoch": 1.472444257467396, "step": 1750}, {"loss": 1.7093, "grad_norm": 0.36502841114997864, "learning_rate": 0.0002, "epoch": 1.4808582246529238, "step": 1760}, {"loss": 1.6597, "grad_norm": 0.35458803176879883, "learning_rate": 0.0002, "epoch": 1.489272191838452, "step": 1770}, {"loss": 1.675, "grad_norm": 0.4462839663028717, "learning_rate": 0.0002, "epoch": 1.4976861590239798, "step": 1780}, {"loss": 1.7267, "grad_norm": 0.34836092591285706, "learning_rate": 0.0002, "epoch": 1.5061001262095077, "step": 1790}, {"loss": 1.7295, "grad_norm": 0.3445749282836914, "learning_rate": 0.0002, "epoch": 1.5145140933950358, "step": 1800}, {"loss": 1.7386, "grad_norm": 0.36012160778045654, "learning_rate": 0.0002, "epoch": 1.5229280605805637, "step": 1810}, {"loss": 1.6594, "grad_norm": 0.4052616059780121, "learning_rate": 0.0002, "epoch": 1.5313420277660916, "step": 1820}, {"loss": 1.72, "grad_norm": 0.3966905474662781, "learning_rate": 0.0002, "epoch": 1.5397559949516197, "step": 1830}, {"loss": 1.7595, "grad_norm": 0.35028719902038574, "learning_rate": 0.0002, "epoch": 1.5481699621371476, "step": 1840}, {"loss": 1.6829, "grad_norm": 0.3936742842197418, "learning_rate": 0.0002, "epoch": 1.5565839293226755, "step": 1850}, {"loss": 1.7579, "grad_norm": 0.34473296999931335, "learning_rate": 0.0002, "epoch": 1.5649978965082036, "step": 1860}, {"loss": 1.7207, "grad_norm": 0.4328365623950958, "learning_rate": 0.0002, "epoch": 1.5734118636937318, "step": 1870}, {"loss": 1.7098, "grad_norm": 0.3566315472126007, "learning_rate": 0.0002, "epoch": 1.5818258308792594, "step": 1880}, {"loss": 1.6095, "grad_norm": 0.3301256597042084, "learning_rate": 0.0002, "epoch": 1.5902397980647875, "step": 1890}, {"loss": 1.748, "grad_norm": 0.3743041455745697, "learning_rate": 0.0002, "epoch": 1.5986537652503157, "step": 1900}, {"loss": 1.7259, "grad_norm": 0.3735344707965851, "learning_rate": 0.0002, "epoch": 1.6070677324358436, "step": 1910}, {"loss": 1.7445, "grad_norm": 0.42191144824028015, "learning_rate": 0.0002, "epoch": 1.6154816996213714, "step": 1920}, {"loss": 1.6978, "grad_norm": 0.3787207305431366, "learning_rate": 0.0002, "epoch": 1.6238956668068996, "step": 1930}, {"loss": 1.6893, "grad_norm": 0.35647350549697876, "learning_rate": 0.0002, "epoch": 1.6323096339924275, "step": 1940}, {"loss": 1.7825, "grad_norm": 0.39791446924209595, "learning_rate": 0.0002, "epoch": 1.6407236011779553, "step": 1950}, {"loss": 1.7293, "grad_norm": 0.37341275811195374, "learning_rate": 0.0002, "epoch": 1.6491375683634835, "step": 1960}, {"loss": 1.6781, "grad_norm": 0.3722686469554901, "learning_rate": 0.0002, "epoch": 1.6575515355490114, "step": 1970}, {"loss": 1.6383, "grad_norm": 0.37467387318611145, "learning_rate": 0.0002, "epoch": 1.6659655027345392, "step": 1980}, {"loss": 1.7439, "grad_norm": 0.37109461426734924, "learning_rate": 0.0002, "epoch": 1.6743794699200674, "step": 1990}, {"loss": 1.7206, "grad_norm": 0.4008837044239044, "learning_rate": 0.0002, "epoch": 1.6827934371055953, "step": 2000}, {"loss": 1.7604, "grad_norm": 0.3316999673843384, "learning_rate": 0.0002, "epoch": 1.6912074042911232, "step": 2010}, {"loss": 1.7325, "grad_norm": 0.3683805465698242, "learning_rate": 0.0002, "epoch": 1.6996213714766513, "step": 2020}, {"loss": 1.7451, "grad_norm": 0.4163658320903778, "learning_rate": 0.0002, "epoch": 1.7080353386621794, "step": 2030}, {"loss": 1.741, "grad_norm": 0.4245431125164032, "learning_rate": 0.0002, "epoch": 1.716449305847707, "step": 2040}, {"loss": 1.7184, "grad_norm": 0.36732038855552673, "learning_rate": 0.0002, "epoch": 1.7248632730332352, "step": 2050}, {"loss": 1.7031, "grad_norm": 0.34981656074523926, "learning_rate": 0.0002, "epoch": 1.7332772402187633, "step": 2060}, {"loss": 1.7545, "grad_norm": 0.38588812947273254, "learning_rate": 0.0002, "epoch": 1.7416912074042912, "step": 2070}, {"loss": 1.7728, "grad_norm": 0.39914557337760925, "learning_rate": 0.0002, "epoch": 1.750105174589819, "step": 2080}, {"loss": 1.7049, "grad_norm": 0.36068692803382874, "learning_rate": 0.0002, "epoch": 1.7585191417753472, "step": 2090}, {"loss": 1.7537, "grad_norm": 0.3983287215232849, "learning_rate": 0.0002, "epoch": 1.766933108960875, "step": 2100}, {"loss": 1.7016, "grad_norm": 0.45008400082588196, "learning_rate": 0.0002, "epoch": 1.775347076146403, "step": 2110}, {"loss": 1.7163, "grad_norm": 0.3618052303791046, "learning_rate": 0.0002, "epoch": 1.783761043331931, "step": 2120}, {"loss": 1.7335, "grad_norm": 0.38745400309562683, "learning_rate": 0.0002, "epoch": 1.792175010517459, "step": 2130}, {"loss": 1.7387, "grad_norm": 0.3413826525211334, "learning_rate": 0.0002, "epoch": 1.8005889777029869, "step": 2140}, {"loss": 1.7414, "grad_norm": 0.35983747243881226, "learning_rate": 0.0002, "epoch": 1.809002944888515, "step": 2150}, {"loss": 1.7892, "grad_norm": 0.40926849842071533, "learning_rate": 0.0002, "epoch": 1.8174169120740429, "step": 2160}, {"loss": 1.6823, "grad_norm": 0.3543093800544739, "learning_rate": 0.0002, "epoch": 1.8258308792595708, "step": 2170}, {"loss": 1.7812, "grad_norm": 0.42690935730934143, "learning_rate": 0.0002, "epoch": 1.8342448464450989, "step": 2180}, {"loss": 1.7471, "grad_norm": 0.40282756090164185, "learning_rate": 0.0002, "epoch": 1.842658813630627, "step": 2190}, {"loss": 1.7411, "grad_norm": 0.36568400263786316, "learning_rate": 0.0002, "epoch": 1.8510727808161547, "step": 2200}, {"loss": 1.7024, "grad_norm": 0.43159013986587524, "learning_rate": 0.0002, "epoch": 1.8594867480016828, "step": 2210}, {"loss": 1.7298, "grad_norm": 0.3554118573665619, "learning_rate": 0.0002, "epoch": 1.867900715187211, "step": 2220}, {"loss": 1.7157, "grad_norm": 0.43349072337150574, "learning_rate": 0.0002, "epoch": 1.8763146823727388, "step": 2230}, {"loss": 1.7302, "grad_norm": 0.36486536264419556, "learning_rate": 0.0002, "epoch": 1.8847286495582667, "step": 2240}, {"loss": 1.6901, "grad_norm": 0.39260047674179077, "learning_rate": 0.0002, "epoch": 1.8931426167437948, "step": 2250}, {"loss": 1.6691, "grad_norm": 0.3741776943206787, "learning_rate": 0.0002, "epoch": 1.9015565839293227, "step": 2260}, {"loss": 1.6931, "grad_norm": 0.3961946964263916, "learning_rate": 0.0002, "epoch": 1.9099705511148506, "step": 2270}, {"loss": 1.737, "grad_norm": 0.3659731149673462, "learning_rate": 0.0002, "epoch": 1.9183845183003787, "step": 2280}, {"loss": 1.7342, "grad_norm": 0.34744107723236084, "learning_rate": 0.0002, "epoch": 1.9267984854859066, "step": 2290}, {"loss": 1.7162, "grad_norm": 0.3607442378997803, "learning_rate": 0.0002, "epoch": 1.9352124526714345, "step": 2300}, {"loss": 1.6673, "grad_norm": 0.331464558839798, "learning_rate": 0.0002, "epoch": 1.9436264198569626, "step": 2310}, {"loss": 1.7101, "grad_norm": 0.3904414474964142, "learning_rate": 0.0002, "epoch": 1.9520403870424905, "step": 2320}, {"loss": 1.7327, "grad_norm": 0.37584832310676575, "learning_rate": 0.0002, "epoch": 1.9604543542280184, "step": 2330}, {"loss": 1.7586, "grad_norm": 0.3698684275150299, "learning_rate": 0.0002, "epoch": 1.9688683214135465, "step": 2340}, {"loss": 1.7764, "grad_norm": 0.40571412444114685, "learning_rate": 0.0002, "epoch": 1.9772822885990746, "step": 2350}, {"loss": 1.744, "grad_norm": 0.40059587359428406, "learning_rate": 0.0002, "epoch": 1.9856962557846023, "step": 2360}, {"loss": 1.7033, "grad_norm": 0.4168248474597931, "learning_rate": 0.0002, "epoch": 1.9941102229701304, "step": 2370}]}
+{"epoch": 2.9995793016407237, "step": 3565, "epoch_duration": 1456.9628908634186, "total_accumulated_duration": 4765.486236572266, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "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_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-0.8-num-5623-sd-42/checkpoint-2377", "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": 2.56, "grad_norm": 0.5458821654319763, "learning_rate": 0.0002, "epoch": 0.008413967185527976, "step": 10}, {"loss": 2.3235, "grad_norm": 0.7293308973312378, "learning_rate": 0.0002, "epoch": 0.016827934371055953, "step": 20}, {"loss": 2.0815, "grad_norm": 0.47792306542396545, "learning_rate": 0.0002, "epoch": 0.02524190155658393, "step": 30}, {"loss": 1.9718, "grad_norm": 0.5944402813911438, "learning_rate": 0.0002, "epoch": 0.033655868742111905, "step": 40}, {"loss": 1.8848, "grad_norm": 0.5415359735488892, "learning_rate": 0.0002, "epoch": 0.04206983592763988, "step": 50}, {"loss": 1.8953, "grad_norm": 0.535713791847229, "learning_rate": 0.0002, "epoch": 0.05048380311316786, "step": 60}, {"loss": 1.937, "grad_norm": 0.5184146761894226, "learning_rate": 0.0002, "epoch": 0.058897770298695834, "step": 70}, {"loss": 1.8396, "grad_norm": 0.458926796913147, "learning_rate": 0.0002, "epoch": 0.06731173748422381, "step": 80}, {"loss": 1.8677, "grad_norm": 0.4780142307281494, "learning_rate": 0.0002, "epoch": 0.07572570466975179, "step": 90}, {"loss": 1.8593, "grad_norm": 0.79965740442276, "learning_rate": 0.0002, "epoch": 0.08413967185527976, "step": 100}, {"loss": 1.9081, "grad_norm": 0.4498862028121948, "learning_rate": 0.0002, "epoch": 0.09255363904080774, "step": 110}, {"loss": 1.8503, "grad_norm": 0.39338430762290955, "learning_rate": 0.0002, "epoch": 0.10096760622633572, "step": 120}, {"loss": 1.8637, "grad_norm": 0.9588953852653503, "learning_rate": 0.0002, "epoch": 0.10938157341186369, "step": 130}, {"loss": 1.8676, "grad_norm": 0.41675639152526855, "learning_rate": 0.0002, "epoch": 0.11779554059739167, "step": 140}, {"loss": 1.8904, "grad_norm": 0.44519832730293274, "learning_rate": 0.0002, "epoch": 0.12620950778291964, "step": 150}, {"loss": 1.798, "grad_norm": 0.4176260530948639, "learning_rate": 0.0002, "epoch": 0.13462347496844762, "step": 160}, {"loss": 1.8398, "grad_norm": 0.35840365290641785, "learning_rate": 0.0002, "epoch": 0.1430374421539756, "step": 170}, {"loss": 1.8666, "grad_norm": 0.3794495463371277, "learning_rate": 0.0002, "epoch": 0.15145140933950357, "step": 180}, {"loss": 1.8111, "grad_norm": 0.4563522934913635, "learning_rate": 0.0002, "epoch": 0.15986537652503155, "step": 190}, {"loss": 1.8893, "grad_norm": 0.37057486176490784, "learning_rate": 0.0002, "epoch": 0.16827934371055953, "step": 200}, {"loss": 1.7995, "grad_norm": 0.44081518054008484, "learning_rate": 0.0002, "epoch": 0.1766933108960875, "step": 210}, {"loss": 1.9048, "grad_norm": 0.46078577637672424, "learning_rate": 0.0002, "epoch": 0.18510727808161548, "step": 220}, {"loss": 1.8403, "grad_norm": 0.36132094264030457, "learning_rate": 0.0002, "epoch": 0.19352124526714345, "step": 230}, {"loss": 1.8827, "grad_norm": 0.3747289180755615, "learning_rate": 0.0002, "epoch": 0.20193521245267143, "step": 240}, {"loss": 1.8382, "grad_norm": 0.3540179133415222, "learning_rate": 0.0002, "epoch": 0.2103491796381994, "step": 250}, {"loss": 1.8196, "grad_norm": 0.3461375832557678, "learning_rate": 0.0002, "epoch": 0.21876314682372738, "step": 260}, {"loss": 1.8509, "grad_norm": 0.3436960279941559, "learning_rate": 0.0002, "epoch": 0.22717711400925536, "step": 270}, {"loss": 1.8285, "grad_norm": 0.35403719544410706, "learning_rate": 0.0002, "epoch": 0.23559108119478334, "step": 280}, {"loss": 1.8369, "grad_norm": 0.37142616510391235, "learning_rate": 0.0002, "epoch": 0.2440050483803113, "step": 290}, {"loss": 1.8044, "grad_norm": 0.3307955861091614, "learning_rate": 0.0002, "epoch": 0.2524190155658393, "step": 300}, {"loss": 1.817, "grad_norm": 0.32855314016342163, "learning_rate": 0.0002, "epoch": 0.2608329827513673, "step": 310}, {"loss": 1.7803, "grad_norm": 0.3299003839492798, "learning_rate": 0.0002, "epoch": 0.26924694993689524, "step": 320}, {"loss": 1.8129, "grad_norm": 0.44311287999153137, "learning_rate": 0.0002, "epoch": 0.27766091712242325, "step": 330}, {"loss": 1.8232, "grad_norm": 0.32989758253097534, "learning_rate": 0.0002, "epoch": 0.2860748843079512, "step": 340}, {"loss": 1.7716, "grad_norm": 0.34400200843811035, "learning_rate": 0.0002, "epoch": 0.2944888514934792, "step": 350}, {"loss": 1.7619, "grad_norm": 0.36286211013793945, "learning_rate": 0.0002, "epoch": 0.30290281867900715, "step": 360}, {"loss": 1.8025, "grad_norm": 0.406827837228775, "learning_rate": 0.0002, "epoch": 0.31131678586453515, "step": 370}, {"loss": 1.7515, "grad_norm": 0.36299195885658264, "learning_rate": 0.0002, "epoch": 0.3197307530500631, "step": 380}, {"loss": 1.837, "grad_norm": 0.3477257192134857, "learning_rate": 0.0002, "epoch": 0.3281447202355911, "step": 390}, {"loss": 1.7767, "grad_norm": 0.3730369210243225, "learning_rate": 0.0002, "epoch": 0.33655868742111905, "step": 400}, {"loss": 1.7747, "grad_norm": 0.4644559919834137, "learning_rate": 0.0002, "epoch": 0.34497265460664706, "step": 410}, {"loss": 1.7538, "grad_norm": 0.406576544046402, "learning_rate": 0.0002, "epoch": 0.353386621792175, "step": 420}, {"loss": 1.7501, "grad_norm": 0.3612699508666992, "learning_rate": 0.0002, "epoch": 0.361800588977703, "step": 430}, {"loss": 1.7473, "grad_norm": 0.3243742287158966, "learning_rate": 0.0002, "epoch": 0.37021455616323096, "step": 440}, {"loss": 1.8851, "grad_norm": 0.36671221256256104, "learning_rate": 0.0002, "epoch": 0.37862852334875896, "step": 450}, {"loss": 1.8853, "grad_norm": 0.3565002381801605, "learning_rate": 0.0002, "epoch": 0.3870424905342869, "step": 460}, {"loss": 1.8923, "grad_norm": 0.34630221128463745, "learning_rate": 0.0002, "epoch": 0.3954564577198149, "step": 470}, {"loss": 1.8234, "grad_norm": 0.3353537321090698, "learning_rate": 0.0002, "epoch": 0.40387042490534286, "step": 480}, {"loss": 1.7135, "grad_norm": 0.4015921950340271, "learning_rate": 0.0002, "epoch": 0.41228439209087087, "step": 490}, {"loss": 1.7815, "grad_norm": 0.5489419102668762, "learning_rate": 0.0002, "epoch": 0.4206983592763988, "step": 500}, {"loss": 1.7903, "grad_norm": 0.4193589985370636, "learning_rate": 0.0002, "epoch": 0.4291123264619268, "step": 510}, {"loss": 1.8416, "grad_norm": 0.3418922424316406, "learning_rate": 0.0002, "epoch": 0.43752629364745477, "step": 520}, {"loss": 1.7982, "grad_norm": 0.32668185234069824, "learning_rate": 0.0002, "epoch": 0.44594026083298277, "step": 530}, {"loss": 1.7501, "grad_norm": 0.3094325661659241, "learning_rate": 0.0002, "epoch": 0.4543542280185107, "step": 540}, {"loss": 1.7438, "grad_norm": 0.3743017315864563, "learning_rate": 0.0002, "epoch": 0.4627681952040387, "step": 550}, {"loss": 1.8451, "grad_norm": 0.3295630216598511, "learning_rate": 0.0002, "epoch": 0.47118216238956667, "step": 560}, {"loss": 1.7529, "grad_norm": 1.6124513149261475, "learning_rate": 0.0002, "epoch": 0.4795961295750947, "step": 570}, {"loss": 1.8028, "grad_norm": 0.3245585858821869, "learning_rate": 0.0002, "epoch": 0.4880100967606226, "step": 580}, {"loss": 1.7976, "grad_norm": 0.3332934081554413, "learning_rate": 0.0002, "epoch": 0.49642406394615063, "step": 590}, {"loss": 1.7912, "grad_norm": 0.3836138844490051, "learning_rate": 0.0002, "epoch": 0.5048380311316786, "step": 600}, {"loss": 1.8347, "grad_norm": 0.32953888177871704, "learning_rate": 0.0002, "epoch": 0.5132519983172066, "step": 610}, {"loss": 1.7729, "grad_norm": 0.36291512846946716, "learning_rate": 0.0002, "epoch": 0.5216659655027346, "step": 620}, {"loss": 1.7758, "grad_norm": 0.3237783908843994, "learning_rate": 0.0002, "epoch": 0.5300799326882625, "step": 630}, {"loss": 1.8352, "grad_norm": 0.38882696628570557, "learning_rate": 0.0002, "epoch": 0.5384938998737905, "step": 640}, {"loss": 1.8624, "grad_norm": 0.37821972370147705, "learning_rate": 0.0002, "epoch": 0.5469078670593185, "step": 650}, {"loss": 1.8075, "grad_norm": 0.3556285500526428, "learning_rate": 0.0002, "epoch": 0.5553218342448465, "step": 660}, {"loss": 1.778, "grad_norm": 0.347499281167984, "learning_rate": 0.0002, "epoch": 0.5637358014303744, "step": 670}, {"loss": 1.8066, "grad_norm": 0.3176489472389221, "learning_rate": 0.0002, "epoch": 0.5721497686159024, "step": 680}, {"loss": 1.7257, "grad_norm": 0.30220088362693787, "learning_rate": 0.0002, "epoch": 0.5805637358014304, "step": 690}, {"loss": 1.8415, "grad_norm": 0.3711601793766022, "learning_rate": 0.0002, "epoch": 0.5889777029869584, "step": 700}, {"loss": 1.7906, "grad_norm": 0.3311759829521179, "learning_rate": 0.0002, "epoch": 0.5973916701724863, "step": 710}, {"loss": 1.7712, "grad_norm": 0.34824270009994507, "learning_rate": 0.0002, "epoch": 0.6058056373580143, "step": 720}, {"loss": 1.7954, "grad_norm": 0.29668381810188293, "learning_rate": 0.0002, "epoch": 0.6142196045435423, "step": 730}, {"loss": 1.8321, "grad_norm": 0.36087489128112793, "learning_rate": 0.0002, "epoch": 0.6226335717290703, "step": 740}, {"loss": 1.7956, "grad_norm": 0.31590089201927185, "learning_rate": 0.0002, "epoch": 0.6310475389145982, "step": 750}, {"loss": 1.7343, "grad_norm": 0.37632957100868225, "learning_rate": 0.0002, "epoch": 0.6394615061001262, "step": 760}, {"loss": 1.8499, "grad_norm": 0.3360748589038849, "learning_rate": 0.0002, "epoch": 0.6478754732856542, "step": 770}, {"loss": 1.8076, "grad_norm": 0.3420640528202057, "learning_rate": 0.0002, "epoch": 0.6562894404711822, "step": 780}, {"loss": 1.8353, "grad_norm": 0.5734959244728088, "learning_rate": 0.0002, "epoch": 0.6647034076567101, "step": 790}, {"loss": 1.7746, "grad_norm": 0.36440837383270264, "learning_rate": 0.0002, "epoch": 0.6731173748422381, "step": 800}, {"loss": 1.7532, "grad_norm": 0.3179708421230316, "learning_rate": 0.0002, "epoch": 0.6815313420277661, "step": 810}, {"loss": 1.7815, "grad_norm": 0.34122881293296814, "learning_rate": 0.0002, "epoch": 0.6899453092132941, "step": 820}, {"loss": 1.8167, "grad_norm": 0.31886112689971924, "learning_rate": 0.0002, "epoch": 0.698359276398822, "step": 830}, {"loss": 1.7505, "grad_norm": 0.31782326102256775, "learning_rate": 0.0002, "epoch": 0.70677324358435, "step": 840}, {"loss": 1.7588, "grad_norm": 0.36052989959716797, "learning_rate": 0.0002, "epoch": 0.715187210769878, "step": 850}, {"loss": 1.7891, "grad_norm": 0.28946155309677124, "learning_rate": 0.0002, "epoch": 0.723601177955406, "step": 860}, {"loss": 1.7923, "grad_norm": 0.3095663785934448, "learning_rate": 0.0002, "epoch": 0.7320151451409339, "step": 870}, {"loss": 1.785, "grad_norm": 0.3317491412162781, "learning_rate": 0.0002, "epoch": 0.7404291123264619, "step": 880}, {"loss": 1.7709, "grad_norm": 0.31324660778045654, "learning_rate": 0.0002, "epoch": 0.7488430795119899, "step": 890}, {"loss": 1.8753, "grad_norm": 0.3290475606918335, "learning_rate": 0.0002, "epoch": 0.7572570466975179, "step": 900}, {"loss": 1.7679, "grad_norm": 0.35690343379974365, "learning_rate": 0.0002, "epoch": 0.7656710138830458, "step": 910}, {"loss": 1.826, "grad_norm": 0.39558273553848267, "learning_rate": 0.0002, "epoch": 0.7740849810685738, "step": 920}, {"loss": 1.8722, "grad_norm": 0.34254348278045654, "learning_rate": 0.0002, "epoch": 0.7824989482541018, "step": 930}, {"loss": 1.7603, "grad_norm": 0.3560165464878082, "learning_rate": 0.0002, "epoch": 0.7909129154396298, "step": 940}, {"loss": 1.7992, "grad_norm": 0.30693164467811584, "learning_rate": 0.0002, "epoch": 0.7993268826251577, "step": 950}, {"loss": 1.8029, "grad_norm": 0.3394823372364044, "learning_rate": 0.0002, "epoch": 0.8077408498106857, "step": 960}, {"loss": 1.8105, "grad_norm": 0.3741514980792999, "learning_rate": 0.0002, "epoch": 0.8161548169962137, "step": 970}, {"loss": 1.7849, "grad_norm": 0.3655228316783905, "learning_rate": 0.0002, "epoch": 0.8245687841817417, "step": 980}, {"loss": 1.8449, "grad_norm": 0.3586033880710602, "learning_rate": 0.0002, "epoch": 0.8329827513672696, "step": 990}, {"loss": 1.7033, "grad_norm": 0.3459678888320923, "learning_rate": 0.0002, "epoch": 0.8413967185527976, "step": 1000}, {"loss": 1.8498, "grad_norm": 0.3184349834918976, "learning_rate": 0.0002, "epoch": 0.8498106857383256, "step": 1010}, {"loss": 1.7632, "grad_norm": 0.3099786043167114, "learning_rate": 0.0002, "epoch": 0.8582246529238536, "step": 1020}, {"loss": 1.8067, "grad_norm": 0.30300915241241455, "learning_rate": 0.0002, "epoch": 0.8666386201093815, "step": 1030}, {"loss": 1.7923, "grad_norm": 0.3128705620765686, "learning_rate": 0.0002, "epoch": 0.8750525872949095, "step": 1040}, {"loss": 1.8252, "grad_norm": 0.3336263597011566, "learning_rate": 0.0002, "epoch": 0.8834665544804375, "step": 1050}, {"loss": 1.8375, "grad_norm": 0.3801328241825104, "learning_rate": 0.0002, "epoch": 0.8918805216659655, "step": 1060}, {"loss": 1.7757, "grad_norm": 0.3122096359729767, "learning_rate": 0.0002, "epoch": 0.9002944888514934, "step": 1070}, {"loss": 1.8251, "grad_norm": 0.35990869998931885, "learning_rate": 0.0002, "epoch": 0.9087084560370214, "step": 1080}, {"loss": 1.7343, "grad_norm": 0.3321819305419922, "learning_rate": 0.0002, "epoch": 0.9171224232225494, "step": 1090}, {"loss": 1.7595, "grad_norm": 0.4202139377593994, "learning_rate": 0.0002, "epoch": 0.9255363904080774, "step": 1100}, {"loss": 1.8056, "grad_norm": 0.32559722661972046, "learning_rate": 0.0002, "epoch": 0.9339503575936053, "step": 1110}, {"loss": 1.812, "grad_norm": 0.3098459839820862, "learning_rate": 0.0002, "epoch": 0.9423643247791333, "step": 1120}, {"loss": 1.8252, "grad_norm": 0.33917108178138733, "learning_rate": 0.0002, "epoch": 0.9507782919646613, "step": 1130}, {"loss": 1.7709, "grad_norm": 0.4055837094783783, "learning_rate": 0.0002, "epoch": 0.9591922591501894, "step": 1140}, {"loss": 1.8259, "grad_norm": 0.32508623600006104, "learning_rate": 0.0002, "epoch": 0.9676062263357172, "step": 1150}, {"loss": 1.782, "grad_norm": 0.30150601267814636, "learning_rate": 0.0002, "epoch": 0.9760201935212452, "step": 1160}, {"loss": 1.8291, "grad_norm": 0.3042563199996948, "learning_rate": 0.0002, "epoch": 0.9844341607067733, "step": 1170}, {"loss": 1.7847, "grad_norm": 0.33254584670066833, "learning_rate": 0.0002, "epoch": 0.9928481278923013, "step": 1180}, {"eval_loss": 1.8077726364135742, "eval_runtime": 38.4359, "eval_samples_per_second": 13.399, "eval_steps_per_second": 1.691, "epoch": 0.9995793016407236, "step": 1188}, {"loss": 1.7414, "grad_norm": 0.35073035955429077, "learning_rate": 0.0002, "epoch": 1.0012620950778293, "step": 1190}, {"loss": 1.7483, "grad_norm": 0.3217269778251648, "learning_rate": 0.0002, "epoch": 1.0096760622633572, "step": 1200}, {"loss": 1.7517, "grad_norm": 0.3635033369064331, "learning_rate": 0.0002, "epoch": 1.018090029448885, "step": 1210}, {"loss": 1.6949, "grad_norm": 0.32468414306640625, "learning_rate": 0.0002, "epoch": 1.0265039966344132, "step": 1220}, {"loss": 1.711, "grad_norm": 0.3307163417339325, "learning_rate": 0.0002, "epoch": 1.034917963819941, "step": 1230}, {"loss": 1.7881, "grad_norm": 0.34381359815597534, "learning_rate": 0.0002, "epoch": 1.0433319310054692, "step": 1240}, {"loss": 1.612, "grad_norm": 0.35874804854393005, "learning_rate": 0.0002, "epoch": 1.051745898190997, "step": 1250}, {"loss": 1.7314, "grad_norm": 0.3615919351577759, "learning_rate": 0.0002, "epoch": 1.060159865376525, "step": 1260}, {"loss": 1.7517, "grad_norm": 0.32835808396339417, "learning_rate": 0.0002, "epoch": 1.068573832562053, "step": 1270}, {"loss": 1.7193, "grad_norm": 0.3876388370990753, "learning_rate": 0.0002, "epoch": 1.076987799747581, "step": 1280}, {"loss": 1.7442, "grad_norm": 0.39895930886268616, "learning_rate": 0.0002, "epoch": 1.0854017669331089, "step": 1290}, {"loss": 1.6601, "grad_norm": 0.39081698656082153, "learning_rate": 0.0002, "epoch": 1.093815734118637, "step": 1300}, {"loss": 1.7623, "grad_norm": 0.39974215626716614, "learning_rate": 0.0002, "epoch": 1.1022297013041649, "step": 1310}, {"loss": 1.7506, "grad_norm": 0.3887332081794739, "learning_rate": 0.0002, "epoch": 1.110643668489693, "step": 1320}, {"loss": 1.7381, "grad_norm": 0.36216408014297485, "learning_rate": 0.0002, "epoch": 1.1190576356752209, "step": 1330}, {"loss": 1.762, "grad_norm": 0.36979028582572937, "learning_rate": 0.0002, "epoch": 1.1274716028607488, "step": 1340}, {"loss": 1.7515, "grad_norm": 0.34052133560180664, "learning_rate": 0.0002, "epoch": 1.1358855700462769, "step": 1350}, {"loss": 1.7513, "grad_norm": 0.3467716574668884, "learning_rate": 0.0002, "epoch": 1.1442995372318048, "step": 1360}, {"loss": 1.7086, "grad_norm": 0.35528799891471863, "learning_rate": 0.0002, "epoch": 1.1527135044173327, "step": 1370}, {"loss": 1.794, "grad_norm": 0.36282262206077576, "learning_rate": 0.0002, "epoch": 1.1611274716028608, "step": 1380}, {"loss": 1.7731, "grad_norm": 0.37355899810791016, "learning_rate": 0.0002, "epoch": 1.1695414387883887, "step": 1390}, {"loss": 1.7483, "grad_norm": 0.37292736768722534, "learning_rate": 0.0002, "epoch": 1.1779554059739168, "step": 1400}, {"loss": 1.6916, "grad_norm": 0.5892812013626099, "learning_rate": 0.0002, "epoch": 1.1863693731594447, "step": 1410}, {"loss": 1.7302, "grad_norm": 0.3712292015552521, "learning_rate": 0.0002, "epoch": 1.1947833403449726, "step": 1420}, {"loss": 1.7709, "grad_norm": 0.3349577486515045, "learning_rate": 0.0002, "epoch": 1.2031973075305007, "step": 1430}, {"loss": 1.7412, "grad_norm": 0.32591062784194946, "learning_rate": 0.0002, "epoch": 1.2116112747160286, "step": 1440}, {"loss": 1.7406, "grad_norm": 0.3840635418891907, "learning_rate": 0.0002, "epoch": 1.2200252419015567, "step": 1450}, {"loss": 1.7276, "grad_norm": 0.37238365411758423, "learning_rate": 0.0002, "epoch": 1.2284392090870846, "step": 1460}, {"loss": 1.7052, "grad_norm": 0.3731217682361603, "learning_rate": 0.0002, "epoch": 1.2368531762726125, "step": 1470}, {"loss": 1.7255, "grad_norm": 0.3318967819213867, "learning_rate": 0.0002, "epoch": 1.2452671434581406, "step": 1480}, {"loss": 1.7463, "grad_norm": 0.3784034848213196, "learning_rate": 0.0002, "epoch": 1.2536811106436685, "step": 1490}, {"loss": 1.6862, "grad_norm": 0.3541383147239685, "learning_rate": 0.0002, "epoch": 1.2620950778291964, "step": 1500}, {"loss": 1.8394, "grad_norm": 0.35312485694885254, "learning_rate": 0.0002, "epoch": 1.2705090450147245, "step": 1510}, {"loss": 1.7029, "grad_norm": 0.35272929072380066, "learning_rate": 0.0002, "epoch": 1.2789230122002524, "step": 1520}, {"loss": 1.7016, "grad_norm": 0.40988272428512573, "learning_rate": 0.0002, "epoch": 1.2873369793857803, "step": 1530}, {"loss": 1.6912, "grad_norm": 0.3543946146965027, "learning_rate": 0.0002, "epoch": 1.2957509465713084, "step": 1540}, {"loss": 1.6757, "grad_norm": 0.35639145970344543, "learning_rate": 0.0002, "epoch": 1.3041649137568363, "step": 1550}, {"loss": 1.6814, "grad_norm": 0.3290826678276062, "learning_rate": 0.0002, "epoch": 1.3125788809423642, "step": 1560}, {"loss": 1.7369, "grad_norm": 0.39264336228370667, "learning_rate": 0.0002, "epoch": 1.3209928481278923, "step": 1570}, {"loss": 1.6804, "grad_norm": 0.5390415191650391, "learning_rate": 0.0002, "epoch": 1.3294068153134202, "step": 1580}, {"loss": 1.708, "grad_norm": 0.5188116431236267, "learning_rate": 0.0002, "epoch": 1.3378207824989483, "step": 1590}, {"loss": 1.6763, "grad_norm": 0.37445148825645447, "learning_rate": 0.0002, "epoch": 1.3462347496844762, "step": 1600}, {"loss": 1.7386, "grad_norm": 0.3296085298061371, "learning_rate": 0.0002, "epoch": 1.3546487168700043, "step": 1610}, {"loss": 1.8107, "grad_norm": 0.39879581332206726, "learning_rate": 0.0002, "epoch": 1.3630626840555322, "step": 1620}, {"loss": 1.6744, "grad_norm": 0.36092764139175415, "learning_rate": 0.0002, "epoch": 1.37147665124106, "step": 1630}, {"loss": 1.7144, "grad_norm": 0.37011823058128357, "learning_rate": 0.0002, "epoch": 1.3798906184265882, "step": 1640}, {"loss": 1.7396, "grad_norm": 0.40863534808158875, "learning_rate": 0.0002, "epoch": 1.3883045856121161, "step": 1650}, {"loss": 1.7901, "grad_norm": 0.337001770734787, "learning_rate": 0.0002, "epoch": 1.396718552797644, "step": 1660}, {"loss": 1.7044, "grad_norm": 0.35596707463264465, "learning_rate": 0.0002, "epoch": 1.4051325199831721, "step": 1670}, {"loss": 1.7717, "grad_norm": 0.3857671916484833, "learning_rate": 0.0002, "epoch": 1.4135464871687, "step": 1680}, {"loss": 1.7015, "grad_norm": 0.419502317905426, "learning_rate": 0.0002, "epoch": 1.421960454354228, "step": 1690}, {"loss": 1.7261, "grad_norm": 0.35459452867507935, "learning_rate": 0.0002, "epoch": 1.430374421539756, "step": 1700}, {"loss": 1.7361, "grad_norm": 0.37246978282928467, "learning_rate": 0.0002, "epoch": 1.438788388725284, "step": 1710}, {"loss": 1.6762, "grad_norm": 0.33091893792152405, "learning_rate": 0.0002, "epoch": 1.4472023559108118, "step": 1720}, {"loss": 1.7044, "grad_norm": 0.37029674649238586, "learning_rate": 0.0002, "epoch": 1.45561632309634, "step": 1730}, {"loss": 1.7117, "grad_norm": 0.374025821685791, "learning_rate": 0.0002, "epoch": 1.4640302902818678, "step": 1740}, {"loss": 1.7549, "grad_norm": 0.3416315019130707, "learning_rate": 0.0002, "epoch": 1.472444257467396, "step": 1750}, {"loss": 1.7093, "grad_norm": 0.36502841114997864, "learning_rate": 0.0002, "epoch": 1.4808582246529238, "step": 1760}, {"loss": 1.6597, "grad_norm": 0.35458803176879883, "learning_rate": 0.0002, "epoch": 1.489272191838452, "step": 1770}, {"loss": 1.675, "grad_norm": 0.4462839663028717, "learning_rate": 0.0002, "epoch": 1.4976861590239798, "step": 1780}, {"loss": 1.7267, "grad_norm": 0.34836092591285706, "learning_rate": 0.0002, "epoch": 1.5061001262095077, "step": 1790}, {"loss": 1.7295, "grad_norm": 0.3445749282836914, "learning_rate": 0.0002, "epoch": 1.5145140933950358, "step": 1800}, {"loss": 1.7386, "grad_norm": 0.36012160778045654, "learning_rate": 0.0002, "epoch": 1.5229280605805637, "step": 1810}, {"loss": 1.6594, "grad_norm": 0.4052616059780121, "learning_rate": 0.0002, "epoch": 1.5313420277660916, "step": 1820}, {"loss": 1.72, "grad_norm": 0.3966905474662781, "learning_rate": 0.0002, "epoch": 1.5397559949516197, "step": 1830}, {"loss": 1.7595, "grad_norm": 0.35028719902038574, "learning_rate": 0.0002, "epoch": 1.5481699621371476, "step": 1840}, {"loss": 1.6829, "grad_norm": 0.3936742842197418, "learning_rate": 0.0002, "epoch": 1.5565839293226755, "step": 1850}, {"loss": 1.7579, "grad_norm": 0.34473296999931335, "learning_rate": 0.0002, "epoch": 1.5649978965082036, "step": 1860}, {"loss": 1.7207, "grad_norm": 0.4328365623950958, "learning_rate": 0.0002, "epoch": 1.5734118636937318, "step": 1870}, {"loss": 1.7098, "grad_norm": 0.3566315472126007, "learning_rate": 0.0002, "epoch": 1.5818258308792594, "step": 1880}, {"loss": 1.6095, "grad_norm": 0.3301256597042084, "learning_rate": 0.0002, "epoch": 1.5902397980647875, "step": 1890}, {"loss": 1.748, "grad_norm": 0.3743041455745697, "learning_rate": 0.0002, "epoch": 1.5986537652503157, "step": 1900}, {"loss": 1.7259, "grad_norm": 0.3735344707965851, "learning_rate": 0.0002, "epoch": 1.6070677324358436, "step": 1910}, {"loss": 1.7445, "grad_norm": 0.42191144824028015, "learning_rate": 0.0002, "epoch": 1.6154816996213714, "step": 1920}, {"loss": 1.6978, "grad_norm": 0.3787207305431366, "learning_rate": 0.0002, "epoch": 1.6238956668068996, "step": 1930}, {"loss": 1.6893, "grad_norm": 0.35647350549697876, "learning_rate": 0.0002, "epoch": 1.6323096339924275, "step": 1940}, {"loss": 1.7825, "grad_norm": 0.39791446924209595, "learning_rate": 0.0002, "epoch": 1.6407236011779553, "step": 1950}, {"loss": 1.7293, "grad_norm": 0.37341275811195374, "learning_rate": 0.0002, "epoch": 1.6491375683634835, "step": 1960}, {"loss": 1.6781, "grad_norm": 0.3722686469554901, "learning_rate": 0.0002, "epoch": 1.6575515355490114, "step": 1970}, {"loss": 1.6383, "grad_norm": 0.37467387318611145, "learning_rate": 0.0002, "epoch": 1.6659655027345392, "step": 1980}, {"loss": 1.7439, "grad_norm": 0.37109461426734924, "learning_rate": 0.0002, "epoch": 1.6743794699200674, "step": 1990}, {"loss": 1.7206, "grad_norm": 0.4008837044239044, "learning_rate": 0.0002, "epoch": 1.6827934371055953, "step": 2000}, {"loss": 1.7604, "grad_norm": 0.3316999673843384, "learning_rate": 0.0002, "epoch": 1.6912074042911232, "step": 2010}, {"loss": 1.7325, "grad_norm": 0.3683805465698242, "learning_rate": 0.0002, "epoch": 1.6996213714766513, "step": 2020}, {"loss": 1.7451, "grad_norm": 0.4163658320903778, "learning_rate": 0.0002, "epoch": 1.7080353386621794, "step": 2030}, {"loss": 1.741, "grad_norm": 0.4245431125164032, "learning_rate": 0.0002, "epoch": 1.716449305847707, "step": 2040}, {"loss": 1.7184, "grad_norm": 0.36732038855552673, "learning_rate": 0.0002, "epoch": 1.7248632730332352, "step": 2050}, {"loss": 1.7031, "grad_norm": 0.34981656074523926, "learning_rate": 0.0002, "epoch": 1.7332772402187633, "step": 2060}, {"loss": 1.7545, "grad_norm": 0.38588812947273254, "learning_rate": 0.0002, "epoch": 1.7416912074042912, "step": 2070}, {"loss": 1.7728, "grad_norm": 0.39914557337760925, "learning_rate": 0.0002, "epoch": 1.750105174589819, "step": 2080}, {"loss": 1.7049, "grad_norm": 0.36068692803382874, "learning_rate": 0.0002, "epoch": 1.7585191417753472, "step": 2090}, {"loss": 1.7537, "grad_norm": 0.3983287215232849, "learning_rate": 0.0002, "epoch": 1.766933108960875, "step": 2100}, {"loss": 1.7016, "grad_norm": 0.45008400082588196, "learning_rate": 0.0002, "epoch": 1.775347076146403, "step": 2110}, {"loss": 1.7163, "grad_norm": 0.3618052303791046, "learning_rate": 0.0002, "epoch": 1.783761043331931, "step": 2120}, {"loss": 1.7335, "grad_norm": 0.38745400309562683, "learning_rate": 0.0002, "epoch": 1.792175010517459, "step": 2130}, {"loss": 1.7387, "grad_norm": 0.3413826525211334, "learning_rate": 0.0002, "epoch": 1.8005889777029869, "step": 2140}, {"loss": 1.7414, "grad_norm": 0.35983747243881226, "learning_rate": 0.0002, "epoch": 1.809002944888515, "step": 2150}, {"loss": 1.7892, "grad_norm": 0.40926849842071533, "learning_rate": 0.0002, "epoch": 1.8174169120740429, "step": 2160}, {"loss": 1.6823, "grad_norm": 0.3543093800544739, "learning_rate": 0.0002, "epoch": 1.8258308792595708, "step": 2170}, {"loss": 1.7812, "grad_norm": 0.42690935730934143, "learning_rate": 0.0002, "epoch": 1.8342448464450989, "step": 2180}, {"loss": 1.7471, "grad_norm": 0.40282756090164185, "learning_rate": 0.0002, "epoch": 1.842658813630627, "step": 2190}, {"loss": 1.7411, "grad_norm": 0.36568400263786316, "learning_rate": 0.0002, "epoch": 1.8510727808161547, "step": 2200}, {"loss": 1.7024, "grad_norm": 0.43159013986587524, "learning_rate": 0.0002, "epoch": 1.8594867480016828, "step": 2210}, {"loss": 1.7298, "grad_norm": 0.3554118573665619, "learning_rate": 0.0002, "epoch": 1.867900715187211, "step": 2220}, {"loss": 1.7157, "grad_norm": 0.43349072337150574, "learning_rate": 0.0002, "epoch": 1.8763146823727388, "step": 2230}, {"loss": 1.7302, "grad_norm": 0.36486536264419556, "learning_rate": 0.0002, "epoch": 1.8847286495582667, "step": 2240}, {"loss": 1.6901, "grad_norm": 0.39260047674179077, "learning_rate": 0.0002, "epoch": 1.8931426167437948, "step": 2250}, {"loss": 1.6691, "grad_norm": 0.3741776943206787, "learning_rate": 0.0002, "epoch": 1.9015565839293227, "step": 2260}, {"loss": 1.6931, "grad_norm": 0.3961946964263916, "learning_rate": 0.0002, "epoch": 1.9099705511148506, "step": 2270}, {"loss": 1.737, "grad_norm": 0.3659731149673462, "learning_rate": 0.0002, "epoch": 1.9183845183003787, "step": 2280}, {"loss": 1.7342, "grad_norm": 0.34744107723236084, "learning_rate": 0.0002, "epoch": 1.9267984854859066, "step": 2290}, {"loss": 1.7162, "grad_norm": 0.3607442378997803, "learning_rate": 0.0002, "epoch": 1.9352124526714345, "step": 2300}, {"loss": 1.6673, "grad_norm": 0.331464558839798, "learning_rate": 0.0002, "epoch": 1.9436264198569626, "step": 2310}, {"loss": 1.7101, "grad_norm": 0.3904414474964142, "learning_rate": 0.0002, "epoch": 1.9520403870424905, "step": 2320}, {"loss": 1.7327, "grad_norm": 0.37584832310676575, "learning_rate": 0.0002, "epoch": 1.9604543542280184, "step": 2330}, {"loss": 1.7586, "grad_norm": 0.3698684275150299, "learning_rate": 0.0002, "epoch": 1.9688683214135465, "step": 2340}, {"loss": 1.7764, "grad_norm": 0.40571412444114685, "learning_rate": 0.0002, "epoch": 1.9772822885990746, "step": 2350}, {"loss": 1.744, "grad_norm": 0.40059587359428406, "learning_rate": 0.0002, "epoch": 1.9856962557846023, "step": 2360}, {"loss": 1.7033, "grad_norm": 0.4168248474597931, "learning_rate": 0.0002, "epoch": 1.9941102229701304, "step": 2370}, {"eval_loss": 1.8055059909820557, "eval_runtime": 38.422, "eval_samples_per_second": 13.404, "eval_steps_per_second": 1.692, "epoch": 2.0, "step": 2377}, {"loss": 1.7673, "grad_norm": 0.35205352306365967, "learning_rate": 0.0002, "epoch": 2.0025241901556585, "step": 2380}, {"loss": 1.6556, "grad_norm": 0.3979377746582031, "learning_rate": 0.0002, "epoch": 2.010938157341186, "step": 2390}, {"loss": 1.6421, "grad_norm": 0.396491676568985, "learning_rate": 0.0002, "epoch": 2.0193521245267143, "step": 2400}, {"loss": 1.6847, "grad_norm": 0.44712209701538086, "learning_rate": 0.0002, "epoch": 2.0277660917122424, "step": 2410}, {"loss": 1.6877, "grad_norm": 0.4454420208930969, "learning_rate": 0.0002, "epoch": 2.03618005889777, "step": 2420}, {"loss": 1.6635, "grad_norm": 0.4170038402080536, "learning_rate": 0.0002, "epoch": 2.044594026083298, "step": 2430}, {"loss": 1.6512, "grad_norm": 0.4309595227241516, "learning_rate": 0.0002, "epoch": 2.0530079932688263, "step": 2440}, {"loss": 1.6223, "grad_norm": 0.4241602122783661, "learning_rate": 0.0002, "epoch": 2.0614219604543544, "step": 2450}, {"loss": 1.6162, "grad_norm": 0.4370540678501129, "learning_rate": 0.0002, "epoch": 2.069835927639882, "step": 2460}, {"loss": 1.6354, "grad_norm": 0.43985554575920105, "learning_rate": 0.0002, "epoch": 2.0782498948254102, "step": 2470}, {"loss": 1.6954, "grad_norm": 0.4158105254173279, "learning_rate": 0.0002, "epoch": 2.0866638620109383, "step": 2480}, {"loss": 1.6114, "grad_norm": 0.441549152135849, "learning_rate": 0.0002, "epoch": 2.095077829196466, "step": 2490}, {"loss": 1.5485, "grad_norm": 0.385718435049057, "learning_rate": 0.0002, "epoch": 2.103491796381994, "step": 2500}, {"loss": 1.5894, "grad_norm": 0.43146514892578125, "learning_rate": 0.0002, "epoch": 2.1119057635675222, "step": 2510}, {"loss": 1.6414, "grad_norm": 0.41663315892219543, "learning_rate": 0.0002, "epoch": 2.12031973075305, "step": 2520}, {"loss": 1.6527, "grad_norm": 0.4410698115825653, "learning_rate": 0.0002, "epoch": 2.128733697938578, "step": 2530}, {"loss": 1.6124, "grad_norm": 0.4472278952598572, "learning_rate": 0.0002, "epoch": 2.137147665124106, "step": 2540}, {"loss": 1.6257, "grad_norm": 0.3879167437553406, "learning_rate": 0.0002, "epoch": 2.145561632309634, "step": 2550}, {"loss": 1.6682, "grad_norm": 0.4212203025817871, "learning_rate": 0.0002, "epoch": 2.153975599495162, "step": 2560}, {"loss": 1.6036, "grad_norm": 0.42841723561286926, "learning_rate": 0.0002, "epoch": 2.16238956668069, "step": 2570}, {"loss": 1.5962, "grad_norm": 0.39272481203079224, "learning_rate": 0.0002, "epoch": 2.1708035338662177, "step": 2580}, {"loss": 1.681, "grad_norm": 0.4075261354446411, "learning_rate": 0.0002, "epoch": 2.179217501051746, "step": 2590}, {"loss": 1.6601, "grad_norm": 0.5358437895774841, "learning_rate": 0.0002, "epoch": 2.187631468237274, "step": 2600}, {"loss": 1.6423, "grad_norm": 0.4738350212574005, "learning_rate": 0.0002, "epoch": 2.1960454354228016, "step": 2610}, {"loss": 1.6386, "grad_norm": 0.446789026260376, "learning_rate": 0.0002, "epoch": 2.2044594026083297, "step": 2620}, {"loss": 1.6246, "grad_norm": 0.4615374505519867, "learning_rate": 0.0002, "epoch": 2.212873369793858, "step": 2630}, {"loss": 1.6205, "grad_norm": 0.46901994943618774, "learning_rate": 0.0002, "epoch": 2.221287336979386, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.46267789602279663, "learning_rate": 0.0002, "epoch": 2.2297013041649136, "step": 2650}, {"loss": 1.6584, "grad_norm": 0.4383080005645752, "learning_rate": 0.0002, "epoch": 2.2381152713504417, "step": 2660}, {"loss": 1.5745, "grad_norm": 0.4070609509944916, "learning_rate": 0.0002, "epoch": 2.24652923853597, "step": 2670}, {"loss": 1.6125, "grad_norm": 0.4572339951992035, "learning_rate": 0.0002, "epoch": 2.2549432057214975, "step": 2680}, {"loss": 1.5671, "grad_norm": 0.393265038728714, "learning_rate": 0.0002, "epoch": 2.2633571729070256, "step": 2690}, {"loss": 1.6239, "grad_norm": 0.46144717931747437, "learning_rate": 0.0002, "epoch": 2.2717711400925538, "step": 2700}, {"loss": 1.5992, "grad_norm": 0.45077767968177795, "learning_rate": 0.0002, "epoch": 2.2801851072780814, "step": 2710}, {"loss": 1.6261, "grad_norm": 0.5697639584541321, "learning_rate": 0.0002, "epoch": 2.2885990744636096, "step": 2720}, {"loss": 1.6192, "grad_norm": 0.4855510890483856, "learning_rate": 0.0002, "epoch": 2.2970130416491377, "step": 2730}, {"loss": 1.7419, "grad_norm": 0.4440622627735138, "learning_rate": 0.0002, "epoch": 2.3054270088346653, "step": 2740}, {"loss": 1.6496, "grad_norm": 0.3904096782207489, "learning_rate": 0.0002, "epoch": 2.3138409760201935, "step": 2750}, {"loss": 1.5888, "grad_norm": 0.5225510597229004, "learning_rate": 0.0002, "epoch": 2.3222549432057216, "step": 2760}, {"loss": 1.6082, "grad_norm": 0.44866397976875305, "learning_rate": 0.0002, "epoch": 2.3306689103912497, "step": 2770}, {"loss": 1.6087, "grad_norm": 0.5167056322097778, "learning_rate": 0.0002, "epoch": 2.3390828775767774, "step": 2780}, {"loss": 1.6136, "grad_norm": 0.45913267135620117, "learning_rate": 0.0002, "epoch": 2.3474968447623055, "step": 2790}, {"loss": 1.6564, "grad_norm": 0.45787590742111206, "learning_rate": 0.0002, "epoch": 2.3559108119478336, "step": 2800}, {"loss": 1.6868, "grad_norm": 0.4633352756500244, "learning_rate": 0.0002, "epoch": 2.3643247791333613, "step": 2810}, {"loss": 1.6316, "grad_norm": 0.46390071511268616, "learning_rate": 0.0002, "epoch": 2.3727387463188894, "step": 2820}, {"loss": 1.6039, "grad_norm": 0.4261005222797394, "learning_rate": 0.0002, "epoch": 2.3811527135044175, "step": 2830}, {"loss": 1.6364, "grad_norm": 0.4283634424209595, "learning_rate": 0.0002, "epoch": 2.389566680689945, "step": 2840}, {"loss": 1.6382, "grad_norm": 0.4955291450023651, "learning_rate": 0.0002, "epoch": 2.3979806478754733, "step": 2850}, {"loss": 1.6173, "grad_norm": 0.4740189015865326, "learning_rate": 0.0002, "epoch": 2.4063946150610014, "step": 2860}, {"loss": 1.6403, "grad_norm": 0.4222276508808136, "learning_rate": 0.0002, "epoch": 2.414808582246529, "step": 2870}, {"loss": 1.5602, "grad_norm": 0.4982149004936218, "learning_rate": 0.0002, "epoch": 2.423222549432057, "step": 2880}, {"loss": 1.6313, "grad_norm": 0.5217409133911133, "learning_rate": 0.0002, "epoch": 2.4316365166175853, "step": 2890}, {"loss": 1.5804, "grad_norm": 0.4555884897708893, "learning_rate": 0.0002, "epoch": 2.4400504838031134, "step": 2900}, {"loss": 1.6189, "grad_norm": 0.43178579211235046, "learning_rate": 0.0002, "epoch": 2.448464450988641, "step": 2910}, {"loss": 1.6824, "grad_norm": 0.4788478910923004, "learning_rate": 0.0002, "epoch": 2.456878418174169, "step": 2920}, {"loss": 1.6829, "grad_norm": 0.43689873814582825, "learning_rate": 0.0002, "epoch": 2.465292385359697, "step": 2930}, {"loss": 1.6196, "grad_norm": 0.5115197896957397, "learning_rate": 0.0002, "epoch": 2.473706352545225, "step": 2940}, {"loss": 1.689, "grad_norm": 0.5290159583091736, "learning_rate": 0.0002, "epoch": 2.482120319730753, "step": 2950}, {"loss": 1.6499, "grad_norm": 0.46042463183403015, "learning_rate": 0.0002, "epoch": 2.490534286916281, "step": 2960}, {"loss": 1.6664, "grad_norm": 0.4359915852546692, "learning_rate": 0.0002, "epoch": 2.498948254101809, "step": 2970}, {"loss": 1.5812, "grad_norm": 0.46352964639663696, "learning_rate": 0.0002, "epoch": 2.507362221287337, "step": 2980}, {"loss": 1.6501, "grad_norm": 0.5324268341064453, "learning_rate": 0.0002, "epoch": 2.515776188472865, "step": 2990}, {"loss": 1.6115, "grad_norm": 0.5929607152938843, "learning_rate": 0.0002, "epoch": 2.5241901556583928, "step": 3000}, {"loss": 1.6772, "grad_norm": 0.4811333417892456, "learning_rate": 0.0002, "epoch": 2.532604122843921, "step": 3010}, {"loss": 1.7023, "grad_norm": 0.4662701487541199, "learning_rate": 0.0002, "epoch": 2.541018090029449, "step": 3020}, {"loss": 1.5426, "grad_norm": 0.4582270681858063, "learning_rate": 0.0002, "epoch": 2.549432057214977, "step": 3030}, {"loss": 1.6737, "grad_norm": 0.4679982662200928, "learning_rate": 0.0002, "epoch": 2.557846024400505, "step": 3040}, {"loss": 1.5442, "grad_norm": 0.4380294680595398, "learning_rate": 0.0002, "epoch": 2.566259991586033, "step": 3050}, {"loss": 1.6055, "grad_norm": 0.44295763969421387, "learning_rate": 0.0002, "epoch": 2.5746739587715606, "step": 3060}, {"loss": 1.5775, "grad_norm": 0.5131027698516846, "learning_rate": 0.0002, "epoch": 2.5830879259570887, "step": 3070}, {"loss": 1.546, "grad_norm": 0.47567516565322876, "learning_rate": 0.0002, "epoch": 2.591501893142617, "step": 3080}, {"loss": 1.5671, "grad_norm": 0.49002596735954285, "learning_rate": 0.0002, "epoch": 2.599915860328145, "step": 3090}, {"loss": 1.5445, "grad_norm": 0.44856327772140503, "learning_rate": 0.0002, "epoch": 2.6083298275136726, "step": 3100}, {"loss": 1.5797, "grad_norm": 0.4480142593383789, "learning_rate": 0.0002, "epoch": 2.6167437946992007, "step": 3110}, {"loss": 1.7132, "grad_norm": 0.4317494034767151, "learning_rate": 0.0002, "epoch": 2.6251577618847284, "step": 3120}, {"loss": 1.6321, "grad_norm": 0.42580848932266235, "learning_rate": 0.0002, "epoch": 2.6335717290702565, "step": 3130}, {"loss": 1.6483, "grad_norm": 0.4516814947128296, "learning_rate": 0.0002, "epoch": 2.6419856962557846, "step": 3140}, {"loss": 1.695, "grad_norm": 0.4438435733318329, "learning_rate": 0.0002, "epoch": 2.6503996634413127, "step": 3150}, {"loss": 1.6938, "grad_norm": 0.4385356307029724, "learning_rate": 0.0002, "epoch": 2.6588136306268404, "step": 3160}, {"loss": 1.6139, "grad_norm": 0.5064112544059753, "learning_rate": 0.0002, "epoch": 2.6672275978123685, "step": 3170}, {"loss": 1.7189, "grad_norm": 0.49163177609443665, "learning_rate": 0.0002, "epoch": 2.6756415649978966, "step": 3180}, {"loss": 1.7323, "grad_norm": 0.49339258670806885, "learning_rate": 0.0002, "epoch": 2.6840555321834243, "step": 3190}, {"loss": 1.6508, "grad_norm": 0.440950870513916, "learning_rate": 0.0002, "epoch": 2.6924694993689524, "step": 3200}, {"loss": 1.6305, "grad_norm": 0.4283970594406128, "learning_rate": 0.0002, "epoch": 2.7008834665544805, "step": 3210}, {"loss": 1.5935, "grad_norm": 0.43875712156295776, "learning_rate": 0.0002, "epoch": 2.7092974337400086, "step": 3220}, {"loss": 1.6129, "grad_norm": 0.49332964420318604, "learning_rate": 0.0002, "epoch": 2.7177114009255363, "step": 3230}, {"loss": 1.642, "grad_norm": 0.5225692391395569, "learning_rate": 0.0002, "epoch": 2.7261253681110644, "step": 3240}, {"loss": 1.6759, "grad_norm": 0.4856489300727844, "learning_rate": 0.0002, "epoch": 2.734539335296592, "step": 3250}, {"loss": 1.6463, "grad_norm": 0.46918296813964844, "learning_rate": 0.0002, "epoch": 2.74295330248212, "step": 3260}, {"loss": 1.6819, "grad_norm": 0.4802931249141693, "learning_rate": 0.0002, "epoch": 2.7513672696676483, "step": 3270}, {"loss": 1.6246, "grad_norm": 0.4485355615615845, "learning_rate": 0.0002, "epoch": 2.7597812368531764, "step": 3280}, {"loss": 1.6251, "grad_norm": 0.43944594264030457, "learning_rate": 0.0002, "epoch": 2.768195204038704, "step": 3290}, {"loss": 1.6501, "grad_norm": 0.46847742795944214, "learning_rate": 0.0002, "epoch": 2.7766091712242322, "step": 3300}, {"loss": 1.5969, "grad_norm": 0.4816027879714966, "learning_rate": 0.0002, "epoch": 2.7850231384097603, "step": 3310}, {"loss": 1.6293, "grad_norm": 0.453960120677948, "learning_rate": 0.0002, "epoch": 2.793437105595288, "step": 3320}, {"loss": 1.6429, "grad_norm": 0.4816017150878906, "learning_rate": 0.0002, "epoch": 2.801851072780816, "step": 3330}, {"loss": 1.6683, "grad_norm": 0.4461034834384918, "learning_rate": 0.0002, "epoch": 2.8102650399663442, "step": 3340}, {"loss": 1.7048, "grad_norm": 0.48821821808815, "learning_rate": 0.0002, "epoch": 2.8186790071518724, "step": 3350}, {"loss": 1.6076, "grad_norm": 0.4574853777885437, "learning_rate": 0.0002, "epoch": 2.8270929743374, "step": 3360}, {"loss": 1.6651, "grad_norm": 0.42062026262283325, "learning_rate": 0.0002, "epoch": 2.835506941522928, "step": 3370}, {"loss": 1.624, "grad_norm": 0.4499834477901459, "learning_rate": 0.0002, "epoch": 2.843920908708456, "step": 3380}, {"loss": 1.621, "grad_norm": 0.4780360758304596, "learning_rate": 0.0002, "epoch": 2.852334875893984, "step": 3390}, {"loss": 1.5882, "grad_norm": 0.45422887802124023, "learning_rate": 0.0002, "epoch": 2.860748843079512, "step": 3400}, {"loss": 1.6028, "grad_norm": 0.4590015709400177, "learning_rate": 0.0002, "epoch": 2.86916281026504, "step": 3410}, {"loss": 1.6746, "grad_norm": 0.45689624547958374, "learning_rate": 0.0002, "epoch": 2.877576777450568, "step": 3420}, {"loss": 1.6326, "grad_norm": 0.46953922510147095, "learning_rate": 0.0002, "epoch": 2.885990744636096, "step": 3430}, {"loss": 1.6015, "grad_norm": 0.4791966378688812, "learning_rate": 0.0002, "epoch": 2.8944047118216236, "step": 3440}, {"loss": 1.694, "grad_norm": 0.4842296242713928, "learning_rate": 0.0002, "epoch": 2.9028186790071517, "step": 3450}, {"loss": 1.6326, "grad_norm": 0.47219768166542053, "learning_rate": 0.0002, "epoch": 2.91123264619268, "step": 3460}, {"loss": 1.6486, "grad_norm": 0.4622127115726471, "learning_rate": 0.0002, "epoch": 2.919646613378208, "step": 3470}, {"loss": 1.6485, "grad_norm": 0.46832820773124695, "learning_rate": 0.0002, "epoch": 2.9280605805637356, "step": 3480}, {"loss": 1.6366, "grad_norm": 0.44582483172416687, "learning_rate": 0.0002, "epoch": 2.9364745477492638, "step": 3490}, {"loss": 1.6859, "grad_norm": 0.4987219274044037, "learning_rate": 0.0002, "epoch": 2.944888514934792, "step": 3500}, {"loss": 1.5991, "grad_norm": 0.43750956654548645, "learning_rate": 0.0002, "epoch": 2.9533024821203195, "step": 3510}, {"loss": 1.6236, "grad_norm": 0.49962925910949707, "learning_rate": 0.0002, "epoch": 2.9617164493058477, "step": 3520}, {"loss": 1.5859, "grad_norm": 0.5189590454101562, "learning_rate": 0.0002, "epoch": 2.9701304164913758, "step": 3530}, {"loss": 1.6688, "grad_norm": 0.391317754983902, "learning_rate": 0.0002, "epoch": 2.978544383676904, "step": 3540}, {"loss": 1.5884, "grad_norm": 0.44934695959091187, "learning_rate": 0.0002, "epoch": 2.9869583508624316, "step": 3550}, {"loss": 1.5688, "grad_norm": 0.4740142226219177, "learning_rate": 0.0002, "epoch": 2.9953723180479597, "step": 3560}]}
+{"epoch": 4.0, "step": 4754, "epoch_duration": 1272.4875209331512, "total_accumulated_duration": 6037.973757505417, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "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_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-0.8-num-5623-sd-42/checkpoint-2377", "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": 2.56, "grad_norm": 0.5458821654319763, "learning_rate": 0.0002, "epoch": 0.008413967185527976, "step": 10}, {"loss": 2.3235, "grad_norm": 0.7293308973312378, "learning_rate": 0.0002, "epoch": 0.016827934371055953, "step": 20}, {"loss": 2.0815, "grad_norm": 0.47792306542396545, "learning_rate": 0.0002, "epoch": 0.02524190155658393, "step": 30}, {"loss": 1.9718, "grad_norm": 0.5944402813911438, "learning_rate": 0.0002, "epoch": 0.033655868742111905, "step": 40}, {"loss": 1.8848, "grad_norm": 0.5415359735488892, "learning_rate": 0.0002, "epoch": 0.04206983592763988, "step": 50}, {"loss": 1.8953, "grad_norm": 0.535713791847229, "learning_rate": 0.0002, "epoch": 0.05048380311316786, "step": 60}, {"loss": 1.937, "grad_norm": 0.5184146761894226, "learning_rate": 0.0002, "epoch": 0.058897770298695834, "step": 70}, {"loss": 1.8396, "grad_norm": 0.458926796913147, "learning_rate": 0.0002, "epoch": 0.06731173748422381, "step": 80}, {"loss": 1.8677, "grad_norm": 0.4780142307281494, "learning_rate": 0.0002, "epoch": 0.07572570466975179, "step": 90}, {"loss": 1.8593, "grad_norm": 0.79965740442276, "learning_rate": 0.0002, "epoch": 0.08413967185527976, "step": 100}, {"loss": 1.9081, "grad_norm": 0.4498862028121948, "learning_rate": 0.0002, "epoch": 0.09255363904080774, "step": 110}, {"loss": 1.8503, "grad_norm": 0.39338430762290955, "learning_rate": 0.0002, "epoch": 0.10096760622633572, "step": 120}, {"loss": 1.8637, "grad_norm": 0.9588953852653503, "learning_rate": 0.0002, "epoch": 0.10938157341186369, "step": 130}, {"loss": 1.8676, "grad_norm": 0.41675639152526855, "learning_rate": 0.0002, "epoch": 0.11779554059739167, "step": 140}, {"loss": 1.8904, "grad_norm": 0.44519832730293274, "learning_rate": 0.0002, "epoch": 0.12620950778291964, "step": 150}, {"loss": 1.798, "grad_norm": 0.4176260530948639, "learning_rate": 0.0002, "epoch": 0.13462347496844762, "step": 160}, {"loss": 1.8398, "grad_norm": 0.35840365290641785, "learning_rate": 0.0002, "epoch": 0.1430374421539756, "step": 170}, {"loss": 1.8666, "grad_norm": 0.3794495463371277, "learning_rate": 0.0002, "epoch": 0.15145140933950357, "step": 180}, {"loss": 1.8111, "grad_norm": 0.4563522934913635, "learning_rate": 0.0002, "epoch": 0.15986537652503155, "step": 190}, {"loss": 1.8893, "grad_norm": 0.37057486176490784, "learning_rate": 0.0002, "epoch": 0.16827934371055953, "step": 200}, {"loss": 1.7995, "grad_norm": 0.44081518054008484, "learning_rate": 0.0002, "epoch": 0.1766933108960875, "step": 210}, {"loss": 1.9048, "grad_norm": 0.46078577637672424, "learning_rate": 0.0002, "epoch": 0.18510727808161548, "step": 220}, {"loss": 1.8403, "grad_norm": 0.36132094264030457, "learning_rate": 0.0002, "epoch": 0.19352124526714345, "step": 230}, {"loss": 1.8827, "grad_norm": 0.3747289180755615, "learning_rate": 0.0002, "epoch": 0.20193521245267143, "step": 240}, {"loss": 1.8382, "grad_norm": 0.3540179133415222, "learning_rate": 0.0002, "epoch": 0.2103491796381994, "step": 250}, {"loss": 1.8196, "grad_norm": 0.3461375832557678, "learning_rate": 0.0002, "epoch": 0.21876314682372738, "step": 260}, {"loss": 1.8509, "grad_norm": 0.3436960279941559, "learning_rate": 0.0002, "epoch": 0.22717711400925536, "step": 270}, {"loss": 1.8285, "grad_norm": 0.35403719544410706, "learning_rate": 0.0002, "epoch": 0.23559108119478334, "step": 280}, {"loss": 1.8369, "grad_norm": 0.37142616510391235, "learning_rate": 0.0002, "epoch": 0.2440050483803113, "step": 290}, {"loss": 1.8044, "grad_norm": 0.3307955861091614, "learning_rate": 0.0002, "epoch": 0.2524190155658393, "step": 300}, {"loss": 1.817, "grad_norm": 0.32855314016342163, "learning_rate": 0.0002, "epoch": 0.2608329827513673, "step": 310}, {"loss": 1.7803, "grad_norm": 0.3299003839492798, "learning_rate": 0.0002, "epoch": 0.26924694993689524, "step": 320}, {"loss": 1.8129, "grad_norm": 0.44311287999153137, "learning_rate": 0.0002, "epoch": 0.27766091712242325, "step": 330}, {"loss": 1.8232, "grad_norm": 0.32989758253097534, "learning_rate": 0.0002, "epoch": 0.2860748843079512, "step": 340}, {"loss": 1.7716, "grad_norm": 0.34400200843811035, "learning_rate": 0.0002, "epoch": 0.2944888514934792, "step": 350}, {"loss": 1.7619, "grad_norm": 0.36286211013793945, "learning_rate": 0.0002, "epoch": 0.30290281867900715, "step": 360}, {"loss": 1.8025, "grad_norm": 0.406827837228775, "learning_rate": 0.0002, "epoch": 0.31131678586453515, "step": 370}, {"loss": 1.7515, "grad_norm": 0.36299195885658264, "learning_rate": 0.0002, "epoch": 0.3197307530500631, "step": 380}, {"loss": 1.837, "grad_norm": 0.3477257192134857, "learning_rate": 0.0002, "epoch": 0.3281447202355911, "step": 390}, {"loss": 1.7767, "grad_norm": 0.3730369210243225, "learning_rate": 0.0002, "epoch": 0.33655868742111905, "step": 400}, {"loss": 1.7747, "grad_norm": 0.4644559919834137, "learning_rate": 0.0002, "epoch": 0.34497265460664706, "step": 410}, {"loss": 1.7538, "grad_norm": 0.406576544046402, "learning_rate": 0.0002, "epoch": 0.353386621792175, "step": 420}, {"loss": 1.7501, "grad_norm": 0.3612699508666992, "learning_rate": 0.0002, "epoch": 0.361800588977703, "step": 430}, {"loss": 1.7473, "grad_norm": 0.3243742287158966, "learning_rate": 0.0002, "epoch": 0.37021455616323096, "step": 440}, {"loss": 1.8851, "grad_norm": 0.36671221256256104, "learning_rate": 0.0002, "epoch": 0.37862852334875896, "step": 450}, {"loss": 1.8853, "grad_norm": 0.3565002381801605, "learning_rate": 0.0002, "epoch": 0.3870424905342869, "step": 460}, {"loss": 1.8923, "grad_norm": 0.34630221128463745, "learning_rate": 0.0002, "epoch": 0.3954564577198149, "step": 470}, {"loss": 1.8234, "grad_norm": 0.3353537321090698, "learning_rate": 0.0002, "epoch": 0.40387042490534286, "step": 480}, {"loss": 1.7135, "grad_norm": 0.4015921950340271, "learning_rate": 0.0002, "epoch": 0.41228439209087087, "step": 490}, {"loss": 1.7815, "grad_norm": 0.5489419102668762, "learning_rate": 0.0002, "epoch": 0.4206983592763988, "step": 500}, {"loss": 1.7903, "grad_norm": 0.4193589985370636, "learning_rate": 0.0002, "epoch": 0.4291123264619268, "step": 510}, {"loss": 1.8416, "grad_norm": 0.3418922424316406, "learning_rate": 0.0002, "epoch": 0.43752629364745477, "step": 520}, {"loss": 1.7982, "grad_norm": 0.32668185234069824, "learning_rate": 0.0002, "epoch": 0.44594026083298277, "step": 530}, {"loss": 1.7501, "grad_norm": 0.3094325661659241, "learning_rate": 0.0002, "epoch": 0.4543542280185107, "step": 540}, {"loss": 1.7438, "grad_norm": 0.3743017315864563, "learning_rate": 0.0002, "epoch": 0.4627681952040387, "step": 550}, {"loss": 1.8451, "grad_norm": 0.3295630216598511, "learning_rate": 0.0002, "epoch": 0.47118216238956667, "step": 560}, {"loss": 1.7529, "grad_norm": 1.6124513149261475, "learning_rate": 0.0002, "epoch": 0.4795961295750947, "step": 570}, {"loss": 1.8028, "grad_norm": 0.3245585858821869, "learning_rate": 0.0002, "epoch": 0.4880100967606226, "step": 580}, {"loss": 1.7976, "grad_norm": 0.3332934081554413, "learning_rate": 0.0002, "epoch": 0.49642406394615063, "step": 590}, {"loss": 1.7912, "grad_norm": 0.3836138844490051, "learning_rate": 0.0002, "epoch": 0.5048380311316786, "step": 600}, {"loss": 1.8347, "grad_norm": 0.32953888177871704, "learning_rate": 0.0002, "epoch": 0.5132519983172066, "step": 610}, {"loss": 1.7729, "grad_norm": 0.36291512846946716, "learning_rate": 0.0002, "epoch": 0.5216659655027346, "step": 620}, {"loss": 1.7758, "grad_norm": 0.3237783908843994, "learning_rate": 0.0002, "epoch": 0.5300799326882625, "step": 630}, {"loss": 1.8352, "grad_norm": 0.38882696628570557, "learning_rate": 0.0002, "epoch": 0.5384938998737905, "step": 640}, {"loss": 1.8624, "grad_norm": 0.37821972370147705, "learning_rate": 0.0002, "epoch": 0.5469078670593185, "step": 650}, {"loss": 1.8075, "grad_norm": 0.3556285500526428, "learning_rate": 0.0002, "epoch": 0.5553218342448465, "step": 660}, {"loss": 1.778, "grad_norm": 0.347499281167984, "learning_rate": 0.0002, "epoch": 0.5637358014303744, "step": 670}, {"loss": 1.8066, "grad_norm": 0.3176489472389221, "learning_rate": 0.0002, "epoch": 0.5721497686159024, "step": 680}, {"loss": 1.7257, "grad_norm": 0.30220088362693787, "learning_rate": 0.0002, "epoch": 0.5805637358014304, "step": 690}, {"loss": 1.8415, "grad_norm": 0.3711601793766022, "learning_rate": 0.0002, "epoch": 0.5889777029869584, "step": 700}, {"loss": 1.7906, "grad_norm": 0.3311759829521179, "learning_rate": 0.0002, "epoch": 0.5973916701724863, "step": 710}, {"loss": 1.7712, "grad_norm": 0.34824270009994507, "learning_rate": 0.0002, "epoch": 0.6058056373580143, "step": 720}, {"loss": 1.7954, "grad_norm": 0.29668381810188293, "learning_rate": 0.0002, "epoch": 0.6142196045435423, "step": 730}, {"loss": 1.8321, "grad_norm": 0.36087489128112793, "learning_rate": 0.0002, "epoch": 0.6226335717290703, "step": 740}, {"loss": 1.7956, "grad_norm": 0.31590089201927185, "learning_rate": 0.0002, "epoch": 0.6310475389145982, "step": 750}, {"loss": 1.7343, "grad_norm": 0.37632957100868225, "learning_rate": 0.0002, "epoch": 0.6394615061001262, "step": 760}, {"loss": 1.8499, "grad_norm": 0.3360748589038849, "learning_rate": 0.0002, "epoch": 0.6478754732856542, "step": 770}, {"loss": 1.8076, "grad_norm": 0.3420640528202057, "learning_rate": 0.0002, "epoch": 0.6562894404711822, "step": 780}, {"loss": 1.8353, "grad_norm": 0.5734959244728088, "learning_rate": 0.0002, "epoch": 0.6647034076567101, "step": 790}, {"loss": 1.7746, "grad_norm": 0.36440837383270264, "learning_rate": 0.0002, "epoch": 0.6731173748422381, "step": 800}, {"loss": 1.7532, "grad_norm": 0.3179708421230316, "learning_rate": 0.0002, "epoch": 0.6815313420277661, "step": 810}, {"loss": 1.7815, "grad_norm": 0.34122881293296814, "learning_rate": 0.0002, "epoch": 0.6899453092132941, "step": 820}, {"loss": 1.8167, "grad_norm": 0.31886112689971924, "learning_rate": 0.0002, "epoch": 0.698359276398822, "step": 830}, {"loss": 1.7505, "grad_norm": 0.31782326102256775, "learning_rate": 0.0002, "epoch": 0.70677324358435, "step": 840}, {"loss": 1.7588, "grad_norm": 0.36052989959716797, "learning_rate": 0.0002, "epoch": 0.715187210769878, "step": 850}, {"loss": 1.7891, "grad_norm": 0.28946155309677124, "learning_rate": 0.0002, "epoch": 0.723601177955406, "step": 860}, {"loss": 1.7923, "grad_norm": 0.3095663785934448, "learning_rate": 0.0002, "epoch": 0.7320151451409339, "step": 870}, {"loss": 1.785, "grad_norm": 0.3317491412162781, "learning_rate": 0.0002, "epoch": 0.7404291123264619, "step": 880}, {"loss": 1.7709, "grad_norm": 0.31324660778045654, "learning_rate": 0.0002, "epoch": 0.7488430795119899, "step": 890}, {"loss": 1.8753, "grad_norm": 0.3290475606918335, "learning_rate": 0.0002, "epoch": 0.7572570466975179, "step": 900}, {"loss": 1.7679, "grad_norm": 0.35690343379974365, "learning_rate": 0.0002, "epoch": 0.7656710138830458, "step": 910}, {"loss": 1.826, "grad_norm": 0.39558273553848267, "learning_rate": 0.0002, "epoch": 0.7740849810685738, "step": 920}, {"loss": 1.8722, "grad_norm": 0.34254348278045654, "learning_rate": 0.0002, "epoch": 0.7824989482541018, "step": 930}, {"loss": 1.7603, "grad_norm": 0.3560165464878082, "learning_rate": 0.0002, "epoch": 0.7909129154396298, "step": 940}, {"loss": 1.7992, "grad_norm": 0.30693164467811584, "learning_rate": 0.0002, "epoch": 0.7993268826251577, "step": 950}, {"loss": 1.8029, "grad_norm": 0.3394823372364044, "learning_rate": 0.0002, "epoch": 0.8077408498106857, "step": 960}, {"loss": 1.8105, "grad_norm": 0.3741514980792999, "learning_rate": 0.0002, "epoch": 0.8161548169962137, "step": 970}, {"loss": 1.7849, "grad_norm": 0.3655228316783905, "learning_rate": 0.0002, "epoch": 0.8245687841817417, "step": 980}, {"loss": 1.8449, "grad_norm": 0.3586033880710602, "learning_rate": 0.0002, "epoch": 0.8329827513672696, "step": 990}, {"loss": 1.7033, "grad_norm": 0.3459678888320923, "learning_rate": 0.0002, "epoch": 0.8413967185527976, "step": 1000}, {"loss": 1.8498, "grad_norm": 0.3184349834918976, "learning_rate": 0.0002, "epoch": 0.8498106857383256, "step": 1010}, {"loss": 1.7632, "grad_norm": 0.3099786043167114, "learning_rate": 0.0002, "epoch": 0.8582246529238536, "step": 1020}, {"loss": 1.8067, "grad_norm": 0.30300915241241455, "learning_rate": 0.0002, "epoch": 0.8666386201093815, "step": 1030}, {"loss": 1.7923, "grad_norm": 0.3128705620765686, "learning_rate": 0.0002, "epoch": 0.8750525872949095, "step": 1040}, {"loss": 1.8252, "grad_norm": 0.3336263597011566, "learning_rate": 0.0002, "epoch": 0.8834665544804375, "step": 1050}, {"loss": 1.8375, "grad_norm": 0.3801328241825104, "learning_rate": 0.0002, "epoch": 0.8918805216659655, "step": 1060}, {"loss": 1.7757, "grad_norm": 0.3122096359729767, "learning_rate": 0.0002, "epoch": 0.9002944888514934, "step": 1070}, {"loss": 1.8251, "grad_norm": 0.35990869998931885, "learning_rate": 0.0002, "epoch": 0.9087084560370214, "step": 1080}, {"loss": 1.7343, "grad_norm": 0.3321819305419922, "learning_rate": 0.0002, "epoch": 0.9171224232225494, "step": 1090}, {"loss": 1.7595, "grad_norm": 0.4202139377593994, "learning_rate": 0.0002, "epoch": 0.9255363904080774, "step": 1100}, {"loss": 1.8056, "grad_norm": 0.32559722661972046, "learning_rate": 0.0002, "epoch": 0.9339503575936053, "step": 1110}, {"loss": 1.812, "grad_norm": 0.3098459839820862, "learning_rate": 0.0002, "epoch": 0.9423643247791333, "step": 1120}, {"loss": 1.8252, "grad_norm": 0.33917108178138733, "learning_rate": 0.0002, "epoch": 0.9507782919646613, "step": 1130}, {"loss": 1.7709, "grad_norm": 0.4055837094783783, "learning_rate": 0.0002, "epoch": 0.9591922591501894, "step": 1140}, {"loss": 1.8259, "grad_norm": 0.32508623600006104, "learning_rate": 0.0002, "epoch": 0.9676062263357172, "step": 1150}, {"loss": 1.782, "grad_norm": 0.30150601267814636, "learning_rate": 0.0002, "epoch": 0.9760201935212452, "step": 1160}, {"loss": 1.8291, "grad_norm": 0.3042563199996948, "learning_rate": 0.0002, "epoch": 0.9844341607067733, "step": 1170}, {"loss": 1.7847, "grad_norm": 0.33254584670066833, "learning_rate": 0.0002, "epoch": 0.9928481278923013, "step": 1180}, {"eval_loss": 1.8077726364135742, "eval_runtime": 38.4359, "eval_samples_per_second": 13.399, "eval_steps_per_second": 1.691, "epoch": 0.9995793016407236, "step": 1188}, {"loss": 1.7414, "grad_norm": 0.35073035955429077, "learning_rate": 0.0002, "epoch": 1.0012620950778293, "step": 1190}, {"loss": 1.7483, "grad_norm": 0.3217269778251648, "learning_rate": 0.0002, "epoch": 1.0096760622633572, "step": 1200}, {"loss": 1.7517, "grad_norm": 0.3635033369064331, "learning_rate": 0.0002, "epoch": 1.018090029448885, "step": 1210}, {"loss": 1.6949, "grad_norm": 0.32468414306640625, "learning_rate": 0.0002, "epoch": 1.0265039966344132, "step": 1220}, {"loss": 1.711, "grad_norm": 0.3307163417339325, "learning_rate": 0.0002, "epoch": 1.034917963819941, "step": 1230}, {"loss": 1.7881, "grad_norm": 0.34381359815597534, "learning_rate": 0.0002, "epoch": 1.0433319310054692, "step": 1240}, {"loss": 1.612, "grad_norm": 0.35874804854393005, "learning_rate": 0.0002, "epoch": 1.051745898190997, "step": 1250}, {"loss": 1.7314, "grad_norm": 0.3615919351577759, "learning_rate": 0.0002, "epoch": 1.060159865376525, "step": 1260}, {"loss": 1.7517, "grad_norm": 0.32835808396339417, "learning_rate": 0.0002, "epoch": 1.068573832562053, "step": 1270}, {"loss": 1.7193, "grad_norm": 0.3876388370990753, "learning_rate": 0.0002, "epoch": 1.076987799747581, "step": 1280}, {"loss": 1.7442, "grad_norm": 0.39895930886268616, "learning_rate": 0.0002, "epoch": 1.0854017669331089, "step": 1290}, {"loss": 1.6601, "grad_norm": 0.39081698656082153, "learning_rate": 0.0002, "epoch": 1.093815734118637, "step": 1300}, {"loss": 1.7623, "grad_norm": 0.39974215626716614, "learning_rate": 0.0002, "epoch": 1.1022297013041649, "step": 1310}, {"loss": 1.7506, "grad_norm": 0.3887332081794739, "learning_rate": 0.0002, "epoch": 1.110643668489693, "step": 1320}, {"loss": 1.7381, "grad_norm": 0.36216408014297485, "learning_rate": 0.0002, "epoch": 1.1190576356752209, "step": 1330}, {"loss": 1.762, "grad_norm": 0.36979028582572937, "learning_rate": 0.0002, "epoch": 1.1274716028607488, "step": 1340}, {"loss": 1.7515, "grad_norm": 0.34052133560180664, "learning_rate": 0.0002, "epoch": 1.1358855700462769, "step": 1350}, {"loss": 1.7513, "grad_norm": 0.3467716574668884, "learning_rate": 0.0002, "epoch": 1.1442995372318048, "step": 1360}, {"loss": 1.7086, "grad_norm": 0.35528799891471863, "learning_rate": 0.0002, "epoch": 1.1527135044173327, "step": 1370}, {"loss": 1.794, "grad_norm": 0.36282262206077576, "learning_rate": 0.0002, "epoch": 1.1611274716028608, "step": 1380}, {"loss": 1.7731, "grad_norm": 0.37355899810791016, "learning_rate": 0.0002, "epoch": 1.1695414387883887, "step": 1390}, {"loss": 1.7483, "grad_norm": 0.37292736768722534, "learning_rate": 0.0002, "epoch": 1.1779554059739168, "step": 1400}, {"loss": 1.6916, "grad_norm": 0.5892812013626099, "learning_rate": 0.0002, "epoch": 1.1863693731594447, "step": 1410}, {"loss": 1.7302, "grad_norm": 0.3712292015552521, "learning_rate": 0.0002, "epoch": 1.1947833403449726, "step": 1420}, {"loss": 1.7709, "grad_norm": 0.3349577486515045, "learning_rate": 0.0002, "epoch": 1.2031973075305007, "step": 1430}, {"loss": 1.7412, "grad_norm": 0.32591062784194946, "learning_rate": 0.0002, "epoch": 1.2116112747160286, "step": 1440}, {"loss": 1.7406, "grad_norm": 0.3840635418891907, "learning_rate": 0.0002, "epoch": 1.2200252419015567, "step": 1450}, {"loss": 1.7276, "grad_norm": 0.37238365411758423, "learning_rate": 0.0002, "epoch": 1.2284392090870846, "step": 1460}, {"loss": 1.7052, "grad_norm": 0.3731217682361603, "learning_rate": 0.0002, "epoch": 1.2368531762726125, "step": 1470}, {"loss": 1.7255, "grad_norm": 0.3318967819213867, "learning_rate": 0.0002, "epoch": 1.2452671434581406, "step": 1480}, {"loss": 1.7463, "grad_norm": 0.3784034848213196, "learning_rate": 0.0002, "epoch": 1.2536811106436685, "step": 1490}, {"loss": 1.6862, "grad_norm": 0.3541383147239685, "learning_rate": 0.0002, "epoch": 1.2620950778291964, "step": 1500}, {"loss": 1.8394, "grad_norm": 0.35312485694885254, "learning_rate": 0.0002, "epoch": 1.2705090450147245, "step": 1510}, {"loss": 1.7029, "grad_norm": 0.35272929072380066, "learning_rate": 0.0002, "epoch": 1.2789230122002524, "step": 1520}, {"loss": 1.7016, "grad_norm": 0.40988272428512573, "learning_rate": 0.0002, "epoch": 1.2873369793857803, "step": 1530}, {"loss": 1.6912, "grad_norm": 0.3543946146965027, "learning_rate": 0.0002, "epoch": 1.2957509465713084, "step": 1540}, {"loss": 1.6757, "grad_norm": 0.35639145970344543, "learning_rate": 0.0002, "epoch": 1.3041649137568363, "step": 1550}, {"loss": 1.6814, "grad_norm": 0.3290826678276062, "learning_rate": 0.0002, "epoch": 1.3125788809423642, "step": 1560}, {"loss": 1.7369, "grad_norm": 0.39264336228370667, "learning_rate": 0.0002, "epoch": 1.3209928481278923, "step": 1570}, {"loss": 1.6804, "grad_norm": 0.5390415191650391, "learning_rate": 0.0002, "epoch": 1.3294068153134202, "step": 1580}, {"loss": 1.708, "grad_norm": 0.5188116431236267, "learning_rate": 0.0002, "epoch": 1.3378207824989483, "step": 1590}, {"loss": 1.6763, "grad_norm": 0.37445148825645447, "learning_rate": 0.0002, "epoch": 1.3462347496844762, "step": 1600}, {"loss": 1.7386, "grad_norm": 0.3296085298061371, "learning_rate": 0.0002, "epoch": 1.3546487168700043, "step": 1610}, {"loss": 1.8107, "grad_norm": 0.39879581332206726, "learning_rate": 0.0002, "epoch": 1.3630626840555322, "step": 1620}, {"loss": 1.6744, "grad_norm": 0.36092764139175415, "learning_rate": 0.0002, "epoch": 1.37147665124106, "step": 1630}, {"loss": 1.7144, "grad_norm": 0.37011823058128357, "learning_rate": 0.0002, "epoch": 1.3798906184265882, "step": 1640}, {"loss": 1.7396, "grad_norm": 0.40863534808158875, "learning_rate": 0.0002, "epoch": 1.3883045856121161, "step": 1650}, {"loss": 1.7901, "grad_norm": 0.337001770734787, "learning_rate": 0.0002, "epoch": 1.396718552797644, "step": 1660}, {"loss": 1.7044, "grad_norm": 0.35596707463264465, "learning_rate": 0.0002, "epoch": 1.4051325199831721, "step": 1670}, {"loss": 1.7717, "grad_norm": 0.3857671916484833, "learning_rate": 0.0002, "epoch": 1.4135464871687, "step": 1680}, {"loss": 1.7015, "grad_norm": 0.419502317905426, "learning_rate": 0.0002, "epoch": 1.421960454354228, "step": 1690}, {"loss": 1.7261, "grad_norm": 0.35459452867507935, "learning_rate": 0.0002, "epoch": 1.430374421539756, "step": 1700}, {"loss": 1.7361, "grad_norm": 0.37246978282928467, "learning_rate": 0.0002, "epoch": 1.438788388725284, "step": 1710}, {"loss": 1.6762, "grad_norm": 0.33091893792152405, "learning_rate": 0.0002, "epoch": 1.4472023559108118, "step": 1720}, {"loss": 1.7044, "grad_norm": 0.37029674649238586, "learning_rate": 0.0002, "epoch": 1.45561632309634, "step": 1730}, {"loss": 1.7117, "grad_norm": 0.374025821685791, "learning_rate": 0.0002, "epoch": 1.4640302902818678, "step": 1740}, {"loss": 1.7549, "grad_norm": 0.3416315019130707, "learning_rate": 0.0002, "epoch": 1.472444257467396, "step": 1750}, {"loss": 1.7093, "grad_norm": 0.36502841114997864, "learning_rate": 0.0002, "epoch": 1.4808582246529238, "step": 1760}, {"loss": 1.6597, "grad_norm": 0.35458803176879883, "learning_rate": 0.0002, "epoch": 1.489272191838452, "step": 1770}, {"loss": 1.675, "grad_norm": 0.4462839663028717, "learning_rate": 0.0002, "epoch": 1.4976861590239798, "step": 1780}, {"loss": 1.7267, "grad_norm": 0.34836092591285706, "learning_rate": 0.0002, "epoch": 1.5061001262095077, "step": 1790}, {"loss": 1.7295, "grad_norm": 0.3445749282836914, "learning_rate": 0.0002, "epoch": 1.5145140933950358, "step": 1800}, {"loss": 1.7386, "grad_norm": 0.36012160778045654, "learning_rate": 0.0002, "epoch": 1.5229280605805637, "step": 1810}, {"loss": 1.6594, "grad_norm": 0.4052616059780121, "learning_rate": 0.0002, "epoch": 1.5313420277660916, "step": 1820}, {"loss": 1.72, "grad_norm": 0.3966905474662781, "learning_rate": 0.0002, "epoch": 1.5397559949516197, "step": 1830}, {"loss": 1.7595, "grad_norm": 0.35028719902038574, "learning_rate": 0.0002, "epoch": 1.5481699621371476, "step": 1840}, {"loss": 1.6829, "grad_norm": 0.3936742842197418, "learning_rate": 0.0002, "epoch": 1.5565839293226755, "step": 1850}, {"loss": 1.7579, "grad_norm": 0.34473296999931335, "learning_rate": 0.0002, "epoch": 1.5649978965082036, "step": 1860}, {"loss": 1.7207, "grad_norm": 0.4328365623950958, "learning_rate": 0.0002, "epoch": 1.5734118636937318, "step": 1870}, {"loss": 1.7098, "grad_norm": 0.3566315472126007, "learning_rate": 0.0002, "epoch": 1.5818258308792594, "step": 1880}, {"loss": 1.6095, "grad_norm": 0.3301256597042084, "learning_rate": 0.0002, "epoch": 1.5902397980647875, "step": 1890}, {"loss": 1.748, "grad_norm": 0.3743041455745697, "learning_rate": 0.0002, "epoch": 1.5986537652503157, "step": 1900}, {"loss": 1.7259, "grad_norm": 0.3735344707965851, "learning_rate": 0.0002, "epoch": 1.6070677324358436, "step": 1910}, {"loss": 1.7445, "grad_norm": 0.42191144824028015, "learning_rate": 0.0002, "epoch": 1.6154816996213714, "step": 1920}, {"loss": 1.6978, "grad_norm": 0.3787207305431366, "learning_rate": 0.0002, "epoch": 1.6238956668068996, "step": 1930}, {"loss": 1.6893, "grad_norm": 0.35647350549697876, "learning_rate": 0.0002, "epoch": 1.6323096339924275, "step": 1940}, {"loss": 1.7825, "grad_norm": 0.39791446924209595, "learning_rate": 0.0002, "epoch": 1.6407236011779553, "step": 1950}, {"loss": 1.7293, "grad_norm": 0.37341275811195374, "learning_rate": 0.0002, "epoch": 1.6491375683634835, "step": 1960}, {"loss": 1.6781, "grad_norm": 0.3722686469554901, "learning_rate": 0.0002, "epoch": 1.6575515355490114, "step": 1970}, {"loss": 1.6383, "grad_norm": 0.37467387318611145, "learning_rate": 0.0002, "epoch": 1.6659655027345392, "step": 1980}, {"loss": 1.7439, "grad_norm": 0.37109461426734924, "learning_rate": 0.0002, "epoch": 1.6743794699200674, "step": 1990}, {"loss": 1.7206, "grad_norm": 0.4008837044239044, "learning_rate": 0.0002, "epoch": 1.6827934371055953, "step": 2000}, {"loss": 1.7604, "grad_norm": 0.3316999673843384, "learning_rate": 0.0002, "epoch": 1.6912074042911232, "step": 2010}, {"loss": 1.7325, "grad_norm": 0.3683805465698242, "learning_rate": 0.0002, "epoch": 1.6996213714766513, "step": 2020}, {"loss": 1.7451, "grad_norm": 0.4163658320903778, "learning_rate": 0.0002, "epoch": 1.7080353386621794, "step": 2030}, {"loss": 1.741, "grad_norm": 0.4245431125164032, "learning_rate": 0.0002, "epoch": 1.716449305847707, "step": 2040}, {"loss": 1.7184, "grad_norm": 0.36732038855552673, "learning_rate": 0.0002, "epoch": 1.7248632730332352, "step": 2050}, {"loss": 1.7031, "grad_norm": 0.34981656074523926, "learning_rate": 0.0002, "epoch": 1.7332772402187633, "step": 2060}, {"loss": 1.7545, "grad_norm": 0.38588812947273254, "learning_rate": 0.0002, "epoch": 1.7416912074042912, "step": 2070}, {"loss": 1.7728, "grad_norm": 0.39914557337760925, "learning_rate": 0.0002, "epoch": 1.750105174589819, "step": 2080}, {"loss": 1.7049, "grad_norm": 0.36068692803382874, "learning_rate": 0.0002, "epoch": 1.7585191417753472, "step": 2090}, {"loss": 1.7537, "grad_norm": 0.3983287215232849, "learning_rate": 0.0002, "epoch": 1.766933108960875, "step": 2100}, {"loss": 1.7016, "grad_norm": 0.45008400082588196, "learning_rate": 0.0002, "epoch": 1.775347076146403, "step": 2110}, {"loss": 1.7163, "grad_norm": 0.3618052303791046, "learning_rate": 0.0002, "epoch": 1.783761043331931, "step": 2120}, {"loss": 1.7335, "grad_norm": 0.38745400309562683, "learning_rate": 0.0002, "epoch": 1.792175010517459, "step": 2130}, {"loss": 1.7387, "grad_norm": 0.3413826525211334, "learning_rate": 0.0002, "epoch": 1.8005889777029869, "step": 2140}, {"loss": 1.7414, "grad_norm": 0.35983747243881226, "learning_rate": 0.0002, "epoch": 1.809002944888515, "step": 2150}, {"loss": 1.7892, "grad_norm": 0.40926849842071533, "learning_rate": 0.0002, "epoch": 1.8174169120740429, "step": 2160}, {"loss": 1.6823, "grad_norm": 0.3543093800544739, "learning_rate": 0.0002, "epoch": 1.8258308792595708, "step": 2170}, {"loss": 1.7812, "grad_norm": 0.42690935730934143, "learning_rate": 0.0002, "epoch": 1.8342448464450989, "step": 2180}, {"loss": 1.7471, "grad_norm": 0.40282756090164185, "learning_rate": 0.0002, "epoch": 1.842658813630627, "step": 2190}, {"loss": 1.7411, "grad_norm": 0.36568400263786316, "learning_rate": 0.0002, "epoch": 1.8510727808161547, "step": 2200}, {"loss": 1.7024, "grad_norm": 0.43159013986587524, "learning_rate": 0.0002, "epoch": 1.8594867480016828, "step": 2210}, {"loss": 1.7298, "grad_norm": 0.3554118573665619, "learning_rate": 0.0002, "epoch": 1.867900715187211, "step": 2220}, {"loss": 1.7157, "grad_norm": 0.43349072337150574, "learning_rate": 0.0002, "epoch": 1.8763146823727388, "step": 2230}, {"loss": 1.7302, "grad_norm": 0.36486536264419556, "learning_rate": 0.0002, "epoch": 1.8847286495582667, "step": 2240}, {"loss": 1.6901, "grad_norm": 0.39260047674179077, "learning_rate": 0.0002, "epoch": 1.8931426167437948, "step": 2250}, {"loss": 1.6691, "grad_norm": 0.3741776943206787, "learning_rate": 0.0002, "epoch": 1.9015565839293227, "step": 2260}, {"loss": 1.6931, "grad_norm": 0.3961946964263916, "learning_rate": 0.0002, "epoch": 1.9099705511148506, "step": 2270}, {"loss": 1.737, "grad_norm": 0.3659731149673462, "learning_rate": 0.0002, "epoch": 1.9183845183003787, "step": 2280}, {"loss": 1.7342, "grad_norm": 0.34744107723236084, "learning_rate": 0.0002, "epoch": 1.9267984854859066, "step": 2290}, {"loss": 1.7162, "grad_norm": 0.3607442378997803, "learning_rate": 0.0002, "epoch": 1.9352124526714345, "step": 2300}, {"loss": 1.6673, "grad_norm": 0.331464558839798, "learning_rate": 0.0002, "epoch": 1.9436264198569626, "step": 2310}, {"loss": 1.7101, "grad_norm": 0.3904414474964142, "learning_rate": 0.0002, "epoch": 1.9520403870424905, "step": 2320}, {"loss": 1.7327, "grad_norm": 0.37584832310676575, "learning_rate": 0.0002, "epoch": 1.9604543542280184, "step": 2330}, {"loss": 1.7586, "grad_norm": 0.3698684275150299, "learning_rate": 0.0002, "epoch": 1.9688683214135465, "step": 2340}, {"loss": 1.7764, "grad_norm": 0.40571412444114685, "learning_rate": 0.0002, "epoch": 1.9772822885990746, "step": 2350}, {"loss": 1.744, "grad_norm": 0.40059587359428406, "learning_rate": 0.0002, "epoch": 1.9856962557846023, "step": 2360}, {"loss": 1.7033, "grad_norm": 0.4168248474597931, "learning_rate": 0.0002, "epoch": 1.9941102229701304, "step": 2370}, {"eval_loss": 1.8055059909820557, "eval_runtime": 38.422, "eval_samples_per_second": 13.404, "eval_steps_per_second": 1.692, "epoch": 2.0, "step": 2377}, {"loss": 1.7673, "grad_norm": 0.35205352306365967, "learning_rate": 0.0002, "epoch": 2.0025241901556585, "step": 2380}, {"loss": 1.6556, "grad_norm": 0.3979377746582031, "learning_rate": 0.0002, "epoch": 2.010938157341186, "step": 2390}, {"loss": 1.6421, "grad_norm": 0.396491676568985, "learning_rate": 0.0002, "epoch": 2.0193521245267143, "step": 2400}, {"loss": 1.6847, "grad_norm": 0.44712209701538086, "learning_rate": 0.0002, "epoch": 2.0277660917122424, "step": 2410}, {"loss": 1.6877, "grad_norm": 0.4454420208930969, "learning_rate": 0.0002, "epoch": 2.03618005889777, "step": 2420}, {"loss": 1.6635, "grad_norm": 0.4170038402080536, "learning_rate": 0.0002, "epoch": 2.044594026083298, "step": 2430}, {"loss": 1.6512, "grad_norm": 0.4309595227241516, "learning_rate": 0.0002, "epoch": 2.0530079932688263, "step": 2440}, {"loss": 1.6223, "grad_norm": 0.4241602122783661, "learning_rate": 0.0002, "epoch": 2.0614219604543544, "step": 2450}, {"loss": 1.6162, "grad_norm": 0.4370540678501129, "learning_rate": 0.0002, "epoch": 2.069835927639882, "step": 2460}, {"loss": 1.6354, "grad_norm": 0.43985554575920105, "learning_rate": 0.0002, "epoch": 2.0782498948254102, "step": 2470}, {"loss": 1.6954, "grad_norm": 0.4158105254173279, "learning_rate": 0.0002, "epoch": 2.0866638620109383, "step": 2480}, {"loss": 1.6114, "grad_norm": 0.441549152135849, "learning_rate": 0.0002, "epoch": 2.095077829196466, "step": 2490}, {"loss": 1.5485, "grad_norm": 0.385718435049057, "learning_rate": 0.0002, "epoch": 2.103491796381994, "step": 2500}, {"loss": 1.5894, "grad_norm": 0.43146514892578125, "learning_rate": 0.0002, "epoch": 2.1119057635675222, "step": 2510}, {"loss": 1.6414, "grad_norm": 0.41663315892219543, "learning_rate": 0.0002, "epoch": 2.12031973075305, "step": 2520}, {"loss": 1.6527, "grad_norm": 0.4410698115825653, "learning_rate": 0.0002, "epoch": 2.128733697938578, "step": 2530}, {"loss": 1.6124, "grad_norm": 0.4472278952598572, "learning_rate": 0.0002, "epoch": 2.137147665124106, "step": 2540}, {"loss": 1.6257, "grad_norm": 0.3879167437553406, "learning_rate": 0.0002, "epoch": 2.145561632309634, "step": 2550}, {"loss": 1.6682, "grad_norm": 0.4212203025817871, "learning_rate": 0.0002, "epoch": 2.153975599495162, "step": 2560}, {"loss": 1.6036, "grad_norm": 0.42841723561286926, "learning_rate": 0.0002, "epoch": 2.16238956668069, "step": 2570}, {"loss": 1.5962, "grad_norm": 0.39272481203079224, "learning_rate": 0.0002, "epoch": 2.1708035338662177, "step": 2580}, {"loss": 1.681, "grad_norm": 0.4075261354446411, "learning_rate": 0.0002, "epoch": 2.179217501051746, "step": 2590}, {"loss": 1.6601, "grad_norm": 0.5358437895774841, "learning_rate": 0.0002, "epoch": 2.187631468237274, "step": 2600}, {"loss": 1.6423, "grad_norm": 0.4738350212574005, "learning_rate": 0.0002, "epoch": 2.1960454354228016, "step": 2610}, {"loss": 1.6386, "grad_norm": 0.446789026260376, "learning_rate": 0.0002, "epoch": 2.2044594026083297, "step": 2620}, {"loss": 1.6246, "grad_norm": 0.4615374505519867, "learning_rate": 0.0002, "epoch": 2.212873369793858, "step": 2630}, {"loss": 1.6205, "grad_norm": 0.46901994943618774, "learning_rate": 0.0002, "epoch": 2.221287336979386, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.46267789602279663, "learning_rate": 0.0002, "epoch": 2.2297013041649136, "step": 2650}, {"loss": 1.6584, "grad_norm": 0.4383080005645752, "learning_rate": 0.0002, "epoch": 2.2381152713504417, "step": 2660}, {"loss": 1.5745, "grad_norm": 0.4070609509944916, "learning_rate": 0.0002, "epoch": 2.24652923853597, "step": 2670}, {"loss": 1.6125, "grad_norm": 0.4572339951992035, "learning_rate": 0.0002, "epoch": 2.2549432057214975, "step": 2680}, {"loss": 1.5671, "grad_norm": 0.393265038728714, "learning_rate": 0.0002, "epoch": 2.2633571729070256, "step": 2690}, {"loss": 1.6239, "grad_norm": 0.46144717931747437, "learning_rate": 0.0002, "epoch": 2.2717711400925538, "step": 2700}, {"loss": 1.5992, "grad_norm": 0.45077767968177795, "learning_rate": 0.0002, "epoch": 2.2801851072780814, "step": 2710}, {"loss": 1.6261, "grad_norm": 0.5697639584541321, "learning_rate": 0.0002, "epoch": 2.2885990744636096, "step": 2720}, {"loss": 1.6192, "grad_norm": 0.4855510890483856, "learning_rate": 0.0002, "epoch": 2.2970130416491377, "step": 2730}, {"loss": 1.7419, "grad_norm": 0.4440622627735138, "learning_rate": 0.0002, "epoch": 2.3054270088346653, "step": 2740}, {"loss": 1.6496, "grad_norm": 0.3904096782207489, "learning_rate": 0.0002, "epoch": 2.3138409760201935, "step": 2750}, {"loss": 1.5888, "grad_norm": 0.5225510597229004, "learning_rate": 0.0002, "epoch": 2.3222549432057216, "step": 2760}, {"loss": 1.6082, "grad_norm": 0.44866397976875305, "learning_rate": 0.0002, "epoch": 2.3306689103912497, "step": 2770}, {"loss": 1.6087, "grad_norm": 0.5167056322097778, "learning_rate": 0.0002, "epoch": 2.3390828775767774, "step": 2780}, {"loss": 1.6136, "grad_norm": 0.45913267135620117, "learning_rate": 0.0002, "epoch": 2.3474968447623055, "step": 2790}, {"loss": 1.6564, "grad_norm": 0.45787590742111206, "learning_rate": 0.0002, "epoch": 2.3559108119478336, "step": 2800}, {"loss": 1.6868, "grad_norm": 0.4633352756500244, "learning_rate": 0.0002, "epoch": 2.3643247791333613, "step": 2810}, {"loss": 1.6316, "grad_norm": 0.46390071511268616, "learning_rate": 0.0002, "epoch": 2.3727387463188894, "step": 2820}, {"loss": 1.6039, "grad_norm": 0.4261005222797394, "learning_rate": 0.0002, "epoch": 2.3811527135044175, "step": 2830}, {"loss": 1.6364, "grad_norm": 0.4283634424209595, "learning_rate": 0.0002, "epoch": 2.389566680689945, "step": 2840}, {"loss": 1.6382, "grad_norm": 0.4955291450023651, "learning_rate": 0.0002, "epoch": 2.3979806478754733, "step": 2850}, {"loss": 1.6173, "grad_norm": 0.4740189015865326, "learning_rate": 0.0002, "epoch": 2.4063946150610014, "step": 2860}, {"loss": 1.6403, "grad_norm": 0.4222276508808136, "learning_rate": 0.0002, "epoch": 2.414808582246529, "step": 2870}, {"loss": 1.5602, "grad_norm": 0.4982149004936218, "learning_rate": 0.0002, "epoch": 2.423222549432057, "step": 2880}, {"loss": 1.6313, "grad_norm": 0.5217409133911133, "learning_rate": 0.0002, "epoch": 2.4316365166175853, "step": 2890}, {"loss": 1.5804, "grad_norm": 0.4555884897708893, "learning_rate": 0.0002, "epoch": 2.4400504838031134, "step": 2900}, {"loss": 1.6189, "grad_norm": 0.43178579211235046, "learning_rate": 0.0002, "epoch": 2.448464450988641, "step": 2910}, {"loss": 1.6824, "grad_norm": 0.4788478910923004, "learning_rate": 0.0002, "epoch": 2.456878418174169, "step": 2920}, {"loss": 1.6829, "grad_norm": 0.43689873814582825, "learning_rate": 0.0002, "epoch": 2.465292385359697, "step": 2930}, {"loss": 1.6196, "grad_norm": 0.5115197896957397, "learning_rate": 0.0002, "epoch": 2.473706352545225, "step": 2940}, {"loss": 1.689, "grad_norm": 0.5290159583091736, "learning_rate": 0.0002, "epoch": 2.482120319730753, "step": 2950}, {"loss": 1.6499, "grad_norm": 0.46042463183403015, "learning_rate": 0.0002, "epoch": 2.490534286916281, "step": 2960}, {"loss": 1.6664, "grad_norm": 0.4359915852546692, "learning_rate": 0.0002, "epoch": 2.498948254101809, "step": 2970}, {"loss": 1.5812, "grad_norm": 0.46352964639663696, "learning_rate": 0.0002, "epoch": 2.507362221287337, "step": 2980}, {"loss": 1.6501, "grad_norm": 0.5324268341064453, "learning_rate": 0.0002, "epoch": 2.515776188472865, "step": 2990}, {"loss": 1.6115, "grad_norm": 0.5929607152938843, "learning_rate": 0.0002, "epoch": 2.5241901556583928, "step": 3000}, {"loss": 1.6772, "grad_norm": 0.4811333417892456, "learning_rate": 0.0002, "epoch": 2.532604122843921, "step": 3010}, {"loss": 1.7023, "grad_norm": 0.4662701487541199, "learning_rate": 0.0002, "epoch": 2.541018090029449, "step": 3020}, {"loss": 1.5426, "grad_norm": 0.4582270681858063, "learning_rate": 0.0002, "epoch": 2.549432057214977, "step": 3030}, {"loss": 1.6737, "grad_norm": 0.4679982662200928, "learning_rate": 0.0002, "epoch": 2.557846024400505, "step": 3040}, {"loss": 1.5442, "grad_norm": 0.4380294680595398, "learning_rate": 0.0002, "epoch": 2.566259991586033, "step": 3050}, {"loss": 1.6055, "grad_norm": 0.44295763969421387, "learning_rate": 0.0002, "epoch": 2.5746739587715606, "step": 3060}, {"loss": 1.5775, "grad_norm": 0.5131027698516846, "learning_rate": 0.0002, "epoch": 2.5830879259570887, "step": 3070}, {"loss": 1.546, "grad_norm": 0.47567516565322876, "learning_rate": 0.0002, "epoch": 2.591501893142617, "step": 3080}, {"loss": 1.5671, "grad_norm": 0.49002596735954285, "learning_rate": 0.0002, "epoch": 2.599915860328145, "step": 3090}, {"loss": 1.5445, "grad_norm": 0.44856327772140503, "learning_rate": 0.0002, "epoch": 2.6083298275136726, "step": 3100}, {"loss": 1.5797, "grad_norm": 0.4480142593383789, "learning_rate": 0.0002, "epoch": 2.6167437946992007, "step": 3110}, {"loss": 1.7132, "grad_norm": 0.4317494034767151, "learning_rate": 0.0002, "epoch": 2.6251577618847284, "step": 3120}, {"loss": 1.6321, "grad_norm": 0.42580848932266235, "learning_rate": 0.0002, "epoch": 2.6335717290702565, "step": 3130}, {"loss": 1.6483, "grad_norm": 0.4516814947128296, "learning_rate": 0.0002, "epoch": 2.6419856962557846, "step": 3140}, {"loss": 1.695, "grad_norm": 0.4438435733318329, "learning_rate": 0.0002, "epoch": 2.6503996634413127, "step": 3150}, {"loss": 1.6938, "grad_norm": 0.4385356307029724, "learning_rate": 0.0002, "epoch": 2.6588136306268404, "step": 3160}, {"loss": 1.6139, "grad_norm": 0.5064112544059753, "learning_rate": 0.0002, "epoch": 2.6672275978123685, "step": 3170}, {"loss": 1.7189, "grad_norm": 0.49163177609443665, "learning_rate": 0.0002, "epoch": 2.6756415649978966, "step": 3180}, {"loss": 1.7323, "grad_norm": 0.49339258670806885, "learning_rate": 0.0002, "epoch": 2.6840555321834243, "step": 3190}, {"loss": 1.6508, "grad_norm": 0.440950870513916, "learning_rate": 0.0002, "epoch": 2.6924694993689524, "step": 3200}, {"loss": 1.6305, "grad_norm": 0.4283970594406128, "learning_rate": 0.0002, "epoch": 2.7008834665544805, "step": 3210}, {"loss": 1.5935, "grad_norm": 0.43875712156295776, "learning_rate": 0.0002, "epoch": 2.7092974337400086, "step": 3220}, {"loss": 1.6129, "grad_norm": 0.49332964420318604, "learning_rate": 0.0002, "epoch": 2.7177114009255363, "step": 3230}, {"loss": 1.642, "grad_norm": 0.5225692391395569, "learning_rate": 0.0002, "epoch": 2.7261253681110644, "step": 3240}, {"loss": 1.6759, "grad_norm": 0.4856489300727844, "learning_rate": 0.0002, "epoch": 2.734539335296592, "step": 3250}, {"loss": 1.6463, "grad_norm": 0.46918296813964844, "learning_rate": 0.0002, "epoch": 2.74295330248212, "step": 3260}, {"loss": 1.6819, "grad_norm": 0.4802931249141693, "learning_rate": 0.0002, "epoch": 2.7513672696676483, "step": 3270}, {"loss": 1.6246, "grad_norm": 0.4485355615615845, "learning_rate": 0.0002, "epoch": 2.7597812368531764, "step": 3280}, {"loss": 1.6251, "grad_norm": 0.43944594264030457, "learning_rate": 0.0002, "epoch": 2.768195204038704, "step": 3290}, {"loss": 1.6501, "grad_norm": 0.46847742795944214, "learning_rate": 0.0002, "epoch": 2.7766091712242322, "step": 3300}, {"loss": 1.5969, "grad_norm": 0.4816027879714966, "learning_rate": 0.0002, "epoch": 2.7850231384097603, "step": 3310}, {"loss": 1.6293, "grad_norm": 0.453960120677948, "learning_rate": 0.0002, "epoch": 2.793437105595288, "step": 3320}, {"loss": 1.6429, "grad_norm": 0.4816017150878906, "learning_rate": 0.0002, "epoch": 2.801851072780816, "step": 3330}, {"loss": 1.6683, "grad_norm": 0.4461034834384918, "learning_rate": 0.0002, "epoch": 2.8102650399663442, "step": 3340}, {"loss": 1.7048, "grad_norm": 0.48821821808815, "learning_rate": 0.0002, "epoch": 2.8186790071518724, "step": 3350}, {"loss": 1.6076, "grad_norm": 0.4574853777885437, "learning_rate": 0.0002, "epoch": 2.8270929743374, "step": 3360}, {"loss": 1.6651, "grad_norm": 0.42062026262283325, "learning_rate": 0.0002, "epoch": 2.835506941522928, "step": 3370}, {"loss": 1.624, "grad_norm": 0.4499834477901459, "learning_rate": 0.0002, "epoch": 2.843920908708456, "step": 3380}, {"loss": 1.621, "grad_norm": 0.4780360758304596, "learning_rate": 0.0002, "epoch": 2.852334875893984, "step": 3390}, {"loss": 1.5882, "grad_norm": 0.45422887802124023, "learning_rate": 0.0002, "epoch": 2.860748843079512, "step": 3400}, {"loss": 1.6028, "grad_norm": 0.4590015709400177, "learning_rate": 0.0002, "epoch": 2.86916281026504, "step": 3410}, {"loss": 1.6746, "grad_norm": 0.45689624547958374, "learning_rate": 0.0002, "epoch": 2.877576777450568, "step": 3420}, {"loss": 1.6326, "grad_norm": 0.46953922510147095, "learning_rate": 0.0002, "epoch": 2.885990744636096, "step": 3430}, {"loss": 1.6015, "grad_norm": 0.4791966378688812, "learning_rate": 0.0002, "epoch": 2.8944047118216236, "step": 3440}, {"loss": 1.694, "grad_norm": 0.4842296242713928, "learning_rate": 0.0002, "epoch": 2.9028186790071517, "step": 3450}, {"loss": 1.6326, "grad_norm": 0.47219768166542053, "learning_rate": 0.0002, "epoch": 2.91123264619268, "step": 3460}, {"loss": 1.6486, "grad_norm": 0.4622127115726471, "learning_rate": 0.0002, "epoch": 2.919646613378208, "step": 3470}, {"loss": 1.6485, "grad_norm": 0.46832820773124695, "learning_rate": 0.0002, "epoch": 2.9280605805637356, "step": 3480}, {"loss": 1.6366, "grad_norm": 0.44582483172416687, "learning_rate": 0.0002, "epoch": 2.9364745477492638, "step": 3490}, {"loss": 1.6859, "grad_norm": 0.4987219274044037, "learning_rate": 0.0002, "epoch": 2.944888514934792, "step": 3500}, {"loss": 1.5991, "grad_norm": 0.43750956654548645, "learning_rate": 0.0002, "epoch": 2.9533024821203195, "step": 3510}, {"loss": 1.6236, "grad_norm": 0.49962925910949707, "learning_rate": 0.0002, "epoch": 2.9617164493058477, "step": 3520}, {"loss": 1.5859, "grad_norm": 0.5189590454101562, "learning_rate": 0.0002, "epoch": 2.9701304164913758, "step": 3530}, {"loss": 1.6688, "grad_norm": 0.391317754983902, "learning_rate": 0.0002, "epoch": 2.978544383676904, "step": 3540}, {"loss": 1.5884, "grad_norm": 0.44934695959091187, "learning_rate": 0.0002, "epoch": 2.9869583508624316, "step": 3550}, {"loss": 1.5688, "grad_norm": 0.4740142226219177, "learning_rate": 0.0002, "epoch": 2.9953723180479597, "step": 3560}, {"eval_loss": 1.8266887664794922, "eval_runtime": 37.9445, "eval_samples_per_second": 13.572, "eval_steps_per_second": 1.713, "epoch": 2.9995793016407237, "step": 3565}, {"loss": 1.5939, "grad_norm": 0.4523724615573883, "learning_rate": 0.0002, "epoch": 3.003786285233488, "step": 3570}, {"loss": 1.526, "grad_norm": 0.5261380076408386, "learning_rate": 0.0002, "epoch": 3.0122002524190155, "step": 3580}, {"loss": 1.4946, "grad_norm": 0.48664888739585876, "learning_rate": 0.0002, "epoch": 3.0206142196045436, "step": 3590}, {"loss": 1.5193, "grad_norm": 0.5070882439613342, "learning_rate": 0.0002, "epoch": 3.0290281867900717, "step": 3600}, {"loss": 1.5316, "grad_norm": 0.5816011428833008, "learning_rate": 0.0002, "epoch": 3.0374421539755994, "step": 3610}, {"loss": 1.5682, "grad_norm": 0.6610211730003357, "learning_rate": 0.0002, "epoch": 3.0458561211611275, "step": 3620}, {"loss": 1.5699, "grad_norm": 0.5257703065872192, "learning_rate": 0.0002, "epoch": 3.0542700883466556, "step": 3630}, {"loss": 1.4438, "grad_norm": 0.5574390888214111, "learning_rate": 0.0002, "epoch": 3.0626840555321833, "step": 3640}, {"loss": 1.547, "grad_norm": 0.5682297348976135, "learning_rate": 0.0002, "epoch": 3.0710980227177114, "step": 3650}, {"loss": 1.5743, "grad_norm": 0.5798383355140686, "learning_rate": 0.0002, "epoch": 3.0795119899032395, "step": 3660}, {"loss": 1.4339, "grad_norm": 0.5458289980888367, "learning_rate": 0.0002, "epoch": 3.087925957088767, "step": 3670}, {"loss": 1.46, "grad_norm": 0.5599102973937988, "learning_rate": 0.0002, "epoch": 3.0963399242742953, "step": 3680}, {"loss": 1.4589, "grad_norm": 0.5023021697998047, "learning_rate": 0.0002, "epoch": 3.1047538914598234, "step": 3690}, {"loss": 1.5114, "grad_norm": 0.5448206067085266, "learning_rate": 0.0002, "epoch": 3.113167858645351, "step": 3700}, {"loss": 1.4692, "grad_norm": 0.5760458707809448, "learning_rate": 0.0002, "epoch": 3.121581825830879, "step": 3710}, {"loss": 1.4789, "grad_norm": 0.6018968224525452, "learning_rate": 0.0002, "epoch": 3.1299957930164073, "step": 3720}, {"loss": 1.5518, "grad_norm": 0.5767101049423218, "learning_rate": 0.0002, "epoch": 3.1384097602019354, "step": 3730}, {"loss": 1.5032, "grad_norm": 0.5333963632583618, "learning_rate": 0.0002, "epoch": 3.146823727387463, "step": 3740}, {"loss": 1.4812, "grad_norm": 0.5918396711349487, "learning_rate": 0.0002, "epoch": 3.155237694572991, "step": 3750}, {"loss": 1.4618, "grad_norm": 0.5931203365325928, "learning_rate": 0.0002, "epoch": 3.1636516617585193, "step": 3760}, {"loss": 1.5592, "grad_norm": 0.6562168598175049, "learning_rate": 0.0002, "epoch": 3.172065628944047, "step": 3770}, {"loss": 1.4932, "grad_norm": 0.5820156335830688, "learning_rate": 0.0002, "epoch": 3.180479596129575, "step": 3780}, {"loss": 1.4523, "grad_norm": 0.5784737467765808, "learning_rate": 0.0002, "epoch": 3.188893563315103, "step": 3790}, {"loss": 1.498, "grad_norm": 0.5506529808044434, "learning_rate": 0.0002, "epoch": 3.197307530500631, "step": 3800}, {"loss": 1.4819, "grad_norm": 0.6101595163345337, "learning_rate": 0.0002, "epoch": 3.205721497686159, "step": 3810}, {"loss": 1.5185, "grad_norm": 0.5597806572914124, "learning_rate": 0.0002, "epoch": 3.214135464871687, "step": 3820}, {"loss": 1.5664, "grad_norm": 0.5641011595726013, "learning_rate": 0.0002, "epoch": 3.222549432057215, "step": 3830}, {"loss": 1.4702, "grad_norm": 0.5892080068588257, "learning_rate": 0.0002, "epoch": 3.230963399242743, "step": 3840}, {"loss": 1.4194, "grad_norm": 0.6034760475158691, "learning_rate": 0.0002, "epoch": 3.239377366428271, "step": 3850}, {"loss": 1.5499, "grad_norm": 0.5112439393997192, "learning_rate": 0.0002, "epoch": 3.247791333613799, "step": 3860}, {"loss": 1.5132, "grad_norm": 0.56565922498703, "learning_rate": 0.0002, "epoch": 3.256205300799327, "step": 3870}, {"loss": 1.4892, "grad_norm": 0.6155247092247009, "learning_rate": 0.0002, "epoch": 3.264619267984855, "step": 3880}, {"loss": 1.5118, "grad_norm": 0.6064623594284058, "learning_rate": 0.0002, "epoch": 3.273033235170383, "step": 3890}, {"loss": 1.5236, "grad_norm": 0.6313768029212952, "learning_rate": 0.0002, "epoch": 3.2814472023559107, "step": 3900}, {"loss": 1.5551, "grad_norm": 0.5903939008712769, "learning_rate": 0.0002, "epoch": 3.289861169541439, "step": 3910}, {"loss": 1.5703, "grad_norm": 0.5770667195320129, "learning_rate": 0.0002, "epoch": 3.298275136726967, "step": 3920}, {"loss": 1.5159, "grad_norm": 0.5785196423530579, "learning_rate": 0.0002, "epoch": 3.3066891039124946, "step": 3930}, {"loss": 1.5277, "grad_norm": 0.6468310356140137, "learning_rate": 0.0002, "epoch": 3.3151030710980227, "step": 3940}, {"loss": 1.6002, "grad_norm": 0.6200279593467712, "learning_rate": 0.0002, "epoch": 3.323517038283551, "step": 3950}, {"loss": 1.5264, "grad_norm": 0.5779302716255188, "learning_rate": 0.0002, "epoch": 3.3319310054690785, "step": 3960}, {"loss": 1.4861, "grad_norm": 0.5463796854019165, "learning_rate": 0.0002, "epoch": 3.3403449726546066, "step": 3970}, {"loss": 1.541, "grad_norm": 0.6117855906486511, "learning_rate": 0.0002, "epoch": 3.3487589398401347, "step": 3980}, {"loss": 1.5566, "grad_norm": 0.5554766058921814, "learning_rate": 0.0002, "epoch": 3.357172907025663, "step": 3990}, {"loss": 1.5004, "grad_norm": 0.6012870073318481, "learning_rate": 0.0002, "epoch": 3.3655868742111905, "step": 4000}, {"loss": 1.473, "grad_norm": 0.5443974137306213, "learning_rate": 0.0002, "epoch": 3.3740008413967186, "step": 4010}, {"loss": 1.5139, "grad_norm": 0.6636057496070862, "learning_rate": 0.0002, "epoch": 3.3824148085822463, "step": 4020}, {"loss": 1.5141, "grad_norm": 0.5801246166229248, "learning_rate": 0.0002, "epoch": 3.3908287757677744, "step": 4030}, {"loss": 1.5026, "grad_norm": 0.5668839812278748, "learning_rate": 0.0002, "epoch": 3.3992427429533025, "step": 4040}, {"loss": 1.523, "grad_norm": 0.7763481736183167, "learning_rate": 0.0002, "epoch": 3.4076567101388306, "step": 4050}, {"loss": 1.4932, "grad_norm": 0.6675992608070374, "learning_rate": 0.0002, "epoch": 3.4160706773243583, "step": 4060}, {"loss": 1.4959, "grad_norm": 0.6290077567100525, "learning_rate": 0.0002, "epoch": 3.4244846445098864, "step": 4070}, {"loss": 1.5766, "grad_norm": 0.6040239930152893, "learning_rate": 0.0002, "epoch": 3.4328986116954145, "step": 4080}, {"loss": 1.5711, "grad_norm": 0.6237877607345581, "learning_rate": 0.0002, "epoch": 3.441312578880942, "step": 4090}, {"loss": 1.4961, "grad_norm": 0.5343508124351501, "learning_rate": 0.0002, "epoch": 3.4497265460664703, "step": 4100}, {"loss": 1.5123, "grad_norm": 0.6817412972450256, "learning_rate": 0.0002, "epoch": 3.4581405132519984, "step": 4110}, {"loss": 1.5377, "grad_norm": 0.7115170359611511, "learning_rate": 0.0002, "epoch": 3.466554480437526, "step": 4120}, {"loss": 1.5275, "grad_norm": 0.6127332448959351, "learning_rate": 0.0002, "epoch": 3.4749684476230542, "step": 4130}, {"loss": 1.557, "grad_norm": 0.5745994448661804, "learning_rate": 0.0002, "epoch": 3.4833824148085824, "step": 4140}, {"loss": 1.4873, "grad_norm": 0.6248795390129089, "learning_rate": 0.0002, "epoch": 3.49179638199411, "step": 4150}, {"loss": 1.4885, "grad_norm": 0.5821124911308289, "learning_rate": 0.0002, "epoch": 3.500210349179638, "step": 4160}, {"loss": 1.4937, "grad_norm": 0.561416506767273, "learning_rate": 0.0002, "epoch": 3.5086243163651663, "step": 4170}, {"loss": 1.5453, "grad_norm": 0.5848962664604187, "learning_rate": 0.0002, "epoch": 3.5170382835506944, "step": 4180}, {"loss": 1.5892, "grad_norm": 0.5335569977760315, "learning_rate": 0.0002, "epoch": 3.525452250736222, "step": 4190}, {"loss": 1.5152, "grad_norm": 0.547964870929718, "learning_rate": 0.0002, "epoch": 3.53386621792175, "step": 4200}, {"loss": 1.4887, "grad_norm": 0.6157727241516113, "learning_rate": 0.0002, "epoch": 3.542280185107278, "step": 4210}, {"loss": 1.5484, "grad_norm": 0.6163121461868286, "learning_rate": 0.0002, "epoch": 3.550694152292806, "step": 4220}, {"loss": 1.5833, "grad_norm": 0.5844616293907166, "learning_rate": 0.0002, "epoch": 3.559108119478334, "step": 4230}, {"loss": 1.5305, "grad_norm": 0.7104926109313965, "learning_rate": 0.0002, "epoch": 3.567522086663862, "step": 4240}, {"loss": 1.5161, "grad_norm": 0.5055213570594788, "learning_rate": 0.0002, "epoch": 3.57593605384939, "step": 4250}, {"loss": 1.482, "grad_norm": 0.611676812171936, "learning_rate": 0.0002, "epoch": 3.584350021034918, "step": 4260}, {"loss": 1.5048, "grad_norm": 0.6326440572738647, "learning_rate": 0.0002, "epoch": 3.592763988220446, "step": 4270}, {"loss": 1.5122, "grad_norm": 0.6290925741195679, "learning_rate": 0.0002, "epoch": 3.6011779554059737, "step": 4280}, {"loss": 1.5654, "grad_norm": 0.5691978931427002, "learning_rate": 0.0002, "epoch": 3.609591922591502, "step": 4290}, {"loss": 1.4854, "grad_norm": 0.6071329116821289, "learning_rate": 0.0002, "epoch": 3.61800588977703, "step": 4300}, {"loss": 1.5336, "grad_norm": 0.606573224067688, "learning_rate": 0.0002, "epoch": 3.626419856962558, "step": 4310}, {"loss": 1.6437, "grad_norm": 0.5515419244766235, "learning_rate": 0.0002, "epoch": 3.6348338241480858, "step": 4320}, {"loss": 1.498, "grad_norm": 0.5964660048484802, "learning_rate": 0.0002, "epoch": 3.643247791333614, "step": 4330}, {"loss": 1.544, "grad_norm": 0.5774146914482117, "learning_rate": 0.0002, "epoch": 3.6516617585191415, "step": 4340}, {"loss": 1.5566, "grad_norm": 0.5732731223106384, "learning_rate": 0.0002, "epoch": 3.6600757257046697, "step": 4350}, {"loss": 1.5682, "grad_norm": 0.7354163527488708, "learning_rate": 0.0002, "epoch": 3.6684896928901978, "step": 4360}, {"loss": 1.5225, "grad_norm": 0.6220902800559998, "learning_rate": 0.0002, "epoch": 3.676903660075726, "step": 4370}, {"loss": 1.4838, "grad_norm": 0.6053991317749023, "learning_rate": 0.0002, "epoch": 3.6853176272612536, "step": 4380}, {"loss": 1.5161, "grad_norm": 0.67010897397995, "learning_rate": 0.0002, "epoch": 3.6937315944467817, "step": 4390}, {"loss": 1.5381, "grad_norm": 0.6139186024665833, "learning_rate": 0.0002, "epoch": 3.70214556163231, "step": 4400}, {"loss": 1.5088, "grad_norm": 0.5433071851730347, "learning_rate": 0.0002, "epoch": 3.7105595288178375, "step": 4410}, {"loss": 1.5337, "grad_norm": 0.5453870296478271, "learning_rate": 0.0002, "epoch": 3.7189734960033656, "step": 4420}, {"loss": 1.4549, "grad_norm": 0.6401727199554443, "learning_rate": 0.0002, "epoch": 3.7273874631888937, "step": 4430}, {"loss": 1.503, "grad_norm": 0.6049367189407349, "learning_rate": 0.0002, "epoch": 3.735801430374422, "step": 4440}, {"loss": 1.5268, "grad_norm": 0.5740529298782349, "learning_rate": 0.0002, "epoch": 3.7442153975599495, "step": 4450}, {"loss": 1.5183, "grad_norm": 0.6521880626678467, "learning_rate": 0.0002, "epoch": 3.7526293647454776, "step": 4460}, {"loss": 1.5741, "grad_norm": 0.7096368074417114, "learning_rate": 0.0002, "epoch": 3.7610433319310053, "step": 4470}, {"loss": 1.5786, "grad_norm": 0.5886474251747131, "learning_rate": 0.0002, "epoch": 3.7694572991165334, "step": 4480}, {"loss": 1.5887, "grad_norm": 0.5821043252944946, "learning_rate": 0.0002, "epoch": 3.7778712663020615, "step": 4490}, {"loss": 1.5777, "grad_norm": 0.628892183303833, "learning_rate": 0.0002, "epoch": 3.7862852334875896, "step": 4500}, {"loss": 1.4708, "grad_norm": 0.5962669849395752, "learning_rate": 0.0002, "epoch": 3.7946992006731173, "step": 4510}, {"loss": 1.5267, "grad_norm": 0.6635549068450928, "learning_rate": 0.0002, "epoch": 3.8031131678586454, "step": 4520}, {"loss": 1.5058, "grad_norm": 0.6010760068893433, "learning_rate": 0.0002, "epoch": 3.811527135044173, "step": 4530}, {"loss": 1.6228, "grad_norm": 0.6322658658027649, "learning_rate": 0.0002, "epoch": 3.819941102229701, "step": 4540}, {"loss": 1.5029, "grad_norm": 0.5893137454986572, "learning_rate": 0.0002, "epoch": 3.8283550694152293, "step": 4550}, {"loss": 1.5435, "grad_norm": 0.7829602360725403, "learning_rate": 0.0002, "epoch": 3.8367690366007574, "step": 4560}, {"loss": 1.5453, "grad_norm": 0.6190396547317505, "learning_rate": 0.0002, "epoch": 3.845183003786285, "step": 4570}, {"loss": 1.5292, "grad_norm": 0.6662813425064087, "learning_rate": 0.0002, "epoch": 3.853596970971813, "step": 4580}, {"loss": 1.5065, "grad_norm": 0.5809855461120605, "learning_rate": 0.0002, "epoch": 3.8620109381573413, "step": 4590}, {"loss": 1.5041, "grad_norm": 0.5779069662094116, "learning_rate": 0.0002, "epoch": 3.870424905342869, "step": 4600}, {"loss": 1.498, "grad_norm": 0.5603038668632507, "learning_rate": 0.0002, "epoch": 3.878838872528397, "step": 4610}, {"loss": 1.5372, "grad_norm": 0.6274181008338928, "learning_rate": 0.0002, "epoch": 3.887252839713925, "step": 4620}, {"loss": 1.4996, "grad_norm": 0.6810959577560425, "learning_rate": 0.0002, "epoch": 3.8956668068994533, "step": 4630}, {"loss": 1.4956, "grad_norm": 0.5647315979003906, "learning_rate": 0.0002, "epoch": 3.904080774084981, "step": 4640}, {"loss": 1.5424, "grad_norm": 0.6830295324325562, "learning_rate": 0.0002, "epoch": 3.912494741270509, "step": 4650}, {"loss": 1.535, "grad_norm": 0.652565598487854, "learning_rate": 0.0002, "epoch": 3.920908708456037, "step": 4660}, {"loss": 1.4772, "grad_norm": 0.5806284546852112, "learning_rate": 0.0002, "epoch": 3.929322675641565, "step": 4670}, {"loss": 1.5812, "grad_norm": 0.6825073957443237, "learning_rate": 0.0002, "epoch": 3.937736642827093, "step": 4680}, {"loss": 1.5516, "grad_norm": 0.6149451732635498, "learning_rate": 0.0002, "epoch": 3.946150610012621, "step": 4690}, {"loss": 1.5608, "grad_norm": 0.6152557134628296, "learning_rate": 0.0002, "epoch": 3.954564577198149, "step": 4700}, {"loss": 1.4897, "grad_norm": 0.6239011883735657, "learning_rate": 0.0002, "epoch": 3.962978544383677, "step": 4710}, {"loss": 1.538, "grad_norm": 0.6485443115234375, "learning_rate": 0.0002, "epoch": 3.971392511569205, "step": 4720}, {"loss": 1.5226, "grad_norm": 0.6449228525161743, "learning_rate": 0.0002, "epoch": 3.9798064787547327, "step": 4730}, {"loss": 1.5087, "grad_norm": 0.6526407599449158, "learning_rate": 0.0002, "epoch": 3.988220445940261, "step": 4740}, {"loss": 1.5026, "grad_norm": 0.6277706027030945, "learning_rate": 0.0002, "epoch": 3.996634413125789, "step": 4750}]}
+{"epoch": 4.999579301640724, "step": 5942, "epoch_duration": 1272.9212999343872, "total_accumulated_duration": 7310.895057439804, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "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_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-0.8-num-5623-sd-42/checkpoint-2377", "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": 2.56, "grad_norm": 0.5458821654319763, "learning_rate": 0.0002, "epoch": 0.008413967185527976, "step": 10}, {"loss": 2.3235, "grad_norm": 0.7293308973312378, "learning_rate": 0.0002, "epoch": 0.016827934371055953, "step": 20}, {"loss": 2.0815, "grad_norm": 0.47792306542396545, "learning_rate": 0.0002, "epoch": 0.02524190155658393, "step": 30}, {"loss": 1.9718, "grad_norm": 0.5944402813911438, "learning_rate": 0.0002, "epoch": 0.033655868742111905, "step": 40}, {"loss": 1.8848, "grad_norm": 0.5415359735488892, "learning_rate": 0.0002, "epoch": 0.04206983592763988, "step": 50}, {"loss": 1.8953, "grad_norm": 0.535713791847229, "learning_rate": 0.0002, "epoch": 0.05048380311316786, "step": 60}, {"loss": 1.937, "grad_norm": 0.5184146761894226, "learning_rate": 0.0002, "epoch": 0.058897770298695834, "step": 70}, {"loss": 1.8396, "grad_norm": 0.458926796913147, "learning_rate": 0.0002, "epoch": 0.06731173748422381, "step": 80}, {"loss": 1.8677, "grad_norm": 0.4780142307281494, "learning_rate": 0.0002, "epoch": 0.07572570466975179, "step": 90}, {"loss": 1.8593, "grad_norm": 0.79965740442276, "learning_rate": 0.0002, "epoch": 0.08413967185527976, "step": 100}, {"loss": 1.9081, "grad_norm": 0.4498862028121948, "learning_rate": 0.0002, "epoch": 0.09255363904080774, "step": 110}, {"loss": 1.8503, "grad_norm": 0.39338430762290955, "learning_rate": 0.0002, "epoch": 0.10096760622633572, "step": 120}, {"loss": 1.8637, "grad_norm": 0.9588953852653503, "learning_rate": 0.0002, "epoch": 0.10938157341186369, "step": 130}, {"loss": 1.8676, "grad_norm": 0.41675639152526855, "learning_rate": 0.0002, "epoch": 0.11779554059739167, "step": 140}, {"loss": 1.8904, "grad_norm": 0.44519832730293274, "learning_rate": 0.0002, "epoch": 0.12620950778291964, "step": 150}, {"loss": 1.798, "grad_norm": 0.4176260530948639, "learning_rate": 0.0002, "epoch": 0.13462347496844762, "step": 160}, {"loss": 1.8398, "grad_norm": 0.35840365290641785, "learning_rate": 0.0002, "epoch": 0.1430374421539756, "step": 170}, {"loss": 1.8666, "grad_norm": 0.3794495463371277, "learning_rate": 0.0002, "epoch": 0.15145140933950357, "step": 180}, {"loss": 1.8111, "grad_norm": 0.4563522934913635, "learning_rate": 0.0002, "epoch": 0.15986537652503155, "step": 190}, {"loss": 1.8893, "grad_norm": 0.37057486176490784, "learning_rate": 0.0002, "epoch": 0.16827934371055953, "step": 200}, {"loss": 1.7995, "grad_norm": 0.44081518054008484, "learning_rate": 0.0002, "epoch": 0.1766933108960875, "step": 210}, {"loss": 1.9048, "grad_norm": 0.46078577637672424, "learning_rate": 0.0002, "epoch": 0.18510727808161548, "step": 220}, {"loss": 1.8403, "grad_norm": 0.36132094264030457, "learning_rate": 0.0002, "epoch": 0.19352124526714345, "step": 230}, {"loss": 1.8827, "grad_norm": 0.3747289180755615, "learning_rate": 0.0002, "epoch": 0.20193521245267143, "step": 240}, {"loss": 1.8382, "grad_norm": 0.3540179133415222, "learning_rate": 0.0002, "epoch": 0.2103491796381994, "step": 250}, {"loss": 1.8196, "grad_norm": 0.3461375832557678, "learning_rate": 0.0002, "epoch": 0.21876314682372738, "step": 260}, {"loss": 1.8509, "grad_norm": 0.3436960279941559, "learning_rate": 0.0002, "epoch": 0.22717711400925536, "step": 270}, {"loss": 1.8285, "grad_norm": 0.35403719544410706, "learning_rate": 0.0002, "epoch": 0.23559108119478334, "step": 280}, {"loss": 1.8369, "grad_norm": 0.37142616510391235, "learning_rate": 0.0002, "epoch": 0.2440050483803113, "step": 290}, {"loss": 1.8044, "grad_norm": 0.3307955861091614, "learning_rate": 0.0002, "epoch": 0.2524190155658393, "step": 300}, {"loss": 1.817, "grad_norm": 0.32855314016342163, "learning_rate": 0.0002, "epoch": 0.2608329827513673, "step": 310}, {"loss": 1.7803, "grad_norm": 0.3299003839492798, "learning_rate": 0.0002, "epoch": 0.26924694993689524, "step": 320}, {"loss": 1.8129, "grad_norm": 0.44311287999153137, "learning_rate": 0.0002, "epoch": 0.27766091712242325, "step": 330}, {"loss": 1.8232, "grad_norm": 0.32989758253097534, "learning_rate": 0.0002, "epoch": 0.2860748843079512, "step": 340}, {"loss": 1.7716, "grad_norm": 0.34400200843811035, "learning_rate": 0.0002, "epoch": 0.2944888514934792, "step": 350}, {"loss": 1.7619, "grad_norm": 0.36286211013793945, "learning_rate": 0.0002, "epoch": 0.30290281867900715, "step": 360}, {"loss": 1.8025, "grad_norm": 0.406827837228775, "learning_rate": 0.0002, "epoch": 0.31131678586453515, "step": 370}, {"loss": 1.7515, "grad_norm": 0.36299195885658264, "learning_rate": 0.0002, "epoch": 0.3197307530500631, "step": 380}, {"loss": 1.837, "grad_norm": 0.3477257192134857, "learning_rate": 0.0002, "epoch": 0.3281447202355911, "step": 390}, {"loss": 1.7767, "grad_norm": 0.3730369210243225, "learning_rate": 0.0002, "epoch": 0.33655868742111905, "step": 400}, {"loss": 1.7747, "grad_norm": 0.4644559919834137, "learning_rate": 0.0002, "epoch": 0.34497265460664706, "step": 410}, {"loss": 1.7538, "grad_norm": 0.406576544046402, "learning_rate": 0.0002, "epoch": 0.353386621792175, "step": 420}, {"loss": 1.7501, "grad_norm": 0.3612699508666992, "learning_rate": 0.0002, "epoch": 0.361800588977703, "step": 430}, {"loss": 1.7473, "grad_norm": 0.3243742287158966, "learning_rate": 0.0002, "epoch": 0.37021455616323096, "step": 440}, {"loss": 1.8851, "grad_norm": 0.36671221256256104, "learning_rate": 0.0002, "epoch": 0.37862852334875896, "step": 450}, {"loss": 1.8853, "grad_norm": 0.3565002381801605, "learning_rate": 0.0002, "epoch": 0.3870424905342869, "step": 460}, {"loss": 1.8923, "grad_norm": 0.34630221128463745, "learning_rate": 0.0002, "epoch": 0.3954564577198149, "step": 470}, {"loss": 1.8234, "grad_norm": 0.3353537321090698, "learning_rate": 0.0002, "epoch": 0.40387042490534286, "step": 480}, {"loss": 1.7135, "grad_norm": 0.4015921950340271, "learning_rate": 0.0002, "epoch": 0.41228439209087087, "step": 490}, {"loss": 1.7815, "grad_norm": 0.5489419102668762, "learning_rate": 0.0002, "epoch": 0.4206983592763988, "step": 500}, {"loss": 1.7903, "grad_norm": 0.4193589985370636, "learning_rate": 0.0002, "epoch": 0.4291123264619268, "step": 510}, {"loss": 1.8416, "grad_norm": 0.3418922424316406, "learning_rate": 0.0002, "epoch": 0.43752629364745477, "step": 520}, {"loss": 1.7982, "grad_norm": 0.32668185234069824, "learning_rate": 0.0002, "epoch": 0.44594026083298277, "step": 530}, {"loss": 1.7501, "grad_norm": 0.3094325661659241, "learning_rate": 0.0002, "epoch": 0.4543542280185107, "step": 540}, {"loss": 1.7438, "grad_norm": 0.3743017315864563, "learning_rate": 0.0002, "epoch": 0.4627681952040387, "step": 550}, {"loss": 1.8451, "grad_norm": 0.3295630216598511, "learning_rate": 0.0002, "epoch": 0.47118216238956667, "step": 560}, {"loss": 1.7529, "grad_norm": 1.6124513149261475, "learning_rate": 0.0002, "epoch": 0.4795961295750947, "step": 570}, {"loss": 1.8028, "grad_norm": 0.3245585858821869, "learning_rate": 0.0002, "epoch": 0.4880100967606226, "step": 580}, {"loss": 1.7976, "grad_norm": 0.3332934081554413, "learning_rate": 0.0002, "epoch": 0.49642406394615063, "step": 590}, {"loss": 1.7912, "grad_norm": 0.3836138844490051, "learning_rate": 0.0002, "epoch": 0.5048380311316786, "step": 600}, {"loss": 1.8347, "grad_norm": 0.32953888177871704, "learning_rate": 0.0002, "epoch": 0.5132519983172066, "step": 610}, {"loss": 1.7729, "grad_norm": 0.36291512846946716, "learning_rate": 0.0002, "epoch": 0.5216659655027346, "step": 620}, {"loss": 1.7758, "grad_norm": 0.3237783908843994, "learning_rate": 0.0002, "epoch": 0.5300799326882625, "step": 630}, {"loss": 1.8352, "grad_norm": 0.38882696628570557, "learning_rate": 0.0002, "epoch": 0.5384938998737905, "step": 640}, {"loss": 1.8624, "grad_norm": 0.37821972370147705, "learning_rate": 0.0002, "epoch": 0.5469078670593185, "step": 650}, {"loss": 1.8075, "grad_norm": 0.3556285500526428, "learning_rate": 0.0002, "epoch": 0.5553218342448465, "step": 660}, {"loss": 1.778, "grad_norm": 0.347499281167984, "learning_rate": 0.0002, "epoch": 0.5637358014303744, "step": 670}, {"loss": 1.8066, "grad_norm": 0.3176489472389221, "learning_rate": 0.0002, "epoch": 0.5721497686159024, "step": 680}, {"loss": 1.7257, "grad_norm": 0.30220088362693787, "learning_rate": 0.0002, "epoch": 0.5805637358014304, "step": 690}, {"loss": 1.8415, "grad_norm": 0.3711601793766022, "learning_rate": 0.0002, "epoch": 0.5889777029869584, "step": 700}, {"loss": 1.7906, "grad_norm": 0.3311759829521179, "learning_rate": 0.0002, "epoch": 0.5973916701724863, "step": 710}, {"loss": 1.7712, "grad_norm": 0.34824270009994507, "learning_rate": 0.0002, "epoch": 0.6058056373580143, "step": 720}, {"loss": 1.7954, "grad_norm": 0.29668381810188293, "learning_rate": 0.0002, "epoch": 0.6142196045435423, "step": 730}, {"loss": 1.8321, "grad_norm": 0.36087489128112793, "learning_rate": 0.0002, "epoch": 0.6226335717290703, "step": 740}, {"loss": 1.7956, "grad_norm": 0.31590089201927185, "learning_rate": 0.0002, "epoch": 0.6310475389145982, "step": 750}, {"loss": 1.7343, "grad_norm": 0.37632957100868225, "learning_rate": 0.0002, "epoch": 0.6394615061001262, "step": 760}, {"loss": 1.8499, "grad_norm": 0.3360748589038849, "learning_rate": 0.0002, "epoch": 0.6478754732856542, "step": 770}, {"loss": 1.8076, "grad_norm": 0.3420640528202057, "learning_rate": 0.0002, "epoch": 0.6562894404711822, "step": 780}, {"loss": 1.8353, "grad_norm": 0.5734959244728088, "learning_rate": 0.0002, "epoch": 0.6647034076567101, "step": 790}, {"loss": 1.7746, "grad_norm": 0.36440837383270264, "learning_rate": 0.0002, "epoch": 0.6731173748422381, "step": 800}, {"loss": 1.7532, "grad_norm": 0.3179708421230316, "learning_rate": 0.0002, "epoch": 0.6815313420277661, "step": 810}, {"loss": 1.7815, "grad_norm": 0.34122881293296814, "learning_rate": 0.0002, "epoch": 0.6899453092132941, "step": 820}, {"loss": 1.8167, "grad_norm": 0.31886112689971924, "learning_rate": 0.0002, "epoch": 0.698359276398822, "step": 830}, {"loss": 1.7505, "grad_norm": 0.31782326102256775, "learning_rate": 0.0002, "epoch": 0.70677324358435, "step": 840}, {"loss": 1.7588, "grad_norm": 0.36052989959716797, "learning_rate": 0.0002, "epoch": 0.715187210769878, "step": 850}, {"loss": 1.7891, "grad_norm": 0.28946155309677124, "learning_rate": 0.0002, "epoch": 0.723601177955406, "step": 860}, {"loss": 1.7923, "grad_norm": 0.3095663785934448, "learning_rate": 0.0002, "epoch": 0.7320151451409339, "step": 870}, {"loss": 1.785, "grad_norm": 0.3317491412162781, "learning_rate": 0.0002, "epoch": 0.7404291123264619, "step": 880}, {"loss": 1.7709, "grad_norm": 0.31324660778045654, "learning_rate": 0.0002, "epoch": 0.7488430795119899, "step": 890}, {"loss": 1.8753, "grad_norm": 0.3290475606918335, "learning_rate": 0.0002, "epoch": 0.7572570466975179, "step": 900}, {"loss": 1.7679, "grad_norm": 0.35690343379974365, "learning_rate": 0.0002, "epoch": 0.7656710138830458, "step": 910}, {"loss": 1.826, "grad_norm": 0.39558273553848267, "learning_rate": 0.0002, "epoch": 0.7740849810685738, "step": 920}, {"loss": 1.8722, "grad_norm": 0.34254348278045654, "learning_rate": 0.0002, "epoch": 0.7824989482541018, "step": 930}, {"loss": 1.7603, "grad_norm": 0.3560165464878082, "learning_rate": 0.0002, "epoch": 0.7909129154396298, "step": 940}, {"loss": 1.7992, "grad_norm": 0.30693164467811584, "learning_rate": 0.0002, "epoch": 0.7993268826251577, "step": 950}, {"loss": 1.8029, "grad_norm": 0.3394823372364044, "learning_rate": 0.0002, "epoch": 0.8077408498106857, "step": 960}, {"loss": 1.8105, "grad_norm": 0.3741514980792999, "learning_rate": 0.0002, "epoch": 0.8161548169962137, "step": 970}, {"loss": 1.7849, "grad_norm": 0.3655228316783905, "learning_rate": 0.0002, "epoch": 0.8245687841817417, "step": 980}, {"loss": 1.8449, "grad_norm": 0.3586033880710602, "learning_rate": 0.0002, "epoch": 0.8329827513672696, "step": 990}, {"loss": 1.7033, "grad_norm": 0.3459678888320923, "learning_rate": 0.0002, "epoch": 0.8413967185527976, "step": 1000}, {"loss": 1.8498, "grad_norm": 0.3184349834918976, "learning_rate": 0.0002, "epoch": 0.8498106857383256, "step": 1010}, {"loss": 1.7632, "grad_norm": 0.3099786043167114, "learning_rate": 0.0002, "epoch": 0.8582246529238536, "step": 1020}, {"loss": 1.8067, "grad_norm": 0.30300915241241455, "learning_rate": 0.0002, "epoch": 0.8666386201093815, "step": 1030}, {"loss": 1.7923, "grad_norm": 0.3128705620765686, "learning_rate": 0.0002, "epoch": 0.8750525872949095, "step": 1040}, {"loss": 1.8252, "grad_norm": 0.3336263597011566, "learning_rate": 0.0002, "epoch": 0.8834665544804375, "step": 1050}, {"loss": 1.8375, "grad_norm": 0.3801328241825104, "learning_rate": 0.0002, "epoch": 0.8918805216659655, "step": 1060}, {"loss": 1.7757, "grad_norm": 0.3122096359729767, "learning_rate": 0.0002, "epoch": 0.9002944888514934, "step": 1070}, {"loss": 1.8251, "grad_norm": 0.35990869998931885, "learning_rate": 0.0002, "epoch": 0.9087084560370214, "step": 1080}, {"loss": 1.7343, "grad_norm": 0.3321819305419922, "learning_rate": 0.0002, "epoch": 0.9171224232225494, "step": 1090}, {"loss": 1.7595, "grad_norm": 0.4202139377593994, "learning_rate": 0.0002, "epoch": 0.9255363904080774, "step": 1100}, {"loss": 1.8056, "grad_norm": 0.32559722661972046, "learning_rate": 0.0002, "epoch": 0.9339503575936053, "step": 1110}, {"loss": 1.812, "grad_norm": 0.3098459839820862, "learning_rate": 0.0002, "epoch": 0.9423643247791333, "step": 1120}, {"loss": 1.8252, "grad_norm": 0.33917108178138733, "learning_rate": 0.0002, "epoch": 0.9507782919646613, "step": 1130}, {"loss": 1.7709, "grad_norm": 0.4055837094783783, "learning_rate": 0.0002, "epoch": 0.9591922591501894, "step": 1140}, {"loss": 1.8259, "grad_norm": 0.32508623600006104, "learning_rate": 0.0002, "epoch": 0.9676062263357172, "step": 1150}, {"loss": 1.782, "grad_norm": 0.30150601267814636, "learning_rate": 0.0002, "epoch": 0.9760201935212452, "step": 1160}, {"loss": 1.8291, "grad_norm": 0.3042563199996948, "learning_rate": 0.0002, "epoch": 0.9844341607067733, "step": 1170}, {"loss": 1.7847, "grad_norm": 0.33254584670066833, "learning_rate": 0.0002, "epoch": 0.9928481278923013, "step": 1180}, {"eval_loss": 1.8077726364135742, "eval_runtime": 38.4359, "eval_samples_per_second": 13.399, "eval_steps_per_second": 1.691, "epoch": 0.9995793016407236, "step": 1188}, {"loss": 1.7414, "grad_norm": 0.35073035955429077, "learning_rate": 0.0002, "epoch": 1.0012620950778293, "step": 1190}, {"loss": 1.7483, "grad_norm": 0.3217269778251648, "learning_rate": 0.0002, "epoch": 1.0096760622633572, "step": 1200}, {"loss": 1.7517, "grad_norm": 0.3635033369064331, "learning_rate": 0.0002, "epoch": 1.018090029448885, "step": 1210}, {"loss": 1.6949, "grad_norm": 0.32468414306640625, "learning_rate": 0.0002, "epoch": 1.0265039966344132, "step": 1220}, {"loss": 1.711, "grad_norm": 0.3307163417339325, "learning_rate": 0.0002, "epoch": 1.034917963819941, "step": 1230}, {"loss": 1.7881, "grad_norm": 0.34381359815597534, "learning_rate": 0.0002, "epoch": 1.0433319310054692, "step": 1240}, {"loss": 1.612, "grad_norm": 0.35874804854393005, "learning_rate": 0.0002, "epoch": 1.051745898190997, "step": 1250}, {"loss": 1.7314, "grad_norm": 0.3615919351577759, "learning_rate": 0.0002, "epoch": 1.060159865376525, "step": 1260}, {"loss": 1.7517, "grad_norm": 0.32835808396339417, "learning_rate": 0.0002, "epoch": 1.068573832562053, "step": 1270}, {"loss": 1.7193, "grad_norm": 0.3876388370990753, "learning_rate": 0.0002, "epoch": 1.076987799747581, "step": 1280}, {"loss": 1.7442, "grad_norm": 0.39895930886268616, "learning_rate": 0.0002, "epoch": 1.0854017669331089, "step": 1290}, {"loss": 1.6601, "grad_norm": 0.39081698656082153, "learning_rate": 0.0002, "epoch": 1.093815734118637, "step": 1300}, {"loss": 1.7623, "grad_norm": 0.39974215626716614, "learning_rate": 0.0002, "epoch": 1.1022297013041649, "step": 1310}, {"loss": 1.7506, "grad_norm": 0.3887332081794739, "learning_rate": 0.0002, "epoch": 1.110643668489693, "step": 1320}, {"loss": 1.7381, "grad_norm": 0.36216408014297485, "learning_rate": 0.0002, "epoch": 1.1190576356752209, "step": 1330}, {"loss": 1.762, "grad_norm": 0.36979028582572937, "learning_rate": 0.0002, "epoch": 1.1274716028607488, "step": 1340}, {"loss": 1.7515, "grad_norm": 0.34052133560180664, "learning_rate": 0.0002, "epoch": 1.1358855700462769, "step": 1350}, {"loss": 1.7513, "grad_norm": 0.3467716574668884, "learning_rate": 0.0002, "epoch": 1.1442995372318048, "step": 1360}, {"loss": 1.7086, "grad_norm": 0.35528799891471863, "learning_rate": 0.0002, "epoch": 1.1527135044173327, "step": 1370}, {"loss": 1.794, "grad_norm": 0.36282262206077576, "learning_rate": 0.0002, "epoch": 1.1611274716028608, "step": 1380}, {"loss": 1.7731, "grad_norm": 0.37355899810791016, "learning_rate": 0.0002, "epoch": 1.1695414387883887, "step": 1390}, {"loss": 1.7483, "grad_norm": 0.37292736768722534, "learning_rate": 0.0002, "epoch": 1.1779554059739168, "step": 1400}, {"loss": 1.6916, "grad_norm": 0.5892812013626099, "learning_rate": 0.0002, "epoch": 1.1863693731594447, "step": 1410}, {"loss": 1.7302, "grad_norm": 0.3712292015552521, "learning_rate": 0.0002, "epoch": 1.1947833403449726, "step": 1420}, {"loss": 1.7709, "grad_norm": 0.3349577486515045, "learning_rate": 0.0002, "epoch": 1.2031973075305007, "step": 1430}, {"loss": 1.7412, "grad_norm": 0.32591062784194946, "learning_rate": 0.0002, "epoch": 1.2116112747160286, "step": 1440}, {"loss": 1.7406, "grad_norm": 0.3840635418891907, "learning_rate": 0.0002, "epoch": 1.2200252419015567, "step": 1450}, {"loss": 1.7276, "grad_norm": 0.37238365411758423, "learning_rate": 0.0002, "epoch": 1.2284392090870846, "step": 1460}, {"loss": 1.7052, "grad_norm": 0.3731217682361603, "learning_rate": 0.0002, "epoch": 1.2368531762726125, "step": 1470}, {"loss": 1.7255, "grad_norm": 0.3318967819213867, "learning_rate": 0.0002, "epoch": 1.2452671434581406, "step": 1480}, {"loss": 1.7463, "grad_norm": 0.3784034848213196, "learning_rate": 0.0002, "epoch": 1.2536811106436685, "step": 1490}, {"loss": 1.6862, "grad_norm": 0.3541383147239685, "learning_rate": 0.0002, "epoch": 1.2620950778291964, "step": 1500}, {"loss": 1.8394, "grad_norm": 0.35312485694885254, "learning_rate": 0.0002, "epoch": 1.2705090450147245, "step": 1510}, {"loss": 1.7029, "grad_norm": 0.35272929072380066, "learning_rate": 0.0002, "epoch": 1.2789230122002524, "step": 1520}, {"loss": 1.7016, "grad_norm": 0.40988272428512573, "learning_rate": 0.0002, "epoch": 1.2873369793857803, "step": 1530}, {"loss": 1.6912, "grad_norm": 0.3543946146965027, "learning_rate": 0.0002, "epoch": 1.2957509465713084, "step": 1540}, {"loss": 1.6757, "grad_norm": 0.35639145970344543, "learning_rate": 0.0002, "epoch": 1.3041649137568363, "step": 1550}, {"loss": 1.6814, "grad_norm": 0.3290826678276062, "learning_rate": 0.0002, "epoch": 1.3125788809423642, "step": 1560}, {"loss": 1.7369, "grad_norm": 0.39264336228370667, "learning_rate": 0.0002, "epoch": 1.3209928481278923, "step": 1570}, {"loss": 1.6804, "grad_norm": 0.5390415191650391, "learning_rate": 0.0002, "epoch": 1.3294068153134202, "step": 1580}, {"loss": 1.708, "grad_norm": 0.5188116431236267, "learning_rate": 0.0002, "epoch": 1.3378207824989483, "step": 1590}, {"loss": 1.6763, "grad_norm": 0.37445148825645447, "learning_rate": 0.0002, "epoch": 1.3462347496844762, "step": 1600}, {"loss": 1.7386, "grad_norm": 0.3296085298061371, "learning_rate": 0.0002, "epoch": 1.3546487168700043, "step": 1610}, {"loss": 1.8107, "grad_norm": 0.39879581332206726, "learning_rate": 0.0002, "epoch": 1.3630626840555322, "step": 1620}, {"loss": 1.6744, "grad_norm": 0.36092764139175415, "learning_rate": 0.0002, "epoch": 1.37147665124106, "step": 1630}, {"loss": 1.7144, "grad_norm": 0.37011823058128357, "learning_rate": 0.0002, "epoch": 1.3798906184265882, "step": 1640}, {"loss": 1.7396, "grad_norm": 0.40863534808158875, "learning_rate": 0.0002, "epoch": 1.3883045856121161, "step": 1650}, {"loss": 1.7901, "grad_norm": 0.337001770734787, "learning_rate": 0.0002, "epoch": 1.396718552797644, "step": 1660}, {"loss": 1.7044, "grad_norm": 0.35596707463264465, "learning_rate": 0.0002, "epoch": 1.4051325199831721, "step": 1670}, {"loss": 1.7717, "grad_norm": 0.3857671916484833, "learning_rate": 0.0002, "epoch": 1.4135464871687, "step": 1680}, {"loss": 1.7015, "grad_norm": 0.419502317905426, "learning_rate": 0.0002, "epoch": 1.421960454354228, "step": 1690}, {"loss": 1.7261, "grad_norm": 0.35459452867507935, "learning_rate": 0.0002, "epoch": 1.430374421539756, "step": 1700}, {"loss": 1.7361, "grad_norm": 0.37246978282928467, "learning_rate": 0.0002, "epoch": 1.438788388725284, "step": 1710}, {"loss": 1.6762, "grad_norm": 0.33091893792152405, "learning_rate": 0.0002, "epoch": 1.4472023559108118, "step": 1720}, {"loss": 1.7044, "grad_norm": 0.37029674649238586, "learning_rate": 0.0002, "epoch": 1.45561632309634, "step": 1730}, {"loss": 1.7117, "grad_norm": 0.374025821685791, "learning_rate": 0.0002, "epoch": 1.4640302902818678, "step": 1740}, {"loss": 1.7549, "grad_norm": 0.3416315019130707, "learning_rate": 0.0002, "epoch": 1.472444257467396, "step": 1750}, {"loss": 1.7093, "grad_norm": 0.36502841114997864, "learning_rate": 0.0002, "epoch": 1.4808582246529238, "step": 1760}, {"loss": 1.6597, "grad_norm": 0.35458803176879883, "learning_rate": 0.0002, "epoch": 1.489272191838452, "step": 1770}, {"loss": 1.675, "grad_norm": 0.4462839663028717, "learning_rate": 0.0002, "epoch": 1.4976861590239798, "step": 1780}, {"loss": 1.7267, "grad_norm": 0.34836092591285706, "learning_rate": 0.0002, "epoch": 1.5061001262095077, "step": 1790}, {"loss": 1.7295, "grad_norm": 0.3445749282836914, "learning_rate": 0.0002, "epoch": 1.5145140933950358, "step": 1800}, {"loss": 1.7386, "grad_norm": 0.36012160778045654, "learning_rate": 0.0002, "epoch": 1.5229280605805637, "step": 1810}, {"loss": 1.6594, "grad_norm": 0.4052616059780121, "learning_rate": 0.0002, "epoch": 1.5313420277660916, "step": 1820}, {"loss": 1.72, "grad_norm": 0.3966905474662781, "learning_rate": 0.0002, "epoch": 1.5397559949516197, "step": 1830}, {"loss": 1.7595, "grad_norm": 0.35028719902038574, "learning_rate": 0.0002, "epoch": 1.5481699621371476, "step": 1840}, {"loss": 1.6829, "grad_norm": 0.3936742842197418, "learning_rate": 0.0002, "epoch": 1.5565839293226755, "step": 1850}, {"loss": 1.7579, "grad_norm": 0.34473296999931335, "learning_rate": 0.0002, "epoch": 1.5649978965082036, "step": 1860}, {"loss": 1.7207, "grad_norm": 0.4328365623950958, "learning_rate": 0.0002, "epoch": 1.5734118636937318, "step": 1870}, {"loss": 1.7098, "grad_norm": 0.3566315472126007, "learning_rate": 0.0002, "epoch": 1.5818258308792594, "step": 1880}, {"loss": 1.6095, "grad_norm": 0.3301256597042084, "learning_rate": 0.0002, "epoch": 1.5902397980647875, "step": 1890}, {"loss": 1.748, "grad_norm": 0.3743041455745697, "learning_rate": 0.0002, "epoch": 1.5986537652503157, "step": 1900}, {"loss": 1.7259, "grad_norm": 0.3735344707965851, "learning_rate": 0.0002, "epoch": 1.6070677324358436, "step": 1910}, {"loss": 1.7445, "grad_norm": 0.42191144824028015, "learning_rate": 0.0002, "epoch": 1.6154816996213714, "step": 1920}, {"loss": 1.6978, "grad_norm": 0.3787207305431366, "learning_rate": 0.0002, "epoch": 1.6238956668068996, "step": 1930}, {"loss": 1.6893, "grad_norm": 0.35647350549697876, "learning_rate": 0.0002, "epoch": 1.6323096339924275, "step": 1940}, {"loss": 1.7825, "grad_norm": 0.39791446924209595, "learning_rate": 0.0002, "epoch": 1.6407236011779553, "step": 1950}, {"loss": 1.7293, "grad_norm": 0.37341275811195374, "learning_rate": 0.0002, "epoch": 1.6491375683634835, "step": 1960}, {"loss": 1.6781, "grad_norm": 0.3722686469554901, "learning_rate": 0.0002, "epoch": 1.6575515355490114, "step": 1970}, {"loss": 1.6383, "grad_norm": 0.37467387318611145, "learning_rate": 0.0002, "epoch": 1.6659655027345392, "step": 1980}, {"loss": 1.7439, "grad_norm": 0.37109461426734924, "learning_rate": 0.0002, "epoch": 1.6743794699200674, "step": 1990}, {"loss": 1.7206, "grad_norm": 0.4008837044239044, "learning_rate": 0.0002, "epoch": 1.6827934371055953, "step": 2000}, {"loss": 1.7604, "grad_norm": 0.3316999673843384, "learning_rate": 0.0002, "epoch": 1.6912074042911232, "step": 2010}, {"loss": 1.7325, "grad_norm": 0.3683805465698242, "learning_rate": 0.0002, "epoch": 1.6996213714766513, "step": 2020}, {"loss": 1.7451, "grad_norm": 0.4163658320903778, "learning_rate": 0.0002, "epoch": 1.7080353386621794, "step": 2030}, {"loss": 1.741, "grad_norm": 0.4245431125164032, "learning_rate": 0.0002, "epoch": 1.716449305847707, "step": 2040}, {"loss": 1.7184, "grad_norm": 0.36732038855552673, "learning_rate": 0.0002, "epoch": 1.7248632730332352, "step": 2050}, {"loss": 1.7031, "grad_norm": 0.34981656074523926, "learning_rate": 0.0002, "epoch": 1.7332772402187633, "step": 2060}, {"loss": 1.7545, "grad_norm": 0.38588812947273254, "learning_rate": 0.0002, "epoch": 1.7416912074042912, "step": 2070}, {"loss": 1.7728, "grad_norm": 0.39914557337760925, "learning_rate": 0.0002, "epoch": 1.750105174589819, "step": 2080}, {"loss": 1.7049, "grad_norm": 0.36068692803382874, "learning_rate": 0.0002, "epoch": 1.7585191417753472, "step": 2090}, {"loss": 1.7537, "grad_norm": 0.3983287215232849, "learning_rate": 0.0002, "epoch": 1.766933108960875, "step": 2100}, {"loss": 1.7016, "grad_norm": 0.45008400082588196, "learning_rate": 0.0002, "epoch": 1.775347076146403, "step": 2110}, {"loss": 1.7163, "grad_norm": 0.3618052303791046, "learning_rate": 0.0002, "epoch": 1.783761043331931, "step": 2120}, {"loss": 1.7335, "grad_norm": 0.38745400309562683, "learning_rate": 0.0002, "epoch": 1.792175010517459, "step": 2130}, {"loss": 1.7387, "grad_norm": 0.3413826525211334, "learning_rate": 0.0002, "epoch": 1.8005889777029869, "step": 2140}, {"loss": 1.7414, "grad_norm": 0.35983747243881226, "learning_rate": 0.0002, "epoch": 1.809002944888515, "step": 2150}, {"loss": 1.7892, "grad_norm": 0.40926849842071533, "learning_rate": 0.0002, "epoch": 1.8174169120740429, "step": 2160}, {"loss": 1.6823, "grad_norm": 0.3543093800544739, "learning_rate": 0.0002, "epoch": 1.8258308792595708, "step": 2170}, {"loss": 1.7812, "grad_norm": 0.42690935730934143, "learning_rate": 0.0002, "epoch": 1.8342448464450989, "step": 2180}, {"loss": 1.7471, "grad_norm": 0.40282756090164185, "learning_rate": 0.0002, "epoch": 1.842658813630627, "step": 2190}, {"loss": 1.7411, "grad_norm": 0.36568400263786316, "learning_rate": 0.0002, "epoch": 1.8510727808161547, "step": 2200}, {"loss": 1.7024, "grad_norm": 0.43159013986587524, "learning_rate": 0.0002, "epoch": 1.8594867480016828, "step": 2210}, {"loss": 1.7298, "grad_norm": 0.3554118573665619, "learning_rate": 0.0002, "epoch": 1.867900715187211, "step": 2220}, {"loss": 1.7157, "grad_norm": 0.43349072337150574, "learning_rate": 0.0002, "epoch": 1.8763146823727388, "step": 2230}, {"loss": 1.7302, "grad_norm": 0.36486536264419556, "learning_rate": 0.0002, "epoch": 1.8847286495582667, "step": 2240}, {"loss": 1.6901, "grad_norm": 0.39260047674179077, "learning_rate": 0.0002, "epoch": 1.8931426167437948, "step": 2250}, {"loss": 1.6691, "grad_norm": 0.3741776943206787, "learning_rate": 0.0002, "epoch": 1.9015565839293227, "step": 2260}, {"loss": 1.6931, "grad_norm": 0.3961946964263916, "learning_rate": 0.0002, "epoch": 1.9099705511148506, "step": 2270}, {"loss": 1.737, "grad_norm": 0.3659731149673462, "learning_rate": 0.0002, "epoch": 1.9183845183003787, "step": 2280}, {"loss": 1.7342, "grad_norm": 0.34744107723236084, "learning_rate": 0.0002, "epoch": 1.9267984854859066, "step": 2290}, {"loss": 1.7162, "grad_norm": 0.3607442378997803, "learning_rate": 0.0002, "epoch": 1.9352124526714345, "step": 2300}, {"loss": 1.6673, "grad_norm": 0.331464558839798, "learning_rate": 0.0002, "epoch": 1.9436264198569626, "step": 2310}, {"loss": 1.7101, "grad_norm": 0.3904414474964142, "learning_rate": 0.0002, "epoch": 1.9520403870424905, "step": 2320}, {"loss": 1.7327, "grad_norm": 0.37584832310676575, "learning_rate": 0.0002, "epoch": 1.9604543542280184, "step": 2330}, {"loss": 1.7586, "grad_norm": 0.3698684275150299, "learning_rate": 0.0002, "epoch": 1.9688683214135465, "step": 2340}, {"loss": 1.7764, "grad_norm": 0.40571412444114685, "learning_rate": 0.0002, "epoch": 1.9772822885990746, "step": 2350}, {"loss": 1.744, "grad_norm": 0.40059587359428406, "learning_rate": 0.0002, "epoch": 1.9856962557846023, "step": 2360}, {"loss": 1.7033, "grad_norm": 0.4168248474597931, "learning_rate": 0.0002, "epoch": 1.9941102229701304, "step": 2370}, {"eval_loss": 1.8055059909820557, "eval_runtime": 38.422, "eval_samples_per_second": 13.404, "eval_steps_per_second": 1.692, "epoch": 2.0, "step": 2377}, {"loss": 1.7673, "grad_norm": 0.35205352306365967, "learning_rate": 0.0002, "epoch": 2.0025241901556585, "step": 2380}, {"loss": 1.6556, "grad_norm": 0.3979377746582031, "learning_rate": 0.0002, "epoch": 2.010938157341186, "step": 2390}, {"loss": 1.6421, "grad_norm": 0.396491676568985, "learning_rate": 0.0002, "epoch": 2.0193521245267143, "step": 2400}, {"loss": 1.6847, "grad_norm": 0.44712209701538086, "learning_rate": 0.0002, "epoch": 2.0277660917122424, "step": 2410}, {"loss": 1.6877, "grad_norm": 0.4454420208930969, "learning_rate": 0.0002, "epoch": 2.03618005889777, "step": 2420}, {"loss": 1.6635, "grad_norm": 0.4170038402080536, "learning_rate": 0.0002, "epoch": 2.044594026083298, "step": 2430}, {"loss": 1.6512, "grad_norm": 0.4309595227241516, "learning_rate": 0.0002, "epoch": 2.0530079932688263, "step": 2440}, {"loss": 1.6223, "grad_norm": 0.4241602122783661, "learning_rate": 0.0002, "epoch": 2.0614219604543544, "step": 2450}, {"loss": 1.6162, "grad_norm": 0.4370540678501129, "learning_rate": 0.0002, "epoch": 2.069835927639882, "step": 2460}, {"loss": 1.6354, "grad_norm": 0.43985554575920105, "learning_rate": 0.0002, "epoch": 2.0782498948254102, "step": 2470}, {"loss": 1.6954, "grad_norm": 0.4158105254173279, "learning_rate": 0.0002, "epoch": 2.0866638620109383, "step": 2480}, {"loss": 1.6114, "grad_norm": 0.441549152135849, "learning_rate": 0.0002, "epoch": 2.095077829196466, "step": 2490}, {"loss": 1.5485, "grad_norm": 0.385718435049057, "learning_rate": 0.0002, "epoch": 2.103491796381994, "step": 2500}, {"loss": 1.5894, "grad_norm": 0.43146514892578125, "learning_rate": 0.0002, "epoch": 2.1119057635675222, "step": 2510}, {"loss": 1.6414, "grad_norm": 0.41663315892219543, "learning_rate": 0.0002, "epoch": 2.12031973075305, "step": 2520}, {"loss": 1.6527, "grad_norm": 0.4410698115825653, "learning_rate": 0.0002, "epoch": 2.128733697938578, "step": 2530}, {"loss": 1.6124, "grad_norm": 0.4472278952598572, "learning_rate": 0.0002, "epoch": 2.137147665124106, "step": 2540}, {"loss": 1.6257, "grad_norm": 0.3879167437553406, "learning_rate": 0.0002, "epoch": 2.145561632309634, "step": 2550}, {"loss": 1.6682, "grad_norm": 0.4212203025817871, "learning_rate": 0.0002, "epoch": 2.153975599495162, "step": 2560}, {"loss": 1.6036, "grad_norm": 0.42841723561286926, "learning_rate": 0.0002, "epoch": 2.16238956668069, "step": 2570}, {"loss": 1.5962, "grad_norm": 0.39272481203079224, "learning_rate": 0.0002, "epoch": 2.1708035338662177, "step": 2580}, {"loss": 1.681, "grad_norm": 0.4075261354446411, "learning_rate": 0.0002, "epoch": 2.179217501051746, "step": 2590}, {"loss": 1.6601, "grad_norm": 0.5358437895774841, "learning_rate": 0.0002, "epoch": 2.187631468237274, "step": 2600}, {"loss": 1.6423, "grad_norm": 0.4738350212574005, "learning_rate": 0.0002, "epoch": 2.1960454354228016, "step": 2610}, {"loss": 1.6386, "grad_norm": 0.446789026260376, "learning_rate": 0.0002, "epoch": 2.2044594026083297, "step": 2620}, {"loss": 1.6246, "grad_norm": 0.4615374505519867, "learning_rate": 0.0002, "epoch": 2.212873369793858, "step": 2630}, {"loss": 1.6205, "grad_norm": 0.46901994943618774, "learning_rate": 0.0002, "epoch": 2.221287336979386, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.46267789602279663, "learning_rate": 0.0002, "epoch": 2.2297013041649136, "step": 2650}, {"loss": 1.6584, "grad_norm": 0.4383080005645752, "learning_rate": 0.0002, "epoch": 2.2381152713504417, "step": 2660}, {"loss": 1.5745, "grad_norm": 0.4070609509944916, "learning_rate": 0.0002, "epoch": 2.24652923853597, "step": 2670}, {"loss": 1.6125, "grad_norm": 0.4572339951992035, "learning_rate": 0.0002, "epoch": 2.2549432057214975, "step": 2680}, {"loss": 1.5671, "grad_norm": 0.393265038728714, "learning_rate": 0.0002, "epoch": 2.2633571729070256, "step": 2690}, {"loss": 1.6239, "grad_norm": 0.46144717931747437, "learning_rate": 0.0002, "epoch": 2.2717711400925538, "step": 2700}, {"loss": 1.5992, "grad_norm": 0.45077767968177795, "learning_rate": 0.0002, "epoch": 2.2801851072780814, "step": 2710}, {"loss": 1.6261, "grad_norm": 0.5697639584541321, "learning_rate": 0.0002, "epoch": 2.2885990744636096, "step": 2720}, {"loss": 1.6192, "grad_norm": 0.4855510890483856, "learning_rate": 0.0002, "epoch": 2.2970130416491377, "step": 2730}, {"loss": 1.7419, "grad_norm": 0.4440622627735138, "learning_rate": 0.0002, "epoch": 2.3054270088346653, "step": 2740}, {"loss": 1.6496, "grad_norm": 0.3904096782207489, "learning_rate": 0.0002, "epoch": 2.3138409760201935, "step": 2750}, {"loss": 1.5888, "grad_norm": 0.5225510597229004, "learning_rate": 0.0002, "epoch": 2.3222549432057216, "step": 2760}, {"loss": 1.6082, "grad_norm": 0.44866397976875305, "learning_rate": 0.0002, "epoch": 2.3306689103912497, "step": 2770}, {"loss": 1.6087, "grad_norm": 0.5167056322097778, "learning_rate": 0.0002, "epoch": 2.3390828775767774, "step": 2780}, {"loss": 1.6136, "grad_norm": 0.45913267135620117, "learning_rate": 0.0002, "epoch": 2.3474968447623055, "step": 2790}, {"loss": 1.6564, "grad_norm": 0.45787590742111206, "learning_rate": 0.0002, "epoch": 2.3559108119478336, "step": 2800}, {"loss": 1.6868, "grad_norm": 0.4633352756500244, "learning_rate": 0.0002, "epoch": 2.3643247791333613, "step": 2810}, {"loss": 1.6316, "grad_norm": 0.46390071511268616, "learning_rate": 0.0002, "epoch": 2.3727387463188894, "step": 2820}, {"loss": 1.6039, "grad_norm": 0.4261005222797394, "learning_rate": 0.0002, "epoch": 2.3811527135044175, "step": 2830}, {"loss": 1.6364, "grad_norm": 0.4283634424209595, "learning_rate": 0.0002, "epoch": 2.389566680689945, "step": 2840}, {"loss": 1.6382, "grad_norm": 0.4955291450023651, "learning_rate": 0.0002, "epoch": 2.3979806478754733, "step": 2850}, {"loss": 1.6173, "grad_norm": 0.4740189015865326, "learning_rate": 0.0002, "epoch": 2.4063946150610014, "step": 2860}, {"loss": 1.6403, "grad_norm": 0.4222276508808136, "learning_rate": 0.0002, "epoch": 2.414808582246529, "step": 2870}, {"loss": 1.5602, "grad_norm": 0.4982149004936218, "learning_rate": 0.0002, "epoch": 2.423222549432057, "step": 2880}, {"loss": 1.6313, "grad_norm": 0.5217409133911133, "learning_rate": 0.0002, "epoch": 2.4316365166175853, "step": 2890}, {"loss": 1.5804, "grad_norm": 0.4555884897708893, "learning_rate": 0.0002, "epoch": 2.4400504838031134, "step": 2900}, {"loss": 1.6189, "grad_norm": 0.43178579211235046, "learning_rate": 0.0002, "epoch": 2.448464450988641, "step": 2910}, {"loss": 1.6824, "grad_norm": 0.4788478910923004, "learning_rate": 0.0002, "epoch": 2.456878418174169, "step": 2920}, {"loss": 1.6829, "grad_norm": 0.43689873814582825, "learning_rate": 0.0002, "epoch": 2.465292385359697, "step": 2930}, {"loss": 1.6196, "grad_norm": 0.5115197896957397, "learning_rate": 0.0002, "epoch": 2.473706352545225, "step": 2940}, {"loss": 1.689, "grad_norm": 0.5290159583091736, "learning_rate": 0.0002, "epoch": 2.482120319730753, "step": 2950}, {"loss": 1.6499, "grad_norm": 0.46042463183403015, "learning_rate": 0.0002, "epoch": 2.490534286916281, "step": 2960}, {"loss": 1.6664, "grad_norm": 0.4359915852546692, "learning_rate": 0.0002, "epoch": 2.498948254101809, "step": 2970}, {"loss": 1.5812, "grad_norm": 0.46352964639663696, "learning_rate": 0.0002, "epoch": 2.507362221287337, "step": 2980}, {"loss": 1.6501, "grad_norm": 0.5324268341064453, "learning_rate": 0.0002, "epoch": 2.515776188472865, "step": 2990}, {"loss": 1.6115, "grad_norm": 0.5929607152938843, "learning_rate": 0.0002, "epoch": 2.5241901556583928, "step": 3000}, {"loss": 1.6772, "grad_norm": 0.4811333417892456, "learning_rate": 0.0002, "epoch": 2.532604122843921, "step": 3010}, {"loss": 1.7023, "grad_norm": 0.4662701487541199, "learning_rate": 0.0002, "epoch": 2.541018090029449, "step": 3020}, {"loss": 1.5426, "grad_norm": 0.4582270681858063, "learning_rate": 0.0002, "epoch": 2.549432057214977, "step": 3030}, {"loss": 1.6737, "grad_norm": 0.4679982662200928, "learning_rate": 0.0002, "epoch": 2.557846024400505, "step": 3040}, {"loss": 1.5442, "grad_norm": 0.4380294680595398, "learning_rate": 0.0002, "epoch": 2.566259991586033, "step": 3050}, {"loss": 1.6055, "grad_norm": 0.44295763969421387, "learning_rate": 0.0002, "epoch": 2.5746739587715606, "step": 3060}, {"loss": 1.5775, "grad_norm": 0.5131027698516846, "learning_rate": 0.0002, "epoch": 2.5830879259570887, "step": 3070}, {"loss": 1.546, "grad_norm": 0.47567516565322876, "learning_rate": 0.0002, "epoch": 2.591501893142617, "step": 3080}, {"loss": 1.5671, "grad_norm": 0.49002596735954285, "learning_rate": 0.0002, "epoch": 2.599915860328145, "step": 3090}, {"loss": 1.5445, "grad_norm": 0.44856327772140503, "learning_rate": 0.0002, "epoch": 2.6083298275136726, "step": 3100}, {"loss": 1.5797, "grad_norm": 0.4480142593383789, "learning_rate": 0.0002, "epoch": 2.6167437946992007, "step": 3110}, {"loss": 1.7132, "grad_norm": 0.4317494034767151, "learning_rate": 0.0002, "epoch": 2.6251577618847284, "step": 3120}, {"loss": 1.6321, "grad_norm": 0.42580848932266235, "learning_rate": 0.0002, "epoch": 2.6335717290702565, "step": 3130}, {"loss": 1.6483, "grad_norm": 0.4516814947128296, "learning_rate": 0.0002, "epoch": 2.6419856962557846, "step": 3140}, {"loss": 1.695, "grad_norm": 0.4438435733318329, "learning_rate": 0.0002, "epoch": 2.6503996634413127, "step": 3150}, {"loss": 1.6938, "grad_norm": 0.4385356307029724, "learning_rate": 0.0002, "epoch": 2.6588136306268404, "step": 3160}, {"loss": 1.6139, "grad_norm": 0.5064112544059753, "learning_rate": 0.0002, "epoch": 2.6672275978123685, "step": 3170}, {"loss": 1.7189, "grad_norm": 0.49163177609443665, "learning_rate": 0.0002, "epoch": 2.6756415649978966, "step": 3180}, {"loss": 1.7323, "grad_norm": 0.49339258670806885, "learning_rate": 0.0002, "epoch": 2.6840555321834243, "step": 3190}, {"loss": 1.6508, "grad_norm": 0.440950870513916, "learning_rate": 0.0002, "epoch": 2.6924694993689524, "step": 3200}, {"loss": 1.6305, "grad_norm": 0.4283970594406128, "learning_rate": 0.0002, "epoch": 2.7008834665544805, "step": 3210}, {"loss": 1.5935, "grad_norm": 0.43875712156295776, "learning_rate": 0.0002, "epoch": 2.7092974337400086, "step": 3220}, {"loss": 1.6129, "grad_norm": 0.49332964420318604, "learning_rate": 0.0002, "epoch": 2.7177114009255363, "step": 3230}, {"loss": 1.642, "grad_norm": 0.5225692391395569, "learning_rate": 0.0002, "epoch": 2.7261253681110644, "step": 3240}, {"loss": 1.6759, "grad_norm": 0.4856489300727844, "learning_rate": 0.0002, "epoch": 2.734539335296592, "step": 3250}, {"loss": 1.6463, "grad_norm": 0.46918296813964844, "learning_rate": 0.0002, "epoch": 2.74295330248212, "step": 3260}, {"loss": 1.6819, "grad_norm": 0.4802931249141693, "learning_rate": 0.0002, "epoch": 2.7513672696676483, "step": 3270}, {"loss": 1.6246, "grad_norm": 0.4485355615615845, "learning_rate": 0.0002, "epoch": 2.7597812368531764, "step": 3280}, {"loss": 1.6251, "grad_norm": 0.43944594264030457, "learning_rate": 0.0002, "epoch": 2.768195204038704, "step": 3290}, {"loss": 1.6501, "grad_norm": 0.46847742795944214, "learning_rate": 0.0002, "epoch": 2.7766091712242322, "step": 3300}, {"loss": 1.5969, "grad_norm": 0.4816027879714966, "learning_rate": 0.0002, "epoch": 2.7850231384097603, "step": 3310}, {"loss": 1.6293, "grad_norm": 0.453960120677948, "learning_rate": 0.0002, "epoch": 2.793437105595288, "step": 3320}, {"loss": 1.6429, "grad_norm": 0.4816017150878906, "learning_rate": 0.0002, "epoch": 2.801851072780816, "step": 3330}, {"loss": 1.6683, "grad_norm": 0.4461034834384918, "learning_rate": 0.0002, "epoch": 2.8102650399663442, "step": 3340}, {"loss": 1.7048, "grad_norm": 0.48821821808815, "learning_rate": 0.0002, "epoch": 2.8186790071518724, "step": 3350}, {"loss": 1.6076, "grad_norm": 0.4574853777885437, "learning_rate": 0.0002, "epoch": 2.8270929743374, "step": 3360}, {"loss": 1.6651, "grad_norm": 0.42062026262283325, "learning_rate": 0.0002, "epoch": 2.835506941522928, "step": 3370}, {"loss": 1.624, "grad_norm": 0.4499834477901459, "learning_rate": 0.0002, "epoch": 2.843920908708456, "step": 3380}, {"loss": 1.621, "grad_norm": 0.4780360758304596, "learning_rate": 0.0002, "epoch": 2.852334875893984, "step": 3390}, {"loss": 1.5882, "grad_norm": 0.45422887802124023, "learning_rate": 0.0002, "epoch": 2.860748843079512, "step": 3400}, {"loss": 1.6028, "grad_norm": 0.4590015709400177, "learning_rate": 0.0002, "epoch": 2.86916281026504, "step": 3410}, {"loss": 1.6746, "grad_norm": 0.45689624547958374, "learning_rate": 0.0002, "epoch": 2.877576777450568, "step": 3420}, {"loss": 1.6326, "grad_norm": 0.46953922510147095, "learning_rate": 0.0002, "epoch": 2.885990744636096, "step": 3430}, {"loss": 1.6015, "grad_norm": 0.4791966378688812, "learning_rate": 0.0002, "epoch": 2.8944047118216236, "step": 3440}, {"loss": 1.694, "grad_norm": 0.4842296242713928, "learning_rate": 0.0002, "epoch": 2.9028186790071517, "step": 3450}, {"loss": 1.6326, "grad_norm": 0.47219768166542053, "learning_rate": 0.0002, "epoch": 2.91123264619268, "step": 3460}, {"loss": 1.6486, "grad_norm": 0.4622127115726471, "learning_rate": 0.0002, "epoch": 2.919646613378208, "step": 3470}, {"loss": 1.6485, "grad_norm": 0.46832820773124695, "learning_rate": 0.0002, "epoch": 2.9280605805637356, "step": 3480}, {"loss": 1.6366, "grad_norm": 0.44582483172416687, "learning_rate": 0.0002, "epoch": 2.9364745477492638, "step": 3490}, {"loss": 1.6859, "grad_norm": 0.4987219274044037, "learning_rate": 0.0002, "epoch": 2.944888514934792, "step": 3500}, {"loss": 1.5991, "grad_norm": 0.43750956654548645, "learning_rate": 0.0002, "epoch": 2.9533024821203195, "step": 3510}, {"loss": 1.6236, "grad_norm": 0.49962925910949707, "learning_rate": 0.0002, "epoch": 2.9617164493058477, "step": 3520}, {"loss": 1.5859, "grad_norm": 0.5189590454101562, "learning_rate": 0.0002, "epoch": 2.9701304164913758, "step": 3530}, {"loss": 1.6688, "grad_norm": 0.391317754983902, "learning_rate": 0.0002, "epoch": 2.978544383676904, "step": 3540}, {"loss": 1.5884, "grad_norm": 0.44934695959091187, "learning_rate": 0.0002, "epoch": 2.9869583508624316, "step": 3550}, {"loss": 1.5688, "grad_norm": 0.4740142226219177, "learning_rate": 0.0002, "epoch": 2.9953723180479597, "step": 3560}, {"eval_loss": 1.8266887664794922, "eval_runtime": 37.9445, "eval_samples_per_second": 13.572, "eval_steps_per_second": 1.713, "epoch": 2.9995793016407237, "step": 3565}, {"loss": 1.5939, "grad_norm": 0.4523724615573883, "learning_rate": 0.0002, "epoch": 3.003786285233488, "step": 3570}, {"loss": 1.526, "grad_norm": 0.5261380076408386, "learning_rate": 0.0002, "epoch": 3.0122002524190155, "step": 3580}, {"loss": 1.4946, "grad_norm": 0.48664888739585876, "learning_rate": 0.0002, "epoch": 3.0206142196045436, "step": 3590}, {"loss": 1.5193, "grad_norm": 0.5070882439613342, "learning_rate": 0.0002, "epoch": 3.0290281867900717, "step": 3600}, {"loss": 1.5316, "grad_norm": 0.5816011428833008, "learning_rate": 0.0002, "epoch": 3.0374421539755994, "step": 3610}, {"loss": 1.5682, "grad_norm": 0.6610211730003357, "learning_rate": 0.0002, "epoch": 3.0458561211611275, "step": 3620}, {"loss": 1.5699, "grad_norm": 0.5257703065872192, "learning_rate": 0.0002, "epoch": 3.0542700883466556, "step": 3630}, {"loss": 1.4438, "grad_norm": 0.5574390888214111, "learning_rate": 0.0002, "epoch": 3.0626840555321833, "step": 3640}, {"loss": 1.547, "grad_norm": 0.5682297348976135, "learning_rate": 0.0002, "epoch": 3.0710980227177114, "step": 3650}, {"loss": 1.5743, "grad_norm": 0.5798383355140686, "learning_rate": 0.0002, "epoch": 3.0795119899032395, "step": 3660}, {"loss": 1.4339, "grad_norm": 0.5458289980888367, "learning_rate": 0.0002, "epoch": 3.087925957088767, "step": 3670}, {"loss": 1.46, "grad_norm": 0.5599102973937988, "learning_rate": 0.0002, "epoch": 3.0963399242742953, "step": 3680}, {"loss": 1.4589, "grad_norm": 0.5023021697998047, "learning_rate": 0.0002, "epoch": 3.1047538914598234, "step": 3690}, {"loss": 1.5114, "grad_norm": 0.5448206067085266, "learning_rate": 0.0002, "epoch": 3.113167858645351, "step": 3700}, {"loss": 1.4692, "grad_norm": 0.5760458707809448, "learning_rate": 0.0002, "epoch": 3.121581825830879, "step": 3710}, {"loss": 1.4789, "grad_norm": 0.6018968224525452, "learning_rate": 0.0002, "epoch": 3.1299957930164073, "step": 3720}, {"loss": 1.5518, "grad_norm": 0.5767101049423218, "learning_rate": 0.0002, "epoch": 3.1384097602019354, "step": 3730}, {"loss": 1.5032, "grad_norm": 0.5333963632583618, "learning_rate": 0.0002, "epoch": 3.146823727387463, "step": 3740}, {"loss": 1.4812, "grad_norm": 0.5918396711349487, "learning_rate": 0.0002, "epoch": 3.155237694572991, "step": 3750}, {"loss": 1.4618, "grad_norm": 0.5931203365325928, "learning_rate": 0.0002, "epoch": 3.1636516617585193, "step": 3760}, {"loss": 1.5592, "grad_norm": 0.6562168598175049, "learning_rate": 0.0002, "epoch": 3.172065628944047, "step": 3770}, {"loss": 1.4932, "grad_norm": 0.5820156335830688, "learning_rate": 0.0002, "epoch": 3.180479596129575, "step": 3780}, {"loss": 1.4523, "grad_norm": 0.5784737467765808, "learning_rate": 0.0002, "epoch": 3.188893563315103, "step": 3790}, {"loss": 1.498, "grad_norm": 0.5506529808044434, "learning_rate": 0.0002, "epoch": 3.197307530500631, "step": 3800}, {"loss": 1.4819, "grad_norm": 0.6101595163345337, "learning_rate": 0.0002, "epoch": 3.205721497686159, "step": 3810}, {"loss": 1.5185, "grad_norm": 0.5597806572914124, "learning_rate": 0.0002, "epoch": 3.214135464871687, "step": 3820}, {"loss": 1.5664, "grad_norm": 0.5641011595726013, "learning_rate": 0.0002, "epoch": 3.222549432057215, "step": 3830}, {"loss": 1.4702, "grad_norm": 0.5892080068588257, "learning_rate": 0.0002, "epoch": 3.230963399242743, "step": 3840}, {"loss": 1.4194, "grad_norm": 0.6034760475158691, "learning_rate": 0.0002, "epoch": 3.239377366428271, "step": 3850}, {"loss": 1.5499, "grad_norm": 0.5112439393997192, "learning_rate": 0.0002, "epoch": 3.247791333613799, "step": 3860}, {"loss": 1.5132, "grad_norm": 0.56565922498703, "learning_rate": 0.0002, "epoch": 3.256205300799327, "step": 3870}, {"loss": 1.4892, "grad_norm": 0.6155247092247009, "learning_rate": 0.0002, "epoch": 3.264619267984855, "step": 3880}, {"loss": 1.5118, "grad_norm": 0.6064623594284058, "learning_rate": 0.0002, "epoch": 3.273033235170383, "step": 3890}, {"loss": 1.5236, "grad_norm": 0.6313768029212952, "learning_rate": 0.0002, "epoch": 3.2814472023559107, "step": 3900}, {"loss": 1.5551, "grad_norm": 0.5903939008712769, "learning_rate": 0.0002, "epoch": 3.289861169541439, "step": 3910}, {"loss": 1.5703, "grad_norm": 0.5770667195320129, "learning_rate": 0.0002, "epoch": 3.298275136726967, "step": 3920}, {"loss": 1.5159, "grad_norm": 0.5785196423530579, "learning_rate": 0.0002, "epoch": 3.3066891039124946, "step": 3930}, {"loss": 1.5277, "grad_norm": 0.6468310356140137, "learning_rate": 0.0002, "epoch": 3.3151030710980227, "step": 3940}, {"loss": 1.6002, "grad_norm": 0.6200279593467712, "learning_rate": 0.0002, "epoch": 3.323517038283551, "step": 3950}, {"loss": 1.5264, "grad_norm": 0.5779302716255188, "learning_rate": 0.0002, "epoch": 3.3319310054690785, "step": 3960}, {"loss": 1.4861, "grad_norm": 0.5463796854019165, "learning_rate": 0.0002, "epoch": 3.3403449726546066, "step": 3970}, {"loss": 1.541, "grad_norm": 0.6117855906486511, "learning_rate": 0.0002, "epoch": 3.3487589398401347, "step": 3980}, {"loss": 1.5566, "grad_norm": 0.5554766058921814, "learning_rate": 0.0002, "epoch": 3.357172907025663, "step": 3990}, {"loss": 1.5004, "grad_norm": 0.6012870073318481, "learning_rate": 0.0002, "epoch": 3.3655868742111905, "step": 4000}, {"loss": 1.473, "grad_norm": 0.5443974137306213, "learning_rate": 0.0002, "epoch": 3.3740008413967186, "step": 4010}, {"loss": 1.5139, "grad_norm": 0.6636057496070862, "learning_rate": 0.0002, "epoch": 3.3824148085822463, "step": 4020}, {"loss": 1.5141, "grad_norm": 0.5801246166229248, "learning_rate": 0.0002, "epoch": 3.3908287757677744, "step": 4030}, {"loss": 1.5026, "grad_norm": 0.5668839812278748, "learning_rate": 0.0002, "epoch": 3.3992427429533025, "step": 4040}, {"loss": 1.523, "grad_norm": 0.7763481736183167, "learning_rate": 0.0002, "epoch": 3.4076567101388306, "step": 4050}, {"loss": 1.4932, "grad_norm": 0.6675992608070374, "learning_rate": 0.0002, "epoch": 3.4160706773243583, "step": 4060}, {"loss": 1.4959, "grad_norm": 0.6290077567100525, "learning_rate": 0.0002, "epoch": 3.4244846445098864, "step": 4070}, {"loss": 1.5766, "grad_norm": 0.6040239930152893, "learning_rate": 0.0002, "epoch": 3.4328986116954145, "step": 4080}, {"loss": 1.5711, "grad_norm": 0.6237877607345581, "learning_rate": 0.0002, "epoch": 3.441312578880942, "step": 4090}, {"loss": 1.4961, "grad_norm": 0.5343508124351501, "learning_rate": 0.0002, "epoch": 3.4497265460664703, "step": 4100}, {"loss": 1.5123, "grad_norm": 0.6817412972450256, "learning_rate": 0.0002, "epoch": 3.4581405132519984, "step": 4110}, {"loss": 1.5377, "grad_norm": 0.7115170359611511, "learning_rate": 0.0002, "epoch": 3.466554480437526, "step": 4120}, {"loss": 1.5275, "grad_norm": 0.6127332448959351, "learning_rate": 0.0002, "epoch": 3.4749684476230542, "step": 4130}, {"loss": 1.557, "grad_norm": 0.5745994448661804, "learning_rate": 0.0002, "epoch": 3.4833824148085824, "step": 4140}, {"loss": 1.4873, "grad_norm": 0.6248795390129089, "learning_rate": 0.0002, "epoch": 3.49179638199411, "step": 4150}, {"loss": 1.4885, "grad_norm": 0.5821124911308289, "learning_rate": 0.0002, "epoch": 3.500210349179638, "step": 4160}, {"loss": 1.4937, "grad_norm": 0.561416506767273, "learning_rate": 0.0002, "epoch": 3.5086243163651663, "step": 4170}, {"loss": 1.5453, "grad_norm": 0.5848962664604187, "learning_rate": 0.0002, "epoch": 3.5170382835506944, "step": 4180}, {"loss": 1.5892, "grad_norm": 0.5335569977760315, "learning_rate": 0.0002, "epoch": 3.525452250736222, "step": 4190}, {"loss": 1.5152, "grad_norm": 0.547964870929718, "learning_rate": 0.0002, "epoch": 3.53386621792175, "step": 4200}, {"loss": 1.4887, "grad_norm": 0.6157727241516113, "learning_rate": 0.0002, "epoch": 3.542280185107278, "step": 4210}, {"loss": 1.5484, "grad_norm": 0.6163121461868286, "learning_rate": 0.0002, "epoch": 3.550694152292806, "step": 4220}, {"loss": 1.5833, "grad_norm": 0.5844616293907166, "learning_rate": 0.0002, "epoch": 3.559108119478334, "step": 4230}, {"loss": 1.5305, "grad_norm": 0.7104926109313965, "learning_rate": 0.0002, "epoch": 3.567522086663862, "step": 4240}, {"loss": 1.5161, "grad_norm": 0.5055213570594788, "learning_rate": 0.0002, "epoch": 3.57593605384939, "step": 4250}, {"loss": 1.482, "grad_norm": 0.611676812171936, "learning_rate": 0.0002, "epoch": 3.584350021034918, "step": 4260}, {"loss": 1.5048, "grad_norm": 0.6326440572738647, "learning_rate": 0.0002, "epoch": 3.592763988220446, "step": 4270}, {"loss": 1.5122, "grad_norm": 0.6290925741195679, "learning_rate": 0.0002, "epoch": 3.6011779554059737, "step": 4280}, {"loss": 1.5654, "grad_norm": 0.5691978931427002, "learning_rate": 0.0002, "epoch": 3.609591922591502, "step": 4290}, {"loss": 1.4854, "grad_norm": 0.6071329116821289, "learning_rate": 0.0002, "epoch": 3.61800588977703, "step": 4300}, {"loss": 1.5336, "grad_norm": 0.606573224067688, "learning_rate": 0.0002, "epoch": 3.626419856962558, "step": 4310}, {"loss": 1.6437, "grad_norm": 0.5515419244766235, "learning_rate": 0.0002, "epoch": 3.6348338241480858, "step": 4320}, {"loss": 1.498, "grad_norm": 0.5964660048484802, "learning_rate": 0.0002, "epoch": 3.643247791333614, "step": 4330}, {"loss": 1.544, "grad_norm": 0.5774146914482117, "learning_rate": 0.0002, "epoch": 3.6516617585191415, "step": 4340}, {"loss": 1.5566, "grad_norm": 0.5732731223106384, "learning_rate": 0.0002, "epoch": 3.6600757257046697, "step": 4350}, {"loss": 1.5682, "grad_norm": 0.7354163527488708, "learning_rate": 0.0002, "epoch": 3.6684896928901978, "step": 4360}, {"loss": 1.5225, "grad_norm": 0.6220902800559998, "learning_rate": 0.0002, "epoch": 3.676903660075726, "step": 4370}, {"loss": 1.4838, "grad_norm": 0.6053991317749023, "learning_rate": 0.0002, "epoch": 3.6853176272612536, "step": 4380}, {"loss": 1.5161, "grad_norm": 0.67010897397995, "learning_rate": 0.0002, "epoch": 3.6937315944467817, "step": 4390}, {"loss": 1.5381, "grad_norm": 0.6139186024665833, "learning_rate": 0.0002, "epoch": 3.70214556163231, "step": 4400}, {"loss": 1.5088, "grad_norm": 0.5433071851730347, "learning_rate": 0.0002, "epoch": 3.7105595288178375, "step": 4410}, {"loss": 1.5337, "grad_norm": 0.5453870296478271, "learning_rate": 0.0002, "epoch": 3.7189734960033656, "step": 4420}, {"loss": 1.4549, "grad_norm": 0.6401727199554443, "learning_rate": 0.0002, "epoch": 3.7273874631888937, "step": 4430}, {"loss": 1.503, "grad_norm": 0.6049367189407349, "learning_rate": 0.0002, "epoch": 3.735801430374422, "step": 4440}, {"loss": 1.5268, "grad_norm": 0.5740529298782349, "learning_rate": 0.0002, "epoch": 3.7442153975599495, "step": 4450}, {"loss": 1.5183, "grad_norm": 0.6521880626678467, "learning_rate": 0.0002, "epoch": 3.7526293647454776, "step": 4460}, {"loss": 1.5741, "grad_norm": 0.7096368074417114, "learning_rate": 0.0002, "epoch": 3.7610433319310053, "step": 4470}, {"loss": 1.5786, "grad_norm": 0.5886474251747131, "learning_rate": 0.0002, "epoch": 3.7694572991165334, "step": 4480}, {"loss": 1.5887, "grad_norm": 0.5821043252944946, "learning_rate": 0.0002, "epoch": 3.7778712663020615, "step": 4490}, {"loss": 1.5777, "grad_norm": 0.628892183303833, "learning_rate": 0.0002, "epoch": 3.7862852334875896, "step": 4500}, {"loss": 1.4708, "grad_norm": 0.5962669849395752, "learning_rate": 0.0002, "epoch": 3.7946992006731173, "step": 4510}, {"loss": 1.5267, "grad_norm": 0.6635549068450928, "learning_rate": 0.0002, "epoch": 3.8031131678586454, "step": 4520}, {"loss": 1.5058, "grad_norm": 0.6010760068893433, "learning_rate": 0.0002, "epoch": 3.811527135044173, "step": 4530}, {"loss": 1.6228, "grad_norm": 0.6322658658027649, "learning_rate": 0.0002, "epoch": 3.819941102229701, "step": 4540}, {"loss": 1.5029, "grad_norm": 0.5893137454986572, "learning_rate": 0.0002, "epoch": 3.8283550694152293, "step": 4550}, {"loss": 1.5435, "grad_norm": 0.7829602360725403, "learning_rate": 0.0002, "epoch": 3.8367690366007574, "step": 4560}, {"loss": 1.5453, "grad_norm": 0.6190396547317505, "learning_rate": 0.0002, "epoch": 3.845183003786285, "step": 4570}, {"loss": 1.5292, "grad_norm": 0.6662813425064087, "learning_rate": 0.0002, "epoch": 3.853596970971813, "step": 4580}, {"loss": 1.5065, "grad_norm": 0.5809855461120605, "learning_rate": 0.0002, "epoch": 3.8620109381573413, "step": 4590}, {"loss": 1.5041, "grad_norm": 0.5779069662094116, "learning_rate": 0.0002, "epoch": 3.870424905342869, "step": 4600}, {"loss": 1.498, "grad_norm": 0.5603038668632507, "learning_rate": 0.0002, "epoch": 3.878838872528397, "step": 4610}, {"loss": 1.5372, "grad_norm": 0.6274181008338928, "learning_rate": 0.0002, "epoch": 3.887252839713925, "step": 4620}, {"loss": 1.4996, "grad_norm": 0.6810959577560425, "learning_rate": 0.0002, "epoch": 3.8956668068994533, "step": 4630}, {"loss": 1.4956, "grad_norm": 0.5647315979003906, "learning_rate": 0.0002, "epoch": 3.904080774084981, "step": 4640}, {"loss": 1.5424, "grad_norm": 0.6830295324325562, "learning_rate": 0.0002, "epoch": 3.912494741270509, "step": 4650}, {"loss": 1.535, "grad_norm": 0.652565598487854, "learning_rate": 0.0002, "epoch": 3.920908708456037, "step": 4660}, {"loss": 1.4772, "grad_norm": 0.5806284546852112, "learning_rate": 0.0002, "epoch": 3.929322675641565, "step": 4670}, {"loss": 1.5812, "grad_norm": 0.6825073957443237, "learning_rate": 0.0002, "epoch": 3.937736642827093, "step": 4680}, {"loss": 1.5516, "grad_norm": 0.6149451732635498, "learning_rate": 0.0002, "epoch": 3.946150610012621, "step": 4690}, {"loss": 1.5608, "grad_norm": 0.6152557134628296, "learning_rate": 0.0002, "epoch": 3.954564577198149, "step": 4700}, {"loss": 1.4897, "grad_norm": 0.6239011883735657, "learning_rate": 0.0002, "epoch": 3.962978544383677, "step": 4710}, {"loss": 1.538, "grad_norm": 0.6485443115234375, "learning_rate": 0.0002, "epoch": 3.971392511569205, "step": 4720}, {"loss": 1.5226, "grad_norm": 0.6449228525161743, "learning_rate": 0.0002, "epoch": 3.9798064787547327, "step": 4730}, {"loss": 1.5087, "grad_norm": 0.6526407599449158, "learning_rate": 0.0002, "epoch": 3.988220445940261, "step": 4740}, {"loss": 1.5026, "grad_norm": 0.6277706027030945, "learning_rate": 0.0002, "epoch": 3.996634413125789, "step": 4750}, {"eval_loss": 1.871641755104065, "eval_runtime": 37.9637, "eval_samples_per_second": 13.566, "eval_steps_per_second": 1.712, "epoch": 4.0, "step": 4754}, {"loss": 1.4744, "grad_norm": 0.6994837522506714, "learning_rate": 0.0002, "epoch": 4.005048380311317, "step": 4760}, {"loss": 1.4433, "grad_norm": 0.8728373050689697, "learning_rate": 0.0002, "epoch": 4.013462347496845, "step": 4770}, {"loss": 1.3329, "grad_norm": 0.688679575920105, "learning_rate": 0.0002, "epoch": 4.021876314682372, "step": 4780}, {"loss": 1.3999, "grad_norm": 0.6313387155532837, "learning_rate": 0.0002, "epoch": 4.0302902818679005, "step": 4790}, {"loss": 1.3346, "grad_norm": 0.6577984690666199, "learning_rate": 0.0002, "epoch": 4.038704249053429, "step": 4800}, {"loss": 1.3403, "grad_norm": 0.7938185930252075, "learning_rate": 0.0002, "epoch": 4.047118216238957, "step": 4810}, {"loss": 1.3716, "grad_norm": 0.760399580001831, "learning_rate": 0.0002, "epoch": 4.055532183424485, "step": 4820}, {"loss": 1.4321, "grad_norm": 0.7329602241516113, "learning_rate": 0.0002, "epoch": 4.063946150610013, "step": 4830}, {"loss": 1.4133, "grad_norm": 0.7778576016426086, "learning_rate": 0.0002, "epoch": 4.07236011779554, "step": 4840}, {"loss": 1.4372, "grad_norm": 0.8235865235328674, "learning_rate": 0.0002, "epoch": 4.080774084981068, "step": 4850}, {"loss": 1.3719, "grad_norm": 0.7743754386901855, "learning_rate": 0.0002, "epoch": 4.089188052166596, "step": 4860}, {"loss": 1.3787, "grad_norm": 0.8145367503166199, "learning_rate": 0.0002, "epoch": 4.0976020193521245, "step": 4870}, {"loss": 1.356, "grad_norm": 0.8517307639122009, "learning_rate": 0.0002, "epoch": 4.106015986537653, "step": 4880}, {"loss": 1.4191, "grad_norm": 0.8208953142166138, "learning_rate": 0.0002, "epoch": 4.114429953723181, "step": 4890}, {"loss": 1.3189, "grad_norm": 0.8437790870666504, "learning_rate": 0.0002, "epoch": 4.122843920908709, "step": 4900}, {"loss": 1.3987, "grad_norm": 0.716672420501709, "learning_rate": 0.0002, "epoch": 4.131257888094236, "step": 4910}, {"loss": 1.4392, "grad_norm": 0.7656235098838806, "learning_rate": 0.0002, "epoch": 4.139671855279764, "step": 4920}, {"loss": 1.3408, "grad_norm": 0.7209306955337524, "learning_rate": 0.0002, "epoch": 4.148085822465292, "step": 4930}, {"loss": 1.3639, "grad_norm": 0.7731267809867859, "learning_rate": 0.0002, "epoch": 4.1564997896508205, "step": 4940}, {"loss": 1.4151, "grad_norm": 0.7477553486824036, "learning_rate": 0.0002, "epoch": 4.164913756836349, "step": 4950}, {"loss": 1.3485, "grad_norm": 0.7372981309890747, "learning_rate": 0.0002, "epoch": 4.173327724021877, "step": 4960}, {"loss": 1.3901, "grad_norm": 0.6582154035568237, "learning_rate": 0.0002, "epoch": 4.181741691207404, "step": 4970}, {"loss": 1.3343, "grad_norm": 0.7003206610679626, "learning_rate": 0.0002, "epoch": 4.190155658392932, "step": 4980}, {"loss": 1.4098, "grad_norm": 0.735223650932312, "learning_rate": 0.0002, "epoch": 4.19856962557846, "step": 4990}, {"loss": 1.3564, "grad_norm": 0.7832302451133728, "learning_rate": 0.0002, "epoch": 4.206983592763988, "step": 5000}, {"loss": 1.3622, "grad_norm": 0.8819546103477478, "learning_rate": 0.0002, "epoch": 4.215397559949516, "step": 5010}, {"loss": 1.4438, "grad_norm": 0.9325336813926697, "learning_rate": 0.0002, "epoch": 4.2238115271350445, "step": 5020}, {"loss": 1.3886, "grad_norm": 0.7007517218589783, "learning_rate": 0.0002, "epoch": 4.232225494320572, "step": 5030}, {"loss": 1.3683, "grad_norm": 0.7118321061134338, "learning_rate": 0.0002, "epoch": 4.2406394615061, "step": 5040}, {"loss": 1.2365, "grad_norm": 0.6578946709632874, "learning_rate": 0.0002, "epoch": 4.249053428691628, "step": 5050}, {"loss": 1.3696, "grad_norm": 0.9438983798027039, "learning_rate": 0.0002, "epoch": 4.257467395877156, "step": 5060}, {"loss": 1.3868, "grad_norm": 0.703037679195404, "learning_rate": 0.0002, "epoch": 4.265881363062684, "step": 5070}, {"loss": 1.3687, "grad_norm": 0.7286025285720825, "learning_rate": 0.0002, "epoch": 4.274295330248212, "step": 5080}, {"loss": 1.3605, "grad_norm": 0.750689685344696, "learning_rate": 0.0002, "epoch": 4.28270929743374, "step": 5090}, {"loss": 1.5089, "grad_norm": 0.869753360748291, "learning_rate": 0.0002, "epoch": 4.291123264619268, "step": 5100}, {"loss": 1.4128, "grad_norm": 0.8712980151176453, "learning_rate": 0.0002, "epoch": 4.299537231804796, "step": 5110}, {"loss": 1.3977, "grad_norm": 0.690263569355011, "learning_rate": 0.0002, "epoch": 4.307951198990324, "step": 5120}, {"loss": 1.4088, "grad_norm": 0.7114760279655457, "learning_rate": 0.0002, "epoch": 4.316365166175852, "step": 5130}, {"loss": 1.363, "grad_norm": 0.7588112354278564, "learning_rate": 0.0002, "epoch": 4.32477913336138, "step": 5140}, {"loss": 1.4408, "grad_norm": 0.7556202411651611, "learning_rate": 0.0002, "epoch": 4.333193100546908, "step": 5150}, {"loss": 1.4203, "grad_norm": 0.8357610702514648, "learning_rate": 0.0002, "epoch": 4.341607067732435, "step": 5160}, {"loss": 1.3348, "grad_norm": 0.8054035902023315, "learning_rate": 0.0002, "epoch": 4.3500210349179635, "step": 5170}, {"loss": 1.3109, "grad_norm": 0.7637107968330383, "learning_rate": 0.0002, "epoch": 4.358435002103492, "step": 5180}, {"loss": 1.3744, "grad_norm": 0.757481038570404, "learning_rate": 0.0002, "epoch": 4.36684896928902, "step": 5190}, {"loss": 1.3622, "grad_norm": 0.7185863852500916, "learning_rate": 0.0002, "epoch": 4.375262936474548, "step": 5200}, {"loss": 1.3896, "grad_norm": 0.7326455116271973, "learning_rate": 0.0002, "epoch": 4.383676903660076, "step": 5210}, {"loss": 1.4098, "grad_norm": 0.7980523109436035, "learning_rate": 0.0002, "epoch": 4.392090870845603, "step": 5220}, {"loss": 1.3783, "grad_norm": 0.8526999354362488, "learning_rate": 0.0002, "epoch": 4.400504838031131, "step": 5230}, {"loss": 1.4022, "grad_norm": 0.7012337446212769, "learning_rate": 0.0002, "epoch": 4.4089188052166595, "step": 5240}, {"loss": 1.3552, "grad_norm": 0.8217827677726746, "learning_rate": 0.0002, "epoch": 4.417332772402188, "step": 5250}, {"loss": 1.3482, "grad_norm": 0.7141005396842957, "learning_rate": 0.0002, "epoch": 4.425746739587716, "step": 5260}, {"loss": 1.3699, "grad_norm": 0.7094302177429199, "learning_rate": 0.0002, "epoch": 4.434160706773244, "step": 5270}, {"loss": 1.3527, "grad_norm": 0.7234613299369812, "learning_rate": 0.0002, "epoch": 4.442574673958772, "step": 5280}, {"loss": 1.4769, "grad_norm": 0.7530457973480225, "learning_rate": 0.0002, "epoch": 4.450988641144299, "step": 5290}, {"loss": 1.3944, "grad_norm": 0.7300912141799927, "learning_rate": 0.0002, "epoch": 4.459402608329827, "step": 5300}, {"loss": 1.3844, "grad_norm": 0.825443685054779, "learning_rate": 0.0002, "epoch": 4.467816575515355, "step": 5310}, {"loss": 1.3648, "grad_norm": 0.7559658885002136, "learning_rate": 0.0002, "epoch": 4.4762305427008835, "step": 5320}, {"loss": 1.4364, "grad_norm": 0.8817561268806458, "learning_rate": 0.0002, "epoch": 4.484644509886412, "step": 5330}, {"loss": 1.3618, "grad_norm": 0.8203575611114502, "learning_rate": 0.0002, "epoch": 4.49305847707194, "step": 5340}, {"loss": 1.3996, "grad_norm": 0.7677690982818604, "learning_rate": 0.0002, "epoch": 4.501472444257468, "step": 5350}, {"loss": 1.4142, "grad_norm": 0.657085120677948, "learning_rate": 0.0002, "epoch": 4.509886411442995, "step": 5360}, {"loss": 1.3722, "grad_norm": 0.7939504384994507, "learning_rate": 0.0002, "epoch": 4.518300378628523, "step": 5370}, {"loss": 1.4361, "grad_norm": 0.6971889138221741, "learning_rate": 0.0002, "epoch": 4.526714345814051, "step": 5380}, {"loss": 1.3637, "grad_norm": 0.6984175443649292, "learning_rate": 0.0002, "epoch": 4.535128312999579, "step": 5390}, {"loss": 1.341, "grad_norm": 0.8504858613014221, "learning_rate": 0.0002, "epoch": 4.5435422801851075, "step": 5400}, {"loss": 1.4026, "grad_norm": 0.9134073853492737, "learning_rate": 0.0002, "epoch": 4.551956247370635, "step": 5410}, {"loss": 1.4375, "grad_norm": 0.7765598893165588, "learning_rate": 0.0002, "epoch": 4.560370214556163, "step": 5420}, {"loss": 1.4832, "grad_norm": 0.6991009712219238, "learning_rate": 0.0002, "epoch": 4.568784181741691, "step": 5430}, {"loss": 1.4021, "grad_norm": 0.8393039107322693, "learning_rate": 0.0002, "epoch": 4.577198148927219, "step": 5440}, {"loss": 1.3976, "grad_norm": 0.7685918211936951, "learning_rate": 0.0002, "epoch": 4.585612116112747, "step": 5450}, {"loss": 1.3883, "grad_norm": 0.7135679721832275, "learning_rate": 0.0002, "epoch": 4.594026083298275, "step": 5460}, {"loss": 1.4083, "grad_norm": 0.6728870868682861, "learning_rate": 0.0002, "epoch": 4.6024400504838034, "step": 5470}, {"loss": 1.3698, "grad_norm": 0.7139479517936707, "learning_rate": 0.0002, "epoch": 4.610854017669331, "step": 5480}, {"loss": 1.3498, "grad_norm": 0.8476598858833313, "learning_rate": 0.0002, "epoch": 4.619267984854859, "step": 5490}, {"loss": 1.3389, "grad_norm": 0.8034361004829407, "learning_rate": 0.0002, "epoch": 4.627681952040387, "step": 5500}, {"loss": 1.4179, "grad_norm": 0.7452183961868286, "learning_rate": 0.0002, "epoch": 4.636095919225915, "step": 5510}, {"loss": 1.4031, "grad_norm": 0.8394148945808411, "learning_rate": 0.0002, "epoch": 4.644509886411443, "step": 5520}, {"loss": 1.4561, "grad_norm": 0.7480153441429138, "learning_rate": 0.0002, "epoch": 4.652923853596971, "step": 5530}, {"loss": 1.378, "grad_norm": 0.7781714797019958, "learning_rate": 0.0002, "epoch": 4.661337820782499, "step": 5540}, {"loss": 1.3924, "grad_norm": 1.0058213472366333, "learning_rate": 0.0002, "epoch": 4.669751787968027, "step": 5550}, {"loss": 1.4198, "grad_norm": 0.7403179407119751, "learning_rate": 0.0002, "epoch": 4.678165755153555, "step": 5560}, {"loss": 1.4328, "grad_norm": 0.7270476818084717, "learning_rate": 0.0002, "epoch": 4.686579722339083, "step": 5570}, {"loss": 1.378, "grad_norm": 0.760877788066864, "learning_rate": 0.0002, "epoch": 4.694993689524611, "step": 5580}, {"loss": 1.387, "grad_norm": 0.8097004890441895, "learning_rate": 0.0002, "epoch": 4.703407656710139, "step": 5590}, {"loss": 1.3661, "grad_norm": 0.9096523523330688, "learning_rate": 0.0002, "epoch": 4.711821623895667, "step": 5600}, {"loss": 1.4012, "grad_norm": 0.7262444496154785, "learning_rate": 0.0002, "epoch": 4.720235591081195, "step": 5610}, {"loss": 1.422, "grad_norm": 0.8207762837409973, "learning_rate": 0.0002, "epoch": 4.7286495582667225, "step": 5620}, {"loss": 1.4017, "grad_norm": 0.8089601993560791, "learning_rate": 0.0002, "epoch": 4.737063525452251, "step": 5630}, {"loss": 1.3675, "grad_norm": 0.7609543800354004, "learning_rate": 0.0002, "epoch": 4.745477492637779, "step": 5640}, {"loss": 1.4085, "grad_norm": 0.7273501753807068, "learning_rate": 0.0002, "epoch": 4.753891459823307, "step": 5650}, {"loss": 1.3849, "grad_norm": 0.7800219058990479, "learning_rate": 0.0002, "epoch": 4.762305427008835, "step": 5660}, {"loss": 1.4319, "grad_norm": 0.8558377623558044, "learning_rate": 0.0002, "epoch": 4.770719394194362, "step": 5670}, {"loss": 1.3831, "grad_norm": 0.7131547927856445, "learning_rate": 0.0002, "epoch": 4.77913336137989, "step": 5680}, {"loss": 1.407, "grad_norm": 0.7651025056838989, "learning_rate": 0.0002, "epoch": 4.787547328565418, "step": 5690}, {"loss": 1.3882, "grad_norm": 0.8129976391792297, "learning_rate": 0.0002, "epoch": 4.7959612957509465, "step": 5700}, {"loss": 1.4347, "grad_norm": 0.8019895553588867, "learning_rate": 0.0002, "epoch": 4.804375262936475, "step": 5710}, {"loss": 1.3961, "grad_norm": 0.7692018151283264, "learning_rate": 0.0002, "epoch": 4.812789230122003, "step": 5720}, {"loss": 1.419, "grad_norm": 0.6893943548202515, "learning_rate": 0.0002, "epoch": 4.821203197307531, "step": 5730}, {"loss": 1.4453, "grad_norm": 0.6881810426712036, "learning_rate": 0.0002, "epoch": 4.829617164493058, "step": 5740}, {"loss": 1.4775, "grad_norm": 0.7838267683982849, "learning_rate": 0.0002, "epoch": 4.838031131678586, "step": 5750}, {"loss": 1.3857, "grad_norm": 0.727799117565155, "learning_rate": 0.0002, "epoch": 4.846445098864114, "step": 5760}, {"loss": 1.4685, "grad_norm": 0.7458277344703674, "learning_rate": 0.0002, "epoch": 4.8548590660496425, "step": 5770}, {"loss": 1.4426, "grad_norm": 0.903802216053009, "learning_rate": 0.0002, "epoch": 4.863273033235171, "step": 5780}, {"loss": 1.451, "grad_norm": 0.7983472347259521, "learning_rate": 0.0002, "epoch": 4.871687000420699, "step": 5790}, {"loss": 1.4534, "grad_norm": 0.6894361972808838, "learning_rate": 0.0002, "epoch": 4.880100967606227, "step": 5800}, {"loss": 1.4486, "grad_norm": 0.7499409317970276, "learning_rate": 0.0002, "epoch": 4.888514934791754, "step": 5810}, {"loss": 1.4253, "grad_norm": 0.7362820506095886, "learning_rate": 0.0002, "epoch": 4.896928901977282, "step": 5820}, {"loss": 1.3763, "grad_norm": 0.8341619968414307, "learning_rate": 0.0002, "epoch": 4.90534286916281, "step": 5830}, {"loss": 1.3748, "grad_norm": 0.9604470133781433, "learning_rate": 0.0002, "epoch": 4.913756836348338, "step": 5840}, {"loss": 1.3658, "grad_norm": 0.8916844129562378, "learning_rate": 0.0002, "epoch": 4.9221708035338665, "step": 5850}, {"loss": 1.363, "grad_norm": 0.8519647121429443, "learning_rate": 0.0002, "epoch": 4.930584770719394, "step": 5860}, {"loss": 1.424, "grad_norm": 0.7946906089782715, "learning_rate": 0.0002, "epoch": 4.938998737904922, "step": 5870}, {"loss": 1.4071, "grad_norm": 0.7843789458274841, "learning_rate": 0.0002, "epoch": 4.94741270509045, "step": 5880}, {"loss": 1.4021, "grad_norm": 0.707618772983551, "learning_rate": 0.0002, "epoch": 4.955826672275978, "step": 5890}, {"loss": 1.502, "grad_norm": 0.7704206109046936, "learning_rate": 0.0002, "epoch": 4.964240639461506, "step": 5900}, {"loss": 1.4456, "grad_norm": 0.7160256505012512, "learning_rate": 0.0002, "epoch": 4.972654606647034, "step": 5910}, {"loss": 1.3874, "grad_norm": 0.7020420432090759, "learning_rate": 0.0002, "epoch": 4.981068573832562, "step": 5920}, {"loss": 1.4037, "grad_norm": 0.7576286792755127, "learning_rate": 0.0002, "epoch": 4.98948254101809, "step": 5930}, {"loss": 1.414, "grad_norm": 0.8573036789894104, "learning_rate": 0.0002, "epoch": 4.997896508203618, "step": 5940}]}
+{"epoch": 6.0, "step": 7131, "epoch_duration": 1254.0791096687317, "total_accumulated_duration": 8564.974167108536, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "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_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-0.8-num-5623-sd-42/checkpoint-2377", "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": 2.56, "grad_norm": 0.5458821654319763, "learning_rate": 0.0002, "epoch": 0.008413967185527976, "step": 10}, {"loss": 2.3235, "grad_norm": 0.7293308973312378, "learning_rate": 0.0002, "epoch": 0.016827934371055953, "step": 20}, {"loss": 2.0815, "grad_norm": 0.47792306542396545, "learning_rate": 0.0002, "epoch": 0.02524190155658393, "step": 30}, {"loss": 1.9718, "grad_norm": 0.5944402813911438, "learning_rate": 0.0002, "epoch": 0.033655868742111905, "step": 40}, {"loss": 1.8848, "grad_norm": 0.5415359735488892, "learning_rate": 0.0002, "epoch": 0.04206983592763988, "step": 50}, {"loss": 1.8953, "grad_norm": 0.535713791847229, "learning_rate": 0.0002, "epoch": 0.05048380311316786, "step": 60}, {"loss": 1.937, "grad_norm": 0.5184146761894226, "learning_rate": 0.0002, "epoch": 0.058897770298695834, "step": 70}, {"loss": 1.8396, "grad_norm": 0.458926796913147, "learning_rate": 0.0002, "epoch": 0.06731173748422381, "step": 80}, {"loss": 1.8677, "grad_norm": 0.4780142307281494, "learning_rate": 0.0002, "epoch": 0.07572570466975179, "step": 90}, {"loss": 1.8593, "grad_norm": 0.79965740442276, "learning_rate": 0.0002, "epoch": 0.08413967185527976, "step": 100}, {"loss": 1.9081, "grad_norm": 0.4498862028121948, "learning_rate": 0.0002, "epoch": 0.09255363904080774, "step": 110}, {"loss": 1.8503, "grad_norm": 0.39338430762290955, "learning_rate": 0.0002, "epoch": 0.10096760622633572, "step": 120}, {"loss": 1.8637, "grad_norm": 0.9588953852653503, "learning_rate": 0.0002, "epoch": 0.10938157341186369, "step": 130}, {"loss": 1.8676, "grad_norm": 0.41675639152526855, "learning_rate": 0.0002, "epoch": 0.11779554059739167, "step": 140}, {"loss": 1.8904, "grad_norm": 0.44519832730293274, "learning_rate": 0.0002, "epoch": 0.12620950778291964, "step": 150}, {"loss": 1.798, "grad_norm": 0.4176260530948639, "learning_rate": 0.0002, "epoch": 0.13462347496844762, "step": 160}, {"loss": 1.8398, "grad_norm": 0.35840365290641785, "learning_rate": 0.0002, "epoch": 0.1430374421539756, "step": 170}, {"loss": 1.8666, "grad_norm": 0.3794495463371277, "learning_rate": 0.0002, "epoch": 0.15145140933950357, "step": 180}, {"loss": 1.8111, "grad_norm": 0.4563522934913635, "learning_rate": 0.0002, "epoch": 0.15986537652503155, "step": 190}, {"loss": 1.8893, "grad_norm": 0.37057486176490784, "learning_rate": 0.0002, "epoch": 0.16827934371055953, "step": 200}, {"loss": 1.7995, "grad_norm": 0.44081518054008484, "learning_rate": 0.0002, "epoch": 0.1766933108960875, "step": 210}, {"loss": 1.9048, "grad_norm": 0.46078577637672424, "learning_rate": 0.0002, "epoch": 0.18510727808161548, "step": 220}, {"loss": 1.8403, "grad_norm": 0.36132094264030457, "learning_rate": 0.0002, "epoch": 0.19352124526714345, "step": 230}, {"loss": 1.8827, "grad_norm": 0.3747289180755615, "learning_rate": 0.0002, "epoch": 0.20193521245267143, "step": 240}, {"loss": 1.8382, "grad_norm": 0.3540179133415222, "learning_rate": 0.0002, "epoch": 0.2103491796381994, "step": 250}, {"loss": 1.8196, "grad_norm": 0.3461375832557678, "learning_rate": 0.0002, "epoch": 0.21876314682372738, "step": 260}, {"loss": 1.8509, "grad_norm": 0.3436960279941559, "learning_rate": 0.0002, "epoch": 0.22717711400925536, "step": 270}, {"loss": 1.8285, "grad_norm": 0.35403719544410706, "learning_rate": 0.0002, "epoch": 0.23559108119478334, "step": 280}, {"loss": 1.8369, "grad_norm": 0.37142616510391235, "learning_rate": 0.0002, "epoch": 0.2440050483803113, "step": 290}, {"loss": 1.8044, "grad_norm": 0.3307955861091614, "learning_rate": 0.0002, "epoch": 0.2524190155658393, "step": 300}, {"loss": 1.817, "grad_norm": 0.32855314016342163, "learning_rate": 0.0002, "epoch": 0.2608329827513673, "step": 310}, {"loss": 1.7803, "grad_norm": 0.3299003839492798, "learning_rate": 0.0002, "epoch": 0.26924694993689524, "step": 320}, {"loss": 1.8129, "grad_norm": 0.44311287999153137, "learning_rate": 0.0002, "epoch": 0.27766091712242325, "step": 330}, {"loss": 1.8232, "grad_norm": 0.32989758253097534, "learning_rate": 0.0002, "epoch": 0.2860748843079512, "step": 340}, {"loss": 1.7716, "grad_norm": 0.34400200843811035, "learning_rate": 0.0002, "epoch": 0.2944888514934792, "step": 350}, {"loss": 1.7619, "grad_norm": 0.36286211013793945, "learning_rate": 0.0002, "epoch": 0.30290281867900715, "step": 360}, {"loss": 1.8025, "grad_norm": 0.406827837228775, "learning_rate": 0.0002, "epoch": 0.31131678586453515, "step": 370}, {"loss": 1.7515, "grad_norm": 0.36299195885658264, "learning_rate": 0.0002, "epoch": 0.3197307530500631, "step": 380}, {"loss": 1.837, "grad_norm": 0.3477257192134857, "learning_rate": 0.0002, "epoch": 0.3281447202355911, "step": 390}, {"loss": 1.7767, "grad_norm": 0.3730369210243225, "learning_rate": 0.0002, "epoch": 0.33655868742111905, "step": 400}, {"loss": 1.7747, "grad_norm": 0.4644559919834137, "learning_rate": 0.0002, "epoch": 0.34497265460664706, "step": 410}, {"loss": 1.7538, "grad_norm": 0.406576544046402, "learning_rate": 0.0002, "epoch": 0.353386621792175, "step": 420}, {"loss": 1.7501, "grad_norm": 0.3612699508666992, "learning_rate": 0.0002, "epoch": 0.361800588977703, "step": 430}, {"loss": 1.7473, "grad_norm": 0.3243742287158966, "learning_rate": 0.0002, "epoch": 0.37021455616323096, "step": 440}, {"loss": 1.8851, "grad_norm": 0.36671221256256104, "learning_rate": 0.0002, "epoch": 0.37862852334875896, "step": 450}, {"loss": 1.8853, "grad_norm": 0.3565002381801605, "learning_rate": 0.0002, "epoch": 0.3870424905342869, "step": 460}, {"loss": 1.8923, "grad_norm": 0.34630221128463745, "learning_rate": 0.0002, "epoch": 0.3954564577198149, "step": 470}, {"loss": 1.8234, "grad_norm": 0.3353537321090698, "learning_rate": 0.0002, "epoch": 0.40387042490534286, "step": 480}, {"loss": 1.7135, "grad_norm": 0.4015921950340271, "learning_rate": 0.0002, "epoch": 0.41228439209087087, "step": 490}, {"loss": 1.7815, "grad_norm": 0.5489419102668762, "learning_rate": 0.0002, "epoch": 0.4206983592763988, "step": 500}, {"loss": 1.7903, "grad_norm": 0.4193589985370636, "learning_rate": 0.0002, "epoch": 0.4291123264619268, "step": 510}, {"loss": 1.8416, "grad_norm": 0.3418922424316406, "learning_rate": 0.0002, "epoch": 0.43752629364745477, "step": 520}, {"loss": 1.7982, "grad_norm": 0.32668185234069824, "learning_rate": 0.0002, "epoch": 0.44594026083298277, "step": 530}, {"loss": 1.7501, "grad_norm": 0.3094325661659241, "learning_rate": 0.0002, "epoch": 0.4543542280185107, "step": 540}, {"loss": 1.7438, "grad_norm": 0.3743017315864563, "learning_rate": 0.0002, "epoch": 0.4627681952040387, "step": 550}, {"loss": 1.8451, "grad_norm": 0.3295630216598511, "learning_rate": 0.0002, "epoch": 0.47118216238956667, "step": 560}, {"loss": 1.7529, "grad_norm": 1.6124513149261475, "learning_rate": 0.0002, "epoch": 0.4795961295750947, "step": 570}, {"loss": 1.8028, "grad_norm": 0.3245585858821869, "learning_rate": 0.0002, "epoch": 0.4880100967606226, "step": 580}, {"loss": 1.7976, "grad_norm": 0.3332934081554413, "learning_rate": 0.0002, "epoch": 0.49642406394615063, "step": 590}, {"loss": 1.7912, "grad_norm": 0.3836138844490051, "learning_rate": 0.0002, "epoch": 0.5048380311316786, "step": 600}, {"loss": 1.8347, "grad_norm": 0.32953888177871704, "learning_rate": 0.0002, "epoch": 0.5132519983172066, "step": 610}, {"loss": 1.7729, "grad_norm": 0.36291512846946716, "learning_rate": 0.0002, "epoch": 0.5216659655027346, "step": 620}, {"loss": 1.7758, "grad_norm": 0.3237783908843994, "learning_rate": 0.0002, "epoch": 0.5300799326882625, "step": 630}, {"loss": 1.8352, "grad_norm": 0.38882696628570557, "learning_rate": 0.0002, "epoch": 0.5384938998737905, "step": 640}, {"loss": 1.8624, "grad_norm": 0.37821972370147705, "learning_rate": 0.0002, "epoch": 0.5469078670593185, "step": 650}, {"loss": 1.8075, "grad_norm": 0.3556285500526428, "learning_rate": 0.0002, "epoch": 0.5553218342448465, "step": 660}, {"loss": 1.778, "grad_norm": 0.347499281167984, "learning_rate": 0.0002, "epoch": 0.5637358014303744, "step": 670}, {"loss": 1.8066, "grad_norm": 0.3176489472389221, "learning_rate": 0.0002, "epoch": 0.5721497686159024, "step": 680}, {"loss": 1.7257, "grad_norm": 0.30220088362693787, "learning_rate": 0.0002, "epoch": 0.5805637358014304, "step": 690}, {"loss": 1.8415, "grad_norm": 0.3711601793766022, "learning_rate": 0.0002, "epoch": 0.5889777029869584, "step": 700}, {"loss": 1.7906, "grad_norm": 0.3311759829521179, "learning_rate": 0.0002, "epoch": 0.5973916701724863, "step": 710}, {"loss": 1.7712, "grad_norm": 0.34824270009994507, "learning_rate": 0.0002, "epoch": 0.6058056373580143, "step": 720}, {"loss": 1.7954, "grad_norm": 0.29668381810188293, "learning_rate": 0.0002, "epoch": 0.6142196045435423, "step": 730}, {"loss": 1.8321, "grad_norm": 0.36087489128112793, "learning_rate": 0.0002, "epoch": 0.6226335717290703, "step": 740}, {"loss": 1.7956, "grad_norm": 0.31590089201927185, "learning_rate": 0.0002, "epoch": 0.6310475389145982, "step": 750}, {"loss": 1.7343, "grad_norm": 0.37632957100868225, "learning_rate": 0.0002, "epoch": 0.6394615061001262, "step": 760}, {"loss": 1.8499, "grad_norm": 0.3360748589038849, "learning_rate": 0.0002, "epoch": 0.6478754732856542, "step": 770}, {"loss": 1.8076, "grad_norm": 0.3420640528202057, "learning_rate": 0.0002, "epoch": 0.6562894404711822, "step": 780}, {"loss": 1.8353, "grad_norm": 0.5734959244728088, "learning_rate": 0.0002, "epoch": 0.6647034076567101, "step": 790}, {"loss": 1.7746, "grad_norm": 0.36440837383270264, "learning_rate": 0.0002, "epoch": 0.6731173748422381, "step": 800}, {"loss": 1.7532, "grad_norm": 0.3179708421230316, "learning_rate": 0.0002, "epoch": 0.6815313420277661, "step": 810}, {"loss": 1.7815, "grad_norm": 0.34122881293296814, "learning_rate": 0.0002, "epoch": 0.6899453092132941, "step": 820}, {"loss": 1.8167, "grad_norm": 0.31886112689971924, "learning_rate": 0.0002, "epoch": 0.698359276398822, "step": 830}, {"loss": 1.7505, "grad_norm": 0.31782326102256775, "learning_rate": 0.0002, "epoch": 0.70677324358435, "step": 840}, {"loss": 1.7588, "grad_norm": 0.36052989959716797, "learning_rate": 0.0002, "epoch": 0.715187210769878, "step": 850}, {"loss": 1.7891, "grad_norm": 0.28946155309677124, "learning_rate": 0.0002, "epoch": 0.723601177955406, "step": 860}, {"loss": 1.7923, "grad_norm": 0.3095663785934448, "learning_rate": 0.0002, "epoch": 0.7320151451409339, "step": 870}, {"loss": 1.785, "grad_norm": 0.3317491412162781, "learning_rate": 0.0002, "epoch": 0.7404291123264619, "step": 880}, {"loss": 1.7709, "grad_norm": 0.31324660778045654, "learning_rate": 0.0002, "epoch": 0.7488430795119899, "step": 890}, {"loss": 1.8753, "grad_norm": 0.3290475606918335, "learning_rate": 0.0002, "epoch": 0.7572570466975179, "step": 900}, {"loss": 1.7679, "grad_norm": 0.35690343379974365, "learning_rate": 0.0002, "epoch": 0.7656710138830458, "step": 910}, {"loss": 1.826, "grad_norm": 0.39558273553848267, "learning_rate": 0.0002, "epoch": 0.7740849810685738, "step": 920}, {"loss": 1.8722, "grad_norm": 0.34254348278045654, "learning_rate": 0.0002, "epoch": 0.7824989482541018, "step": 930}, {"loss": 1.7603, "grad_norm": 0.3560165464878082, "learning_rate": 0.0002, "epoch": 0.7909129154396298, "step": 940}, {"loss": 1.7992, "grad_norm": 0.30693164467811584, "learning_rate": 0.0002, "epoch": 0.7993268826251577, "step": 950}, {"loss": 1.8029, "grad_norm": 0.3394823372364044, "learning_rate": 0.0002, "epoch": 0.8077408498106857, "step": 960}, {"loss": 1.8105, "grad_norm": 0.3741514980792999, "learning_rate": 0.0002, "epoch": 0.8161548169962137, "step": 970}, {"loss": 1.7849, "grad_norm": 0.3655228316783905, "learning_rate": 0.0002, "epoch": 0.8245687841817417, "step": 980}, {"loss": 1.8449, "grad_norm": 0.3586033880710602, "learning_rate": 0.0002, "epoch": 0.8329827513672696, "step": 990}, {"loss": 1.7033, "grad_norm": 0.3459678888320923, "learning_rate": 0.0002, "epoch": 0.8413967185527976, "step": 1000}, {"loss": 1.8498, "grad_norm": 0.3184349834918976, "learning_rate": 0.0002, "epoch": 0.8498106857383256, "step": 1010}, {"loss": 1.7632, "grad_norm": 0.3099786043167114, "learning_rate": 0.0002, "epoch": 0.8582246529238536, "step": 1020}, {"loss": 1.8067, "grad_norm": 0.30300915241241455, "learning_rate": 0.0002, "epoch": 0.8666386201093815, "step": 1030}, {"loss": 1.7923, "grad_norm": 0.3128705620765686, "learning_rate": 0.0002, "epoch": 0.8750525872949095, "step": 1040}, {"loss": 1.8252, "grad_norm": 0.3336263597011566, "learning_rate": 0.0002, "epoch": 0.8834665544804375, "step": 1050}, {"loss": 1.8375, "grad_norm": 0.3801328241825104, "learning_rate": 0.0002, "epoch": 0.8918805216659655, "step": 1060}, {"loss": 1.7757, "grad_norm": 0.3122096359729767, "learning_rate": 0.0002, "epoch": 0.9002944888514934, "step": 1070}, {"loss": 1.8251, "grad_norm": 0.35990869998931885, "learning_rate": 0.0002, "epoch": 0.9087084560370214, "step": 1080}, {"loss": 1.7343, "grad_norm": 0.3321819305419922, "learning_rate": 0.0002, "epoch": 0.9171224232225494, "step": 1090}, {"loss": 1.7595, "grad_norm": 0.4202139377593994, "learning_rate": 0.0002, "epoch": 0.9255363904080774, "step": 1100}, {"loss": 1.8056, "grad_norm": 0.32559722661972046, "learning_rate": 0.0002, "epoch": 0.9339503575936053, "step": 1110}, {"loss": 1.812, "grad_norm": 0.3098459839820862, "learning_rate": 0.0002, "epoch": 0.9423643247791333, "step": 1120}, {"loss": 1.8252, "grad_norm": 0.33917108178138733, "learning_rate": 0.0002, "epoch": 0.9507782919646613, "step": 1130}, {"loss": 1.7709, "grad_norm": 0.4055837094783783, "learning_rate": 0.0002, "epoch": 0.9591922591501894, "step": 1140}, {"loss": 1.8259, "grad_norm": 0.32508623600006104, "learning_rate": 0.0002, "epoch": 0.9676062263357172, "step": 1150}, {"loss": 1.782, "grad_norm": 0.30150601267814636, "learning_rate": 0.0002, "epoch": 0.9760201935212452, "step": 1160}, {"loss": 1.8291, "grad_norm": 0.3042563199996948, "learning_rate": 0.0002, "epoch": 0.9844341607067733, "step": 1170}, {"loss": 1.7847, "grad_norm": 0.33254584670066833, "learning_rate": 0.0002, "epoch": 0.9928481278923013, "step": 1180}, {"eval_loss": 1.8077726364135742, "eval_runtime": 38.4359, "eval_samples_per_second": 13.399, "eval_steps_per_second": 1.691, "epoch": 0.9995793016407236, "step": 1188}, {"loss": 1.7414, "grad_norm": 0.35073035955429077, "learning_rate": 0.0002, "epoch": 1.0012620950778293, "step": 1190}, {"loss": 1.7483, "grad_norm": 0.3217269778251648, "learning_rate": 0.0002, "epoch": 1.0096760622633572, "step": 1200}, {"loss": 1.7517, "grad_norm": 0.3635033369064331, "learning_rate": 0.0002, "epoch": 1.018090029448885, "step": 1210}, {"loss": 1.6949, "grad_norm": 0.32468414306640625, "learning_rate": 0.0002, "epoch": 1.0265039966344132, "step": 1220}, {"loss": 1.711, "grad_norm": 0.3307163417339325, "learning_rate": 0.0002, "epoch": 1.034917963819941, "step": 1230}, {"loss": 1.7881, "grad_norm": 0.34381359815597534, "learning_rate": 0.0002, "epoch": 1.0433319310054692, "step": 1240}, {"loss": 1.612, "grad_norm": 0.35874804854393005, "learning_rate": 0.0002, "epoch": 1.051745898190997, "step": 1250}, {"loss": 1.7314, "grad_norm": 0.3615919351577759, "learning_rate": 0.0002, "epoch": 1.060159865376525, "step": 1260}, {"loss": 1.7517, "grad_norm": 0.32835808396339417, "learning_rate": 0.0002, "epoch": 1.068573832562053, "step": 1270}, {"loss": 1.7193, "grad_norm": 0.3876388370990753, "learning_rate": 0.0002, "epoch": 1.076987799747581, "step": 1280}, {"loss": 1.7442, "grad_norm": 0.39895930886268616, "learning_rate": 0.0002, "epoch": 1.0854017669331089, "step": 1290}, {"loss": 1.6601, "grad_norm": 0.39081698656082153, "learning_rate": 0.0002, "epoch": 1.093815734118637, "step": 1300}, {"loss": 1.7623, "grad_norm": 0.39974215626716614, "learning_rate": 0.0002, "epoch": 1.1022297013041649, "step": 1310}, {"loss": 1.7506, "grad_norm": 0.3887332081794739, "learning_rate": 0.0002, "epoch": 1.110643668489693, "step": 1320}, {"loss": 1.7381, "grad_norm": 0.36216408014297485, "learning_rate": 0.0002, "epoch": 1.1190576356752209, "step": 1330}, {"loss": 1.762, "grad_norm": 0.36979028582572937, "learning_rate": 0.0002, "epoch": 1.1274716028607488, "step": 1340}, {"loss": 1.7515, "grad_norm": 0.34052133560180664, "learning_rate": 0.0002, "epoch": 1.1358855700462769, "step": 1350}, {"loss": 1.7513, "grad_norm": 0.3467716574668884, "learning_rate": 0.0002, "epoch": 1.1442995372318048, "step": 1360}, {"loss": 1.7086, "grad_norm": 0.35528799891471863, "learning_rate": 0.0002, "epoch": 1.1527135044173327, "step": 1370}, {"loss": 1.794, "grad_norm": 0.36282262206077576, "learning_rate": 0.0002, "epoch": 1.1611274716028608, "step": 1380}, {"loss": 1.7731, "grad_norm": 0.37355899810791016, "learning_rate": 0.0002, "epoch": 1.1695414387883887, "step": 1390}, {"loss": 1.7483, "grad_norm": 0.37292736768722534, "learning_rate": 0.0002, "epoch": 1.1779554059739168, "step": 1400}, {"loss": 1.6916, "grad_norm": 0.5892812013626099, "learning_rate": 0.0002, "epoch": 1.1863693731594447, "step": 1410}, {"loss": 1.7302, "grad_norm": 0.3712292015552521, "learning_rate": 0.0002, "epoch": 1.1947833403449726, "step": 1420}, {"loss": 1.7709, "grad_norm": 0.3349577486515045, "learning_rate": 0.0002, "epoch": 1.2031973075305007, "step": 1430}, {"loss": 1.7412, "grad_norm": 0.32591062784194946, "learning_rate": 0.0002, "epoch": 1.2116112747160286, "step": 1440}, {"loss": 1.7406, "grad_norm": 0.3840635418891907, "learning_rate": 0.0002, "epoch": 1.2200252419015567, "step": 1450}, {"loss": 1.7276, "grad_norm": 0.37238365411758423, "learning_rate": 0.0002, "epoch": 1.2284392090870846, "step": 1460}, {"loss": 1.7052, "grad_norm": 0.3731217682361603, "learning_rate": 0.0002, "epoch": 1.2368531762726125, "step": 1470}, {"loss": 1.7255, "grad_norm": 0.3318967819213867, "learning_rate": 0.0002, "epoch": 1.2452671434581406, "step": 1480}, {"loss": 1.7463, "grad_norm": 0.3784034848213196, "learning_rate": 0.0002, "epoch": 1.2536811106436685, "step": 1490}, {"loss": 1.6862, "grad_norm": 0.3541383147239685, "learning_rate": 0.0002, "epoch": 1.2620950778291964, "step": 1500}, {"loss": 1.8394, "grad_norm": 0.35312485694885254, "learning_rate": 0.0002, "epoch": 1.2705090450147245, "step": 1510}, {"loss": 1.7029, "grad_norm": 0.35272929072380066, "learning_rate": 0.0002, "epoch": 1.2789230122002524, "step": 1520}, {"loss": 1.7016, "grad_norm": 0.40988272428512573, "learning_rate": 0.0002, "epoch": 1.2873369793857803, "step": 1530}, {"loss": 1.6912, "grad_norm": 0.3543946146965027, "learning_rate": 0.0002, "epoch": 1.2957509465713084, "step": 1540}, {"loss": 1.6757, "grad_norm": 0.35639145970344543, "learning_rate": 0.0002, "epoch": 1.3041649137568363, "step": 1550}, {"loss": 1.6814, "grad_norm": 0.3290826678276062, "learning_rate": 0.0002, "epoch": 1.3125788809423642, "step": 1560}, {"loss": 1.7369, "grad_norm": 0.39264336228370667, "learning_rate": 0.0002, "epoch": 1.3209928481278923, "step": 1570}, {"loss": 1.6804, "grad_norm": 0.5390415191650391, "learning_rate": 0.0002, "epoch": 1.3294068153134202, "step": 1580}, {"loss": 1.708, "grad_norm": 0.5188116431236267, "learning_rate": 0.0002, "epoch": 1.3378207824989483, "step": 1590}, {"loss": 1.6763, "grad_norm": 0.37445148825645447, "learning_rate": 0.0002, "epoch": 1.3462347496844762, "step": 1600}, {"loss": 1.7386, "grad_norm": 0.3296085298061371, "learning_rate": 0.0002, "epoch": 1.3546487168700043, "step": 1610}, {"loss": 1.8107, "grad_norm": 0.39879581332206726, "learning_rate": 0.0002, "epoch": 1.3630626840555322, "step": 1620}, {"loss": 1.6744, "grad_norm": 0.36092764139175415, "learning_rate": 0.0002, "epoch": 1.37147665124106, "step": 1630}, {"loss": 1.7144, "grad_norm": 0.37011823058128357, "learning_rate": 0.0002, "epoch": 1.3798906184265882, "step": 1640}, {"loss": 1.7396, "grad_norm": 0.40863534808158875, "learning_rate": 0.0002, "epoch": 1.3883045856121161, "step": 1650}, {"loss": 1.7901, "grad_norm": 0.337001770734787, "learning_rate": 0.0002, "epoch": 1.396718552797644, "step": 1660}, {"loss": 1.7044, "grad_norm": 0.35596707463264465, "learning_rate": 0.0002, "epoch": 1.4051325199831721, "step": 1670}, {"loss": 1.7717, "grad_norm": 0.3857671916484833, "learning_rate": 0.0002, "epoch": 1.4135464871687, "step": 1680}, {"loss": 1.7015, "grad_norm": 0.419502317905426, "learning_rate": 0.0002, "epoch": 1.421960454354228, "step": 1690}, {"loss": 1.7261, "grad_norm": 0.35459452867507935, "learning_rate": 0.0002, "epoch": 1.430374421539756, "step": 1700}, {"loss": 1.7361, "grad_norm": 0.37246978282928467, "learning_rate": 0.0002, "epoch": 1.438788388725284, "step": 1710}, {"loss": 1.6762, "grad_norm": 0.33091893792152405, "learning_rate": 0.0002, "epoch": 1.4472023559108118, "step": 1720}, {"loss": 1.7044, "grad_norm": 0.37029674649238586, "learning_rate": 0.0002, "epoch": 1.45561632309634, "step": 1730}, {"loss": 1.7117, "grad_norm": 0.374025821685791, "learning_rate": 0.0002, "epoch": 1.4640302902818678, "step": 1740}, {"loss": 1.7549, "grad_norm": 0.3416315019130707, "learning_rate": 0.0002, "epoch": 1.472444257467396, "step": 1750}, {"loss": 1.7093, "grad_norm": 0.36502841114997864, "learning_rate": 0.0002, "epoch": 1.4808582246529238, "step": 1760}, {"loss": 1.6597, "grad_norm": 0.35458803176879883, "learning_rate": 0.0002, "epoch": 1.489272191838452, "step": 1770}, {"loss": 1.675, "grad_norm": 0.4462839663028717, "learning_rate": 0.0002, "epoch": 1.4976861590239798, "step": 1780}, {"loss": 1.7267, "grad_norm": 0.34836092591285706, "learning_rate": 0.0002, "epoch": 1.5061001262095077, "step": 1790}, {"loss": 1.7295, "grad_norm": 0.3445749282836914, "learning_rate": 0.0002, "epoch": 1.5145140933950358, "step": 1800}, {"loss": 1.7386, "grad_norm": 0.36012160778045654, "learning_rate": 0.0002, "epoch": 1.5229280605805637, "step": 1810}, {"loss": 1.6594, "grad_norm": 0.4052616059780121, "learning_rate": 0.0002, "epoch": 1.5313420277660916, "step": 1820}, {"loss": 1.72, "grad_norm": 0.3966905474662781, "learning_rate": 0.0002, "epoch": 1.5397559949516197, "step": 1830}, {"loss": 1.7595, "grad_norm": 0.35028719902038574, "learning_rate": 0.0002, "epoch": 1.5481699621371476, "step": 1840}, {"loss": 1.6829, "grad_norm": 0.3936742842197418, "learning_rate": 0.0002, "epoch": 1.5565839293226755, "step": 1850}, {"loss": 1.7579, "grad_norm": 0.34473296999931335, "learning_rate": 0.0002, "epoch": 1.5649978965082036, "step": 1860}, {"loss": 1.7207, "grad_norm": 0.4328365623950958, "learning_rate": 0.0002, "epoch": 1.5734118636937318, "step": 1870}, {"loss": 1.7098, "grad_norm": 0.3566315472126007, "learning_rate": 0.0002, "epoch": 1.5818258308792594, "step": 1880}, {"loss": 1.6095, "grad_norm": 0.3301256597042084, "learning_rate": 0.0002, "epoch": 1.5902397980647875, "step": 1890}, {"loss": 1.748, "grad_norm": 0.3743041455745697, "learning_rate": 0.0002, "epoch": 1.5986537652503157, "step": 1900}, {"loss": 1.7259, "grad_norm": 0.3735344707965851, "learning_rate": 0.0002, "epoch": 1.6070677324358436, "step": 1910}, {"loss": 1.7445, "grad_norm": 0.42191144824028015, "learning_rate": 0.0002, "epoch": 1.6154816996213714, "step": 1920}, {"loss": 1.6978, "grad_norm": 0.3787207305431366, "learning_rate": 0.0002, "epoch": 1.6238956668068996, "step": 1930}, {"loss": 1.6893, "grad_norm": 0.35647350549697876, "learning_rate": 0.0002, "epoch": 1.6323096339924275, "step": 1940}, {"loss": 1.7825, "grad_norm": 0.39791446924209595, "learning_rate": 0.0002, "epoch": 1.6407236011779553, "step": 1950}, {"loss": 1.7293, "grad_norm": 0.37341275811195374, "learning_rate": 0.0002, "epoch": 1.6491375683634835, "step": 1960}, {"loss": 1.6781, "grad_norm": 0.3722686469554901, "learning_rate": 0.0002, "epoch": 1.6575515355490114, "step": 1970}, {"loss": 1.6383, "grad_norm": 0.37467387318611145, "learning_rate": 0.0002, "epoch": 1.6659655027345392, "step": 1980}, {"loss": 1.7439, "grad_norm": 0.37109461426734924, "learning_rate": 0.0002, "epoch": 1.6743794699200674, "step": 1990}, {"loss": 1.7206, "grad_norm": 0.4008837044239044, "learning_rate": 0.0002, "epoch": 1.6827934371055953, "step": 2000}, {"loss": 1.7604, "grad_norm": 0.3316999673843384, "learning_rate": 0.0002, "epoch": 1.6912074042911232, "step": 2010}, {"loss": 1.7325, "grad_norm": 0.3683805465698242, "learning_rate": 0.0002, "epoch": 1.6996213714766513, "step": 2020}, {"loss": 1.7451, "grad_norm": 0.4163658320903778, "learning_rate": 0.0002, "epoch": 1.7080353386621794, "step": 2030}, {"loss": 1.741, "grad_norm": 0.4245431125164032, "learning_rate": 0.0002, "epoch": 1.716449305847707, "step": 2040}, {"loss": 1.7184, "grad_norm": 0.36732038855552673, "learning_rate": 0.0002, "epoch": 1.7248632730332352, "step": 2050}, {"loss": 1.7031, "grad_norm": 0.34981656074523926, "learning_rate": 0.0002, "epoch": 1.7332772402187633, "step": 2060}, {"loss": 1.7545, "grad_norm": 0.38588812947273254, "learning_rate": 0.0002, "epoch": 1.7416912074042912, "step": 2070}, {"loss": 1.7728, "grad_norm": 0.39914557337760925, "learning_rate": 0.0002, "epoch": 1.750105174589819, "step": 2080}, {"loss": 1.7049, "grad_norm": 0.36068692803382874, "learning_rate": 0.0002, "epoch": 1.7585191417753472, "step": 2090}, {"loss": 1.7537, "grad_norm": 0.3983287215232849, "learning_rate": 0.0002, "epoch": 1.766933108960875, "step": 2100}, {"loss": 1.7016, "grad_norm": 0.45008400082588196, "learning_rate": 0.0002, "epoch": 1.775347076146403, "step": 2110}, {"loss": 1.7163, "grad_norm": 0.3618052303791046, "learning_rate": 0.0002, "epoch": 1.783761043331931, "step": 2120}, {"loss": 1.7335, "grad_norm": 0.38745400309562683, "learning_rate": 0.0002, "epoch": 1.792175010517459, "step": 2130}, {"loss": 1.7387, "grad_norm": 0.3413826525211334, "learning_rate": 0.0002, "epoch": 1.8005889777029869, "step": 2140}, {"loss": 1.7414, "grad_norm": 0.35983747243881226, "learning_rate": 0.0002, "epoch": 1.809002944888515, "step": 2150}, {"loss": 1.7892, "grad_norm": 0.40926849842071533, "learning_rate": 0.0002, "epoch": 1.8174169120740429, "step": 2160}, {"loss": 1.6823, "grad_norm": 0.3543093800544739, "learning_rate": 0.0002, "epoch": 1.8258308792595708, "step": 2170}, {"loss": 1.7812, "grad_norm": 0.42690935730934143, "learning_rate": 0.0002, "epoch": 1.8342448464450989, "step": 2180}, {"loss": 1.7471, "grad_norm": 0.40282756090164185, "learning_rate": 0.0002, "epoch": 1.842658813630627, "step": 2190}, {"loss": 1.7411, "grad_norm": 0.36568400263786316, "learning_rate": 0.0002, "epoch": 1.8510727808161547, "step": 2200}, {"loss": 1.7024, "grad_norm": 0.43159013986587524, "learning_rate": 0.0002, "epoch": 1.8594867480016828, "step": 2210}, {"loss": 1.7298, "grad_norm": 0.3554118573665619, "learning_rate": 0.0002, "epoch": 1.867900715187211, "step": 2220}, {"loss": 1.7157, "grad_norm": 0.43349072337150574, "learning_rate": 0.0002, "epoch": 1.8763146823727388, "step": 2230}, {"loss": 1.7302, "grad_norm": 0.36486536264419556, "learning_rate": 0.0002, "epoch": 1.8847286495582667, "step": 2240}, {"loss": 1.6901, "grad_norm": 0.39260047674179077, "learning_rate": 0.0002, "epoch": 1.8931426167437948, "step": 2250}, {"loss": 1.6691, "grad_norm": 0.3741776943206787, "learning_rate": 0.0002, "epoch": 1.9015565839293227, "step": 2260}, {"loss": 1.6931, "grad_norm": 0.3961946964263916, "learning_rate": 0.0002, "epoch": 1.9099705511148506, "step": 2270}, {"loss": 1.737, "grad_norm": 0.3659731149673462, "learning_rate": 0.0002, "epoch": 1.9183845183003787, "step": 2280}, {"loss": 1.7342, "grad_norm": 0.34744107723236084, "learning_rate": 0.0002, "epoch": 1.9267984854859066, "step": 2290}, {"loss": 1.7162, "grad_norm": 0.3607442378997803, "learning_rate": 0.0002, "epoch": 1.9352124526714345, "step": 2300}, {"loss": 1.6673, "grad_norm": 0.331464558839798, "learning_rate": 0.0002, "epoch": 1.9436264198569626, "step": 2310}, {"loss": 1.7101, "grad_norm": 0.3904414474964142, "learning_rate": 0.0002, "epoch": 1.9520403870424905, "step": 2320}, {"loss": 1.7327, "grad_norm": 0.37584832310676575, "learning_rate": 0.0002, "epoch": 1.9604543542280184, "step": 2330}, {"loss": 1.7586, "grad_norm": 0.3698684275150299, "learning_rate": 0.0002, "epoch": 1.9688683214135465, "step": 2340}, {"loss": 1.7764, "grad_norm": 0.40571412444114685, "learning_rate": 0.0002, "epoch": 1.9772822885990746, "step": 2350}, {"loss": 1.744, "grad_norm": 0.40059587359428406, "learning_rate": 0.0002, "epoch": 1.9856962557846023, "step": 2360}, {"loss": 1.7033, "grad_norm": 0.4168248474597931, "learning_rate": 0.0002, "epoch": 1.9941102229701304, "step": 2370}, {"eval_loss": 1.8055059909820557, "eval_runtime": 38.422, "eval_samples_per_second": 13.404, "eval_steps_per_second": 1.692, "epoch": 2.0, "step": 2377}, {"loss": 1.7673, "grad_norm": 0.35205352306365967, "learning_rate": 0.0002, "epoch": 2.0025241901556585, "step": 2380}, {"loss": 1.6556, "grad_norm": 0.3979377746582031, "learning_rate": 0.0002, "epoch": 2.010938157341186, "step": 2390}, {"loss": 1.6421, "grad_norm": 0.396491676568985, "learning_rate": 0.0002, "epoch": 2.0193521245267143, "step": 2400}, {"loss": 1.6847, "grad_norm": 0.44712209701538086, "learning_rate": 0.0002, "epoch": 2.0277660917122424, "step": 2410}, {"loss": 1.6877, "grad_norm": 0.4454420208930969, "learning_rate": 0.0002, "epoch": 2.03618005889777, "step": 2420}, {"loss": 1.6635, "grad_norm": 0.4170038402080536, "learning_rate": 0.0002, "epoch": 2.044594026083298, "step": 2430}, {"loss": 1.6512, "grad_norm": 0.4309595227241516, "learning_rate": 0.0002, "epoch": 2.0530079932688263, "step": 2440}, {"loss": 1.6223, "grad_norm": 0.4241602122783661, "learning_rate": 0.0002, "epoch": 2.0614219604543544, "step": 2450}, {"loss": 1.6162, "grad_norm": 0.4370540678501129, "learning_rate": 0.0002, "epoch": 2.069835927639882, "step": 2460}, {"loss": 1.6354, "grad_norm": 0.43985554575920105, "learning_rate": 0.0002, "epoch": 2.0782498948254102, "step": 2470}, {"loss": 1.6954, "grad_norm": 0.4158105254173279, "learning_rate": 0.0002, "epoch": 2.0866638620109383, "step": 2480}, {"loss": 1.6114, "grad_norm": 0.441549152135849, "learning_rate": 0.0002, "epoch": 2.095077829196466, "step": 2490}, {"loss": 1.5485, "grad_norm": 0.385718435049057, "learning_rate": 0.0002, "epoch": 2.103491796381994, "step": 2500}, {"loss": 1.5894, "grad_norm": 0.43146514892578125, "learning_rate": 0.0002, "epoch": 2.1119057635675222, "step": 2510}, {"loss": 1.6414, "grad_norm": 0.41663315892219543, "learning_rate": 0.0002, "epoch": 2.12031973075305, "step": 2520}, {"loss": 1.6527, "grad_norm": 0.4410698115825653, "learning_rate": 0.0002, "epoch": 2.128733697938578, "step": 2530}, {"loss": 1.6124, "grad_norm": 0.4472278952598572, "learning_rate": 0.0002, "epoch": 2.137147665124106, "step": 2540}, {"loss": 1.6257, "grad_norm": 0.3879167437553406, "learning_rate": 0.0002, "epoch": 2.145561632309634, "step": 2550}, {"loss": 1.6682, "grad_norm": 0.4212203025817871, "learning_rate": 0.0002, "epoch": 2.153975599495162, "step": 2560}, {"loss": 1.6036, "grad_norm": 0.42841723561286926, "learning_rate": 0.0002, "epoch": 2.16238956668069, "step": 2570}, {"loss": 1.5962, "grad_norm": 0.39272481203079224, "learning_rate": 0.0002, "epoch": 2.1708035338662177, "step": 2580}, {"loss": 1.681, "grad_norm": 0.4075261354446411, "learning_rate": 0.0002, "epoch": 2.179217501051746, "step": 2590}, {"loss": 1.6601, "grad_norm": 0.5358437895774841, "learning_rate": 0.0002, "epoch": 2.187631468237274, "step": 2600}, {"loss": 1.6423, "grad_norm": 0.4738350212574005, "learning_rate": 0.0002, "epoch": 2.1960454354228016, "step": 2610}, {"loss": 1.6386, "grad_norm": 0.446789026260376, "learning_rate": 0.0002, "epoch": 2.2044594026083297, "step": 2620}, {"loss": 1.6246, "grad_norm": 0.4615374505519867, "learning_rate": 0.0002, "epoch": 2.212873369793858, "step": 2630}, {"loss": 1.6205, "grad_norm": 0.46901994943618774, "learning_rate": 0.0002, "epoch": 2.221287336979386, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.46267789602279663, "learning_rate": 0.0002, "epoch": 2.2297013041649136, "step": 2650}, {"loss": 1.6584, "grad_norm": 0.4383080005645752, "learning_rate": 0.0002, "epoch": 2.2381152713504417, "step": 2660}, {"loss": 1.5745, "grad_norm": 0.4070609509944916, "learning_rate": 0.0002, "epoch": 2.24652923853597, "step": 2670}, {"loss": 1.6125, "grad_norm": 0.4572339951992035, "learning_rate": 0.0002, "epoch": 2.2549432057214975, "step": 2680}, {"loss": 1.5671, "grad_norm": 0.393265038728714, "learning_rate": 0.0002, "epoch": 2.2633571729070256, "step": 2690}, {"loss": 1.6239, "grad_norm": 0.46144717931747437, "learning_rate": 0.0002, "epoch": 2.2717711400925538, "step": 2700}, {"loss": 1.5992, "grad_norm": 0.45077767968177795, "learning_rate": 0.0002, "epoch": 2.2801851072780814, "step": 2710}, {"loss": 1.6261, "grad_norm": 0.5697639584541321, "learning_rate": 0.0002, "epoch": 2.2885990744636096, "step": 2720}, {"loss": 1.6192, "grad_norm": 0.4855510890483856, "learning_rate": 0.0002, "epoch": 2.2970130416491377, "step": 2730}, {"loss": 1.7419, "grad_norm": 0.4440622627735138, "learning_rate": 0.0002, "epoch": 2.3054270088346653, "step": 2740}, {"loss": 1.6496, "grad_norm": 0.3904096782207489, "learning_rate": 0.0002, "epoch": 2.3138409760201935, "step": 2750}, {"loss": 1.5888, "grad_norm": 0.5225510597229004, "learning_rate": 0.0002, "epoch": 2.3222549432057216, "step": 2760}, {"loss": 1.6082, "grad_norm": 0.44866397976875305, "learning_rate": 0.0002, "epoch": 2.3306689103912497, "step": 2770}, {"loss": 1.6087, "grad_norm": 0.5167056322097778, "learning_rate": 0.0002, "epoch": 2.3390828775767774, "step": 2780}, {"loss": 1.6136, "grad_norm": 0.45913267135620117, "learning_rate": 0.0002, "epoch": 2.3474968447623055, "step": 2790}, {"loss": 1.6564, "grad_norm": 0.45787590742111206, "learning_rate": 0.0002, "epoch": 2.3559108119478336, "step": 2800}, {"loss": 1.6868, "grad_norm": 0.4633352756500244, "learning_rate": 0.0002, "epoch": 2.3643247791333613, "step": 2810}, {"loss": 1.6316, "grad_norm": 0.46390071511268616, "learning_rate": 0.0002, "epoch": 2.3727387463188894, "step": 2820}, {"loss": 1.6039, "grad_norm": 0.4261005222797394, "learning_rate": 0.0002, "epoch": 2.3811527135044175, "step": 2830}, {"loss": 1.6364, "grad_norm": 0.4283634424209595, "learning_rate": 0.0002, "epoch": 2.389566680689945, "step": 2840}, {"loss": 1.6382, "grad_norm": 0.4955291450023651, "learning_rate": 0.0002, "epoch": 2.3979806478754733, "step": 2850}, {"loss": 1.6173, "grad_norm": 0.4740189015865326, "learning_rate": 0.0002, "epoch": 2.4063946150610014, "step": 2860}, {"loss": 1.6403, "grad_norm": 0.4222276508808136, "learning_rate": 0.0002, "epoch": 2.414808582246529, "step": 2870}, {"loss": 1.5602, "grad_norm": 0.4982149004936218, "learning_rate": 0.0002, "epoch": 2.423222549432057, "step": 2880}, {"loss": 1.6313, "grad_norm": 0.5217409133911133, "learning_rate": 0.0002, "epoch": 2.4316365166175853, "step": 2890}, {"loss": 1.5804, "grad_norm": 0.4555884897708893, "learning_rate": 0.0002, "epoch": 2.4400504838031134, "step": 2900}, {"loss": 1.6189, "grad_norm": 0.43178579211235046, "learning_rate": 0.0002, "epoch": 2.448464450988641, "step": 2910}, {"loss": 1.6824, "grad_norm": 0.4788478910923004, "learning_rate": 0.0002, "epoch": 2.456878418174169, "step": 2920}, {"loss": 1.6829, "grad_norm": 0.43689873814582825, "learning_rate": 0.0002, "epoch": 2.465292385359697, "step": 2930}, {"loss": 1.6196, "grad_norm": 0.5115197896957397, "learning_rate": 0.0002, "epoch": 2.473706352545225, "step": 2940}, {"loss": 1.689, "grad_norm": 0.5290159583091736, "learning_rate": 0.0002, "epoch": 2.482120319730753, "step": 2950}, {"loss": 1.6499, "grad_norm": 0.46042463183403015, "learning_rate": 0.0002, "epoch": 2.490534286916281, "step": 2960}, {"loss": 1.6664, "grad_norm": 0.4359915852546692, "learning_rate": 0.0002, "epoch": 2.498948254101809, "step": 2970}, {"loss": 1.5812, "grad_norm": 0.46352964639663696, "learning_rate": 0.0002, "epoch": 2.507362221287337, "step": 2980}, {"loss": 1.6501, "grad_norm": 0.5324268341064453, "learning_rate": 0.0002, "epoch": 2.515776188472865, "step": 2990}, {"loss": 1.6115, "grad_norm": 0.5929607152938843, "learning_rate": 0.0002, "epoch": 2.5241901556583928, "step": 3000}, {"loss": 1.6772, "grad_norm": 0.4811333417892456, "learning_rate": 0.0002, "epoch": 2.532604122843921, "step": 3010}, {"loss": 1.7023, "grad_norm": 0.4662701487541199, "learning_rate": 0.0002, "epoch": 2.541018090029449, "step": 3020}, {"loss": 1.5426, "grad_norm": 0.4582270681858063, "learning_rate": 0.0002, "epoch": 2.549432057214977, "step": 3030}, {"loss": 1.6737, "grad_norm": 0.4679982662200928, "learning_rate": 0.0002, "epoch": 2.557846024400505, "step": 3040}, {"loss": 1.5442, "grad_norm": 0.4380294680595398, "learning_rate": 0.0002, "epoch": 2.566259991586033, "step": 3050}, {"loss": 1.6055, "grad_norm": 0.44295763969421387, "learning_rate": 0.0002, "epoch": 2.5746739587715606, "step": 3060}, {"loss": 1.5775, "grad_norm": 0.5131027698516846, "learning_rate": 0.0002, "epoch": 2.5830879259570887, "step": 3070}, {"loss": 1.546, "grad_norm": 0.47567516565322876, "learning_rate": 0.0002, "epoch": 2.591501893142617, "step": 3080}, {"loss": 1.5671, "grad_norm": 0.49002596735954285, "learning_rate": 0.0002, "epoch": 2.599915860328145, "step": 3090}, {"loss": 1.5445, "grad_norm": 0.44856327772140503, "learning_rate": 0.0002, "epoch": 2.6083298275136726, "step": 3100}, {"loss": 1.5797, "grad_norm": 0.4480142593383789, "learning_rate": 0.0002, "epoch": 2.6167437946992007, "step": 3110}, {"loss": 1.7132, "grad_norm": 0.4317494034767151, "learning_rate": 0.0002, "epoch": 2.6251577618847284, "step": 3120}, {"loss": 1.6321, "grad_norm": 0.42580848932266235, "learning_rate": 0.0002, "epoch": 2.6335717290702565, "step": 3130}, {"loss": 1.6483, "grad_norm": 0.4516814947128296, "learning_rate": 0.0002, "epoch": 2.6419856962557846, "step": 3140}, {"loss": 1.695, "grad_norm": 0.4438435733318329, "learning_rate": 0.0002, "epoch": 2.6503996634413127, "step": 3150}, {"loss": 1.6938, "grad_norm": 0.4385356307029724, "learning_rate": 0.0002, "epoch": 2.6588136306268404, "step": 3160}, {"loss": 1.6139, "grad_norm": 0.5064112544059753, "learning_rate": 0.0002, "epoch": 2.6672275978123685, "step": 3170}, {"loss": 1.7189, "grad_norm": 0.49163177609443665, "learning_rate": 0.0002, "epoch": 2.6756415649978966, "step": 3180}, {"loss": 1.7323, "grad_norm": 0.49339258670806885, "learning_rate": 0.0002, "epoch": 2.6840555321834243, "step": 3190}, {"loss": 1.6508, "grad_norm": 0.440950870513916, "learning_rate": 0.0002, "epoch": 2.6924694993689524, "step": 3200}, {"loss": 1.6305, "grad_norm": 0.4283970594406128, "learning_rate": 0.0002, "epoch": 2.7008834665544805, "step": 3210}, {"loss": 1.5935, "grad_norm": 0.43875712156295776, "learning_rate": 0.0002, "epoch": 2.7092974337400086, "step": 3220}, {"loss": 1.6129, "grad_norm": 0.49332964420318604, "learning_rate": 0.0002, "epoch": 2.7177114009255363, "step": 3230}, {"loss": 1.642, "grad_norm": 0.5225692391395569, "learning_rate": 0.0002, "epoch": 2.7261253681110644, "step": 3240}, {"loss": 1.6759, "grad_norm": 0.4856489300727844, "learning_rate": 0.0002, "epoch": 2.734539335296592, "step": 3250}, {"loss": 1.6463, "grad_norm": 0.46918296813964844, "learning_rate": 0.0002, "epoch": 2.74295330248212, "step": 3260}, {"loss": 1.6819, "grad_norm": 0.4802931249141693, "learning_rate": 0.0002, "epoch": 2.7513672696676483, "step": 3270}, {"loss": 1.6246, "grad_norm": 0.4485355615615845, "learning_rate": 0.0002, "epoch": 2.7597812368531764, "step": 3280}, {"loss": 1.6251, "grad_norm": 0.43944594264030457, "learning_rate": 0.0002, "epoch": 2.768195204038704, "step": 3290}, {"loss": 1.6501, "grad_norm": 0.46847742795944214, "learning_rate": 0.0002, "epoch": 2.7766091712242322, "step": 3300}, {"loss": 1.5969, "grad_norm": 0.4816027879714966, "learning_rate": 0.0002, "epoch": 2.7850231384097603, "step": 3310}, {"loss": 1.6293, "grad_norm": 0.453960120677948, "learning_rate": 0.0002, "epoch": 2.793437105595288, "step": 3320}, {"loss": 1.6429, "grad_norm": 0.4816017150878906, "learning_rate": 0.0002, "epoch": 2.801851072780816, "step": 3330}, {"loss": 1.6683, "grad_norm": 0.4461034834384918, "learning_rate": 0.0002, "epoch": 2.8102650399663442, "step": 3340}, {"loss": 1.7048, "grad_norm": 0.48821821808815, "learning_rate": 0.0002, "epoch": 2.8186790071518724, "step": 3350}, {"loss": 1.6076, "grad_norm": 0.4574853777885437, "learning_rate": 0.0002, "epoch": 2.8270929743374, "step": 3360}, {"loss": 1.6651, "grad_norm": 0.42062026262283325, "learning_rate": 0.0002, "epoch": 2.835506941522928, "step": 3370}, {"loss": 1.624, "grad_norm": 0.4499834477901459, "learning_rate": 0.0002, "epoch": 2.843920908708456, "step": 3380}, {"loss": 1.621, "grad_norm": 0.4780360758304596, "learning_rate": 0.0002, "epoch": 2.852334875893984, "step": 3390}, {"loss": 1.5882, "grad_norm": 0.45422887802124023, "learning_rate": 0.0002, "epoch": 2.860748843079512, "step": 3400}, {"loss": 1.6028, "grad_norm": 0.4590015709400177, "learning_rate": 0.0002, "epoch": 2.86916281026504, "step": 3410}, {"loss": 1.6746, "grad_norm": 0.45689624547958374, "learning_rate": 0.0002, "epoch": 2.877576777450568, "step": 3420}, {"loss": 1.6326, "grad_norm": 0.46953922510147095, "learning_rate": 0.0002, "epoch": 2.885990744636096, "step": 3430}, {"loss": 1.6015, "grad_norm": 0.4791966378688812, "learning_rate": 0.0002, "epoch": 2.8944047118216236, "step": 3440}, {"loss": 1.694, "grad_norm": 0.4842296242713928, "learning_rate": 0.0002, "epoch": 2.9028186790071517, "step": 3450}, {"loss": 1.6326, "grad_norm": 0.47219768166542053, "learning_rate": 0.0002, "epoch": 2.91123264619268, "step": 3460}, {"loss": 1.6486, "grad_norm": 0.4622127115726471, "learning_rate": 0.0002, "epoch": 2.919646613378208, "step": 3470}, {"loss": 1.6485, "grad_norm": 0.46832820773124695, "learning_rate": 0.0002, "epoch": 2.9280605805637356, "step": 3480}, {"loss": 1.6366, "grad_norm": 0.44582483172416687, "learning_rate": 0.0002, "epoch": 2.9364745477492638, "step": 3490}, {"loss": 1.6859, "grad_norm": 0.4987219274044037, "learning_rate": 0.0002, "epoch": 2.944888514934792, "step": 3500}, {"loss": 1.5991, "grad_norm": 0.43750956654548645, "learning_rate": 0.0002, "epoch": 2.9533024821203195, "step": 3510}, {"loss": 1.6236, "grad_norm": 0.49962925910949707, "learning_rate": 0.0002, "epoch": 2.9617164493058477, "step": 3520}, {"loss": 1.5859, "grad_norm": 0.5189590454101562, "learning_rate": 0.0002, "epoch": 2.9701304164913758, "step": 3530}, {"loss": 1.6688, "grad_norm": 0.391317754983902, "learning_rate": 0.0002, "epoch": 2.978544383676904, "step": 3540}, {"loss": 1.5884, "grad_norm": 0.44934695959091187, "learning_rate": 0.0002, "epoch": 2.9869583508624316, "step": 3550}, {"loss": 1.5688, "grad_norm": 0.4740142226219177, "learning_rate": 0.0002, "epoch": 2.9953723180479597, "step": 3560}, {"eval_loss": 1.8266887664794922, "eval_runtime": 37.9445, "eval_samples_per_second": 13.572, "eval_steps_per_second": 1.713, "epoch": 2.9995793016407237, "step": 3565}, {"loss": 1.5939, "grad_norm": 0.4523724615573883, "learning_rate": 0.0002, "epoch": 3.003786285233488, "step": 3570}, {"loss": 1.526, "grad_norm": 0.5261380076408386, "learning_rate": 0.0002, "epoch": 3.0122002524190155, "step": 3580}, {"loss": 1.4946, "grad_norm": 0.48664888739585876, "learning_rate": 0.0002, "epoch": 3.0206142196045436, "step": 3590}, {"loss": 1.5193, "grad_norm": 0.5070882439613342, "learning_rate": 0.0002, "epoch": 3.0290281867900717, "step": 3600}, {"loss": 1.5316, "grad_norm": 0.5816011428833008, "learning_rate": 0.0002, "epoch": 3.0374421539755994, "step": 3610}, {"loss": 1.5682, "grad_norm": 0.6610211730003357, "learning_rate": 0.0002, "epoch": 3.0458561211611275, "step": 3620}, {"loss": 1.5699, "grad_norm": 0.5257703065872192, "learning_rate": 0.0002, "epoch": 3.0542700883466556, "step": 3630}, {"loss": 1.4438, "grad_norm": 0.5574390888214111, "learning_rate": 0.0002, "epoch": 3.0626840555321833, "step": 3640}, {"loss": 1.547, "grad_norm": 0.5682297348976135, "learning_rate": 0.0002, "epoch": 3.0710980227177114, "step": 3650}, {"loss": 1.5743, "grad_norm": 0.5798383355140686, "learning_rate": 0.0002, "epoch": 3.0795119899032395, "step": 3660}, {"loss": 1.4339, "grad_norm": 0.5458289980888367, "learning_rate": 0.0002, "epoch": 3.087925957088767, "step": 3670}, {"loss": 1.46, "grad_norm": 0.5599102973937988, "learning_rate": 0.0002, "epoch": 3.0963399242742953, "step": 3680}, {"loss": 1.4589, "grad_norm": 0.5023021697998047, "learning_rate": 0.0002, "epoch": 3.1047538914598234, "step": 3690}, {"loss": 1.5114, "grad_norm": 0.5448206067085266, "learning_rate": 0.0002, "epoch": 3.113167858645351, "step": 3700}, {"loss": 1.4692, "grad_norm": 0.5760458707809448, "learning_rate": 0.0002, "epoch": 3.121581825830879, "step": 3710}, {"loss": 1.4789, "grad_norm": 0.6018968224525452, "learning_rate": 0.0002, "epoch": 3.1299957930164073, "step": 3720}, {"loss": 1.5518, "grad_norm": 0.5767101049423218, "learning_rate": 0.0002, "epoch": 3.1384097602019354, "step": 3730}, {"loss": 1.5032, "grad_norm": 0.5333963632583618, "learning_rate": 0.0002, "epoch": 3.146823727387463, "step": 3740}, {"loss": 1.4812, "grad_norm": 0.5918396711349487, "learning_rate": 0.0002, "epoch": 3.155237694572991, "step": 3750}, {"loss": 1.4618, "grad_norm": 0.5931203365325928, "learning_rate": 0.0002, "epoch": 3.1636516617585193, "step": 3760}, {"loss": 1.5592, "grad_norm": 0.6562168598175049, "learning_rate": 0.0002, "epoch": 3.172065628944047, "step": 3770}, {"loss": 1.4932, "grad_norm": 0.5820156335830688, "learning_rate": 0.0002, "epoch": 3.180479596129575, "step": 3780}, {"loss": 1.4523, "grad_norm": 0.5784737467765808, "learning_rate": 0.0002, "epoch": 3.188893563315103, "step": 3790}, {"loss": 1.498, "grad_norm": 0.5506529808044434, "learning_rate": 0.0002, "epoch": 3.197307530500631, "step": 3800}, {"loss": 1.4819, "grad_norm": 0.6101595163345337, "learning_rate": 0.0002, "epoch": 3.205721497686159, "step": 3810}, {"loss": 1.5185, "grad_norm": 0.5597806572914124, "learning_rate": 0.0002, "epoch": 3.214135464871687, "step": 3820}, {"loss": 1.5664, "grad_norm": 0.5641011595726013, "learning_rate": 0.0002, "epoch": 3.222549432057215, "step": 3830}, {"loss": 1.4702, "grad_norm": 0.5892080068588257, "learning_rate": 0.0002, "epoch": 3.230963399242743, "step": 3840}, {"loss": 1.4194, "grad_norm": 0.6034760475158691, "learning_rate": 0.0002, "epoch": 3.239377366428271, "step": 3850}, {"loss": 1.5499, "grad_norm": 0.5112439393997192, "learning_rate": 0.0002, "epoch": 3.247791333613799, "step": 3860}, {"loss": 1.5132, "grad_norm": 0.56565922498703, "learning_rate": 0.0002, "epoch": 3.256205300799327, "step": 3870}, {"loss": 1.4892, "grad_norm": 0.6155247092247009, "learning_rate": 0.0002, "epoch": 3.264619267984855, "step": 3880}, {"loss": 1.5118, "grad_norm": 0.6064623594284058, "learning_rate": 0.0002, "epoch": 3.273033235170383, "step": 3890}, {"loss": 1.5236, "grad_norm": 0.6313768029212952, "learning_rate": 0.0002, "epoch": 3.2814472023559107, "step": 3900}, {"loss": 1.5551, "grad_norm": 0.5903939008712769, "learning_rate": 0.0002, "epoch": 3.289861169541439, "step": 3910}, {"loss": 1.5703, "grad_norm": 0.5770667195320129, "learning_rate": 0.0002, "epoch": 3.298275136726967, "step": 3920}, {"loss": 1.5159, "grad_norm": 0.5785196423530579, "learning_rate": 0.0002, "epoch": 3.3066891039124946, "step": 3930}, {"loss": 1.5277, "grad_norm": 0.6468310356140137, "learning_rate": 0.0002, "epoch": 3.3151030710980227, "step": 3940}, {"loss": 1.6002, "grad_norm": 0.6200279593467712, "learning_rate": 0.0002, "epoch": 3.323517038283551, "step": 3950}, {"loss": 1.5264, "grad_norm": 0.5779302716255188, "learning_rate": 0.0002, "epoch": 3.3319310054690785, "step": 3960}, {"loss": 1.4861, "grad_norm": 0.5463796854019165, "learning_rate": 0.0002, "epoch": 3.3403449726546066, "step": 3970}, {"loss": 1.541, "grad_norm": 0.6117855906486511, "learning_rate": 0.0002, "epoch": 3.3487589398401347, "step": 3980}, {"loss": 1.5566, "grad_norm": 0.5554766058921814, "learning_rate": 0.0002, "epoch": 3.357172907025663, "step": 3990}, {"loss": 1.5004, "grad_norm": 0.6012870073318481, "learning_rate": 0.0002, "epoch": 3.3655868742111905, "step": 4000}, {"loss": 1.473, "grad_norm": 0.5443974137306213, "learning_rate": 0.0002, "epoch": 3.3740008413967186, "step": 4010}, {"loss": 1.5139, "grad_norm": 0.6636057496070862, "learning_rate": 0.0002, "epoch": 3.3824148085822463, "step": 4020}, {"loss": 1.5141, "grad_norm": 0.5801246166229248, "learning_rate": 0.0002, "epoch": 3.3908287757677744, "step": 4030}, {"loss": 1.5026, "grad_norm": 0.5668839812278748, "learning_rate": 0.0002, "epoch": 3.3992427429533025, "step": 4040}, {"loss": 1.523, "grad_norm": 0.7763481736183167, "learning_rate": 0.0002, "epoch": 3.4076567101388306, "step": 4050}, {"loss": 1.4932, "grad_norm": 0.6675992608070374, "learning_rate": 0.0002, "epoch": 3.4160706773243583, "step": 4060}, {"loss": 1.4959, "grad_norm": 0.6290077567100525, "learning_rate": 0.0002, "epoch": 3.4244846445098864, "step": 4070}, {"loss": 1.5766, "grad_norm": 0.6040239930152893, "learning_rate": 0.0002, "epoch": 3.4328986116954145, "step": 4080}, {"loss": 1.5711, "grad_norm": 0.6237877607345581, "learning_rate": 0.0002, "epoch": 3.441312578880942, "step": 4090}, {"loss": 1.4961, "grad_norm": 0.5343508124351501, "learning_rate": 0.0002, "epoch": 3.4497265460664703, "step": 4100}, {"loss": 1.5123, "grad_norm": 0.6817412972450256, "learning_rate": 0.0002, "epoch": 3.4581405132519984, "step": 4110}, {"loss": 1.5377, "grad_norm": 0.7115170359611511, "learning_rate": 0.0002, "epoch": 3.466554480437526, "step": 4120}, {"loss": 1.5275, "grad_norm": 0.6127332448959351, "learning_rate": 0.0002, "epoch": 3.4749684476230542, "step": 4130}, {"loss": 1.557, "grad_norm": 0.5745994448661804, "learning_rate": 0.0002, "epoch": 3.4833824148085824, "step": 4140}, {"loss": 1.4873, "grad_norm": 0.6248795390129089, "learning_rate": 0.0002, "epoch": 3.49179638199411, "step": 4150}, {"loss": 1.4885, "grad_norm": 0.5821124911308289, "learning_rate": 0.0002, "epoch": 3.500210349179638, "step": 4160}, {"loss": 1.4937, "grad_norm": 0.561416506767273, "learning_rate": 0.0002, "epoch": 3.5086243163651663, "step": 4170}, {"loss": 1.5453, "grad_norm": 0.5848962664604187, "learning_rate": 0.0002, "epoch": 3.5170382835506944, "step": 4180}, {"loss": 1.5892, "grad_norm": 0.5335569977760315, "learning_rate": 0.0002, "epoch": 3.525452250736222, "step": 4190}, {"loss": 1.5152, "grad_norm": 0.547964870929718, "learning_rate": 0.0002, "epoch": 3.53386621792175, "step": 4200}, {"loss": 1.4887, "grad_norm": 0.6157727241516113, "learning_rate": 0.0002, "epoch": 3.542280185107278, "step": 4210}, {"loss": 1.5484, "grad_norm": 0.6163121461868286, "learning_rate": 0.0002, "epoch": 3.550694152292806, "step": 4220}, {"loss": 1.5833, "grad_norm": 0.5844616293907166, "learning_rate": 0.0002, "epoch": 3.559108119478334, "step": 4230}, {"loss": 1.5305, "grad_norm": 0.7104926109313965, "learning_rate": 0.0002, "epoch": 3.567522086663862, "step": 4240}, {"loss": 1.5161, "grad_norm": 0.5055213570594788, "learning_rate": 0.0002, "epoch": 3.57593605384939, "step": 4250}, {"loss": 1.482, "grad_norm": 0.611676812171936, "learning_rate": 0.0002, "epoch": 3.584350021034918, "step": 4260}, {"loss": 1.5048, "grad_norm": 0.6326440572738647, "learning_rate": 0.0002, "epoch": 3.592763988220446, "step": 4270}, {"loss": 1.5122, "grad_norm": 0.6290925741195679, "learning_rate": 0.0002, "epoch": 3.6011779554059737, "step": 4280}, {"loss": 1.5654, "grad_norm": 0.5691978931427002, "learning_rate": 0.0002, "epoch": 3.609591922591502, "step": 4290}, {"loss": 1.4854, "grad_norm": 0.6071329116821289, "learning_rate": 0.0002, "epoch": 3.61800588977703, "step": 4300}, {"loss": 1.5336, "grad_norm": 0.606573224067688, "learning_rate": 0.0002, "epoch": 3.626419856962558, "step": 4310}, {"loss": 1.6437, "grad_norm": 0.5515419244766235, "learning_rate": 0.0002, "epoch": 3.6348338241480858, "step": 4320}, {"loss": 1.498, "grad_norm": 0.5964660048484802, "learning_rate": 0.0002, "epoch": 3.643247791333614, "step": 4330}, {"loss": 1.544, "grad_norm": 0.5774146914482117, "learning_rate": 0.0002, "epoch": 3.6516617585191415, "step": 4340}, {"loss": 1.5566, "grad_norm": 0.5732731223106384, "learning_rate": 0.0002, "epoch": 3.6600757257046697, "step": 4350}, {"loss": 1.5682, "grad_norm": 0.7354163527488708, "learning_rate": 0.0002, "epoch": 3.6684896928901978, "step": 4360}, {"loss": 1.5225, "grad_norm": 0.6220902800559998, "learning_rate": 0.0002, "epoch": 3.676903660075726, "step": 4370}, {"loss": 1.4838, "grad_norm": 0.6053991317749023, "learning_rate": 0.0002, "epoch": 3.6853176272612536, "step": 4380}, {"loss": 1.5161, "grad_norm": 0.67010897397995, "learning_rate": 0.0002, "epoch": 3.6937315944467817, "step": 4390}, {"loss": 1.5381, "grad_norm": 0.6139186024665833, "learning_rate": 0.0002, "epoch": 3.70214556163231, "step": 4400}, {"loss": 1.5088, "grad_norm": 0.5433071851730347, "learning_rate": 0.0002, "epoch": 3.7105595288178375, "step": 4410}, {"loss": 1.5337, "grad_norm": 0.5453870296478271, "learning_rate": 0.0002, "epoch": 3.7189734960033656, "step": 4420}, {"loss": 1.4549, "grad_norm": 0.6401727199554443, "learning_rate": 0.0002, "epoch": 3.7273874631888937, "step": 4430}, {"loss": 1.503, "grad_norm": 0.6049367189407349, "learning_rate": 0.0002, "epoch": 3.735801430374422, "step": 4440}, {"loss": 1.5268, "grad_norm": 0.5740529298782349, "learning_rate": 0.0002, "epoch": 3.7442153975599495, "step": 4450}, {"loss": 1.5183, "grad_norm": 0.6521880626678467, "learning_rate": 0.0002, "epoch": 3.7526293647454776, "step": 4460}, {"loss": 1.5741, "grad_norm": 0.7096368074417114, "learning_rate": 0.0002, "epoch": 3.7610433319310053, "step": 4470}, {"loss": 1.5786, "grad_norm": 0.5886474251747131, "learning_rate": 0.0002, "epoch": 3.7694572991165334, "step": 4480}, {"loss": 1.5887, "grad_norm": 0.5821043252944946, "learning_rate": 0.0002, "epoch": 3.7778712663020615, "step": 4490}, {"loss": 1.5777, "grad_norm": 0.628892183303833, "learning_rate": 0.0002, "epoch": 3.7862852334875896, "step": 4500}, {"loss": 1.4708, "grad_norm": 0.5962669849395752, "learning_rate": 0.0002, "epoch": 3.7946992006731173, "step": 4510}, {"loss": 1.5267, "grad_norm": 0.6635549068450928, "learning_rate": 0.0002, "epoch": 3.8031131678586454, "step": 4520}, {"loss": 1.5058, "grad_norm": 0.6010760068893433, "learning_rate": 0.0002, "epoch": 3.811527135044173, "step": 4530}, {"loss": 1.6228, "grad_norm": 0.6322658658027649, "learning_rate": 0.0002, "epoch": 3.819941102229701, "step": 4540}, {"loss": 1.5029, "grad_norm": 0.5893137454986572, "learning_rate": 0.0002, "epoch": 3.8283550694152293, "step": 4550}, {"loss": 1.5435, "grad_norm": 0.7829602360725403, "learning_rate": 0.0002, "epoch": 3.8367690366007574, "step": 4560}, {"loss": 1.5453, "grad_norm": 0.6190396547317505, "learning_rate": 0.0002, "epoch": 3.845183003786285, "step": 4570}, {"loss": 1.5292, "grad_norm": 0.6662813425064087, "learning_rate": 0.0002, "epoch": 3.853596970971813, "step": 4580}, {"loss": 1.5065, "grad_norm": 0.5809855461120605, "learning_rate": 0.0002, "epoch": 3.8620109381573413, "step": 4590}, {"loss": 1.5041, "grad_norm": 0.5779069662094116, "learning_rate": 0.0002, "epoch": 3.870424905342869, "step": 4600}, {"loss": 1.498, "grad_norm": 0.5603038668632507, "learning_rate": 0.0002, "epoch": 3.878838872528397, "step": 4610}, {"loss": 1.5372, "grad_norm": 0.6274181008338928, "learning_rate": 0.0002, "epoch": 3.887252839713925, "step": 4620}, {"loss": 1.4996, "grad_norm": 0.6810959577560425, "learning_rate": 0.0002, "epoch": 3.8956668068994533, "step": 4630}, {"loss": 1.4956, "grad_norm": 0.5647315979003906, "learning_rate": 0.0002, "epoch": 3.904080774084981, "step": 4640}, {"loss": 1.5424, "grad_norm": 0.6830295324325562, "learning_rate": 0.0002, "epoch": 3.912494741270509, "step": 4650}, {"loss": 1.535, "grad_norm": 0.652565598487854, "learning_rate": 0.0002, "epoch": 3.920908708456037, "step": 4660}, {"loss": 1.4772, "grad_norm": 0.5806284546852112, "learning_rate": 0.0002, "epoch": 3.929322675641565, "step": 4670}, {"loss": 1.5812, "grad_norm": 0.6825073957443237, "learning_rate": 0.0002, "epoch": 3.937736642827093, "step": 4680}, {"loss": 1.5516, "grad_norm": 0.6149451732635498, "learning_rate": 0.0002, "epoch": 3.946150610012621, "step": 4690}, {"loss": 1.5608, "grad_norm": 0.6152557134628296, "learning_rate": 0.0002, "epoch": 3.954564577198149, "step": 4700}, {"loss": 1.4897, "grad_norm": 0.6239011883735657, "learning_rate": 0.0002, "epoch": 3.962978544383677, "step": 4710}, {"loss": 1.538, "grad_norm": 0.6485443115234375, "learning_rate": 0.0002, "epoch": 3.971392511569205, "step": 4720}, {"loss": 1.5226, "grad_norm": 0.6449228525161743, "learning_rate": 0.0002, "epoch": 3.9798064787547327, "step": 4730}, {"loss": 1.5087, "grad_norm": 0.6526407599449158, "learning_rate": 0.0002, "epoch": 3.988220445940261, "step": 4740}, {"loss": 1.5026, "grad_norm": 0.6277706027030945, "learning_rate": 0.0002, "epoch": 3.996634413125789, "step": 4750}, {"eval_loss": 1.871641755104065, "eval_runtime": 37.9637, "eval_samples_per_second": 13.566, "eval_steps_per_second": 1.712, "epoch": 4.0, "step": 4754}, {"loss": 1.4744, "grad_norm": 0.6994837522506714, "learning_rate": 0.0002, "epoch": 4.005048380311317, "step": 4760}, {"loss": 1.4433, "grad_norm": 0.8728373050689697, "learning_rate": 0.0002, "epoch": 4.013462347496845, "step": 4770}, {"loss": 1.3329, "grad_norm": 0.688679575920105, "learning_rate": 0.0002, "epoch": 4.021876314682372, "step": 4780}, {"loss": 1.3999, "grad_norm": 0.6313387155532837, "learning_rate": 0.0002, "epoch": 4.0302902818679005, "step": 4790}, {"loss": 1.3346, "grad_norm": 0.6577984690666199, "learning_rate": 0.0002, "epoch": 4.038704249053429, "step": 4800}, {"loss": 1.3403, "grad_norm": 0.7938185930252075, "learning_rate": 0.0002, "epoch": 4.047118216238957, "step": 4810}, {"loss": 1.3716, "grad_norm": 0.760399580001831, "learning_rate": 0.0002, "epoch": 4.055532183424485, "step": 4820}, {"loss": 1.4321, "grad_norm": 0.7329602241516113, "learning_rate": 0.0002, "epoch": 4.063946150610013, "step": 4830}, {"loss": 1.4133, "grad_norm": 0.7778576016426086, "learning_rate": 0.0002, "epoch": 4.07236011779554, "step": 4840}, {"loss": 1.4372, "grad_norm": 0.8235865235328674, "learning_rate": 0.0002, "epoch": 4.080774084981068, "step": 4850}, {"loss": 1.3719, "grad_norm": 0.7743754386901855, "learning_rate": 0.0002, "epoch": 4.089188052166596, "step": 4860}, {"loss": 1.3787, "grad_norm": 0.8145367503166199, "learning_rate": 0.0002, "epoch": 4.0976020193521245, "step": 4870}, {"loss": 1.356, "grad_norm": 0.8517307639122009, "learning_rate": 0.0002, "epoch": 4.106015986537653, "step": 4880}, {"loss": 1.4191, "grad_norm": 0.8208953142166138, "learning_rate": 0.0002, "epoch": 4.114429953723181, "step": 4890}, {"loss": 1.3189, "grad_norm": 0.8437790870666504, "learning_rate": 0.0002, "epoch": 4.122843920908709, "step": 4900}, {"loss": 1.3987, "grad_norm": 0.716672420501709, "learning_rate": 0.0002, "epoch": 4.131257888094236, "step": 4910}, {"loss": 1.4392, "grad_norm": 0.7656235098838806, "learning_rate": 0.0002, "epoch": 4.139671855279764, "step": 4920}, {"loss": 1.3408, "grad_norm": 0.7209306955337524, "learning_rate": 0.0002, "epoch": 4.148085822465292, "step": 4930}, {"loss": 1.3639, "grad_norm": 0.7731267809867859, "learning_rate": 0.0002, "epoch": 4.1564997896508205, "step": 4940}, {"loss": 1.4151, "grad_norm": 0.7477553486824036, "learning_rate": 0.0002, "epoch": 4.164913756836349, "step": 4950}, {"loss": 1.3485, "grad_norm": 0.7372981309890747, "learning_rate": 0.0002, "epoch": 4.173327724021877, "step": 4960}, {"loss": 1.3901, "grad_norm": 0.6582154035568237, "learning_rate": 0.0002, "epoch": 4.181741691207404, "step": 4970}, {"loss": 1.3343, "grad_norm": 0.7003206610679626, "learning_rate": 0.0002, "epoch": 4.190155658392932, "step": 4980}, {"loss": 1.4098, "grad_norm": 0.735223650932312, "learning_rate": 0.0002, "epoch": 4.19856962557846, "step": 4990}, {"loss": 1.3564, "grad_norm": 0.7832302451133728, "learning_rate": 0.0002, "epoch": 4.206983592763988, "step": 5000}, {"loss": 1.3622, "grad_norm": 0.8819546103477478, "learning_rate": 0.0002, "epoch": 4.215397559949516, "step": 5010}, {"loss": 1.4438, "grad_norm": 0.9325336813926697, "learning_rate": 0.0002, "epoch": 4.2238115271350445, "step": 5020}, {"loss": 1.3886, "grad_norm": 0.7007517218589783, "learning_rate": 0.0002, "epoch": 4.232225494320572, "step": 5030}, {"loss": 1.3683, "grad_norm": 0.7118321061134338, "learning_rate": 0.0002, "epoch": 4.2406394615061, "step": 5040}, {"loss": 1.2365, "grad_norm": 0.6578946709632874, "learning_rate": 0.0002, "epoch": 4.249053428691628, "step": 5050}, {"loss": 1.3696, "grad_norm": 0.9438983798027039, "learning_rate": 0.0002, "epoch": 4.257467395877156, "step": 5060}, {"loss": 1.3868, "grad_norm": 0.703037679195404, "learning_rate": 0.0002, "epoch": 4.265881363062684, "step": 5070}, {"loss": 1.3687, "grad_norm": 0.7286025285720825, "learning_rate": 0.0002, "epoch": 4.274295330248212, "step": 5080}, {"loss": 1.3605, "grad_norm": 0.750689685344696, "learning_rate": 0.0002, "epoch": 4.28270929743374, "step": 5090}, {"loss": 1.5089, "grad_norm": 0.869753360748291, "learning_rate": 0.0002, "epoch": 4.291123264619268, "step": 5100}, {"loss": 1.4128, "grad_norm": 0.8712980151176453, "learning_rate": 0.0002, "epoch": 4.299537231804796, "step": 5110}, {"loss": 1.3977, "grad_norm": 0.690263569355011, "learning_rate": 0.0002, "epoch": 4.307951198990324, "step": 5120}, {"loss": 1.4088, "grad_norm": 0.7114760279655457, "learning_rate": 0.0002, "epoch": 4.316365166175852, "step": 5130}, {"loss": 1.363, "grad_norm": 0.7588112354278564, "learning_rate": 0.0002, "epoch": 4.32477913336138, "step": 5140}, {"loss": 1.4408, "grad_norm": 0.7556202411651611, "learning_rate": 0.0002, "epoch": 4.333193100546908, "step": 5150}, {"loss": 1.4203, "grad_norm": 0.8357610702514648, "learning_rate": 0.0002, "epoch": 4.341607067732435, "step": 5160}, {"loss": 1.3348, "grad_norm": 0.8054035902023315, "learning_rate": 0.0002, "epoch": 4.3500210349179635, "step": 5170}, {"loss": 1.3109, "grad_norm": 0.7637107968330383, "learning_rate": 0.0002, "epoch": 4.358435002103492, "step": 5180}, {"loss": 1.3744, "grad_norm": 0.757481038570404, "learning_rate": 0.0002, "epoch": 4.36684896928902, "step": 5190}, {"loss": 1.3622, "grad_norm": 0.7185863852500916, "learning_rate": 0.0002, "epoch": 4.375262936474548, "step": 5200}, {"loss": 1.3896, "grad_norm": 0.7326455116271973, "learning_rate": 0.0002, "epoch": 4.383676903660076, "step": 5210}, {"loss": 1.4098, "grad_norm": 0.7980523109436035, "learning_rate": 0.0002, "epoch": 4.392090870845603, "step": 5220}, {"loss": 1.3783, "grad_norm": 0.8526999354362488, "learning_rate": 0.0002, "epoch": 4.400504838031131, "step": 5230}, {"loss": 1.4022, "grad_norm": 0.7012337446212769, "learning_rate": 0.0002, "epoch": 4.4089188052166595, "step": 5240}, {"loss": 1.3552, "grad_norm": 0.8217827677726746, "learning_rate": 0.0002, "epoch": 4.417332772402188, "step": 5250}, {"loss": 1.3482, "grad_norm": 0.7141005396842957, "learning_rate": 0.0002, "epoch": 4.425746739587716, "step": 5260}, {"loss": 1.3699, "grad_norm": 0.7094302177429199, "learning_rate": 0.0002, "epoch": 4.434160706773244, "step": 5270}, {"loss": 1.3527, "grad_norm": 0.7234613299369812, "learning_rate": 0.0002, "epoch": 4.442574673958772, "step": 5280}, {"loss": 1.4769, "grad_norm": 0.7530457973480225, "learning_rate": 0.0002, "epoch": 4.450988641144299, "step": 5290}, {"loss": 1.3944, "grad_norm": 0.7300912141799927, "learning_rate": 0.0002, "epoch": 4.459402608329827, "step": 5300}, {"loss": 1.3844, "grad_norm": 0.825443685054779, "learning_rate": 0.0002, "epoch": 4.467816575515355, "step": 5310}, {"loss": 1.3648, "grad_norm": 0.7559658885002136, "learning_rate": 0.0002, "epoch": 4.4762305427008835, "step": 5320}, {"loss": 1.4364, "grad_norm": 0.8817561268806458, "learning_rate": 0.0002, "epoch": 4.484644509886412, "step": 5330}, {"loss": 1.3618, "grad_norm": 0.8203575611114502, "learning_rate": 0.0002, "epoch": 4.49305847707194, "step": 5340}, {"loss": 1.3996, "grad_norm": 0.7677690982818604, "learning_rate": 0.0002, "epoch": 4.501472444257468, "step": 5350}, {"loss": 1.4142, "grad_norm": 0.657085120677948, "learning_rate": 0.0002, "epoch": 4.509886411442995, "step": 5360}, {"loss": 1.3722, "grad_norm": 0.7939504384994507, "learning_rate": 0.0002, "epoch": 4.518300378628523, "step": 5370}, {"loss": 1.4361, "grad_norm": 0.6971889138221741, "learning_rate": 0.0002, "epoch": 4.526714345814051, "step": 5380}, {"loss": 1.3637, "grad_norm": 0.6984175443649292, "learning_rate": 0.0002, "epoch": 4.535128312999579, "step": 5390}, {"loss": 1.341, "grad_norm": 0.8504858613014221, "learning_rate": 0.0002, "epoch": 4.5435422801851075, "step": 5400}, {"loss": 1.4026, "grad_norm": 0.9134073853492737, "learning_rate": 0.0002, "epoch": 4.551956247370635, "step": 5410}, {"loss": 1.4375, "grad_norm": 0.7765598893165588, "learning_rate": 0.0002, "epoch": 4.560370214556163, "step": 5420}, {"loss": 1.4832, "grad_norm": 0.6991009712219238, "learning_rate": 0.0002, "epoch": 4.568784181741691, "step": 5430}, {"loss": 1.4021, "grad_norm": 0.8393039107322693, "learning_rate": 0.0002, "epoch": 4.577198148927219, "step": 5440}, {"loss": 1.3976, "grad_norm": 0.7685918211936951, "learning_rate": 0.0002, "epoch": 4.585612116112747, "step": 5450}, {"loss": 1.3883, "grad_norm": 0.7135679721832275, "learning_rate": 0.0002, "epoch": 4.594026083298275, "step": 5460}, {"loss": 1.4083, "grad_norm": 0.6728870868682861, "learning_rate": 0.0002, "epoch": 4.6024400504838034, "step": 5470}, {"loss": 1.3698, "grad_norm": 0.7139479517936707, "learning_rate": 0.0002, "epoch": 4.610854017669331, "step": 5480}, {"loss": 1.3498, "grad_norm": 0.8476598858833313, "learning_rate": 0.0002, "epoch": 4.619267984854859, "step": 5490}, {"loss": 1.3389, "grad_norm": 0.8034361004829407, "learning_rate": 0.0002, "epoch": 4.627681952040387, "step": 5500}, {"loss": 1.4179, "grad_norm": 0.7452183961868286, "learning_rate": 0.0002, "epoch": 4.636095919225915, "step": 5510}, {"loss": 1.4031, "grad_norm": 0.8394148945808411, "learning_rate": 0.0002, "epoch": 4.644509886411443, "step": 5520}, {"loss": 1.4561, "grad_norm": 0.7480153441429138, "learning_rate": 0.0002, "epoch": 4.652923853596971, "step": 5530}, {"loss": 1.378, "grad_norm": 0.7781714797019958, "learning_rate": 0.0002, "epoch": 4.661337820782499, "step": 5540}, {"loss": 1.3924, "grad_norm": 1.0058213472366333, "learning_rate": 0.0002, "epoch": 4.669751787968027, "step": 5550}, {"loss": 1.4198, "grad_norm": 0.7403179407119751, "learning_rate": 0.0002, "epoch": 4.678165755153555, "step": 5560}, {"loss": 1.4328, "grad_norm": 0.7270476818084717, "learning_rate": 0.0002, "epoch": 4.686579722339083, "step": 5570}, {"loss": 1.378, "grad_norm": 0.760877788066864, "learning_rate": 0.0002, "epoch": 4.694993689524611, "step": 5580}, {"loss": 1.387, "grad_norm": 0.8097004890441895, "learning_rate": 0.0002, "epoch": 4.703407656710139, "step": 5590}, {"loss": 1.3661, "grad_norm": 0.9096523523330688, "learning_rate": 0.0002, "epoch": 4.711821623895667, "step": 5600}, {"loss": 1.4012, "grad_norm": 0.7262444496154785, "learning_rate": 0.0002, "epoch": 4.720235591081195, "step": 5610}, {"loss": 1.422, "grad_norm": 0.8207762837409973, "learning_rate": 0.0002, "epoch": 4.7286495582667225, "step": 5620}, {"loss": 1.4017, "grad_norm": 0.8089601993560791, "learning_rate": 0.0002, "epoch": 4.737063525452251, "step": 5630}, {"loss": 1.3675, "grad_norm": 0.7609543800354004, "learning_rate": 0.0002, "epoch": 4.745477492637779, "step": 5640}, {"loss": 1.4085, "grad_norm": 0.7273501753807068, "learning_rate": 0.0002, "epoch": 4.753891459823307, "step": 5650}, {"loss": 1.3849, "grad_norm": 0.7800219058990479, "learning_rate": 0.0002, "epoch": 4.762305427008835, "step": 5660}, {"loss": 1.4319, "grad_norm": 0.8558377623558044, "learning_rate": 0.0002, "epoch": 4.770719394194362, "step": 5670}, {"loss": 1.3831, "grad_norm": 0.7131547927856445, "learning_rate": 0.0002, "epoch": 4.77913336137989, "step": 5680}, {"loss": 1.407, "grad_norm": 0.7651025056838989, "learning_rate": 0.0002, "epoch": 4.787547328565418, "step": 5690}, {"loss": 1.3882, "grad_norm": 0.8129976391792297, "learning_rate": 0.0002, "epoch": 4.7959612957509465, "step": 5700}, {"loss": 1.4347, "grad_norm": 0.8019895553588867, "learning_rate": 0.0002, "epoch": 4.804375262936475, "step": 5710}, {"loss": 1.3961, "grad_norm": 0.7692018151283264, "learning_rate": 0.0002, "epoch": 4.812789230122003, "step": 5720}, {"loss": 1.419, "grad_norm": 0.6893943548202515, "learning_rate": 0.0002, "epoch": 4.821203197307531, "step": 5730}, {"loss": 1.4453, "grad_norm": 0.6881810426712036, "learning_rate": 0.0002, "epoch": 4.829617164493058, "step": 5740}, {"loss": 1.4775, "grad_norm": 0.7838267683982849, "learning_rate": 0.0002, "epoch": 4.838031131678586, "step": 5750}, {"loss": 1.3857, "grad_norm": 0.727799117565155, "learning_rate": 0.0002, "epoch": 4.846445098864114, "step": 5760}, {"loss": 1.4685, "grad_norm": 0.7458277344703674, "learning_rate": 0.0002, "epoch": 4.8548590660496425, "step": 5770}, {"loss": 1.4426, "grad_norm": 0.903802216053009, "learning_rate": 0.0002, "epoch": 4.863273033235171, "step": 5780}, {"loss": 1.451, "grad_norm": 0.7983472347259521, "learning_rate": 0.0002, "epoch": 4.871687000420699, "step": 5790}, {"loss": 1.4534, "grad_norm": 0.6894361972808838, "learning_rate": 0.0002, "epoch": 4.880100967606227, "step": 5800}, {"loss": 1.4486, "grad_norm": 0.7499409317970276, "learning_rate": 0.0002, "epoch": 4.888514934791754, "step": 5810}, {"loss": 1.4253, "grad_norm": 0.7362820506095886, "learning_rate": 0.0002, "epoch": 4.896928901977282, "step": 5820}, {"loss": 1.3763, "grad_norm": 0.8341619968414307, "learning_rate": 0.0002, "epoch": 4.90534286916281, "step": 5830}, {"loss": 1.3748, "grad_norm": 0.9604470133781433, "learning_rate": 0.0002, "epoch": 4.913756836348338, "step": 5840}, {"loss": 1.3658, "grad_norm": 0.8916844129562378, "learning_rate": 0.0002, "epoch": 4.9221708035338665, "step": 5850}, {"loss": 1.363, "grad_norm": 0.8519647121429443, "learning_rate": 0.0002, "epoch": 4.930584770719394, "step": 5860}, {"loss": 1.424, "grad_norm": 0.7946906089782715, "learning_rate": 0.0002, "epoch": 4.938998737904922, "step": 5870}, {"loss": 1.4071, "grad_norm": 0.7843789458274841, "learning_rate": 0.0002, "epoch": 4.94741270509045, "step": 5880}, {"loss": 1.4021, "grad_norm": 0.707618772983551, "learning_rate": 0.0002, "epoch": 4.955826672275978, "step": 5890}, {"loss": 1.502, "grad_norm": 0.7704206109046936, "learning_rate": 0.0002, "epoch": 4.964240639461506, "step": 5900}, {"loss": 1.4456, "grad_norm": 0.7160256505012512, "learning_rate": 0.0002, "epoch": 4.972654606647034, "step": 5910}, {"loss": 1.3874, "grad_norm": 0.7020420432090759, "learning_rate": 0.0002, "epoch": 4.981068573832562, "step": 5920}, {"loss": 1.4037, "grad_norm": 0.7576286792755127, "learning_rate": 0.0002, "epoch": 4.98948254101809, "step": 5930}, {"loss": 1.414, "grad_norm": 0.8573036789894104, "learning_rate": 0.0002, "epoch": 4.997896508203618, "step": 5940}, {"eval_loss": 1.9353811740875244, "eval_runtime": 37.9208, "eval_samples_per_second": 13.581, "eval_steps_per_second": 1.714, "epoch": 4.999579301640724, "step": 5942}, {"loss": 1.2418, "grad_norm": 0.8204267621040344, "learning_rate": 0.0002, "epoch": 5.006310475389146, "step": 5950}, {"loss": 1.235, "grad_norm": 0.976840615272522, "learning_rate": 0.0002, "epoch": 5.014724442574674, "step": 5960}, {"loss": 1.2134, "grad_norm": 0.8765613436698914, "learning_rate": 0.0002, "epoch": 5.023138409760202, "step": 5970}, {"loss": 1.2748, "grad_norm": 1.1793042421340942, "learning_rate": 0.0002, "epoch": 5.03155237694573, "step": 5980}, {"loss": 1.2412, "grad_norm": 0.971062958240509, "learning_rate": 0.0002, "epoch": 5.039966344131258, "step": 5990}, {"loss": 1.1819, "grad_norm": 0.8649757504463196, "learning_rate": 0.0002, "epoch": 5.0483803113167856, "step": 6000}, {"loss": 1.1654, "grad_norm": 0.9563034176826477, "learning_rate": 0.0002, "epoch": 5.056794278502314, "step": 6010}, {"loss": 1.2238, "grad_norm": 1.0093994140625, "learning_rate": 0.0002, "epoch": 5.065208245687842, "step": 6020}, {"loss": 1.2519, "grad_norm": 1.004213571548462, "learning_rate": 0.0002, "epoch": 5.07362221287337, "step": 6030}, {"loss": 1.2379, "grad_norm": 0.8307787179946899, "learning_rate": 0.0002, "epoch": 5.082036180058898, "step": 6040}, {"loss": 1.2282, "grad_norm": 0.9117848873138428, "learning_rate": 0.0002, "epoch": 5.090450147244426, "step": 6050}, {"loss": 1.2582, "grad_norm": 1.0269840955734253, "learning_rate": 0.0002, "epoch": 5.098864114429953, "step": 6060}, {"loss": 1.1836, "grad_norm": 0.9079542756080627, "learning_rate": 0.0002, "epoch": 5.1072780816154815, "step": 6070}, {"loss": 1.215, "grad_norm": 0.885702908039093, "learning_rate": 0.0002, "epoch": 5.11569204880101, "step": 6080}, {"loss": 1.2406, "grad_norm": 0.9976128339767456, "learning_rate": 0.0002, "epoch": 5.124106015986538, "step": 6090}, {"loss": 1.3082, "grad_norm": 0.8472117185592651, "learning_rate": 0.0002, "epoch": 5.132519983172066, "step": 6100}, {"loss": 1.226, "grad_norm": 1.0385161638259888, "learning_rate": 0.0002, "epoch": 5.140933950357594, "step": 6110}, {"loss": 1.213, "grad_norm": 0.8948383927345276, "learning_rate": 0.0002, "epoch": 5.149347917543121, "step": 6120}, {"loss": 1.2213, "grad_norm": 1.2613716125488281, "learning_rate": 0.0002, "epoch": 5.157761884728649, "step": 6130}, {"loss": 1.2632, "grad_norm": 0.9933410286903381, "learning_rate": 0.0002, "epoch": 5.166175851914177, "step": 6140}, {"loss": 1.1715, "grad_norm": 0.9673663973808289, "learning_rate": 0.0002, "epoch": 5.1745898190997055, "step": 6150}, {"loss": 1.2947, "grad_norm": 0.9969648122787476, "learning_rate": 0.0002, "epoch": 5.183003786285234, "step": 6160}, {"loss": 1.2416, "grad_norm": 1.2163258790969849, "learning_rate": 0.0002, "epoch": 5.191417753470762, "step": 6170}, {"loss": 1.2221, "grad_norm": 0.9163419604301453, "learning_rate": 0.0002, "epoch": 5.19983172065629, "step": 6180}, {"loss": 1.2624, "grad_norm": 0.9225585460662842, "learning_rate": 0.0002, "epoch": 5.208245687841817, "step": 6190}, {"loss": 1.2932, "grad_norm": 0.9205296635627747, "learning_rate": 0.0002, "epoch": 5.216659655027345, "step": 6200}, {"loss": 1.1825, "grad_norm": 1.0655443668365479, "learning_rate": 0.0002, "epoch": 5.225073622212873, "step": 6210}, {"loss": 1.2613, "grad_norm": 1.0854865312576294, "learning_rate": 0.0002, "epoch": 5.233487589398401, "step": 6220}, {"loss": 1.3045, "grad_norm": 0.8489186763763428, "learning_rate": 0.0002, "epoch": 5.2419015565839295, "step": 6230}, {"loss": 1.2708, "grad_norm": 0.910391628742218, "learning_rate": 0.0002, "epoch": 5.250315523769458, "step": 6240}, {"loss": 1.1914, "grad_norm": 0.925507128238678, "learning_rate": 0.0002, "epoch": 5.258729490954985, "step": 6250}, {"loss": 1.3368, "grad_norm": 1.1069735288619995, "learning_rate": 0.0002, "epoch": 5.267143458140513, "step": 6260}, {"loss": 1.2505, "grad_norm": 0.9705119132995605, "learning_rate": 0.0002, "epoch": 5.275557425326041, "step": 6270}, {"loss": 1.2602, "grad_norm": 0.9752426147460938, "learning_rate": 0.0002, "epoch": 5.283971392511569, "step": 6280}, {"loss": 1.2043, "grad_norm": 1.021359920501709, "learning_rate": 0.0002, "epoch": 5.292385359697097, "step": 6290}, {"loss": 1.2848, "grad_norm": 1.148606300354004, "learning_rate": 0.0002, "epoch": 5.3007993268826255, "step": 6300}, {"loss": 1.2201, "grad_norm": 0.8909247517585754, "learning_rate": 0.0002, "epoch": 5.309213294068153, "step": 6310}, {"loss": 1.2376, "grad_norm": 0.9879156351089478, "learning_rate": 0.0002, "epoch": 5.317627261253681, "step": 6320}, {"loss": 1.2638, "grad_norm": 0.9473357200622559, "learning_rate": 0.0002, "epoch": 5.326041228439209, "step": 6330}, {"loss": 1.232, "grad_norm": 1.1422028541564941, "learning_rate": 0.0002, "epoch": 5.334455195624737, "step": 6340}, {"loss": 1.263, "grad_norm": 0.9942235350608826, "learning_rate": 0.0002, "epoch": 5.342869162810265, "step": 6350}, {"loss": 1.3032, "grad_norm": 0.9535723924636841, "learning_rate": 0.0002, "epoch": 5.351283129995793, "step": 6360}, {"loss": 1.2908, "grad_norm": 0.9020892381668091, "learning_rate": 0.0002, "epoch": 5.359697097181321, "step": 6370}, {"loss": 1.2023, "grad_norm": 1.0626472234725952, "learning_rate": 0.0002, "epoch": 5.368111064366849, "step": 6380}, {"loss": 1.2555, "grad_norm": 1.1395848989486694, "learning_rate": 0.0002, "epoch": 5.376525031552377, "step": 6390}, {"loss": 1.2839, "grad_norm": 0.9274451732635498, "learning_rate": 0.0002, "epoch": 5.384938998737905, "step": 6400}, {"loss": 1.2819, "grad_norm": 0.8108699917793274, "learning_rate": 0.0002, "epoch": 5.393352965923433, "step": 6410}, {"loss": 1.2589, "grad_norm": 1.1805564165115356, "learning_rate": 0.0002, "epoch": 5.401766933108961, "step": 6420}, {"loss": 1.3549, "grad_norm": 0.8321298360824585, "learning_rate": 0.0002, "epoch": 5.410180900294489, "step": 6430}, {"loss": 1.2925, "grad_norm": 0.8981925249099731, "learning_rate": 0.0002, "epoch": 5.418594867480017, "step": 6440}, {"loss": 1.258, "grad_norm": 1.0730986595153809, "learning_rate": 0.0002, "epoch": 5.4270088346655445, "step": 6450}, {"loss": 1.26, "grad_norm": 1.0584609508514404, "learning_rate": 0.0002, "epoch": 5.435422801851073, "step": 6460}, {"loss": 1.2847, "grad_norm": 1.0792299509048462, "learning_rate": 0.0002, "epoch": 5.443836769036601, "step": 6470}, {"loss": 1.2035, "grad_norm": 0.9101872444152832, "learning_rate": 0.0002, "epoch": 5.452250736222129, "step": 6480}, {"loss": 1.2574, "grad_norm": 0.9910100698471069, "learning_rate": 0.0002, "epoch": 5.460664703407657, "step": 6490}, {"loss": 1.3098, "grad_norm": 1.041412353515625, "learning_rate": 0.0002, "epoch": 5.469078670593185, "step": 6500}, {"loss": 1.2812, "grad_norm": 1.0091687440872192, "learning_rate": 0.0002, "epoch": 5.477492637778712, "step": 6510}, {"loss": 1.2523, "grad_norm": 0.8755383491516113, "learning_rate": 0.0002, "epoch": 5.48590660496424, "step": 6520}, {"loss": 1.3042, "grad_norm": 0.980212390422821, "learning_rate": 0.0002, "epoch": 5.4943205721497685, "step": 6530}, {"loss": 1.2873, "grad_norm": 0.9356869459152222, "learning_rate": 0.0002, "epoch": 5.502734539335297, "step": 6540}, {"loss": 1.2254, "grad_norm": 0.9008095264434814, "learning_rate": 0.0002, "epoch": 5.511148506520825, "step": 6550}, {"loss": 1.2818, "grad_norm": 0.8908938765525818, "learning_rate": 0.0002, "epoch": 5.519562473706353, "step": 6560}, {"loss": 1.2212, "grad_norm": 1.1423932313919067, "learning_rate": 0.0002, "epoch": 5.52797644089188, "step": 6570}, {"loss": 1.3039, "grad_norm": 1.0508161783218384, "learning_rate": 0.0002, "epoch": 5.536390408077408, "step": 6580}, {"loss": 1.2446, "grad_norm": 0.8357517719268799, "learning_rate": 0.0002, "epoch": 5.544804375262936, "step": 6590}, {"loss": 1.3037, "grad_norm": 0.9892540574073792, "learning_rate": 0.0002, "epoch": 5.5532183424484645, "step": 6600}, {"loss": 1.3028, "grad_norm": 1.0048326253890991, "learning_rate": 0.0002, "epoch": 5.561632309633993, "step": 6610}, {"loss": 1.2152, "grad_norm": 0.9801995158195496, "learning_rate": 0.0002, "epoch": 5.570046276819521, "step": 6620}, {"loss": 1.2606, "grad_norm": 0.9899214506149292, "learning_rate": 0.0002, "epoch": 5.578460244005049, "step": 6630}, {"loss": 1.2043, "grad_norm": 1.1911814212799072, "learning_rate": 0.0002, "epoch": 5.586874211190576, "step": 6640}, {"loss": 1.3458, "grad_norm": 1.0368894338607788, "learning_rate": 0.0002, "epoch": 5.595288178376104, "step": 6650}, {"loss": 1.2595, "grad_norm": 1.1248382329940796, "learning_rate": 0.0002, "epoch": 5.603702145561632, "step": 6660}, {"loss": 1.2548, "grad_norm": 0.9765539765357971, "learning_rate": 0.0002, "epoch": 5.61211611274716, "step": 6670}, {"loss": 1.3451, "grad_norm": 0.9810206890106201, "learning_rate": 0.0002, "epoch": 5.6205300799326885, "step": 6680}, {"loss": 1.2952, "grad_norm": 1.100386619567871, "learning_rate": 0.0002, "epoch": 5.628944047118217, "step": 6690}, {"loss": 1.2467, "grad_norm": 0.8824519515037537, "learning_rate": 0.0002, "epoch": 5.637358014303744, "step": 6700}, {"loss": 1.25, "grad_norm": 1.0864064693450928, "learning_rate": 0.0002, "epoch": 5.645771981489272, "step": 6710}, {"loss": 1.2479, "grad_norm": 1.1614511013031006, "learning_rate": 0.0002, "epoch": 5.6541859486748, "step": 6720}, {"loss": 1.2753, "grad_norm": 1.0762972831726074, "learning_rate": 0.0002, "epoch": 5.662599915860328, "step": 6730}, {"loss": 1.2741, "grad_norm": 0.9408974647521973, "learning_rate": 0.0002, "epoch": 5.671013883045856, "step": 6740}, {"loss": 1.2431, "grad_norm": 0.8906030058860779, "learning_rate": 0.0002, "epoch": 5.679427850231384, "step": 6750}, {"loss": 1.2643, "grad_norm": 0.9527303576469421, "learning_rate": 0.0002, "epoch": 5.687841817416912, "step": 6760}, {"loss": 1.322, "grad_norm": 0.9471196532249451, "learning_rate": 0.0002, "epoch": 5.69625578460244, "step": 6770}, {"loss": 1.2514, "grad_norm": 0.9186838865280151, "learning_rate": 0.0002, "epoch": 5.704669751787968, "step": 6780}, {"loss": 1.2347, "grad_norm": 0.9225441813468933, "learning_rate": 0.0002, "epoch": 5.713083718973496, "step": 6790}, {"loss": 1.1849, "grad_norm": 0.9712982773780823, "learning_rate": 0.0002, "epoch": 5.721497686159024, "step": 6800}, {"loss": 1.2431, "grad_norm": 1.0743170976638794, "learning_rate": 0.0002, "epoch": 5.729911653344552, "step": 6810}, {"loss": 1.2136, "grad_norm": 1.2738113403320312, "learning_rate": 0.0002, "epoch": 5.73832562053008, "step": 6820}, {"loss": 1.2176, "grad_norm": 0.9386790990829468, "learning_rate": 0.0002, "epoch": 5.7467395877156076, "step": 6830}, {"loss": 1.285, "grad_norm": 1.0817769765853882, "learning_rate": 0.0002, "epoch": 5.755153554901136, "step": 6840}, {"loss": 1.2247, "grad_norm": 1.1040263175964355, "learning_rate": 0.0002, "epoch": 5.763567522086664, "step": 6850}, {"loss": 1.2507, "grad_norm": 1.0656492710113525, "learning_rate": 0.0002, "epoch": 5.771981489272192, "step": 6860}, {"loss": 1.2999, "grad_norm": 0.9550157189369202, "learning_rate": 0.0002, "epoch": 5.78039545645772, "step": 6870}, {"loss": 1.3201, "grad_norm": 1.0130870342254639, "learning_rate": 0.0002, "epoch": 5.788809423643248, "step": 6880}, {"loss": 1.3392, "grad_norm": 1.0675787925720215, "learning_rate": 0.0002, "epoch": 5.797223390828776, "step": 6890}, {"loss": 1.2949, "grad_norm": 0.9537774920463562, "learning_rate": 0.0002, "epoch": 5.8056373580143035, "step": 6900}, {"loss": 1.2658, "grad_norm": 0.9640319347381592, "learning_rate": 0.0002, "epoch": 5.814051325199832, "step": 6910}, {"loss": 1.2199, "grad_norm": 0.8917992115020752, "learning_rate": 0.0002, "epoch": 5.82246529238536, "step": 6920}, {"loss": 1.373, "grad_norm": 0.9881822466850281, "learning_rate": 0.0002, "epoch": 5.830879259570888, "step": 6930}, {"loss": 1.323, "grad_norm": 0.9136882424354553, "learning_rate": 0.0002, "epoch": 5.839293226756416, "step": 6940}, {"loss": 1.3159, "grad_norm": 0.9086098074913025, "learning_rate": 0.0002, "epoch": 5.847707193941943, "step": 6950}, {"loss": 1.2624, "grad_norm": 0.9443018436431885, "learning_rate": 0.0002, "epoch": 5.856121161127471, "step": 6960}, {"loss": 1.3224, "grad_norm": 0.9915381669998169, "learning_rate": 0.0002, "epoch": 5.864535128312999, "step": 6970}, {"loss": 1.337, "grad_norm": 0.8939146995544434, "learning_rate": 0.0002, "epoch": 5.8729490954985275, "step": 6980}, {"loss": 1.2611, "grad_norm": 1.3672245740890503, "learning_rate": 0.0002, "epoch": 5.881363062684056, "step": 6990}, {"loss": 1.3012, "grad_norm": 1.0116257667541504, "learning_rate": 0.0002, "epoch": 5.889777029869584, "step": 7000}, {"loss": 1.3128, "grad_norm": 1.1561565399169922, "learning_rate": 0.0002, "epoch": 5.898190997055112, "step": 7010}, {"loss": 1.2301, "grad_norm": 0.9900678992271423, "learning_rate": 0.0002, "epoch": 5.906604964240639, "step": 7020}, {"loss": 1.2845, "grad_norm": 0.9297345876693726, "learning_rate": 0.0002, "epoch": 5.915018931426167, "step": 7030}, {"loss": 1.2317, "grad_norm": 0.9357825517654419, "learning_rate": 0.0002, "epoch": 5.923432898611695, "step": 7040}, {"loss": 1.2303, "grad_norm": 1.049317717552185, "learning_rate": 0.0002, "epoch": 5.931846865797223, "step": 7050}, {"loss": 1.3243, "grad_norm": 0.950633704662323, "learning_rate": 0.0002, "epoch": 5.9402608329827515, "step": 7060}, {"loss": 1.2758, "grad_norm": 0.854581892490387, "learning_rate": 0.0002, "epoch": 5.94867480016828, "step": 7070}, {"loss": 1.3252, "grad_norm": 0.9097039699554443, "learning_rate": 0.0002, "epoch": 5.957088767353808, "step": 7080}, {"loss": 1.291, "grad_norm": 0.9072173237800598, "learning_rate": 0.0002, "epoch": 5.965502734539335, "step": 7090}, {"loss": 1.2724, "grad_norm": 1.0470727682113647, "learning_rate": 0.0002, "epoch": 5.973916701724863, "step": 7100}, {"loss": 1.3324, "grad_norm": 1.2628462314605713, "learning_rate": 0.0002, "epoch": 5.982330668910391, "step": 7110}, {"loss": 1.2701, "grad_norm": 1.055279016494751, "learning_rate": 0.0002, "epoch": 5.990744636095919, "step": 7120}, {"loss": 1.3234, "grad_norm": 0.966194212436676, "learning_rate": 0.0002, "epoch": 5.9991586032814475, "step": 7130}]}
+{"epoch": 6.999579301640724, "step": 8319, "epoch_duration": 1250.1847441196442, "total_accumulated_duration": 9815.15891122818, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "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_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-0.8-num-5623-sd-42/checkpoint-2377", "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": 2.56, "grad_norm": 0.5458821654319763, "learning_rate": 0.0002, "epoch": 0.008413967185527976, "step": 10}, {"loss": 2.3235, "grad_norm": 0.7293308973312378, "learning_rate": 0.0002, "epoch": 0.016827934371055953, "step": 20}, {"loss": 2.0815, "grad_norm": 0.47792306542396545, "learning_rate": 0.0002, "epoch": 0.02524190155658393, "step": 30}, {"loss": 1.9718, "grad_norm": 0.5944402813911438, "learning_rate": 0.0002, "epoch": 0.033655868742111905, "step": 40}, {"loss": 1.8848, "grad_norm": 0.5415359735488892, "learning_rate": 0.0002, "epoch": 0.04206983592763988, "step": 50}, {"loss": 1.8953, "grad_norm": 0.535713791847229, "learning_rate": 0.0002, "epoch": 0.05048380311316786, "step": 60}, {"loss": 1.937, "grad_norm": 0.5184146761894226, "learning_rate": 0.0002, "epoch": 0.058897770298695834, "step": 70}, {"loss": 1.8396, "grad_norm": 0.458926796913147, "learning_rate": 0.0002, "epoch": 0.06731173748422381, "step": 80}, {"loss": 1.8677, "grad_norm": 0.4780142307281494, "learning_rate": 0.0002, "epoch": 0.07572570466975179, "step": 90}, {"loss": 1.8593, "grad_norm": 0.79965740442276, "learning_rate": 0.0002, "epoch": 0.08413967185527976, "step": 100}, {"loss": 1.9081, "grad_norm": 0.4498862028121948, "learning_rate": 0.0002, "epoch": 0.09255363904080774, "step": 110}, {"loss": 1.8503, "grad_norm": 0.39338430762290955, "learning_rate": 0.0002, "epoch": 0.10096760622633572, "step": 120}, {"loss": 1.8637, "grad_norm": 0.9588953852653503, "learning_rate": 0.0002, "epoch": 0.10938157341186369, "step": 130}, {"loss": 1.8676, "grad_norm": 0.41675639152526855, "learning_rate": 0.0002, "epoch": 0.11779554059739167, "step": 140}, {"loss": 1.8904, "grad_norm": 0.44519832730293274, "learning_rate": 0.0002, "epoch": 0.12620950778291964, "step": 150}, {"loss": 1.798, "grad_norm": 0.4176260530948639, "learning_rate": 0.0002, "epoch": 0.13462347496844762, "step": 160}, {"loss": 1.8398, "grad_norm": 0.35840365290641785, "learning_rate": 0.0002, "epoch": 0.1430374421539756, "step": 170}, {"loss": 1.8666, "grad_norm": 0.3794495463371277, "learning_rate": 0.0002, "epoch": 0.15145140933950357, "step": 180}, {"loss": 1.8111, "grad_norm": 0.4563522934913635, "learning_rate": 0.0002, "epoch": 0.15986537652503155, "step": 190}, {"loss": 1.8893, "grad_norm": 0.37057486176490784, "learning_rate": 0.0002, "epoch": 0.16827934371055953, "step": 200}, {"loss": 1.7995, "grad_norm": 0.44081518054008484, "learning_rate": 0.0002, "epoch": 0.1766933108960875, "step": 210}, {"loss": 1.9048, "grad_norm": 0.46078577637672424, "learning_rate": 0.0002, "epoch": 0.18510727808161548, "step": 220}, {"loss": 1.8403, "grad_norm": 0.36132094264030457, "learning_rate": 0.0002, "epoch": 0.19352124526714345, "step": 230}, {"loss": 1.8827, "grad_norm": 0.3747289180755615, "learning_rate": 0.0002, "epoch": 0.20193521245267143, "step": 240}, {"loss": 1.8382, "grad_norm": 0.3540179133415222, "learning_rate": 0.0002, "epoch": 0.2103491796381994, "step": 250}, {"loss": 1.8196, "grad_norm": 0.3461375832557678, "learning_rate": 0.0002, "epoch": 0.21876314682372738, "step": 260}, {"loss": 1.8509, "grad_norm": 0.3436960279941559, "learning_rate": 0.0002, "epoch": 0.22717711400925536, "step": 270}, {"loss": 1.8285, "grad_norm": 0.35403719544410706, "learning_rate": 0.0002, "epoch": 0.23559108119478334, "step": 280}, {"loss": 1.8369, "grad_norm": 0.37142616510391235, "learning_rate": 0.0002, "epoch": 0.2440050483803113, "step": 290}, {"loss": 1.8044, "grad_norm": 0.3307955861091614, "learning_rate": 0.0002, "epoch": 0.2524190155658393, "step": 300}, {"loss": 1.817, "grad_norm": 0.32855314016342163, "learning_rate": 0.0002, "epoch": 0.2608329827513673, "step": 310}, {"loss": 1.7803, "grad_norm": 0.3299003839492798, "learning_rate": 0.0002, "epoch": 0.26924694993689524, "step": 320}, {"loss": 1.8129, "grad_norm": 0.44311287999153137, "learning_rate": 0.0002, "epoch": 0.27766091712242325, "step": 330}, {"loss": 1.8232, "grad_norm": 0.32989758253097534, "learning_rate": 0.0002, "epoch": 0.2860748843079512, "step": 340}, {"loss": 1.7716, "grad_norm": 0.34400200843811035, "learning_rate": 0.0002, "epoch": 0.2944888514934792, "step": 350}, {"loss": 1.7619, "grad_norm": 0.36286211013793945, "learning_rate": 0.0002, "epoch": 0.30290281867900715, "step": 360}, {"loss": 1.8025, "grad_norm": 0.406827837228775, "learning_rate": 0.0002, "epoch": 0.31131678586453515, "step": 370}, {"loss": 1.7515, "grad_norm": 0.36299195885658264, "learning_rate": 0.0002, "epoch": 0.3197307530500631, "step": 380}, {"loss": 1.837, "grad_norm": 0.3477257192134857, "learning_rate": 0.0002, "epoch": 0.3281447202355911, "step": 390}, {"loss": 1.7767, "grad_norm": 0.3730369210243225, "learning_rate": 0.0002, "epoch": 0.33655868742111905, "step": 400}, {"loss": 1.7747, "grad_norm": 0.4644559919834137, "learning_rate": 0.0002, "epoch": 0.34497265460664706, "step": 410}, {"loss": 1.7538, "grad_norm": 0.406576544046402, "learning_rate": 0.0002, "epoch": 0.353386621792175, "step": 420}, {"loss": 1.7501, "grad_norm": 0.3612699508666992, "learning_rate": 0.0002, "epoch": 0.361800588977703, "step": 430}, {"loss": 1.7473, "grad_norm": 0.3243742287158966, "learning_rate": 0.0002, "epoch": 0.37021455616323096, "step": 440}, {"loss": 1.8851, "grad_norm": 0.36671221256256104, "learning_rate": 0.0002, "epoch": 0.37862852334875896, "step": 450}, {"loss": 1.8853, "grad_norm": 0.3565002381801605, "learning_rate": 0.0002, "epoch": 0.3870424905342869, "step": 460}, {"loss": 1.8923, "grad_norm": 0.34630221128463745, "learning_rate": 0.0002, "epoch": 0.3954564577198149, "step": 470}, {"loss": 1.8234, "grad_norm": 0.3353537321090698, "learning_rate": 0.0002, "epoch": 0.40387042490534286, "step": 480}, {"loss": 1.7135, "grad_norm": 0.4015921950340271, "learning_rate": 0.0002, "epoch": 0.41228439209087087, "step": 490}, {"loss": 1.7815, "grad_norm": 0.5489419102668762, "learning_rate": 0.0002, "epoch": 0.4206983592763988, "step": 500}, {"loss": 1.7903, "grad_norm": 0.4193589985370636, "learning_rate": 0.0002, "epoch": 0.4291123264619268, "step": 510}, {"loss": 1.8416, "grad_norm": 0.3418922424316406, "learning_rate": 0.0002, "epoch": 0.43752629364745477, "step": 520}, {"loss": 1.7982, "grad_norm": 0.32668185234069824, "learning_rate": 0.0002, "epoch": 0.44594026083298277, "step": 530}, {"loss": 1.7501, "grad_norm": 0.3094325661659241, "learning_rate": 0.0002, "epoch": 0.4543542280185107, "step": 540}, {"loss": 1.7438, "grad_norm": 0.3743017315864563, "learning_rate": 0.0002, "epoch": 0.4627681952040387, "step": 550}, {"loss": 1.8451, "grad_norm": 0.3295630216598511, "learning_rate": 0.0002, "epoch": 0.47118216238956667, "step": 560}, {"loss": 1.7529, "grad_norm": 1.6124513149261475, "learning_rate": 0.0002, "epoch": 0.4795961295750947, "step": 570}, {"loss": 1.8028, "grad_norm": 0.3245585858821869, "learning_rate": 0.0002, "epoch": 0.4880100967606226, "step": 580}, {"loss": 1.7976, "grad_norm": 0.3332934081554413, "learning_rate": 0.0002, "epoch": 0.49642406394615063, "step": 590}, {"loss": 1.7912, "grad_norm": 0.3836138844490051, "learning_rate": 0.0002, "epoch": 0.5048380311316786, "step": 600}, {"loss": 1.8347, "grad_norm": 0.32953888177871704, "learning_rate": 0.0002, "epoch": 0.5132519983172066, "step": 610}, {"loss": 1.7729, "grad_norm": 0.36291512846946716, "learning_rate": 0.0002, "epoch": 0.5216659655027346, "step": 620}, {"loss": 1.7758, "grad_norm": 0.3237783908843994, "learning_rate": 0.0002, "epoch": 0.5300799326882625, "step": 630}, {"loss": 1.8352, "grad_norm": 0.38882696628570557, "learning_rate": 0.0002, "epoch": 0.5384938998737905, "step": 640}, {"loss": 1.8624, "grad_norm": 0.37821972370147705, "learning_rate": 0.0002, "epoch": 0.5469078670593185, "step": 650}, {"loss": 1.8075, "grad_norm": 0.3556285500526428, "learning_rate": 0.0002, "epoch": 0.5553218342448465, "step": 660}, {"loss": 1.778, "grad_norm": 0.347499281167984, "learning_rate": 0.0002, "epoch": 0.5637358014303744, "step": 670}, {"loss": 1.8066, "grad_norm": 0.3176489472389221, "learning_rate": 0.0002, "epoch": 0.5721497686159024, "step": 680}, {"loss": 1.7257, "grad_norm": 0.30220088362693787, "learning_rate": 0.0002, "epoch": 0.5805637358014304, "step": 690}, {"loss": 1.8415, "grad_norm": 0.3711601793766022, "learning_rate": 0.0002, "epoch": 0.5889777029869584, "step": 700}, {"loss": 1.7906, "grad_norm": 0.3311759829521179, "learning_rate": 0.0002, "epoch": 0.5973916701724863, "step": 710}, {"loss": 1.7712, "grad_norm": 0.34824270009994507, "learning_rate": 0.0002, "epoch": 0.6058056373580143, "step": 720}, {"loss": 1.7954, "grad_norm": 0.29668381810188293, "learning_rate": 0.0002, "epoch": 0.6142196045435423, "step": 730}, {"loss": 1.8321, "grad_norm": 0.36087489128112793, "learning_rate": 0.0002, "epoch": 0.6226335717290703, "step": 740}, {"loss": 1.7956, "grad_norm": 0.31590089201927185, "learning_rate": 0.0002, "epoch": 0.6310475389145982, "step": 750}, {"loss": 1.7343, "grad_norm": 0.37632957100868225, "learning_rate": 0.0002, "epoch": 0.6394615061001262, "step": 760}, {"loss": 1.8499, "grad_norm": 0.3360748589038849, "learning_rate": 0.0002, "epoch": 0.6478754732856542, "step": 770}, {"loss": 1.8076, "grad_norm": 0.3420640528202057, "learning_rate": 0.0002, "epoch": 0.6562894404711822, "step": 780}, {"loss": 1.8353, "grad_norm": 0.5734959244728088, "learning_rate": 0.0002, "epoch": 0.6647034076567101, "step": 790}, {"loss": 1.7746, "grad_norm": 0.36440837383270264, "learning_rate": 0.0002, "epoch": 0.6731173748422381, "step": 800}, {"loss": 1.7532, "grad_norm": 0.3179708421230316, "learning_rate": 0.0002, "epoch": 0.6815313420277661, "step": 810}, {"loss": 1.7815, "grad_norm": 0.34122881293296814, "learning_rate": 0.0002, "epoch": 0.6899453092132941, "step": 820}, {"loss": 1.8167, "grad_norm": 0.31886112689971924, "learning_rate": 0.0002, "epoch": 0.698359276398822, "step": 830}, {"loss": 1.7505, "grad_norm": 0.31782326102256775, "learning_rate": 0.0002, "epoch": 0.70677324358435, "step": 840}, {"loss": 1.7588, "grad_norm": 0.36052989959716797, "learning_rate": 0.0002, "epoch": 0.715187210769878, "step": 850}, {"loss": 1.7891, "grad_norm": 0.28946155309677124, "learning_rate": 0.0002, "epoch": 0.723601177955406, "step": 860}, {"loss": 1.7923, "grad_norm": 0.3095663785934448, "learning_rate": 0.0002, "epoch": 0.7320151451409339, "step": 870}, {"loss": 1.785, "grad_norm": 0.3317491412162781, "learning_rate": 0.0002, "epoch": 0.7404291123264619, "step": 880}, {"loss": 1.7709, "grad_norm": 0.31324660778045654, "learning_rate": 0.0002, "epoch": 0.7488430795119899, "step": 890}, {"loss": 1.8753, "grad_norm": 0.3290475606918335, "learning_rate": 0.0002, "epoch": 0.7572570466975179, "step": 900}, {"loss": 1.7679, "grad_norm": 0.35690343379974365, "learning_rate": 0.0002, "epoch": 0.7656710138830458, "step": 910}, {"loss": 1.826, "grad_norm": 0.39558273553848267, "learning_rate": 0.0002, "epoch": 0.7740849810685738, "step": 920}, {"loss": 1.8722, "grad_norm": 0.34254348278045654, "learning_rate": 0.0002, "epoch": 0.7824989482541018, "step": 930}, {"loss": 1.7603, "grad_norm": 0.3560165464878082, "learning_rate": 0.0002, "epoch": 0.7909129154396298, "step": 940}, {"loss": 1.7992, "grad_norm": 0.30693164467811584, "learning_rate": 0.0002, "epoch": 0.7993268826251577, "step": 950}, {"loss": 1.8029, "grad_norm": 0.3394823372364044, "learning_rate": 0.0002, "epoch": 0.8077408498106857, "step": 960}, {"loss": 1.8105, "grad_norm": 0.3741514980792999, "learning_rate": 0.0002, "epoch": 0.8161548169962137, "step": 970}, {"loss": 1.7849, "grad_norm": 0.3655228316783905, "learning_rate": 0.0002, "epoch": 0.8245687841817417, "step": 980}, {"loss": 1.8449, "grad_norm": 0.3586033880710602, "learning_rate": 0.0002, "epoch": 0.8329827513672696, "step": 990}, {"loss": 1.7033, "grad_norm": 0.3459678888320923, "learning_rate": 0.0002, "epoch": 0.8413967185527976, "step": 1000}, {"loss": 1.8498, "grad_norm": 0.3184349834918976, "learning_rate": 0.0002, "epoch": 0.8498106857383256, "step": 1010}, {"loss": 1.7632, "grad_norm": 0.3099786043167114, "learning_rate": 0.0002, "epoch": 0.8582246529238536, "step": 1020}, {"loss": 1.8067, "grad_norm": 0.30300915241241455, "learning_rate": 0.0002, "epoch": 0.8666386201093815, "step": 1030}, {"loss": 1.7923, "grad_norm": 0.3128705620765686, "learning_rate": 0.0002, "epoch": 0.8750525872949095, "step": 1040}, {"loss": 1.8252, "grad_norm": 0.3336263597011566, "learning_rate": 0.0002, "epoch": 0.8834665544804375, "step": 1050}, {"loss": 1.8375, "grad_norm": 0.3801328241825104, "learning_rate": 0.0002, "epoch": 0.8918805216659655, "step": 1060}, {"loss": 1.7757, "grad_norm": 0.3122096359729767, "learning_rate": 0.0002, "epoch": 0.9002944888514934, "step": 1070}, {"loss": 1.8251, "grad_norm": 0.35990869998931885, "learning_rate": 0.0002, "epoch": 0.9087084560370214, "step": 1080}, {"loss": 1.7343, "grad_norm": 0.3321819305419922, "learning_rate": 0.0002, "epoch": 0.9171224232225494, "step": 1090}, {"loss": 1.7595, "grad_norm": 0.4202139377593994, "learning_rate": 0.0002, "epoch": 0.9255363904080774, "step": 1100}, {"loss": 1.8056, "grad_norm": 0.32559722661972046, "learning_rate": 0.0002, "epoch": 0.9339503575936053, "step": 1110}, {"loss": 1.812, "grad_norm": 0.3098459839820862, "learning_rate": 0.0002, "epoch": 0.9423643247791333, "step": 1120}, {"loss": 1.8252, "grad_norm": 0.33917108178138733, "learning_rate": 0.0002, "epoch": 0.9507782919646613, "step": 1130}, {"loss": 1.7709, "grad_norm": 0.4055837094783783, "learning_rate": 0.0002, "epoch": 0.9591922591501894, "step": 1140}, {"loss": 1.8259, "grad_norm": 0.32508623600006104, "learning_rate": 0.0002, "epoch": 0.9676062263357172, "step": 1150}, {"loss": 1.782, "grad_norm": 0.30150601267814636, "learning_rate": 0.0002, "epoch": 0.9760201935212452, "step": 1160}, {"loss": 1.8291, "grad_norm": 0.3042563199996948, "learning_rate": 0.0002, "epoch": 0.9844341607067733, "step": 1170}, {"loss": 1.7847, "grad_norm": 0.33254584670066833, "learning_rate": 0.0002, "epoch": 0.9928481278923013, "step": 1180}, {"eval_loss": 1.8077726364135742, "eval_runtime": 38.4359, "eval_samples_per_second": 13.399, "eval_steps_per_second": 1.691, "epoch": 0.9995793016407236, "step": 1188}, {"loss": 1.7414, "grad_norm": 0.35073035955429077, "learning_rate": 0.0002, "epoch": 1.0012620950778293, "step": 1190}, {"loss": 1.7483, "grad_norm": 0.3217269778251648, "learning_rate": 0.0002, "epoch": 1.0096760622633572, "step": 1200}, {"loss": 1.7517, "grad_norm": 0.3635033369064331, "learning_rate": 0.0002, "epoch": 1.018090029448885, "step": 1210}, {"loss": 1.6949, "grad_norm": 0.32468414306640625, "learning_rate": 0.0002, "epoch": 1.0265039966344132, "step": 1220}, {"loss": 1.711, "grad_norm": 0.3307163417339325, "learning_rate": 0.0002, "epoch": 1.034917963819941, "step": 1230}, {"loss": 1.7881, "grad_norm": 0.34381359815597534, "learning_rate": 0.0002, "epoch": 1.0433319310054692, "step": 1240}, {"loss": 1.612, "grad_norm": 0.35874804854393005, "learning_rate": 0.0002, "epoch": 1.051745898190997, "step": 1250}, {"loss": 1.7314, "grad_norm": 0.3615919351577759, "learning_rate": 0.0002, "epoch": 1.060159865376525, "step": 1260}, {"loss": 1.7517, "grad_norm": 0.32835808396339417, "learning_rate": 0.0002, "epoch": 1.068573832562053, "step": 1270}, {"loss": 1.7193, "grad_norm": 0.3876388370990753, "learning_rate": 0.0002, "epoch": 1.076987799747581, "step": 1280}, {"loss": 1.7442, "grad_norm": 0.39895930886268616, "learning_rate": 0.0002, "epoch": 1.0854017669331089, "step": 1290}, {"loss": 1.6601, "grad_norm": 0.39081698656082153, "learning_rate": 0.0002, "epoch": 1.093815734118637, "step": 1300}, {"loss": 1.7623, "grad_norm": 0.39974215626716614, "learning_rate": 0.0002, "epoch": 1.1022297013041649, "step": 1310}, {"loss": 1.7506, "grad_norm": 0.3887332081794739, "learning_rate": 0.0002, "epoch": 1.110643668489693, "step": 1320}, {"loss": 1.7381, "grad_norm": 0.36216408014297485, "learning_rate": 0.0002, "epoch": 1.1190576356752209, "step": 1330}, {"loss": 1.762, "grad_norm": 0.36979028582572937, "learning_rate": 0.0002, "epoch": 1.1274716028607488, "step": 1340}, {"loss": 1.7515, "grad_norm": 0.34052133560180664, "learning_rate": 0.0002, "epoch": 1.1358855700462769, "step": 1350}, {"loss": 1.7513, "grad_norm": 0.3467716574668884, "learning_rate": 0.0002, "epoch": 1.1442995372318048, "step": 1360}, {"loss": 1.7086, "grad_norm": 0.35528799891471863, "learning_rate": 0.0002, "epoch": 1.1527135044173327, "step": 1370}, {"loss": 1.794, "grad_norm": 0.36282262206077576, "learning_rate": 0.0002, "epoch": 1.1611274716028608, "step": 1380}, {"loss": 1.7731, "grad_norm": 0.37355899810791016, "learning_rate": 0.0002, "epoch": 1.1695414387883887, "step": 1390}, {"loss": 1.7483, "grad_norm": 0.37292736768722534, "learning_rate": 0.0002, "epoch": 1.1779554059739168, "step": 1400}, {"loss": 1.6916, "grad_norm": 0.5892812013626099, "learning_rate": 0.0002, "epoch": 1.1863693731594447, "step": 1410}, {"loss": 1.7302, "grad_norm": 0.3712292015552521, "learning_rate": 0.0002, "epoch": 1.1947833403449726, "step": 1420}, {"loss": 1.7709, "grad_norm": 0.3349577486515045, "learning_rate": 0.0002, "epoch": 1.2031973075305007, "step": 1430}, {"loss": 1.7412, "grad_norm": 0.32591062784194946, "learning_rate": 0.0002, "epoch": 1.2116112747160286, "step": 1440}, {"loss": 1.7406, "grad_norm": 0.3840635418891907, "learning_rate": 0.0002, "epoch": 1.2200252419015567, "step": 1450}, {"loss": 1.7276, "grad_norm": 0.37238365411758423, "learning_rate": 0.0002, "epoch": 1.2284392090870846, "step": 1460}, {"loss": 1.7052, "grad_norm": 0.3731217682361603, "learning_rate": 0.0002, "epoch": 1.2368531762726125, "step": 1470}, {"loss": 1.7255, "grad_norm": 0.3318967819213867, "learning_rate": 0.0002, "epoch": 1.2452671434581406, "step": 1480}, {"loss": 1.7463, "grad_norm": 0.3784034848213196, "learning_rate": 0.0002, "epoch": 1.2536811106436685, "step": 1490}, {"loss": 1.6862, "grad_norm": 0.3541383147239685, "learning_rate": 0.0002, "epoch": 1.2620950778291964, "step": 1500}, {"loss": 1.8394, "grad_norm": 0.35312485694885254, "learning_rate": 0.0002, "epoch": 1.2705090450147245, "step": 1510}, {"loss": 1.7029, "grad_norm": 0.35272929072380066, "learning_rate": 0.0002, "epoch": 1.2789230122002524, "step": 1520}, {"loss": 1.7016, "grad_norm": 0.40988272428512573, "learning_rate": 0.0002, "epoch": 1.2873369793857803, "step": 1530}, {"loss": 1.6912, "grad_norm": 0.3543946146965027, "learning_rate": 0.0002, "epoch": 1.2957509465713084, "step": 1540}, {"loss": 1.6757, "grad_norm": 0.35639145970344543, "learning_rate": 0.0002, "epoch": 1.3041649137568363, "step": 1550}, {"loss": 1.6814, "grad_norm": 0.3290826678276062, "learning_rate": 0.0002, "epoch": 1.3125788809423642, "step": 1560}, {"loss": 1.7369, "grad_norm": 0.39264336228370667, "learning_rate": 0.0002, "epoch": 1.3209928481278923, "step": 1570}, {"loss": 1.6804, "grad_norm": 0.5390415191650391, "learning_rate": 0.0002, "epoch": 1.3294068153134202, "step": 1580}, {"loss": 1.708, "grad_norm": 0.5188116431236267, "learning_rate": 0.0002, "epoch": 1.3378207824989483, "step": 1590}, {"loss": 1.6763, "grad_norm": 0.37445148825645447, "learning_rate": 0.0002, "epoch": 1.3462347496844762, "step": 1600}, {"loss": 1.7386, "grad_norm": 0.3296085298061371, "learning_rate": 0.0002, "epoch": 1.3546487168700043, "step": 1610}, {"loss": 1.8107, "grad_norm": 0.39879581332206726, "learning_rate": 0.0002, "epoch": 1.3630626840555322, "step": 1620}, {"loss": 1.6744, "grad_norm": 0.36092764139175415, "learning_rate": 0.0002, "epoch": 1.37147665124106, "step": 1630}, {"loss": 1.7144, "grad_norm": 0.37011823058128357, "learning_rate": 0.0002, "epoch": 1.3798906184265882, "step": 1640}, {"loss": 1.7396, "grad_norm": 0.40863534808158875, "learning_rate": 0.0002, "epoch": 1.3883045856121161, "step": 1650}, {"loss": 1.7901, "grad_norm": 0.337001770734787, "learning_rate": 0.0002, "epoch": 1.396718552797644, "step": 1660}, {"loss": 1.7044, "grad_norm": 0.35596707463264465, "learning_rate": 0.0002, "epoch": 1.4051325199831721, "step": 1670}, {"loss": 1.7717, "grad_norm": 0.3857671916484833, "learning_rate": 0.0002, "epoch": 1.4135464871687, "step": 1680}, {"loss": 1.7015, "grad_norm": 0.419502317905426, "learning_rate": 0.0002, "epoch": 1.421960454354228, "step": 1690}, {"loss": 1.7261, "grad_norm": 0.35459452867507935, "learning_rate": 0.0002, "epoch": 1.430374421539756, "step": 1700}, {"loss": 1.7361, "grad_norm": 0.37246978282928467, "learning_rate": 0.0002, "epoch": 1.438788388725284, "step": 1710}, {"loss": 1.6762, "grad_norm": 0.33091893792152405, "learning_rate": 0.0002, "epoch": 1.4472023559108118, "step": 1720}, {"loss": 1.7044, "grad_norm": 0.37029674649238586, "learning_rate": 0.0002, "epoch": 1.45561632309634, "step": 1730}, {"loss": 1.7117, "grad_norm": 0.374025821685791, "learning_rate": 0.0002, "epoch": 1.4640302902818678, "step": 1740}, {"loss": 1.7549, "grad_norm": 0.3416315019130707, "learning_rate": 0.0002, "epoch": 1.472444257467396, "step": 1750}, {"loss": 1.7093, "grad_norm": 0.36502841114997864, "learning_rate": 0.0002, "epoch": 1.4808582246529238, "step": 1760}, {"loss": 1.6597, "grad_norm": 0.35458803176879883, "learning_rate": 0.0002, "epoch": 1.489272191838452, "step": 1770}, {"loss": 1.675, "grad_norm": 0.4462839663028717, "learning_rate": 0.0002, "epoch": 1.4976861590239798, "step": 1780}, {"loss": 1.7267, "grad_norm": 0.34836092591285706, "learning_rate": 0.0002, "epoch": 1.5061001262095077, "step": 1790}, {"loss": 1.7295, "grad_norm": 0.3445749282836914, "learning_rate": 0.0002, "epoch": 1.5145140933950358, "step": 1800}, {"loss": 1.7386, "grad_norm": 0.36012160778045654, "learning_rate": 0.0002, "epoch": 1.5229280605805637, "step": 1810}, {"loss": 1.6594, "grad_norm": 0.4052616059780121, "learning_rate": 0.0002, "epoch": 1.5313420277660916, "step": 1820}, {"loss": 1.72, "grad_norm": 0.3966905474662781, "learning_rate": 0.0002, "epoch": 1.5397559949516197, "step": 1830}, {"loss": 1.7595, "grad_norm": 0.35028719902038574, "learning_rate": 0.0002, "epoch": 1.5481699621371476, "step": 1840}, {"loss": 1.6829, "grad_norm": 0.3936742842197418, "learning_rate": 0.0002, "epoch": 1.5565839293226755, "step": 1850}, {"loss": 1.7579, "grad_norm": 0.34473296999931335, "learning_rate": 0.0002, "epoch": 1.5649978965082036, "step": 1860}, {"loss": 1.7207, "grad_norm": 0.4328365623950958, "learning_rate": 0.0002, "epoch": 1.5734118636937318, "step": 1870}, {"loss": 1.7098, "grad_norm": 0.3566315472126007, "learning_rate": 0.0002, "epoch": 1.5818258308792594, "step": 1880}, {"loss": 1.6095, "grad_norm": 0.3301256597042084, "learning_rate": 0.0002, "epoch": 1.5902397980647875, "step": 1890}, {"loss": 1.748, "grad_norm": 0.3743041455745697, "learning_rate": 0.0002, "epoch": 1.5986537652503157, "step": 1900}, {"loss": 1.7259, "grad_norm": 0.3735344707965851, "learning_rate": 0.0002, "epoch": 1.6070677324358436, "step": 1910}, {"loss": 1.7445, "grad_norm": 0.42191144824028015, "learning_rate": 0.0002, "epoch": 1.6154816996213714, "step": 1920}, {"loss": 1.6978, "grad_norm": 0.3787207305431366, "learning_rate": 0.0002, "epoch": 1.6238956668068996, "step": 1930}, {"loss": 1.6893, "grad_norm": 0.35647350549697876, "learning_rate": 0.0002, "epoch": 1.6323096339924275, "step": 1940}, {"loss": 1.7825, "grad_norm": 0.39791446924209595, "learning_rate": 0.0002, "epoch": 1.6407236011779553, "step": 1950}, {"loss": 1.7293, "grad_norm": 0.37341275811195374, "learning_rate": 0.0002, "epoch": 1.6491375683634835, "step": 1960}, {"loss": 1.6781, "grad_norm": 0.3722686469554901, "learning_rate": 0.0002, "epoch": 1.6575515355490114, "step": 1970}, {"loss": 1.6383, "grad_norm": 0.37467387318611145, "learning_rate": 0.0002, "epoch": 1.6659655027345392, "step": 1980}, {"loss": 1.7439, "grad_norm": 0.37109461426734924, "learning_rate": 0.0002, "epoch": 1.6743794699200674, "step": 1990}, {"loss": 1.7206, "grad_norm": 0.4008837044239044, "learning_rate": 0.0002, "epoch": 1.6827934371055953, "step": 2000}, {"loss": 1.7604, "grad_norm": 0.3316999673843384, "learning_rate": 0.0002, "epoch": 1.6912074042911232, "step": 2010}, {"loss": 1.7325, "grad_norm": 0.3683805465698242, "learning_rate": 0.0002, "epoch": 1.6996213714766513, "step": 2020}, {"loss": 1.7451, "grad_norm": 0.4163658320903778, "learning_rate": 0.0002, "epoch": 1.7080353386621794, "step": 2030}, {"loss": 1.741, "grad_norm": 0.4245431125164032, "learning_rate": 0.0002, "epoch": 1.716449305847707, "step": 2040}, {"loss": 1.7184, "grad_norm": 0.36732038855552673, "learning_rate": 0.0002, "epoch": 1.7248632730332352, "step": 2050}, {"loss": 1.7031, "grad_norm": 0.34981656074523926, "learning_rate": 0.0002, "epoch": 1.7332772402187633, "step": 2060}, {"loss": 1.7545, "grad_norm": 0.38588812947273254, "learning_rate": 0.0002, "epoch": 1.7416912074042912, "step": 2070}, {"loss": 1.7728, "grad_norm": 0.39914557337760925, "learning_rate": 0.0002, "epoch": 1.750105174589819, "step": 2080}, {"loss": 1.7049, "grad_norm": 0.36068692803382874, "learning_rate": 0.0002, "epoch": 1.7585191417753472, "step": 2090}, {"loss": 1.7537, "grad_norm": 0.3983287215232849, "learning_rate": 0.0002, "epoch": 1.766933108960875, "step": 2100}, {"loss": 1.7016, "grad_norm": 0.45008400082588196, "learning_rate": 0.0002, "epoch": 1.775347076146403, "step": 2110}, {"loss": 1.7163, "grad_norm": 0.3618052303791046, "learning_rate": 0.0002, "epoch": 1.783761043331931, "step": 2120}, {"loss": 1.7335, "grad_norm": 0.38745400309562683, "learning_rate": 0.0002, "epoch": 1.792175010517459, "step": 2130}, {"loss": 1.7387, "grad_norm": 0.3413826525211334, "learning_rate": 0.0002, "epoch": 1.8005889777029869, "step": 2140}, {"loss": 1.7414, "grad_norm": 0.35983747243881226, "learning_rate": 0.0002, "epoch": 1.809002944888515, "step": 2150}, {"loss": 1.7892, "grad_norm": 0.40926849842071533, "learning_rate": 0.0002, "epoch": 1.8174169120740429, "step": 2160}, {"loss": 1.6823, "grad_norm": 0.3543093800544739, "learning_rate": 0.0002, "epoch": 1.8258308792595708, "step": 2170}, {"loss": 1.7812, "grad_norm": 0.42690935730934143, "learning_rate": 0.0002, "epoch": 1.8342448464450989, "step": 2180}, {"loss": 1.7471, "grad_norm": 0.40282756090164185, "learning_rate": 0.0002, "epoch": 1.842658813630627, "step": 2190}, {"loss": 1.7411, "grad_norm": 0.36568400263786316, "learning_rate": 0.0002, "epoch": 1.8510727808161547, "step": 2200}, {"loss": 1.7024, "grad_norm": 0.43159013986587524, "learning_rate": 0.0002, "epoch": 1.8594867480016828, "step": 2210}, {"loss": 1.7298, "grad_norm": 0.3554118573665619, "learning_rate": 0.0002, "epoch": 1.867900715187211, "step": 2220}, {"loss": 1.7157, "grad_norm": 0.43349072337150574, "learning_rate": 0.0002, "epoch": 1.8763146823727388, "step": 2230}, {"loss": 1.7302, "grad_norm": 0.36486536264419556, "learning_rate": 0.0002, "epoch": 1.8847286495582667, "step": 2240}, {"loss": 1.6901, "grad_norm": 0.39260047674179077, "learning_rate": 0.0002, "epoch": 1.8931426167437948, "step": 2250}, {"loss": 1.6691, "grad_norm": 0.3741776943206787, "learning_rate": 0.0002, "epoch": 1.9015565839293227, "step": 2260}, {"loss": 1.6931, "grad_norm": 0.3961946964263916, "learning_rate": 0.0002, "epoch": 1.9099705511148506, "step": 2270}, {"loss": 1.737, "grad_norm": 0.3659731149673462, "learning_rate": 0.0002, "epoch": 1.9183845183003787, "step": 2280}, {"loss": 1.7342, "grad_norm": 0.34744107723236084, "learning_rate": 0.0002, "epoch": 1.9267984854859066, "step": 2290}, {"loss": 1.7162, "grad_norm": 0.3607442378997803, "learning_rate": 0.0002, "epoch": 1.9352124526714345, "step": 2300}, {"loss": 1.6673, "grad_norm": 0.331464558839798, "learning_rate": 0.0002, "epoch": 1.9436264198569626, "step": 2310}, {"loss": 1.7101, "grad_norm": 0.3904414474964142, "learning_rate": 0.0002, "epoch": 1.9520403870424905, "step": 2320}, {"loss": 1.7327, "grad_norm": 0.37584832310676575, "learning_rate": 0.0002, "epoch": 1.9604543542280184, "step": 2330}, {"loss": 1.7586, "grad_norm": 0.3698684275150299, "learning_rate": 0.0002, "epoch": 1.9688683214135465, "step": 2340}, {"loss": 1.7764, "grad_norm": 0.40571412444114685, "learning_rate": 0.0002, "epoch": 1.9772822885990746, "step": 2350}, {"loss": 1.744, "grad_norm": 0.40059587359428406, "learning_rate": 0.0002, "epoch": 1.9856962557846023, "step": 2360}, {"loss": 1.7033, "grad_norm": 0.4168248474597931, "learning_rate": 0.0002, "epoch": 1.9941102229701304, "step": 2370}, {"eval_loss": 1.8055059909820557, "eval_runtime": 38.422, "eval_samples_per_second": 13.404, "eval_steps_per_second": 1.692, "epoch": 2.0, "step": 2377}, {"loss": 1.7673, "grad_norm": 0.35205352306365967, "learning_rate": 0.0002, "epoch": 2.0025241901556585, "step": 2380}, {"loss": 1.6556, "grad_norm": 0.3979377746582031, "learning_rate": 0.0002, "epoch": 2.010938157341186, "step": 2390}, {"loss": 1.6421, "grad_norm": 0.396491676568985, "learning_rate": 0.0002, "epoch": 2.0193521245267143, "step": 2400}, {"loss": 1.6847, "grad_norm": 0.44712209701538086, "learning_rate": 0.0002, "epoch": 2.0277660917122424, "step": 2410}, {"loss": 1.6877, "grad_norm": 0.4454420208930969, "learning_rate": 0.0002, "epoch": 2.03618005889777, "step": 2420}, {"loss": 1.6635, "grad_norm": 0.4170038402080536, "learning_rate": 0.0002, "epoch": 2.044594026083298, "step": 2430}, {"loss": 1.6512, "grad_norm": 0.4309595227241516, "learning_rate": 0.0002, "epoch": 2.0530079932688263, "step": 2440}, {"loss": 1.6223, "grad_norm": 0.4241602122783661, "learning_rate": 0.0002, "epoch": 2.0614219604543544, "step": 2450}, {"loss": 1.6162, "grad_norm": 0.4370540678501129, "learning_rate": 0.0002, "epoch": 2.069835927639882, "step": 2460}, {"loss": 1.6354, "grad_norm": 0.43985554575920105, "learning_rate": 0.0002, "epoch": 2.0782498948254102, "step": 2470}, {"loss": 1.6954, "grad_norm": 0.4158105254173279, "learning_rate": 0.0002, "epoch": 2.0866638620109383, "step": 2480}, {"loss": 1.6114, "grad_norm": 0.441549152135849, "learning_rate": 0.0002, "epoch": 2.095077829196466, "step": 2490}, {"loss": 1.5485, "grad_norm": 0.385718435049057, "learning_rate": 0.0002, "epoch": 2.103491796381994, "step": 2500}, {"loss": 1.5894, "grad_norm": 0.43146514892578125, "learning_rate": 0.0002, "epoch": 2.1119057635675222, "step": 2510}, {"loss": 1.6414, "grad_norm": 0.41663315892219543, "learning_rate": 0.0002, "epoch": 2.12031973075305, "step": 2520}, {"loss": 1.6527, "grad_norm": 0.4410698115825653, "learning_rate": 0.0002, "epoch": 2.128733697938578, "step": 2530}, {"loss": 1.6124, "grad_norm": 0.4472278952598572, "learning_rate": 0.0002, "epoch": 2.137147665124106, "step": 2540}, {"loss": 1.6257, "grad_norm": 0.3879167437553406, "learning_rate": 0.0002, "epoch": 2.145561632309634, "step": 2550}, {"loss": 1.6682, "grad_norm": 0.4212203025817871, "learning_rate": 0.0002, "epoch": 2.153975599495162, "step": 2560}, {"loss": 1.6036, "grad_norm": 0.42841723561286926, "learning_rate": 0.0002, "epoch": 2.16238956668069, "step": 2570}, {"loss": 1.5962, "grad_norm": 0.39272481203079224, "learning_rate": 0.0002, "epoch": 2.1708035338662177, "step": 2580}, {"loss": 1.681, "grad_norm": 0.4075261354446411, "learning_rate": 0.0002, "epoch": 2.179217501051746, "step": 2590}, {"loss": 1.6601, "grad_norm": 0.5358437895774841, "learning_rate": 0.0002, "epoch": 2.187631468237274, "step": 2600}, {"loss": 1.6423, "grad_norm": 0.4738350212574005, "learning_rate": 0.0002, "epoch": 2.1960454354228016, "step": 2610}, {"loss": 1.6386, "grad_norm": 0.446789026260376, "learning_rate": 0.0002, "epoch": 2.2044594026083297, "step": 2620}, {"loss": 1.6246, "grad_norm": 0.4615374505519867, "learning_rate": 0.0002, "epoch": 2.212873369793858, "step": 2630}, {"loss": 1.6205, "grad_norm": 0.46901994943618774, "learning_rate": 0.0002, "epoch": 2.221287336979386, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.46267789602279663, "learning_rate": 0.0002, "epoch": 2.2297013041649136, "step": 2650}, {"loss": 1.6584, "grad_norm": 0.4383080005645752, "learning_rate": 0.0002, "epoch": 2.2381152713504417, "step": 2660}, {"loss": 1.5745, "grad_norm": 0.4070609509944916, "learning_rate": 0.0002, "epoch": 2.24652923853597, "step": 2670}, {"loss": 1.6125, "grad_norm": 0.4572339951992035, "learning_rate": 0.0002, "epoch": 2.2549432057214975, "step": 2680}, {"loss": 1.5671, "grad_norm": 0.393265038728714, "learning_rate": 0.0002, "epoch": 2.2633571729070256, "step": 2690}, {"loss": 1.6239, "grad_norm": 0.46144717931747437, "learning_rate": 0.0002, "epoch": 2.2717711400925538, "step": 2700}, {"loss": 1.5992, "grad_norm": 0.45077767968177795, "learning_rate": 0.0002, "epoch": 2.2801851072780814, "step": 2710}, {"loss": 1.6261, "grad_norm": 0.5697639584541321, "learning_rate": 0.0002, "epoch": 2.2885990744636096, "step": 2720}, {"loss": 1.6192, "grad_norm": 0.4855510890483856, "learning_rate": 0.0002, "epoch": 2.2970130416491377, "step": 2730}, {"loss": 1.7419, "grad_norm": 0.4440622627735138, "learning_rate": 0.0002, "epoch": 2.3054270088346653, "step": 2740}, {"loss": 1.6496, "grad_norm": 0.3904096782207489, "learning_rate": 0.0002, "epoch": 2.3138409760201935, "step": 2750}, {"loss": 1.5888, "grad_norm": 0.5225510597229004, "learning_rate": 0.0002, "epoch": 2.3222549432057216, "step": 2760}, {"loss": 1.6082, "grad_norm": 0.44866397976875305, "learning_rate": 0.0002, "epoch": 2.3306689103912497, "step": 2770}, {"loss": 1.6087, "grad_norm": 0.5167056322097778, "learning_rate": 0.0002, "epoch": 2.3390828775767774, "step": 2780}, {"loss": 1.6136, "grad_norm": 0.45913267135620117, "learning_rate": 0.0002, "epoch": 2.3474968447623055, "step": 2790}, {"loss": 1.6564, "grad_norm": 0.45787590742111206, "learning_rate": 0.0002, "epoch": 2.3559108119478336, "step": 2800}, {"loss": 1.6868, "grad_norm": 0.4633352756500244, "learning_rate": 0.0002, "epoch": 2.3643247791333613, "step": 2810}, {"loss": 1.6316, "grad_norm": 0.46390071511268616, "learning_rate": 0.0002, "epoch": 2.3727387463188894, "step": 2820}, {"loss": 1.6039, "grad_norm": 0.4261005222797394, "learning_rate": 0.0002, "epoch": 2.3811527135044175, "step": 2830}, {"loss": 1.6364, "grad_norm": 0.4283634424209595, "learning_rate": 0.0002, "epoch": 2.389566680689945, "step": 2840}, {"loss": 1.6382, "grad_norm": 0.4955291450023651, "learning_rate": 0.0002, "epoch": 2.3979806478754733, "step": 2850}, {"loss": 1.6173, "grad_norm": 0.4740189015865326, "learning_rate": 0.0002, "epoch": 2.4063946150610014, "step": 2860}, {"loss": 1.6403, "grad_norm": 0.4222276508808136, "learning_rate": 0.0002, "epoch": 2.414808582246529, "step": 2870}, {"loss": 1.5602, "grad_norm": 0.4982149004936218, "learning_rate": 0.0002, "epoch": 2.423222549432057, "step": 2880}, {"loss": 1.6313, "grad_norm": 0.5217409133911133, "learning_rate": 0.0002, "epoch": 2.4316365166175853, "step": 2890}, {"loss": 1.5804, "grad_norm": 0.4555884897708893, "learning_rate": 0.0002, "epoch": 2.4400504838031134, "step": 2900}, {"loss": 1.6189, "grad_norm": 0.43178579211235046, "learning_rate": 0.0002, "epoch": 2.448464450988641, "step": 2910}, {"loss": 1.6824, "grad_norm": 0.4788478910923004, "learning_rate": 0.0002, "epoch": 2.456878418174169, "step": 2920}, {"loss": 1.6829, "grad_norm": 0.43689873814582825, "learning_rate": 0.0002, "epoch": 2.465292385359697, "step": 2930}, {"loss": 1.6196, "grad_norm": 0.5115197896957397, "learning_rate": 0.0002, "epoch": 2.473706352545225, "step": 2940}, {"loss": 1.689, "grad_norm": 0.5290159583091736, "learning_rate": 0.0002, "epoch": 2.482120319730753, "step": 2950}, {"loss": 1.6499, "grad_norm": 0.46042463183403015, "learning_rate": 0.0002, "epoch": 2.490534286916281, "step": 2960}, {"loss": 1.6664, "grad_norm": 0.4359915852546692, "learning_rate": 0.0002, "epoch": 2.498948254101809, "step": 2970}, {"loss": 1.5812, "grad_norm": 0.46352964639663696, "learning_rate": 0.0002, "epoch": 2.507362221287337, "step": 2980}, {"loss": 1.6501, "grad_norm": 0.5324268341064453, "learning_rate": 0.0002, "epoch": 2.515776188472865, "step": 2990}, {"loss": 1.6115, "grad_norm": 0.5929607152938843, "learning_rate": 0.0002, "epoch": 2.5241901556583928, "step": 3000}, {"loss": 1.6772, "grad_norm": 0.4811333417892456, "learning_rate": 0.0002, "epoch": 2.532604122843921, "step": 3010}, {"loss": 1.7023, "grad_norm": 0.4662701487541199, "learning_rate": 0.0002, "epoch": 2.541018090029449, "step": 3020}, {"loss": 1.5426, "grad_norm": 0.4582270681858063, "learning_rate": 0.0002, "epoch": 2.549432057214977, "step": 3030}, {"loss": 1.6737, "grad_norm": 0.4679982662200928, "learning_rate": 0.0002, "epoch": 2.557846024400505, "step": 3040}, {"loss": 1.5442, "grad_norm": 0.4380294680595398, "learning_rate": 0.0002, "epoch": 2.566259991586033, "step": 3050}, {"loss": 1.6055, "grad_norm": 0.44295763969421387, "learning_rate": 0.0002, "epoch": 2.5746739587715606, "step": 3060}, {"loss": 1.5775, "grad_norm": 0.5131027698516846, "learning_rate": 0.0002, "epoch": 2.5830879259570887, "step": 3070}, {"loss": 1.546, "grad_norm": 0.47567516565322876, "learning_rate": 0.0002, "epoch": 2.591501893142617, "step": 3080}, {"loss": 1.5671, "grad_norm": 0.49002596735954285, "learning_rate": 0.0002, "epoch": 2.599915860328145, "step": 3090}, {"loss": 1.5445, "grad_norm": 0.44856327772140503, "learning_rate": 0.0002, "epoch": 2.6083298275136726, "step": 3100}, {"loss": 1.5797, "grad_norm": 0.4480142593383789, "learning_rate": 0.0002, "epoch": 2.6167437946992007, "step": 3110}, {"loss": 1.7132, "grad_norm": 0.4317494034767151, "learning_rate": 0.0002, "epoch": 2.6251577618847284, "step": 3120}, {"loss": 1.6321, "grad_norm": 0.42580848932266235, "learning_rate": 0.0002, "epoch": 2.6335717290702565, "step": 3130}, {"loss": 1.6483, "grad_norm": 0.4516814947128296, "learning_rate": 0.0002, "epoch": 2.6419856962557846, "step": 3140}, {"loss": 1.695, "grad_norm": 0.4438435733318329, "learning_rate": 0.0002, "epoch": 2.6503996634413127, "step": 3150}, {"loss": 1.6938, "grad_norm": 0.4385356307029724, "learning_rate": 0.0002, "epoch": 2.6588136306268404, "step": 3160}, {"loss": 1.6139, "grad_norm": 0.5064112544059753, "learning_rate": 0.0002, "epoch": 2.6672275978123685, "step": 3170}, {"loss": 1.7189, "grad_norm": 0.49163177609443665, "learning_rate": 0.0002, "epoch": 2.6756415649978966, "step": 3180}, {"loss": 1.7323, "grad_norm": 0.49339258670806885, "learning_rate": 0.0002, "epoch": 2.6840555321834243, "step": 3190}, {"loss": 1.6508, "grad_norm": 0.440950870513916, "learning_rate": 0.0002, "epoch": 2.6924694993689524, "step": 3200}, {"loss": 1.6305, "grad_norm": 0.4283970594406128, "learning_rate": 0.0002, "epoch": 2.7008834665544805, "step": 3210}, {"loss": 1.5935, "grad_norm": 0.43875712156295776, "learning_rate": 0.0002, "epoch": 2.7092974337400086, "step": 3220}, {"loss": 1.6129, "grad_norm": 0.49332964420318604, "learning_rate": 0.0002, "epoch": 2.7177114009255363, "step": 3230}, {"loss": 1.642, "grad_norm": 0.5225692391395569, "learning_rate": 0.0002, "epoch": 2.7261253681110644, "step": 3240}, {"loss": 1.6759, "grad_norm": 0.4856489300727844, "learning_rate": 0.0002, "epoch": 2.734539335296592, "step": 3250}, {"loss": 1.6463, "grad_norm": 0.46918296813964844, "learning_rate": 0.0002, "epoch": 2.74295330248212, "step": 3260}, {"loss": 1.6819, "grad_norm": 0.4802931249141693, "learning_rate": 0.0002, "epoch": 2.7513672696676483, "step": 3270}, {"loss": 1.6246, "grad_norm": 0.4485355615615845, "learning_rate": 0.0002, "epoch": 2.7597812368531764, "step": 3280}, {"loss": 1.6251, "grad_norm": 0.43944594264030457, "learning_rate": 0.0002, "epoch": 2.768195204038704, "step": 3290}, {"loss": 1.6501, "grad_norm": 0.46847742795944214, "learning_rate": 0.0002, "epoch": 2.7766091712242322, "step": 3300}, {"loss": 1.5969, "grad_norm": 0.4816027879714966, "learning_rate": 0.0002, "epoch": 2.7850231384097603, "step": 3310}, {"loss": 1.6293, "grad_norm": 0.453960120677948, "learning_rate": 0.0002, "epoch": 2.793437105595288, "step": 3320}, {"loss": 1.6429, "grad_norm": 0.4816017150878906, "learning_rate": 0.0002, "epoch": 2.801851072780816, "step": 3330}, {"loss": 1.6683, "grad_norm": 0.4461034834384918, "learning_rate": 0.0002, "epoch": 2.8102650399663442, "step": 3340}, {"loss": 1.7048, "grad_norm": 0.48821821808815, "learning_rate": 0.0002, "epoch": 2.8186790071518724, "step": 3350}, {"loss": 1.6076, "grad_norm": 0.4574853777885437, "learning_rate": 0.0002, "epoch": 2.8270929743374, "step": 3360}, {"loss": 1.6651, "grad_norm": 0.42062026262283325, "learning_rate": 0.0002, "epoch": 2.835506941522928, "step": 3370}, {"loss": 1.624, "grad_norm": 0.4499834477901459, "learning_rate": 0.0002, "epoch": 2.843920908708456, "step": 3380}, {"loss": 1.621, "grad_norm": 0.4780360758304596, "learning_rate": 0.0002, "epoch": 2.852334875893984, "step": 3390}, {"loss": 1.5882, "grad_norm": 0.45422887802124023, "learning_rate": 0.0002, "epoch": 2.860748843079512, "step": 3400}, {"loss": 1.6028, "grad_norm": 0.4590015709400177, "learning_rate": 0.0002, "epoch": 2.86916281026504, "step": 3410}, {"loss": 1.6746, "grad_norm": 0.45689624547958374, "learning_rate": 0.0002, "epoch": 2.877576777450568, "step": 3420}, {"loss": 1.6326, "grad_norm": 0.46953922510147095, "learning_rate": 0.0002, "epoch": 2.885990744636096, "step": 3430}, {"loss": 1.6015, "grad_norm": 0.4791966378688812, "learning_rate": 0.0002, "epoch": 2.8944047118216236, "step": 3440}, {"loss": 1.694, "grad_norm": 0.4842296242713928, "learning_rate": 0.0002, "epoch": 2.9028186790071517, "step": 3450}, {"loss": 1.6326, "grad_norm": 0.47219768166542053, "learning_rate": 0.0002, "epoch": 2.91123264619268, "step": 3460}, {"loss": 1.6486, "grad_norm": 0.4622127115726471, "learning_rate": 0.0002, "epoch": 2.919646613378208, "step": 3470}, {"loss": 1.6485, "grad_norm": 0.46832820773124695, "learning_rate": 0.0002, "epoch": 2.9280605805637356, "step": 3480}, {"loss": 1.6366, "grad_norm": 0.44582483172416687, "learning_rate": 0.0002, "epoch": 2.9364745477492638, "step": 3490}, {"loss": 1.6859, "grad_norm": 0.4987219274044037, "learning_rate": 0.0002, "epoch": 2.944888514934792, "step": 3500}, {"loss": 1.5991, "grad_norm": 0.43750956654548645, "learning_rate": 0.0002, "epoch": 2.9533024821203195, "step": 3510}, {"loss": 1.6236, "grad_norm": 0.49962925910949707, "learning_rate": 0.0002, "epoch": 2.9617164493058477, "step": 3520}, {"loss": 1.5859, "grad_norm": 0.5189590454101562, "learning_rate": 0.0002, "epoch": 2.9701304164913758, "step": 3530}, {"loss": 1.6688, "grad_norm": 0.391317754983902, "learning_rate": 0.0002, "epoch": 2.978544383676904, "step": 3540}, {"loss": 1.5884, "grad_norm": 0.44934695959091187, "learning_rate": 0.0002, "epoch": 2.9869583508624316, "step": 3550}, {"loss": 1.5688, "grad_norm": 0.4740142226219177, "learning_rate": 0.0002, "epoch": 2.9953723180479597, "step": 3560}, {"eval_loss": 1.8266887664794922, "eval_runtime": 37.9445, "eval_samples_per_second": 13.572, "eval_steps_per_second": 1.713, "epoch": 2.9995793016407237, "step": 3565}, {"loss": 1.5939, "grad_norm": 0.4523724615573883, "learning_rate": 0.0002, "epoch": 3.003786285233488, "step": 3570}, {"loss": 1.526, "grad_norm": 0.5261380076408386, "learning_rate": 0.0002, "epoch": 3.0122002524190155, "step": 3580}, {"loss": 1.4946, "grad_norm": 0.48664888739585876, "learning_rate": 0.0002, "epoch": 3.0206142196045436, "step": 3590}, {"loss": 1.5193, "grad_norm": 0.5070882439613342, "learning_rate": 0.0002, "epoch": 3.0290281867900717, "step": 3600}, {"loss": 1.5316, "grad_norm": 0.5816011428833008, "learning_rate": 0.0002, "epoch": 3.0374421539755994, "step": 3610}, {"loss": 1.5682, "grad_norm": 0.6610211730003357, "learning_rate": 0.0002, "epoch": 3.0458561211611275, "step": 3620}, {"loss": 1.5699, "grad_norm": 0.5257703065872192, "learning_rate": 0.0002, "epoch": 3.0542700883466556, "step": 3630}, {"loss": 1.4438, "grad_norm": 0.5574390888214111, "learning_rate": 0.0002, "epoch": 3.0626840555321833, "step": 3640}, {"loss": 1.547, "grad_norm": 0.5682297348976135, "learning_rate": 0.0002, "epoch": 3.0710980227177114, "step": 3650}, {"loss": 1.5743, "grad_norm": 0.5798383355140686, "learning_rate": 0.0002, "epoch": 3.0795119899032395, "step": 3660}, {"loss": 1.4339, "grad_norm": 0.5458289980888367, "learning_rate": 0.0002, "epoch": 3.087925957088767, "step": 3670}, {"loss": 1.46, "grad_norm": 0.5599102973937988, "learning_rate": 0.0002, "epoch": 3.0963399242742953, "step": 3680}, {"loss": 1.4589, "grad_norm": 0.5023021697998047, "learning_rate": 0.0002, "epoch": 3.1047538914598234, "step": 3690}, {"loss": 1.5114, "grad_norm": 0.5448206067085266, "learning_rate": 0.0002, "epoch": 3.113167858645351, "step": 3700}, {"loss": 1.4692, "grad_norm": 0.5760458707809448, "learning_rate": 0.0002, "epoch": 3.121581825830879, "step": 3710}, {"loss": 1.4789, "grad_norm": 0.6018968224525452, "learning_rate": 0.0002, "epoch": 3.1299957930164073, "step": 3720}, {"loss": 1.5518, "grad_norm": 0.5767101049423218, "learning_rate": 0.0002, "epoch": 3.1384097602019354, "step": 3730}, {"loss": 1.5032, "grad_norm": 0.5333963632583618, "learning_rate": 0.0002, "epoch": 3.146823727387463, "step": 3740}, {"loss": 1.4812, "grad_norm": 0.5918396711349487, "learning_rate": 0.0002, "epoch": 3.155237694572991, "step": 3750}, {"loss": 1.4618, "grad_norm": 0.5931203365325928, "learning_rate": 0.0002, "epoch": 3.1636516617585193, "step": 3760}, {"loss": 1.5592, "grad_norm": 0.6562168598175049, "learning_rate": 0.0002, "epoch": 3.172065628944047, "step": 3770}, {"loss": 1.4932, "grad_norm": 0.5820156335830688, "learning_rate": 0.0002, "epoch": 3.180479596129575, "step": 3780}, {"loss": 1.4523, "grad_norm": 0.5784737467765808, "learning_rate": 0.0002, "epoch": 3.188893563315103, "step": 3790}, {"loss": 1.498, "grad_norm": 0.5506529808044434, "learning_rate": 0.0002, "epoch": 3.197307530500631, "step": 3800}, {"loss": 1.4819, "grad_norm": 0.6101595163345337, "learning_rate": 0.0002, "epoch": 3.205721497686159, "step": 3810}, {"loss": 1.5185, "grad_norm": 0.5597806572914124, "learning_rate": 0.0002, "epoch": 3.214135464871687, "step": 3820}, {"loss": 1.5664, "grad_norm": 0.5641011595726013, "learning_rate": 0.0002, "epoch": 3.222549432057215, "step": 3830}, {"loss": 1.4702, "grad_norm": 0.5892080068588257, "learning_rate": 0.0002, "epoch": 3.230963399242743, "step": 3840}, {"loss": 1.4194, "grad_norm": 0.6034760475158691, "learning_rate": 0.0002, "epoch": 3.239377366428271, "step": 3850}, {"loss": 1.5499, "grad_norm": 0.5112439393997192, "learning_rate": 0.0002, "epoch": 3.247791333613799, "step": 3860}, {"loss": 1.5132, "grad_norm": 0.56565922498703, "learning_rate": 0.0002, "epoch": 3.256205300799327, "step": 3870}, {"loss": 1.4892, "grad_norm": 0.6155247092247009, "learning_rate": 0.0002, "epoch": 3.264619267984855, "step": 3880}, {"loss": 1.5118, "grad_norm": 0.6064623594284058, "learning_rate": 0.0002, "epoch": 3.273033235170383, "step": 3890}, {"loss": 1.5236, "grad_norm": 0.6313768029212952, "learning_rate": 0.0002, "epoch": 3.2814472023559107, "step": 3900}, {"loss": 1.5551, "grad_norm": 0.5903939008712769, "learning_rate": 0.0002, "epoch": 3.289861169541439, "step": 3910}, {"loss": 1.5703, "grad_norm": 0.5770667195320129, "learning_rate": 0.0002, "epoch": 3.298275136726967, "step": 3920}, {"loss": 1.5159, "grad_norm": 0.5785196423530579, "learning_rate": 0.0002, "epoch": 3.3066891039124946, "step": 3930}, {"loss": 1.5277, "grad_norm": 0.6468310356140137, "learning_rate": 0.0002, "epoch": 3.3151030710980227, "step": 3940}, {"loss": 1.6002, "grad_norm": 0.6200279593467712, "learning_rate": 0.0002, "epoch": 3.323517038283551, "step": 3950}, {"loss": 1.5264, "grad_norm": 0.5779302716255188, "learning_rate": 0.0002, "epoch": 3.3319310054690785, "step": 3960}, {"loss": 1.4861, "grad_norm": 0.5463796854019165, "learning_rate": 0.0002, "epoch": 3.3403449726546066, "step": 3970}, {"loss": 1.541, "grad_norm": 0.6117855906486511, "learning_rate": 0.0002, "epoch": 3.3487589398401347, "step": 3980}, {"loss": 1.5566, "grad_norm": 0.5554766058921814, "learning_rate": 0.0002, "epoch": 3.357172907025663, "step": 3990}, {"loss": 1.5004, "grad_norm": 0.6012870073318481, "learning_rate": 0.0002, "epoch": 3.3655868742111905, "step": 4000}, {"loss": 1.473, "grad_norm": 0.5443974137306213, "learning_rate": 0.0002, "epoch": 3.3740008413967186, "step": 4010}, {"loss": 1.5139, "grad_norm": 0.6636057496070862, "learning_rate": 0.0002, "epoch": 3.3824148085822463, "step": 4020}, {"loss": 1.5141, "grad_norm": 0.5801246166229248, "learning_rate": 0.0002, "epoch": 3.3908287757677744, "step": 4030}, {"loss": 1.5026, "grad_norm": 0.5668839812278748, "learning_rate": 0.0002, "epoch": 3.3992427429533025, "step": 4040}, {"loss": 1.523, "grad_norm": 0.7763481736183167, "learning_rate": 0.0002, "epoch": 3.4076567101388306, "step": 4050}, {"loss": 1.4932, "grad_norm": 0.6675992608070374, "learning_rate": 0.0002, "epoch": 3.4160706773243583, "step": 4060}, {"loss": 1.4959, "grad_norm": 0.6290077567100525, "learning_rate": 0.0002, "epoch": 3.4244846445098864, "step": 4070}, {"loss": 1.5766, "grad_norm": 0.6040239930152893, "learning_rate": 0.0002, "epoch": 3.4328986116954145, "step": 4080}, {"loss": 1.5711, "grad_norm": 0.6237877607345581, "learning_rate": 0.0002, "epoch": 3.441312578880942, "step": 4090}, {"loss": 1.4961, "grad_norm": 0.5343508124351501, "learning_rate": 0.0002, "epoch": 3.4497265460664703, "step": 4100}, {"loss": 1.5123, "grad_norm": 0.6817412972450256, "learning_rate": 0.0002, "epoch": 3.4581405132519984, "step": 4110}, {"loss": 1.5377, "grad_norm": 0.7115170359611511, "learning_rate": 0.0002, "epoch": 3.466554480437526, "step": 4120}, {"loss": 1.5275, "grad_norm": 0.6127332448959351, "learning_rate": 0.0002, "epoch": 3.4749684476230542, "step": 4130}, {"loss": 1.557, "grad_norm": 0.5745994448661804, "learning_rate": 0.0002, "epoch": 3.4833824148085824, "step": 4140}, {"loss": 1.4873, "grad_norm": 0.6248795390129089, "learning_rate": 0.0002, "epoch": 3.49179638199411, "step": 4150}, {"loss": 1.4885, "grad_norm": 0.5821124911308289, "learning_rate": 0.0002, "epoch": 3.500210349179638, "step": 4160}, {"loss": 1.4937, "grad_norm": 0.561416506767273, "learning_rate": 0.0002, "epoch": 3.5086243163651663, "step": 4170}, {"loss": 1.5453, "grad_norm": 0.5848962664604187, "learning_rate": 0.0002, "epoch": 3.5170382835506944, "step": 4180}, {"loss": 1.5892, "grad_norm": 0.5335569977760315, "learning_rate": 0.0002, "epoch": 3.525452250736222, "step": 4190}, {"loss": 1.5152, "grad_norm": 0.547964870929718, "learning_rate": 0.0002, "epoch": 3.53386621792175, "step": 4200}, {"loss": 1.4887, "grad_norm": 0.6157727241516113, "learning_rate": 0.0002, "epoch": 3.542280185107278, "step": 4210}, {"loss": 1.5484, "grad_norm": 0.6163121461868286, "learning_rate": 0.0002, "epoch": 3.550694152292806, "step": 4220}, {"loss": 1.5833, "grad_norm": 0.5844616293907166, "learning_rate": 0.0002, "epoch": 3.559108119478334, "step": 4230}, {"loss": 1.5305, "grad_norm": 0.7104926109313965, "learning_rate": 0.0002, "epoch": 3.567522086663862, "step": 4240}, {"loss": 1.5161, "grad_norm": 0.5055213570594788, "learning_rate": 0.0002, "epoch": 3.57593605384939, "step": 4250}, {"loss": 1.482, "grad_norm": 0.611676812171936, "learning_rate": 0.0002, "epoch": 3.584350021034918, "step": 4260}, {"loss": 1.5048, "grad_norm": 0.6326440572738647, "learning_rate": 0.0002, "epoch": 3.592763988220446, "step": 4270}, {"loss": 1.5122, "grad_norm": 0.6290925741195679, "learning_rate": 0.0002, "epoch": 3.6011779554059737, "step": 4280}, {"loss": 1.5654, "grad_norm": 0.5691978931427002, "learning_rate": 0.0002, "epoch": 3.609591922591502, "step": 4290}, {"loss": 1.4854, "grad_norm": 0.6071329116821289, "learning_rate": 0.0002, "epoch": 3.61800588977703, "step": 4300}, {"loss": 1.5336, "grad_norm": 0.606573224067688, "learning_rate": 0.0002, "epoch": 3.626419856962558, "step": 4310}, {"loss": 1.6437, "grad_norm": 0.5515419244766235, "learning_rate": 0.0002, "epoch": 3.6348338241480858, "step": 4320}, {"loss": 1.498, "grad_norm": 0.5964660048484802, "learning_rate": 0.0002, "epoch": 3.643247791333614, "step": 4330}, {"loss": 1.544, "grad_norm": 0.5774146914482117, "learning_rate": 0.0002, "epoch": 3.6516617585191415, "step": 4340}, {"loss": 1.5566, "grad_norm": 0.5732731223106384, "learning_rate": 0.0002, "epoch": 3.6600757257046697, "step": 4350}, {"loss": 1.5682, "grad_norm": 0.7354163527488708, "learning_rate": 0.0002, "epoch": 3.6684896928901978, "step": 4360}, {"loss": 1.5225, "grad_norm": 0.6220902800559998, "learning_rate": 0.0002, "epoch": 3.676903660075726, "step": 4370}, {"loss": 1.4838, "grad_norm": 0.6053991317749023, "learning_rate": 0.0002, "epoch": 3.6853176272612536, "step": 4380}, {"loss": 1.5161, "grad_norm": 0.67010897397995, "learning_rate": 0.0002, "epoch": 3.6937315944467817, "step": 4390}, {"loss": 1.5381, "grad_norm": 0.6139186024665833, "learning_rate": 0.0002, "epoch": 3.70214556163231, "step": 4400}, {"loss": 1.5088, "grad_norm": 0.5433071851730347, "learning_rate": 0.0002, "epoch": 3.7105595288178375, "step": 4410}, {"loss": 1.5337, "grad_norm": 0.5453870296478271, "learning_rate": 0.0002, "epoch": 3.7189734960033656, "step": 4420}, {"loss": 1.4549, "grad_norm": 0.6401727199554443, "learning_rate": 0.0002, "epoch": 3.7273874631888937, "step": 4430}, {"loss": 1.503, "grad_norm": 0.6049367189407349, "learning_rate": 0.0002, "epoch": 3.735801430374422, "step": 4440}, {"loss": 1.5268, "grad_norm": 0.5740529298782349, "learning_rate": 0.0002, "epoch": 3.7442153975599495, "step": 4450}, {"loss": 1.5183, "grad_norm": 0.6521880626678467, "learning_rate": 0.0002, "epoch": 3.7526293647454776, "step": 4460}, {"loss": 1.5741, "grad_norm": 0.7096368074417114, "learning_rate": 0.0002, "epoch": 3.7610433319310053, "step": 4470}, {"loss": 1.5786, "grad_norm": 0.5886474251747131, "learning_rate": 0.0002, "epoch": 3.7694572991165334, "step": 4480}, {"loss": 1.5887, "grad_norm": 0.5821043252944946, "learning_rate": 0.0002, "epoch": 3.7778712663020615, "step": 4490}, {"loss": 1.5777, "grad_norm": 0.628892183303833, "learning_rate": 0.0002, "epoch": 3.7862852334875896, "step": 4500}, {"loss": 1.4708, "grad_norm": 0.5962669849395752, "learning_rate": 0.0002, "epoch": 3.7946992006731173, "step": 4510}, {"loss": 1.5267, "grad_norm": 0.6635549068450928, "learning_rate": 0.0002, "epoch": 3.8031131678586454, "step": 4520}, {"loss": 1.5058, "grad_norm": 0.6010760068893433, "learning_rate": 0.0002, "epoch": 3.811527135044173, "step": 4530}, {"loss": 1.6228, "grad_norm": 0.6322658658027649, "learning_rate": 0.0002, "epoch": 3.819941102229701, "step": 4540}, {"loss": 1.5029, "grad_norm": 0.5893137454986572, "learning_rate": 0.0002, "epoch": 3.8283550694152293, "step": 4550}, {"loss": 1.5435, "grad_norm": 0.7829602360725403, "learning_rate": 0.0002, "epoch": 3.8367690366007574, "step": 4560}, {"loss": 1.5453, "grad_norm": 0.6190396547317505, "learning_rate": 0.0002, "epoch": 3.845183003786285, "step": 4570}, {"loss": 1.5292, "grad_norm": 0.6662813425064087, "learning_rate": 0.0002, "epoch": 3.853596970971813, "step": 4580}, {"loss": 1.5065, "grad_norm": 0.5809855461120605, "learning_rate": 0.0002, "epoch": 3.8620109381573413, "step": 4590}, {"loss": 1.5041, "grad_norm": 0.5779069662094116, "learning_rate": 0.0002, "epoch": 3.870424905342869, "step": 4600}, {"loss": 1.498, "grad_norm": 0.5603038668632507, "learning_rate": 0.0002, "epoch": 3.878838872528397, "step": 4610}, {"loss": 1.5372, "grad_norm": 0.6274181008338928, "learning_rate": 0.0002, "epoch": 3.887252839713925, "step": 4620}, {"loss": 1.4996, "grad_norm": 0.6810959577560425, "learning_rate": 0.0002, "epoch": 3.8956668068994533, "step": 4630}, {"loss": 1.4956, "grad_norm": 0.5647315979003906, "learning_rate": 0.0002, "epoch": 3.904080774084981, "step": 4640}, {"loss": 1.5424, "grad_norm": 0.6830295324325562, "learning_rate": 0.0002, "epoch": 3.912494741270509, "step": 4650}, {"loss": 1.535, "grad_norm": 0.652565598487854, "learning_rate": 0.0002, "epoch": 3.920908708456037, "step": 4660}, {"loss": 1.4772, "grad_norm": 0.5806284546852112, "learning_rate": 0.0002, "epoch": 3.929322675641565, "step": 4670}, {"loss": 1.5812, "grad_norm": 0.6825073957443237, "learning_rate": 0.0002, "epoch": 3.937736642827093, "step": 4680}, {"loss": 1.5516, "grad_norm": 0.6149451732635498, "learning_rate": 0.0002, "epoch": 3.946150610012621, "step": 4690}, {"loss": 1.5608, "grad_norm": 0.6152557134628296, "learning_rate": 0.0002, "epoch": 3.954564577198149, "step": 4700}, {"loss": 1.4897, "grad_norm": 0.6239011883735657, "learning_rate": 0.0002, "epoch": 3.962978544383677, "step": 4710}, {"loss": 1.538, "grad_norm": 0.6485443115234375, "learning_rate": 0.0002, "epoch": 3.971392511569205, "step": 4720}, {"loss": 1.5226, "grad_norm": 0.6449228525161743, "learning_rate": 0.0002, "epoch": 3.9798064787547327, "step": 4730}, {"loss": 1.5087, "grad_norm": 0.6526407599449158, "learning_rate": 0.0002, "epoch": 3.988220445940261, "step": 4740}, {"loss": 1.5026, "grad_norm": 0.6277706027030945, "learning_rate": 0.0002, "epoch": 3.996634413125789, "step": 4750}, {"eval_loss": 1.871641755104065, "eval_runtime": 37.9637, "eval_samples_per_second": 13.566, "eval_steps_per_second": 1.712, "epoch": 4.0, "step": 4754}, {"loss": 1.4744, "grad_norm": 0.6994837522506714, "learning_rate": 0.0002, "epoch": 4.005048380311317, "step": 4760}, {"loss": 1.4433, "grad_norm": 0.8728373050689697, "learning_rate": 0.0002, "epoch": 4.013462347496845, "step": 4770}, {"loss": 1.3329, "grad_norm": 0.688679575920105, "learning_rate": 0.0002, "epoch": 4.021876314682372, "step": 4780}, {"loss": 1.3999, "grad_norm": 0.6313387155532837, "learning_rate": 0.0002, "epoch": 4.0302902818679005, "step": 4790}, {"loss": 1.3346, "grad_norm": 0.6577984690666199, "learning_rate": 0.0002, "epoch": 4.038704249053429, "step": 4800}, {"loss": 1.3403, "grad_norm": 0.7938185930252075, "learning_rate": 0.0002, "epoch": 4.047118216238957, "step": 4810}, {"loss": 1.3716, "grad_norm": 0.760399580001831, "learning_rate": 0.0002, "epoch": 4.055532183424485, "step": 4820}, {"loss": 1.4321, "grad_norm": 0.7329602241516113, "learning_rate": 0.0002, "epoch": 4.063946150610013, "step": 4830}, {"loss": 1.4133, "grad_norm": 0.7778576016426086, "learning_rate": 0.0002, "epoch": 4.07236011779554, "step": 4840}, {"loss": 1.4372, "grad_norm": 0.8235865235328674, "learning_rate": 0.0002, "epoch": 4.080774084981068, "step": 4850}, {"loss": 1.3719, "grad_norm": 0.7743754386901855, "learning_rate": 0.0002, "epoch": 4.089188052166596, "step": 4860}, {"loss": 1.3787, "grad_norm": 0.8145367503166199, "learning_rate": 0.0002, "epoch": 4.0976020193521245, "step": 4870}, {"loss": 1.356, "grad_norm": 0.8517307639122009, "learning_rate": 0.0002, "epoch": 4.106015986537653, "step": 4880}, {"loss": 1.4191, "grad_norm": 0.8208953142166138, "learning_rate": 0.0002, "epoch": 4.114429953723181, "step": 4890}, {"loss": 1.3189, "grad_norm": 0.8437790870666504, "learning_rate": 0.0002, "epoch": 4.122843920908709, "step": 4900}, {"loss": 1.3987, "grad_norm": 0.716672420501709, "learning_rate": 0.0002, "epoch": 4.131257888094236, "step": 4910}, {"loss": 1.4392, "grad_norm": 0.7656235098838806, "learning_rate": 0.0002, "epoch": 4.139671855279764, "step": 4920}, {"loss": 1.3408, "grad_norm": 0.7209306955337524, "learning_rate": 0.0002, "epoch": 4.148085822465292, "step": 4930}, {"loss": 1.3639, "grad_norm": 0.7731267809867859, "learning_rate": 0.0002, "epoch": 4.1564997896508205, "step": 4940}, {"loss": 1.4151, "grad_norm": 0.7477553486824036, "learning_rate": 0.0002, "epoch": 4.164913756836349, "step": 4950}, {"loss": 1.3485, "grad_norm": 0.7372981309890747, "learning_rate": 0.0002, "epoch": 4.173327724021877, "step": 4960}, {"loss": 1.3901, "grad_norm": 0.6582154035568237, "learning_rate": 0.0002, "epoch": 4.181741691207404, "step": 4970}, {"loss": 1.3343, "grad_norm": 0.7003206610679626, "learning_rate": 0.0002, "epoch": 4.190155658392932, "step": 4980}, {"loss": 1.4098, "grad_norm": 0.735223650932312, "learning_rate": 0.0002, "epoch": 4.19856962557846, "step": 4990}, {"loss": 1.3564, "grad_norm": 0.7832302451133728, "learning_rate": 0.0002, "epoch": 4.206983592763988, "step": 5000}, {"loss": 1.3622, "grad_norm": 0.8819546103477478, "learning_rate": 0.0002, "epoch": 4.215397559949516, "step": 5010}, {"loss": 1.4438, "grad_norm": 0.9325336813926697, "learning_rate": 0.0002, "epoch": 4.2238115271350445, "step": 5020}, {"loss": 1.3886, "grad_norm": 0.7007517218589783, "learning_rate": 0.0002, "epoch": 4.232225494320572, "step": 5030}, {"loss": 1.3683, "grad_norm": 0.7118321061134338, "learning_rate": 0.0002, "epoch": 4.2406394615061, "step": 5040}, {"loss": 1.2365, "grad_norm": 0.6578946709632874, "learning_rate": 0.0002, "epoch": 4.249053428691628, "step": 5050}, {"loss": 1.3696, "grad_norm": 0.9438983798027039, "learning_rate": 0.0002, "epoch": 4.257467395877156, "step": 5060}, {"loss": 1.3868, "grad_norm": 0.703037679195404, "learning_rate": 0.0002, "epoch": 4.265881363062684, "step": 5070}, {"loss": 1.3687, "grad_norm": 0.7286025285720825, "learning_rate": 0.0002, "epoch": 4.274295330248212, "step": 5080}, {"loss": 1.3605, "grad_norm": 0.750689685344696, "learning_rate": 0.0002, "epoch": 4.28270929743374, "step": 5090}, {"loss": 1.5089, "grad_norm": 0.869753360748291, "learning_rate": 0.0002, "epoch": 4.291123264619268, "step": 5100}, {"loss": 1.4128, "grad_norm": 0.8712980151176453, "learning_rate": 0.0002, "epoch": 4.299537231804796, "step": 5110}, {"loss": 1.3977, "grad_norm": 0.690263569355011, "learning_rate": 0.0002, "epoch": 4.307951198990324, "step": 5120}, {"loss": 1.4088, "grad_norm": 0.7114760279655457, "learning_rate": 0.0002, "epoch": 4.316365166175852, "step": 5130}, {"loss": 1.363, "grad_norm": 0.7588112354278564, "learning_rate": 0.0002, "epoch": 4.32477913336138, "step": 5140}, {"loss": 1.4408, "grad_norm": 0.7556202411651611, "learning_rate": 0.0002, "epoch": 4.333193100546908, "step": 5150}, {"loss": 1.4203, "grad_norm": 0.8357610702514648, "learning_rate": 0.0002, "epoch": 4.341607067732435, "step": 5160}, {"loss": 1.3348, "grad_norm": 0.8054035902023315, "learning_rate": 0.0002, "epoch": 4.3500210349179635, "step": 5170}, {"loss": 1.3109, "grad_norm": 0.7637107968330383, "learning_rate": 0.0002, "epoch": 4.358435002103492, "step": 5180}, {"loss": 1.3744, "grad_norm": 0.757481038570404, "learning_rate": 0.0002, "epoch": 4.36684896928902, "step": 5190}, {"loss": 1.3622, "grad_norm": 0.7185863852500916, "learning_rate": 0.0002, "epoch": 4.375262936474548, "step": 5200}, {"loss": 1.3896, "grad_norm": 0.7326455116271973, "learning_rate": 0.0002, "epoch": 4.383676903660076, "step": 5210}, {"loss": 1.4098, "grad_norm": 0.7980523109436035, "learning_rate": 0.0002, "epoch": 4.392090870845603, "step": 5220}, {"loss": 1.3783, "grad_norm": 0.8526999354362488, "learning_rate": 0.0002, "epoch": 4.400504838031131, "step": 5230}, {"loss": 1.4022, "grad_norm": 0.7012337446212769, "learning_rate": 0.0002, "epoch": 4.4089188052166595, "step": 5240}, {"loss": 1.3552, "grad_norm": 0.8217827677726746, "learning_rate": 0.0002, "epoch": 4.417332772402188, "step": 5250}, {"loss": 1.3482, "grad_norm": 0.7141005396842957, "learning_rate": 0.0002, "epoch": 4.425746739587716, "step": 5260}, {"loss": 1.3699, "grad_norm": 0.7094302177429199, "learning_rate": 0.0002, "epoch": 4.434160706773244, "step": 5270}, {"loss": 1.3527, "grad_norm": 0.7234613299369812, "learning_rate": 0.0002, "epoch": 4.442574673958772, "step": 5280}, {"loss": 1.4769, "grad_norm": 0.7530457973480225, "learning_rate": 0.0002, "epoch": 4.450988641144299, "step": 5290}, {"loss": 1.3944, "grad_norm": 0.7300912141799927, "learning_rate": 0.0002, "epoch": 4.459402608329827, "step": 5300}, {"loss": 1.3844, "grad_norm": 0.825443685054779, "learning_rate": 0.0002, "epoch": 4.467816575515355, "step": 5310}, {"loss": 1.3648, "grad_norm": 0.7559658885002136, "learning_rate": 0.0002, "epoch": 4.4762305427008835, "step": 5320}, {"loss": 1.4364, "grad_norm": 0.8817561268806458, "learning_rate": 0.0002, "epoch": 4.484644509886412, "step": 5330}, {"loss": 1.3618, "grad_norm": 0.8203575611114502, "learning_rate": 0.0002, "epoch": 4.49305847707194, "step": 5340}, {"loss": 1.3996, "grad_norm": 0.7677690982818604, "learning_rate": 0.0002, "epoch": 4.501472444257468, "step": 5350}, {"loss": 1.4142, "grad_norm": 0.657085120677948, "learning_rate": 0.0002, "epoch": 4.509886411442995, "step": 5360}, {"loss": 1.3722, "grad_norm": 0.7939504384994507, "learning_rate": 0.0002, "epoch": 4.518300378628523, "step": 5370}, {"loss": 1.4361, "grad_norm": 0.6971889138221741, "learning_rate": 0.0002, "epoch": 4.526714345814051, "step": 5380}, {"loss": 1.3637, "grad_norm": 0.6984175443649292, "learning_rate": 0.0002, "epoch": 4.535128312999579, "step": 5390}, {"loss": 1.341, "grad_norm": 0.8504858613014221, "learning_rate": 0.0002, "epoch": 4.5435422801851075, "step": 5400}, {"loss": 1.4026, "grad_norm": 0.9134073853492737, "learning_rate": 0.0002, "epoch": 4.551956247370635, "step": 5410}, {"loss": 1.4375, "grad_norm": 0.7765598893165588, "learning_rate": 0.0002, "epoch": 4.560370214556163, "step": 5420}, {"loss": 1.4832, "grad_norm": 0.6991009712219238, "learning_rate": 0.0002, "epoch": 4.568784181741691, "step": 5430}, {"loss": 1.4021, "grad_norm": 0.8393039107322693, "learning_rate": 0.0002, "epoch": 4.577198148927219, "step": 5440}, {"loss": 1.3976, "grad_norm": 0.7685918211936951, "learning_rate": 0.0002, "epoch": 4.585612116112747, "step": 5450}, {"loss": 1.3883, "grad_norm": 0.7135679721832275, "learning_rate": 0.0002, "epoch": 4.594026083298275, "step": 5460}, {"loss": 1.4083, "grad_norm": 0.6728870868682861, "learning_rate": 0.0002, "epoch": 4.6024400504838034, "step": 5470}, {"loss": 1.3698, "grad_norm": 0.7139479517936707, "learning_rate": 0.0002, "epoch": 4.610854017669331, "step": 5480}, {"loss": 1.3498, "grad_norm": 0.8476598858833313, "learning_rate": 0.0002, "epoch": 4.619267984854859, "step": 5490}, {"loss": 1.3389, "grad_norm": 0.8034361004829407, "learning_rate": 0.0002, "epoch": 4.627681952040387, "step": 5500}, {"loss": 1.4179, "grad_norm": 0.7452183961868286, "learning_rate": 0.0002, "epoch": 4.636095919225915, "step": 5510}, {"loss": 1.4031, "grad_norm": 0.8394148945808411, "learning_rate": 0.0002, "epoch": 4.644509886411443, "step": 5520}, {"loss": 1.4561, "grad_norm": 0.7480153441429138, "learning_rate": 0.0002, "epoch": 4.652923853596971, "step": 5530}, {"loss": 1.378, "grad_norm": 0.7781714797019958, "learning_rate": 0.0002, "epoch": 4.661337820782499, "step": 5540}, {"loss": 1.3924, "grad_norm": 1.0058213472366333, "learning_rate": 0.0002, "epoch": 4.669751787968027, "step": 5550}, {"loss": 1.4198, "grad_norm": 0.7403179407119751, "learning_rate": 0.0002, "epoch": 4.678165755153555, "step": 5560}, {"loss": 1.4328, "grad_norm": 0.7270476818084717, "learning_rate": 0.0002, "epoch": 4.686579722339083, "step": 5570}, {"loss": 1.378, "grad_norm": 0.760877788066864, "learning_rate": 0.0002, "epoch": 4.694993689524611, "step": 5580}, {"loss": 1.387, "grad_norm": 0.8097004890441895, "learning_rate": 0.0002, "epoch": 4.703407656710139, "step": 5590}, {"loss": 1.3661, "grad_norm": 0.9096523523330688, "learning_rate": 0.0002, "epoch": 4.711821623895667, "step": 5600}, {"loss": 1.4012, "grad_norm": 0.7262444496154785, "learning_rate": 0.0002, "epoch": 4.720235591081195, "step": 5610}, {"loss": 1.422, "grad_norm": 0.8207762837409973, "learning_rate": 0.0002, "epoch": 4.7286495582667225, "step": 5620}, {"loss": 1.4017, "grad_norm": 0.8089601993560791, "learning_rate": 0.0002, "epoch": 4.737063525452251, "step": 5630}, {"loss": 1.3675, "grad_norm": 0.7609543800354004, "learning_rate": 0.0002, "epoch": 4.745477492637779, "step": 5640}, {"loss": 1.4085, "grad_norm": 0.7273501753807068, "learning_rate": 0.0002, "epoch": 4.753891459823307, "step": 5650}, {"loss": 1.3849, "grad_norm": 0.7800219058990479, "learning_rate": 0.0002, "epoch": 4.762305427008835, "step": 5660}, {"loss": 1.4319, "grad_norm": 0.8558377623558044, "learning_rate": 0.0002, "epoch": 4.770719394194362, "step": 5670}, {"loss": 1.3831, "grad_norm": 0.7131547927856445, "learning_rate": 0.0002, "epoch": 4.77913336137989, "step": 5680}, {"loss": 1.407, "grad_norm": 0.7651025056838989, "learning_rate": 0.0002, "epoch": 4.787547328565418, "step": 5690}, {"loss": 1.3882, "grad_norm": 0.8129976391792297, "learning_rate": 0.0002, "epoch": 4.7959612957509465, "step": 5700}, {"loss": 1.4347, "grad_norm": 0.8019895553588867, "learning_rate": 0.0002, "epoch": 4.804375262936475, "step": 5710}, {"loss": 1.3961, "grad_norm": 0.7692018151283264, "learning_rate": 0.0002, "epoch": 4.812789230122003, "step": 5720}, {"loss": 1.419, "grad_norm": 0.6893943548202515, "learning_rate": 0.0002, "epoch": 4.821203197307531, "step": 5730}, {"loss": 1.4453, "grad_norm": 0.6881810426712036, "learning_rate": 0.0002, "epoch": 4.829617164493058, "step": 5740}, {"loss": 1.4775, "grad_norm": 0.7838267683982849, "learning_rate": 0.0002, "epoch": 4.838031131678586, "step": 5750}, {"loss": 1.3857, "grad_norm": 0.727799117565155, "learning_rate": 0.0002, "epoch": 4.846445098864114, "step": 5760}, {"loss": 1.4685, "grad_norm": 0.7458277344703674, "learning_rate": 0.0002, "epoch": 4.8548590660496425, "step": 5770}, {"loss": 1.4426, "grad_norm": 0.903802216053009, "learning_rate": 0.0002, "epoch": 4.863273033235171, "step": 5780}, {"loss": 1.451, "grad_norm": 0.7983472347259521, "learning_rate": 0.0002, "epoch": 4.871687000420699, "step": 5790}, {"loss": 1.4534, "grad_norm": 0.6894361972808838, "learning_rate": 0.0002, "epoch": 4.880100967606227, "step": 5800}, {"loss": 1.4486, "grad_norm": 0.7499409317970276, "learning_rate": 0.0002, "epoch": 4.888514934791754, "step": 5810}, {"loss": 1.4253, "grad_norm": 0.7362820506095886, "learning_rate": 0.0002, "epoch": 4.896928901977282, "step": 5820}, {"loss": 1.3763, "grad_norm": 0.8341619968414307, "learning_rate": 0.0002, "epoch": 4.90534286916281, "step": 5830}, {"loss": 1.3748, "grad_norm": 0.9604470133781433, "learning_rate": 0.0002, "epoch": 4.913756836348338, "step": 5840}, {"loss": 1.3658, "grad_norm": 0.8916844129562378, "learning_rate": 0.0002, "epoch": 4.9221708035338665, "step": 5850}, {"loss": 1.363, "grad_norm": 0.8519647121429443, "learning_rate": 0.0002, "epoch": 4.930584770719394, "step": 5860}, {"loss": 1.424, "grad_norm": 0.7946906089782715, "learning_rate": 0.0002, "epoch": 4.938998737904922, "step": 5870}, {"loss": 1.4071, "grad_norm": 0.7843789458274841, "learning_rate": 0.0002, "epoch": 4.94741270509045, "step": 5880}, {"loss": 1.4021, "grad_norm": 0.707618772983551, "learning_rate": 0.0002, "epoch": 4.955826672275978, "step": 5890}, {"loss": 1.502, "grad_norm": 0.7704206109046936, "learning_rate": 0.0002, "epoch": 4.964240639461506, "step": 5900}, {"loss": 1.4456, "grad_norm": 0.7160256505012512, "learning_rate": 0.0002, "epoch": 4.972654606647034, "step": 5910}, {"loss": 1.3874, "grad_norm": 0.7020420432090759, "learning_rate": 0.0002, "epoch": 4.981068573832562, "step": 5920}, {"loss": 1.4037, "grad_norm": 0.7576286792755127, "learning_rate": 0.0002, "epoch": 4.98948254101809, "step": 5930}, {"loss": 1.414, "grad_norm": 0.8573036789894104, "learning_rate": 0.0002, "epoch": 4.997896508203618, "step": 5940}, {"eval_loss": 1.9353811740875244, "eval_runtime": 37.9208, "eval_samples_per_second": 13.581, "eval_steps_per_second": 1.714, "epoch": 4.999579301640724, "step": 5942}, {"loss": 1.2418, "grad_norm": 0.8204267621040344, "learning_rate": 0.0002, "epoch": 5.006310475389146, "step": 5950}, {"loss": 1.235, "grad_norm": 0.976840615272522, "learning_rate": 0.0002, "epoch": 5.014724442574674, "step": 5960}, {"loss": 1.2134, "grad_norm": 0.8765613436698914, "learning_rate": 0.0002, "epoch": 5.023138409760202, "step": 5970}, {"loss": 1.2748, "grad_norm": 1.1793042421340942, "learning_rate": 0.0002, "epoch": 5.03155237694573, "step": 5980}, {"loss": 1.2412, "grad_norm": 0.971062958240509, "learning_rate": 0.0002, "epoch": 5.039966344131258, "step": 5990}, {"loss": 1.1819, "grad_norm": 0.8649757504463196, "learning_rate": 0.0002, "epoch": 5.0483803113167856, "step": 6000}, {"loss": 1.1654, "grad_norm": 0.9563034176826477, "learning_rate": 0.0002, "epoch": 5.056794278502314, "step": 6010}, {"loss": 1.2238, "grad_norm": 1.0093994140625, "learning_rate": 0.0002, "epoch": 5.065208245687842, "step": 6020}, {"loss": 1.2519, "grad_norm": 1.004213571548462, "learning_rate": 0.0002, "epoch": 5.07362221287337, "step": 6030}, {"loss": 1.2379, "grad_norm": 0.8307787179946899, "learning_rate": 0.0002, "epoch": 5.082036180058898, "step": 6040}, {"loss": 1.2282, "grad_norm": 0.9117848873138428, "learning_rate": 0.0002, "epoch": 5.090450147244426, "step": 6050}, {"loss": 1.2582, "grad_norm": 1.0269840955734253, "learning_rate": 0.0002, "epoch": 5.098864114429953, "step": 6060}, {"loss": 1.1836, "grad_norm": 0.9079542756080627, "learning_rate": 0.0002, "epoch": 5.1072780816154815, "step": 6070}, {"loss": 1.215, "grad_norm": 0.885702908039093, "learning_rate": 0.0002, "epoch": 5.11569204880101, "step": 6080}, {"loss": 1.2406, "grad_norm": 0.9976128339767456, "learning_rate": 0.0002, "epoch": 5.124106015986538, "step": 6090}, {"loss": 1.3082, "grad_norm": 0.8472117185592651, "learning_rate": 0.0002, "epoch": 5.132519983172066, "step": 6100}, {"loss": 1.226, "grad_norm": 1.0385161638259888, "learning_rate": 0.0002, "epoch": 5.140933950357594, "step": 6110}, {"loss": 1.213, "grad_norm": 0.8948383927345276, "learning_rate": 0.0002, "epoch": 5.149347917543121, "step": 6120}, {"loss": 1.2213, "grad_norm": 1.2613716125488281, "learning_rate": 0.0002, "epoch": 5.157761884728649, "step": 6130}, {"loss": 1.2632, "grad_norm": 0.9933410286903381, "learning_rate": 0.0002, "epoch": 5.166175851914177, "step": 6140}, {"loss": 1.1715, "grad_norm": 0.9673663973808289, "learning_rate": 0.0002, "epoch": 5.1745898190997055, "step": 6150}, {"loss": 1.2947, "grad_norm": 0.9969648122787476, "learning_rate": 0.0002, "epoch": 5.183003786285234, "step": 6160}, {"loss": 1.2416, "grad_norm": 1.2163258790969849, "learning_rate": 0.0002, "epoch": 5.191417753470762, "step": 6170}, {"loss": 1.2221, "grad_norm": 0.9163419604301453, "learning_rate": 0.0002, "epoch": 5.19983172065629, "step": 6180}, {"loss": 1.2624, "grad_norm": 0.9225585460662842, "learning_rate": 0.0002, "epoch": 5.208245687841817, "step": 6190}, {"loss": 1.2932, "grad_norm": 0.9205296635627747, "learning_rate": 0.0002, "epoch": 5.216659655027345, "step": 6200}, {"loss": 1.1825, "grad_norm": 1.0655443668365479, "learning_rate": 0.0002, "epoch": 5.225073622212873, "step": 6210}, {"loss": 1.2613, "grad_norm": 1.0854865312576294, "learning_rate": 0.0002, "epoch": 5.233487589398401, "step": 6220}, {"loss": 1.3045, "grad_norm": 0.8489186763763428, "learning_rate": 0.0002, "epoch": 5.2419015565839295, "step": 6230}, {"loss": 1.2708, "grad_norm": 0.910391628742218, "learning_rate": 0.0002, "epoch": 5.250315523769458, "step": 6240}, {"loss": 1.1914, "grad_norm": 0.925507128238678, "learning_rate": 0.0002, "epoch": 5.258729490954985, "step": 6250}, {"loss": 1.3368, "grad_norm": 1.1069735288619995, "learning_rate": 0.0002, "epoch": 5.267143458140513, "step": 6260}, {"loss": 1.2505, "grad_norm": 0.9705119132995605, "learning_rate": 0.0002, "epoch": 5.275557425326041, "step": 6270}, {"loss": 1.2602, "grad_norm": 0.9752426147460938, "learning_rate": 0.0002, "epoch": 5.283971392511569, "step": 6280}, {"loss": 1.2043, "grad_norm": 1.021359920501709, "learning_rate": 0.0002, "epoch": 5.292385359697097, "step": 6290}, {"loss": 1.2848, "grad_norm": 1.148606300354004, "learning_rate": 0.0002, "epoch": 5.3007993268826255, "step": 6300}, {"loss": 1.2201, "grad_norm": 0.8909247517585754, "learning_rate": 0.0002, "epoch": 5.309213294068153, "step": 6310}, {"loss": 1.2376, "grad_norm": 0.9879156351089478, "learning_rate": 0.0002, "epoch": 5.317627261253681, "step": 6320}, {"loss": 1.2638, "grad_norm": 0.9473357200622559, "learning_rate": 0.0002, "epoch": 5.326041228439209, "step": 6330}, {"loss": 1.232, "grad_norm": 1.1422028541564941, "learning_rate": 0.0002, "epoch": 5.334455195624737, "step": 6340}, {"loss": 1.263, "grad_norm": 0.9942235350608826, "learning_rate": 0.0002, "epoch": 5.342869162810265, "step": 6350}, {"loss": 1.3032, "grad_norm": 0.9535723924636841, "learning_rate": 0.0002, "epoch": 5.351283129995793, "step": 6360}, {"loss": 1.2908, "grad_norm": 0.9020892381668091, "learning_rate": 0.0002, "epoch": 5.359697097181321, "step": 6370}, {"loss": 1.2023, "grad_norm": 1.0626472234725952, "learning_rate": 0.0002, "epoch": 5.368111064366849, "step": 6380}, {"loss": 1.2555, "grad_norm": 1.1395848989486694, "learning_rate": 0.0002, "epoch": 5.376525031552377, "step": 6390}, {"loss": 1.2839, "grad_norm": 0.9274451732635498, "learning_rate": 0.0002, "epoch": 5.384938998737905, "step": 6400}, {"loss": 1.2819, "grad_norm": 0.8108699917793274, "learning_rate": 0.0002, "epoch": 5.393352965923433, "step": 6410}, {"loss": 1.2589, "grad_norm": 1.1805564165115356, "learning_rate": 0.0002, "epoch": 5.401766933108961, "step": 6420}, {"loss": 1.3549, "grad_norm": 0.8321298360824585, "learning_rate": 0.0002, "epoch": 5.410180900294489, "step": 6430}, {"loss": 1.2925, "grad_norm": 0.8981925249099731, "learning_rate": 0.0002, "epoch": 5.418594867480017, "step": 6440}, {"loss": 1.258, "grad_norm": 1.0730986595153809, "learning_rate": 0.0002, "epoch": 5.4270088346655445, "step": 6450}, {"loss": 1.26, "grad_norm": 1.0584609508514404, "learning_rate": 0.0002, "epoch": 5.435422801851073, "step": 6460}, {"loss": 1.2847, "grad_norm": 1.0792299509048462, "learning_rate": 0.0002, "epoch": 5.443836769036601, "step": 6470}, {"loss": 1.2035, "grad_norm": 0.9101872444152832, "learning_rate": 0.0002, "epoch": 5.452250736222129, "step": 6480}, {"loss": 1.2574, "grad_norm": 0.9910100698471069, "learning_rate": 0.0002, "epoch": 5.460664703407657, "step": 6490}, {"loss": 1.3098, "grad_norm": 1.041412353515625, "learning_rate": 0.0002, "epoch": 5.469078670593185, "step": 6500}, {"loss": 1.2812, "grad_norm": 1.0091687440872192, "learning_rate": 0.0002, "epoch": 5.477492637778712, "step": 6510}, {"loss": 1.2523, "grad_norm": 0.8755383491516113, "learning_rate": 0.0002, "epoch": 5.48590660496424, "step": 6520}, {"loss": 1.3042, "grad_norm": 0.980212390422821, "learning_rate": 0.0002, "epoch": 5.4943205721497685, "step": 6530}, {"loss": 1.2873, "grad_norm": 0.9356869459152222, "learning_rate": 0.0002, "epoch": 5.502734539335297, "step": 6540}, {"loss": 1.2254, "grad_norm": 0.9008095264434814, "learning_rate": 0.0002, "epoch": 5.511148506520825, "step": 6550}, {"loss": 1.2818, "grad_norm": 0.8908938765525818, "learning_rate": 0.0002, "epoch": 5.519562473706353, "step": 6560}, {"loss": 1.2212, "grad_norm": 1.1423932313919067, "learning_rate": 0.0002, "epoch": 5.52797644089188, "step": 6570}, {"loss": 1.3039, "grad_norm": 1.0508161783218384, "learning_rate": 0.0002, "epoch": 5.536390408077408, "step": 6580}, {"loss": 1.2446, "grad_norm": 0.8357517719268799, "learning_rate": 0.0002, "epoch": 5.544804375262936, "step": 6590}, {"loss": 1.3037, "grad_norm": 0.9892540574073792, "learning_rate": 0.0002, "epoch": 5.5532183424484645, "step": 6600}, {"loss": 1.3028, "grad_norm": 1.0048326253890991, "learning_rate": 0.0002, "epoch": 5.561632309633993, "step": 6610}, {"loss": 1.2152, "grad_norm": 0.9801995158195496, "learning_rate": 0.0002, "epoch": 5.570046276819521, "step": 6620}, {"loss": 1.2606, "grad_norm": 0.9899214506149292, "learning_rate": 0.0002, "epoch": 5.578460244005049, "step": 6630}, {"loss": 1.2043, "grad_norm": 1.1911814212799072, "learning_rate": 0.0002, "epoch": 5.586874211190576, "step": 6640}, {"loss": 1.3458, "grad_norm": 1.0368894338607788, "learning_rate": 0.0002, "epoch": 5.595288178376104, "step": 6650}, {"loss": 1.2595, "grad_norm": 1.1248382329940796, "learning_rate": 0.0002, "epoch": 5.603702145561632, "step": 6660}, {"loss": 1.2548, "grad_norm": 0.9765539765357971, "learning_rate": 0.0002, "epoch": 5.61211611274716, "step": 6670}, {"loss": 1.3451, "grad_norm": 0.9810206890106201, "learning_rate": 0.0002, "epoch": 5.6205300799326885, "step": 6680}, {"loss": 1.2952, "grad_norm": 1.100386619567871, "learning_rate": 0.0002, "epoch": 5.628944047118217, "step": 6690}, {"loss": 1.2467, "grad_norm": 0.8824519515037537, "learning_rate": 0.0002, "epoch": 5.637358014303744, "step": 6700}, {"loss": 1.25, "grad_norm": 1.0864064693450928, "learning_rate": 0.0002, "epoch": 5.645771981489272, "step": 6710}, {"loss": 1.2479, "grad_norm": 1.1614511013031006, "learning_rate": 0.0002, "epoch": 5.6541859486748, "step": 6720}, {"loss": 1.2753, "grad_norm": 1.0762972831726074, "learning_rate": 0.0002, "epoch": 5.662599915860328, "step": 6730}, {"loss": 1.2741, "grad_norm": 0.9408974647521973, "learning_rate": 0.0002, "epoch": 5.671013883045856, "step": 6740}, {"loss": 1.2431, "grad_norm": 0.8906030058860779, "learning_rate": 0.0002, "epoch": 5.679427850231384, "step": 6750}, {"loss": 1.2643, "grad_norm": 0.9527303576469421, "learning_rate": 0.0002, "epoch": 5.687841817416912, "step": 6760}, {"loss": 1.322, "grad_norm": 0.9471196532249451, "learning_rate": 0.0002, "epoch": 5.69625578460244, "step": 6770}, {"loss": 1.2514, "grad_norm": 0.9186838865280151, "learning_rate": 0.0002, "epoch": 5.704669751787968, "step": 6780}, {"loss": 1.2347, "grad_norm": 0.9225441813468933, "learning_rate": 0.0002, "epoch": 5.713083718973496, "step": 6790}, {"loss": 1.1849, "grad_norm": 0.9712982773780823, "learning_rate": 0.0002, "epoch": 5.721497686159024, "step": 6800}, {"loss": 1.2431, "grad_norm": 1.0743170976638794, "learning_rate": 0.0002, "epoch": 5.729911653344552, "step": 6810}, {"loss": 1.2136, "grad_norm": 1.2738113403320312, "learning_rate": 0.0002, "epoch": 5.73832562053008, "step": 6820}, {"loss": 1.2176, "grad_norm": 0.9386790990829468, "learning_rate": 0.0002, "epoch": 5.7467395877156076, "step": 6830}, {"loss": 1.285, "grad_norm": 1.0817769765853882, "learning_rate": 0.0002, "epoch": 5.755153554901136, "step": 6840}, {"loss": 1.2247, "grad_norm": 1.1040263175964355, "learning_rate": 0.0002, "epoch": 5.763567522086664, "step": 6850}, {"loss": 1.2507, "grad_norm": 1.0656492710113525, "learning_rate": 0.0002, "epoch": 5.771981489272192, "step": 6860}, {"loss": 1.2999, "grad_norm": 0.9550157189369202, "learning_rate": 0.0002, "epoch": 5.78039545645772, "step": 6870}, {"loss": 1.3201, "grad_norm": 1.0130870342254639, "learning_rate": 0.0002, "epoch": 5.788809423643248, "step": 6880}, {"loss": 1.3392, "grad_norm": 1.0675787925720215, "learning_rate": 0.0002, "epoch": 5.797223390828776, "step": 6890}, {"loss": 1.2949, "grad_norm": 0.9537774920463562, "learning_rate": 0.0002, "epoch": 5.8056373580143035, "step": 6900}, {"loss": 1.2658, "grad_norm": 0.9640319347381592, "learning_rate": 0.0002, "epoch": 5.814051325199832, "step": 6910}, {"loss": 1.2199, "grad_norm": 0.8917992115020752, "learning_rate": 0.0002, "epoch": 5.82246529238536, "step": 6920}, {"loss": 1.373, "grad_norm": 0.9881822466850281, "learning_rate": 0.0002, "epoch": 5.830879259570888, "step": 6930}, {"loss": 1.323, "grad_norm": 0.9136882424354553, "learning_rate": 0.0002, "epoch": 5.839293226756416, "step": 6940}, {"loss": 1.3159, "grad_norm": 0.9086098074913025, "learning_rate": 0.0002, "epoch": 5.847707193941943, "step": 6950}, {"loss": 1.2624, "grad_norm": 0.9443018436431885, "learning_rate": 0.0002, "epoch": 5.856121161127471, "step": 6960}, {"loss": 1.3224, "grad_norm": 0.9915381669998169, "learning_rate": 0.0002, "epoch": 5.864535128312999, "step": 6970}, {"loss": 1.337, "grad_norm": 0.8939146995544434, "learning_rate": 0.0002, "epoch": 5.8729490954985275, "step": 6980}, {"loss": 1.2611, "grad_norm": 1.3672245740890503, "learning_rate": 0.0002, "epoch": 5.881363062684056, "step": 6990}, {"loss": 1.3012, "grad_norm": 1.0116257667541504, "learning_rate": 0.0002, "epoch": 5.889777029869584, "step": 7000}, {"loss": 1.3128, "grad_norm": 1.1561565399169922, "learning_rate": 0.0002, "epoch": 5.898190997055112, "step": 7010}, {"loss": 1.2301, "grad_norm": 0.9900678992271423, "learning_rate": 0.0002, "epoch": 5.906604964240639, "step": 7020}, {"loss": 1.2845, "grad_norm": 0.9297345876693726, "learning_rate": 0.0002, "epoch": 5.915018931426167, "step": 7030}, {"loss": 1.2317, "grad_norm": 0.9357825517654419, "learning_rate": 0.0002, "epoch": 5.923432898611695, "step": 7040}, {"loss": 1.2303, "grad_norm": 1.049317717552185, "learning_rate": 0.0002, "epoch": 5.931846865797223, "step": 7050}, {"loss": 1.3243, "grad_norm": 0.950633704662323, "learning_rate": 0.0002, "epoch": 5.9402608329827515, "step": 7060}, {"loss": 1.2758, "grad_norm": 0.854581892490387, "learning_rate": 0.0002, "epoch": 5.94867480016828, "step": 7070}, {"loss": 1.3252, "grad_norm": 0.9097039699554443, "learning_rate": 0.0002, "epoch": 5.957088767353808, "step": 7080}, {"loss": 1.291, "grad_norm": 0.9072173237800598, "learning_rate": 0.0002, "epoch": 5.965502734539335, "step": 7090}, {"loss": 1.2724, "grad_norm": 1.0470727682113647, "learning_rate": 0.0002, "epoch": 5.973916701724863, "step": 7100}, {"loss": 1.3324, "grad_norm": 1.2628462314605713, "learning_rate": 0.0002, "epoch": 5.982330668910391, "step": 7110}, {"loss": 1.2701, "grad_norm": 1.055279016494751, "learning_rate": 0.0002, "epoch": 5.990744636095919, "step": 7120}, {"loss": 1.3234, "grad_norm": 0.966194212436676, "learning_rate": 0.0002, "epoch": 5.9991586032814475, "step": 7130}, {"eval_loss": 2.0427448749542236, "eval_runtime": 37.8426, "eval_samples_per_second": 13.609, "eval_steps_per_second": 1.718, "epoch": 6.0, "step": 7131}, {"loss": 1.1308, "grad_norm": 1.4037928581237793, "learning_rate": 0.0002, "epoch": 6.007572570466976, "step": 7140}, {"loss": 1.047, "grad_norm": 1.1081010103225708, "learning_rate": 0.0002, "epoch": 6.015986537652503, "step": 7150}, {"loss": 1.1368, "grad_norm": 1.1585499048233032, "learning_rate": 0.0002, "epoch": 6.024400504838031, "step": 7160}, {"loss": 1.0192, "grad_norm": 1.0822780132293701, "learning_rate": 0.0002, "epoch": 6.032814472023559, "step": 7170}, {"loss": 1.0755, "grad_norm": 0.9662094712257385, "learning_rate": 0.0002, "epoch": 6.041228439209087, "step": 7180}, {"loss": 1.1366, "grad_norm": 1.063936710357666, "learning_rate": 0.0002, "epoch": 6.049642406394615, "step": 7190}, {"loss": 1.0121, "grad_norm": 1.0349032878875732, "learning_rate": 0.0002, "epoch": 6.058056373580143, "step": 7200}, {"loss": 1.0591, "grad_norm": 1.0312575101852417, "learning_rate": 0.0002, "epoch": 6.066470340765671, "step": 7210}, {"loss": 1.1824, "grad_norm": 1.1942846775054932, "learning_rate": 0.0002, "epoch": 6.074884307951199, "step": 7220}, {"loss": 1.1034, "grad_norm": 1.0816049575805664, "learning_rate": 0.0002, "epoch": 6.083298275136727, "step": 7230}, {"loss": 1.0859, "grad_norm": 0.9985513687133789, "learning_rate": 0.0002, "epoch": 6.091712242322255, "step": 7240}, {"loss": 1.0367, "grad_norm": 1.2573972940444946, "learning_rate": 0.0002, "epoch": 6.100126209507783, "step": 7250}, {"loss": 1.1051, "grad_norm": 1.1182395219802856, "learning_rate": 0.0002, "epoch": 6.108540176693311, "step": 7260}, {"loss": 1.1219, "grad_norm": 0.9679344296455383, "learning_rate": 0.0002, "epoch": 6.116954143878839, "step": 7270}, {"loss": 1.1192, "grad_norm": 1.0913981199264526, "learning_rate": 0.0002, "epoch": 6.1253681110643665, "step": 7280}, {"loss": 1.0411, "grad_norm": 1.1291013956069946, "learning_rate": 0.0002, "epoch": 6.133782078249895, "step": 7290}, {"loss": 1.0963, "grad_norm": 1.2679595947265625, "learning_rate": 0.0002, "epoch": 6.142196045435423, "step": 7300}, {"loss": 1.0875, "grad_norm": 1.2350026369094849, "learning_rate": 0.0002, "epoch": 6.150610012620951, "step": 7310}, {"loss": 1.1139, "grad_norm": 1.3213104009628296, "learning_rate": 0.0002, "epoch": 6.159023979806479, "step": 7320}, {"loss": 1.1167, "grad_norm": 1.1924850940704346, "learning_rate": 0.0002, "epoch": 6.167437946992007, "step": 7330}, {"loss": 1.1242, "grad_norm": 1.1890000104904175, "learning_rate": 0.0002, "epoch": 6.175851914177534, "step": 7340}, {"loss": 1.1341, "grad_norm": 1.3821455240249634, "learning_rate": 0.0002, "epoch": 6.184265881363062, "step": 7350}, {"loss": 1.0748, "grad_norm": 1.1217057704925537, "learning_rate": 0.0002, "epoch": 6.1926798485485905, "step": 7360}, {"loss": 1.159, "grad_norm": 1.2441548109054565, "learning_rate": 0.0002, "epoch": 6.201093815734119, "step": 7370}, {"loss": 1.1199, "grad_norm": 1.0837615728378296, "learning_rate": 0.0002, "epoch": 6.209507782919647, "step": 7380}, {"loss": 1.1641, "grad_norm": 1.164304256439209, "learning_rate": 0.0002, "epoch": 6.217921750105175, "step": 7390}, {"loss": 1.1325, "grad_norm": 1.3129467964172363, "learning_rate": 0.0002, "epoch": 6.226335717290702, "step": 7400}, {"loss": 1.1537, "grad_norm": 1.1938153505325317, "learning_rate": 0.0002, "epoch": 6.23474968447623, "step": 7410}, {"loss": 1.1238, "grad_norm": 1.4348443746566772, "learning_rate": 0.0002, "epoch": 6.243163651661758, "step": 7420}, {"loss": 1.0778, "grad_norm": 1.132301926612854, "learning_rate": 0.0002, "epoch": 6.2515776188472865, "step": 7430}, {"loss": 1.1148, "grad_norm": 1.136966586112976, "learning_rate": 0.0002, "epoch": 6.259991586032815, "step": 7440}, {"loss": 1.096, "grad_norm": 1.12801194190979, "learning_rate": 0.0002, "epoch": 6.268405553218343, "step": 7450}, {"loss": 1.0408, "grad_norm": 1.0246902704238892, "learning_rate": 0.0002, "epoch": 6.276819520403871, "step": 7460}, {"loss": 1.0389, "grad_norm": 1.1066974401474, "learning_rate": 0.0002, "epoch": 6.285233487589398, "step": 7470}, {"loss": 1.1589, "grad_norm": 1.012710690498352, "learning_rate": 0.0002, "epoch": 6.293647454774926, "step": 7480}, {"loss": 1.1049, "grad_norm": 1.2227119207382202, "learning_rate": 0.0002, "epoch": 6.302061421960454, "step": 7490}, {"loss": 1.1376, "grad_norm": 0.9736923575401306, "learning_rate": 0.0002, "epoch": 6.310475389145982, "step": 7500}, {"loss": 1.1017, "grad_norm": 1.2945268154144287, "learning_rate": 0.0002, "epoch": 6.3188893563315105, "step": 7510}, {"loss": 1.0724, "grad_norm": 1.1579312086105347, "learning_rate": 0.0002, "epoch": 6.327303323517039, "step": 7520}, {"loss": 1.0899, "grad_norm": 1.2404558658599854, "learning_rate": 0.0002, "epoch": 6.335717290702567, "step": 7530}, {"loss": 1.1635, "grad_norm": 1.4673258066177368, "learning_rate": 0.0002, "epoch": 6.344131257888094, "step": 7540}, {"loss": 1.128, "grad_norm": 1.2268997430801392, "learning_rate": 0.0002, "epoch": 6.352545225073622, "step": 7550}, {"loss": 1.0932, "grad_norm": 0.9772747159004211, "learning_rate": 0.0002, "epoch": 6.36095919225915, "step": 7560}, {"loss": 1.1214, "grad_norm": 1.0205204486846924, "learning_rate": 0.0002, "epoch": 6.369373159444678, "step": 7570}, {"loss": 1.1095, "grad_norm": 1.2227109670639038, "learning_rate": 0.0002, "epoch": 6.377787126630206, "step": 7580}, {"loss": 1.1115, "grad_norm": 1.0708507299423218, "learning_rate": 0.0002, "epoch": 6.3862010938157345, "step": 7590}, {"loss": 1.1018, "grad_norm": 1.1427522897720337, "learning_rate": 0.0002, "epoch": 6.394615061001262, "step": 7600}, {"loss": 1.1079, "grad_norm": 1.0706431865692139, "learning_rate": 0.0002, "epoch": 6.40302902818679, "step": 7610}, {"loss": 1.0933, "grad_norm": 1.1358282566070557, "learning_rate": 0.0002, "epoch": 6.411442995372318, "step": 7620}, {"loss": 1.1075, "grad_norm": 1.4011822938919067, "learning_rate": 0.0002, "epoch": 6.419856962557846, "step": 7630}, {"loss": 1.1269, "grad_norm": 1.5616450309753418, "learning_rate": 0.0002, "epoch": 6.428270929743374, "step": 7640}, {"loss": 1.0953, "grad_norm": 1.1442687511444092, "learning_rate": 0.0002, "epoch": 6.436684896928902, "step": 7650}, {"loss": 1.1341, "grad_norm": 1.164803147315979, "learning_rate": 0.0002, "epoch": 6.44509886411443, "step": 7660}, {"loss": 1.14, "grad_norm": 1.3184553384780884, "learning_rate": 0.0002, "epoch": 6.453512831299958, "step": 7670}, {"loss": 1.1526, "grad_norm": 1.2701894044876099, "learning_rate": 0.0002, "epoch": 6.461926798485486, "step": 7680}, {"loss": 1.2119, "grad_norm": 1.1998416185379028, "learning_rate": 0.0002, "epoch": 6.470340765671014, "step": 7690}, {"loss": 1.1528, "grad_norm": 1.156459927558899, "learning_rate": 0.0002, "epoch": 6.478754732856542, "step": 7700}, {"loss": 1.2122, "grad_norm": 1.0217190980911255, "learning_rate": 0.0002, "epoch": 6.48716870004207, "step": 7710}, {"loss": 1.0917, "grad_norm": 1.230372428894043, "learning_rate": 0.0002, "epoch": 6.495582667227598, "step": 7720}, {"loss": 1.119, "grad_norm": 1.105675220489502, "learning_rate": 0.0002, "epoch": 6.5039966344131255, "step": 7730}, {"loss": 1.0758, "grad_norm": 1.1623669862747192, "learning_rate": 0.0002, "epoch": 6.512410601598654, "step": 7740}, {"loss": 1.1548, "grad_norm": 1.2884684801101685, "learning_rate": 0.0002, "epoch": 6.520824568784182, "step": 7750}, {"loss": 1.142, "grad_norm": 1.1785279512405396, "learning_rate": 0.0002, "epoch": 6.52923853596971, "step": 7760}, {"loss": 1.1598, "grad_norm": 1.0607101917266846, "learning_rate": 0.0002, "epoch": 6.537652503155238, "step": 7770}, {"loss": 1.1472, "grad_norm": 1.21990168094635, "learning_rate": 0.0002, "epoch": 6.546066470340766, "step": 7780}, {"loss": 1.1468, "grad_norm": 1.1498621702194214, "learning_rate": 0.0002, "epoch": 6.554480437526293, "step": 7790}, {"loss": 1.1847, "grad_norm": 1.263929009437561, "learning_rate": 0.0002, "epoch": 6.562894404711821, "step": 7800}, {"loss": 1.1177, "grad_norm": 1.1580625772476196, "learning_rate": 0.0002, "epoch": 6.5713083718973495, "step": 7810}, {"loss": 1.1313, "grad_norm": 1.4431294202804565, "learning_rate": 0.0002, "epoch": 6.579722339082878, "step": 7820}, {"loss": 1.1944, "grad_norm": 1.1309990882873535, "learning_rate": 0.0002, "epoch": 6.588136306268406, "step": 7830}, {"loss": 1.1156, "grad_norm": 1.0543386936187744, "learning_rate": 0.0002, "epoch": 6.596550273453934, "step": 7840}, {"loss": 1.0945, "grad_norm": 1.2180639505386353, "learning_rate": 0.0002, "epoch": 6.604964240639461, "step": 7850}, {"loss": 1.1318, "grad_norm": 1.0631271600723267, "learning_rate": 0.0002, "epoch": 6.613378207824989, "step": 7860}, {"loss": 1.1792, "grad_norm": 1.138885498046875, "learning_rate": 0.0002, "epoch": 6.621792175010517, "step": 7870}, {"loss": 1.1805, "grad_norm": 1.1117745637893677, "learning_rate": 0.0002, "epoch": 6.630206142196045, "step": 7880}, {"loss": 1.15, "grad_norm": 1.3734886646270752, "learning_rate": 0.0002, "epoch": 6.6386201093815735, "step": 7890}, {"loss": 1.1584, "grad_norm": 1.236003041267395, "learning_rate": 0.0002, "epoch": 6.647034076567102, "step": 7900}, {"loss": 1.1718, "grad_norm": 1.2206000089645386, "learning_rate": 0.0002, "epoch": 6.65544804375263, "step": 7910}, {"loss": 1.1637, "grad_norm": 1.2842656373977661, "learning_rate": 0.0002, "epoch": 6.663862010938157, "step": 7920}, {"loss": 1.2219, "grad_norm": 1.2365005016326904, "learning_rate": 0.0002, "epoch": 6.672275978123685, "step": 7930}, {"loss": 1.0827, "grad_norm": 1.256620168685913, "learning_rate": 0.0002, "epoch": 6.680689945309213, "step": 7940}, {"loss": 1.1788, "grad_norm": 1.3232917785644531, "learning_rate": 0.0002, "epoch": 6.689103912494741, "step": 7950}, {"loss": 1.2042, "grad_norm": 1.2470088005065918, "learning_rate": 0.0002, "epoch": 6.6975178796802695, "step": 7960}, {"loss": 1.0959, "grad_norm": 1.0511926412582397, "learning_rate": 0.0002, "epoch": 6.705931846865798, "step": 7970}, {"loss": 1.118, "grad_norm": 1.107310175895691, "learning_rate": 0.0002, "epoch": 6.714345814051326, "step": 7980}, {"loss": 1.2109, "grad_norm": 1.4069843292236328, "learning_rate": 0.0002, "epoch": 6.722759781236853, "step": 7990}, {"loss": 1.1298, "grad_norm": 1.0800836086273193, "learning_rate": 0.0002, "epoch": 6.731173748422381, "step": 8000}, {"loss": 1.1824, "grad_norm": 1.1676300764083862, "learning_rate": 0.0002, "epoch": 6.739587715607909, "step": 8010}, {"loss": 1.1253, "grad_norm": 1.0579663515090942, "learning_rate": 0.0002, "epoch": 6.748001682793437, "step": 8020}, {"loss": 1.1542, "grad_norm": 1.2770029306411743, "learning_rate": 0.0002, "epoch": 6.756415649978965, "step": 8030}, {"loss": 1.1519, "grad_norm": 1.0981038808822632, "learning_rate": 0.0002, "epoch": 6.764829617164493, "step": 8040}, {"loss": 1.1422, "grad_norm": 1.1194742918014526, "learning_rate": 0.0002, "epoch": 6.773243584350021, "step": 8050}, {"loss": 1.1463, "grad_norm": 1.0130012035369873, "learning_rate": 0.0002, "epoch": 6.781657551535549, "step": 8060}, {"loss": 1.2008, "grad_norm": 1.2051167488098145, "learning_rate": 0.0002, "epoch": 6.790071518721077, "step": 8070}, {"loss": 1.142, "grad_norm": 1.095689058303833, "learning_rate": 0.0002, "epoch": 6.798485485906605, "step": 8080}, {"loss": 1.1352, "grad_norm": 1.2275174856185913, "learning_rate": 0.0002, "epoch": 6.806899453092133, "step": 8090}, {"loss": 1.1453, "grad_norm": 1.1439805030822754, "learning_rate": 0.0002, "epoch": 6.815313420277661, "step": 8100}, {"loss": 1.1624, "grad_norm": 1.276331901550293, "learning_rate": 0.0002, "epoch": 6.8237273874631885, "step": 8110}, {"loss": 1.1686, "grad_norm": 1.0450139045715332, "learning_rate": 0.0002, "epoch": 6.832141354648717, "step": 8120}, {"loss": 1.1783, "grad_norm": 1.1189453601837158, "learning_rate": 0.0002, "epoch": 6.840555321834245, "step": 8130}, {"loss": 1.1093, "grad_norm": 1.194640874862671, "learning_rate": 0.0002, "epoch": 6.848969289019773, "step": 8140}, {"loss": 1.1559, "grad_norm": 1.095372200012207, "learning_rate": 0.0002, "epoch": 6.857383256205301, "step": 8150}, {"loss": 1.165, "grad_norm": 1.2416104078292847, "learning_rate": 0.0002, "epoch": 6.865797223390829, "step": 8160}, {"loss": 1.2174, "grad_norm": 1.2402868270874023, "learning_rate": 0.0002, "epoch": 6.874211190576357, "step": 8170}, {"loss": 1.1306, "grad_norm": 1.1317291259765625, "learning_rate": 0.0002, "epoch": 6.882625157761884, "step": 8180}, {"loss": 1.1944, "grad_norm": 1.0581914186477661, "learning_rate": 0.0002, "epoch": 6.8910391249474126, "step": 8190}, {"loss": 1.1271, "grad_norm": 1.3540890216827393, "learning_rate": 0.0002, "epoch": 6.899453092132941, "step": 8200}, {"loss": 1.2119, "grad_norm": 1.213672399520874, "learning_rate": 0.0002, "epoch": 6.907867059318469, "step": 8210}, {"loss": 1.1406, "grad_norm": 1.2654485702514648, "learning_rate": 0.0002, "epoch": 6.916281026503997, "step": 8220}, {"loss": 1.205, "grad_norm": 1.203903317451477, "learning_rate": 0.0002, "epoch": 6.924694993689524, "step": 8230}, {"loss": 1.1635, "grad_norm": 1.1332030296325684, "learning_rate": 0.0002, "epoch": 6.933108960875052, "step": 8240}, {"loss": 1.1148, "grad_norm": 1.2699192762374878, "learning_rate": 0.0002, "epoch": 6.94152292806058, "step": 8250}, {"loss": 1.1831, "grad_norm": 1.2728958129882812, "learning_rate": 0.0002, "epoch": 6.9499368952461085, "step": 8260}, {"loss": 1.1757, "grad_norm": 1.238410472869873, "learning_rate": 0.0002, "epoch": 6.958350862431637, "step": 8270}, {"loss": 1.1499, "grad_norm": 1.403863549232483, "learning_rate": 0.0002, "epoch": 6.966764829617165, "step": 8280}, {"loss": 1.1515, "grad_norm": 1.1096396446228027, "learning_rate": 0.0002, "epoch": 6.975178796802693, "step": 8290}, {"loss": 1.2049, "grad_norm": 1.1043379306793213, "learning_rate": 0.0002, "epoch": 6.98359276398822, "step": 8300}, {"loss": 1.1255, "grad_norm": 1.391754388809204, "learning_rate": 0.0002, "epoch": 6.992006731173748, "step": 8310}]}
+{"epoch": 7.996634413125789, "step": 9504, "epoch_duration": 1245.9935760498047, "total_accumulated_duration": 11061.152487277985, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "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_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-0.8-num-5623-sd-42/checkpoint-2377", "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": 2.56, "grad_norm": 0.5458821654319763, "learning_rate": 0.0002, "epoch": 0.008413967185527976, "step": 10}, {"loss": 2.3235, "grad_norm": 0.7293308973312378, "learning_rate": 0.0002, "epoch": 0.016827934371055953, "step": 20}, {"loss": 2.0815, "grad_norm": 0.47792306542396545, "learning_rate": 0.0002, "epoch": 0.02524190155658393, "step": 30}, {"loss": 1.9718, "grad_norm": 0.5944402813911438, "learning_rate": 0.0002, "epoch": 0.033655868742111905, "step": 40}, {"loss": 1.8848, "grad_norm": 0.5415359735488892, "learning_rate": 0.0002, "epoch": 0.04206983592763988, "step": 50}, {"loss": 1.8953, "grad_norm": 0.535713791847229, "learning_rate": 0.0002, "epoch": 0.05048380311316786, "step": 60}, {"loss": 1.937, "grad_norm": 0.5184146761894226, "learning_rate": 0.0002, "epoch": 0.058897770298695834, "step": 70}, {"loss": 1.8396, "grad_norm": 0.458926796913147, "learning_rate": 0.0002, "epoch": 0.06731173748422381, "step": 80}, {"loss": 1.8677, "grad_norm": 0.4780142307281494, "learning_rate": 0.0002, "epoch": 0.07572570466975179, "step": 90}, {"loss": 1.8593, "grad_norm": 0.79965740442276, "learning_rate": 0.0002, "epoch": 0.08413967185527976, "step": 100}, {"loss": 1.9081, "grad_norm": 0.4498862028121948, "learning_rate": 0.0002, "epoch": 0.09255363904080774, "step": 110}, {"loss": 1.8503, "grad_norm": 0.39338430762290955, "learning_rate": 0.0002, "epoch": 0.10096760622633572, "step": 120}, {"loss": 1.8637, "grad_norm": 0.9588953852653503, "learning_rate": 0.0002, "epoch": 0.10938157341186369, "step": 130}, {"loss": 1.8676, "grad_norm": 0.41675639152526855, "learning_rate": 0.0002, "epoch": 0.11779554059739167, "step": 140}, {"loss": 1.8904, "grad_norm": 0.44519832730293274, "learning_rate": 0.0002, "epoch": 0.12620950778291964, "step": 150}, {"loss": 1.798, "grad_norm": 0.4176260530948639, "learning_rate": 0.0002, "epoch": 0.13462347496844762, "step": 160}, {"loss": 1.8398, "grad_norm": 0.35840365290641785, "learning_rate": 0.0002, "epoch": 0.1430374421539756, "step": 170}, {"loss": 1.8666, "grad_norm": 0.3794495463371277, "learning_rate": 0.0002, "epoch": 0.15145140933950357, "step": 180}, {"loss": 1.8111, "grad_norm": 0.4563522934913635, "learning_rate": 0.0002, "epoch": 0.15986537652503155, "step": 190}, {"loss": 1.8893, "grad_norm": 0.37057486176490784, "learning_rate": 0.0002, "epoch": 0.16827934371055953, "step": 200}, {"loss": 1.7995, "grad_norm": 0.44081518054008484, "learning_rate": 0.0002, "epoch": 0.1766933108960875, "step": 210}, {"loss": 1.9048, "grad_norm": 0.46078577637672424, "learning_rate": 0.0002, "epoch": 0.18510727808161548, "step": 220}, {"loss": 1.8403, "grad_norm": 0.36132094264030457, "learning_rate": 0.0002, "epoch": 0.19352124526714345, "step": 230}, {"loss": 1.8827, "grad_norm": 0.3747289180755615, "learning_rate": 0.0002, "epoch": 0.20193521245267143, "step": 240}, {"loss": 1.8382, "grad_norm": 0.3540179133415222, "learning_rate": 0.0002, "epoch": 0.2103491796381994, "step": 250}, {"loss": 1.8196, "grad_norm": 0.3461375832557678, "learning_rate": 0.0002, "epoch": 0.21876314682372738, "step": 260}, {"loss": 1.8509, "grad_norm": 0.3436960279941559, "learning_rate": 0.0002, "epoch": 0.22717711400925536, "step": 270}, {"loss": 1.8285, "grad_norm": 0.35403719544410706, "learning_rate": 0.0002, "epoch": 0.23559108119478334, "step": 280}, {"loss": 1.8369, "grad_norm": 0.37142616510391235, "learning_rate": 0.0002, "epoch": 0.2440050483803113, "step": 290}, {"loss": 1.8044, "grad_norm": 0.3307955861091614, "learning_rate": 0.0002, "epoch": 0.2524190155658393, "step": 300}, {"loss": 1.817, "grad_norm": 0.32855314016342163, "learning_rate": 0.0002, "epoch": 0.2608329827513673, "step": 310}, {"loss": 1.7803, "grad_norm": 0.3299003839492798, "learning_rate": 0.0002, "epoch": 0.26924694993689524, "step": 320}, {"loss": 1.8129, "grad_norm": 0.44311287999153137, "learning_rate": 0.0002, "epoch": 0.27766091712242325, "step": 330}, {"loss": 1.8232, "grad_norm": 0.32989758253097534, "learning_rate": 0.0002, "epoch": 0.2860748843079512, "step": 340}, {"loss": 1.7716, "grad_norm": 0.34400200843811035, "learning_rate": 0.0002, "epoch": 0.2944888514934792, "step": 350}, {"loss": 1.7619, "grad_norm": 0.36286211013793945, "learning_rate": 0.0002, "epoch": 0.30290281867900715, "step": 360}, {"loss": 1.8025, "grad_norm": 0.406827837228775, "learning_rate": 0.0002, "epoch": 0.31131678586453515, "step": 370}, {"loss": 1.7515, "grad_norm": 0.36299195885658264, "learning_rate": 0.0002, "epoch": 0.3197307530500631, "step": 380}, {"loss": 1.837, "grad_norm": 0.3477257192134857, "learning_rate": 0.0002, "epoch": 0.3281447202355911, "step": 390}, {"loss": 1.7767, "grad_norm": 0.3730369210243225, "learning_rate": 0.0002, "epoch": 0.33655868742111905, "step": 400}, {"loss": 1.7747, "grad_norm": 0.4644559919834137, "learning_rate": 0.0002, "epoch": 0.34497265460664706, "step": 410}, {"loss": 1.7538, "grad_norm": 0.406576544046402, "learning_rate": 0.0002, "epoch": 0.353386621792175, "step": 420}, {"loss": 1.7501, "grad_norm": 0.3612699508666992, "learning_rate": 0.0002, "epoch": 0.361800588977703, "step": 430}, {"loss": 1.7473, "grad_norm": 0.3243742287158966, "learning_rate": 0.0002, "epoch": 0.37021455616323096, "step": 440}, {"loss": 1.8851, "grad_norm": 0.36671221256256104, "learning_rate": 0.0002, "epoch": 0.37862852334875896, "step": 450}, {"loss": 1.8853, "grad_norm": 0.3565002381801605, "learning_rate": 0.0002, "epoch": 0.3870424905342869, "step": 460}, {"loss": 1.8923, "grad_norm": 0.34630221128463745, "learning_rate": 0.0002, "epoch": 0.3954564577198149, "step": 470}, {"loss": 1.8234, "grad_norm": 0.3353537321090698, "learning_rate": 0.0002, "epoch": 0.40387042490534286, "step": 480}, {"loss": 1.7135, "grad_norm": 0.4015921950340271, "learning_rate": 0.0002, "epoch": 0.41228439209087087, "step": 490}, {"loss": 1.7815, "grad_norm": 0.5489419102668762, "learning_rate": 0.0002, "epoch": 0.4206983592763988, "step": 500}, {"loss": 1.7903, "grad_norm": 0.4193589985370636, "learning_rate": 0.0002, "epoch": 0.4291123264619268, "step": 510}, {"loss": 1.8416, "grad_norm": 0.3418922424316406, "learning_rate": 0.0002, "epoch": 0.43752629364745477, "step": 520}, {"loss": 1.7982, "grad_norm": 0.32668185234069824, "learning_rate": 0.0002, "epoch": 0.44594026083298277, "step": 530}, {"loss": 1.7501, "grad_norm": 0.3094325661659241, "learning_rate": 0.0002, "epoch": 0.4543542280185107, "step": 540}, {"loss": 1.7438, "grad_norm": 0.3743017315864563, "learning_rate": 0.0002, "epoch": 0.4627681952040387, "step": 550}, {"loss": 1.8451, "grad_norm": 0.3295630216598511, "learning_rate": 0.0002, "epoch": 0.47118216238956667, "step": 560}, {"loss": 1.7529, "grad_norm": 1.6124513149261475, "learning_rate": 0.0002, "epoch": 0.4795961295750947, "step": 570}, {"loss": 1.8028, "grad_norm": 0.3245585858821869, "learning_rate": 0.0002, "epoch": 0.4880100967606226, "step": 580}, {"loss": 1.7976, "grad_norm": 0.3332934081554413, "learning_rate": 0.0002, "epoch": 0.49642406394615063, "step": 590}, {"loss": 1.7912, "grad_norm": 0.3836138844490051, "learning_rate": 0.0002, "epoch": 0.5048380311316786, "step": 600}, {"loss": 1.8347, "grad_norm": 0.32953888177871704, "learning_rate": 0.0002, "epoch": 0.5132519983172066, "step": 610}, {"loss": 1.7729, "grad_norm": 0.36291512846946716, "learning_rate": 0.0002, "epoch": 0.5216659655027346, "step": 620}, {"loss": 1.7758, "grad_norm": 0.3237783908843994, "learning_rate": 0.0002, "epoch": 0.5300799326882625, "step": 630}, {"loss": 1.8352, "grad_norm": 0.38882696628570557, "learning_rate": 0.0002, "epoch": 0.5384938998737905, "step": 640}, {"loss": 1.8624, "grad_norm": 0.37821972370147705, "learning_rate": 0.0002, "epoch": 0.5469078670593185, "step": 650}, {"loss": 1.8075, "grad_norm": 0.3556285500526428, "learning_rate": 0.0002, "epoch": 0.5553218342448465, "step": 660}, {"loss": 1.778, "grad_norm": 0.347499281167984, "learning_rate": 0.0002, "epoch": 0.5637358014303744, "step": 670}, {"loss": 1.8066, "grad_norm": 0.3176489472389221, "learning_rate": 0.0002, "epoch": 0.5721497686159024, "step": 680}, {"loss": 1.7257, "grad_norm": 0.30220088362693787, "learning_rate": 0.0002, "epoch": 0.5805637358014304, "step": 690}, {"loss": 1.8415, "grad_norm": 0.3711601793766022, "learning_rate": 0.0002, "epoch": 0.5889777029869584, "step": 700}, {"loss": 1.7906, "grad_norm": 0.3311759829521179, "learning_rate": 0.0002, "epoch": 0.5973916701724863, "step": 710}, {"loss": 1.7712, "grad_norm": 0.34824270009994507, "learning_rate": 0.0002, "epoch": 0.6058056373580143, "step": 720}, {"loss": 1.7954, "grad_norm": 0.29668381810188293, "learning_rate": 0.0002, "epoch": 0.6142196045435423, "step": 730}, {"loss": 1.8321, "grad_norm": 0.36087489128112793, "learning_rate": 0.0002, "epoch": 0.6226335717290703, "step": 740}, {"loss": 1.7956, "grad_norm": 0.31590089201927185, "learning_rate": 0.0002, "epoch": 0.6310475389145982, "step": 750}, {"loss": 1.7343, "grad_norm": 0.37632957100868225, "learning_rate": 0.0002, "epoch": 0.6394615061001262, "step": 760}, {"loss": 1.8499, "grad_norm": 0.3360748589038849, "learning_rate": 0.0002, "epoch": 0.6478754732856542, "step": 770}, {"loss": 1.8076, "grad_norm": 0.3420640528202057, "learning_rate": 0.0002, "epoch": 0.6562894404711822, "step": 780}, {"loss": 1.8353, "grad_norm": 0.5734959244728088, "learning_rate": 0.0002, "epoch": 0.6647034076567101, "step": 790}, {"loss": 1.7746, "grad_norm": 0.36440837383270264, "learning_rate": 0.0002, "epoch": 0.6731173748422381, "step": 800}, {"loss": 1.7532, "grad_norm": 0.3179708421230316, "learning_rate": 0.0002, "epoch": 0.6815313420277661, "step": 810}, {"loss": 1.7815, "grad_norm": 0.34122881293296814, "learning_rate": 0.0002, "epoch": 0.6899453092132941, "step": 820}, {"loss": 1.8167, "grad_norm": 0.31886112689971924, "learning_rate": 0.0002, "epoch": 0.698359276398822, "step": 830}, {"loss": 1.7505, "grad_norm": 0.31782326102256775, "learning_rate": 0.0002, "epoch": 0.70677324358435, "step": 840}, {"loss": 1.7588, "grad_norm": 0.36052989959716797, "learning_rate": 0.0002, "epoch": 0.715187210769878, "step": 850}, {"loss": 1.7891, "grad_norm": 0.28946155309677124, "learning_rate": 0.0002, "epoch": 0.723601177955406, "step": 860}, {"loss": 1.7923, "grad_norm": 0.3095663785934448, "learning_rate": 0.0002, "epoch": 0.7320151451409339, "step": 870}, {"loss": 1.785, "grad_norm": 0.3317491412162781, "learning_rate": 0.0002, "epoch": 0.7404291123264619, "step": 880}, {"loss": 1.7709, "grad_norm": 0.31324660778045654, "learning_rate": 0.0002, "epoch": 0.7488430795119899, "step": 890}, {"loss": 1.8753, "grad_norm": 0.3290475606918335, "learning_rate": 0.0002, "epoch": 0.7572570466975179, "step": 900}, {"loss": 1.7679, "grad_norm": 0.35690343379974365, "learning_rate": 0.0002, "epoch": 0.7656710138830458, "step": 910}, {"loss": 1.826, "grad_norm": 0.39558273553848267, "learning_rate": 0.0002, "epoch": 0.7740849810685738, "step": 920}, {"loss": 1.8722, "grad_norm": 0.34254348278045654, "learning_rate": 0.0002, "epoch": 0.7824989482541018, "step": 930}, {"loss": 1.7603, "grad_norm": 0.3560165464878082, "learning_rate": 0.0002, "epoch": 0.7909129154396298, "step": 940}, {"loss": 1.7992, "grad_norm": 0.30693164467811584, "learning_rate": 0.0002, "epoch": 0.7993268826251577, "step": 950}, {"loss": 1.8029, "grad_norm": 0.3394823372364044, "learning_rate": 0.0002, "epoch": 0.8077408498106857, "step": 960}, {"loss": 1.8105, "grad_norm": 0.3741514980792999, "learning_rate": 0.0002, "epoch": 0.8161548169962137, "step": 970}, {"loss": 1.7849, "grad_norm": 0.3655228316783905, "learning_rate": 0.0002, "epoch": 0.8245687841817417, "step": 980}, {"loss": 1.8449, "grad_norm": 0.3586033880710602, "learning_rate": 0.0002, "epoch": 0.8329827513672696, "step": 990}, {"loss": 1.7033, "grad_norm": 0.3459678888320923, "learning_rate": 0.0002, "epoch": 0.8413967185527976, "step": 1000}, {"loss": 1.8498, "grad_norm": 0.3184349834918976, "learning_rate": 0.0002, "epoch": 0.8498106857383256, "step": 1010}, {"loss": 1.7632, "grad_norm": 0.3099786043167114, "learning_rate": 0.0002, "epoch": 0.8582246529238536, "step": 1020}, {"loss": 1.8067, "grad_norm": 0.30300915241241455, "learning_rate": 0.0002, "epoch": 0.8666386201093815, "step": 1030}, {"loss": 1.7923, "grad_norm": 0.3128705620765686, "learning_rate": 0.0002, "epoch": 0.8750525872949095, "step": 1040}, {"loss": 1.8252, "grad_norm": 0.3336263597011566, "learning_rate": 0.0002, "epoch": 0.8834665544804375, "step": 1050}, {"loss": 1.8375, "grad_norm": 0.3801328241825104, "learning_rate": 0.0002, "epoch": 0.8918805216659655, "step": 1060}, {"loss": 1.7757, "grad_norm": 0.3122096359729767, "learning_rate": 0.0002, "epoch": 0.9002944888514934, "step": 1070}, {"loss": 1.8251, "grad_norm": 0.35990869998931885, "learning_rate": 0.0002, "epoch": 0.9087084560370214, "step": 1080}, {"loss": 1.7343, "grad_norm": 0.3321819305419922, "learning_rate": 0.0002, "epoch": 0.9171224232225494, "step": 1090}, {"loss": 1.7595, "grad_norm": 0.4202139377593994, "learning_rate": 0.0002, "epoch": 0.9255363904080774, "step": 1100}, {"loss": 1.8056, "grad_norm": 0.32559722661972046, "learning_rate": 0.0002, "epoch": 0.9339503575936053, "step": 1110}, {"loss": 1.812, "grad_norm": 0.3098459839820862, "learning_rate": 0.0002, "epoch": 0.9423643247791333, "step": 1120}, {"loss": 1.8252, "grad_norm": 0.33917108178138733, "learning_rate": 0.0002, "epoch": 0.9507782919646613, "step": 1130}, {"loss": 1.7709, "grad_norm": 0.4055837094783783, "learning_rate": 0.0002, "epoch": 0.9591922591501894, "step": 1140}, {"loss": 1.8259, "grad_norm": 0.32508623600006104, "learning_rate": 0.0002, "epoch": 0.9676062263357172, "step": 1150}, {"loss": 1.782, "grad_norm": 0.30150601267814636, "learning_rate": 0.0002, "epoch": 0.9760201935212452, "step": 1160}, {"loss": 1.8291, "grad_norm": 0.3042563199996948, "learning_rate": 0.0002, "epoch": 0.9844341607067733, "step": 1170}, {"loss": 1.7847, "grad_norm": 0.33254584670066833, "learning_rate": 0.0002, "epoch": 0.9928481278923013, "step": 1180}, {"eval_loss": 1.8077726364135742, "eval_runtime": 38.4359, "eval_samples_per_second": 13.399, "eval_steps_per_second": 1.691, "epoch": 0.9995793016407236, "step": 1188}, {"loss": 1.7414, "grad_norm": 0.35073035955429077, "learning_rate": 0.0002, "epoch": 1.0012620950778293, "step": 1190}, {"loss": 1.7483, "grad_norm": 0.3217269778251648, "learning_rate": 0.0002, "epoch": 1.0096760622633572, "step": 1200}, {"loss": 1.7517, "grad_norm": 0.3635033369064331, "learning_rate": 0.0002, "epoch": 1.018090029448885, "step": 1210}, {"loss": 1.6949, "grad_norm": 0.32468414306640625, "learning_rate": 0.0002, "epoch": 1.0265039966344132, "step": 1220}, {"loss": 1.711, "grad_norm": 0.3307163417339325, "learning_rate": 0.0002, "epoch": 1.034917963819941, "step": 1230}, {"loss": 1.7881, "grad_norm": 0.34381359815597534, "learning_rate": 0.0002, "epoch": 1.0433319310054692, "step": 1240}, {"loss": 1.612, "grad_norm": 0.35874804854393005, "learning_rate": 0.0002, "epoch": 1.051745898190997, "step": 1250}, {"loss": 1.7314, "grad_norm": 0.3615919351577759, "learning_rate": 0.0002, "epoch": 1.060159865376525, "step": 1260}, {"loss": 1.7517, "grad_norm": 0.32835808396339417, "learning_rate": 0.0002, "epoch": 1.068573832562053, "step": 1270}, {"loss": 1.7193, "grad_norm": 0.3876388370990753, "learning_rate": 0.0002, "epoch": 1.076987799747581, "step": 1280}, {"loss": 1.7442, "grad_norm": 0.39895930886268616, "learning_rate": 0.0002, "epoch": 1.0854017669331089, "step": 1290}, {"loss": 1.6601, "grad_norm": 0.39081698656082153, "learning_rate": 0.0002, "epoch": 1.093815734118637, "step": 1300}, {"loss": 1.7623, "grad_norm": 0.39974215626716614, "learning_rate": 0.0002, "epoch": 1.1022297013041649, "step": 1310}, {"loss": 1.7506, "grad_norm": 0.3887332081794739, "learning_rate": 0.0002, "epoch": 1.110643668489693, "step": 1320}, {"loss": 1.7381, "grad_norm": 0.36216408014297485, "learning_rate": 0.0002, "epoch": 1.1190576356752209, "step": 1330}, {"loss": 1.762, "grad_norm": 0.36979028582572937, "learning_rate": 0.0002, "epoch": 1.1274716028607488, "step": 1340}, {"loss": 1.7515, "grad_norm": 0.34052133560180664, "learning_rate": 0.0002, "epoch": 1.1358855700462769, "step": 1350}, {"loss": 1.7513, "grad_norm": 0.3467716574668884, "learning_rate": 0.0002, "epoch": 1.1442995372318048, "step": 1360}, {"loss": 1.7086, "grad_norm": 0.35528799891471863, "learning_rate": 0.0002, "epoch": 1.1527135044173327, "step": 1370}, {"loss": 1.794, "grad_norm": 0.36282262206077576, "learning_rate": 0.0002, "epoch": 1.1611274716028608, "step": 1380}, {"loss": 1.7731, "grad_norm": 0.37355899810791016, "learning_rate": 0.0002, "epoch": 1.1695414387883887, "step": 1390}, {"loss": 1.7483, "grad_norm": 0.37292736768722534, "learning_rate": 0.0002, "epoch": 1.1779554059739168, "step": 1400}, {"loss": 1.6916, "grad_norm": 0.5892812013626099, "learning_rate": 0.0002, "epoch": 1.1863693731594447, "step": 1410}, {"loss": 1.7302, "grad_norm": 0.3712292015552521, "learning_rate": 0.0002, "epoch": 1.1947833403449726, "step": 1420}, {"loss": 1.7709, "grad_norm": 0.3349577486515045, "learning_rate": 0.0002, "epoch": 1.2031973075305007, "step": 1430}, {"loss": 1.7412, "grad_norm": 0.32591062784194946, "learning_rate": 0.0002, "epoch": 1.2116112747160286, "step": 1440}, {"loss": 1.7406, "grad_norm": 0.3840635418891907, "learning_rate": 0.0002, "epoch": 1.2200252419015567, "step": 1450}, {"loss": 1.7276, "grad_norm": 0.37238365411758423, "learning_rate": 0.0002, "epoch": 1.2284392090870846, "step": 1460}, {"loss": 1.7052, "grad_norm": 0.3731217682361603, "learning_rate": 0.0002, "epoch": 1.2368531762726125, "step": 1470}, {"loss": 1.7255, "grad_norm": 0.3318967819213867, "learning_rate": 0.0002, "epoch": 1.2452671434581406, "step": 1480}, {"loss": 1.7463, "grad_norm": 0.3784034848213196, "learning_rate": 0.0002, "epoch": 1.2536811106436685, "step": 1490}, {"loss": 1.6862, "grad_norm": 0.3541383147239685, "learning_rate": 0.0002, "epoch": 1.2620950778291964, "step": 1500}, {"loss": 1.8394, "grad_norm": 0.35312485694885254, "learning_rate": 0.0002, "epoch": 1.2705090450147245, "step": 1510}, {"loss": 1.7029, "grad_norm": 0.35272929072380066, "learning_rate": 0.0002, "epoch": 1.2789230122002524, "step": 1520}, {"loss": 1.7016, "grad_norm": 0.40988272428512573, "learning_rate": 0.0002, "epoch": 1.2873369793857803, "step": 1530}, {"loss": 1.6912, "grad_norm": 0.3543946146965027, "learning_rate": 0.0002, "epoch": 1.2957509465713084, "step": 1540}, {"loss": 1.6757, "grad_norm": 0.35639145970344543, "learning_rate": 0.0002, "epoch": 1.3041649137568363, "step": 1550}, {"loss": 1.6814, "grad_norm": 0.3290826678276062, "learning_rate": 0.0002, "epoch": 1.3125788809423642, "step": 1560}, {"loss": 1.7369, "grad_norm": 0.39264336228370667, "learning_rate": 0.0002, "epoch": 1.3209928481278923, "step": 1570}, {"loss": 1.6804, "grad_norm": 0.5390415191650391, "learning_rate": 0.0002, "epoch": 1.3294068153134202, "step": 1580}, {"loss": 1.708, "grad_norm": 0.5188116431236267, "learning_rate": 0.0002, "epoch": 1.3378207824989483, "step": 1590}, {"loss": 1.6763, "grad_norm": 0.37445148825645447, "learning_rate": 0.0002, "epoch": 1.3462347496844762, "step": 1600}, {"loss": 1.7386, "grad_norm": 0.3296085298061371, "learning_rate": 0.0002, "epoch": 1.3546487168700043, "step": 1610}, {"loss": 1.8107, "grad_norm": 0.39879581332206726, "learning_rate": 0.0002, "epoch": 1.3630626840555322, "step": 1620}, {"loss": 1.6744, "grad_norm": 0.36092764139175415, "learning_rate": 0.0002, "epoch": 1.37147665124106, "step": 1630}, {"loss": 1.7144, "grad_norm": 0.37011823058128357, "learning_rate": 0.0002, "epoch": 1.3798906184265882, "step": 1640}, {"loss": 1.7396, "grad_norm": 0.40863534808158875, "learning_rate": 0.0002, "epoch": 1.3883045856121161, "step": 1650}, {"loss": 1.7901, "grad_norm": 0.337001770734787, "learning_rate": 0.0002, "epoch": 1.396718552797644, "step": 1660}, {"loss": 1.7044, "grad_norm": 0.35596707463264465, "learning_rate": 0.0002, "epoch": 1.4051325199831721, "step": 1670}, {"loss": 1.7717, "grad_norm": 0.3857671916484833, "learning_rate": 0.0002, "epoch": 1.4135464871687, "step": 1680}, {"loss": 1.7015, "grad_norm": 0.419502317905426, "learning_rate": 0.0002, "epoch": 1.421960454354228, "step": 1690}, {"loss": 1.7261, "grad_norm": 0.35459452867507935, "learning_rate": 0.0002, "epoch": 1.430374421539756, "step": 1700}, {"loss": 1.7361, "grad_norm": 0.37246978282928467, "learning_rate": 0.0002, "epoch": 1.438788388725284, "step": 1710}, {"loss": 1.6762, "grad_norm": 0.33091893792152405, "learning_rate": 0.0002, "epoch": 1.4472023559108118, "step": 1720}, {"loss": 1.7044, "grad_norm": 0.37029674649238586, "learning_rate": 0.0002, "epoch": 1.45561632309634, "step": 1730}, {"loss": 1.7117, "grad_norm": 0.374025821685791, "learning_rate": 0.0002, "epoch": 1.4640302902818678, "step": 1740}, {"loss": 1.7549, "grad_norm": 0.3416315019130707, "learning_rate": 0.0002, "epoch": 1.472444257467396, "step": 1750}, {"loss": 1.7093, "grad_norm": 0.36502841114997864, "learning_rate": 0.0002, "epoch": 1.4808582246529238, "step": 1760}, {"loss": 1.6597, "grad_norm": 0.35458803176879883, "learning_rate": 0.0002, "epoch": 1.489272191838452, "step": 1770}, {"loss": 1.675, "grad_norm": 0.4462839663028717, "learning_rate": 0.0002, "epoch": 1.4976861590239798, "step": 1780}, {"loss": 1.7267, "grad_norm": 0.34836092591285706, "learning_rate": 0.0002, "epoch": 1.5061001262095077, "step": 1790}, {"loss": 1.7295, "grad_norm": 0.3445749282836914, "learning_rate": 0.0002, "epoch": 1.5145140933950358, "step": 1800}, {"loss": 1.7386, "grad_norm": 0.36012160778045654, "learning_rate": 0.0002, "epoch": 1.5229280605805637, "step": 1810}, {"loss": 1.6594, "grad_norm": 0.4052616059780121, "learning_rate": 0.0002, "epoch": 1.5313420277660916, "step": 1820}, {"loss": 1.72, "grad_norm": 0.3966905474662781, "learning_rate": 0.0002, "epoch": 1.5397559949516197, "step": 1830}, {"loss": 1.7595, "grad_norm": 0.35028719902038574, "learning_rate": 0.0002, "epoch": 1.5481699621371476, "step": 1840}, {"loss": 1.6829, "grad_norm": 0.3936742842197418, "learning_rate": 0.0002, "epoch": 1.5565839293226755, "step": 1850}, {"loss": 1.7579, "grad_norm": 0.34473296999931335, "learning_rate": 0.0002, "epoch": 1.5649978965082036, "step": 1860}, {"loss": 1.7207, "grad_norm": 0.4328365623950958, "learning_rate": 0.0002, "epoch": 1.5734118636937318, "step": 1870}, {"loss": 1.7098, "grad_norm": 0.3566315472126007, "learning_rate": 0.0002, "epoch": 1.5818258308792594, "step": 1880}, {"loss": 1.6095, "grad_norm": 0.3301256597042084, "learning_rate": 0.0002, "epoch": 1.5902397980647875, "step": 1890}, {"loss": 1.748, "grad_norm": 0.3743041455745697, "learning_rate": 0.0002, "epoch": 1.5986537652503157, "step": 1900}, {"loss": 1.7259, "grad_norm": 0.3735344707965851, "learning_rate": 0.0002, "epoch": 1.6070677324358436, "step": 1910}, {"loss": 1.7445, "grad_norm": 0.42191144824028015, "learning_rate": 0.0002, "epoch": 1.6154816996213714, "step": 1920}, {"loss": 1.6978, "grad_norm": 0.3787207305431366, "learning_rate": 0.0002, "epoch": 1.6238956668068996, "step": 1930}, {"loss": 1.6893, "grad_norm": 0.35647350549697876, "learning_rate": 0.0002, "epoch": 1.6323096339924275, "step": 1940}, {"loss": 1.7825, "grad_norm": 0.39791446924209595, "learning_rate": 0.0002, "epoch": 1.6407236011779553, "step": 1950}, {"loss": 1.7293, "grad_norm": 0.37341275811195374, "learning_rate": 0.0002, "epoch": 1.6491375683634835, "step": 1960}, {"loss": 1.6781, "grad_norm": 0.3722686469554901, "learning_rate": 0.0002, "epoch": 1.6575515355490114, "step": 1970}, {"loss": 1.6383, "grad_norm": 0.37467387318611145, "learning_rate": 0.0002, "epoch": 1.6659655027345392, "step": 1980}, {"loss": 1.7439, "grad_norm": 0.37109461426734924, "learning_rate": 0.0002, "epoch": 1.6743794699200674, "step": 1990}, {"loss": 1.7206, "grad_norm": 0.4008837044239044, "learning_rate": 0.0002, "epoch": 1.6827934371055953, "step": 2000}, {"loss": 1.7604, "grad_norm": 0.3316999673843384, "learning_rate": 0.0002, "epoch": 1.6912074042911232, "step": 2010}, {"loss": 1.7325, "grad_norm": 0.3683805465698242, "learning_rate": 0.0002, "epoch": 1.6996213714766513, "step": 2020}, {"loss": 1.7451, "grad_norm": 0.4163658320903778, "learning_rate": 0.0002, "epoch": 1.7080353386621794, "step": 2030}, {"loss": 1.741, "grad_norm": 0.4245431125164032, "learning_rate": 0.0002, "epoch": 1.716449305847707, "step": 2040}, {"loss": 1.7184, "grad_norm": 0.36732038855552673, "learning_rate": 0.0002, "epoch": 1.7248632730332352, "step": 2050}, {"loss": 1.7031, "grad_norm": 0.34981656074523926, "learning_rate": 0.0002, "epoch": 1.7332772402187633, "step": 2060}, {"loss": 1.7545, "grad_norm": 0.38588812947273254, "learning_rate": 0.0002, "epoch": 1.7416912074042912, "step": 2070}, {"loss": 1.7728, "grad_norm": 0.39914557337760925, "learning_rate": 0.0002, "epoch": 1.750105174589819, "step": 2080}, {"loss": 1.7049, "grad_norm": 0.36068692803382874, "learning_rate": 0.0002, "epoch": 1.7585191417753472, "step": 2090}, {"loss": 1.7537, "grad_norm": 0.3983287215232849, "learning_rate": 0.0002, "epoch": 1.766933108960875, "step": 2100}, {"loss": 1.7016, "grad_norm": 0.45008400082588196, "learning_rate": 0.0002, "epoch": 1.775347076146403, "step": 2110}, {"loss": 1.7163, "grad_norm": 0.3618052303791046, "learning_rate": 0.0002, "epoch": 1.783761043331931, "step": 2120}, {"loss": 1.7335, "grad_norm": 0.38745400309562683, "learning_rate": 0.0002, "epoch": 1.792175010517459, "step": 2130}, {"loss": 1.7387, "grad_norm": 0.3413826525211334, "learning_rate": 0.0002, "epoch": 1.8005889777029869, "step": 2140}, {"loss": 1.7414, "grad_norm": 0.35983747243881226, "learning_rate": 0.0002, "epoch": 1.809002944888515, "step": 2150}, {"loss": 1.7892, "grad_norm": 0.40926849842071533, "learning_rate": 0.0002, "epoch": 1.8174169120740429, "step": 2160}, {"loss": 1.6823, "grad_norm": 0.3543093800544739, "learning_rate": 0.0002, "epoch": 1.8258308792595708, "step": 2170}, {"loss": 1.7812, "grad_norm": 0.42690935730934143, "learning_rate": 0.0002, "epoch": 1.8342448464450989, "step": 2180}, {"loss": 1.7471, "grad_norm": 0.40282756090164185, "learning_rate": 0.0002, "epoch": 1.842658813630627, "step": 2190}, {"loss": 1.7411, "grad_norm": 0.36568400263786316, "learning_rate": 0.0002, "epoch": 1.8510727808161547, "step": 2200}, {"loss": 1.7024, "grad_norm": 0.43159013986587524, "learning_rate": 0.0002, "epoch": 1.8594867480016828, "step": 2210}, {"loss": 1.7298, "grad_norm": 0.3554118573665619, "learning_rate": 0.0002, "epoch": 1.867900715187211, "step": 2220}, {"loss": 1.7157, "grad_norm": 0.43349072337150574, "learning_rate": 0.0002, "epoch": 1.8763146823727388, "step": 2230}, {"loss": 1.7302, "grad_norm": 0.36486536264419556, "learning_rate": 0.0002, "epoch": 1.8847286495582667, "step": 2240}, {"loss": 1.6901, "grad_norm": 0.39260047674179077, "learning_rate": 0.0002, "epoch": 1.8931426167437948, "step": 2250}, {"loss": 1.6691, "grad_norm": 0.3741776943206787, "learning_rate": 0.0002, "epoch": 1.9015565839293227, "step": 2260}, {"loss": 1.6931, "grad_norm": 0.3961946964263916, "learning_rate": 0.0002, "epoch": 1.9099705511148506, "step": 2270}, {"loss": 1.737, "grad_norm": 0.3659731149673462, "learning_rate": 0.0002, "epoch": 1.9183845183003787, "step": 2280}, {"loss": 1.7342, "grad_norm": 0.34744107723236084, "learning_rate": 0.0002, "epoch": 1.9267984854859066, "step": 2290}, {"loss": 1.7162, "grad_norm": 0.3607442378997803, "learning_rate": 0.0002, "epoch": 1.9352124526714345, "step": 2300}, {"loss": 1.6673, "grad_norm": 0.331464558839798, "learning_rate": 0.0002, "epoch": 1.9436264198569626, "step": 2310}, {"loss": 1.7101, "grad_norm": 0.3904414474964142, "learning_rate": 0.0002, "epoch": 1.9520403870424905, "step": 2320}, {"loss": 1.7327, "grad_norm": 0.37584832310676575, "learning_rate": 0.0002, "epoch": 1.9604543542280184, "step": 2330}, {"loss": 1.7586, "grad_norm": 0.3698684275150299, "learning_rate": 0.0002, "epoch": 1.9688683214135465, "step": 2340}, {"loss": 1.7764, "grad_norm": 0.40571412444114685, "learning_rate": 0.0002, "epoch": 1.9772822885990746, "step": 2350}, {"loss": 1.744, "grad_norm": 0.40059587359428406, "learning_rate": 0.0002, "epoch": 1.9856962557846023, "step": 2360}, {"loss": 1.7033, "grad_norm": 0.4168248474597931, "learning_rate": 0.0002, "epoch": 1.9941102229701304, "step": 2370}, {"eval_loss": 1.8055059909820557, "eval_runtime": 38.422, "eval_samples_per_second": 13.404, "eval_steps_per_second": 1.692, "epoch": 2.0, "step": 2377}, {"loss": 1.7673, "grad_norm": 0.35205352306365967, "learning_rate": 0.0002, "epoch": 2.0025241901556585, "step": 2380}, {"loss": 1.6556, "grad_norm": 0.3979377746582031, "learning_rate": 0.0002, "epoch": 2.010938157341186, "step": 2390}, {"loss": 1.6421, "grad_norm": 0.396491676568985, "learning_rate": 0.0002, "epoch": 2.0193521245267143, "step": 2400}, {"loss": 1.6847, "grad_norm": 0.44712209701538086, "learning_rate": 0.0002, "epoch": 2.0277660917122424, "step": 2410}, {"loss": 1.6877, "grad_norm": 0.4454420208930969, "learning_rate": 0.0002, "epoch": 2.03618005889777, "step": 2420}, {"loss": 1.6635, "grad_norm": 0.4170038402080536, "learning_rate": 0.0002, "epoch": 2.044594026083298, "step": 2430}, {"loss": 1.6512, "grad_norm": 0.4309595227241516, "learning_rate": 0.0002, "epoch": 2.0530079932688263, "step": 2440}, {"loss": 1.6223, "grad_norm": 0.4241602122783661, "learning_rate": 0.0002, "epoch": 2.0614219604543544, "step": 2450}, {"loss": 1.6162, "grad_norm": 0.4370540678501129, "learning_rate": 0.0002, "epoch": 2.069835927639882, "step": 2460}, {"loss": 1.6354, "grad_norm": 0.43985554575920105, "learning_rate": 0.0002, "epoch": 2.0782498948254102, "step": 2470}, {"loss": 1.6954, "grad_norm": 0.4158105254173279, "learning_rate": 0.0002, "epoch": 2.0866638620109383, "step": 2480}, {"loss": 1.6114, "grad_norm": 0.441549152135849, "learning_rate": 0.0002, "epoch": 2.095077829196466, "step": 2490}, {"loss": 1.5485, "grad_norm": 0.385718435049057, "learning_rate": 0.0002, "epoch": 2.103491796381994, "step": 2500}, {"loss": 1.5894, "grad_norm": 0.43146514892578125, "learning_rate": 0.0002, "epoch": 2.1119057635675222, "step": 2510}, {"loss": 1.6414, "grad_norm": 0.41663315892219543, "learning_rate": 0.0002, "epoch": 2.12031973075305, "step": 2520}, {"loss": 1.6527, "grad_norm": 0.4410698115825653, "learning_rate": 0.0002, "epoch": 2.128733697938578, "step": 2530}, {"loss": 1.6124, "grad_norm": 0.4472278952598572, "learning_rate": 0.0002, "epoch": 2.137147665124106, "step": 2540}, {"loss": 1.6257, "grad_norm": 0.3879167437553406, "learning_rate": 0.0002, "epoch": 2.145561632309634, "step": 2550}, {"loss": 1.6682, "grad_norm": 0.4212203025817871, "learning_rate": 0.0002, "epoch": 2.153975599495162, "step": 2560}, {"loss": 1.6036, "grad_norm": 0.42841723561286926, "learning_rate": 0.0002, "epoch": 2.16238956668069, "step": 2570}, {"loss": 1.5962, "grad_norm": 0.39272481203079224, "learning_rate": 0.0002, "epoch": 2.1708035338662177, "step": 2580}, {"loss": 1.681, "grad_norm": 0.4075261354446411, "learning_rate": 0.0002, "epoch": 2.179217501051746, "step": 2590}, {"loss": 1.6601, "grad_norm": 0.5358437895774841, "learning_rate": 0.0002, "epoch": 2.187631468237274, "step": 2600}, {"loss": 1.6423, "grad_norm": 0.4738350212574005, "learning_rate": 0.0002, "epoch": 2.1960454354228016, "step": 2610}, {"loss": 1.6386, "grad_norm": 0.446789026260376, "learning_rate": 0.0002, "epoch": 2.2044594026083297, "step": 2620}, {"loss": 1.6246, "grad_norm": 0.4615374505519867, "learning_rate": 0.0002, "epoch": 2.212873369793858, "step": 2630}, {"loss": 1.6205, "grad_norm": 0.46901994943618774, "learning_rate": 0.0002, "epoch": 2.221287336979386, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.46267789602279663, "learning_rate": 0.0002, "epoch": 2.2297013041649136, "step": 2650}, {"loss": 1.6584, "grad_norm": 0.4383080005645752, "learning_rate": 0.0002, "epoch": 2.2381152713504417, "step": 2660}, {"loss": 1.5745, "grad_norm": 0.4070609509944916, "learning_rate": 0.0002, "epoch": 2.24652923853597, "step": 2670}, {"loss": 1.6125, "grad_norm": 0.4572339951992035, "learning_rate": 0.0002, "epoch": 2.2549432057214975, "step": 2680}, {"loss": 1.5671, "grad_norm": 0.393265038728714, "learning_rate": 0.0002, "epoch": 2.2633571729070256, "step": 2690}, {"loss": 1.6239, "grad_norm": 0.46144717931747437, "learning_rate": 0.0002, "epoch": 2.2717711400925538, "step": 2700}, {"loss": 1.5992, "grad_norm": 0.45077767968177795, "learning_rate": 0.0002, "epoch": 2.2801851072780814, "step": 2710}, {"loss": 1.6261, "grad_norm": 0.5697639584541321, "learning_rate": 0.0002, "epoch": 2.2885990744636096, "step": 2720}, {"loss": 1.6192, "grad_norm": 0.4855510890483856, "learning_rate": 0.0002, "epoch": 2.2970130416491377, "step": 2730}, {"loss": 1.7419, "grad_norm": 0.4440622627735138, "learning_rate": 0.0002, "epoch": 2.3054270088346653, "step": 2740}, {"loss": 1.6496, "grad_norm": 0.3904096782207489, "learning_rate": 0.0002, "epoch": 2.3138409760201935, "step": 2750}, {"loss": 1.5888, "grad_norm": 0.5225510597229004, "learning_rate": 0.0002, "epoch": 2.3222549432057216, "step": 2760}, {"loss": 1.6082, "grad_norm": 0.44866397976875305, "learning_rate": 0.0002, "epoch": 2.3306689103912497, "step": 2770}, {"loss": 1.6087, "grad_norm": 0.5167056322097778, "learning_rate": 0.0002, "epoch": 2.3390828775767774, "step": 2780}, {"loss": 1.6136, "grad_norm": 0.45913267135620117, "learning_rate": 0.0002, "epoch": 2.3474968447623055, "step": 2790}, {"loss": 1.6564, "grad_norm": 0.45787590742111206, "learning_rate": 0.0002, "epoch": 2.3559108119478336, "step": 2800}, {"loss": 1.6868, "grad_norm": 0.4633352756500244, "learning_rate": 0.0002, "epoch": 2.3643247791333613, "step": 2810}, {"loss": 1.6316, "grad_norm": 0.46390071511268616, "learning_rate": 0.0002, "epoch": 2.3727387463188894, "step": 2820}, {"loss": 1.6039, "grad_norm": 0.4261005222797394, "learning_rate": 0.0002, "epoch": 2.3811527135044175, "step": 2830}, {"loss": 1.6364, "grad_norm": 0.4283634424209595, "learning_rate": 0.0002, "epoch": 2.389566680689945, "step": 2840}, {"loss": 1.6382, "grad_norm": 0.4955291450023651, "learning_rate": 0.0002, "epoch": 2.3979806478754733, "step": 2850}, {"loss": 1.6173, "grad_norm": 0.4740189015865326, "learning_rate": 0.0002, "epoch": 2.4063946150610014, "step": 2860}, {"loss": 1.6403, "grad_norm": 0.4222276508808136, "learning_rate": 0.0002, "epoch": 2.414808582246529, "step": 2870}, {"loss": 1.5602, "grad_norm": 0.4982149004936218, "learning_rate": 0.0002, "epoch": 2.423222549432057, "step": 2880}, {"loss": 1.6313, "grad_norm": 0.5217409133911133, "learning_rate": 0.0002, "epoch": 2.4316365166175853, "step": 2890}, {"loss": 1.5804, "grad_norm": 0.4555884897708893, "learning_rate": 0.0002, "epoch": 2.4400504838031134, "step": 2900}, {"loss": 1.6189, "grad_norm": 0.43178579211235046, "learning_rate": 0.0002, "epoch": 2.448464450988641, "step": 2910}, {"loss": 1.6824, "grad_norm": 0.4788478910923004, "learning_rate": 0.0002, "epoch": 2.456878418174169, "step": 2920}, {"loss": 1.6829, "grad_norm": 0.43689873814582825, "learning_rate": 0.0002, "epoch": 2.465292385359697, "step": 2930}, {"loss": 1.6196, "grad_norm": 0.5115197896957397, "learning_rate": 0.0002, "epoch": 2.473706352545225, "step": 2940}, {"loss": 1.689, "grad_norm": 0.5290159583091736, "learning_rate": 0.0002, "epoch": 2.482120319730753, "step": 2950}, {"loss": 1.6499, "grad_norm": 0.46042463183403015, "learning_rate": 0.0002, "epoch": 2.490534286916281, "step": 2960}, {"loss": 1.6664, "grad_norm": 0.4359915852546692, "learning_rate": 0.0002, "epoch": 2.498948254101809, "step": 2970}, {"loss": 1.5812, "grad_norm": 0.46352964639663696, "learning_rate": 0.0002, "epoch": 2.507362221287337, "step": 2980}, {"loss": 1.6501, "grad_norm": 0.5324268341064453, "learning_rate": 0.0002, "epoch": 2.515776188472865, "step": 2990}, {"loss": 1.6115, "grad_norm": 0.5929607152938843, "learning_rate": 0.0002, "epoch": 2.5241901556583928, "step": 3000}, {"loss": 1.6772, "grad_norm": 0.4811333417892456, "learning_rate": 0.0002, "epoch": 2.532604122843921, "step": 3010}, {"loss": 1.7023, "grad_norm": 0.4662701487541199, "learning_rate": 0.0002, "epoch": 2.541018090029449, "step": 3020}, {"loss": 1.5426, "grad_norm": 0.4582270681858063, "learning_rate": 0.0002, "epoch": 2.549432057214977, "step": 3030}, {"loss": 1.6737, "grad_norm": 0.4679982662200928, "learning_rate": 0.0002, "epoch": 2.557846024400505, "step": 3040}, {"loss": 1.5442, "grad_norm": 0.4380294680595398, "learning_rate": 0.0002, "epoch": 2.566259991586033, "step": 3050}, {"loss": 1.6055, "grad_norm": 0.44295763969421387, "learning_rate": 0.0002, "epoch": 2.5746739587715606, "step": 3060}, {"loss": 1.5775, "grad_norm": 0.5131027698516846, "learning_rate": 0.0002, "epoch": 2.5830879259570887, "step": 3070}, {"loss": 1.546, "grad_norm": 0.47567516565322876, "learning_rate": 0.0002, "epoch": 2.591501893142617, "step": 3080}, {"loss": 1.5671, "grad_norm": 0.49002596735954285, "learning_rate": 0.0002, "epoch": 2.599915860328145, "step": 3090}, {"loss": 1.5445, "grad_norm": 0.44856327772140503, "learning_rate": 0.0002, "epoch": 2.6083298275136726, "step": 3100}, {"loss": 1.5797, "grad_norm": 0.4480142593383789, "learning_rate": 0.0002, "epoch": 2.6167437946992007, "step": 3110}, {"loss": 1.7132, "grad_norm": 0.4317494034767151, "learning_rate": 0.0002, "epoch": 2.6251577618847284, "step": 3120}, {"loss": 1.6321, "grad_norm": 0.42580848932266235, "learning_rate": 0.0002, "epoch": 2.6335717290702565, "step": 3130}, {"loss": 1.6483, "grad_norm": 0.4516814947128296, "learning_rate": 0.0002, "epoch": 2.6419856962557846, "step": 3140}, {"loss": 1.695, "grad_norm": 0.4438435733318329, "learning_rate": 0.0002, "epoch": 2.6503996634413127, "step": 3150}, {"loss": 1.6938, "grad_norm": 0.4385356307029724, "learning_rate": 0.0002, "epoch": 2.6588136306268404, "step": 3160}, {"loss": 1.6139, "grad_norm": 0.5064112544059753, "learning_rate": 0.0002, "epoch": 2.6672275978123685, "step": 3170}, {"loss": 1.7189, "grad_norm": 0.49163177609443665, "learning_rate": 0.0002, "epoch": 2.6756415649978966, "step": 3180}, {"loss": 1.7323, "grad_norm": 0.49339258670806885, "learning_rate": 0.0002, "epoch": 2.6840555321834243, "step": 3190}, {"loss": 1.6508, "grad_norm": 0.440950870513916, "learning_rate": 0.0002, "epoch": 2.6924694993689524, "step": 3200}, {"loss": 1.6305, "grad_norm": 0.4283970594406128, "learning_rate": 0.0002, "epoch": 2.7008834665544805, "step": 3210}, {"loss": 1.5935, "grad_norm": 0.43875712156295776, "learning_rate": 0.0002, "epoch": 2.7092974337400086, "step": 3220}, {"loss": 1.6129, "grad_norm": 0.49332964420318604, "learning_rate": 0.0002, "epoch": 2.7177114009255363, "step": 3230}, {"loss": 1.642, "grad_norm": 0.5225692391395569, "learning_rate": 0.0002, "epoch": 2.7261253681110644, "step": 3240}, {"loss": 1.6759, "grad_norm": 0.4856489300727844, "learning_rate": 0.0002, "epoch": 2.734539335296592, "step": 3250}, {"loss": 1.6463, "grad_norm": 0.46918296813964844, "learning_rate": 0.0002, "epoch": 2.74295330248212, "step": 3260}, {"loss": 1.6819, "grad_norm": 0.4802931249141693, "learning_rate": 0.0002, "epoch": 2.7513672696676483, "step": 3270}, {"loss": 1.6246, "grad_norm": 0.4485355615615845, "learning_rate": 0.0002, "epoch": 2.7597812368531764, "step": 3280}, {"loss": 1.6251, "grad_norm": 0.43944594264030457, "learning_rate": 0.0002, "epoch": 2.768195204038704, "step": 3290}, {"loss": 1.6501, "grad_norm": 0.46847742795944214, "learning_rate": 0.0002, "epoch": 2.7766091712242322, "step": 3300}, {"loss": 1.5969, "grad_norm": 0.4816027879714966, "learning_rate": 0.0002, "epoch": 2.7850231384097603, "step": 3310}, {"loss": 1.6293, "grad_norm": 0.453960120677948, "learning_rate": 0.0002, "epoch": 2.793437105595288, "step": 3320}, {"loss": 1.6429, "grad_norm": 0.4816017150878906, "learning_rate": 0.0002, "epoch": 2.801851072780816, "step": 3330}, {"loss": 1.6683, "grad_norm": 0.4461034834384918, "learning_rate": 0.0002, "epoch": 2.8102650399663442, "step": 3340}, {"loss": 1.7048, "grad_norm": 0.48821821808815, "learning_rate": 0.0002, "epoch": 2.8186790071518724, "step": 3350}, {"loss": 1.6076, "grad_norm": 0.4574853777885437, "learning_rate": 0.0002, "epoch": 2.8270929743374, "step": 3360}, {"loss": 1.6651, "grad_norm": 0.42062026262283325, "learning_rate": 0.0002, "epoch": 2.835506941522928, "step": 3370}, {"loss": 1.624, "grad_norm": 0.4499834477901459, "learning_rate": 0.0002, "epoch": 2.843920908708456, "step": 3380}, {"loss": 1.621, "grad_norm": 0.4780360758304596, "learning_rate": 0.0002, "epoch": 2.852334875893984, "step": 3390}, {"loss": 1.5882, "grad_norm": 0.45422887802124023, "learning_rate": 0.0002, "epoch": 2.860748843079512, "step": 3400}, {"loss": 1.6028, "grad_norm": 0.4590015709400177, "learning_rate": 0.0002, "epoch": 2.86916281026504, "step": 3410}, {"loss": 1.6746, "grad_norm": 0.45689624547958374, "learning_rate": 0.0002, "epoch": 2.877576777450568, "step": 3420}, {"loss": 1.6326, "grad_norm": 0.46953922510147095, "learning_rate": 0.0002, "epoch": 2.885990744636096, "step": 3430}, {"loss": 1.6015, "grad_norm": 0.4791966378688812, "learning_rate": 0.0002, "epoch": 2.8944047118216236, "step": 3440}, {"loss": 1.694, "grad_norm": 0.4842296242713928, "learning_rate": 0.0002, "epoch": 2.9028186790071517, "step": 3450}, {"loss": 1.6326, "grad_norm": 0.47219768166542053, "learning_rate": 0.0002, "epoch": 2.91123264619268, "step": 3460}, {"loss": 1.6486, "grad_norm": 0.4622127115726471, "learning_rate": 0.0002, "epoch": 2.919646613378208, "step": 3470}, {"loss": 1.6485, "grad_norm": 0.46832820773124695, "learning_rate": 0.0002, "epoch": 2.9280605805637356, "step": 3480}, {"loss": 1.6366, "grad_norm": 0.44582483172416687, "learning_rate": 0.0002, "epoch": 2.9364745477492638, "step": 3490}, {"loss": 1.6859, "grad_norm": 0.4987219274044037, "learning_rate": 0.0002, "epoch": 2.944888514934792, "step": 3500}, {"loss": 1.5991, "grad_norm": 0.43750956654548645, "learning_rate": 0.0002, "epoch": 2.9533024821203195, "step": 3510}, {"loss": 1.6236, "grad_norm": 0.49962925910949707, "learning_rate": 0.0002, "epoch": 2.9617164493058477, "step": 3520}, {"loss": 1.5859, "grad_norm": 0.5189590454101562, "learning_rate": 0.0002, "epoch": 2.9701304164913758, "step": 3530}, {"loss": 1.6688, "grad_norm": 0.391317754983902, "learning_rate": 0.0002, "epoch": 2.978544383676904, "step": 3540}, {"loss": 1.5884, "grad_norm": 0.44934695959091187, "learning_rate": 0.0002, "epoch": 2.9869583508624316, "step": 3550}, {"loss": 1.5688, "grad_norm": 0.4740142226219177, "learning_rate": 0.0002, "epoch": 2.9953723180479597, "step": 3560}, {"eval_loss": 1.8266887664794922, "eval_runtime": 37.9445, "eval_samples_per_second": 13.572, "eval_steps_per_second": 1.713, "epoch": 2.9995793016407237, "step": 3565}, {"loss": 1.5939, "grad_norm": 0.4523724615573883, "learning_rate": 0.0002, "epoch": 3.003786285233488, "step": 3570}, {"loss": 1.526, "grad_norm": 0.5261380076408386, "learning_rate": 0.0002, "epoch": 3.0122002524190155, "step": 3580}, {"loss": 1.4946, "grad_norm": 0.48664888739585876, "learning_rate": 0.0002, "epoch": 3.0206142196045436, "step": 3590}, {"loss": 1.5193, "grad_norm": 0.5070882439613342, "learning_rate": 0.0002, "epoch": 3.0290281867900717, "step": 3600}, {"loss": 1.5316, "grad_norm": 0.5816011428833008, "learning_rate": 0.0002, "epoch": 3.0374421539755994, "step": 3610}, {"loss": 1.5682, "grad_norm": 0.6610211730003357, "learning_rate": 0.0002, "epoch": 3.0458561211611275, "step": 3620}, {"loss": 1.5699, "grad_norm": 0.5257703065872192, "learning_rate": 0.0002, "epoch": 3.0542700883466556, "step": 3630}, {"loss": 1.4438, "grad_norm": 0.5574390888214111, "learning_rate": 0.0002, "epoch": 3.0626840555321833, "step": 3640}, {"loss": 1.547, "grad_norm": 0.5682297348976135, "learning_rate": 0.0002, "epoch": 3.0710980227177114, "step": 3650}, {"loss": 1.5743, "grad_norm": 0.5798383355140686, "learning_rate": 0.0002, "epoch": 3.0795119899032395, "step": 3660}, {"loss": 1.4339, "grad_norm": 0.5458289980888367, "learning_rate": 0.0002, "epoch": 3.087925957088767, "step": 3670}, {"loss": 1.46, "grad_norm": 0.5599102973937988, "learning_rate": 0.0002, "epoch": 3.0963399242742953, "step": 3680}, {"loss": 1.4589, "grad_norm": 0.5023021697998047, "learning_rate": 0.0002, "epoch": 3.1047538914598234, "step": 3690}, {"loss": 1.5114, "grad_norm": 0.5448206067085266, "learning_rate": 0.0002, "epoch": 3.113167858645351, "step": 3700}, {"loss": 1.4692, "grad_norm": 0.5760458707809448, "learning_rate": 0.0002, "epoch": 3.121581825830879, "step": 3710}, {"loss": 1.4789, "grad_norm": 0.6018968224525452, "learning_rate": 0.0002, "epoch": 3.1299957930164073, "step": 3720}, {"loss": 1.5518, "grad_norm": 0.5767101049423218, "learning_rate": 0.0002, "epoch": 3.1384097602019354, "step": 3730}, {"loss": 1.5032, "grad_norm": 0.5333963632583618, "learning_rate": 0.0002, "epoch": 3.146823727387463, "step": 3740}, {"loss": 1.4812, "grad_norm": 0.5918396711349487, "learning_rate": 0.0002, "epoch": 3.155237694572991, "step": 3750}, {"loss": 1.4618, "grad_norm": 0.5931203365325928, "learning_rate": 0.0002, "epoch": 3.1636516617585193, "step": 3760}, {"loss": 1.5592, "grad_norm": 0.6562168598175049, "learning_rate": 0.0002, "epoch": 3.172065628944047, "step": 3770}, {"loss": 1.4932, "grad_norm": 0.5820156335830688, "learning_rate": 0.0002, "epoch": 3.180479596129575, "step": 3780}, {"loss": 1.4523, "grad_norm": 0.5784737467765808, "learning_rate": 0.0002, "epoch": 3.188893563315103, "step": 3790}, {"loss": 1.498, "grad_norm": 0.5506529808044434, "learning_rate": 0.0002, "epoch": 3.197307530500631, "step": 3800}, {"loss": 1.4819, "grad_norm": 0.6101595163345337, "learning_rate": 0.0002, "epoch": 3.205721497686159, "step": 3810}, {"loss": 1.5185, "grad_norm": 0.5597806572914124, "learning_rate": 0.0002, "epoch": 3.214135464871687, "step": 3820}, {"loss": 1.5664, "grad_norm": 0.5641011595726013, "learning_rate": 0.0002, "epoch": 3.222549432057215, "step": 3830}, {"loss": 1.4702, "grad_norm": 0.5892080068588257, "learning_rate": 0.0002, "epoch": 3.230963399242743, "step": 3840}, {"loss": 1.4194, "grad_norm": 0.6034760475158691, "learning_rate": 0.0002, "epoch": 3.239377366428271, "step": 3850}, {"loss": 1.5499, "grad_norm": 0.5112439393997192, "learning_rate": 0.0002, "epoch": 3.247791333613799, "step": 3860}, {"loss": 1.5132, "grad_norm": 0.56565922498703, "learning_rate": 0.0002, "epoch": 3.256205300799327, "step": 3870}, {"loss": 1.4892, "grad_norm": 0.6155247092247009, "learning_rate": 0.0002, "epoch": 3.264619267984855, "step": 3880}, {"loss": 1.5118, "grad_norm": 0.6064623594284058, "learning_rate": 0.0002, "epoch": 3.273033235170383, "step": 3890}, {"loss": 1.5236, "grad_norm": 0.6313768029212952, "learning_rate": 0.0002, "epoch": 3.2814472023559107, "step": 3900}, {"loss": 1.5551, "grad_norm": 0.5903939008712769, "learning_rate": 0.0002, "epoch": 3.289861169541439, "step": 3910}, {"loss": 1.5703, "grad_norm": 0.5770667195320129, "learning_rate": 0.0002, "epoch": 3.298275136726967, "step": 3920}, {"loss": 1.5159, "grad_norm": 0.5785196423530579, "learning_rate": 0.0002, "epoch": 3.3066891039124946, "step": 3930}, {"loss": 1.5277, "grad_norm": 0.6468310356140137, "learning_rate": 0.0002, "epoch": 3.3151030710980227, "step": 3940}, {"loss": 1.6002, "grad_norm": 0.6200279593467712, "learning_rate": 0.0002, "epoch": 3.323517038283551, "step": 3950}, {"loss": 1.5264, "grad_norm": 0.5779302716255188, "learning_rate": 0.0002, "epoch": 3.3319310054690785, "step": 3960}, {"loss": 1.4861, "grad_norm": 0.5463796854019165, "learning_rate": 0.0002, "epoch": 3.3403449726546066, "step": 3970}, {"loss": 1.541, "grad_norm": 0.6117855906486511, "learning_rate": 0.0002, "epoch": 3.3487589398401347, "step": 3980}, {"loss": 1.5566, "grad_norm": 0.5554766058921814, "learning_rate": 0.0002, "epoch": 3.357172907025663, "step": 3990}, {"loss": 1.5004, "grad_norm": 0.6012870073318481, "learning_rate": 0.0002, "epoch": 3.3655868742111905, "step": 4000}, {"loss": 1.473, "grad_norm": 0.5443974137306213, "learning_rate": 0.0002, "epoch": 3.3740008413967186, "step": 4010}, {"loss": 1.5139, "grad_norm": 0.6636057496070862, "learning_rate": 0.0002, "epoch": 3.3824148085822463, "step": 4020}, {"loss": 1.5141, "grad_norm": 0.5801246166229248, "learning_rate": 0.0002, "epoch": 3.3908287757677744, "step": 4030}, {"loss": 1.5026, "grad_norm": 0.5668839812278748, "learning_rate": 0.0002, "epoch": 3.3992427429533025, "step": 4040}, {"loss": 1.523, "grad_norm": 0.7763481736183167, "learning_rate": 0.0002, "epoch": 3.4076567101388306, "step": 4050}, {"loss": 1.4932, "grad_norm": 0.6675992608070374, "learning_rate": 0.0002, "epoch": 3.4160706773243583, "step": 4060}, {"loss": 1.4959, "grad_norm": 0.6290077567100525, "learning_rate": 0.0002, "epoch": 3.4244846445098864, "step": 4070}, {"loss": 1.5766, "grad_norm": 0.6040239930152893, "learning_rate": 0.0002, "epoch": 3.4328986116954145, "step": 4080}, {"loss": 1.5711, "grad_norm": 0.6237877607345581, "learning_rate": 0.0002, "epoch": 3.441312578880942, "step": 4090}, {"loss": 1.4961, "grad_norm": 0.5343508124351501, "learning_rate": 0.0002, "epoch": 3.4497265460664703, "step": 4100}, {"loss": 1.5123, "grad_norm": 0.6817412972450256, "learning_rate": 0.0002, "epoch": 3.4581405132519984, "step": 4110}, {"loss": 1.5377, "grad_norm": 0.7115170359611511, "learning_rate": 0.0002, "epoch": 3.466554480437526, "step": 4120}, {"loss": 1.5275, "grad_norm": 0.6127332448959351, "learning_rate": 0.0002, "epoch": 3.4749684476230542, "step": 4130}, {"loss": 1.557, "grad_norm": 0.5745994448661804, "learning_rate": 0.0002, "epoch": 3.4833824148085824, "step": 4140}, {"loss": 1.4873, "grad_norm": 0.6248795390129089, "learning_rate": 0.0002, "epoch": 3.49179638199411, "step": 4150}, {"loss": 1.4885, "grad_norm": 0.5821124911308289, "learning_rate": 0.0002, "epoch": 3.500210349179638, "step": 4160}, {"loss": 1.4937, "grad_norm": 0.561416506767273, "learning_rate": 0.0002, "epoch": 3.5086243163651663, "step": 4170}, {"loss": 1.5453, "grad_norm": 0.5848962664604187, "learning_rate": 0.0002, "epoch": 3.5170382835506944, "step": 4180}, {"loss": 1.5892, "grad_norm": 0.5335569977760315, "learning_rate": 0.0002, "epoch": 3.525452250736222, "step": 4190}, {"loss": 1.5152, "grad_norm": 0.547964870929718, "learning_rate": 0.0002, "epoch": 3.53386621792175, "step": 4200}, {"loss": 1.4887, "grad_norm": 0.6157727241516113, "learning_rate": 0.0002, "epoch": 3.542280185107278, "step": 4210}, {"loss": 1.5484, "grad_norm": 0.6163121461868286, "learning_rate": 0.0002, "epoch": 3.550694152292806, "step": 4220}, {"loss": 1.5833, "grad_norm": 0.5844616293907166, "learning_rate": 0.0002, "epoch": 3.559108119478334, "step": 4230}, {"loss": 1.5305, "grad_norm": 0.7104926109313965, "learning_rate": 0.0002, "epoch": 3.567522086663862, "step": 4240}, {"loss": 1.5161, "grad_norm": 0.5055213570594788, "learning_rate": 0.0002, "epoch": 3.57593605384939, "step": 4250}, {"loss": 1.482, "grad_norm": 0.611676812171936, "learning_rate": 0.0002, "epoch": 3.584350021034918, "step": 4260}, {"loss": 1.5048, "grad_norm": 0.6326440572738647, "learning_rate": 0.0002, "epoch": 3.592763988220446, "step": 4270}, {"loss": 1.5122, "grad_norm": 0.6290925741195679, "learning_rate": 0.0002, "epoch": 3.6011779554059737, "step": 4280}, {"loss": 1.5654, "grad_norm": 0.5691978931427002, "learning_rate": 0.0002, "epoch": 3.609591922591502, "step": 4290}, {"loss": 1.4854, "grad_norm": 0.6071329116821289, "learning_rate": 0.0002, "epoch": 3.61800588977703, "step": 4300}, {"loss": 1.5336, "grad_norm": 0.606573224067688, "learning_rate": 0.0002, "epoch": 3.626419856962558, "step": 4310}, {"loss": 1.6437, "grad_norm": 0.5515419244766235, "learning_rate": 0.0002, "epoch": 3.6348338241480858, "step": 4320}, {"loss": 1.498, "grad_norm": 0.5964660048484802, "learning_rate": 0.0002, "epoch": 3.643247791333614, "step": 4330}, {"loss": 1.544, "grad_norm": 0.5774146914482117, "learning_rate": 0.0002, "epoch": 3.6516617585191415, "step": 4340}, {"loss": 1.5566, "grad_norm": 0.5732731223106384, "learning_rate": 0.0002, "epoch": 3.6600757257046697, "step": 4350}, {"loss": 1.5682, "grad_norm": 0.7354163527488708, "learning_rate": 0.0002, "epoch": 3.6684896928901978, "step": 4360}, {"loss": 1.5225, "grad_norm": 0.6220902800559998, "learning_rate": 0.0002, "epoch": 3.676903660075726, "step": 4370}, {"loss": 1.4838, "grad_norm": 0.6053991317749023, "learning_rate": 0.0002, "epoch": 3.6853176272612536, "step": 4380}, {"loss": 1.5161, "grad_norm": 0.67010897397995, "learning_rate": 0.0002, "epoch": 3.6937315944467817, "step": 4390}, {"loss": 1.5381, "grad_norm": 0.6139186024665833, "learning_rate": 0.0002, "epoch": 3.70214556163231, "step": 4400}, {"loss": 1.5088, "grad_norm": 0.5433071851730347, "learning_rate": 0.0002, "epoch": 3.7105595288178375, "step": 4410}, {"loss": 1.5337, "grad_norm": 0.5453870296478271, "learning_rate": 0.0002, "epoch": 3.7189734960033656, "step": 4420}, {"loss": 1.4549, "grad_norm": 0.6401727199554443, "learning_rate": 0.0002, "epoch": 3.7273874631888937, "step": 4430}, {"loss": 1.503, "grad_norm": 0.6049367189407349, "learning_rate": 0.0002, "epoch": 3.735801430374422, "step": 4440}, {"loss": 1.5268, "grad_norm": 0.5740529298782349, "learning_rate": 0.0002, "epoch": 3.7442153975599495, "step": 4450}, {"loss": 1.5183, "grad_norm": 0.6521880626678467, "learning_rate": 0.0002, "epoch": 3.7526293647454776, "step": 4460}, {"loss": 1.5741, "grad_norm": 0.7096368074417114, "learning_rate": 0.0002, "epoch": 3.7610433319310053, "step": 4470}, {"loss": 1.5786, "grad_norm": 0.5886474251747131, "learning_rate": 0.0002, "epoch": 3.7694572991165334, "step": 4480}, {"loss": 1.5887, "grad_norm": 0.5821043252944946, "learning_rate": 0.0002, "epoch": 3.7778712663020615, "step": 4490}, {"loss": 1.5777, "grad_norm": 0.628892183303833, "learning_rate": 0.0002, "epoch": 3.7862852334875896, "step": 4500}, {"loss": 1.4708, "grad_norm": 0.5962669849395752, "learning_rate": 0.0002, "epoch": 3.7946992006731173, "step": 4510}, {"loss": 1.5267, "grad_norm": 0.6635549068450928, "learning_rate": 0.0002, "epoch": 3.8031131678586454, "step": 4520}, {"loss": 1.5058, "grad_norm": 0.6010760068893433, "learning_rate": 0.0002, "epoch": 3.811527135044173, "step": 4530}, {"loss": 1.6228, "grad_norm": 0.6322658658027649, "learning_rate": 0.0002, "epoch": 3.819941102229701, "step": 4540}, {"loss": 1.5029, "grad_norm": 0.5893137454986572, "learning_rate": 0.0002, "epoch": 3.8283550694152293, "step": 4550}, {"loss": 1.5435, "grad_norm": 0.7829602360725403, "learning_rate": 0.0002, "epoch": 3.8367690366007574, "step": 4560}, {"loss": 1.5453, "grad_norm": 0.6190396547317505, "learning_rate": 0.0002, "epoch": 3.845183003786285, "step": 4570}, {"loss": 1.5292, "grad_norm": 0.6662813425064087, "learning_rate": 0.0002, "epoch": 3.853596970971813, "step": 4580}, {"loss": 1.5065, "grad_norm": 0.5809855461120605, "learning_rate": 0.0002, "epoch": 3.8620109381573413, "step": 4590}, {"loss": 1.5041, "grad_norm": 0.5779069662094116, "learning_rate": 0.0002, "epoch": 3.870424905342869, "step": 4600}, {"loss": 1.498, "grad_norm": 0.5603038668632507, "learning_rate": 0.0002, "epoch": 3.878838872528397, "step": 4610}, {"loss": 1.5372, "grad_norm": 0.6274181008338928, "learning_rate": 0.0002, "epoch": 3.887252839713925, "step": 4620}, {"loss": 1.4996, "grad_norm": 0.6810959577560425, "learning_rate": 0.0002, "epoch": 3.8956668068994533, "step": 4630}, {"loss": 1.4956, "grad_norm": 0.5647315979003906, "learning_rate": 0.0002, "epoch": 3.904080774084981, "step": 4640}, {"loss": 1.5424, "grad_norm": 0.6830295324325562, "learning_rate": 0.0002, "epoch": 3.912494741270509, "step": 4650}, {"loss": 1.535, "grad_norm": 0.652565598487854, "learning_rate": 0.0002, "epoch": 3.920908708456037, "step": 4660}, {"loss": 1.4772, "grad_norm": 0.5806284546852112, "learning_rate": 0.0002, "epoch": 3.929322675641565, "step": 4670}, {"loss": 1.5812, "grad_norm": 0.6825073957443237, "learning_rate": 0.0002, "epoch": 3.937736642827093, "step": 4680}, {"loss": 1.5516, "grad_norm": 0.6149451732635498, "learning_rate": 0.0002, "epoch": 3.946150610012621, "step": 4690}, {"loss": 1.5608, "grad_norm": 0.6152557134628296, "learning_rate": 0.0002, "epoch": 3.954564577198149, "step": 4700}, {"loss": 1.4897, "grad_norm": 0.6239011883735657, "learning_rate": 0.0002, "epoch": 3.962978544383677, "step": 4710}, {"loss": 1.538, "grad_norm": 0.6485443115234375, "learning_rate": 0.0002, "epoch": 3.971392511569205, "step": 4720}, {"loss": 1.5226, "grad_norm": 0.6449228525161743, "learning_rate": 0.0002, "epoch": 3.9798064787547327, "step": 4730}, {"loss": 1.5087, "grad_norm": 0.6526407599449158, "learning_rate": 0.0002, "epoch": 3.988220445940261, "step": 4740}, {"loss": 1.5026, "grad_norm": 0.6277706027030945, "learning_rate": 0.0002, "epoch": 3.996634413125789, "step": 4750}, {"eval_loss": 1.871641755104065, "eval_runtime": 37.9637, "eval_samples_per_second": 13.566, "eval_steps_per_second": 1.712, "epoch": 4.0, "step": 4754}, {"loss": 1.4744, "grad_norm": 0.6994837522506714, "learning_rate": 0.0002, "epoch": 4.005048380311317, "step": 4760}, {"loss": 1.4433, "grad_norm": 0.8728373050689697, "learning_rate": 0.0002, "epoch": 4.013462347496845, "step": 4770}, {"loss": 1.3329, "grad_norm": 0.688679575920105, "learning_rate": 0.0002, "epoch": 4.021876314682372, "step": 4780}, {"loss": 1.3999, "grad_norm": 0.6313387155532837, "learning_rate": 0.0002, "epoch": 4.0302902818679005, "step": 4790}, {"loss": 1.3346, "grad_norm": 0.6577984690666199, "learning_rate": 0.0002, "epoch": 4.038704249053429, "step": 4800}, {"loss": 1.3403, "grad_norm": 0.7938185930252075, "learning_rate": 0.0002, "epoch": 4.047118216238957, "step": 4810}, {"loss": 1.3716, "grad_norm": 0.760399580001831, "learning_rate": 0.0002, "epoch": 4.055532183424485, "step": 4820}, {"loss": 1.4321, "grad_norm": 0.7329602241516113, "learning_rate": 0.0002, "epoch": 4.063946150610013, "step": 4830}, {"loss": 1.4133, "grad_norm": 0.7778576016426086, "learning_rate": 0.0002, "epoch": 4.07236011779554, "step": 4840}, {"loss": 1.4372, "grad_norm": 0.8235865235328674, "learning_rate": 0.0002, "epoch": 4.080774084981068, "step": 4850}, {"loss": 1.3719, "grad_norm": 0.7743754386901855, "learning_rate": 0.0002, "epoch": 4.089188052166596, "step": 4860}, {"loss": 1.3787, "grad_norm": 0.8145367503166199, "learning_rate": 0.0002, "epoch": 4.0976020193521245, "step": 4870}, {"loss": 1.356, "grad_norm": 0.8517307639122009, "learning_rate": 0.0002, "epoch": 4.106015986537653, "step": 4880}, {"loss": 1.4191, "grad_norm": 0.8208953142166138, "learning_rate": 0.0002, "epoch": 4.114429953723181, "step": 4890}, {"loss": 1.3189, "grad_norm": 0.8437790870666504, "learning_rate": 0.0002, "epoch": 4.122843920908709, "step": 4900}, {"loss": 1.3987, "grad_norm": 0.716672420501709, "learning_rate": 0.0002, "epoch": 4.131257888094236, "step": 4910}, {"loss": 1.4392, "grad_norm": 0.7656235098838806, "learning_rate": 0.0002, "epoch": 4.139671855279764, "step": 4920}, {"loss": 1.3408, "grad_norm": 0.7209306955337524, "learning_rate": 0.0002, "epoch": 4.148085822465292, "step": 4930}, {"loss": 1.3639, "grad_norm": 0.7731267809867859, "learning_rate": 0.0002, "epoch": 4.1564997896508205, "step": 4940}, {"loss": 1.4151, "grad_norm": 0.7477553486824036, "learning_rate": 0.0002, "epoch": 4.164913756836349, "step": 4950}, {"loss": 1.3485, "grad_norm": 0.7372981309890747, "learning_rate": 0.0002, "epoch": 4.173327724021877, "step": 4960}, {"loss": 1.3901, "grad_norm": 0.6582154035568237, "learning_rate": 0.0002, "epoch": 4.181741691207404, "step": 4970}, {"loss": 1.3343, "grad_norm": 0.7003206610679626, "learning_rate": 0.0002, "epoch": 4.190155658392932, "step": 4980}, {"loss": 1.4098, "grad_norm": 0.735223650932312, "learning_rate": 0.0002, "epoch": 4.19856962557846, "step": 4990}, {"loss": 1.3564, "grad_norm": 0.7832302451133728, "learning_rate": 0.0002, "epoch": 4.206983592763988, "step": 5000}, {"loss": 1.3622, "grad_norm": 0.8819546103477478, "learning_rate": 0.0002, "epoch": 4.215397559949516, "step": 5010}, {"loss": 1.4438, "grad_norm": 0.9325336813926697, "learning_rate": 0.0002, "epoch": 4.2238115271350445, "step": 5020}, {"loss": 1.3886, "grad_norm": 0.7007517218589783, "learning_rate": 0.0002, "epoch": 4.232225494320572, "step": 5030}, {"loss": 1.3683, "grad_norm": 0.7118321061134338, "learning_rate": 0.0002, "epoch": 4.2406394615061, "step": 5040}, {"loss": 1.2365, "grad_norm": 0.6578946709632874, "learning_rate": 0.0002, "epoch": 4.249053428691628, "step": 5050}, {"loss": 1.3696, "grad_norm": 0.9438983798027039, "learning_rate": 0.0002, "epoch": 4.257467395877156, "step": 5060}, {"loss": 1.3868, "grad_norm": 0.703037679195404, "learning_rate": 0.0002, "epoch": 4.265881363062684, "step": 5070}, {"loss": 1.3687, "grad_norm": 0.7286025285720825, "learning_rate": 0.0002, "epoch": 4.274295330248212, "step": 5080}, {"loss": 1.3605, "grad_norm": 0.750689685344696, "learning_rate": 0.0002, "epoch": 4.28270929743374, "step": 5090}, {"loss": 1.5089, "grad_norm": 0.869753360748291, "learning_rate": 0.0002, "epoch": 4.291123264619268, "step": 5100}, {"loss": 1.4128, "grad_norm": 0.8712980151176453, "learning_rate": 0.0002, "epoch": 4.299537231804796, "step": 5110}, {"loss": 1.3977, "grad_norm": 0.690263569355011, "learning_rate": 0.0002, "epoch": 4.307951198990324, "step": 5120}, {"loss": 1.4088, "grad_norm": 0.7114760279655457, "learning_rate": 0.0002, "epoch": 4.316365166175852, "step": 5130}, {"loss": 1.363, "grad_norm": 0.7588112354278564, "learning_rate": 0.0002, "epoch": 4.32477913336138, "step": 5140}, {"loss": 1.4408, "grad_norm": 0.7556202411651611, "learning_rate": 0.0002, "epoch": 4.333193100546908, "step": 5150}, {"loss": 1.4203, "grad_norm": 0.8357610702514648, "learning_rate": 0.0002, "epoch": 4.341607067732435, "step": 5160}, {"loss": 1.3348, "grad_norm": 0.8054035902023315, "learning_rate": 0.0002, "epoch": 4.3500210349179635, "step": 5170}, {"loss": 1.3109, "grad_norm": 0.7637107968330383, "learning_rate": 0.0002, "epoch": 4.358435002103492, "step": 5180}, {"loss": 1.3744, "grad_norm": 0.757481038570404, "learning_rate": 0.0002, "epoch": 4.36684896928902, "step": 5190}, {"loss": 1.3622, "grad_norm": 0.7185863852500916, "learning_rate": 0.0002, "epoch": 4.375262936474548, "step": 5200}, {"loss": 1.3896, "grad_norm": 0.7326455116271973, "learning_rate": 0.0002, "epoch": 4.383676903660076, "step": 5210}, {"loss": 1.4098, "grad_norm": 0.7980523109436035, "learning_rate": 0.0002, "epoch": 4.392090870845603, "step": 5220}, {"loss": 1.3783, "grad_norm": 0.8526999354362488, "learning_rate": 0.0002, "epoch": 4.400504838031131, "step": 5230}, {"loss": 1.4022, "grad_norm": 0.7012337446212769, "learning_rate": 0.0002, "epoch": 4.4089188052166595, "step": 5240}, {"loss": 1.3552, "grad_norm": 0.8217827677726746, "learning_rate": 0.0002, "epoch": 4.417332772402188, "step": 5250}, {"loss": 1.3482, "grad_norm": 0.7141005396842957, "learning_rate": 0.0002, "epoch": 4.425746739587716, "step": 5260}, {"loss": 1.3699, "grad_norm": 0.7094302177429199, "learning_rate": 0.0002, "epoch": 4.434160706773244, "step": 5270}, {"loss": 1.3527, "grad_norm": 0.7234613299369812, "learning_rate": 0.0002, "epoch": 4.442574673958772, "step": 5280}, {"loss": 1.4769, "grad_norm": 0.7530457973480225, "learning_rate": 0.0002, "epoch": 4.450988641144299, "step": 5290}, {"loss": 1.3944, "grad_norm": 0.7300912141799927, "learning_rate": 0.0002, "epoch": 4.459402608329827, "step": 5300}, {"loss": 1.3844, "grad_norm": 0.825443685054779, "learning_rate": 0.0002, "epoch": 4.467816575515355, "step": 5310}, {"loss": 1.3648, "grad_norm": 0.7559658885002136, "learning_rate": 0.0002, "epoch": 4.4762305427008835, "step": 5320}, {"loss": 1.4364, "grad_norm": 0.8817561268806458, "learning_rate": 0.0002, "epoch": 4.484644509886412, "step": 5330}, {"loss": 1.3618, "grad_norm": 0.8203575611114502, "learning_rate": 0.0002, "epoch": 4.49305847707194, "step": 5340}, {"loss": 1.3996, "grad_norm": 0.7677690982818604, "learning_rate": 0.0002, "epoch": 4.501472444257468, "step": 5350}, {"loss": 1.4142, "grad_norm": 0.657085120677948, "learning_rate": 0.0002, "epoch": 4.509886411442995, "step": 5360}, {"loss": 1.3722, "grad_norm": 0.7939504384994507, "learning_rate": 0.0002, "epoch": 4.518300378628523, "step": 5370}, {"loss": 1.4361, "grad_norm": 0.6971889138221741, "learning_rate": 0.0002, "epoch": 4.526714345814051, "step": 5380}, {"loss": 1.3637, "grad_norm": 0.6984175443649292, "learning_rate": 0.0002, "epoch": 4.535128312999579, "step": 5390}, {"loss": 1.341, "grad_norm": 0.8504858613014221, "learning_rate": 0.0002, "epoch": 4.5435422801851075, "step": 5400}, {"loss": 1.4026, "grad_norm": 0.9134073853492737, "learning_rate": 0.0002, "epoch": 4.551956247370635, "step": 5410}, {"loss": 1.4375, "grad_norm": 0.7765598893165588, "learning_rate": 0.0002, "epoch": 4.560370214556163, "step": 5420}, {"loss": 1.4832, "grad_norm": 0.6991009712219238, "learning_rate": 0.0002, "epoch": 4.568784181741691, "step": 5430}, {"loss": 1.4021, "grad_norm": 0.8393039107322693, "learning_rate": 0.0002, "epoch": 4.577198148927219, "step": 5440}, {"loss": 1.3976, "grad_norm": 0.7685918211936951, "learning_rate": 0.0002, "epoch": 4.585612116112747, "step": 5450}, {"loss": 1.3883, "grad_norm": 0.7135679721832275, "learning_rate": 0.0002, "epoch": 4.594026083298275, "step": 5460}, {"loss": 1.4083, "grad_norm": 0.6728870868682861, "learning_rate": 0.0002, "epoch": 4.6024400504838034, "step": 5470}, {"loss": 1.3698, "grad_norm": 0.7139479517936707, "learning_rate": 0.0002, "epoch": 4.610854017669331, "step": 5480}, {"loss": 1.3498, "grad_norm": 0.8476598858833313, "learning_rate": 0.0002, "epoch": 4.619267984854859, "step": 5490}, {"loss": 1.3389, "grad_norm": 0.8034361004829407, "learning_rate": 0.0002, "epoch": 4.627681952040387, "step": 5500}, {"loss": 1.4179, "grad_norm": 0.7452183961868286, "learning_rate": 0.0002, "epoch": 4.636095919225915, "step": 5510}, {"loss": 1.4031, "grad_norm": 0.8394148945808411, "learning_rate": 0.0002, "epoch": 4.644509886411443, "step": 5520}, {"loss": 1.4561, "grad_norm": 0.7480153441429138, "learning_rate": 0.0002, "epoch": 4.652923853596971, "step": 5530}, {"loss": 1.378, "grad_norm": 0.7781714797019958, "learning_rate": 0.0002, "epoch": 4.661337820782499, "step": 5540}, {"loss": 1.3924, "grad_norm": 1.0058213472366333, "learning_rate": 0.0002, "epoch": 4.669751787968027, "step": 5550}, {"loss": 1.4198, "grad_norm": 0.7403179407119751, "learning_rate": 0.0002, "epoch": 4.678165755153555, "step": 5560}, {"loss": 1.4328, "grad_norm": 0.7270476818084717, "learning_rate": 0.0002, "epoch": 4.686579722339083, "step": 5570}, {"loss": 1.378, "grad_norm": 0.760877788066864, "learning_rate": 0.0002, "epoch": 4.694993689524611, "step": 5580}, {"loss": 1.387, "grad_norm": 0.8097004890441895, "learning_rate": 0.0002, "epoch": 4.703407656710139, "step": 5590}, {"loss": 1.3661, "grad_norm": 0.9096523523330688, "learning_rate": 0.0002, "epoch": 4.711821623895667, "step": 5600}, {"loss": 1.4012, "grad_norm": 0.7262444496154785, "learning_rate": 0.0002, "epoch": 4.720235591081195, "step": 5610}, {"loss": 1.422, "grad_norm": 0.8207762837409973, "learning_rate": 0.0002, "epoch": 4.7286495582667225, "step": 5620}, {"loss": 1.4017, "grad_norm": 0.8089601993560791, "learning_rate": 0.0002, "epoch": 4.737063525452251, "step": 5630}, {"loss": 1.3675, "grad_norm": 0.7609543800354004, "learning_rate": 0.0002, "epoch": 4.745477492637779, "step": 5640}, {"loss": 1.4085, "grad_norm": 0.7273501753807068, "learning_rate": 0.0002, "epoch": 4.753891459823307, "step": 5650}, {"loss": 1.3849, "grad_norm": 0.7800219058990479, "learning_rate": 0.0002, "epoch": 4.762305427008835, "step": 5660}, {"loss": 1.4319, "grad_norm": 0.8558377623558044, "learning_rate": 0.0002, "epoch": 4.770719394194362, "step": 5670}, {"loss": 1.3831, "grad_norm": 0.7131547927856445, "learning_rate": 0.0002, "epoch": 4.77913336137989, "step": 5680}, {"loss": 1.407, "grad_norm": 0.7651025056838989, "learning_rate": 0.0002, "epoch": 4.787547328565418, "step": 5690}, {"loss": 1.3882, "grad_norm": 0.8129976391792297, "learning_rate": 0.0002, "epoch": 4.7959612957509465, "step": 5700}, {"loss": 1.4347, "grad_norm": 0.8019895553588867, "learning_rate": 0.0002, "epoch": 4.804375262936475, "step": 5710}, {"loss": 1.3961, "grad_norm": 0.7692018151283264, "learning_rate": 0.0002, "epoch": 4.812789230122003, "step": 5720}, {"loss": 1.419, "grad_norm": 0.6893943548202515, "learning_rate": 0.0002, "epoch": 4.821203197307531, "step": 5730}, {"loss": 1.4453, "grad_norm": 0.6881810426712036, "learning_rate": 0.0002, "epoch": 4.829617164493058, "step": 5740}, {"loss": 1.4775, "grad_norm": 0.7838267683982849, "learning_rate": 0.0002, "epoch": 4.838031131678586, "step": 5750}, {"loss": 1.3857, "grad_norm": 0.727799117565155, "learning_rate": 0.0002, "epoch": 4.846445098864114, "step": 5760}, {"loss": 1.4685, "grad_norm": 0.7458277344703674, "learning_rate": 0.0002, "epoch": 4.8548590660496425, "step": 5770}, {"loss": 1.4426, "grad_norm": 0.903802216053009, "learning_rate": 0.0002, "epoch": 4.863273033235171, "step": 5780}, {"loss": 1.451, "grad_norm": 0.7983472347259521, "learning_rate": 0.0002, "epoch": 4.871687000420699, "step": 5790}, {"loss": 1.4534, "grad_norm": 0.6894361972808838, "learning_rate": 0.0002, "epoch": 4.880100967606227, "step": 5800}, {"loss": 1.4486, "grad_norm": 0.7499409317970276, "learning_rate": 0.0002, "epoch": 4.888514934791754, "step": 5810}, {"loss": 1.4253, "grad_norm": 0.7362820506095886, "learning_rate": 0.0002, "epoch": 4.896928901977282, "step": 5820}, {"loss": 1.3763, "grad_norm": 0.8341619968414307, "learning_rate": 0.0002, "epoch": 4.90534286916281, "step": 5830}, {"loss": 1.3748, "grad_norm": 0.9604470133781433, "learning_rate": 0.0002, "epoch": 4.913756836348338, "step": 5840}, {"loss": 1.3658, "grad_norm": 0.8916844129562378, "learning_rate": 0.0002, "epoch": 4.9221708035338665, "step": 5850}, {"loss": 1.363, "grad_norm": 0.8519647121429443, "learning_rate": 0.0002, "epoch": 4.930584770719394, "step": 5860}, {"loss": 1.424, "grad_norm": 0.7946906089782715, "learning_rate": 0.0002, "epoch": 4.938998737904922, "step": 5870}, {"loss": 1.4071, "grad_norm": 0.7843789458274841, "learning_rate": 0.0002, "epoch": 4.94741270509045, "step": 5880}, {"loss": 1.4021, "grad_norm": 0.707618772983551, "learning_rate": 0.0002, "epoch": 4.955826672275978, "step": 5890}, {"loss": 1.502, "grad_norm": 0.7704206109046936, "learning_rate": 0.0002, "epoch": 4.964240639461506, "step": 5900}, {"loss": 1.4456, "grad_norm": 0.7160256505012512, "learning_rate": 0.0002, "epoch": 4.972654606647034, "step": 5910}, {"loss": 1.3874, "grad_norm": 0.7020420432090759, "learning_rate": 0.0002, "epoch": 4.981068573832562, "step": 5920}, {"loss": 1.4037, "grad_norm": 0.7576286792755127, "learning_rate": 0.0002, "epoch": 4.98948254101809, "step": 5930}, {"loss": 1.414, "grad_norm": 0.8573036789894104, "learning_rate": 0.0002, "epoch": 4.997896508203618, "step": 5940}, {"eval_loss": 1.9353811740875244, "eval_runtime": 37.9208, "eval_samples_per_second": 13.581, "eval_steps_per_second": 1.714, "epoch": 4.999579301640724, "step": 5942}, {"loss": 1.2418, "grad_norm": 0.8204267621040344, "learning_rate": 0.0002, "epoch": 5.006310475389146, "step": 5950}, {"loss": 1.235, "grad_norm": 0.976840615272522, "learning_rate": 0.0002, "epoch": 5.014724442574674, "step": 5960}, {"loss": 1.2134, "grad_norm": 0.8765613436698914, "learning_rate": 0.0002, "epoch": 5.023138409760202, "step": 5970}, {"loss": 1.2748, "grad_norm": 1.1793042421340942, "learning_rate": 0.0002, "epoch": 5.03155237694573, "step": 5980}, {"loss": 1.2412, "grad_norm": 0.971062958240509, "learning_rate": 0.0002, "epoch": 5.039966344131258, "step": 5990}, {"loss": 1.1819, "grad_norm": 0.8649757504463196, "learning_rate": 0.0002, "epoch": 5.0483803113167856, "step": 6000}, {"loss": 1.1654, "grad_norm": 0.9563034176826477, "learning_rate": 0.0002, "epoch": 5.056794278502314, "step": 6010}, {"loss": 1.2238, "grad_norm": 1.0093994140625, "learning_rate": 0.0002, "epoch": 5.065208245687842, "step": 6020}, {"loss": 1.2519, "grad_norm": 1.004213571548462, "learning_rate": 0.0002, "epoch": 5.07362221287337, "step": 6030}, {"loss": 1.2379, "grad_norm": 0.8307787179946899, "learning_rate": 0.0002, "epoch": 5.082036180058898, "step": 6040}, {"loss": 1.2282, "grad_norm": 0.9117848873138428, "learning_rate": 0.0002, "epoch": 5.090450147244426, "step": 6050}, {"loss": 1.2582, "grad_norm": 1.0269840955734253, "learning_rate": 0.0002, "epoch": 5.098864114429953, "step": 6060}, {"loss": 1.1836, "grad_norm": 0.9079542756080627, "learning_rate": 0.0002, "epoch": 5.1072780816154815, "step": 6070}, {"loss": 1.215, "grad_norm": 0.885702908039093, "learning_rate": 0.0002, "epoch": 5.11569204880101, "step": 6080}, {"loss": 1.2406, "grad_norm": 0.9976128339767456, "learning_rate": 0.0002, "epoch": 5.124106015986538, "step": 6090}, {"loss": 1.3082, "grad_norm": 0.8472117185592651, "learning_rate": 0.0002, "epoch": 5.132519983172066, "step": 6100}, {"loss": 1.226, "grad_norm": 1.0385161638259888, "learning_rate": 0.0002, "epoch": 5.140933950357594, "step": 6110}, {"loss": 1.213, "grad_norm": 0.8948383927345276, "learning_rate": 0.0002, "epoch": 5.149347917543121, "step": 6120}, {"loss": 1.2213, "grad_norm": 1.2613716125488281, "learning_rate": 0.0002, "epoch": 5.157761884728649, "step": 6130}, {"loss": 1.2632, "grad_norm": 0.9933410286903381, "learning_rate": 0.0002, "epoch": 5.166175851914177, "step": 6140}, {"loss": 1.1715, "grad_norm": 0.9673663973808289, "learning_rate": 0.0002, "epoch": 5.1745898190997055, "step": 6150}, {"loss": 1.2947, "grad_norm": 0.9969648122787476, "learning_rate": 0.0002, "epoch": 5.183003786285234, "step": 6160}, {"loss": 1.2416, "grad_norm": 1.2163258790969849, "learning_rate": 0.0002, "epoch": 5.191417753470762, "step": 6170}, {"loss": 1.2221, "grad_norm": 0.9163419604301453, "learning_rate": 0.0002, "epoch": 5.19983172065629, "step": 6180}, {"loss": 1.2624, "grad_norm": 0.9225585460662842, "learning_rate": 0.0002, "epoch": 5.208245687841817, "step": 6190}, {"loss": 1.2932, "grad_norm": 0.9205296635627747, "learning_rate": 0.0002, "epoch": 5.216659655027345, "step": 6200}, {"loss": 1.1825, "grad_norm": 1.0655443668365479, "learning_rate": 0.0002, "epoch": 5.225073622212873, "step": 6210}, {"loss": 1.2613, "grad_norm": 1.0854865312576294, "learning_rate": 0.0002, "epoch": 5.233487589398401, "step": 6220}, {"loss": 1.3045, "grad_norm": 0.8489186763763428, "learning_rate": 0.0002, "epoch": 5.2419015565839295, "step": 6230}, {"loss": 1.2708, "grad_norm": 0.910391628742218, "learning_rate": 0.0002, "epoch": 5.250315523769458, "step": 6240}, {"loss": 1.1914, "grad_norm": 0.925507128238678, "learning_rate": 0.0002, "epoch": 5.258729490954985, "step": 6250}, {"loss": 1.3368, "grad_norm": 1.1069735288619995, "learning_rate": 0.0002, "epoch": 5.267143458140513, "step": 6260}, {"loss": 1.2505, "grad_norm": 0.9705119132995605, "learning_rate": 0.0002, "epoch": 5.275557425326041, "step": 6270}, {"loss": 1.2602, "grad_norm": 0.9752426147460938, "learning_rate": 0.0002, "epoch": 5.283971392511569, "step": 6280}, {"loss": 1.2043, "grad_norm": 1.021359920501709, "learning_rate": 0.0002, "epoch": 5.292385359697097, "step": 6290}, {"loss": 1.2848, "grad_norm": 1.148606300354004, "learning_rate": 0.0002, "epoch": 5.3007993268826255, "step": 6300}, {"loss": 1.2201, "grad_norm": 0.8909247517585754, "learning_rate": 0.0002, "epoch": 5.309213294068153, "step": 6310}, {"loss": 1.2376, "grad_norm": 0.9879156351089478, "learning_rate": 0.0002, "epoch": 5.317627261253681, "step": 6320}, {"loss": 1.2638, "grad_norm": 0.9473357200622559, "learning_rate": 0.0002, "epoch": 5.326041228439209, "step": 6330}, {"loss": 1.232, "grad_norm": 1.1422028541564941, "learning_rate": 0.0002, "epoch": 5.334455195624737, "step": 6340}, {"loss": 1.263, "grad_norm": 0.9942235350608826, "learning_rate": 0.0002, "epoch": 5.342869162810265, "step": 6350}, {"loss": 1.3032, "grad_norm": 0.9535723924636841, "learning_rate": 0.0002, "epoch": 5.351283129995793, "step": 6360}, {"loss": 1.2908, "grad_norm": 0.9020892381668091, "learning_rate": 0.0002, "epoch": 5.359697097181321, "step": 6370}, {"loss": 1.2023, "grad_norm": 1.0626472234725952, "learning_rate": 0.0002, "epoch": 5.368111064366849, "step": 6380}, {"loss": 1.2555, "grad_norm": 1.1395848989486694, "learning_rate": 0.0002, "epoch": 5.376525031552377, "step": 6390}, {"loss": 1.2839, "grad_norm": 0.9274451732635498, "learning_rate": 0.0002, "epoch": 5.384938998737905, "step": 6400}, {"loss": 1.2819, "grad_norm": 0.8108699917793274, "learning_rate": 0.0002, "epoch": 5.393352965923433, "step": 6410}, {"loss": 1.2589, "grad_norm": 1.1805564165115356, "learning_rate": 0.0002, "epoch": 5.401766933108961, "step": 6420}, {"loss": 1.3549, "grad_norm": 0.8321298360824585, "learning_rate": 0.0002, "epoch": 5.410180900294489, "step": 6430}, {"loss": 1.2925, "grad_norm": 0.8981925249099731, "learning_rate": 0.0002, "epoch": 5.418594867480017, "step": 6440}, {"loss": 1.258, "grad_norm": 1.0730986595153809, "learning_rate": 0.0002, "epoch": 5.4270088346655445, "step": 6450}, {"loss": 1.26, "grad_norm": 1.0584609508514404, "learning_rate": 0.0002, "epoch": 5.435422801851073, "step": 6460}, {"loss": 1.2847, "grad_norm": 1.0792299509048462, "learning_rate": 0.0002, "epoch": 5.443836769036601, "step": 6470}, {"loss": 1.2035, "grad_norm": 0.9101872444152832, "learning_rate": 0.0002, "epoch": 5.452250736222129, "step": 6480}, {"loss": 1.2574, "grad_norm": 0.9910100698471069, "learning_rate": 0.0002, "epoch": 5.460664703407657, "step": 6490}, {"loss": 1.3098, "grad_norm": 1.041412353515625, "learning_rate": 0.0002, "epoch": 5.469078670593185, "step": 6500}, {"loss": 1.2812, "grad_norm": 1.0091687440872192, "learning_rate": 0.0002, "epoch": 5.477492637778712, "step": 6510}, {"loss": 1.2523, "grad_norm": 0.8755383491516113, "learning_rate": 0.0002, "epoch": 5.48590660496424, "step": 6520}, {"loss": 1.3042, "grad_norm": 0.980212390422821, "learning_rate": 0.0002, "epoch": 5.4943205721497685, "step": 6530}, {"loss": 1.2873, "grad_norm": 0.9356869459152222, "learning_rate": 0.0002, "epoch": 5.502734539335297, "step": 6540}, {"loss": 1.2254, "grad_norm": 0.9008095264434814, "learning_rate": 0.0002, "epoch": 5.511148506520825, "step": 6550}, {"loss": 1.2818, "grad_norm": 0.8908938765525818, "learning_rate": 0.0002, "epoch": 5.519562473706353, "step": 6560}, {"loss": 1.2212, "grad_norm": 1.1423932313919067, "learning_rate": 0.0002, "epoch": 5.52797644089188, "step": 6570}, {"loss": 1.3039, "grad_norm": 1.0508161783218384, "learning_rate": 0.0002, "epoch": 5.536390408077408, "step": 6580}, {"loss": 1.2446, "grad_norm": 0.8357517719268799, "learning_rate": 0.0002, "epoch": 5.544804375262936, "step": 6590}, {"loss": 1.3037, "grad_norm": 0.9892540574073792, "learning_rate": 0.0002, "epoch": 5.5532183424484645, "step": 6600}, {"loss": 1.3028, "grad_norm": 1.0048326253890991, "learning_rate": 0.0002, "epoch": 5.561632309633993, "step": 6610}, {"loss": 1.2152, "grad_norm": 0.9801995158195496, "learning_rate": 0.0002, "epoch": 5.570046276819521, "step": 6620}, {"loss": 1.2606, "grad_norm": 0.9899214506149292, "learning_rate": 0.0002, "epoch": 5.578460244005049, "step": 6630}, {"loss": 1.2043, "grad_norm": 1.1911814212799072, "learning_rate": 0.0002, "epoch": 5.586874211190576, "step": 6640}, {"loss": 1.3458, "grad_norm": 1.0368894338607788, "learning_rate": 0.0002, "epoch": 5.595288178376104, "step": 6650}, {"loss": 1.2595, "grad_norm": 1.1248382329940796, "learning_rate": 0.0002, "epoch": 5.603702145561632, "step": 6660}, {"loss": 1.2548, "grad_norm": 0.9765539765357971, "learning_rate": 0.0002, "epoch": 5.61211611274716, "step": 6670}, {"loss": 1.3451, "grad_norm": 0.9810206890106201, "learning_rate": 0.0002, "epoch": 5.6205300799326885, "step": 6680}, {"loss": 1.2952, "grad_norm": 1.100386619567871, "learning_rate": 0.0002, "epoch": 5.628944047118217, "step": 6690}, {"loss": 1.2467, "grad_norm": 0.8824519515037537, "learning_rate": 0.0002, "epoch": 5.637358014303744, "step": 6700}, {"loss": 1.25, "grad_norm": 1.0864064693450928, "learning_rate": 0.0002, "epoch": 5.645771981489272, "step": 6710}, {"loss": 1.2479, "grad_norm": 1.1614511013031006, "learning_rate": 0.0002, "epoch": 5.6541859486748, "step": 6720}, {"loss": 1.2753, "grad_norm": 1.0762972831726074, "learning_rate": 0.0002, "epoch": 5.662599915860328, "step": 6730}, {"loss": 1.2741, "grad_norm": 0.9408974647521973, "learning_rate": 0.0002, "epoch": 5.671013883045856, "step": 6740}, {"loss": 1.2431, "grad_norm": 0.8906030058860779, "learning_rate": 0.0002, "epoch": 5.679427850231384, "step": 6750}, {"loss": 1.2643, "grad_norm": 0.9527303576469421, "learning_rate": 0.0002, "epoch": 5.687841817416912, "step": 6760}, {"loss": 1.322, "grad_norm": 0.9471196532249451, "learning_rate": 0.0002, "epoch": 5.69625578460244, "step": 6770}, {"loss": 1.2514, "grad_norm": 0.9186838865280151, "learning_rate": 0.0002, "epoch": 5.704669751787968, "step": 6780}, {"loss": 1.2347, "grad_norm": 0.9225441813468933, "learning_rate": 0.0002, "epoch": 5.713083718973496, "step": 6790}, {"loss": 1.1849, "grad_norm": 0.9712982773780823, "learning_rate": 0.0002, "epoch": 5.721497686159024, "step": 6800}, {"loss": 1.2431, "grad_norm": 1.0743170976638794, "learning_rate": 0.0002, "epoch": 5.729911653344552, "step": 6810}, {"loss": 1.2136, "grad_norm": 1.2738113403320312, "learning_rate": 0.0002, "epoch": 5.73832562053008, "step": 6820}, {"loss": 1.2176, "grad_norm": 0.9386790990829468, "learning_rate": 0.0002, "epoch": 5.7467395877156076, "step": 6830}, {"loss": 1.285, "grad_norm": 1.0817769765853882, "learning_rate": 0.0002, "epoch": 5.755153554901136, "step": 6840}, {"loss": 1.2247, "grad_norm": 1.1040263175964355, "learning_rate": 0.0002, "epoch": 5.763567522086664, "step": 6850}, {"loss": 1.2507, "grad_norm": 1.0656492710113525, "learning_rate": 0.0002, "epoch": 5.771981489272192, "step": 6860}, {"loss": 1.2999, "grad_norm": 0.9550157189369202, "learning_rate": 0.0002, "epoch": 5.78039545645772, "step": 6870}, {"loss": 1.3201, "grad_norm": 1.0130870342254639, "learning_rate": 0.0002, "epoch": 5.788809423643248, "step": 6880}, {"loss": 1.3392, "grad_norm": 1.0675787925720215, "learning_rate": 0.0002, "epoch": 5.797223390828776, "step": 6890}, {"loss": 1.2949, "grad_norm": 0.9537774920463562, "learning_rate": 0.0002, "epoch": 5.8056373580143035, "step": 6900}, {"loss": 1.2658, "grad_norm": 0.9640319347381592, "learning_rate": 0.0002, "epoch": 5.814051325199832, "step": 6910}, {"loss": 1.2199, "grad_norm": 0.8917992115020752, "learning_rate": 0.0002, "epoch": 5.82246529238536, "step": 6920}, {"loss": 1.373, "grad_norm": 0.9881822466850281, "learning_rate": 0.0002, "epoch": 5.830879259570888, "step": 6930}, {"loss": 1.323, "grad_norm": 0.9136882424354553, "learning_rate": 0.0002, "epoch": 5.839293226756416, "step": 6940}, {"loss": 1.3159, "grad_norm": 0.9086098074913025, "learning_rate": 0.0002, "epoch": 5.847707193941943, "step": 6950}, {"loss": 1.2624, "grad_norm": 0.9443018436431885, "learning_rate": 0.0002, "epoch": 5.856121161127471, "step": 6960}, {"loss": 1.3224, "grad_norm": 0.9915381669998169, "learning_rate": 0.0002, "epoch": 5.864535128312999, "step": 6970}, {"loss": 1.337, "grad_norm": 0.8939146995544434, "learning_rate": 0.0002, "epoch": 5.8729490954985275, "step": 6980}, {"loss": 1.2611, "grad_norm": 1.3672245740890503, "learning_rate": 0.0002, "epoch": 5.881363062684056, "step": 6990}, {"loss": 1.3012, "grad_norm": 1.0116257667541504, "learning_rate": 0.0002, "epoch": 5.889777029869584, "step": 7000}, {"loss": 1.3128, "grad_norm": 1.1561565399169922, "learning_rate": 0.0002, "epoch": 5.898190997055112, "step": 7010}, {"loss": 1.2301, "grad_norm": 0.9900678992271423, "learning_rate": 0.0002, "epoch": 5.906604964240639, "step": 7020}, {"loss": 1.2845, "grad_norm": 0.9297345876693726, "learning_rate": 0.0002, "epoch": 5.915018931426167, "step": 7030}, {"loss": 1.2317, "grad_norm": 0.9357825517654419, "learning_rate": 0.0002, "epoch": 5.923432898611695, "step": 7040}, {"loss": 1.2303, "grad_norm": 1.049317717552185, "learning_rate": 0.0002, "epoch": 5.931846865797223, "step": 7050}, {"loss": 1.3243, "grad_norm": 0.950633704662323, "learning_rate": 0.0002, "epoch": 5.9402608329827515, "step": 7060}, {"loss": 1.2758, "grad_norm": 0.854581892490387, "learning_rate": 0.0002, "epoch": 5.94867480016828, "step": 7070}, {"loss": 1.3252, "grad_norm": 0.9097039699554443, "learning_rate": 0.0002, "epoch": 5.957088767353808, "step": 7080}, {"loss": 1.291, "grad_norm": 0.9072173237800598, "learning_rate": 0.0002, "epoch": 5.965502734539335, "step": 7090}, {"loss": 1.2724, "grad_norm": 1.0470727682113647, "learning_rate": 0.0002, "epoch": 5.973916701724863, "step": 7100}, {"loss": 1.3324, "grad_norm": 1.2628462314605713, "learning_rate": 0.0002, "epoch": 5.982330668910391, "step": 7110}, {"loss": 1.2701, "grad_norm": 1.055279016494751, "learning_rate": 0.0002, "epoch": 5.990744636095919, "step": 7120}, {"loss": 1.3234, "grad_norm": 0.966194212436676, "learning_rate": 0.0002, "epoch": 5.9991586032814475, "step": 7130}, {"eval_loss": 2.0427448749542236, "eval_runtime": 37.8426, "eval_samples_per_second": 13.609, "eval_steps_per_second": 1.718, "epoch": 6.0, "step": 7131}, {"loss": 1.1308, "grad_norm": 1.4037928581237793, "learning_rate": 0.0002, "epoch": 6.007572570466976, "step": 7140}, {"loss": 1.047, "grad_norm": 1.1081010103225708, "learning_rate": 0.0002, "epoch": 6.015986537652503, "step": 7150}, {"loss": 1.1368, "grad_norm": 1.1585499048233032, "learning_rate": 0.0002, "epoch": 6.024400504838031, "step": 7160}, {"loss": 1.0192, "grad_norm": 1.0822780132293701, "learning_rate": 0.0002, "epoch": 6.032814472023559, "step": 7170}, {"loss": 1.0755, "grad_norm": 0.9662094712257385, "learning_rate": 0.0002, "epoch": 6.041228439209087, "step": 7180}, {"loss": 1.1366, "grad_norm": 1.063936710357666, "learning_rate": 0.0002, "epoch": 6.049642406394615, "step": 7190}, {"loss": 1.0121, "grad_norm": 1.0349032878875732, "learning_rate": 0.0002, "epoch": 6.058056373580143, "step": 7200}, {"loss": 1.0591, "grad_norm": 1.0312575101852417, "learning_rate": 0.0002, "epoch": 6.066470340765671, "step": 7210}, {"loss": 1.1824, "grad_norm": 1.1942846775054932, "learning_rate": 0.0002, "epoch": 6.074884307951199, "step": 7220}, {"loss": 1.1034, "grad_norm": 1.0816049575805664, "learning_rate": 0.0002, "epoch": 6.083298275136727, "step": 7230}, {"loss": 1.0859, "grad_norm": 0.9985513687133789, "learning_rate": 0.0002, "epoch": 6.091712242322255, "step": 7240}, {"loss": 1.0367, "grad_norm": 1.2573972940444946, "learning_rate": 0.0002, "epoch": 6.100126209507783, "step": 7250}, {"loss": 1.1051, "grad_norm": 1.1182395219802856, "learning_rate": 0.0002, "epoch": 6.108540176693311, "step": 7260}, {"loss": 1.1219, "grad_norm": 0.9679344296455383, "learning_rate": 0.0002, "epoch": 6.116954143878839, "step": 7270}, {"loss": 1.1192, "grad_norm": 1.0913981199264526, "learning_rate": 0.0002, "epoch": 6.1253681110643665, "step": 7280}, {"loss": 1.0411, "grad_norm": 1.1291013956069946, "learning_rate": 0.0002, "epoch": 6.133782078249895, "step": 7290}, {"loss": 1.0963, "grad_norm": 1.2679595947265625, "learning_rate": 0.0002, "epoch": 6.142196045435423, "step": 7300}, {"loss": 1.0875, "grad_norm": 1.2350026369094849, "learning_rate": 0.0002, "epoch": 6.150610012620951, "step": 7310}, {"loss": 1.1139, "grad_norm": 1.3213104009628296, "learning_rate": 0.0002, "epoch": 6.159023979806479, "step": 7320}, {"loss": 1.1167, "grad_norm": 1.1924850940704346, "learning_rate": 0.0002, "epoch": 6.167437946992007, "step": 7330}, {"loss": 1.1242, "grad_norm": 1.1890000104904175, "learning_rate": 0.0002, "epoch": 6.175851914177534, "step": 7340}, {"loss": 1.1341, "grad_norm": 1.3821455240249634, "learning_rate": 0.0002, "epoch": 6.184265881363062, "step": 7350}, {"loss": 1.0748, "grad_norm": 1.1217057704925537, "learning_rate": 0.0002, "epoch": 6.1926798485485905, "step": 7360}, {"loss": 1.159, "grad_norm": 1.2441548109054565, "learning_rate": 0.0002, "epoch": 6.201093815734119, "step": 7370}, {"loss": 1.1199, "grad_norm": 1.0837615728378296, "learning_rate": 0.0002, "epoch": 6.209507782919647, "step": 7380}, {"loss": 1.1641, "grad_norm": 1.164304256439209, "learning_rate": 0.0002, "epoch": 6.217921750105175, "step": 7390}, {"loss": 1.1325, "grad_norm": 1.3129467964172363, "learning_rate": 0.0002, "epoch": 6.226335717290702, "step": 7400}, {"loss": 1.1537, "grad_norm": 1.1938153505325317, "learning_rate": 0.0002, "epoch": 6.23474968447623, "step": 7410}, {"loss": 1.1238, "grad_norm": 1.4348443746566772, "learning_rate": 0.0002, "epoch": 6.243163651661758, "step": 7420}, {"loss": 1.0778, "grad_norm": 1.132301926612854, "learning_rate": 0.0002, "epoch": 6.2515776188472865, "step": 7430}, {"loss": 1.1148, "grad_norm": 1.136966586112976, "learning_rate": 0.0002, "epoch": 6.259991586032815, "step": 7440}, {"loss": 1.096, "grad_norm": 1.12801194190979, "learning_rate": 0.0002, "epoch": 6.268405553218343, "step": 7450}, {"loss": 1.0408, "grad_norm": 1.0246902704238892, "learning_rate": 0.0002, "epoch": 6.276819520403871, "step": 7460}, {"loss": 1.0389, "grad_norm": 1.1066974401474, "learning_rate": 0.0002, "epoch": 6.285233487589398, "step": 7470}, {"loss": 1.1589, "grad_norm": 1.012710690498352, "learning_rate": 0.0002, "epoch": 6.293647454774926, "step": 7480}, {"loss": 1.1049, "grad_norm": 1.2227119207382202, "learning_rate": 0.0002, "epoch": 6.302061421960454, "step": 7490}, {"loss": 1.1376, "grad_norm": 0.9736923575401306, "learning_rate": 0.0002, "epoch": 6.310475389145982, "step": 7500}, {"loss": 1.1017, "grad_norm": 1.2945268154144287, "learning_rate": 0.0002, "epoch": 6.3188893563315105, "step": 7510}, {"loss": 1.0724, "grad_norm": 1.1579312086105347, "learning_rate": 0.0002, "epoch": 6.327303323517039, "step": 7520}, {"loss": 1.0899, "grad_norm": 1.2404558658599854, "learning_rate": 0.0002, "epoch": 6.335717290702567, "step": 7530}, {"loss": 1.1635, "grad_norm": 1.4673258066177368, "learning_rate": 0.0002, "epoch": 6.344131257888094, "step": 7540}, {"loss": 1.128, "grad_norm": 1.2268997430801392, "learning_rate": 0.0002, "epoch": 6.352545225073622, "step": 7550}, {"loss": 1.0932, "grad_norm": 0.9772747159004211, "learning_rate": 0.0002, "epoch": 6.36095919225915, "step": 7560}, {"loss": 1.1214, "grad_norm": 1.0205204486846924, "learning_rate": 0.0002, "epoch": 6.369373159444678, "step": 7570}, {"loss": 1.1095, "grad_norm": 1.2227109670639038, "learning_rate": 0.0002, "epoch": 6.377787126630206, "step": 7580}, {"loss": 1.1115, "grad_norm": 1.0708507299423218, "learning_rate": 0.0002, "epoch": 6.3862010938157345, "step": 7590}, {"loss": 1.1018, "grad_norm": 1.1427522897720337, "learning_rate": 0.0002, "epoch": 6.394615061001262, "step": 7600}, {"loss": 1.1079, "grad_norm": 1.0706431865692139, "learning_rate": 0.0002, "epoch": 6.40302902818679, "step": 7610}, {"loss": 1.0933, "grad_norm": 1.1358282566070557, "learning_rate": 0.0002, "epoch": 6.411442995372318, "step": 7620}, {"loss": 1.1075, "grad_norm": 1.4011822938919067, "learning_rate": 0.0002, "epoch": 6.419856962557846, "step": 7630}, {"loss": 1.1269, "grad_norm": 1.5616450309753418, "learning_rate": 0.0002, "epoch": 6.428270929743374, "step": 7640}, {"loss": 1.0953, "grad_norm": 1.1442687511444092, "learning_rate": 0.0002, "epoch": 6.436684896928902, "step": 7650}, {"loss": 1.1341, "grad_norm": 1.164803147315979, "learning_rate": 0.0002, "epoch": 6.44509886411443, "step": 7660}, {"loss": 1.14, "grad_norm": 1.3184553384780884, "learning_rate": 0.0002, "epoch": 6.453512831299958, "step": 7670}, {"loss": 1.1526, "grad_norm": 1.2701894044876099, "learning_rate": 0.0002, "epoch": 6.461926798485486, "step": 7680}, {"loss": 1.2119, "grad_norm": 1.1998416185379028, "learning_rate": 0.0002, "epoch": 6.470340765671014, "step": 7690}, {"loss": 1.1528, "grad_norm": 1.156459927558899, "learning_rate": 0.0002, "epoch": 6.478754732856542, "step": 7700}, {"loss": 1.2122, "grad_norm": 1.0217190980911255, "learning_rate": 0.0002, "epoch": 6.48716870004207, "step": 7710}, {"loss": 1.0917, "grad_norm": 1.230372428894043, "learning_rate": 0.0002, "epoch": 6.495582667227598, "step": 7720}, {"loss": 1.119, "grad_norm": 1.105675220489502, "learning_rate": 0.0002, "epoch": 6.5039966344131255, "step": 7730}, {"loss": 1.0758, "grad_norm": 1.1623669862747192, "learning_rate": 0.0002, "epoch": 6.512410601598654, "step": 7740}, {"loss": 1.1548, "grad_norm": 1.2884684801101685, "learning_rate": 0.0002, "epoch": 6.520824568784182, "step": 7750}, {"loss": 1.142, "grad_norm": 1.1785279512405396, "learning_rate": 0.0002, "epoch": 6.52923853596971, "step": 7760}, {"loss": 1.1598, "grad_norm": 1.0607101917266846, "learning_rate": 0.0002, "epoch": 6.537652503155238, "step": 7770}, {"loss": 1.1472, "grad_norm": 1.21990168094635, "learning_rate": 0.0002, "epoch": 6.546066470340766, "step": 7780}, {"loss": 1.1468, "grad_norm": 1.1498621702194214, "learning_rate": 0.0002, "epoch": 6.554480437526293, "step": 7790}, {"loss": 1.1847, "grad_norm": 1.263929009437561, "learning_rate": 0.0002, "epoch": 6.562894404711821, "step": 7800}, {"loss": 1.1177, "grad_norm": 1.1580625772476196, "learning_rate": 0.0002, "epoch": 6.5713083718973495, "step": 7810}, {"loss": 1.1313, "grad_norm": 1.4431294202804565, "learning_rate": 0.0002, "epoch": 6.579722339082878, "step": 7820}, {"loss": 1.1944, "grad_norm": 1.1309990882873535, "learning_rate": 0.0002, "epoch": 6.588136306268406, "step": 7830}, {"loss": 1.1156, "grad_norm": 1.0543386936187744, "learning_rate": 0.0002, "epoch": 6.596550273453934, "step": 7840}, {"loss": 1.0945, "grad_norm": 1.2180639505386353, "learning_rate": 0.0002, "epoch": 6.604964240639461, "step": 7850}, {"loss": 1.1318, "grad_norm": 1.0631271600723267, "learning_rate": 0.0002, "epoch": 6.613378207824989, "step": 7860}, {"loss": 1.1792, "grad_norm": 1.138885498046875, "learning_rate": 0.0002, "epoch": 6.621792175010517, "step": 7870}, {"loss": 1.1805, "grad_norm": 1.1117745637893677, "learning_rate": 0.0002, "epoch": 6.630206142196045, "step": 7880}, {"loss": 1.15, "grad_norm": 1.3734886646270752, "learning_rate": 0.0002, "epoch": 6.6386201093815735, "step": 7890}, {"loss": 1.1584, "grad_norm": 1.236003041267395, "learning_rate": 0.0002, "epoch": 6.647034076567102, "step": 7900}, {"loss": 1.1718, "grad_norm": 1.2206000089645386, "learning_rate": 0.0002, "epoch": 6.65544804375263, "step": 7910}, {"loss": 1.1637, "grad_norm": 1.2842656373977661, "learning_rate": 0.0002, "epoch": 6.663862010938157, "step": 7920}, {"loss": 1.2219, "grad_norm": 1.2365005016326904, "learning_rate": 0.0002, "epoch": 6.672275978123685, "step": 7930}, {"loss": 1.0827, "grad_norm": 1.256620168685913, "learning_rate": 0.0002, "epoch": 6.680689945309213, "step": 7940}, {"loss": 1.1788, "grad_norm": 1.3232917785644531, "learning_rate": 0.0002, "epoch": 6.689103912494741, "step": 7950}, {"loss": 1.2042, "grad_norm": 1.2470088005065918, "learning_rate": 0.0002, "epoch": 6.6975178796802695, "step": 7960}, {"loss": 1.0959, "grad_norm": 1.0511926412582397, "learning_rate": 0.0002, "epoch": 6.705931846865798, "step": 7970}, {"loss": 1.118, "grad_norm": 1.107310175895691, "learning_rate": 0.0002, "epoch": 6.714345814051326, "step": 7980}, {"loss": 1.2109, "grad_norm": 1.4069843292236328, "learning_rate": 0.0002, "epoch": 6.722759781236853, "step": 7990}, {"loss": 1.1298, "grad_norm": 1.0800836086273193, "learning_rate": 0.0002, "epoch": 6.731173748422381, "step": 8000}, {"loss": 1.1824, "grad_norm": 1.1676300764083862, "learning_rate": 0.0002, "epoch": 6.739587715607909, "step": 8010}, {"loss": 1.1253, "grad_norm": 1.0579663515090942, "learning_rate": 0.0002, "epoch": 6.748001682793437, "step": 8020}, {"loss": 1.1542, "grad_norm": 1.2770029306411743, "learning_rate": 0.0002, "epoch": 6.756415649978965, "step": 8030}, {"loss": 1.1519, "grad_norm": 1.0981038808822632, "learning_rate": 0.0002, "epoch": 6.764829617164493, "step": 8040}, {"loss": 1.1422, "grad_norm": 1.1194742918014526, "learning_rate": 0.0002, "epoch": 6.773243584350021, "step": 8050}, {"loss": 1.1463, "grad_norm": 1.0130012035369873, "learning_rate": 0.0002, "epoch": 6.781657551535549, "step": 8060}, {"loss": 1.2008, "grad_norm": 1.2051167488098145, "learning_rate": 0.0002, "epoch": 6.790071518721077, "step": 8070}, {"loss": 1.142, "grad_norm": 1.095689058303833, "learning_rate": 0.0002, "epoch": 6.798485485906605, "step": 8080}, {"loss": 1.1352, "grad_norm": 1.2275174856185913, "learning_rate": 0.0002, "epoch": 6.806899453092133, "step": 8090}, {"loss": 1.1453, "grad_norm": 1.1439805030822754, "learning_rate": 0.0002, "epoch": 6.815313420277661, "step": 8100}, {"loss": 1.1624, "grad_norm": 1.276331901550293, "learning_rate": 0.0002, "epoch": 6.8237273874631885, "step": 8110}, {"loss": 1.1686, "grad_norm": 1.0450139045715332, "learning_rate": 0.0002, "epoch": 6.832141354648717, "step": 8120}, {"loss": 1.1783, "grad_norm": 1.1189453601837158, "learning_rate": 0.0002, "epoch": 6.840555321834245, "step": 8130}, {"loss": 1.1093, "grad_norm": 1.194640874862671, "learning_rate": 0.0002, "epoch": 6.848969289019773, "step": 8140}, {"loss": 1.1559, "grad_norm": 1.095372200012207, "learning_rate": 0.0002, "epoch": 6.857383256205301, "step": 8150}, {"loss": 1.165, "grad_norm": 1.2416104078292847, "learning_rate": 0.0002, "epoch": 6.865797223390829, "step": 8160}, {"loss": 1.2174, "grad_norm": 1.2402868270874023, "learning_rate": 0.0002, "epoch": 6.874211190576357, "step": 8170}, {"loss": 1.1306, "grad_norm": 1.1317291259765625, "learning_rate": 0.0002, "epoch": 6.882625157761884, "step": 8180}, {"loss": 1.1944, "grad_norm": 1.0581914186477661, "learning_rate": 0.0002, "epoch": 6.8910391249474126, "step": 8190}, {"loss": 1.1271, "grad_norm": 1.3540890216827393, "learning_rate": 0.0002, "epoch": 6.899453092132941, "step": 8200}, {"loss": 1.2119, "grad_norm": 1.213672399520874, "learning_rate": 0.0002, "epoch": 6.907867059318469, "step": 8210}, {"loss": 1.1406, "grad_norm": 1.2654485702514648, "learning_rate": 0.0002, "epoch": 6.916281026503997, "step": 8220}, {"loss": 1.205, "grad_norm": 1.203903317451477, "learning_rate": 0.0002, "epoch": 6.924694993689524, "step": 8230}, {"loss": 1.1635, "grad_norm": 1.1332030296325684, "learning_rate": 0.0002, "epoch": 6.933108960875052, "step": 8240}, {"loss": 1.1148, "grad_norm": 1.2699192762374878, "learning_rate": 0.0002, "epoch": 6.94152292806058, "step": 8250}, {"loss": 1.1831, "grad_norm": 1.2728958129882812, "learning_rate": 0.0002, "epoch": 6.9499368952461085, "step": 8260}, {"loss": 1.1757, "grad_norm": 1.238410472869873, "learning_rate": 0.0002, "epoch": 6.958350862431637, "step": 8270}, {"loss": 1.1499, "grad_norm": 1.403863549232483, "learning_rate": 0.0002, "epoch": 6.966764829617165, "step": 8280}, {"loss": 1.1515, "grad_norm": 1.1096396446228027, "learning_rate": 0.0002, "epoch": 6.975178796802693, "step": 8290}, {"loss": 1.2049, "grad_norm": 1.1043379306793213, "learning_rate": 0.0002, "epoch": 6.98359276398822, "step": 8300}, {"loss": 1.1255, "grad_norm": 1.391754388809204, "learning_rate": 0.0002, "epoch": 6.992006731173748, "step": 8310}, {"eval_loss": 2.1421656608581543, "eval_runtime": 37.8262, "eval_samples_per_second": 13.615, "eval_steps_per_second": 1.718, "epoch": 6.999579301640724, "step": 8319}, {"loss": 1.1107, "grad_norm": 1.1739230155944824, "learning_rate": 0.0002, "epoch": 7.000420698359276, "step": 8320}, {"loss": 1.0066, "grad_norm": 1.5428645610809326, "learning_rate": 0.0002, "epoch": 7.008834665544804, "step": 8330}, {"loss": 0.9885, "grad_norm": 1.307463526725769, "learning_rate": 0.0002, "epoch": 7.0172486327303325, "step": 8340}, {"loss": 0.9098, "grad_norm": 1.4964789152145386, "learning_rate": 0.0002, "epoch": 7.025662599915861, "step": 8350}, {"loss": 0.8976, "grad_norm": 1.2289477586746216, "learning_rate": 0.0002, "epoch": 7.034076567101389, "step": 8360}, {"loss": 0.9254, "grad_norm": 1.325327754020691, "learning_rate": 0.0002, "epoch": 7.042490534286916, "step": 8370}, {"loss": 0.8967, "grad_norm": 1.4672988653182983, "learning_rate": 0.0002, "epoch": 7.050904501472444, "step": 8380}, {"loss": 0.8927, "grad_norm": 1.4184634685516357, "learning_rate": 0.0002, "epoch": 7.059318468657972, "step": 8390}, {"loss": 0.9129, "grad_norm": 1.3103536367416382, "learning_rate": 0.0002, "epoch": 7.0677324358435, "step": 8400}, {"loss": 0.997, "grad_norm": 1.2364518642425537, "learning_rate": 0.0002, "epoch": 7.076146403029028, "step": 8410}, {"loss": 0.8776, "grad_norm": 1.3712464570999146, "learning_rate": 0.0002, "epoch": 7.0845603702145565, "step": 8420}, {"loss": 0.9685, "grad_norm": 1.4655892848968506, "learning_rate": 0.0002, "epoch": 7.092974337400084, "step": 8430}, {"loss": 0.9276, "grad_norm": 1.3276227712631226, "learning_rate": 0.0002, "epoch": 7.101388304585612, "step": 8440}, {"loss": 0.9695, "grad_norm": 1.1355878114700317, "learning_rate": 0.0002, "epoch": 7.10980227177114, "step": 8450}, {"loss": 0.9673, "grad_norm": 1.2767117023468018, "learning_rate": 0.0002, "epoch": 7.118216238956668, "step": 8460}, {"loss": 0.9296, "grad_norm": 1.4915258884429932, "learning_rate": 0.0002, "epoch": 7.126630206142196, "step": 8470}, {"loss": 0.9469, "grad_norm": 1.355043649673462, "learning_rate": 0.0002, "epoch": 7.135044173327724, "step": 8480}, {"loss": 0.9626, "grad_norm": 1.0848617553710938, "learning_rate": 0.0002, "epoch": 7.143458140513252, "step": 8490}, {"loss": 0.92, "grad_norm": 1.5321701765060425, "learning_rate": 0.0002, "epoch": 7.15187210769878, "step": 8500}, {"loss": 0.9787, "grad_norm": 1.4917421340942383, "learning_rate": 0.0002, "epoch": 7.160286074884308, "step": 8510}, {"loss": 0.9709, "grad_norm": 1.4249778985977173, "learning_rate": 0.0002, "epoch": 7.168700042069836, "step": 8520}, {"loss": 0.9023, "grad_norm": 1.5257216691970825, "learning_rate": 0.0002, "epoch": 7.177114009255364, "step": 8530}, {"loss": 0.9818, "grad_norm": 1.4094327688217163, "learning_rate": 0.0002, "epoch": 7.185527976440892, "step": 8540}, {"loss": 0.9676, "grad_norm": 1.5506917238235474, "learning_rate": 0.0002, "epoch": 7.19394194362642, "step": 8550}, {"loss": 1.0494, "grad_norm": 1.336599588394165, "learning_rate": 0.0002, "epoch": 7.2023559108119475, "step": 8560}, {"loss": 0.9902, "grad_norm": 1.2018364667892456, "learning_rate": 0.0002, "epoch": 7.210769877997476, "step": 8570}, {"loss": 0.9329, "grad_norm": 1.198525071144104, "learning_rate": 0.0002, "epoch": 7.219183845183004, "step": 8580}, {"loss": 0.8954, "grad_norm": 1.4427133798599243, "learning_rate": 0.0002, "epoch": 7.227597812368532, "step": 8590}, {"loss": 0.9827, "grad_norm": 1.3134386539459229, "learning_rate": 0.0002, "epoch": 7.23601177955406, "step": 8600}, {"loss": 1.025, "grad_norm": 1.4141706228256226, "learning_rate": 0.0002, "epoch": 7.244425746739588, "step": 8610}, {"loss": 1.023, "grad_norm": 1.4951153993606567, "learning_rate": 0.0002, "epoch": 7.252839713925115, "step": 8620}, {"loss": 0.9595, "grad_norm": 1.383599042892456, "learning_rate": 0.0002, "epoch": 7.261253681110643, "step": 8630}, {"loss": 0.9775, "grad_norm": 1.2315951585769653, "learning_rate": 0.0002, "epoch": 7.2696676482961715, "step": 8640}, {"loss": 0.9946, "grad_norm": 1.253337025642395, "learning_rate": 0.0002, "epoch": 7.2780816154817, "step": 8650}, {"loss": 1.0381, "grad_norm": 1.2234476804733276, "learning_rate": 0.0002, "epoch": 7.286495582667228, "step": 8660}, {"loss": 0.9774, "grad_norm": 1.395650863647461, "learning_rate": 0.0002, "epoch": 7.294909549852756, "step": 8670}, {"loss": 0.9234, "grad_norm": 1.2411445379257202, "learning_rate": 0.0002, "epoch": 7.303323517038283, "step": 8680}, {"loss": 0.975, "grad_norm": 1.22808837890625, "learning_rate": 0.0002, "epoch": 7.311737484223811, "step": 8690}, {"loss": 0.9808, "grad_norm": 1.5197114944458008, "learning_rate": 0.0002, "epoch": 7.320151451409339, "step": 8700}, {"loss": 0.96, "grad_norm": 1.3072681427001953, "learning_rate": 0.0002, "epoch": 7.328565418594867, "step": 8710}, {"loss": 0.9386, "grad_norm": 1.3035615682601929, "learning_rate": 0.0002, "epoch": 7.3369793857803955, "step": 8720}, {"loss": 0.9666, "grad_norm": 1.2765713930130005, "learning_rate": 0.0002, "epoch": 7.345393352965924, "step": 8730}, {"loss": 0.9581, "grad_norm": 1.419601321220398, "learning_rate": 0.0002, "epoch": 7.353807320151452, "step": 8740}, {"loss": 1.0378, "grad_norm": 1.376158595085144, "learning_rate": 0.0002, "epoch": 7.362221287336979, "step": 8750}, {"loss": 0.9947, "grad_norm": 1.3880754709243774, "learning_rate": 0.0002, "epoch": 7.370635254522507, "step": 8760}, {"loss": 1.0512, "grad_norm": 1.2978262901306152, "learning_rate": 0.0002, "epoch": 7.379049221708035, "step": 8770}, {"loss": 1.0312, "grad_norm": 1.5811840295791626, "learning_rate": 0.0002, "epoch": 7.387463188893563, "step": 8780}, {"loss": 0.9977, "grad_norm": 1.3790863752365112, "learning_rate": 0.0002, "epoch": 7.3958771560790915, "step": 8790}, {"loss": 1.008, "grad_norm": 1.475306510925293, "learning_rate": 0.0002, "epoch": 7.40429112326462, "step": 8800}, {"loss": 0.9752, "grad_norm": 1.1038212776184082, "learning_rate": 0.0002, "epoch": 7.412705090450148, "step": 8810}, {"loss": 1.0048, "grad_norm": 1.5204451084136963, "learning_rate": 0.0002, "epoch": 7.421119057635675, "step": 8820}, {"loss": 1.019, "grad_norm": 1.7151343822479248, "learning_rate": 0.0002, "epoch": 7.429533024821203, "step": 8830}, {"loss": 1.0038, "grad_norm": 1.128046989440918, "learning_rate": 0.0002, "epoch": 7.437946992006731, "step": 8840}, {"loss": 1.0377, "grad_norm": 1.5780670642852783, "learning_rate": 0.0002, "epoch": 7.446360959192259, "step": 8850}, {"loss": 1.0584, "grad_norm": 1.3571979999542236, "learning_rate": 0.0002, "epoch": 7.454774926377787, "step": 8860}, {"loss": 1.0141, "grad_norm": 1.2764537334442139, "learning_rate": 0.0002, "epoch": 7.4631888935633155, "step": 8870}, {"loss": 0.9982, "grad_norm": 1.3429038524627686, "learning_rate": 0.0002, "epoch": 7.471602860748843, "step": 8880}, {"loss": 0.9671, "grad_norm": 1.3288369178771973, "learning_rate": 0.0002, "epoch": 7.480016827934371, "step": 8890}, {"loss": 0.9461, "grad_norm": 1.360141396522522, "learning_rate": 0.0002, "epoch": 7.488430795119899, "step": 8900}, {"loss": 1.0278, "grad_norm": 1.31229829788208, "learning_rate": 0.0002, "epoch": 7.496844762305427, "step": 8910}, {"loss": 0.9945, "grad_norm": 1.530605435371399, "learning_rate": 0.0002, "epoch": 7.505258729490955, "step": 8920}, {"loss": 1.0442, "grad_norm": 1.2880185842514038, "learning_rate": 0.0002, "epoch": 7.513672696676483, "step": 8930}, {"loss": 0.9859, "grad_norm": 1.3219470977783203, "learning_rate": 0.0002, "epoch": 7.5220866638620105, "step": 8940}, {"loss": 1.0664, "grad_norm": 1.565633773803711, "learning_rate": 0.0002, "epoch": 7.530500631047539, "step": 8950}, {"loss": 1.0089, "grad_norm": 1.4392317533493042, "learning_rate": 0.0002, "epoch": 7.538914598233067, "step": 8960}, {"loss": 1.0214, "grad_norm": 1.4557991027832031, "learning_rate": 0.0002, "epoch": 7.547328565418595, "step": 8970}, {"loss": 1.0247, "grad_norm": 1.3411110639572144, "learning_rate": 0.0002, "epoch": 7.555742532604123, "step": 8980}, {"loss": 1.0738, "grad_norm": 1.333378791809082, "learning_rate": 0.0002, "epoch": 7.564156499789651, "step": 8990}, {"loss": 1.0429, "grad_norm": 1.4422006607055664, "learning_rate": 0.0002, "epoch": 7.572570466975179, "step": 9000}, {"loss": 1.0401, "grad_norm": 1.2519633769989014, "learning_rate": 0.0002, "epoch": 7.580984434160706, "step": 9010}, {"loss": 1.0028, "grad_norm": 1.3628246784210205, "learning_rate": 0.0002, "epoch": 7.589398401346235, "step": 9020}, {"loss": 0.9883, "grad_norm": 1.35457181930542, "learning_rate": 0.0002, "epoch": 7.597812368531763, "step": 9030}, {"loss": 0.9929, "grad_norm": 1.4441956281661987, "learning_rate": 0.0002, "epoch": 7.606226335717291, "step": 9040}, {"loss": 0.9987, "grad_norm": 1.3812335729599, "learning_rate": 0.0002, "epoch": 7.614640302902819, "step": 9050}, {"loss": 0.9692, "grad_norm": 1.3576860427856445, "learning_rate": 0.0002, "epoch": 7.623054270088347, "step": 9060}, {"loss": 1.0259, "grad_norm": 1.350433588027954, "learning_rate": 0.0002, "epoch": 7.631468237273874, "step": 9070}, {"loss": 1.0292, "grad_norm": 1.3413814306259155, "learning_rate": 0.0002, "epoch": 7.639882204459402, "step": 9080}, {"loss": 1.016, "grad_norm": 1.2727786302566528, "learning_rate": 0.0002, "epoch": 7.6482961716449305, "step": 9090}, {"loss": 1.0046, "grad_norm": 1.1601275205612183, "learning_rate": 0.0002, "epoch": 7.656710138830459, "step": 9100}, {"loss": 1.1032, "grad_norm": 1.5492266416549683, "learning_rate": 0.0002, "epoch": 7.665124106015987, "step": 9110}, {"loss": 1.0174, "grad_norm": 1.4239033460617065, "learning_rate": 0.0002, "epoch": 7.673538073201515, "step": 9120}, {"loss": 0.9972, "grad_norm": 1.4212028980255127, "learning_rate": 0.0002, "epoch": 7.681952040387042, "step": 9130}, {"loss": 1.0802, "grad_norm": 1.116467833518982, "learning_rate": 0.0002, "epoch": 7.69036600757257, "step": 9140}, {"loss": 1.0311, "grad_norm": 1.299910545349121, "learning_rate": 0.0002, "epoch": 7.698779974758098, "step": 9150}, {"loss": 1.0262, "grad_norm": 1.404690146446228, "learning_rate": 0.0002, "epoch": 7.707193941943626, "step": 9160}, {"loss": 0.9633, "grad_norm": 1.383244276046753, "learning_rate": 0.0002, "epoch": 7.7156079091291545, "step": 9170}, {"loss": 1.0563, "grad_norm": 1.5001360177993774, "learning_rate": 0.0002, "epoch": 7.724021876314683, "step": 9180}, {"loss": 1.0731, "grad_norm": 1.4455186128616333, "learning_rate": 0.0002, "epoch": 7.732435843500211, "step": 9190}, {"loss": 1.0667, "grad_norm": 1.294964075088501, "learning_rate": 0.0002, "epoch": 7.740849810685738, "step": 9200}, {"loss": 0.9649, "grad_norm": 1.31305730342865, "learning_rate": 0.0002, "epoch": 7.749263777871266, "step": 9210}, {"loss": 0.9883, "grad_norm": 1.3849674463272095, "learning_rate": 0.0002, "epoch": 7.757677745056794, "step": 9220}, {"loss": 1.0219, "grad_norm": 1.6689352989196777, "learning_rate": 0.0002, "epoch": 7.766091712242322, "step": 9230}, {"loss": 1.03, "grad_norm": 1.416099190711975, "learning_rate": 0.0002, "epoch": 7.77450567942785, "step": 9240}, {"loss": 1.0429, "grad_norm": 1.5212045907974243, "learning_rate": 0.0002, "epoch": 7.7829196466133785, "step": 9250}, {"loss": 1.0607, "grad_norm": 1.3623390197753906, "learning_rate": 0.0002, "epoch": 7.791333613798907, "step": 9260}, {"loss": 1.0469, "grad_norm": 1.304148554801941, "learning_rate": 0.0002, "epoch": 7.799747580984434, "step": 9270}, {"loss": 1.0316, "grad_norm": 1.3833202123641968, "learning_rate": 0.0002, "epoch": 7.808161548169962, "step": 9280}, {"loss": 1.0122, "grad_norm": 1.3440886735916138, "learning_rate": 0.0002, "epoch": 7.81657551535549, "step": 9290}, {"loss": 1.0268, "grad_norm": 1.2798155546188354, "learning_rate": 0.0002, "epoch": 7.824989482541018, "step": 9300}, {"loss": 1.0521, "grad_norm": 1.3755156993865967, "learning_rate": 0.0002, "epoch": 7.833403449726546, "step": 9310}, {"loss": 1.0571, "grad_norm": 1.3145397901535034, "learning_rate": 0.0002, "epoch": 7.841817416912074, "step": 9320}, {"loss": 1.0684, "grad_norm": 1.6102794408798218, "learning_rate": 0.0002, "epoch": 7.850231384097602, "step": 9330}, {"loss": 1.1878, "grad_norm": 1.3959331512451172, "learning_rate": 0.0002, "epoch": 7.85864535128313, "step": 9340}, {"loss": 1.05, "grad_norm": 1.4965628385543823, "learning_rate": 0.0002, "epoch": 7.867059318468658, "step": 9350}, {"loss": 1.085, "grad_norm": 1.194201946258545, "learning_rate": 0.0002, "epoch": 7.875473285654186, "step": 9360}, {"loss": 1.0712, "grad_norm": 1.4831446409225464, "learning_rate": 0.0002, "epoch": 7.883887252839714, "step": 9370}, {"loss": 1.0568, "grad_norm": 1.3473302125930786, "learning_rate": 0.0002, "epoch": 7.892301220025242, "step": 9380}, {"loss": 1.0172, "grad_norm": 1.4373382329940796, "learning_rate": 0.0002, "epoch": 7.9007151872107695, "step": 9390}, {"loss": 0.9892, "grad_norm": 1.4341524839401245, "learning_rate": 0.0002, "epoch": 7.909129154396298, "step": 9400}, {"loss": 1.0428, "grad_norm": 1.3210171461105347, "learning_rate": 0.0002, "epoch": 7.917543121581826, "step": 9410}, {"loss": 1.0543, "grad_norm": 1.2708462476730347, "learning_rate": 0.0002, "epoch": 7.925957088767354, "step": 9420}, {"loss": 1.0789, "grad_norm": 1.4132758378982544, "learning_rate": 0.0002, "epoch": 7.934371055952882, "step": 9430}, {"loss": 1.095, "grad_norm": 1.5193610191345215, "learning_rate": 0.0002, "epoch": 7.94278502313841, "step": 9440}, {"loss": 0.967, "grad_norm": 1.427832841873169, "learning_rate": 0.0002, "epoch": 7.951198990323938, "step": 9450}, {"loss": 1.0052, "grad_norm": 1.380478024482727, "learning_rate": 0.0002, "epoch": 7.959612957509465, "step": 9460}, {"loss": 1.1032, "grad_norm": 1.3083926439285278, "learning_rate": 0.0002, "epoch": 7.9680269246949935, "step": 9470}, {"loss": 1.0883, "grad_norm": 1.3049120903015137, "learning_rate": 0.0002, "epoch": 7.976440891880522, "step": 9480}, {"loss": 1.0123, "grad_norm": 1.42048978805542, "learning_rate": 0.0002, "epoch": 7.98485485906605, "step": 9490}, {"loss": 1.094, "grad_norm": 1.2492578029632568, "learning_rate": 0.0002, "epoch": 7.993268826251578, "step": 9500}]}