diff --git a/.gitattributes b/.gitattributes
index e7c9b529c9d0119cdbfc2f190fb6ee13d1624895..f1c84864ebdf27f0f951808b46e41461ed9fae79 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -938,3 +938,12 @@ gemma-2b-it_int4_arc_challenge-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32
 gemma-2b-it_int4_arc_challenge-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-308-sd-42/checkpoint-64/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2b-it_int4_arc_challenge-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-308-sd-42/checkpoint-96/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2b-it_int4_arc_challenge-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.3-num-308-sd-42/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-sd-42/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f512b3457d4ef0662b10279451f78e55458a70a9
--- /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-0.9-num-6326-sd-42/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8767e0d0604ccff5925fd06b1ebf5a468436644640dabefecdff8b512facf8aa
+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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..118a6342a9d425a40112200b0ade39aeb26e6074
--- /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-0.9-num-6326-sd-42/checkpoint-10704/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abcdfcac6f09b8d04b847f0786435c773d18d77630561c9d325f90ae90a09178
+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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1ef2f91a640c4bd3bb25d9cc246e46c9c1b847b2
--- /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-0.9-num-6326-sd-42/checkpoint-10704/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4baeb9cc32972d690c5b7a9b98b0eb28c6b8afd2ce9c908058754805719d6dd
+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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..b65f38e866f2e7644f8c295eb82181a90805765c
--- /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-0.9-num-6326-sd-42/checkpoint-10704/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26f3421092a5ba4c2db1ce51d3b93698e8a2c6e0239127b2a377690d0d1f4cc1
+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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ddf6305008abf7ffbeb48eb5fde5e3982aded719
--- /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-0.9-num-6326-sd-42/checkpoint-10704/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9340a5bf0cf4c7a2611e19a4b084d5b6085acada4269b8e0cfcdade92431d60d
+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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..3552e32a37010bd9bf54689b68d9f664311ca981
--- /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-0.9-num-6326-sd-42/checkpoint-10704/trainer_state.json
@@ -0,0 +1,7587 @@
+{
+  "best_metric": 1.8046749830245972,
+  "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-0.9-num-6326-sd-42/checkpoint-2677",
+  "epoch": 7.997011580127008,
+  "eval_steps": 10,
+  "global_step": 10704,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007471049682480389,
+      "grad_norm": 0.4912872612476349,
+      "learning_rate": 0.0002,
+      "loss": 2.6181,
+      "step": 10
+    },
+    {
+      "epoch": 0.014942099364960777,
+      "grad_norm": 0.4856316149234772,
+      "learning_rate": 0.0002,
+      "loss": 2.2606,
+      "step": 20
+    },
+    {
+      "epoch": 0.022413149047441166,
+      "grad_norm": 0.47683125734329224,
+      "learning_rate": 0.0002,
+      "loss": 2.0957,
+      "step": 30
+    },
+    {
+      "epoch": 0.029884198729921554,
+      "grad_norm": 0.515082597732544,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 40
+    },
+    {
+      "epoch": 0.03735524841240194,
+      "grad_norm": 0.5299215316772461,
+      "learning_rate": 0.0002,
+      "loss": 1.9704,
+      "step": 50
+    },
+    {
+      "epoch": 0.04482629809488233,
+      "grad_norm": 0.4951399862766266,
+      "learning_rate": 0.0002,
+      "loss": 1.9225,
+      "step": 60
+    },
+    {
+      "epoch": 0.05229734777736272,
+      "grad_norm": 0.48079821467399597,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05976839745984311,
+      "grad_norm": 0.49402132630348206,
+      "learning_rate": 0.0002,
+      "loss": 1.9466,
+      "step": 80
+    },
+    {
+      "epoch": 0.0672394471423235,
+      "grad_norm": 0.4778193235397339,
+      "learning_rate": 0.0002,
+      "loss": 1.8691,
+      "step": 90
+    },
+    {
+      "epoch": 0.07471049682480388,
+      "grad_norm": 0.42472657561302185,
+      "learning_rate": 0.0002,
+      "loss": 1.8455,
+      "step": 100
+    },
+    {
+      "epoch": 0.08218154650728428,
+      "grad_norm": 0.4433092474937439,
+      "learning_rate": 0.0002,
+      "loss": 1.8744,
+      "step": 110
+    },
+    {
+      "epoch": 0.08965259618976466,
+      "grad_norm": 0.4472862780094147,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 120
+    },
+    {
+      "epoch": 0.09712364587224505,
+      "grad_norm": 0.42596298456192017,
+      "learning_rate": 0.0002,
+      "loss": 1.9256,
+      "step": 130
+    },
+    {
+      "epoch": 0.10459469555472543,
+      "grad_norm": 0.46645811200141907,
+      "learning_rate": 0.0002,
+      "loss": 1.8015,
+      "step": 140
+    },
+    {
+      "epoch": 0.11206574523720583,
+      "grad_norm": 0.41041234135627747,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 150
+    },
+    {
+      "epoch": 0.11953679491968622,
+      "grad_norm": 0.5329819917678833,
+      "learning_rate": 0.0002,
+      "loss": 1.8276,
+      "step": 160
+    },
+    {
+      "epoch": 0.1270078446021666,
+      "grad_norm": 0.4065922200679779,
+      "learning_rate": 0.0002,
+      "loss": 1.8118,
+      "step": 170
+    },
+    {
+      "epoch": 0.134478894284647,
+      "grad_norm": 0.38406994938850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8559,
+      "step": 180
+    },
+    {
+      "epoch": 0.14194994396712737,
+      "grad_norm": 0.4246881306171417,
+      "learning_rate": 0.0002,
+      "loss": 1.8647,
+      "step": 190
+    },
+    {
+      "epoch": 0.14942099364960776,
+      "grad_norm": 0.35136649012565613,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 200
+    },
+    {
+      "epoch": 0.15689204333208817,
+      "grad_norm": 0.43252742290496826,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.16436309301456856,
+      "grad_norm": 0.39236941933631897,
+      "learning_rate": 0.0002,
+      "loss": 1.7823,
+      "step": 220
+    },
+    {
+      "epoch": 0.17183414269704894,
+      "grad_norm": 0.3748249113559723,
+      "learning_rate": 0.0002,
+      "loss": 1.818,
+      "step": 230
+    },
+    {
+      "epoch": 0.17930519237952933,
+      "grad_norm": 0.6432855725288391,
+      "learning_rate": 0.0002,
+      "loss": 1.866,
+      "step": 240
+    },
+    {
+      "epoch": 0.1867762420620097,
+      "grad_norm": 0.34874802827835083,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 250
+    },
+    {
+      "epoch": 0.1942472917444901,
+      "grad_norm": 0.3721984326839447,
+      "learning_rate": 0.0002,
+      "loss": 1.79,
+      "step": 260
+    },
+    {
+      "epoch": 0.20171834142697048,
+      "grad_norm": 0.4339311420917511,
+      "learning_rate": 0.0002,
+      "loss": 1.8464,
+      "step": 270
+    },
+    {
+      "epoch": 0.20918939110945087,
+      "grad_norm": 0.4018215537071228,
+      "learning_rate": 0.0002,
+      "loss": 1.8665,
+      "step": 280
+    },
+    {
+      "epoch": 0.21666044079193125,
+      "grad_norm": 0.3278839886188507,
+      "learning_rate": 0.0002,
+      "loss": 1.8048,
+      "step": 290
+    },
+    {
+      "epoch": 0.22413149047441167,
+      "grad_norm": 0.36146077513694763,
+      "learning_rate": 0.0002,
+      "loss": 1.7395,
+      "step": 300
+    },
+    {
+      "epoch": 0.23160254015689205,
+      "grad_norm": 0.38175010681152344,
+      "learning_rate": 0.0002,
+      "loss": 1.7916,
+      "step": 310
+    },
+    {
+      "epoch": 0.23907358983937244,
+      "grad_norm": 0.44776618480682373,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 320
+    },
+    {
+      "epoch": 0.24654463952185282,
+      "grad_norm": 0.3933652937412262,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 330
+    },
+    {
+      "epoch": 0.2540156892043332,
+      "grad_norm": 0.3515005111694336,
+      "learning_rate": 0.0002,
+      "loss": 1.8393,
+      "step": 340
+    },
+    {
+      "epoch": 0.2614867388868136,
+      "grad_norm": 0.6683304309844971,
+      "learning_rate": 0.0002,
+      "loss": 1.8653,
+      "step": 350
+    },
+    {
+      "epoch": 0.268957788569294,
+      "grad_norm": 0.37093454599380493,
+      "learning_rate": 0.0002,
+      "loss": 1.8797,
+      "step": 360
+    },
+    {
+      "epoch": 0.2764288382517744,
+      "grad_norm": 0.3450651168823242,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 370
+    },
+    {
+      "epoch": 0.28389988793425475,
+      "grad_norm": 0.5140917301177979,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 380
+    },
+    {
+      "epoch": 0.29137093761673516,
+      "grad_norm": 0.32885563373565674,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 390
+    },
+    {
+      "epoch": 0.2988419872992155,
+      "grad_norm": 0.33962297439575195,
+      "learning_rate": 0.0002,
+      "loss": 1.8174,
+      "step": 400
+    },
+    {
+      "epoch": 0.30631303698169593,
+      "grad_norm": 0.3723141849040985,
+      "learning_rate": 0.0002,
+      "loss": 1.7467,
+      "step": 410
+    },
+    {
+      "epoch": 0.31378408666417634,
+      "grad_norm": 0.37173134088516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8459,
+      "step": 420
+    },
+    {
+      "epoch": 0.3212551363466567,
+      "grad_norm": 0.33736956119537354,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 430
+    },
+    {
+      "epoch": 0.3287261860291371,
+      "grad_norm": 0.3602448105812073,
+      "learning_rate": 0.0002,
+      "loss": 1.8367,
+      "step": 440
+    },
+    {
+      "epoch": 0.33619723571161747,
+      "grad_norm": 0.3569699227809906,
+      "learning_rate": 0.0002,
+      "loss": 1.8058,
+      "step": 450
+    },
+    {
+      "epoch": 0.3436682853940979,
+      "grad_norm": 0.31009167432785034,
+      "learning_rate": 0.0002,
+      "loss": 1.8086,
+      "step": 460
+    },
+    {
+      "epoch": 0.35113933507657824,
+      "grad_norm": 0.5278693437576294,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 470
+    },
+    {
+      "epoch": 0.35861038475905865,
+      "grad_norm": 0.3587537109851837,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 480
+    },
+    {
+      "epoch": 0.366081434441539,
+      "grad_norm": 0.3859670162200928,
+      "learning_rate": 0.0002,
+      "loss": 1.8046,
+      "step": 490
+    },
+    {
+      "epoch": 0.3735524841240194,
+      "grad_norm": 0.395913690328598,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 500
+    },
+    {
+      "epoch": 0.38102353380649984,
+      "grad_norm": 0.35052940249443054,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 510
+    },
+    {
+      "epoch": 0.3884945834889802,
+      "grad_norm": 0.2979494333267212,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 520
+    },
+    {
+      "epoch": 0.3959656331714606,
+      "grad_norm": 0.3062683343887329,
+      "learning_rate": 0.0002,
+      "loss": 1.8641,
+      "step": 530
+    },
+    {
+      "epoch": 0.40343668285394096,
+      "grad_norm": 0.3172847330570221,
+      "learning_rate": 0.0002,
+      "loss": 1.7651,
+      "step": 540
+    },
+    {
+      "epoch": 0.4109077325364214,
+      "grad_norm": 0.360435426235199,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 550
+    },
+    {
+      "epoch": 0.41837878221890173,
+      "grad_norm": 0.3427872359752655,
+      "learning_rate": 0.0002,
+      "loss": 1.9054,
+      "step": 560
+    },
+    {
+      "epoch": 0.42584983190138215,
+      "grad_norm": 0.34036558866500854,
+      "learning_rate": 0.0002,
+      "loss": 1.7562,
+      "step": 570
+    },
+    {
+      "epoch": 0.4333208815838625,
+      "grad_norm": 0.3365345299243927,
+      "learning_rate": 0.0002,
+      "loss": 1.7254,
+      "step": 580
+    },
+    {
+      "epoch": 0.4407919312663429,
+      "grad_norm": 0.35619041323661804,
+      "learning_rate": 0.0002,
+      "loss": 1.8328,
+      "step": 590
+    },
+    {
+      "epoch": 0.44826298094882333,
+      "grad_norm": 0.3569088280200958,
+      "learning_rate": 0.0002,
+      "loss": 1.8114,
+      "step": 600
+    },
+    {
+      "epoch": 0.4557340306313037,
+      "grad_norm": 0.3581278622150421,
+      "learning_rate": 0.0002,
+      "loss": 1.8599,
+      "step": 610
+    },
+    {
+      "epoch": 0.4632050803137841,
+      "grad_norm": 0.43197110295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.7078,
+      "step": 620
+    },
+    {
+      "epoch": 0.47067612999626446,
+      "grad_norm": 0.33966198563575745,
+      "learning_rate": 0.0002,
+      "loss": 1.8257,
+      "step": 630
+    },
+    {
+      "epoch": 0.47814717967874487,
+      "grad_norm": 0.3343866467475891,
+      "learning_rate": 0.0002,
+      "loss": 1.7528,
+      "step": 640
+    },
+    {
+      "epoch": 0.48561822936122523,
+      "grad_norm": 0.33878564834594727,
+      "learning_rate": 0.0002,
+      "loss": 1.8191,
+      "step": 650
+    },
+    {
+      "epoch": 0.49308927904370564,
+      "grad_norm": 0.387195885181427,
+      "learning_rate": 0.0002,
+      "loss": 1.8801,
+      "step": 660
+    },
+    {
+      "epoch": 0.500560328726186,
+      "grad_norm": 0.3755440413951874,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 670
+    },
+    {
+      "epoch": 0.5080313784086664,
+      "grad_norm": 0.3272816836833954,
+      "learning_rate": 0.0002,
+      "loss": 1.8057,
+      "step": 680
+    },
+    {
+      "epoch": 0.5155024280911468,
+      "grad_norm": 0.36063864827156067,
+      "learning_rate": 0.0002,
+      "loss": 1.8156,
+      "step": 690
+    },
+    {
+      "epoch": 0.5229734777736272,
+      "grad_norm": 0.35317373275756836,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 700
+    },
+    {
+      "epoch": 0.5304445274561076,
+      "grad_norm": 0.3561195433139801,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 710
+    },
+    {
+      "epoch": 0.537915577138588,
+      "grad_norm": 0.31124624609947205,
+      "learning_rate": 0.0002,
+      "loss": 1.8149,
+      "step": 720
+    },
+    {
+      "epoch": 0.5453866268210683,
+      "grad_norm": 0.3294544517993927,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 730
+    },
+    {
+      "epoch": 0.5528576765035488,
+      "grad_norm": 0.31933900713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.8027,
+      "step": 740
+    },
+    {
+      "epoch": 0.5603287261860291,
+      "grad_norm": 0.3226020634174347,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 750
+    },
+    {
+      "epoch": 0.5677997758685095,
+      "grad_norm": 0.3147525489330292,
+      "learning_rate": 0.0002,
+      "loss": 1.7862,
+      "step": 760
+    },
+    {
+      "epoch": 0.57527082555099,
+      "grad_norm": 0.32234328985214233,
+      "learning_rate": 0.0002,
+      "loss": 1.9028,
+      "step": 770
+    },
+    {
+      "epoch": 0.5827418752334703,
+      "grad_norm": 0.3258664309978485,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 780
+    },
+    {
+      "epoch": 0.5902129249159507,
+      "grad_norm": 0.3166961967945099,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 790
+    },
+    {
+      "epoch": 0.597683974598431,
+      "grad_norm": 0.35621458292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 800
+    },
+    {
+      "epoch": 0.6051550242809115,
+      "grad_norm": 0.3236999213695526,
+      "learning_rate": 0.0002,
+      "loss": 1.8313,
+      "step": 810
+    },
+    {
+      "epoch": 0.6126260739633919,
+      "grad_norm": 0.2892923653125763,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 820
+    },
+    {
+      "epoch": 0.6200971236458722,
+      "grad_norm": 0.4098321497440338,
+      "learning_rate": 0.0002,
+      "loss": 1.8709,
+      "step": 830
+    },
+    {
+      "epoch": 0.6275681733283527,
+      "grad_norm": 0.3337118923664093,
+      "learning_rate": 0.0002,
+      "loss": 1.7637,
+      "step": 840
+    },
+    {
+      "epoch": 0.635039223010833,
+      "grad_norm": 0.30416029691696167,
+      "learning_rate": 0.0002,
+      "loss": 1.7375,
+      "step": 850
+    },
+    {
+      "epoch": 0.6425102726933134,
+      "grad_norm": 0.3361026346683502,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 860
+    },
+    {
+      "epoch": 0.6499813223757938,
+      "grad_norm": 0.3537365198135376,
+      "learning_rate": 0.0002,
+      "loss": 1.732,
+      "step": 870
+    },
+    {
+      "epoch": 0.6574523720582742,
+      "grad_norm": 0.33854469656944275,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 880
+    },
+    {
+      "epoch": 0.6649234217407546,
+      "grad_norm": 0.3332272469997406,
+      "learning_rate": 0.0002,
+      "loss": 1.7561,
+      "step": 890
+    },
+    {
+      "epoch": 0.6723944714232349,
+      "grad_norm": 0.34954726696014404,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 900
+    },
+    {
+      "epoch": 0.6798655211057153,
+      "grad_norm": 0.2921750247478485,
+      "learning_rate": 0.0002,
+      "loss": 1.7917,
+      "step": 910
+    },
+    {
+      "epoch": 0.6873365707881958,
+      "grad_norm": 0.30508682131767273,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 920
+    },
+    {
+      "epoch": 0.6948076204706761,
+      "grad_norm": 0.32268425822257996,
+      "learning_rate": 0.0002,
+      "loss": 1.8082,
+      "step": 930
+    },
+    {
+      "epoch": 0.7022786701531565,
+      "grad_norm": 0.2844390869140625,
+      "learning_rate": 0.0002,
+      "loss": 1.8283,
+      "step": 940
+    },
+    {
+      "epoch": 0.709749719835637,
+      "grad_norm": 0.31263890862464905,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 950
+    },
+    {
+      "epoch": 0.7172207695181173,
+      "grad_norm": 0.3626808822154999,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 960
+    },
+    {
+      "epoch": 0.7246918192005977,
+      "grad_norm": 0.3322749733924866,
+      "learning_rate": 0.0002,
+      "loss": 1.853,
+      "step": 970
+    },
+    {
+      "epoch": 0.732162868883078,
+      "grad_norm": 0.29177871346473694,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 980
+    },
+    {
+      "epoch": 0.7396339185655585,
+      "grad_norm": 0.35405513644218445,
+      "learning_rate": 0.0002,
+      "loss": 1.8447,
+      "step": 990
+    },
+    {
+      "epoch": 0.7471049682480388,
+      "grad_norm": 0.39318400621414185,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7545760179305192,
+      "grad_norm": 0.29401418566703796,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7620470676129997,
+      "grad_norm": 0.3271748721599579,
+      "learning_rate": 0.0002,
+      "loss": 1.7649,
+      "step": 1020
+    },
+    {
+      "epoch": 0.76951811729548,
+      "grad_norm": 0.30883970856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7769891669779604,
+      "grad_norm": 0.3411838412284851,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7844602166604407,
+      "grad_norm": 0.30608129501342773,
+      "learning_rate": 0.0002,
+      "loss": 1.829,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7919312663429212,
+      "grad_norm": 0.30899080634117126,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7994023160254016,
+      "grad_norm": 0.3160453140735626,
+      "learning_rate": 0.0002,
+      "loss": 1.7625,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8068733657078819,
+      "grad_norm": 0.30947187542915344,
+      "learning_rate": 0.0002,
+      "loss": 1.8452,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8143444153903624,
+      "grad_norm": 0.3103134036064148,
+      "learning_rate": 0.0002,
+      "loss": 1.7418,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8218154650728428,
+      "grad_norm": 0.31771138310432434,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8292865147553231,
+      "grad_norm": 0.5860997438430786,
+      "learning_rate": 0.0002,
+      "loss": 1.7918,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8367575644378035,
+      "grad_norm": 0.3230148255825043,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8442286141202839,
+      "grad_norm": 0.29611510038375854,
+      "learning_rate": 0.0002,
+      "loss": 1.8478,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8516996638027643,
+      "grad_norm": 0.3373654782772064,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8591707134852447,
+      "grad_norm": 0.3474279046058655,
+      "learning_rate": 0.0002,
+      "loss": 1.7997,
+      "step": 1150
+    },
+    {
+      "epoch": 0.866641763167725,
+      "grad_norm": 0.35057875514030457,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8741128128502055,
+      "grad_norm": 0.39537495374679565,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8815838625326858,
+      "grad_norm": 0.3714233636856079,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8890549122151662,
+      "grad_norm": 0.2950296998023987,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8965259618976467,
+      "grad_norm": 0.38182979822158813,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 1200
+    },
+    {
+      "epoch": 0.903997011580127,
+      "grad_norm": 0.27883678674697876,
+      "learning_rate": 0.0002,
+      "loss": 1.827,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9114680612626074,
+      "grad_norm": 0.33874374628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9189391109450877,
+      "grad_norm": 0.3014272153377533,
+      "learning_rate": 0.0002,
+      "loss": 1.7334,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9264101606275682,
+      "grad_norm": 0.3194271922111511,
+      "learning_rate": 0.0002,
+      "loss": 1.8235,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9338812103100486,
+      "grad_norm": 0.3049403429031372,
+      "learning_rate": 0.0002,
+      "loss": 1.7924,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9413522599925289,
+      "grad_norm": 0.30621254444122314,
+      "learning_rate": 0.0002,
+      "loss": 1.7535,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9488233096750094,
+      "grad_norm": 0.28675132989883423,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9562943593574897,
+      "grad_norm": 0.3322032690048218,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9637654090399701,
+      "grad_norm": 0.35408294200897217,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9712364587224505,
+      "grad_norm": 0.36386919021606445,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9787075084049309,
+      "grad_norm": 0.32338324189186096,
+      "learning_rate": 0.0002,
+      "loss": 1.8633,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9861785580874113,
+      "grad_norm": 0.3714013993740082,
+      "learning_rate": 0.0002,
+      "loss": 1.7724,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9936496077698916,
+      "grad_norm": 0.3133082389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.7766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9996264475158759,
+      "eval_loss": 1.8051470518112183,
+      "eval_runtime": 38.6332,
+      "eval_samples_per_second": 13.331,
+      "eval_steps_per_second": 1.682,
+      "step": 1338
+    },
+    {
+      "epoch": 1.001120657452372,
+      "grad_norm": 0.31595754623413086,
+      "learning_rate": 0.0002,
+      "loss": 1.8035,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0085917071348525,
+      "grad_norm": 0.3095700144767761,
+      "learning_rate": 0.0002,
+      "loss": 1.7486,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0160627568173328,
+      "grad_norm": 0.34677496552467346,
+      "learning_rate": 0.0002,
+      "loss": 1.6981,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0235338064998132,
+      "grad_norm": 0.29108840227127075,
+      "learning_rate": 0.0002,
+      "loss": 1.7377,
+      "step": 1370
+    },
+    {
+      "epoch": 1.0310048561822935,
+      "grad_norm": 0.32356950640678406,
+      "learning_rate": 0.0002,
+      "loss": 1.7194,
+      "step": 1380
+    },
+    {
+      "epoch": 1.038475905864774,
+      "grad_norm": 0.4200669229030609,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0459469555472545,
+      "grad_norm": 0.3283711373806,
+      "learning_rate": 0.0002,
+      "loss": 1.797,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0534180052297348,
+      "grad_norm": 0.32898256182670593,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0608890549122152,
+      "grad_norm": 0.38790300488471985,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0683601045946955,
+      "grad_norm": 0.339800089597702,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1430
+    },
+    {
+      "epoch": 1.075831154277176,
+      "grad_norm": 0.3548751175403595,
+      "learning_rate": 0.0002,
+      "loss": 1.7076,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0833022039596563,
+      "grad_norm": 0.35114359855651855,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0907732536421366,
+      "grad_norm": 0.35226720571517944,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0982443033246172,
+      "grad_norm": 0.33665576577186584,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1057153530070976,
+      "grad_norm": 0.363889217376709,
+      "learning_rate": 0.0002,
+      "loss": 1.6699,
+      "step": 1480
+    },
+    {
+      "epoch": 1.113186402689578,
+      "grad_norm": 0.3826201856136322,
+      "learning_rate": 0.0002,
+      "loss": 1.7933,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1206574523720583,
+      "grad_norm": 0.34058740735054016,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1281285020545386,
+      "grad_norm": 0.3462134301662445,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 1510
+    },
+    {
+      "epoch": 1.135599551737019,
+      "grad_norm": 0.3396756052970886,
+      "learning_rate": 0.0002,
+      "loss": 1.7147,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1430706014194993,
+      "grad_norm": 0.32004743814468384,
+      "learning_rate": 0.0002,
+      "loss": 1.7219,
+      "step": 1530
+    },
+    {
+      "epoch": 1.15054165110198,
+      "grad_norm": 0.3397733271121979,
+      "learning_rate": 0.0002,
+      "loss": 1.743,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1580127007844603,
+      "grad_norm": 0.3783262073993683,
+      "learning_rate": 0.0002,
+      "loss": 1.7333,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1654837504669406,
+      "grad_norm": 0.35121291875839233,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 1560
+    },
+    {
+      "epoch": 1.172954800149421,
+      "grad_norm": 0.35816895961761475,
+      "learning_rate": 0.0002,
+      "loss": 1.678,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1804258498319014,
+      "grad_norm": 0.33843839168548584,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1878968995143817,
+      "grad_norm": 0.3371972143650055,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 1590
+    },
+    {
+      "epoch": 1.195367949196862,
+      "grad_norm": 0.36016878485679626,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2028389988793426,
+      "grad_norm": 0.40879473090171814,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 1610
+    },
+    {
+      "epoch": 1.210310048561823,
+      "grad_norm": 0.3216715455055237,
+      "learning_rate": 0.0002,
+      "loss": 1.6955,
+      "step": 1620
+    },
+    {
+      "epoch": 1.2177810982443034,
+      "grad_norm": 0.4482610821723938,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2252521479267837,
+      "grad_norm": 0.3257700502872467,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1640
+    },
+    {
+      "epoch": 1.232723197609264,
+      "grad_norm": 0.38646459579467773,
+      "learning_rate": 0.0002,
+      "loss": 1.7177,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2401942472917444,
+      "grad_norm": 0.4081360697746277,
+      "learning_rate": 0.0002,
+      "loss": 1.7081,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2476652969742248,
+      "grad_norm": 0.4326848089694977,
+      "learning_rate": 0.0002,
+      "loss": 1.7519,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2551363466567054,
+      "grad_norm": 0.346401572227478,
+      "learning_rate": 0.0002,
+      "loss": 1.6752,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2626073963391857,
+      "grad_norm": 0.34536251425743103,
+      "learning_rate": 0.0002,
+      "loss": 1.7425,
+      "step": 1690
+    },
+    {
+      "epoch": 1.270078446021666,
+      "grad_norm": 0.41359591484069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7061,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2775494957041464,
+      "grad_norm": 0.3530874252319336,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2850205453866268,
+      "grad_norm": 0.3702719211578369,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2924915950691072,
+      "grad_norm": 0.3703329563140869,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2999626447515875,
+      "grad_norm": 0.37919729948043823,
+      "learning_rate": 0.0002,
+      "loss": 1.7221,
+      "step": 1740
+    },
+    {
+      "epoch": 1.307433694434068,
+      "grad_norm": 0.32526856660842896,
+      "learning_rate": 0.0002,
+      "loss": 1.7859,
+      "step": 1750
+    },
+    {
+      "epoch": 1.3149047441165485,
+      "grad_norm": 0.36752620339393616,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3223757937990288,
+      "grad_norm": 0.3398192524909973,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3298468434815092,
+      "grad_norm": 0.37435585260391235,
+      "learning_rate": 0.0002,
+      "loss": 1.7492,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3373178931639895,
+      "grad_norm": 0.35793280601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3447889428464699,
+      "grad_norm": 0.35481882095336914,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3522599925289502,
+      "grad_norm": 0.3786393105983734,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3597310422114308,
+      "grad_norm": 0.33245593309402466,
+      "learning_rate": 0.0002,
+      "loss": 1.7169,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3672020918939112,
+      "grad_norm": 0.35388344526290894,
+      "learning_rate": 0.0002,
+      "loss": 1.7577,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3746731415763915,
+      "grad_norm": 0.3695325553417206,
+      "learning_rate": 0.0002,
+      "loss": 1.6968,
+      "step": 1840
+    },
+    {
+      "epoch": 1.382144191258872,
+      "grad_norm": 0.3683604598045349,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3896152409413522,
+      "grad_norm": 0.3753012418746948,
+      "learning_rate": 0.0002,
+      "loss": 1.7878,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3970862906238326,
+      "grad_norm": 0.3331069350242615,
+      "learning_rate": 0.0002,
+      "loss": 1.6969,
+      "step": 1870
+    },
+    {
+      "epoch": 1.404557340306313,
+      "grad_norm": 0.3877500295639038,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4120283899887935,
+      "grad_norm": 0.33525151014328003,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4194994396712737,
+      "grad_norm": 0.3697299659252167,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4269704893537543,
+      "grad_norm": 0.4029286205768585,
+      "learning_rate": 0.0002,
+      "loss": 1.6956,
+      "step": 1910
+    },
+    {
+      "epoch": 1.4344415390362346,
+      "grad_norm": 0.3596203029155731,
+      "learning_rate": 0.0002,
+      "loss": 1.6897,
+      "step": 1920
+    },
+    {
+      "epoch": 1.441912588718715,
+      "grad_norm": 0.450783908367157,
+      "learning_rate": 0.0002,
+      "loss": 1.7139,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4493836384011953,
+      "grad_norm": 0.3651481866836548,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4568546880836757,
+      "grad_norm": 0.3608424663543701,
+      "learning_rate": 0.0002,
+      "loss": 1.6637,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4643257377661563,
+      "grad_norm": 0.39684420824050903,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4717967874486364,
+      "grad_norm": 0.34618663787841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7514,
+      "step": 1970
+    },
+    {
+      "epoch": 1.479267837131117,
+      "grad_norm": 0.4150386452674866,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4867388868135973,
+      "grad_norm": 0.35500776767730713,
+      "learning_rate": 0.0002,
+      "loss": 1.7021,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4942099364960777,
+      "grad_norm": 0.344144344329834,
+      "learning_rate": 0.0002,
+      "loss": 1.7322,
+      "step": 2000
+    },
+    {
+      "epoch": 1.501680986178558,
+      "grad_norm": 0.3340149223804474,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5091520358610384,
+      "grad_norm": 0.37685006856918335,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 2020
+    },
+    {
+      "epoch": 1.516623085543519,
+      "grad_norm": 0.3699876368045807,
+      "learning_rate": 0.0002,
+      "loss": 1.8299,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5240941352259991,
+      "grad_norm": 0.3370307385921478,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5315651849084797,
+      "grad_norm": 0.37780630588531494,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 2050
+    },
+    {
+      "epoch": 1.53903623459096,
+      "grad_norm": 0.370259165763855,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5465072842734404,
+      "grad_norm": 0.3440011441707611,
+      "learning_rate": 0.0002,
+      "loss": 1.7398,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5539783339559208,
+      "grad_norm": 0.40382063388824463,
+      "learning_rate": 0.0002,
+      "loss": 1.7105,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5614493836384011,
+      "grad_norm": 0.38002029061317444,
+      "learning_rate": 0.0002,
+      "loss": 1.7071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5689204333208817,
+      "grad_norm": 0.3658451437950134,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5763914830033618,
+      "grad_norm": 0.354842871427536,
+      "learning_rate": 0.0002,
+      "loss": 1.7598,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5838625326858424,
+      "grad_norm": 0.34735530614852905,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5913335823683228,
+      "grad_norm": 0.377581924200058,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5988046320508031,
+      "grad_norm": 0.41254034638404846,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6062756817332835,
+      "grad_norm": 0.3630715310573578,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6137467314157639,
+      "grad_norm": 0.36980143189430237,
+      "learning_rate": 0.0002,
+      "loss": 1.7531,
+      "step": 2160
+    },
+    {
+      "epoch": 1.6212177810982444,
+      "grad_norm": 0.3634769320487976,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6286888307807246,
+      "grad_norm": 0.3794139623641968,
+      "learning_rate": 0.0002,
+      "loss": 1.6367,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6361598804632052,
+      "grad_norm": 0.359742134809494,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6436309301456855,
+      "grad_norm": 0.3770543932914734,
+      "learning_rate": 0.0002,
+      "loss": 1.7027,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6511019798281659,
+      "grad_norm": 0.3797036409378052,
+      "learning_rate": 0.0002,
+      "loss": 1.784,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6585730295106462,
+      "grad_norm": 0.35622093081474304,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6660440791931266,
+      "grad_norm": 0.34552520513534546,
+      "learning_rate": 0.0002,
+      "loss": 1.6615,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6735151288756072,
+      "grad_norm": 0.379926860332489,
+      "learning_rate": 0.0002,
+      "loss": 1.7522,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6809861785580873,
+      "grad_norm": 0.37083810567855835,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6884572282405679,
+      "grad_norm": 0.42746543884277344,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6959282779230482,
+      "grad_norm": 0.3372884690761566,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7033993276055286,
+      "grad_norm": 0.35220256447792053,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2280
+    },
+    {
+      "epoch": 1.710870377288009,
+      "grad_norm": 0.3659130930900574,
+      "learning_rate": 0.0002,
+      "loss": 1.7154,
+      "step": 2290
+    },
+    {
+      "epoch": 1.7183414269704893,
+      "grad_norm": 0.37629297375679016,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7258124766529699,
+      "grad_norm": 0.36312398314476013,
+      "learning_rate": 0.0002,
+      "loss": 1.7212,
+      "step": 2310
+    },
+    {
+      "epoch": 1.73328352633545,
+      "grad_norm": 0.467709481716156,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7407545760179306,
+      "grad_norm": 0.38685527443885803,
+      "learning_rate": 0.0002,
+      "loss": 1.696,
+      "step": 2330
+    },
+    {
+      "epoch": 1.748225625700411,
+      "grad_norm": 0.3578338325023651,
+      "learning_rate": 0.0002,
+      "loss": 1.7041,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7556966753828913,
+      "grad_norm": 0.36057502031326294,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7631677250653717,
+      "grad_norm": 0.3615196645259857,
+      "learning_rate": 0.0002,
+      "loss": 1.6853,
+      "step": 2360
+    },
+    {
+      "epoch": 1.770638774747852,
+      "grad_norm": 0.4118947684764862,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7781098244303326,
+      "grad_norm": 0.4067276120185852,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7855808741128127,
+      "grad_norm": 0.3979823887348175,
+      "learning_rate": 0.0002,
+      "loss": 1.712,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7930519237952933,
+      "grad_norm": 0.44045883417129517,
+      "learning_rate": 0.0002,
+      "loss": 1.7644,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8005229734777737,
+      "grad_norm": 0.3998069167137146,
+      "learning_rate": 0.0002,
+      "loss": 1.7251,
+      "step": 2410
+    },
+    {
+      "epoch": 1.807994023160254,
+      "grad_norm": 0.3450094759464264,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8154650728427344,
+      "grad_norm": 0.3759009838104248,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2430
+    },
+    {
+      "epoch": 1.8229361225252148,
+      "grad_norm": 0.34347015619277954,
+      "learning_rate": 0.0002,
+      "loss": 1.7706,
+      "step": 2440
+    },
+    {
+      "epoch": 1.8304071722076953,
+      "grad_norm": 0.3511228859424591,
+      "learning_rate": 0.0002,
+      "loss": 1.7345,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8378782218901755,
+      "grad_norm": 0.36853715777397156,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 2460
+    },
+    {
+      "epoch": 1.845349271572656,
+      "grad_norm": 0.40659376978874207,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8528203212551362,
+      "grad_norm": 0.39621320366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8602913709376168,
+      "grad_norm": 0.3753979504108429,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8677624206200971,
+      "grad_norm": 0.3811938464641571,
+      "learning_rate": 0.0002,
+      "loss": 1.6622,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8752334703025775,
+      "grad_norm": 0.3432596027851105,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 2510
+    },
+    {
+      "epoch": 1.882704519985058,
+      "grad_norm": 0.3670712113380432,
+      "learning_rate": 0.0002,
+      "loss": 1.7488,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8901755696675382,
+      "grad_norm": 0.40907177329063416,
+      "learning_rate": 0.0002,
+      "loss": 1.705,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8976466193500188,
+      "grad_norm": 0.3821999728679657,
+      "learning_rate": 0.0002,
+      "loss": 1.7148,
+      "step": 2540
+    },
+    {
+      "epoch": 1.905117669032499,
+      "grad_norm": 0.36173978447914124,
+      "learning_rate": 0.0002,
+      "loss": 1.7934,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9125887187149795,
+      "grad_norm": 0.38990336656570435,
+      "learning_rate": 0.0002,
+      "loss": 1.6939,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9200597683974598,
+      "grad_norm": 0.35242322087287903,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 2570
+    },
+    {
+      "epoch": 1.9275308180799402,
+      "grad_norm": 0.3506428003311157,
+      "learning_rate": 0.0002,
+      "loss": 1.7268,
+      "step": 2580
+    },
+    {
+      "epoch": 1.9350018677624208,
+      "grad_norm": 0.39540135860443115,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2590
+    },
+    {
+      "epoch": 1.942472917444901,
+      "grad_norm": 0.3444725573062897,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9499439671273815,
+      "grad_norm": 0.3963521718978882,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9574150168098616,
+      "grad_norm": 0.3689815402030945,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9648860664923422,
+      "grad_norm": 0.3482626676559448,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9723571161748226,
+      "grad_norm": 0.35832616686820984,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 2640
+    },
+    {
+      "epoch": 1.979828165857303,
+      "grad_norm": 0.4776208996772766,
+      "learning_rate": 0.0002,
+      "loss": 1.6681,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9872992155397835,
+      "grad_norm": 0.32570165395736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6696,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9947702652222636,
+      "grad_norm": 0.3380725085735321,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8046749830245972,
+      "eval_runtime": 38.5096,
+      "eval_samples_per_second": 13.373,
+      "eval_steps_per_second": 1.688,
+      "step": 2677
+    },
+    {
+      "epoch": 2.002241314904744,
+      "grad_norm": 0.36817631125450134,
+      "learning_rate": 0.0002,
+      "loss": 1.7265,
+      "step": 2680
+    },
+    {
+      "epoch": 2.0097123645872244,
+      "grad_norm": 0.4056456685066223,
+      "learning_rate": 0.0002,
+      "loss": 1.548,
+      "step": 2690
+    },
+    {
+      "epoch": 2.017183414269705,
+      "grad_norm": 0.37416863441467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5515,
+      "step": 2700
+    },
+    {
+      "epoch": 2.024654463952185,
+      "grad_norm": 0.4273638427257538,
+      "learning_rate": 0.0002,
+      "loss": 1.5895,
+      "step": 2710
+    },
+    {
+      "epoch": 2.0321255136346656,
+      "grad_norm": 0.36497923731803894,
+      "learning_rate": 0.0002,
+      "loss": 1.5884,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0395965633171462,
+      "grad_norm": 0.5021994113922119,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 2730
+    },
+    {
+      "epoch": 2.0470676129996264,
+      "grad_norm": 0.45896220207214355,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 2740
+    },
+    {
+      "epoch": 2.054538662682107,
+      "grad_norm": 0.3973815143108368,
+      "learning_rate": 0.0002,
+      "loss": 1.6305,
+      "step": 2750
+    },
+    {
+      "epoch": 2.062009712364587,
+      "grad_norm": 0.4521815776824951,
+      "learning_rate": 0.0002,
+      "loss": 1.6301,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0694807620470677,
+      "grad_norm": 0.42775002121925354,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2770
+    },
+    {
+      "epoch": 2.076951811729548,
+      "grad_norm": 0.48158586025238037,
+      "learning_rate": 0.0002,
+      "loss": 1.6491,
+      "step": 2780
+    },
+    {
+      "epoch": 2.0844228614120284,
+      "grad_norm": 0.4612371623516083,
+      "learning_rate": 0.0002,
+      "loss": 1.6301,
+      "step": 2790
+    },
+    {
+      "epoch": 2.091893911094509,
+      "grad_norm": 0.42536866664886475,
+      "learning_rate": 0.0002,
+      "loss": 1.6327,
+      "step": 2800
+    },
+    {
+      "epoch": 2.099364960776989,
+      "grad_norm": 0.48515772819519043,
+      "learning_rate": 0.0002,
+      "loss": 1.651,
+      "step": 2810
+    },
+    {
+      "epoch": 2.1068360104594697,
+      "grad_norm": 0.41418662667274475,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 2820
+    },
+    {
+      "epoch": 2.11430706014195,
+      "grad_norm": 0.4683697819709778,
+      "learning_rate": 0.0002,
+      "loss": 1.6266,
+      "step": 2830
+    },
+    {
+      "epoch": 2.1217781098244304,
+      "grad_norm": 0.4484657049179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6586,
+      "step": 2840
+    },
+    {
+      "epoch": 2.1292491595069105,
+      "grad_norm": 0.6621400713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.6483,
+      "step": 2850
+    },
+    {
+      "epoch": 2.136720209189391,
+      "grad_norm": 0.45074811577796936,
+      "learning_rate": 0.0002,
+      "loss": 1.5755,
+      "step": 2860
+    },
+    {
+      "epoch": 2.1441912588718717,
+      "grad_norm": 0.3513113558292389,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2870
+    },
+    {
+      "epoch": 2.151662308554352,
+      "grad_norm": 0.40411314368247986,
+      "learning_rate": 0.0002,
+      "loss": 1.6081,
+      "step": 2880
+    },
+    {
+      "epoch": 2.1591333582368324,
+      "grad_norm": 0.4121065139770508,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 2890
+    },
+    {
+      "epoch": 2.1666044079193125,
+      "grad_norm": 0.44723689556121826,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 2900
+    },
+    {
+      "epoch": 2.174075457601793,
+      "grad_norm": 0.4226122498512268,
+      "learning_rate": 0.0002,
+      "loss": 1.5699,
+      "step": 2910
+    },
+    {
+      "epoch": 2.1815465072842732,
+      "grad_norm": 0.46617650985717773,
+      "learning_rate": 0.0002,
+      "loss": 1.5652,
+      "step": 2920
+    },
+    {
+      "epoch": 2.189017556966754,
+      "grad_norm": 0.4506422281265259,
+      "learning_rate": 0.0002,
+      "loss": 1.6378,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1964886066492344,
+      "grad_norm": 0.4892672896385193,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 2940
+    },
+    {
+      "epoch": 2.2039596563317145,
+      "grad_norm": 0.44095516204833984,
+      "learning_rate": 0.0002,
+      "loss": 1.6176,
+      "step": 2950
+    },
+    {
+      "epoch": 2.211430706014195,
+      "grad_norm": 0.41522109508514404,
+      "learning_rate": 0.0002,
+      "loss": 1.6058,
+      "step": 2960
+    },
+    {
+      "epoch": 2.2189017556966752,
+      "grad_norm": 0.4860858917236328,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2970
+    },
+    {
+      "epoch": 2.226372805379156,
+      "grad_norm": 0.42662516236305237,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2980
+    },
+    {
+      "epoch": 2.233843855061636,
+      "grad_norm": 0.4390648305416107,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 2990
+    },
+    {
+      "epoch": 2.2413149047441165,
+      "grad_norm": 0.47515565156936646,
+      "learning_rate": 0.0002,
+      "loss": 1.5992,
+      "step": 3000
+    },
+    {
+      "epoch": 2.248785954426597,
+      "grad_norm": 0.4104543924331665,
+      "learning_rate": 0.0002,
+      "loss": 1.5563,
+      "step": 3010
+    },
+    {
+      "epoch": 2.2562570041090773,
+      "grad_norm": 0.4404028654098511,
+      "learning_rate": 0.0002,
+      "loss": 1.6895,
+      "step": 3020
+    },
+    {
+      "epoch": 2.263728053791558,
+      "grad_norm": 0.4717366695404053,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 3030
+    },
+    {
+      "epoch": 2.271199103474038,
+      "grad_norm": 0.48345857858657837,
+      "learning_rate": 0.0002,
+      "loss": 1.7287,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2786701531565186,
+      "grad_norm": 0.5312452912330627,
+      "learning_rate": 0.0002,
+      "loss": 1.681,
+      "step": 3050
+    },
+    {
+      "epoch": 2.2861412028389987,
+      "grad_norm": 0.5073099732398987,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2936122525214793,
+      "grad_norm": 0.5027463436126709,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 3070
+    },
+    {
+      "epoch": 2.30108330220396,
+      "grad_norm": 0.5436304807662964,
+      "learning_rate": 0.0002,
+      "loss": 1.5862,
+      "step": 3080
+    },
+    {
+      "epoch": 2.30855435188644,
+      "grad_norm": 0.4701065123081207,
+      "learning_rate": 0.0002,
+      "loss": 1.5763,
+      "step": 3090
+    },
+    {
+      "epoch": 2.3160254015689206,
+      "grad_norm": 0.46988746523857117,
+      "learning_rate": 0.0002,
+      "loss": 1.6177,
+      "step": 3100
+    },
+    {
+      "epoch": 2.3234964512514007,
+      "grad_norm": 0.45112869143486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6502,
+      "step": 3110
+    },
+    {
+      "epoch": 2.3309675009338813,
+      "grad_norm": 0.5173566937446594,
+      "learning_rate": 0.0002,
+      "loss": 1.6291,
+      "step": 3120
+    },
+    {
+      "epoch": 2.3384385506163614,
+      "grad_norm": 0.40345850586891174,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 3130
+    },
+    {
+      "epoch": 2.345909600298842,
+      "grad_norm": 0.4218924939632416,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 3140
+    },
+    {
+      "epoch": 2.3533806499813226,
+      "grad_norm": 0.41857317090034485,
+      "learning_rate": 0.0002,
+      "loss": 1.6341,
+      "step": 3150
+    },
+    {
+      "epoch": 2.3608516996638027,
+      "grad_norm": 0.4197218418121338,
+      "learning_rate": 0.0002,
+      "loss": 1.6087,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3683227493462833,
+      "grad_norm": 0.4260677397251129,
+      "learning_rate": 0.0002,
+      "loss": 1.6572,
+      "step": 3170
+    },
+    {
+      "epoch": 2.3757937990287634,
+      "grad_norm": 0.4209042191505432,
+      "learning_rate": 0.0002,
+      "loss": 1.6376,
+      "step": 3180
+    },
+    {
+      "epoch": 2.383264848711244,
+      "grad_norm": 0.4092234969139099,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 3190
+    },
+    {
+      "epoch": 2.390735898393724,
+      "grad_norm": 0.4928431510925293,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3982069480762047,
+      "grad_norm": 0.49252402782440186,
+      "learning_rate": 0.0002,
+      "loss": 1.6015,
+      "step": 3210
+    },
+    {
+      "epoch": 2.4056779977586853,
+      "grad_norm": 0.4368397295475006,
+      "learning_rate": 0.0002,
+      "loss": 1.5773,
+      "step": 3220
+    },
+    {
+      "epoch": 2.4131490474411654,
+      "grad_norm": 0.46122390031814575,
+      "learning_rate": 0.0002,
+      "loss": 1.6629,
+      "step": 3230
+    },
+    {
+      "epoch": 2.420620097123646,
+      "grad_norm": 0.4272301197052002,
+      "learning_rate": 0.0002,
+      "loss": 1.6224,
+      "step": 3240
+    },
+    {
+      "epoch": 2.428091146806126,
+      "grad_norm": 0.41480937600135803,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 3250
+    },
+    {
+      "epoch": 2.4355621964886067,
+      "grad_norm": 0.48911941051483154,
+      "learning_rate": 0.0002,
+      "loss": 1.6281,
+      "step": 3260
+    },
+    {
+      "epoch": 2.443033246171087,
+      "grad_norm": 0.4444098472595215,
+      "learning_rate": 0.0002,
+      "loss": 1.6846,
+      "step": 3270
+    },
+    {
+      "epoch": 2.4505042958535674,
+      "grad_norm": 0.5111684799194336,
+      "learning_rate": 0.0002,
+      "loss": 1.6961,
+      "step": 3280
+    },
+    {
+      "epoch": 2.457975345536048,
+      "grad_norm": 0.5058825016021729,
+      "learning_rate": 0.0002,
+      "loss": 1.6152,
+      "step": 3290
+    },
+    {
+      "epoch": 2.465446395218528,
+      "grad_norm": 0.44173210859298706,
+      "learning_rate": 0.0002,
+      "loss": 1.625,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4729174449010087,
+      "grad_norm": 0.4659745991230011,
+      "learning_rate": 0.0002,
+      "loss": 1.6491,
+      "step": 3310
+    },
+    {
+      "epoch": 2.480388494583489,
+      "grad_norm": 0.47237497568130493,
+      "learning_rate": 0.0002,
+      "loss": 1.6114,
+      "step": 3320
+    },
+    {
+      "epoch": 2.4878595442659694,
+      "grad_norm": 0.47303131222724915,
+      "learning_rate": 0.0002,
+      "loss": 1.6193,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4953305939484496,
+      "grad_norm": 0.4522389769554138,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 3340
+    },
+    {
+      "epoch": 2.50280164363093,
+      "grad_norm": 0.4467332363128662,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3350
+    },
+    {
+      "epoch": 2.5102726933134107,
+      "grad_norm": 0.4413762092590332,
+      "learning_rate": 0.0002,
+      "loss": 1.6108,
+      "step": 3360
+    },
+    {
+      "epoch": 2.517743742995891,
+      "grad_norm": 0.495514452457428,
+      "learning_rate": 0.0002,
+      "loss": 1.537,
+      "step": 3370
+    },
+    {
+      "epoch": 2.5252147926783715,
+      "grad_norm": 0.4429773986339569,
+      "learning_rate": 0.0002,
+      "loss": 1.5839,
+      "step": 3380
+    },
+    {
+      "epoch": 2.5326858423608516,
+      "grad_norm": 0.4589079022407532,
+      "learning_rate": 0.0002,
+      "loss": 1.6522,
+      "step": 3390
+    },
+    {
+      "epoch": 2.540156892043332,
+      "grad_norm": 0.4683997333049774,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 3400
+    },
+    {
+      "epoch": 2.5476279417258123,
+      "grad_norm": 0.4651731252670288,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 3410
+    },
+    {
+      "epoch": 2.555098991408293,
+      "grad_norm": 0.45818084478378296,
+      "learning_rate": 0.0002,
+      "loss": 1.5918,
+      "step": 3420
+    },
+    {
+      "epoch": 2.5625700410907735,
+      "grad_norm": 0.45209529995918274,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3430
+    },
+    {
+      "epoch": 2.5700410907732536,
+      "grad_norm": 0.4344733655452728,
+      "learning_rate": 0.0002,
+      "loss": 1.5606,
+      "step": 3440
+    },
+    {
+      "epoch": 2.577512140455734,
+      "grad_norm": 0.47435566782951355,
+      "learning_rate": 0.0002,
+      "loss": 1.6748,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5849831901382143,
+      "grad_norm": 0.43841999769210815,
+      "learning_rate": 0.0002,
+      "loss": 1.6237,
+      "step": 3460
+    },
+    {
+      "epoch": 2.592454239820695,
+      "grad_norm": 0.4323869049549103,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 3470
+    },
+    {
+      "epoch": 2.599925289503175,
+      "grad_norm": 0.44355881214141846,
+      "learning_rate": 0.0002,
+      "loss": 1.5494,
+      "step": 3480
+    },
+    {
+      "epoch": 2.6073963391856556,
+      "grad_norm": 0.45847779512405396,
+      "learning_rate": 0.0002,
+      "loss": 1.665,
+      "step": 3490
+    },
+    {
+      "epoch": 2.614867388868136,
+      "grad_norm": 0.4411061704158783,
+      "learning_rate": 0.0002,
+      "loss": 1.6006,
+      "step": 3500
+    },
+    {
+      "epoch": 2.6223384385506163,
+      "grad_norm": 0.4446796178817749,
+      "learning_rate": 0.0002,
+      "loss": 1.5868,
+      "step": 3510
+    },
+    {
+      "epoch": 2.629809488233097,
+      "grad_norm": 0.41969653964042664,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 3520
+    },
+    {
+      "epoch": 2.637280537915577,
+      "grad_norm": 0.5263747572898865,
+      "learning_rate": 0.0002,
+      "loss": 1.6798,
+      "step": 3530
+    },
+    {
+      "epoch": 2.6447515875980576,
+      "grad_norm": 0.47719451785087585,
+      "learning_rate": 0.0002,
+      "loss": 1.6309,
+      "step": 3540
+    },
+    {
+      "epoch": 2.6522226372805378,
+      "grad_norm": 0.46574118733406067,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 3550
+    },
+    {
+      "epoch": 2.6596936869630183,
+      "grad_norm": 0.46867135167121887,
+      "learning_rate": 0.0002,
+      "loss": 1.618,
+      "step": 3560
+    },
+    {
+      "epoch": 2.667164736645499,
+      "grad_norm": 0.4441198706626892,
+      "learning_rate": 0.0002,
+      "loss": 1.5885,
+      "step": 3570
+    },
+    {
+      "epoch": 2.674635786327979,
+      "grad_norm": 0.4871319830417633,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6821068360104596,
+      "grad_norm": 0.43900373578071594,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6895778856929398,
+      "grad_norm": 0.42509549856185913,
+      "learning_rate": 0.0002,
+      "loss": 1.6071,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6970489353754203,
+      "grad_norm": 0.4691086709499359,
+      "learning_rate": 0.0002,
+      "loss": 1.5651,
+      "step": 3610
+    },
+    {
+      "epoch": 2.7045199850579005,
+      "grad_norm": 0.46318942308425903,
+      "learning_rate": 0.0002,
+      "loss": 1.5491,
+      "step": 3620
+    },
+    {
+      "epoch": 2.711991034740381,
+      "grad_norm": 0.44631096720695496,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 3630
+    },
+    {
+      "epoch": 2.7194620844228616,
+      "grad_norm": 0.42315489053726196,
+      "learning_rate": 0.0002,
+      "loss": 1.6831,
+      "step": 3640
+    },
+    {
+      "epoch": 2.7269331341053418,
+      "grad_norm": 0.4971241056919098,
+      "learning_rate": 0.0002,
+      "loss": 1.6008,
+      "step": 3650
+    },
+    {
+      "epoch": 2.7344041837878224,
+      "grad_norm": 0.4578486382961273,
+      "learning_rate": 0.0002,
+      "loss": 1.6042,
+      "step": 3660
+    },
+    {
+      "epoch": 2.7418752334703025,
+      "grad_norm": 0.46584776043891907,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 3670
+    },
+    {
+      "epoch": 2.749346283152783,
+      "grad_norm": 0.4951731264591217,
+      "learning_rate": 0.0002,
+      "loss": 1.6809,
+      "step": 3680
+    },
+    {
+      "epoch": 2.756817332835263,
+      "grad_norm": 0.4935225546360016,
+      "learning_rate": 0.0002,
+      "loss": 1.6226,
+      "step": 3690
+    },
+    {
+      "epoch": 2.764288382517744,
+      "grad_norm": 0.41805586218833923,
+      "learning_rate": 0.0002,
+      "loss": 1.5878,
+      "step": 3700
+    },
+    {
+      "epoch": 2.7717594322002244,
+      "grad_norm": 0.4417555630207062,
+      "learning_rate": 0.0002,
+      "loss": 1.7173,
+      "step": 3710
+    },
+    {
+      "epoch": 2.7792304818827045,
+      "grad_norm": 0.48229655623435974,
+      "learning_rate": 0.0002,
+      "loss": 1.6398,
+      "step": 3720
+    },
+    {
+      "epoch": 2.786701531565185,
+      "grad_norm": 0.48562315106391907,
+      "learning_rate": 0.0002,
+      "loss": 1.6074,
+      "step": 3730
+    },
+    {
+      "epoch": 2.794172581247665,
+      "grad_norm": 0.4473940432071686,
+      "learning_rate": 0.0002,
+      "loss": 1.607,
+      "step": 3740
+    },
+    {
+      "epoch": 2.801643630930146,
+      "grad_norm": 0.4626813232898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6065,
+      "step": 3750
+    },
+    {
+      "epoch": 2.809114680612626,
+      "grad_norm": 0.4339792728424072,
+      "learning_rate": 0.0002,
+      "loss": 1.6296,
+      "step": 3760
+    },
+    {
+      "epoch": 2.8165857302951065,
+      "grad_norm": 0.5250858068466187,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 3770
+    },
+    {
+      "epoch": 2.824056779977587,
+      "grad_norm": 0.4537523090839386,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 3780
+    },
+    {
+      "epoch": 2.831527829660067,
+      "grad_norm": 0.5646113157272339,
+      "learning_rate": 0.0002,
+      "loss": 1.6535,
+      "step": 3790
+    },
+    {
+      "epoch": 2.8389988793425474,
+      "grad_norm": 0.44243332743644714,
+      "learning_rate": 0.0002,
+      "loss": 1.5712,
+      "step": 3800
+    },
+    {
+      "epoch": 2.846469929025028,
+      "grad_norm": 0.4585791826248169,
+      "learning_rate": 0.0002,
+      "loss": 1.6478,
+      "step": 3810
+    },
+    {
+      "epoch": 2.8539409787075085,
+      "grad_norm": 0.489702045917511,
+      "learning_rate": 0.0002,
+      "loss": 1.6854,
+      "step": 3820
+    },
+    {
+      "epoch": 2.8614120283899886,
+      "grad_norm": 0.502470850944519,
+      "learning_rate": 0.0002,
+      "loss": 1.7066,
+      "step": 3830
+    },
+    {
+      "epoch": 2.8688830780724692,
+      "grad_norm": 0.4395960867404938,
+      "learning_rate": 0.0002,
+      "loss": 1.5785,
+      "step": 3840
+    },
+    {
+      "epoch": 2.87635412775495,
+      "grad_norm": 0.4348670244216919,
+      "learning_rate": 0.0002,
+      "loss": 1.6434,
+      "step": 3850
+    },
+    {
+      "epoch": 2.88382517743743,
+      "grad_norm": 0.48852720856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 3860
+    },
+    {
+      "epoch": 2.89129622711991,
+      "grad_norm": 0.45317450165748596,
+      "learning_rate": 0.0002,
+      "loss": 1.5916,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8987672768023907,
+      "grad_norm": 0.4732758700847626,
+      "learning_rate": 0.0002,
+      "loss": 1.6486,
+      "step": 3880
+    },
+    {
+      "epoch": 2.9062383264848712,
+      "grad_norm": 0.45238012075424194,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 3890
+    },
+    {
+      "epoch": 2.9137093761673514,
+      "grad_norm": 0.48838064074516296,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 3900
+    },
+    {
+      "epoch": 2.921180425849832,
+      "grad_norm": 0.43496349453926086,
+      "learning_rate": 0.0002,
+      "loss": 1.658,
+      "step": 3910
+    },
+    {
+      "epoch": 2.9286514755323125,
+      "grad_norm": 0.47963935136795044,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 3920
+    },
+    {
+      "epoch": 2.9361225252147927,
+      "grad_norm": 0.4544987976551056,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 3930
+    },
+    {
+      "epoch": 2.943593574897273,
+      "grad_norm": 0.4622892141342163,
+      "learning_rate": 0.0002,
+      "loss": 1.6192,
+      "step": 3940
+    },
+    {
+      "epoch": 2.9510646245797534,
+      "grad_norm": 0.47026222944259644,
+      "learning_rate": 0.0002,
+      "loss": 1.6178,
+      "step": 3950
+    },
+    {
+      "epoch": 2.958535674262234,
+      "grad_norm": 0.4549552798271179,
+      "learning_rate": 0.0002,
+      "loss": 1.6612,
+      "step": 3960
+    },
+    {
+      "epoch": 2.966006723944714,
+      "grad_norm": 0.46647515892982483,
+      "learning_rate": 0.0002,
+      "loss": 1.6458,
+      "step": 3970
+    },
+    {
+      "epoch": 2.9734777736271947,
+      "grad_norm": 0.45095112919807434,
+      "learning_rate": 0.0002,
+      "loss": 1.6051,
+      "step": 3980
+    },
+    {
+      "epoch": 2.9809488233096753,
+      "grad_norm": 0.4690017104148865,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 3990
+    },
+    {
+      "epoch": 2.9884198729921554,
+      "grad_norm": 0.4603444039821625,
+      "learning_rate": 0.0002,
+      "loss": 1.6061,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9958909226746355,
+      "grad_norm": 0.4743294417858124,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 4010
+    },
+    {
+      "epoch": 2.999626447515876,
+      "eval_loss": 1.8252571821212769,
+      "eval_runtime": 38.7853,
+      "eval_samples_per_second": 13.278,
+      "eval_steps_per_second": 1.676,
+      "step": 4015
+    },
+    {
+      "epoch": 3.003361972357116,
+      "grad_norm": 0.4919724464416504,
+      "learning_rate": 0.0002,
+      "loss": 1.6512,
+      "step": 4020
+    },
+    {
+      "epoch": 3.0108330220395967,
+      "grad_norm": 0.4747185707092285,
+      "learning_rate": 0.0002,
+      "loss": 1.5354,
+      "step": 4030
+    },
+    {
+      "epoch": 3.018304071722077,
+      "grad_norm": 0.4797595143318176,
+      "learning_rate": 0.0002,
+      "loss": 1.568,
+      "step": 4040
+    },
+    {
+      "epoch": 3.0257751214045574,
+      "grad_norm": 0.5450999140739441,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 4050
+    },
+    {
+      "epoch": 3.0332461710870375,
+      "grad_norm": 0.49058812856674194,
+      "learning_rate": 0.0002,
+      "loss": 1.5065,
+      "step": 4060
+    },
+    {
+      "epoch": 3.040717220769518,
+      "grad_norm": 0.5219563841819763,
+      "learning_rate": 0.0002,
+      "loss": 1.4884,
+      "step": 4070
+    },
+    {
+      "epoch": 3.0481882704519987,
+      "grad_norm": 0.515628457069397,
+      "learning_rate": 0.0002,
+      "loss": 1.4742,
+      "step": 4080
+    },
+    {
+      "epoch": 3.055659320134479,
+      "grad_norm": 0.6145984530448914,
+      "learning_rate": 0.0002,
+      "loss": 1.5313,
+      "step": 4090
+    },
+    {
+      "epoch": 3.0631303698169594,
+      "grad_norm": 0.6067144274711609,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 4100
+    },
+    {
+      "epoch": 3.0706014194994395,
+      "grad_norm": 0.5773133039474487,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 4110
+    },
+    {
+      "epoch": 3.07807246918192,
+      "grad_norm": 0.6894241571426392,
+      "learning_rate": 0.0002,
+      "loss": 1.5374,
+      "step": 4120
+    },
+    {
+      "epoch": 3.0855435188644003,
+      "grad_norm": 0.6422514915466309,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 4130
+    },
+    {
+      "epoch": 3.093014568546881,
+      "grad_norm": 0.6119855046272278,
+      "learning_rate": 0.0002,
+      "loss": 1.4724,
+      "step": 4140
+    },
+    {
+      "epoch": 3.1004856182293614,
+      "grad_norm": 0.5847280025482178,
+      "learning_rate": 0.0002,
+      "loss": 1.5361,
+      "step": 4150
+    },
+    {
+      "epoch": 3.1079566679118416,
+      "grad_norm": 0.5401515960693359,
+      "learning_rate": 0.0002,
+      "loss": 1.5151,
+      "step": 4160
+    },
+    {
+      "epoch": 3.115427717594322,
+      "grad_norm": 0.6501587629318237,
+      "learning_rate": 0.0002,
+      "loss": 1.502,
+      "step": 4170
+    },
+    {
+      "epoch": 3.1228987672768023,
+      "grad_norm": 0.5988039374351501,
+      "learning_rate": 0.0002,
+      "loss": 1.4952,
+      "step": 4180
+    },
+    {
+      "epoch": 3.130369816959283,
+      "grad_norm": 0.4982665181159973,
+      "learning_rate": 0.0002,
+      "loss": 1.5287,
+      "step": 4190
+    },
+    {
+      "epoch": 3.137840866641763,
+      "grad_norm": 0.5548039078712463,
+      "learning_rate": 0.0002,
+      "loss": 1.5078,
+      "step": 4200
+    },
+    {
+      "epoch": 3.1453119163242436,
+      "grad_norm": 0.5920777320861816,
+      "learning_rate": 0.0002,
+      "loss": 1.4904,
+      "step": 4210
+    },
+    {
+      "epoch": 3.152782966006724,
+      "grad_norm": 0.6965190172195435,
+      "learning_rate": 0.0002,
+      "loss": 1.442,
+      "step": 4220
+    },
+    {
+      "epoch": 3.1602540156892043,
+      "grad_norm": 0.5196244716644287,
+      "learning_rate": 0.0002,
+      "loss": 1.557,
+      "step": 4230
+    },
+    {
+      "epoch": 3.167725065371685,
+      "grad_norm": 0.6942682266235352,
+      "learning_rate": 0.0002,
+      "loss": 1.5706,
+      "step": 4240
+    },
+    {
+      "epoch": 3.175196115054165,
+      "grad_norm": 0.5765156149864197,
+      "learning_rate": 0.0002,
+      "loss": 1.5407,
+      "step": 4250
+    },
+    {
+      "epoch": 3.1826671647366456,
+      "grad_norm": 0.5801976919174194,
+      "learning_rate": 0.0002,
+      "loss": 1.4963,
+      "step": 4260
+    },
+    {
+      "epoch": 3.1901382144191257,
+      "grad_norm": 0.6260752081871033,
+      "learning_rate": 0.0002,
+      "loss": 1.4988,
+      "step": 4270
+    },
+    {
+      "epoch": 3.1976092641016063,
+      "grad_norm": 0.6610770225524902,
+      "learning_rate": 0.0002,
+      "loss": 1.5074,
+      "step": 4280
+    },
+    {
+      "epoch": 3.205080313784087,
+      "grad_norm": 0.5762143135070801,
+      "learning_rate": 0.0002,
+      "loss": 1.4657,
+      "step": 4290
+    },
+    {
+      "epoch": 3.212551363466567,
+      "grad_norm": 0.5926990509033203,
+      "learning_rate": 0.0002,
+      "loss": 1.5181,
+      "step": 4300
+    },
+    {
+      "epoch": 3.2200224131490476,
+      "grad_norm": 0.7373854517936707,
+      "learning_rate": 0.0002,
+      "loss": 1.5492,
+      "step": 4310
+    },
+    {
+      "epoch": 3.2274934628315277,
+      "grad_norm": 0.5963311195373535,
+      "learning_rate": 0.0002,
+      "loss": 1.4648,
+      "step": 4320
+    },
+    {
+      "epoch": 3.2349645125140083,
+      "grad_norm": 0.5754616856575012,
+      "learning_rate": 0.0002,
+      "loss": 1.5262,
+      "step": 4330
+    },
+    {
+      "epoch": 3.2424355621964884,
+      "grad_norm": 0.6116095781326294,
+      "learning_rate": 0.0002,
+      "loss": 1.4767,
+      "step": 4340
+    },
+    {
+      "epoch": 3.249906611878969,
+      "grad_norm": 0.6001536846160889,
+      "learning_rate": 0.0002,
+      "loss": 1.5008,
+      "step": 4350
+    },
+    {
+      "epoch": 3.257377661561449,
+      "grad_norm": 0.5270227789878845,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 4360
+    },
+    {
+      "epoch": 3.2648487112439297,
+      "grad_norm": 0.6666602492332458,
+      "learning_rate": 0.0002,
+      "loss": 1.5235,
+      "step": 4370
+    },
+    {
+      "epoch": 3.2723197609264103,
+      "grad_norm": 0.520310640335083,
+      "learning_rate": 0.0002,
+      "loss": 1.5665,
+      "step": 4380
+    },
+    {
+      "epoch": 3.2797908106088904,
+      "grad_norm": 0.5165975093841553,
+      "learning_rate": 0.0002,
+      "loss": 1.542,
+      "step": 4390
+    },
+    {
+      "epoch": 3.287261860291371,
+      "grad_norm": 0.6080228686332703,
+      "learning_rate": 0.0002,
+      "loss": 1.4746,
+      "step": 4400
+    },
+    {
+      "epoch": 3.294732909973851,
+      "grad_norm": 0.670122504234314,
+      "learning_rate": 0.0002,
+      "loss": 1.4901,
+      "step": 4410
+    },
+    {
+      "epoch": 3.3022039596563317,
+      "grad_norm": 0.6019457578659058,
+      "learning_rate": 0.0002,
+      "loss": 1.4677,
+      "step": 4420
+    },
+    {
+      "epoch": 3.309675009338812,
+      "grad_norm": 0.5519300103187561,
+      "learning_rate": 0.0002,
+      "loss": 1.4249,
+      "step": 4430
+    },
+    {
+      "epoch": 3.3171460590212924,
+      "grad_norm": 0.5958521962165833,
+      "learning_rate": 0.0002,
+      "loss": 1.555,
+      "step": 4440
+    },
+    {
+      "epoch": 3.324617108703773,
+      "grad_norm": 0.5552705526351929,
+      "learning_rate": 0.0002,
+      "loss": 1.5067,
+      "step": 4450
+    },
+    {
+      "epoch": 3.332088158386253,
+      "grad_norm": 0.6583784818649292,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 4460
+    },
+    {
+      "epoch": 3.3395592080687337,
+      "grad_norm": 0.5815939903259277,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 4470
+    },
+    {
+      "epoch": 3.347030257751214,
+      "grad_norm": 1.3342205286026,
+      "learning_rate": 0.0002,
+      "loss": 1.5942,
+      "step": 4480
+    },
+    {
+      "epoch": 3.3545013074336945,
+      "grad_norm": 0.6341500878334045,
+      "learning_rate": 0.0002,
+      "loss": 1.484,
+      "step": 4490
+    },
+    {
+      "epoch": 3.3619723571161746,
+      "grad_norm": 0.6384079456329346,
+      "learning_rate": 0.0002,
+      "loss": 1.5219,
+      "step": 4500
+    },
+    {
+      "epoch": 3.369443406798655,
+      "grad_norm": 0.6098346710205078,
+      "learning_rate": 0.0002,
+      "loss": 1.5222,
+      "step": 4510
+    },
+    {
+      "epoch": 3.3769144564811358,
+      "grad_norm": 0.5958296656608582,
+      "learning_rate": 0.0002,
+      "loss": 1.5475,
+      "step": 4520
+    },
+    {
+      "epoch": 3.384385506163616,
+      "grad_norm": 0.6157881617546082,
+      "learning_rate": 0.0002,
+      "loss": 1.5171,
+      "step": 4530
+    },
+    {
+      "epoch": 3.3918565558460965,
+      "grad_norm": 0.5671007037162781,
+      "learning_rate": 0.0002,
+      "loss": 1.569,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3993276055285766,
+      "grad_norm": 0.6203294992446899,
+      "learning_rate": 0.0002,
+      "loss": 1.604,
+      "step": 4550
+    },
+    {
+      "epoch": 3.406798655211057,
+      "grad_norm": 0.6743317246437073,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 4560
+    },
+    {
+      "epoch": 3.4142697048935373,
+      "grad_norm": 0.731765627861023,
+      "learning_rate": 0.0002,
+      "loss": 1.5034,
+      "step": 4570
+    },
+    {
+      "epoch": 3.421740754576018,
+      "grad_norm": 0.6285187602043152,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 4580
+    },
+    {
+      "epoch": 3.4292118042584985,
+      "grad_norm": 0.612680196762085,
+      "learning_rate": 0.0002,
+      "loss": 1.5296,
+      "step": 4590
+    },
+    {
+      "epoch": 3.4366828539409786,
+      "grad_norm": 0.6413681507110596,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 4600
+    },
+    {
+      "epoch": 3.444153903623459,
+      "grad_norm": 0.6240990161895752,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4610
+    },
+    {
+      "epoch": 3.4516249533059393,
+      "grad_norm": 0.5095735192298889,
+      "learning_rate": 0.0002,
+      "loss": 1.5887,
+      "step": 4620
+    },
+    {
+      "epoch": 3.45909600298842,
+      "grad_norm": 0.5699611902236938,
+      "learning_rate": 0.0002,
+      "loss": 1.4906,
+      "step": 4630
+    },
+    {
+      "epoch": 3.4665670526709,
+      "grad_norm": 0.7289775609970093,
+      "learning_rate": 0.0002,
+      "loss": 1.5176,
+      "step": 4640
+    },
+    {
+      "epoch": 3.4740381023533806,
+      "grad_norm": 0.6211609840393066,
+      "learning_rate": 0.0002,
+      "loss": 1.5467,
+      "step": 4650
+    },
+    {
+      "epoch": 3.481509152035861,
+      "grad_norm": 0.5714802145957947,
+      "learning_rate": 0.0002,
+      "loss": 1.533,
+      "step": 4660
+    },
+    {
+      "epoch": 3.4889802017183413,
+      "grad_norm": 0.6287049651145935,
+      "learning_rate": 0.0002,
+      "loss": 1.5096,
+      "step": 4670
+    },
+    {
+      "epoch": 3.496451251400822,
+      "grad_norm": 0.5480595827102661,
+      "learning_rate": 0.0002,
+      "loss": 1.4212,
+      "step": 4680
+    },
+    {
+      "epoch": 3.503922301083302,
+      "grad_norm": 0.5683253407478333,
+      "learning_rate": 0.0002,
+      "loss": 1.4746,
+      "step": 4690
+    },
+    {
+      "epoch": 3.5113933507657826,
+      "grad_norm": 0.601140558719635,
+      "learning_rate": 0.0002,
+      "loss": 1.5012,
+      "step": 4700
+    },
+    {
+      "epoch": 3.5188644004482628,
+      "grad_norm": 0.5344498157501221,
+      "learning_rate": 0.0002,
+      "loss": 1.5383,
+      "step": 4710
+    },
+    {
+      "epoch": 3.5263354501307433,
+      "grad_norm": 0.5739690661430359,
+      "learning_rate": 0.0002,
+      "loss": 1.5428,
+      "step": 4720
+    },
+    {
+      "epoch": 3.533806499813224,
+      "grad_norm": 0.5640085935592651,
+      "learning_rate": 0.0002,
+      "loss": 1.5589,
+      "step": 4730
+    },
+    {
+      "epoch": 3.541277549495704,
+      "grad_norm": 0.5967805981636047,
+      "learning_rate": 0.0002,
+      "loss": 1.487,
+      "step": 4740
+    },
+    {
+      "epoch": 3.5487485991781846,
+      "grad_norm": 0.6138835549354553,
+      "learning_rate": 0.0002,
+      "loss": 1.5461,
+      "step": 4750
+    },
+    {
+      "epoch": 3.5562196488606648,
+      "grad_norm": 0.6779900193214417,
+      "learning_rate": 0.0002,
+      "loss": 1.5502,
+      "step": 4760
+    },
+    {
+      "epoch": 3.5636906985431454,
+      "grad_norm": 0.6122010350227356,
+      "learning_rate": 0.0002,
+      "loss": 1.4917,
+      "step": 4770
+    },
+    {
+      "epoch": 3.5711617482256255,
+      "grad_norm": 0.5685241222381592,
+      "learning_rate": 0.0002,
+      "loss": 1.5405,
+      "step": 4780
+    },
+    {
+      "epoch": 3.578632797908106,
+      "grad_norm": 0.604583203792572,
+      "learning_rate": 0.0002,
+      "loss": 1.5427,
+      "step": 4790
+    },
+    {
+      "epoch": 3.5861038475905866,
+      "grad_norm": 0.651165246963501,
+      "learning_rate": 0.0002,
+      "loss": 1.4514,
+      "step": 4800
+    },
+    {
+      "epoch": 3.593574897273067,
+      "grad_norm": 0.6398511528968811,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 4810
+    },
+    {
+      "epoch": 3.6010459469555474,
+      "grad_norm": 0.6444641351699829,
+      "learning_rate": 0.0002,
+      "loss": 1.4261,
+      "step": 4820
+    },
+    {
+      "epoch": 3.6085169966380275,
+      "grad_norm": 0.6018481850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.5274,
+      "step": 4830
+    },
+    {
+      "epoch": 3.615988046320508,
+      "grad_norm": 0.6025291085243225,
+      "learning_rate": 0.0002,
+      "loss": 1.4647,
+      "step": 4840
+    },
+    {
+      "epoch": 3.623459096002988,
+      "grad_norm": 0.6810156106948853,
+      "learning_rate": 0.0002,
+      "loss": 1.5609,
+      "step": 4850
+    },
+    {
+      "epoch": 3.630930145685469,
+      "grad_norm": 0.6408044695854187,
+      "learning_rate": 0.0002,
+      "loss": 1.5299,
+      "step": 4860
+    },
+    {
+      "epoch": 3.6384011953679494,
+      "grad_norm": 0.5608272552490234,
+      "learning_rate": 0.0002,
+      "loss": 1.5366,
+      "step": 4870
+    },
+    {
+      "epoch": 3.6458722450504295,
+      "grad_norm": 0.6136814951896667,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 4880
+    },
+    {
+      "epoch": 3.65334329473291,
+      "grad_norm": 0.5927900075912476,
+      "learning_rate": 0.0002,
+      "loss": 1.5021,
+      "step": 4890
+    },
+    {
+      "epoch": 3.66081434441539,
+      "grad_norm": 0.5336901545524597,
+      "learning_rate": 0.0002,
+      "loss": 1.6084,
+      "step": 4900
+    },
+    {
+      "epoch": 3.668285394097871,
+      "grad_norm": 0.7823320627212524,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 4910
+    },
+    {
+      "epoch": 3.675756443780351,
+      "grad_norm": 0.6703504323959351,
+      "learning_rate": 0.0002,
+      "loss": 1.4881,
+      "step": 4920
+    },
+    {
+      "epoch": 3.6832274934628315,
+      "grad_norm": 0.6061160564422607,
+      "learning_rate": 0.0002,
+      "loss": 1.5332,
+      "step": 4930
+    },
+    {
+      "epoch": 3.690698543145312,
+      "grad_norm": 0.6237227916717529,
+      "learning_rate": 0.0002,
+      "loss": 1.5405,
+      "step": 4940
+    },
+    {
+      "epoch": 3.6981695928277922,
+      "grad_norm": 0.5985278487205505,
+      "learning_rate": 0.0002,
+      "loss": 1.497,
+      "step": 4950
+    },
+    {
+      "epoch": 3.705640642510273,
+      "grad_norm": 0.6483839750289917,
+      "learning_rate": 0.0002,
+      "loss": 1.5132,
+      "step": 4960
+    },
+    {
+      "epoch": 3.713111692192753,
+      "grad_norm": 0.5788805484771729,
+      "learning_rate": 0.0002,
+      "loss": 1.5338,
+      "step": 4970
+    },
+    {
+      "epoch": 3.7205827418752335,
+      "grad_norm": 0.5609974265098572,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 4980
+    },
+    {
+      "epoch": 3.7280537915577137,
+      "grad_norm": 0.5681300759315491,
+      "learning_rate": 0.0002,
+      "loss": 1.4759,
+      "step": 4990
+    },
+    {
+      "epoch": 3.7355248412401942,
+      "grad_norm": 0.5860186219215393,
+      "learning_rate": 0.0002,
+      "loss": 1.6018,
+      "step": 5000
+    },
+    {
+      "epoch": 3.742995890922675,
+      "grad_norm": 0.5718157291412354,
+      "learning_rate": 0.0002,
+      "loss": 1.58,
+      "step": 5010
+    },
+    {
+      "epoch": 3.750466940605155,
+      "grad_norm": 0.6173721551895142,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 5020
+    },
+    {
+      "epoch": 3.7579379902876355,
+      "grad_norm": 0.629152238368988,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 5030
+    },
+    {
+      "epoch": 3.7654090399701157,
+      "grad_norm": 0.5666284561157227,
+      "learning_rate": 0.0002,
+      "loss": 1.519,
+      "step": 5040
+    },
+    {
+      "epoch": 3.7728800896525962,
+      "grad_norm": 0.6053005456924438,
+      "learning_rate": 0.0002,
+      "loss": 1.5329,
+      "step": 5050
+    },
+    {
+      "epoch": 3.7803511393350764,
+      "grad_norm": 0.5870583057403564,
+      "learning_rate": 0.0002,
+      "loss": 1.5404,
+      "step": 5060
+    },
+    {
+      "epoch": 3.787822189017557,
+      "grad_norm": 0.5422009229660034,
+      "learning_rate": 0.0002,
+      "loss": 1.4444,
+      "step": 5070
+    },
+    {
+      "epoch": 3.7952932387000375,
+      "grad_norm": 0.5396918058395386,
+      "learning_rate": 0.0002,
+      "loss": 1.5308,
+      "step": 5080
+    },
+    {
+      "epoch": 3.8027642883825177,
+      "grad_norm": 0.5544713139533997,
+      "learning_rate": 0.0002,
+      "loss": 1.464,
+      "step": 5090
+    },
+    {
+      "epoch": 3.8102353380649983,
+      "grad_norm": 0.5983749628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5100
+    },
+    {
+      "epoch": 3.8177063877474784,
+      "grad_norm": 0.5702024102210999,
+      "learning_rate": 0.0002,
+      "loss": 1.4972,
+      "step": 5110
+    },
+    {
+      "epoch": 3.825177437429959,
+      "grad_norm": 0.5436882376670837,
+      "learning_rate": 0.0002,
+      "loss": 1.5471,
+      "step": 5120
+    },
+    {
+      "epoch": 3.832648487112439,
+      "grad_norm": 0.5453617572784424,
+      "learning_rate": 0.0002,
+      "loss": 1.5118,
+      "step": 5130
+    },
+    {
+      "epoch": 3.8401195367949197,
+      "grad_norm": 0.6269069314002991,
+      "learning_rate": 0.0002,
+      "loss": 1.5732,
+      "step": 5140
+    },
+    {
+      "epoch": 3.8475905864774003,
+      "grad_norm": 0.6189185380935669,
+      "learning_rate": 0.0002,
+      "loss": 1.4959,
+      "step": 5150
+    },
+    {
+      "epoch": 3.8550616361598804,
+      "grad_norm": 0.6653388142585754,
+      "learning_rate": 0.0002,
+      "loss": 1.4999,
+      "step": 5160
+    },
+    {
+      "epoch": 3.862532685842361,
+      "grad_norm": 0.5771768689155579,
+      "learning_rate": 0.0002,
+      "loss": 1.5075,
+      "step": 5170
+    },
+    {
+      "epoch": 3.870003735524841,
+      "grad_norm": 0.6052790880203247,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5180
+    },
+    {
+      "epoch": 3.8774747852073217,
+      "grad_norm": 0.6572316884994507,
+      "learning_rate": 0.0002,
+      "loss": 1.4987,
+      "step": 5190
+    },
+    {
+      "epoch": 3.884945834889802,
+      "grad_norm": 0.670576810836792,
+      "learning_rate": 0.0002,
+      "loss": 1.5241,
+      "step": 5200
+    },
+    {
+      "epoch": 3.8924168845722824,
+      "grad_norm": 0.5728798508644104,
+      "learning_rate": 0.0002,
+      "loss": 1.4777,
+      "step": 5210
+    },
+    {
+      "epoch": 3.899887934254763,
+      "grad_norm": 0.6340774297714233,
+      "learning_rate": 0.0002,
+      "loss": 1.5351,
+      "step": 5220
+    },
+    {
+      "epoch": 3.907358983937243,
+      "grad_norm": 0.5981315970420837,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 5230
+    },
+    {
+      "epoch": 3.9148300336197237,
+      "grad_norm": 0.6212025880813599,
+      "learning_rate": 0.0002,
+      "loss": 1.4875,
+      "step": 5240
+    },
+    {
+      "epoch": 3.922301083302204,
+      "grad_norm": 0.6202296018600464,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5250
+    },
+    {
+      "epoch": 3.9297721329846844,
+      "grad_norm": 0.6159142255783081,
+      "learning_rate": 0.0002,
+      "loss": 1.5765,
+      "step": 5260
+    },
+    {
+      "epoch": 3.9372431826671646,
+      "grad_norm": 0.6519438624382019,
+      "learning_rate": 0.0002,
+      "loss": 1.4938,
+      "step": 5270
+    },
+    {
+      "epoch": 3.944714232349645,
+      "grad_norm": 0.539813756942749,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 5280
+    },
+    {
+      "epoch": 3.9521852820321257,
+      "grad_norm": 0.6443665027618408,
+      "learning_rate": 0.0002,
+      "loss": 1.5921,
+      "step": 5290
+    },
+    {
+      "epoch": 3.959656331714606,
+      "grad_norm": 0.6635757684707642,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 5300
+    },
+    {
+      "epoch": 3.9671273813970864,
+      "grad_norm": 0.589363157749176,
+      "learning_rate": 0.0002,
+      "loss": 1.5485,
+      "step": 5310
+    },
+    {
+      "epoch": 3.9745984310795666,
+      "grad_norm": 0.5788735747337341,
+      "learning_rate": 0.0002,
+      "loss": 1.5498,
+      "step": 5320
+    },
+    {
+      "epoch": 3.982069480762047,
+      "grad_norm": 0.5976864695549011,
+      "learning_rate": 0.0002,
+      "loss": 1.5607,
+      "step": 5330
+    },
+    {
+      "epoch": 3.9895405304445273,
+      "grad_norm": 0.6624067425727844,
+      "learning_rate": 0.0002,
+      "loss": 1.5302,
+      "step": 5340
+    },
+    {
+      "epoch": 3.997011580127008,
+      "grad_norm": 0.6738956570625305,
+      "learning_rate": 0.0002,
+      "loss": 1.5904,
+      "step": 5350
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.868006944656372,
+      "eval_runtime": 38.5153,
+      "eval_samples_per_second": 13.371,
+      "eval_steps_per_second": 1.688,
+      "step": 5354
+    },
+    {
+      "epoch": 4.004482629809488,
+      "grad_norm": 0.6023468971252441,
+      "learning_rate": 0.0002,
+      "loss": 1.4535,
+      "step": 5360
+    },
+    {
+      "epoch": 4.011953679491969,
+      "grad_norm": 0.8589285612106323,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 5370
+    },
+    {
+      "epoch": 4.019424729174449,
+      "grad_norm": 0.7477491497993469,
+      "learning_rate": 0.0002,
+      "loss": 1.3952,
+      "step": 5380
+    },
+    {
+      "epoch": 4.02689577885693,
+      "grad_norm": 0.7601922154426575,
+      "learning_rate": 0.0002,
+      "loss": 1.3745,
+      "step": 5390
+    },
+    {
+      "epoch": 4.03436682853941,
+      "grad_norm": 0.8115614056587219,
+      "learning_rate": 0.0002,
+      "loss": 1.4133,
+      "step": 5400
+    },
+    {
+      "epoch": 4.04183787822189,
+      "grad_norm": 0.669925332069397,
+      "learning_rate": 0.0002,
+      "loss": 1.3748,
+      "step": 5410
+    },
+    {
+      "epoch": 4.04930892790437,
+      "grad_norm": 0.8091904520988464,
+      "learning_rate": 0.0002,
+      "loss": 1.2835,
+      "step": 5420
+    },
+    {
+      "epoch": 4.056779977586851,
+      "grad_norm": 0.709405779838562,
+      "learning_rate": 0.0002,
+      "loss": 1.3615,
+      "step": 5430
+    },
+    {
+      "epoch": 4.064251027269331,
+      "grad_norm": 1.0006179809570312,
+      "learning_rate": 0.0002,
+      "loss": 1.3558,
+      "step": 5440
+    },
+    {
+      "epoch": 4.071722076951811,
+      "grad_norm": 0.7017965912818909,
+      "learning_rate": 0.0002,
+      "loss": 1.3491,
+      "step": 5450
+    },
+    {
+      "epoch": 4.0791931266342925,
+      "grad_norm": 0.8991572260856628,
+      "learning_rate": 0.0002,
+      "loss": 1.3642,
+      "step": 5460
+    },
+    {
+      "epoch": 4.086664176316773,
+      "grad_norm": 0.9064797759056091,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 5470
+    },
+    {
+      "epoch": 4.094135225999253,
+      "grad_norm": 0.7981749176979065,
+      "learning_rate": 0.0002,
+      "loss": 1.3425,
+      "step": 5480
+    },
+    {
+      "epoch": 4.101606275681733,
+      "grad_norm": 0.7280883193016052,
+      "learning_rate": 0.0002,
+      "loss": 1.3826,
+      "step": 5490
+    },
+    {
+      "epoch": 4.109077325364214,
+      "grad_norm": 0.7419600486755371,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 5500
+    },
+    {
+      "epoch": 4.116548375046694,
+      "grad_norm": 0.8019949197769165,
+      "learning_rate": 0.0002,
+      "loss": 1.3199,
+      "step": 5510
+    },
+    {
+      "epoch": 4.124019424729174,
+      "grad_norm": 0.7501229047775269,
+      "learning_rate": 0.0002,
+      "loss": 1.3133,
+      "step": 5520
+    },
+    {
+      "epoch": 4.131490474411655,
+      "grad_norm": 0.8166249990463257,
+      "learning_rate": 0.0002,
+      "loss": 1.4432,
+      "step": 5530
+    },
+    {
+      "epoch": 4.138961524094135,
+      "grad_norm": 0.9728496074676514,
+      "learning_rate": 0.0002,
+      "loss": 1.3901,
+      "step": 5540
+    },
+    {
+      "epoch": 4.1464325737766154,
+      "grad_norm": 0.7590922117233276,
+      "learning_rate": 0.0002,
+      "loss": 1.3538,
+      "step": 5550
+    },
+    {
+      "epoch": 4.153903623459096,
+      "grad_norm": 0.7759010791778564,
+      "learning_rate": 0.0002,
+      "loss": 1.4368,
+      "step": 5560
+    },
+    {
+      "epoch": 4.161374673141577,
+      "grad_norm": 0.9057986736297607,
+      "learning_rate": 0.0002,
+      "loss": 1.3635,
+      "step": 5570
+    },
+    {
+      "epoch": 4.168845722824057,
+      "grad_norm": 0.8853937983512878,
+      "learning_rate": 0.0002,
+      "loss": 1.4152,
+      "step": 5580
+    },
+    {
+      "epoch": 4.176316772506537,
+      "grad_norm": 0.7070684432983398,
+      "learning_rate": 0.0002,
+      "loss": 1.3633,
+      "step": 5590
+    },
+    {
+      "epoch": 4.183787822189018,
+      "grad_norm": 0.7649410963058472,
+      "learning_rate": 0.0002,
+      "loss": 1.3218,
+      "step": 5600
+    },
+    {
+      "epoch": 4.191258871871498,
+      "grad_norm": 1.2048029899597168,
+      "learning_rate": 0.0002,
+      "loss": 1.3857,
+      "step": 5610
+    },
+    {
+      "epoch": 4.198729921553978,
+      "grad_norm": 0.7986605763435364,
+      "learning_rate": 0.0002,
+      "loss": 1.3629,
+      "step": 5620
+    },
+    {
+      "epoch": 4.206200971236458,
+      "grad_norm": 0.8151885867118835,
+      "learning_rate": 0.0002,
+      "loss": 1.3995,
+      "step": 5630
+    },
+    {
+      "epoch": 4.213672020918939,
+      "grad_norm": 0.7719064354896545,
+      "learning_rate": 0.0002,
+      "loss": 1.3782,
+      "step": 5640
+    },
+    {
+      "epoch": 4.2211430706014195,
+      "grad_norm": 0.8422448039054871,
+      "learning_rate": 0.0002,
+      "loss": 1.3852,
+      "step": 5650
+    },
+    {
+      "epoch": 4.2286141202839,
+      "grad_norm": 0.7017164826393127,
+      "learning_rate": 0.0002,
+      "loss": 1.3321,
+      "step": 5660
+    },
+    {
+      "epoch": 4.236085169966381,
+      "grad_norm": 0.8559677600860596,
+      "learning_rate": 0.0002,
+      "loss": 1.4105,
+      "step": 5670
+    },
+    {
+      "epoch": 4.243556219648861,
+      "grad_norm": 0.8216157555580139,
+      "learning_rate": 0.0002,
+      "loss": 1.3701,
+      "step": 5680
+    },
+    {
+      "epoch": 4.251027269331341,
+      "grad_norm": 0.7681755423545837,
+      "learning_rate": 0.0002,
+      "loss": 1.3565,
+      "step": 5690
+    },
+    {
+      "epoch": 4.258498319013821,
+      "grad_norm": 0.811665952205658,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 5700
+    },
+    {
+      "epoch": 4.265969368696302,
+      "grad_norm": 0.7242204546928406,
+      "learning_rate": 0.0002,
+      "loss": 1.4161,
+      "step": 5710
+    },
+    {
+      "epoch": 4.273440418378782,
+      "grad_norm": 0.7570181488990784,
+      "learning_rate": 0.0002,
+      "loss": 1.2958,
+      "step": 5720
+    },
+    {
+      "epoch": 4.280911468061262,
+      "grad_norm": 0.8951969146728516,
+      "learning_rate": 0.0002,
+      "loss": 1.4265,
+      "step": 5730
+    },
+    {
+      "epoch": 4.288382517743743,
+      "grad_norm": 0.7222902178764343,
+      "learning_rate": 0.0002,
+      "loss": 1.3895,
+      "step": 5740
+    },
+    {
+      "epoch": 4.2958535674262235,
+      "grad_norm": 0.8508469462394714,
+      "learning_rate": 0.0002,
+      "loss": 1.4155,
+      "step": 5750
+    },
+    {
+      "epoch": 4.303324617108704,
+      "grad_norm": 0.7215430736541748,
+      "learning_rate": 0.0002,
+      "loss": 1.365,
+      "step": 5760
+    },
+    {
+      "epoch": 4.310795666791184,
+      "grad_norm": 0.8774884939193726,
+      "learning_rate": 0.0002,
+      "loss": 1.4472,
+      "step": 5770
+    },
+    {
+      "epoch": 4.318266716473665,
+      "grad_norm": 0.8354552984237671,
+      "learning_rate": 0.0002,
+      "loss": 1.427,
+      "step": 5780
+    },
+    {
+      "epoch": 4.325737766156145,
+      "grad_norm": 0.6938814520835876,
+      "learning_rate": 0.0002,
+      "loss": 1.3222,
+      "step": 5790
+    },
+    {
+      "epoch": 4.333208815838625,
+      "grad_norm": 0.78675377368927,
+      "learning_rate": 0.0002,
+      "loss": 1.3589,
+      "step": 5800
+    },
+    {
+      "epoch": 4.340679865521106,
+      "grad_norm": 0.7147697806358337,
+      "learning_rate": 0.0002,
+      "loss": 1.3662,
+      "step": 5810
+    },
+    {
+      "epoch": 4.348150915203586,
+      "grad_norm": 0.7693623304367065,
+      "learning_rate": 0.0002,
+      "loss": 1.3597,
+      "step": 5820
+    },
+    {
+      "epoch": 4.355621964886066,
+      "grad_norm": 0.856517493724823,
+      "learning_rate": 0.0002,
+      "loss": 1.2944,
+      "step": 5830
+    },
+    {
+      "epoch": 4.3630930145685465,
+      "grad_norm": 0.7200973033905029,
+      "learning_rate": 0.0002,
+      "loss": 1.4307,
+      "step": 5840
+    },
+    {
+      "epoch": 4.3705640642510275,
+      "grad_norm": 0.743281364440918,
+      "learning_rate": 0.0002,
+      "loss": 1.442,
+      "step": 5850
+    },
+    {
+      "epoch": 4.378035113933508,
+      "grad_norm": 0.7627727389335632,
+      "learning_rate": 0.0002,
+      "loss": 1.3999,
+      "step": 5860
+    },
+    {
+      "epoch": 4.385506163615988,
+      "grad_norm": 0.7238836884498596,
+      "learning_rate": 0.0002,
+      "loss": 1.4082,
+      "step": 5870
+    },
+    {
+      "epoch": 4.392977213298469,
+      "grad_norm": 0.7253410816192627,
+      "learning_rate": 0.0002,
+      "loss": 1.4292,
+      "step": 5880
+    },
+    {
+      "epoch": 4.400448262980949,
+      "grad_norm": 0.8232238292694092,
+      "learning_rate": 0.0002,
+      "loss": 1.3774,
+      "step": 5890
+    },
+    {
+      "epoch": 4.407919312663429,
+      "grad_norm": 0.8778504729270935,
+      "learning_rate": 0.0002,
+      "loss": 1.3757,
+      "step": 5900
+    },
+    {
+      "epoch": 4.415390362345909,
+      "grad_norm": 0.7639474868774414,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 5910
+    },
+    {
+      "epoch": 4.42286141202839,
+      "grad_norm": 0.7666519284248352,
+      "learning_rate": 0.0002,
+      "loss": 1.3862,
+      "step": 5920
+    },
+    {
+      "epoch": 4.43033246171087,
+      "grad_norm": 0.867132842540741,
+      "learning_rate": 0.0002,
+      "loss": 1.4168,
+      "step": 5930
+    },
+    {
+      "epoch": 4.4378035113933505,
+      "grad_norm": 0.7571166753768921,
+      "learning_rate": 0.0002,
+      "loss": 1.4772,
+      "step": 5940
+    },
+    {
+      "epoch": 4.4452745610758315,
+      "grad_norm": 0.7911370992660522,
+      "learning_rate": 0.0002,
+      "loss": 1.4401,
+      "step": 5950
+    },
+    {
+      "epoch": 4.452745610758312,
+      "grad_norm": 0.8844250440597534,
+      "learning_rate": 0.0002,
+      "loss": 1.4516,
+      "step": 5960
+    },
+    {
+      "epoch": 4.460216660440792,
+      "grad_norm": 0.7336231470108032,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 5970
+    },
+    {
+      "epoch": 4.467687710123272,
+      "grad_norm": 0.8162738084793091,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 5980
+    },
+    {
+      "epoch": 4.475158759805753,
+      "grad_norm": 0.7413017153739929,
+      "learning_rate": 0.0002,
+      "loss": 1.393,
+      "step": 5990
+    },
+    {
+      "epoch": 4.482629809488233,
+      "grad_norm": 0.7215432524681091,
+      "learning_rate": 0.0002,
+      "loss": 1.3712,
+      "step": 6000
+    },
+    {
+      "epoch": 4.490100859170713,
+      "grad_norm": 0.8943389058113098,
+      "learning_rate": 0.0002,
+      "loss": 1.3521,
+      "step": 6010
+    },
+    {
+      "epoch": 4.497571908853194,
+      "grad_norm": 0.7850823998451233,
+      "learning_rate": 0.0002,
+      "loss": 1.4172,
+      "step": 6020
+    },
+    {
+      "epoch": 4.505042958535674,
+      "grad_norm": 0.8117504119873047,
+      "learning_rate": 0.0002,
+      "loss": 1.3582,
+      "step": 6030
+    },
+    {
+      "epoch": 4.5125140082181545,
+      "grad_norm": 0.8381605744361877,
+      "learning_rate": 0.0002,
+      "loss": 1.4272,
+      "step": 6040
+    },
+    {
+      "epoch": 4.519985057900635,
+      "grad_norm": 0.7964059710502625,
+      "learning_rate": 0.0002,
+      "loss": 1.3829,
+      "step": 6050
+    },
+    {
+      "epoch": 4.527456107583116,
+      "grad_norm": 0.7935128211975098,
+      "learning_rate": 0.0002,
+      "loss": 1.3555,
+      "step": 6060
+    },
+    {
+      "epoch": 4.534927157265596,
+      "grad_norm": 0.8725124597549438,
+      "learning_rate": 0.0002,
+      "loss": 1.3994,
+      "step": 6070
+    },
+    {
+      "epoch": 4.542398206948076,
+      "grad_norm": 0.880325198173523,
+      "learning_rate": 0.0002,
+      "loss": 1.3923,
+      "step": 6080
+    },
+    {
+      "epoch": 4.549869256630557,
+      "grad_norm": 0.7220637202262878,
+      "learning_rate": 0.0002,
+      "loss": 1.4459,
+      "step": 6090
+    },
+    {
+      "epoch": 4.557340306313037,
+      "grad_norm": 0.6908547878265381,
+      "learning_rate": 0.0002,
+      "loss": 1.3281,
+      "step": 6100
+    },
+    {
+      "epoch": 4.564811355995517,
+      "grad_norm": 0.797931969165802,
+      "learning_rate": 0.0002,
+      "loss": 1.437,
+      "step": 6110
+    },
+    {
+      "epoch": 4.572282405677997,
+      "grad_norm": 0.7056134343147278,
+      "learning_rate": 0.0002,
+      "loss": 1.4023,
+      "step": 6120
+    },
+    {
+      "epoch": 4.579753455360478,
+      "grad_norm": 0.7850478887557983,
+      "learning_rate": 0.0002,
+      "loss": 1.3814,
+      "step": 6130
+    },
+    {
+      "epoch": 4.5872245050429585,
+      "grad_norm": 0.8112621307373047,
+      "learning_rate": 0.0002,
+      "loss": 1.3579,
+      "step": 6140
+    },
+    {
+      "epoch": 4.594695554725439,
+      "grad_norm": 0.7040849328041077,
+      "learning_rate": 0.0002,
+      "loss": 1.3523,
+      "step": 6150
+    },
+    {
+      "epoch": 4.60216660440792,
+      "grad_norm": 0.7214553952217102,
+      "learning_rate": 0.0002,
+      "loss": 1.3526,
+      "step": 6160
+    },
+    {
+      "epoch": 4.6096376540904,
+      "grad_norm": 0.8616511821746826,
+      "learning_rate": 0.0002,
+      "loss": 1.3932,
+      "step": 6170
+    },
+    {
+      "epoch": 4.61710870377288,
+      "grad_norm": 0.8374658226966858,
+      "learning_rate": 0.0002,
+      "loss": 1.4622,
+      "step": 6180
+    },
+    {
+      "epoch": 4.62457975345536,
+      "grad_norm": 0.6761606931686401,
+      "learning_rate": 0.0002,
+      "loss": 1.3703,
+      "step": 6190
+    },
+    {
+      "epoch": 4.632050803137841,
+      "grad_norm": 0.768028199672699,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 6200
+    },
+    {
+      "epoch": 4.639521852820321,
+      "grad_norm": 0.9372717142105103,
+      "learning_rate": 0.0002,
+      "loss": 1.3772,
+      "step": 6210
+    },
+    {
+      "epoch": 4.646992902502801,
+      "grad_norm": 0.7906546592712402,
+      "learning_rate": 0.0002,
+      "loss": 1.4098,
+      "step": 6220
+    },
+    {
+      "epoch": 4.654463952185282,
+      "grad_norm": 0.7376723289489746,
+      "learning_rate": 0.0002,
+      "loss": 1.3962,
+      "step": 6230
+    },
+    {
+      "epoch": 4.6619350018677626,
+      "grad_norm": 0.8972630500793457,
+      "learning_rate": 0.0002,
+      "loss": 1.4529,
+      "step": 6240
+    },
+    {
+      "epoch": 4.669406051550243,
+      "grad_norm": 0.8261756300926208,
+      "learning_rate": 0.0002,
+      "loss": 1.4668,
+      "step": 6250
+    },
+    {
+      "epoch": 4.676877101232723,
+      "grad_norm": 0.7512393593788147,
+      "learning_rate": 0.0002,
+      "loss": 1.3267,
+      "step": 6260
+    },
+    {
+      "epoch": 4.684348150915204,
+      "grad_norm": 0.7132362127304077,
+      "learning_rate": 0.0002,
+      "loss": 1.4278,
+      "step": 6270
+    },
+    {
+      "epoch": 4.691819200597684,
+      "grad_norm": 0.7690575122833252,
+      "learning_rate": 0.0002,
+      "loss": 1.4299,
+      "step": 6280
+    },
+    {
+      "epoch": 4.699290250280164,
+      "grad_norm": 0.9886258840560913,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 6290
+    },
+    {
+      "epoch": 4.706761299962645,
+      "grad_norm": 0.9502435922622681,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 6300
+    },
+    {
+      "epoch": 4.714232349645125,
+      "grad_norm": 0.702255129814148,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 6310
+    },
+    {
+      "epoch": 4.721703399327605,
+      "grad_norm": 0.7713103890419006,
+      "learning_rate": 0.0002,
+      "loss": 1.4447,
+      "step": 6320
+    },
+    {
+      "epoch": 4.7291744490100855,
+      "grad_norm": 0.7778580784797668,
+      "learning_rate": 0.0002,
+      "loss": 1.4392,
+      "step": 6330
+    },
+    {
+      "epoch": 4.736645498692567,
+      "grad_norm": 0.7275111079216003,
+      "learning_rate": 0.0002,
+      "loss": 1.4169,
+      "step": 6340
+    },
+    {
+      "epoch": 4.744116548375047,
+      "grad_norm": 0.7728744149208069,
+      "learning_rate": 0.0002,
+      "loss": 1.4429,
+      "step": 6350
+    },
+    {
+      "epoch": 4.751587598057527,
+      "grad_norm": 0.9724260568618774,
+      "learning_rate": 0.0002,
+      "loss": 1.3756,
+      "step": 6360
+    },
+    {
+      "epoch": 4.759058647740007,
+      "grad_norm": 0.7505622506141663,
+      "learning_rate": 0.0002,
+      "loss": 1.3358,
+      "step": 6370
+    },
+    {
+      "epoch": 4.766529697422488,
+      "grad_norm": 0.7994682788848877,
+      "learning_rate": 0.0002,
+      "loss": 1.379,
+      "step": 6380
+    },
+    {
+      "epoch": 4.774000747104968,
+      "grad_norm": 0.8432038426399231,
+      "learning_rate": 0.0002,
+      "loss": 1.4275,
+      "step": 6390
+    },
+    {
+      "epoch": 4.781471796787448,
+      "grad_norm": 0.7436022758483887,
+      "learning_rate": 0.0002,
+      "loss": 1.4606,
+      "step": 6400
+    },
+    {
+      "epoch": 4.788942846469929,
+      "grad_norm": 0.7709194421768188,
+      "learning_rate": 0.0002,
+      "loss": 1.3461,
+      "step": 6410
+    },
+    {
+      "epoch": 4.796413896152409,
+      "grad_norm": 0.8798436522483826,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 6420
+    },
+    {
+      "epoch": 4.80388494583489,
+      "grad_norm": 0.790189266204834,
+      "learning_rate": 0.0002,
+      "loss": 1.3761,
+      "step": 6430
+    },
+    {
+      "epoch": 4.811355995517371,
+      "grad_norm": 0.6824303865432739,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 6440
+    },
+    {
+      "epoch": 4.818827045199851,
+      "grad_norm": 0.7501044869422913,
+      "learning_rate": 0.0002,
+      "loss": 1.3877,
+      "step": 6450
+    },
+    {
+      "epoch": 4.826298094882331,
+      "grad_norm": 0.8840398192405701,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 6460
+    },
+    {
+      "epoch": 4.833769144564811,
+      "grad_norm": 0.7812688946723938,
+      "learning_rate": 0.0002,
+      "loss": 1.4412,
+      "step": 6470
+    },
+    {
+      "epoch": 4.841240194247292,
+      "grad_norm": 0.7429926991462708,
+      "learning_rate": 0.0002,
+      "loss": 1.4299,
+      "step": 6480
+    },
+    {
+      "epoch": 4.848711243929772,
+      "grad_norm": 0.7778021693229675,
+      "learning_rate": 0.0002,
+      "loss": 1.5062,
+      "step": 6490
+    },
+    {
+      "epoch": 4.856182293612252,
+      "grad_norm": 0.8270702362060547,
+      "learning_rate": 0.0002,
+      "loss": 1.4589,
+      "step": 6500
+    },
+    {
+      "epoch": 4.863653343294732,
+      "grad_norm": 0.6960513591766357,
+      "learning_rate": 0.0002,
+      "loss": 1.4091,
+      "step": 6510
+    },
+    {
+      "epoch": 4.8711243929772134,
+      "grad_norm": 0.7728942632675171,
+      "learning_rate": 0.0002,
+      "loss": 1.376,
+      "step": 6520
+    },
+    {
+      "epoch": 4.878595442659694,
+      "grad_norm": 0.7377303838729858,
+      "learning_rate": 0.0002,
+      "loss": 1.4852,
+      "step": 6530
+    },
+    {
+      "epoch": 4.886066492342174,
+      "grad_norm": 0.7257253527641296,
+      "learning_rate": 0.0002,
+      "loss": 1.3846,
+      "step": 6540
+    },
+    {
+      "epoch": 4.893537542024655,
+      "grad_norm": 0.7875821590423584,
+      "learning_rate": 0.0002,
+      "loss": 1.4166,
+      "step": 6550
+    },
+    {
+      "epoch": 4.901008591707135,
+      "grad_norm": 0.8346304297447205,
+      "learning_rate": 0.0002,
+      "loss": 1.357,
+      "step": 6560
+    },
+    {
+      "epoch": 4.908479641389615,
+      "grad_norm": 0.7710739374160767,
+      "learning_rate": 0.0002,
+      "loss": 1.4522,
+      "step": 6570
+    },
+    {
+      "epoch": 4.915950691072096,
+      "grad_norm": 0.7015138268470764,
+      "learning_rate": 0.0002,
+      "loss": 1.4465,
+      "step": 6580
+    },
+    {
+      "epoch": 4.923421740754576,
+      "grad_norm": 0.8707432150840759,
+      "learning_rate": 0.0002,
+      "loss": 1.435,
+      "step": 6590
+    },
+    {
+      "epoch": 4.930892790437056,
+      "grad_norm": 0.786601185798645,
+      "learning_rate": 0.0002,
+      "loss": 1.2968,
+      "step": 6600
+    },
+    {
+      "epoch": 4.938363840119536,
+      "grad_norm": 0.978519082069397,
+      "learning_rate": 0.0002,
+      "loss": 1.4385,
+      "step": 6610
+    },
+    {
+      "epoch": 4.9458348898020175,
+      "grad_norm": 0.8102927207946777,
+      "learning_rate": 0.0002,
+      "loss": 1.3997,
+      "step": 6620
+    },
+    {
+      "epoch": 4.953305939484498,
+      "grad_norm": 0.7628704309463501,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 6630
+    },
+    {
+      "epoch": 4.960776989166978,
+      "grad_norm": 0.8053455352783203,
+      "learning_rate": 0.0002,
+      "loss": 1.3774,
+      "step": 6640
+    },
+    {
+      "epoch": 4.968248038849458,
+      "grad_norm": 0.8680412173271179,
+      "learning_rate": 0.0002,
+      "loss": 1.5092,
+      "step": 6650
+    },
+    {
+      "epoch": 4.975719088531939,
+      "grad_norm": 0.7415758371353149,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 6660
+    },
+    {
+      "epoch": 4.983190138214419,
+      "grad_norm": 0.7730312347412109,
+      "learning_rate": 0.0002,
+      "loss": 1.3793,
+      "step": 6670
+    },
+    {
+      "epoch": 4.990661187896899,
+      "grad_norm": 0.7924041152000427,
+      "learning_rate": 0.0002,
+      "loss": 1.4863,
+      "step": 6680
+    },
+    {
+      "epoch": 4.99813223757938,
+      "grad_norm": 0.8677893877029419,
+      "learning_rate": 0.0002,
+      "loss": 1.4137,
+      "step": 6690
+    },
+    {
+      "epoch": 4.999626447515876,
+      "eval_loss": 1.9444633722305298,
+      "eval_runtime": 39.3488,
+      "eval_samples_per_second": 13.088,
+      "eval_steps_per_second": 1.652,
+      "step": 6692
+    },
+    {
+      "epoch": 5.00560328726186,
+      "grad_norm": 0.7102245092391968,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 6700
+    },
+    {
+      "epoch": 5.0130743369443405,
+      "grad_norm": 1.0425463914871216,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 6710
+    },
+    {
+      "epoch": 5.0205453866268215,
+      "grad_norm": 0.9320756793022156,
+      "learning_rate": 0.0002,
+      "loss": 1.181,
+      "step": 6720
+    },
+    {
+      "epoch": 5.028016436309302,
+      "grad_norm": 0.8797217607498169,
+      "learning_rate": 0.0002,
+      "loss": 1.1786,
+      "step": 6730
+    },
+    {
+      "epoch": 5.035487485991782,
+      "grad_norm": 2.135707139968872,
+      "learning_rate": 0.0002,
+      "loss": 1.2097,
+      "step": 6740
+    },
+    {
+      "epoch": 5.042958535674262,
+      "grad_norm": 0.8747734427452087,
+      "learning_rate": 0.0002,
+      "loss": 1.1761,
+      "step": 6750
+    },
+    {
+      "epoch": 5.050429585356743,
+      "grad_norm": 0.9981076717376709,
+      "learning_rate": 0.0002,
+      "loss": 1.1675,
+      "step": 6760
+    },
+    {
+      "epoch": 5.057900635039223,
+      "grad_norm": 0.985078752040863,
+      "learning_rate": 0.0002,
+      "loss": 1.1976,
+      "step": 6770
+    },
+    {
+      "epoch": 5.065371684721703,
+      "grad_norm": 1.0974019765853882,
+      "learning_rate": 0.0002,
+      "loss": 1.2688,
+      "step": 6780
+    },
+    {
+      "epoch": 5.072842734404184,
+      "grad_norm": 0.9823219180107117,
+      "learning_rate": 0.0002,
+      "loss": 1.1982,
+      "step": 6790
+    },
+    {
+      "epoch": 5.080313784086664,
+      "grad_norm": 1.122605562210083,
+      "learning_rate": 0.0002,
+      "loss": 1.2586,
+      "step": 6800
+    },
+    {
+      "epoch": 5.0877848337691445,
+      "grad_norm": 0.8556802272796631,
+      "learning_rate": 0.0002,
+      "loss": 1.2069,
+      "step": 6810
+    },
+    {
+      "epoch": 5.095255883451625,
+      "grad_norm": 1.1699262857437134,
+      "learning_rate": 0.0002,
+      "loss": 1.1908,
+      "step": 6820
+    },
+    {
+      "epoch": 5.102726933134106,
+      "grad_norm": 1.0440590381622314,
+      "learning_rate": 0.0002,
+      "loss": 1.1869,
+      "step": 6830
+    },
+    {
+      "epoch": 5.110197982816586,
+      "grad_norm": 1.0445300340652466,
+      "learning_rate": 0.0002,
+      "loss": 1.1655,
+      "step": 6840
+    },
+    {
+      "epoch": 5.117669032499066,
+      "grad_norm": 0.8289563059806824,
+      "learning_rate": 0.0002,
+      "loss": 1.2392,
+      "step": 6850
+    },
+    {
+      "epoch": 5.125140082181547,
+      "grad_norm": 1.1051193475723267,
+      "learning_rate": 0.0002,
+      "loss": 1.1687,
+      "step": 6860
+    },
+    {
+      "epoch": 5.132611131864027,
+      "grad_norm": 0.9345614910125732,
+      "learning_rate": 0.0002,
+      "loss": 1.2737,
+      "step": 6870
+    },
+    {
+      "epoch": 5.140082181546507,
+      "grad_norm": 1.1222996711730957,
+      "learning_rate": 0.0002,
+      "loss": 1.3021,
+      "step": 6880
+    },
+    {
+      "epoch": 5.147553231228987,
+      "grad_norm": 0.9405338764190674,
+      "learning_rate": 0.0002,
+      "loss": 1.2408,
+      "step": 6890
+    },
+    {
+      "epoch": 5.155024280911468,
+      "grad_norm": 1.0935171842575073,
+      "learning_rate": 0.0002,
+      "loss": 1.2367,
+      "step": 6900
+    },
+    {
+      "epoch": 5.1624953305939485,
+      "grad_norm": 1.0438612699508667,
+      "learning_rate": 0.0002,
+      "loss": 1.2458,
+      "step": 6910
+    },
+    {
+      "epoch": 5.169966380276429,
+      "grad_norm": 1.1189004182815552,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 6920
+    },
+    {
+      "epoch": 5.17743742995891,
+      "grad_norm": 1.0533215999603271,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 6930
+    },
+    {
+      "epoch": 5.18490847964139,
+      "grad_norm": 0.9779648780822754,
+      "learning_rate": 0.0002,
+      "loss": 1.2974,
+      "step": 6940
+    },
+    {
+      "epoch": 5.19237952932387,
+      "grad_norm": 0.8920868635177612,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 6950
+    },
+    {
+      "epoch": 5.19985057900635,
+      "grad_norm": 0.8374548554420471,
+      "learning_rate": 0.0002,
+      "loss": 1.283,
+      "step": 6960
+    },
+    {
+      "epoch": 5.207321628688831,
+      "grad_norm": 1.0490682125091553,
+      "learning_rate": 0.0002,
+      "loss": 1.2775,
+      "step": 6970
+    },
+    {
+      "epoch": 5.214792678371311,
+      "grad_norm": 0.9658287167549133,
+      "learning_rate": 0.0002,
+      "loss": 1.1826,
+      "step": 6980
+    },
+    {
+      "epoch": 5.222263728053791,
+      "grad_norm": 0.9652056097984314,
+      "learning_rate": 0.0002,
+      "loss": 1.2647,
+      "step": 6990
+    },
+    {
+      "epoch": 5.229734777736272,
+      "grad_norm": 0.9141794443130493,
+      "learning_rate": 0.0002,
+      "loss": 1.3023,
+      "step": 7000
+    },
+    {
+      "epoch": 5.2372058274187525,
+      "grad_norm": 0.9831376671791077,
+      "learning_rate": 0.0002,
+      "loss": 1.2456,
+      "step": 7010
+    },
+    {
+      "epoch": 5.244676877101233,
+      "grad_norm": 1.0198718309402466,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 7020
+    },
+    {
+      "epoch": 5.252147926783713,
+      "grad_norm": 0.9647888541221619,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 7030
+    },
+    {
+      "epoch": 5.259618976466194,
+      "grad_norm": 1.3941649198532104,
+      "learning_rate": 0.0002,
+      "loss": 1.2106,
+      "step": 7040
+    },
+    {
+      "epoch": 5.267090026148674,
+      "grad_norm": 1.0305466651916504,
+      "learning_rate": 0.0002,
+      "loss": 1.2885,
+      "step": 7050
+    },
+    {
+      "epoch": 5.274561075831154,
+      "grad_norm": 0.9577859044075012,
+      "learning_rate": 0.0002,
+      "loss": 1.2362,
+      "step": 7060
+    },
+    {
+      "epoch": 5.282032125513634,
+      "grad_norm": 1.149092197418213,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 7070
+    },
+    {
+      "epoch": 5.289503175196115,
+      "grad_norm": 1.2582733631134033,
+      "learning_rate": 0.0002,
+      "loss": 1.2986,
+      "step": 7080
+    },
+    {
+      "epoch": 5.296974224878595,
+      "grad_norm": 1.1777442693710327,
+      "learning_rate": 0.0002,
+      "loss": 1.2307,
+      "step": 7090
+    },
+    {
+      "epoch": 5.3044452745610755,
+      "grad_norm": 1.0076404809951782,
+      "learning_rate": 0.0002,
+      "loss": 1.24,
+      "step": 7100
+    },
+    {
+      "epoch": 5.3119163242435565,
+      "grad_norm": 0.9037365913391113,
+      "learning_rate": 0.0002,
+      "loss": 1.1407,
+      "step": 7110
+    },
+    {
+      "epoch": 5.319387373926037,
+      "grad_norm": 0.9428724646568298,
+      "learning_rate": 0.0002,
+      "loss": 1.238,
+      "step": 7120
+    },
+    {
+      "epoch": 5.326858423608517,
+      "grad_norm": 0.9935154318809509,
+      "learning_rate": 0.0002,
+      "loss": 1.2571,
+      "step": 7130
+    },
+    {
+      "epoch": 5.334329473290998,
+      "grad_norm": 1.087500810623169,
+      "learning_rate": 0.0002,
+      "loss": 1.2833,
+      "step": 7140
+    },
+    {
+      "epoch": 5.341800522973478,
+      "grad_norm": 0.8543072938919067,
+      "learning_rate": 0.0002,
+      "loss": 1.2304,
+      "step": 7150
+    },
+    {
+      "epoch": 5.349271572655958,
+      "grad_norm": 0.9323700070381165,
+      "learning_rate": 0.0002,
+      "loss": 1.2755,
+      "step": 7160
+    },
+    {
+      "epoch": 5.356742622338438,
+      "grad_norm": 1.0037827491760254,
+      "learning_rate": 0.0002,
+      "loss": 1.2769,
+      "step": 7170
+    },
+    {
+      "epoch": 5.364213672020919,
+      "grad_norm": 0.8746469616889954,
+      "learning_rate": 0.0002,
+      "loss": 1.3204,
+      "step": 7180
+    },
+    {
+      "epoch": 5.371684721703399,
+      "grad_norm": 0.9516328573226929,
+      "learning_rate": 0.0002,
+      "loss": 1.2759,
+      "step": 7190
+    },
+    {
+      "epoch": 5.3791557713858795,
+      "grad_norm": 0.9395177364349365,
+      "learning_rate": 0.0002,
+      "loss": 1.2428,
+      "step": 7200
+    },
+    {
+      "epoch": 5.38662682106836,
+      "grad_norm": 1.000369906425476,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 7210
+    },
+    {
+      "epoch": 5.394097870750841,
+      "grad_norm": 1.0845502614974976,
+      "learning_rate": 0.0002,
+      "loss": 1.2337,
+      "step": 7220
+    },
+    {
+      "epoch": 5.401568920433321,
+      "grad_norm": 0.8975145220756531,
+      "learning_rate": 0.0002,
+      "loss": 1.2776,
+      "step": 7230
+    },
+    {
+      "epoch": 5.409039970115801,
+      "grad_norm": 1.040077805519104,
+      "learning_rate": 0.0002,
+      "loss": 1.2306,
+      "step": 7240
+    },
+    {
+      "epoch": 5.416511019798282,
+      "grad_norm": 1.0729942321777344,
+      "learning_rate": 0.0002,
+      "loss": 1.2277,
+      "step": 7250
+    },
+    {
+      "epoch": 5.423982069480762,
+      "grad_norm": 0.8322232961654663,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 7260
+    },
+    {
+      "epoch": 5.431453119163242,
+      "grad_norm": 1.0654641389846802,
+      "learning_rate": 0.0002,
+      "loss": 1.3036,
+      "step": 7270
+    },
+    {
+      "epoch": 5.438924168845723,
+      "grad_norm": 1.0445852279663086,
+      "learning_rate": 0.0002,
+      "loss": 1.268,
+      "step": 7280
+    },
+    {
+      "epoch": 5.446395218528203,
+      "grad_norm": 1.0762956142425537,
+      "learning_rate": 0.0002,
+      "loss": 1.2743,
+      "step": 7290
+    },
+    {
+      "epoch": 5.4538662682106835,
+      "grad_norm": 0.9721953868865967,
+      "learning_rate": 0.0002,
+      "loss": 1.2887,
+      "step": 7300
+    },
+    {
+      "epoch": 5.461337317893164,
+      "grad_norm": 0.9238539338111877,
+      "learning_rate": 0.0002,
+      "loss": 1.2833,
+      "step": 7310
+    },
+    {
+      "epoch": 5.468808367575645,
+      "grad_norm": 0.9912874102592468,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 7320
+    },
+    {
+      "epoch": 5.476279417258125,
+      "grad_norm": 1.0727077722549438,
+      "learning_rate": 0.0002,
+      "loss": 1.2557,
+      "step": 7330
+    },
+    {
+      "epoch": 5.483750466940605,
+      "grad_norm": 0.8633865118026733,
+      "learning_rate": 0.0002,
+      "loss": 1.3471,
+      "step": 7340
+    },
+    {
+      "epoch": 5.491221516623085,
+      "grad_norm": 0.9396262764930725,
+      "learning_rate": 0.0002,
+      "loss": 1.3155,
+      "step": 7350
+    },
+    {
+      "epoch": 5.498692566305566,
+      "grad_norm": 1.0253715515136719,
+      "learning_rate": 0.0002,
+      "loss": 1.3146,
+      "step": 7360
+    },
+    {
+      "epoch": 5.506163615988046,
+      "grad_norm": 1.006047010421753,
+      "learning_rate": 0.0002,
+      "loss": 1.3156,
+      "step": 7370
+    },
+    {
+      "epoch": 5.513634665670526,
+      "grad_norm": 0.9781233072280884,
+      "learning_rate": 0.0002,
+      "loss": 1.3107,
+      "step": 7380
+    },
+    {
+      "epoch": 5.521105715353007,
+      "grad_norm": 0.9945126175880432,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 7390
+    },
+    {
+      "epoch": 5.528576765035488,
+      "grad_norm": 0.9081175327301025,
+      "learning_rate": 0.0002,
+      "loss": 1.1936,
+      "step": 7400
+    },
+    {
+      "epoch": 5.536047814717968,
+      "grad_norm": 1.2215938568115234,
+      "learning_rate": 0.0002,
+      "loss": 1.2651,
+      "step": 7410
+    },
+    {
+      "epoch": 5.543518864400449,
+      "grad_norm": 1.0724077224731445,
+      "learning_rate": 0.0002,
+      "loss": 1.2484,
+      "step": 7420
+    },
+    {
+      "epoch": 5.550989914082929,
+      "grad_norm": 1.106955885887146,
+      "learning_rate": 0.0002,
+      "loss": 1.3083,
+      "step": 7430
+    },
+    {
+      "epoch": 5.558460963765409,
+      "grad_norm": 1.0657650232315063,
+      "learning_rate": 0.0002,
+      "loss": 1.2125,
+      "step": 7440
+    },
+    {
+      "epoch": 5.565932013447889,
+      "grad_norm": 0.9725455641746521,
+      "learning_rate": 0.0002,
+      "loss": 1.2576,
+      "step": 7450
+    },
+    {
+      "epoch": 5.57340306313037,
+      "grad_norm": 0.8604224324226379,
+      "learning_rate": 0.0002,
+      "loss": 1.3297,
+      "step": 7460
+    },
+    {
+      "epoch": 5.58087411281285,
+      "grad_norm": 0.9913371205329895,
+      "learning_rate": 0.0002,
+      "loss": 1.3084,
+      "step": 7470
+    },
+    {
+      "epoch": 5.58834516249533,
+      "grad_norm": 1.012073040008545,
+      "learning_rate": 0.0002,
+      "loss": 1.3371,
+      "step": 7480
+    },
+    {
+      "epoch": 5.5958162121778106,
+      "grad_norm": 1.1003159284591675,
+      "learning_rate": 0.0002,
+      "loss": 1.2526,
+      "step": 7490
+    },
+    {
+      "epoch": 5.603287261860292,
+      "grad_norm": 0.9104593992233276,
+      "learning_rate": 0.0002,
+      "loss": 1.2577,
+      "step": 7500
+    },
+    {
+      "epoch": 5.610758311542772,
+      "grad_norm": 0.9480831623077393,
+      "learning_rate": 0.0002,
+      "loss": 1.2578,
+      "step": 7510
+    },
+    {
+      "epoch": 5.618229361225252,
+      "grad_norm": 1.0826456546783447,
+      "learning_rate": 0.0002,
+      "loss": 1.3056,
+      "step": 7520
+    },
+    {
+      "epoch": 5.625700410907733,
+      "grad_norm": 0.8286259174346924,
+      "learning_rate": 0.0002,
+      "loss": 1.2931,
+      "step": 7530
+    },
+    {
+      "epoch": 5.633171460590213,
+      "grad_norm": 0.9145061373710632,
+      "learning_rate": 0.0002,
+      "loss": 1.2918,
+      "step": 7540
+    },
+    {
+      "epoch": 5.640642510272693,
+      "grad_norm": 0.9363601803779602,
+      "learning_rate": 0.0002,
+      "loss": 1.1736,
+      "step": 7550
+    },
+    {
+      "epoch": 5.648113559955174,
+      "grad_norm": 0.9553244709968567,
+      "learning_rate": 0.0002,
+      "loss": 1.2265,
+      "step": 7560
+    },
+    {
+      "epoch": 5.655584609637654,
+      "grad_norm": 1.0343557596206665,
+      "learning_rate": 0.0002,
+      "loss": 1.2356,
+      "step": 7570
+    },
+    {
+      "epoch": 5.663055659320134,
+      "grad_norm": 0.8734238743782043,
+      "learning_rate": 0.0002,
+      "loss": 1.3171,
+      "step": 7580
+    },
+    {
+      "epoch": 5.670526709002615,
+      "grad_norm": 1.0230586528778076,
+      "learning_rate": 0.0002,
+      "loss": 1.2785,
+      "step": 7590
+    },
+    {
+      "epoch": 5.677997758685096,
+      "grad_norm": 1.0063409805297852,
+      "learning_rate": 0.0002,
+      "loss": 1.2936,
+      "step": 7600
+    },
+    {
+      "epoch": 5.685468808367576,
+      "grad_norm": 1.0104626417160034,
+      "learning_rate": 0.0002,
+      "loss": 1.2396,
+      "step": 7610
+    },
+    {
+      "epoch": 5.692939858050056,
+      "grad_norm": 0.9528168439865112,
+      "learning_rate": 0.0002,
+      "loss": 1.2581,
+      "step": 7620
+    },
+    {
+      "epoch": 5.700410907732536,
+      "grad_norm": 0.9799878597259521,
+      "learning_rate": 0.0002,
+      "loss": 1.3116,
+      "step": 7630
+    },
+    {
+      "epoch": 5.707881957415017,
+      "grad_norm": 0.969351589679718,
+      "learning_rate": 0.0002,
+      "loss": 1.2632,
+      "step": 7640
+    },
+    {
+      "epoch": 5.715353007097497,
+      "grad_norm": 1.3037652969360352,
+      "learning_rate": 0.0002,
+      "loss": 1.3055,
+      "step": 7650
+    },
+    {
+      "epoch": 5.722824056779977,
+      "grad_norm": 1.0640486478805542,
+      "learning_rate": 0.0002,
+      "loss": 1.3126,
+      "step": 7660
+    },
+    {
+      "epoch": 5.730295106462458,
+      "grad_norm": 1.0416420698165894,
+      "learning_rate": 0.0002,
+      "loss": 1.3325,
+      "step": 7670
+    },
+    {
+      "epoch": 5.7377661561449385,
+      "grad_norm": 0.8893619775772095,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 7680
+    },
+    {
+      "epoch": 5.745237205827419,
+      "grad_norm": 0.8512844443321228,
+      "learning_rate": 0.0002,
+      "loss": 1.319,
+      "step": 7690
+    },
+    {
+      "epoch": 5.7527082555099,
+      "grad_norm": 0.9955748319625854,
+      "learning_rate": 0.0002,
+      "loss": 1.3328,
+      "step": 7700
+    },
+    {
+      "epoch": 5.76017930519238,
+      "grad_norm": 1.0409910678863525,
+      "learning_rate": 0.0002,
+      "loss": 1.294,
+      "step": 7710
+    },
+    {
+      "epoch": 5.76765035487486,
+      "grad_norm": 1.010097861289978,
+      "learning_rate": 0.0002,
+      "loss": 1.3518,
+      "step": 7720
+    },
+    {
+      "epoch": 5.77512140455734,
+      "grad_norm": 0.8974892497062683,
+      "learning_rate": 0.0002,
+      "loss": 1.2106,
+      "step": 7730
+    },
+    {
+      "epoch": 5.782592454239821,
+      "grad_norm": 0.972835123538971,
+      "learning_rate": 0.0002,
+      "loss": 1.2743,
+      "step": 7740
+    },
+    {
+      "epoch": 5.790063503922301,
+      "grad_norm": 0.9607440829277039,
+      "learning_rate": 0.0002,
+      "loss": 1.3549,
+      "step": 7750
+    },
+    {
+      "epoch": 5.797534553604781,
+      "grad_norm": 0.9426500797271729,
+      "learning_rate": 0.0002,
+      "loss": 1.29,
+      "step": 7760
+    },
+    {
+      "epoch": 5.8050056032872615,
+      "grad_norm": 0.8745320439338684,
+      "learning_rate": 0.0002,
+      "loss": 1.274,
+      "step": 7770
+    },
+    {
+      "epoch": 5.8124766529697425,
+      "grad_norm": 1.0117204189300537,
+      "learning_rate": 0.0002,
+      "loss": 1.3009,
+      "step": 7780
+    },
+    {
+      "epoch": 5.819947702652223,
+      "grad_norm": 1.0387755632400513,
+      "learning_rate": 0.0002,
+      "loss": 1.3135,
+      "step": 7790
+    },
+    {
+      "epoch": 5.827418752334703,
+      "grad_norm": 1.0709784030914307,
+      "learning_rate": 0.0002,
+      "loss": 1.2709,
+      "step": 7800
+    },
+    {
+      "epoch": 5.834889802017184,
+      "grad_norm": 0.9512667655944824,
+      "learning_rate": 0.0002,
+      "loss": 1.225,
+      "step": 7810
+    },
+    {
+      "epoch": 5.842360851699664,
+      "grad_norm": 1.021094560623169,
+      "learning_rate": 0.0002,
+      "loss": 1.3284,
+      "step": 7820
+    },
+    {
+      "epoch": 5.849831901382144,
+      "grad_norm": 1.117491364479065,
+      "learning_rate": 0.0002,
+      "loss": 1.2794,
+      "step": 7830
+    },
+    {
+      "epoch": 5.857302951064625,
+      "grad_norm": 0.9252554178237915,
+      "learning_rate": 0.0002,
+      "loss": 1.3646,
+      "step": 7840
+    },
+    {
+      "epoch": 5.864774000747105,
+      "grad_norm": 1.1416207551956177,
+      "learning_rate": 0.0002,
+      "loss": 1.2976,
+      "step": 7850
+    },
+    {
+      "epoch": 5.872245050429585,
+      "grad_norm": 1.1219907999038696,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 7860
+    },
+    {
+      "epoch": 5.8797161001120655,
+      "grad_norm": 0.8300467729568481,
+      "learning_rate": 0.0002,
+      "loss": 1.2334,
+      "step": 7870
+    },
+    {
+      "epoch": 5.8871871497945465,
+      "grad_norm": 1.00551438331604,
+      "learning_rate": 0.0002,
+      "loss": 1.3132,
+      "step": 7880
+    },
+    {
+      "epoch": 5.894658199477027,
+      "grad_norm": 0.8981153964996338,
+      "learning_rate": 0.0002,
+      "loss": 1.2609,
+      "step": 7890
+    },
+    {
+      "epoch": 5.902129249159507,
+      "grad_norm": 1.0247976779937744,
+      "learning_rate": 0.0002,
+      "loss": 1.2817,
+      "step": 7900
+    },
+    {
+      "epoch": 5.909600298841987,
+      "grad_norm": 1.0820319652557373,
+      "learning_rate": 0.0002,
+      "loss": 1.2866,
+      "step": 7910
+    },
+    {
+      "epoch": 5.917071348524468,
+      "grad_norm": 0.952675461769104,
+      "learning_rate": 0.0002,
+      "loss": 1.2941,
+      "step": 7920
+    },
+    {
+      "epoch": 5.924542398206948,
+      "grad_norm": 0.8666740655899048,
+      "learning_rate": 0.0002,
+      "loss": 1.307,
+      "step": 7930
+    },
+    {
+      "epoch": 5.932013447889428,
+      "grad_norm": 0.8640421032905579,
+      "learning_rate": 0.0002,
+      "loss": 1.2752,
+      "step": 7940
+    },
+    {
+      "epoch": 5.939484497571909,
+      "grad_norm": 1.2343276739120483,
+      "learning_rate": 0.0002,
+      "loss": 1.2386,
+      "step": 7950
+    },
+    {
+      "epoch": 5.946955547254389,
+      "grad_norm": 0.958046555519104,
+      "learning_rate": 0.0002,
+      "loss": 1.2333,
+      "step": 7960
+    },
+    {
+      "epoch": 5.9544265969368695,
+      "grad_norm": 1.0538510084152222,
+      "learning_rate": 0.0002,
+      "loss": 1.2352,
+      "step": 7970
+    },
+    {
+      "epoch": 5.9618976466193505,
+      "grad_norm": 1.2681571245193481,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 7980
+    },
+    {
+      "epoch": 5.969368696301831,
+      "grad_norm": 0.8171183466911316,
+      "learning_rate": 0.0002,
+      "loss": 1.2514,
+      "step": 7990
+    },
+    {
+      "epoch": 5.976839745984311,
+      "grad_norm": 0.9109523892402649,
+      "learning_rate": 0.0002,
+      "loss": 1.3412,
+      "step": 8000
+    },
+    {
+      "epoch": 5.984310795666791,
+      "grad_norm": 1.0040639638900757,
+      "learning_rate": 0.0002,
+      "loss": 1.3497,
+      "step": 8010
+    },
+    {
+      "epoch": 5.991781845349272,
+      "grad_norm": 0.9596554040908813,
+      "learning_rate": 0.0002,
+      "loss": 1.3299,
+      "step": 8020
+    },
+    {
+      "epoch": 5.999252895031752,
+      "grad_norm": 0.9782963991165161,
+      "learning_rate": 0.0002,
+      "loss": 1.3109,
+      "step": 8030
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.0417845249176025,
+      "eval_runtime": 38.8465,
+      "eval_samples_per_second": 13.257,
+      "eval_steps_per_second": 1.673,
+      "step": 8031
+    },
+    {
+      "epoch": 6.006723944714232,
+      "grad_norm": 1.380823016166687,
+      "learning_rate": 0.0002,
+      "loss": 1.0886,
+      "step": 8040
+    },
+    {
+      "epoch": 6.014194994396712,
+      "grad_norm": 1.067636251449585,
+      "learning_rate": 0.0002,
+      "loss": 1.0413,
+      "step": 8050
+    },
+    {
+      "epoch": 6.021666044079193,
+      "grad_norm": 1.363402009010315,
+      "learning_rate": 0.0002,
+      "loss": 1.0686,
+      "step": 8060
+    },
+    {
+      "epoch": 6.0291370937616735,
+      "grad_norm": 0.9901054501533508,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 8070
+    },
+    {
+      "epoch": 6.036608143444154,
+      "grad_norm": 1.1545379161834717,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 8080
+    },
+    {
+      "epoch": 6.044079193126635,
+      "grad_norm": 1.2259265184402466,
+      "learning_rate": 0.0002,
+      "loss": 1.0644,
+      "step": 8090
+    },
+    {
+      "epoch": 6.051550242809115,
+      "grad_norm": 1.1237425804138184,
+      "learning_rate": 0.0002,
+      "loss": 1.1273,
+      "step": 8100
+    },
+    {
+      "epoch": 6.059021292491595,
+      "grad_norm": 1.2805622816085815,
+      "learning_rate": 0.0002,
+      "loss": 1.1001,
+      "step": 8110
+    },
+    {
+      "epoch": 6.066492342174075,
+      "grad_norm": 1.2270452976226807,
+      "learning_rate": 0.0002,
+      "loss": 1.0731,
+      "step": 8120
+    },
+    {
+      "epoch": 6.073963391856556,
+      "grad_norm": 1.1924101114273071,
+      "learning_rate": 0.0002,
+      "loss": 1.0692,
+      "step": 8130
+    },
+    {
+      "epoch": 6.081434441539036,
+      "grad_norm": 1.2543894052505493,
+      "learning_rate": 0.0002,
+      "loss": 1.1698,
+      "step": 8140
+    },
+    {
+      "epoch": 6.088905491221516,
+      "grad_norm": 1.1821149587631226,
+      "learning_rate": 0.0002,
+      "loss": 1.069,
+      "step": 8150
+    },
+    {
+      "epoch": 6.096376540903997,
+      "grad_norm": 1.2202836275100708,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 8160
+    },
+    {
+      "epoch": 6.1038475905864775,
+      "grad_norm": 1.0576019287109375,
+      "learning_rate": 0.0002,
+      "loss": 1.136,
+      "step": 8170
+    },
+    {
+      "epoch": 6.111318640268958,
+      "grad_norm": 1.31708824634552,
+      "learning_rate": 0.0002,
+      "loss": 1.1395,
+      "step": 8180
+    },
+    {
+      "epoch": 6.118789689951438,
+      "grad_norm": 1.0479495525360107,
+      "learning_rate": 0.0002,
+      "loss": 1.0887,
+      "step": 8190
+    },
+    {
+      "epoch": 6.126260739633919,
+      "grad_norm": 1.285003423690796,
+      "learning_rate": 0.0002,
+      "loss": 1.0764,
+      "step": 8200
+    },
+    {
+      "epoch": 6.133731789316399,
+      "grad_norm": 1.0989165306091309,
+      "learning_rate": 0.0002,
+      "loss": 1.0642,
+      "step": 8210
+    },
+    {
+      "epoch": 6.141202838998879,
+      "grad_norm": 1.1659013032913208,
+      "learning_rate": 0.0002,
+      "loss": 1.0981,
+      "step": 8220
+    },
+    {
+      "epoch": 6.14867388868136,
+      "grad_norm": 1.2796376943588257,
+      "learning_rate": 0.0002,
+      "loss": 1.1138,
+      "step": 8230
+    },
+    {
+      "epoch": 6.15614493836384,
+      "grad_norm": 1.060564637184143,
+      "learning_rate": 0.0002,
+      "loss": 1.1116,
+      "step": 8240
+    },
+    {
+      "epoch": 6.16361598804632,
+      "grad_norm": 1.3884605169296265,
+      "learning_rate": 0.0002,
+      "loss": 1.1493,
+      "step": 8250
+    },
+    {
+      "epoch": 6.1710870377288005,
+      "grad_norm": 1.1570569276809692,
+      "learning_rate": 0.0002,
+      "loss": 1.0504,
+      "step": 8260
+    },
+    {
+      "epoch": 6.1785580874112815,
+      "grad_norm": 1.4136502742767334,
+      "learning_rate": 0.0002,
+      "loss": 1.0386,
+      "step": 8270
+    },
+    {
+      "epoch": 6.186029137093762,
+      "grad_norm": 1.3396095037460327,
+      "learning_rate": 0.0002,
+      "loss": 1.0882,
+      "step": 8280
+    },
+    {
+      "epoch": 6.193500186776242,
+      "grad_norm": 1.2549997568130493,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 8290
+    },
+    {
+      "epoch": 6.200971236458723,
+      "grad_norm": 1.3629751205444336,
+      "learning_rate": 0.0002,
+      "loss": 1.0626,
+      "step": 8300
+    },
+    {
+      "epoch": 6.208442286141203,
+      "grad_norm": 1.1029163599014282,
+      "learning_rate": 0.0002,
+      "loss": 1.1343,
+      "step": 8310
+    },
+    {
+      "epoch": 6.215913335823683,
+      "grad_norm": 1.1992450952529907,
+      "learning_rate": 0.0002,
+      "loss": 1.0895,
+      "step": 8320
+    },
+    {
+      "epoch": 6.223384385506163,
+      "grad_norm": 1.3317986726760864,
+      "learning_rate": 0.0002,
+      "loss": 1.1417,
+      "step": 8330
+    },
+    {
+      "epoch": 6.230855435188644,
+      "grad_norm": 1.0538336038589478,
+      "learning_rate": 0.0002,
+      "loss": 1.0958,
+      "step": 8340
+    },
+    {
+      "epoch": 6.238326484871124,
+      "grad_norm": 1.1767704486846924,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 8350
+    },
+    {
+      "epoch": 6.2457975345536045,
+      "grad_norm": 1.1213016510009766,
+      "learning_rate": 0.0002,
+      "loss": 1.1038,
+      "step": 8360
+    },
+    {
+      "epoch": 6.253268584236086,
+      "grad_norm": 1.1895716190338135,
+      "learning_rate": 0.0002,
+      "loss": 1.1241,
+      "step": 8370
+    },
+    {
+      "epoch": 6.260739633918566,
+      "grad_norm": 1.1078153848648071,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 8380
+    },
+    {
+      "epoch": 6.268210683601046,
+      "grad_norm": 1.1662801504135132,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 8390
+    },
+    {
+      "epoch": 6.275681733283526,
+      "grad_norm": 1.2071197032928467,
+      "learning_rate": 0.0002,
+      "loss": 1.125,
+      "step": 8400
+    },
+    {
+      "epoch": 6.283152782966007,
+      "grad_norm": 1.2653778791427612,
+      "learning_rate": 0.0002,
+      "loss": 1.0625,
+      "step": 8410
+    },
+    {
+      "epoch": 6.290623832648487,
+      "grad_norm": 1.6128872632980347,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 8420
+    },
+    {
+      "epoch": 6.298094882330967,
+      "grad_norm": 1.4993070363998413,
+      "learning_rate": 0.0002,
+      "loss": 1.1212,
+      "step": 8430
+    },
+    {
+      "epoch": 6.305565932013448,
+      "grad_norm": 1.16339910030365,
+      "learning_rate": 0.0002,
+      "loss": 1.1516,
+      "step": 8440
+    },
+    {
+      "epoch": 6.313036981695928,
+      "grad_norm": 1.256822943687439,
+      "learning_rate": 0.0002,
+      "loss": 1.0662,
+      "step": 8450
+    },
+    {
+      "epoch": 6.3205080313784086,
+      "grad_norm": 1.1352964639663696,
+      "learning_rate": 0.0002,
+      "loss": 1.1566,
+      "step": 8460
+    },
+    {
+      "epoch": 6.327979081060889,
+      "grad_norm": 1.0061070919036865,
+      "learning_rate": 0.0002,
+      "loss": 1.1297,
+      "step": 8470
+    },
+    {
+      "epoch": 6.33545013074337,
+      "grad_norm": 1.1901768445968628,
+      "learning_rate": 0.0002,
+      "loss": 1.0967,
+      "step": 8480
+    },
+    {
+      "epoch": 6.34292118042585,
+      "grad_norm": 1.2715139389038086,
+      "learning_rate": 0.0002,
+      "loss": 1.1463,
+      "step": 8490
+    },
+    {
+      "epoch": 6.35039223010833,
+      "grad_norm": 1.1583346128463745,
+      "learning_rate": 0.0002,
+      "loss": 1.2143,
+      "step": 8500
+    },
+    {
+      "epoch": 6.357863279790811,
+      "grad_norm": 1.1427477598190308,
+      "learning_rate": 0.0002,
+      "loss": 1.1072,
+      "step": 8510
+    },
+    {
+      "epoch": 6.365334329473291,
+      "grad_norm": 1.1952263116836548,
+      "learning_rate": 0.0002,
+      "loss": 1.1119,
+      "step": 8520
+    },
+    {
+      "epoch": 6.372805379155771,
+      "grad_norm": 1.0599623918533325,
+      "learning_rate": 0.0002,
+      "loss": 1.0797,
+      "step": 8530
+    },
+    {
+      "epoch": 6.380276428838251,
+      "grad_norm": 1.3511574268341064,
+      "learning_rate": 0.0002,
+      "loss": 1.1091,
+      "step": 8540
+    },
+    {
+      "epoch": 6.387747478520732,
+      "grad_norm": 1.171126127243042,
+      "learning_rate": 0.0002,
+      "loss": 1.1272,
+      "step": 8550
+    },
+    {
+      "epoch": 6.395218528203213,
+      "grad_norm": 1.285474419593811,
+      "learning_rate": 0.0002,
+      "loss": 1.1615,
+      "step": 8560
+    },
+    {
+      "epoch": 6.402689577885693,
+      "grad_norm": 0.9751279950141907,
+      "learning_rate": 0.0002,
+      "loss": 1.1505,
+      "step": 8570
+    },
+    {
+      "epoch": 6.410160627568174,
+      "grad_norm": 1.2194149494171143,
+      "learning_rate": 0.0002,
+      "loss": 1.1502,
+      "step": 8580
+    },
+    {
+      "epoch": 6.417631677250654,
+      "grad_norm": 1.255888819694519,
+      "learning_rate": 0.0002,
+      "loss": 1.138,
+      "step": 8590
+    },
+    {
+      "epoch": 6.425102726933134,
+      "grad_norm": 1.1636122465133667,
+      "learning_rate": 0.0002,
+      "loss": 1.1308,
+      "step": 8600
+    },
+    {
+      "epoch": 6.432573776615614,
+      "grad_norm": 1.0769859552383423,
+      "learning_rate": 0.0002,
+      "loss": 1.1398,
+      "step": 8610
+    },
+    {
+      "epoch": 6.440044826298095,
+      "grad_norm": 1.151778221130371,
+      "learning_rate": 0.0002,
+      "loss": 1.1183,
+      "step": 8620
+    },
+    {
+      "epoch": 6.447515875980575,
+      "grad_norm": 1.2749944925308228,
+      "learning_rate": 0.0002,
+      "loss": 1.0706,
+      "step": 8630
+    },
+    {
+      "epoch": 6.454986925663055,
+      "grad_norm": 1.1925828456878662,
+      "learning_rate": 0.0002,
+      "loss": 1.1011,
+      "step": 8640
+    },
+    {
+      "epoch": 6.4624579753455365,
+      "grad_norm": 1.166107416152954,
+      "learning_rate": 0.0002,
+      "loss": 1.1581,
+      "step": 8650
+    },
+    {
+      "epoch": 6.469929025028017,
+      "grad_norm": 1.0372248888015747,
+      "learning_rate": 0.0002,
+      "loss": 1.105,
+      "step": 8660
+    },
+    {
+      "epoch": 6.477400074710497,
+      "grad_norm": 1.26933753490448,
+      "learning_rate": 0.0002,
+      "loss": 1.1546,
+      "step": 8670
+    },
+    {
+      "epoch": 6.484871124392977,
+      "grad_norm": 1.2154223918914795,
+      "learning_rate": 0.0002,
+      "loss": 1.2362,
+      "step": 8680
+    },
+    {
+      "epoch": 6.492342174075458,
+      "grad_norm": 1.09475839138031,
+      "learning_rate": 0.0002,
+      "loss": 1.1096,
+      "step": 8690
+    },
+    {
+      "epoch": 6.499813223757938,
+      "grad_norm": 1.0763037204742432,
+      "learning_rate": 0.0002,
+      "loss": 1.1168,
+      "step": 8700
+    },
+    {
+      "epoch": 6.507284273440418,
+      "grad_norm": 1.1882896423339844,
+      "learning_rate": 0.0002,
+      "loss": 1.1993,
+      "step": 8710
+    },
+    {
+      "epoch": 6.514755323122898,
+      "grad_norm": 1.1662089824676514,
+      "learning_rate": 0.0002,
+      "loss": 1.1498,
+      "step": 8720
+    },
+    {
+      "epoch": 6.522226372805379,
+      "grad_norm": 1.3259495496749878,
+      "learning_rate": 0.0002,
+      "loss": 1.2008,
+      "step": 8730
+    },
+    {
+      "epoch": 6.5296974224878594,
+      "grad_norm": 1.0858017206192017,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 8740
+    },
+    {
+      "epoch": 6.53716847217034,
+      "grad_norm": 1.240337610244751,
+      "learning_rate": 0.0002,
+      "loss": 1.1335,
+      "step": 8750
+    },
+    {
+      "epoch": 6.544639521852821,
+      "grad_norm": 1.1381462812423706,
+      "learning_rate": 0.0002,
+      "loss": 1.1479,
+      "step": 8760
+    },
+    {
+      "epoch": 6.552110571535301,
+      "grad_norm": 1.2220063209533691,
+      "learning_rate": 0.0002,
+      "loss": 1.0991,
+      "step": 8770
+    },
+    {
+      "epoch": 6.559581621217781,
+      "grad_norm": 1.1553083658218384,
+      "learning_rate": 0.0002,
+      "loss": 1.159,
+      "step": 8780
+    },
+    {
+      "epoch": 6.567052670900262,
+      "grad_norm": 1.1383219957351685,
+      "learning_rate": 0.0002,
+      "loss": 1.0996,
+      "step": 8790
+    },
+    {
+      "epoch": 6.574523720582742,
+      "grad_norm": 1.0379676818847656,
+      "learning_rate": 0.0002,
+      "loss": 1.1355,
+      "step": 8800
+    },
+    {
+      "epoch": 6.581994770265222,
+      "grad_norm": 1.376488447189331,
+      "learning_rate": 0.0002,
+      "loss": 1.1704,
+      "step": 8810
+    },
+    {
+      "epoch": 6.589465819947702,
+      "grad_norm": 1.1586211919784546,
+      "learning_rate": 0.0002,
+      "loss": 1.1265,
+      "step": 8820
+    },
+    {
+      "epoch": 6.596936869630183,
+      "grad_norm": 1.28152334690094,
+      "learning_rate": 0.0002,
+      "loss": 1.1904,
+      "step": 8830
+    },
+    {
+      "epoch": 6.6044079193126635,
+      "grad_norm": 1.2656810283660889,
+      "learning_rate": 0.0002,
+      "loss": 1.1646,
+      "step": 8840
+    },
+    {
+      "epoch": 6.611878968995144,
+      "grad_norm": 1.0636502504348755,
+      "learning_rate": 0.0002,
+      "loss": 1.1865,
+      "step": 8850
+    },
+    {
+      "epoch": 6.619350018677624,
+      "grad_norm": 1.273239254951477,
+      "learning_rate": 0.0002,
+      "loss": 1.125,
+      "step": 8860
+    },
+    {
+      "epoch": 6.626821068360105,
+      "grad_norm": 1.1055482625961304,
+      "learning_rate": 0.0002,
+      "loss": 1.1443,
+      "step": 8870
+    },
+    {
+      "epoch": 6.634292118042585,
+      "grad_norm": 1.1934176683425903,
+      "learning_rate": 0.0002,
+      "loss": 1.0877,
+      "step": 8880
+    },
+    {
+      "epoch": 6.641763167725065,
+      "grad_norm": 1.2248114347457886,
+      "learning_rate": 0.0002,
+      "loss": 1.194,
+      "step": 8890
+    },
+    {
+      "epoch": 6.649234217407546,
+      "grad_norm": 1.1950982809066772,
+      "learning_rate": 0.0002,
+      "loss": 1.1609,
+      "step": 8900
+    },
+    {
+      "epoch": 6.656705267090026,
+      "grad_norm": 1.0821784734725952,
+      "learning_rate": 0.0002,
+      "loss": 1.169,
+      "step": 8910
+    },
+    {
+      "epoch": 6.664176316772506,
+      "grad_norm": 1.0062463283538818,
+      "learning_rate": 0.0002,
+      "loss": 1.1337,
+      "step": 8920
+    },
+    {
+      "epoch": 6.671647366454987,
+      "grad_norm": 1.2373089790344238,
+      "learning_rate": 0.0002,
+      "loss": 1.1403,
+      "step": 8930
+    },
+    {
+      "epoch": 6.6791184161374675,
+      "grad_norm": 1.1821746826171875,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 8940
+    },
+    {
+      "epoch": 6.686589465819948,
+      "grad_norm": 1.2350659370422363,
+      "learning_rate": 0.0002,
+      "loss": 1.1214,
+      "step": 8950
+    },
+    {
+      "epoch": 6.694060515502428,
+      "grad_norm": 1.1012883186340332,
+      "learning_rate": 0.0002,
+      "loss": 1.225,
+      "step": 8960
+    },
+    {
+      "epoch": 6.701531565184909,
+      "grad_norm": 1.2008943557739258,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 8970
+    },
+    {
+      "epoch": 6.709002614867389,
+      "grad_norm": 1.2355504035949707,
+      "learning_rate": 0.0002,
+      "loss": 1.1769,
+      "step": 8980
+    },
+    {
+      "epoch": 6.716473664549869,
+      "grad_norm": 1.2367502450942993,
+      "learning_rate": 0.0002,
+      "loss": 1.1323,
+      "step": 8990
+    },
+    {
+      "epoch": 6.723944714232349,
+      "grad_norm": 1.1075866222381592,
+      "learning_rate": 0.0002,
+      "loss": 1.1235,
+      "step": 9000
+    },
+    {
+      "epoch": 6.73141576391483,
+      "grad_norm": 1.246480941772461,
+      "learning_rate": 0.0002,
+      "loss": 1.1239,
+      "step": 9010
+    },
+    {
+      "epoch": 6.73888681359731,
+      "grad_norm": 1.1252824068069458,
+      "learning_rate": 0.0002,
+      "loss": 1.2154,
+      "step": 9020
+    },
+    {
+      "epoch": 6.7463578632797905,
+      "grad_norm": 1.0706887245178223,
+      "learning_rate": 0.0002,
+      "loss": 1.1762,
+      "step": 9030
+    },
+    {
+      "epoch": 6.7538289129622715,
+      "grad_norm": 1.0874755382537842,
+      "learning_rate": 0.0002,
+      "loss": 1.1961,
+      "step": 9040
+    },
+    {
+      "epoch": 6.761299962644752,
+      "grad_norm": 1.121434211730957,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 9050
+    },
+    {
+      "epoch": 6.768771012327232,
+      "grad_norm": 1.1517996788024902,
+      "learning_rate": 0.0002,
+      "loss": 1.2018,
+      "step": 9060
+    },
+    {
+      "epoch": 6.776242062009713,
+      "grad_norm": 1.2484540939331055,
+      "learning_rate": 0.0002,
+      "loss": 1.1593,
+      "step": 9070
+    },
+    {
+      "epoch": 6.783713111692193,
+      "grad_norm": 1.023059368133545,
+      "learning_rate": 0.0002,
+      "loss": 1.13,
+      "step": 9080
+    },
+    {
+      "epoch": 6.791184161374673,
+      "grad_norm": 1.1334631443023682,
+      "learning_rate": 0.0002,
+      "loss": 1.1929,
+      "step": 9090
+    },
+    {
+      "epoch": 6.798655211057153,
+      "grad_norm": 1.2991816997528076,
+      "learning_rate": 0.0002,
+      "loss": 1.18,
+      "step": 9100
+    },
+    {
+      "epoch": 6.806126260739634,
+      "grad_norm": 1.4147199392318726,
+      "learning_rate": 0.0002,
+      "loss": 1.2398,
+      "step": 9110
+    },
+    {
+      "epoch": 6.813597310422114,
+      "grad_norm": 1.1353832483291626,
+      "learning_rate": 0.0002,
+      "loss": 1.0958,
+      "step": 9120
+    },
+    {
+      "epoch": 6.8210683601045945,
+      "grad_norm": 1.0332539081573486,
+      "learning_rate": 0.0002,
+      "loss": 1.1379,
+      "step": 9130
+    },
+    {
+      "epoch": 6.828539409787075,
+      "grad_norm": 1.2208142280578613,
+      "learning_rate": 0.0002,
+      "loss": 1.1652,
+      "step": 9140
+    },
+    {
+      "epoch": 6.836010459469556,
+      "grad_norm": 1.3033398389816284,
+      "learning_rate": 0.0002,
+      "loss": 1.1463,
+      "step": 9150
+    },
+    {
+      "epoch": 6.843481509152036,
+      "grad_norm": 1.2676737308502197,
+      "learning_rate": 0.0002,
+      "loss": 1.1834,
+      "step": 9160
+    },
+    {
+      "epoch": 6.850952558834516,
+      "grad_norm": 1.1668603420257568,
+      "learning_rate": 0.0002,
+      "loss": 1.1786,
+      "step": 9170
+    },
+    {
+      "epoch": 6.858423608516997,
+      "grad_norm": 1.1994788646697998,
+      "learning_rate": 0.0002,
+      "loss": 1.1801,
+      "step": 9180
+    },
+    {
+      "epoch": 6.865894658199477,
+      "grad_norm": 1.231873869895935,
+      "learning_rate": 0.0002,
+      "loss": 1.2131,
+      "step": 9190
+    },
+    {
+      "epoch": 6.873365707881957,
+      "grad_norm": 0.9981484413146973,
+      "learning_rate": 0.0002,
+      "loss": 1.2109,
+      "step": 9200
+    },
+    {
+      "epoch": 6.880836757564438,
+      "grad_norm": 1.2799428701400757,
+      "learning_rate": 0.0002,
+      "loss": 1.1084,
+      "step": 9210
+    },
+    {
+      "epoch": 6.888307807246918,
+      "grad_norm": 1.2042057514190674,
+      "learning_rate": 0.0002,
+      "loss": 1.2004,
+      "step": 9220
+    },
+    {
+      "epoch": 6.8957788569293985,
+      "grad_norm": 1.070420265197754,
+      "learning_rate": 0.0002,
+      "loss": 1.1567,
+      "step": 9230
+    },
+    {
+      "epoch": 6.903249906611879,
+      "grad_norm": 1.327160358428955,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 9240
+    },
+    {
+      "epoch": 6.91072095629436,
+      "grad_norm": 1.1109007596969604,
+      "learning_rate": 0.0002,
+      "loss": 1.1945,
+      "step": 9250
+    },
+    {
+      "epoch": 6.91819200597684,
+      "grad_norm": 1.1669930219650269,
+      "learning_rate": 0.0002,
+      "loss": 1.1701,
+      "step": 9260
+    },
+    {
+      "epoch": 6.92566305565932,
+      "grad_norm": 1.034532904624939,
+      "learning_rate": 0.0002,
+      "loss": 1.1854,
+      "step": 9270
+    },
+    {
+      "epoch": 6.9331341053418,
+      "grad_norm": 1.1035540103912354,
+      "learning_rate": 0.0002,
+      "loss": 1.1712,
+      "step": 9280
+    },
+    {
+      "epoch": 6.940605155024281,
+      "grad_norm": 1.366254448890686,
+      "learning_rate": 0.0002,
+      "loss": 1.1767,
+      "step": 9290
+    },
+    {
+      "epoch": 6.948076204706761,
+      "grad_norm": 1.094214677810669,
+      "learning_rate": 0.0002,
+      "loss": 1.1591,
+      "step": 9300
+    },
+    {
+      "epoch": 6.955547254389241,
+      "grad_norm": 1.131238579750061,
+      "learning_rate": 0.0002,
+      "loss": 1.18,
+      "step": 9310
+    },
+    {
+      "epoch": 6.963018304071722,
+      "grad_norm": 1.202369213104248,
+      "learning_rate": 0.0002,
+      "loss": 1.2513,
+      "step": 9320
+    },
+    {
+      "epoch": 6.9704893537542025,
+      "grad_norm": 1.1067225933074951,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 9330
+    },
+    {
+      "epoch": 6.977960403436683,
+      "grad_norm": 1.0258643627166748,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 9340
+    },
+    {
+      "epoch": 6.985431453119164,
+      "grad_norm": 1.3311655521392822,
+      "learning_rate": 0.0002,
+      "loss": 1.2053,
+      "step": 9350
+    },
+    {
+      "epoch": 6.992902502801644,
+      "grad_norm": 1.1245559453964233,
+      "learning_rate": 0.0002,
+      "loss": 1.1778,
+      "step": 9360
+    },
+    {
+      "epoch": 6.999626447515876,
+      "eval_loss": 2.128103017807007,
+      "eval_runtime": 39.1339,
+      "eval_samples_per_second": 13.16,
+      "eval_steps_per_second": 1.661,
+      "step": 9369
+    },
+    {
+      "epoch": 7.000373552484124,
+      "grad_norm": 1.0868251323699951,
+      "learning_rate": 0.0002,
+      "loss": 1.1782,
+      "step": 9370
+    },
+    {
+      "epoch": 7.007844602166604,
+      "grad_norm": 1.5252128839492798,
+      "learning_rate": 0.0002,
+      "loss": 1.0133,
+      "step": 9380
+    },
+    {
+      "epoch": 7.015315651849085,
+      "grad_norm": 1.1230034828186035,
+      "learning_rate": 0.0002,
+      "loss": 0.9364,
+      "step": 9390
+    },
+    {
+      "epoch": 7.022786701531565,
+      "grad_norm": 1.275871992111206,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 9400
+    },
+    {
+      "epoch": 7.030257751214045,
+      "grad_norm": 1.462963342666626,
+      "learning_rate": 0.0002,
+      "loss": 0.9305,
+      "step": 9410
+    },
+    {
+      "epoch": 7.0377288008965255,
+      "grad_norm": 1.0506054162979126,
+      "learning_rate": 0.0002,
+      "loss": 0.9329,
+      "step": 9420
+    },
+    {
+      "epoch": 7.0451998505790066,
+      "grad_norm": 1.4315128326416016,
+      "learning_rate": 0.0002,
+      "loss": 0.9398,
+      "step": 9430
+    },
+    {
+      "epoch": 7.052670900261487,
+      "grad_norm": 1.5143473148345947,
+      "learning_rate": 0.0002,
+      "loss": 0.9086,
+      "step": 9440
+    },
+    {
+      "epoch": 7.060141949943967,
+      "grad_norm": 1.2537293434143066,
+      "learning_rate": 0.0002,
+      "loss": 0.9712,
+      "step": 9450
+    },
+    {
+      "epoch": 7.067612999626448,
+      "grad_norm": 1.36807382106781,
+      "learning_rate": 0.0002,
+      "loss": 0.9591,
+      "step": 9460
+    },
+    {
+      "epoch": 7.075084049308928,
+      "grad_norm": 1.5365028381347656,
+      "learning_rate": 0.0002,
+      "loss": 1.0046,
+      "step": 9470
+    },
+    {
+      "epoch": 7.082555098991408,
+      "grad_norm": 1.227250576019287,
+      "learning_rate": 0.0002,
+      "loss": 1.0045,
+      "step": 9480
+    },
+    {
+      "epoch": 7.090026148673888,
+      "grad_norm": 1.6941372156143188,
+      "learning_rate": 0.0002,
+      "loss": 0.9745,
+      "step": 9490
+    },
+    {
+      "epoch": 7.097497198356369,
+      "grad_norm": 1.587410569190979,
+      "learning_rate": 0.0002,
+      "loss": 0.9203,
+      "step": 9500
+    },
+    {
+      "epoch": 7.104968248038849,
+      "grad_norm": 1.481272578239441,
+      "learning_rate": 0.0002,
+      "loss": 0.9713,
+      "step": 9510
+    },
+    {
+      "epoch": 7.1124392977213295,
+      "grad_norm": 1.2331953048706055,
+      "learning_rate": 0.0002,
+      "loss": 1.0066,
+      "step": 9520
+    },
+    {
+      "epoch": 7.119910347403811,
+      "grad_norm": 1.6446775197982788,
+      "learning_rate": 0.0002,
+      "loss": 0.9422,
+      "step": 9530
+    },
+    {
+      "epoch": 7.127381397086291,
+      "grad_norm": 1.2055929899215698,
+      "learning_rate": 0.0002,
+      "loss": 0.901,
+      "step": 9540
+    },
+    {
+      "epoch": 7.134852446768771,
+      "grad_norm": 1.119033932685852,
+      "learning_rate": 0.0002,
+      "loss": 0.8959,
+      "step": 9550
+    },
+    {
+      "epoch": 7.142323496451251,
+      "grad_norm": 1.712833046913147,
+      "learning_rate": 0.0002,
+      "loss": 0.9586,
+      "step": 9560
+    },
+    {
+      "epoch": 7.149794546133732,
+      "grad_norm": 1.2007980346679688,
+      "learning_rate": 0.0002,
+      "loss": 0.9326,
+      "step": 9570
+    },
+    {
+      "epoch": 7.157265595816212,
+      "grad_norm": 1.3251731395721436,
+      "learning_rate": 0.0002,
+      "loss": 0.9926,
+      "step": 9580
+    },
+    {
+      "epoch": 7.164736645498692,
+      "grad_norm": 1.1897934675216675,
+      "learning_rate": 0.0002,
+      "loss": 0.973,
+      "step": 9590
+    },
+    {
+      "epoch": 7.172207695181173,
+      "grad_norm": 1.348583698272705,
+      "learning_rate": 0.0002,
+      "loss": 0.9401,
+      "step": 9600
+    },
+    {
+      "epoch": 7.179678744863653,
+      "grad_norm": 1.1588937044143677,
+      "learning_rate": 0.0002,
+      "loss": 0.9931,
+      "step": 9610
+    },
+    {
+      "epoch": 7.187149794546134,
+      "grad_norm": 1.3808276653289795,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 9620
+    },
+    {
+      "epoch": 7.194620844228614,
+      "grad_norm": 1.552425503730774,
+      "learning_rate": 0.0002,
+      "loss": 0.9955,
+      "step": 9630
+    },
+    {
+      "epoch": 7.202091893911095,
+      "grad_norm": 1.3649828433990479,
+      "learning_rate": 0.0002,
+      "loss": 1.0255,
+      "step": 9640
+    },
+    {
+      "epoch": 7.209562943593575,
+      "grad_norm": 1.3196533918380737,
+      "learning_rate": 0.0002,
+      "loss": 0.973,
+      "step": 9650
+    },
+    {
+      "epoch": 7.217033993276055,
+      "grad_norm": 1.4874017238616943,
+      "learning_rate": 0.0002,
+      "loss": 1.0119,
+      "step": 9660
+    },
+    {
+      "epoch": 7.224505042958536,
+      "grad_norm": 1.2448325157165527,
+      "learning_rate": 0.0002,
+      "loss": 0.9778,
+      "step": 9670
+    },
+    {
+      "epoch": 7.231976092641016,
+      "grad_norm": 1.4631818532943726,
+      "learning_rate": 0.0002,
+      "loss": 0.9697,
+      "step": 9680
+    },
+    {
+      "epoch": 7.239447142323496,
+      "grad_norm": 1.2041361331939697,
+      "learning_rate": 0.0002,
+      "loss": 0.9827,
+      "step": 9690
+    },
+    {
+      "epoch": 7.246918192005976,
+      "grad_norm": 1.559156060218811,
+      "learning_rate": 0.0002,
+      "loss": 0.9417,
+      "step": 9700
+    },
+    {
+      "epoch": 7.2543892416884574,
+      "grad_norm": 1.3939464092254639,
+      "learning_rate": 0.0002,
+      "loss": 1.0232,
+      "step": 9710
+    },
+    {
+      "epoch": 7.261860291370938,
+      "grad_norm": 1.347583293914795,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 9720
+    },
+    {
+      "epoch": 7.269331341053418,
+      "grad_norm": 1.302850365638733,
+      "learning_rate": 0.0002,
+      "loss": 0.9256,
+      "step": 9730
+    },
+    {
+      "epoch": 7.276802390735899,
+      "grad_norm": 1.1425062417984009,
+      "learning_rate": 0.0002,
+      "loss": 0.956,
+      "step": 9740
+    },
+    {
+      "epoch": 7.284273440418379,
+      "grad_norm": 1.2865869998931885,
+      "learning_rate": 0.0002,
+      "loss": 0.9978,
+      "step": 9750
+    },
+    {
+      "epoch": 7.291744490100859,
+      "grad_norm": 1.3773187398910522,
+      "learning_rate": 0.0002,
+      "loss": 0.9841,
+      "step": 9760
+    },
+    {
+      "epoch": 7.299215539783339,
+      "grad_norm": 1.2692701816558838,
+      "learning_rate": 0.0002,
+      "loss": 1.0063,
+      "step": 9770
+    },
+    {
+      "epoch": 7.30668658946582,
+      "grad_norm": 1.38542902469635,
+      "learning_rate": 0.0002,
+      "loss": 1.0347,
+      "step": 9780
+    },
+    {
+      "epoch": 7.3141576391483,
+      "grad_norm": 1.2204844951629639,
+      "learning_rate": 0.0002,
+      "loss": 0.9606,
+      "step": 9790
+    },
+    {
+      "epoch": 7.32162868883078,
+      "grad_norm": 1.4863795042037964,
+      "learning_rate": 0.0002,
+      "loss": 1.0225,
+      "step": 9800
+    },
+    {
+      "epoch": 7.3290997385132615,
+      "grad_norm": 1.2458586692810059,
+      "learning_rate": 0.0002,
+      "loss": 0.9933,
+      "step": 9810
+    },
+    {
+      "epoch": 7.336570788195742,
+      "grad_norm": 1.3530622720718384,
+      "learning_rate": 0.0002,
+      "loss": 1.0336,
+      "step": 9820
+    },
+    {
+      "epoch": 7.344041837878222,
+      "grad_norm": 1.2571991682052612,
+      "learning_rate": 0.0002,
+      "loss": 0.9319,
+      "step": 9830
+    },
+    {
+      "epoch": 7.351512887560702,
+      "grad_norm": 1.3074439764022827,
+      "learning_rate": 0.0002,
+      "loss": 1.0042,
+      "step": 9840
+    },
+    {
+      "epoch": 7.358983937243183,
+      "grad_norm": 1.2986950874328613,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 9850
+    },
+    {
+      "epoch": 7.366454986925663,
+      "grad_norm": 1.4233403205871582,
+      "learning_rate": 0.0002,
+      "loss": 1.0078,
+      "step": 9860
+    },
+    {
+      "epoch": 7.373926036608143,
+      "grad_norm": 1.468161702156067,
+      "learning_rate": 0.0002,
+      "loss": 0.9359,
+      "step": 9870
+    },
+    {
+      "epoch": 7.381397086290624,
+      "grad_norm": 1.354690432548523,
+      "learning_rate": 0.0002,
+      "loss": 1.1074,
+      "step": 9880
+    },
+    {
+      "epoch": 7.388868135973104,
+      "grad_norm": 1.4891324043273926,
+      "learning_rate": 0.0002,
+      "loss": 1.0153,
+      "step": 9890
+    },
+    {
+      "epoch": 7.3963391856555845,
+      "grad_norm": 1.3470090627670288,
+      "learning_rate": 0.0002,
+      "loss": 1.0234,
+      "step": 9900
+    },
+    {
+      "epoch": 7.403810235338065,
+      "grad_norm": 1.373061180114746,
+      "learning_rate": 0.0002,
+      "loss": 1.063,
+      "step": 9910
+    },
+    {
+      "epoch": 7.411281285020546,
+      "grad_norm": 1.4181641340255737,
+      "learning_rate": 0.0002,
+      "loss": 1.0109,
+      "step": 9920
+    },
+    {
+      "epoch": 7.418752334703026,
+      "grad_norm": 1.3284671306610107,
+      "learning_rate": 0.0002,
+      "loss": 0.9801,
+      "step": 9930
+    },
+    {
+      "epoch": 7.426223384385506,
+      "grad_norm": 1.333896517753601,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 9940
+    },
+    {
+      "epoch": 7.433694434067987,
+      "grad_norm": 1.6348158121109009,
+      "learning_rate": 0.0002,
+      "loss": 1.0056,
+      "step": 9950
+    },
+    {
+      "epoch": 7.441165483750467,
+      "grad_norm": 1.364643931388855,
+      "learning_rate": 0.0002,
+      "loss": 1.0173,
+      "step": 9960
+    },
+    {
+      "epoch": 7.448636533432947,
+      "grad_norm": 1.3974874019622803,
+      "learning_rate": 0.0002,
+      "loss": 1.0076,
+      "step": 9970
+    },
+    {
+      "epoch": 7.456107583115427,
+      "grad_norm": 1.5207233428955078,
+      "learning_rate": 0.0002,
+      "loss": 0.9918,
+      "step": 9980
+    },
+    {
+      "epoch": 7.463578632797908,
+      "grad_norm": 1.541517734527588,
+      "learning_rate": 0.0002,
+      "loss": 1.019,
+      "step": 9990
+    },
+    {
+      "epoch": 7.4710496824803885,
+      "grad_norm": 1.3563939332962036,
+      "learning_rate": 0.0002,
+      "loss": 0.9904,
+      "step": 10000
+    },
+    {
+      "epoch": 7.478520732162869,
+      "grad_norm": 1.3443987369537354,
+      "learning_rate": 0.0002,
+      "loss": 1.0285,
+      "step": 10010
+    },
+    {
+      "epoch": 7.48599178184535,
+      "grad_norm": 1.2904508113861084,
+      "learning_rate": 0.0002,
+      "loss": 1.0028,
+      "step": 10020
+    },
+    {
+      "epoch": 7.49346283152783,
+      "grad_norm": 1.434145450592041,
+      "learning_rate": 0.0002,
+      "loss": 0.9949,
+      "step": 10030
+    },
+    {
+      "epoch": 7.50093388121031,
+      "grad_norm": 1.4659384489059448,
+      "learning_rate": 0.0002,
+      "loss": 1.0837,
+      "step": 10040
+    },
+    {
+      "epoch": 7.508404930892791,
+      "grad_norm": 1.3430006504058838,
+      "learning_rate": 0.0002,
+      "loss": 1.0063,
+      "step": 10050
+    },
+    {
+      "epoch": 7.515875980575271,
+      "grad_norm": 1.3595343828201294,
+      "learning_rate": 0.0002,
+      "loss": 1.0168,
+      "step": 10060
+    },
+    {
+      "epoch": 7.523347030257751,
+      "grad_norm": 1.7456434965133667,
+      "learning_rate": 0.0002,
+      "loss": 0.9609,
+      "step": 10070
+    },
+    {
+      "epoch": 7.530818079940231,
+      "grad_norm": 1.329853892326355,
+      "learning_rate": 0.0002,
+      "loss": 1.0843,
+      "step": 10080
+    },
+    {
+      "epoch": 7.538289129622712,
+      "grad_norm": 1.548466682434082,
+      "learning_rate": 0.0002,
+      "loss": 1.0879,
+      "step": 10090
+    },
+    {
+      "epoch": 7.5457601793051925,
+      "grad_norm": 1.2951644659042358,
+      "learning_rate": 0.0002,
+      "loss": 1.0221,
+      "step": 10100
+    },
+    {
+      "epoch": 7.553231228987673,
+      "grad_norm": 1.3988010883331299,
+      "learning_rate": 0.0002,
+      "loss": 1.007,
+      "step": 10110
+    },
+    {
+      "epoch": 7.560702278670153,
+      "grad_norm": 1.211068868637085,
+      "learning_rate": 0.0002,
+      "loss": 1.0353,
+      "step": 10120
+    },
+    {
+      "epoch": 7.568173328352634,
+      "grad_norm": 1.2159098386764526,
+      "learning_rate": 0.0002,
+      "loss": 0.9962,
+      "step": 10130
+    },
+    {
+      "epoch": 7.575644378035114,
+      "grad_norm": 1.3533744812011719,
+      "learning_rate": 0.0002,
+      "loss": 0.9928,
+      "step": 10140
+    },
+    {
+      "epoch": 7.583115427717594,
+      "grad_norm": 1.3153362274169922,
+      "learning_rate": 0.0002,
+      "loss": 1.0386,
+      "step": 10150
+    },
+    {
+      "epoch": 7.590586477400075,
+      "grad_norm": 1.535762906074524,
+      "learning_rate": 0.0002,
+      "loss": 1.0017,
+      "step": 10160
+    },
+    {
+      "epoch": 7.598057527082555,
+      "grad_norm": 1.5531504154205322,
+      "learning_rate": 0.0002,
+      "loss": 1.0592,
+      "step": 10170
+    },
+    {
+      "epoch": 7.605528576765035,
+      "grad_norm": 1.3588606119155884,
+      "learning_rate": 0.0002,
+      "loss": 1.0684,
+      "step": 10180
+    },
+    {
+      "epoch": 7.6129996264475155,
+      "grad_norm": 1.6648331880569458,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 10190
+    },
+    {
+      "epoch": 7.6204706761299965,
+      "grad_norm": 1.250205159187317,
+      "learning_rate": 0.0002,
+      "loss": 1.04,
+      "step": 10200
+    },
+    {
+      "epoch": 7.627941725812477,
+      "grad_norm": 1.2442443370819092,
+      "learning_rate": 0.0002,
+      "loss": 1.0188,
+      "step": 10210
+    },
+    {
+      "epoch": 7.635412775494957,
+      "grad_norm": 1.386197805404663,
+      "learning_rate": 0.0002,
+      "loss": 1.0297,
+      "step": 10220
+    },
+    {
+      "epoch": 7.642883825177438,
+      "grad_norm": 1.3478381633758545,
+      "learning_rate": 0.0002,
+      "loss": 1.0233,
+      "step": 10230
+    },
+    {
+      "epoch": 7.650354874859918,
+      "grad_norm": 1.2800627946853638,
+      "learning_rate": 0.0002,
+      "loss": 1.0313,
+      "step": 10240
+    },
+    {
+      "epoch": 7.657825924542398,
+      "grad_norm": 1.4082499742507935,
+      "learning_rate": 0.0002,
+      "loss": 1.0252,
+      "step": 10250
+    },
+    {
+      "epoch": 7.665296974224878,
+      "grad_norm": 1.321529746055603,
+      "learning_rate": 0.0002,
+      "loss": 1.0462,
+      "step": 10260
+    },
+    {
+      "epoch": 7.672768023907359,
+      "grad_norm": 1.4213372468948364,
+      "learning_rate": 0.0002,
+      "loss": 1.0727,
+      "step": 10270
+    },
+    {
+      "epoch": 7.680239073589839,
+      "grad_norm": 1.5585565567016602,
+      "learning_rate": 0.0002,
+      "loss": 0.9938,
+      "step": 10280
+    },
+    {
+      "epoch": 7.6877101232723195,
+      "grad_norm": 1.4025108814239502,
+      "learning_rate": 0.0002,
+      "loss": 1.021,
+      "step": 10290
+    },
+    {
+      "epoch": 7.6951811729548005,
+      "grad_norm": 1.344456434249878,
+      "learning_rate": 0.0002,
+      "loss": 1.0136,
+      "step": 10300
+    },
+    {
+      "epoch": 7.702652222637281,
+      "grad_norm": 1.4962990283966064,
+      "learning_rate": 0.0002,
+      "loss": 1.0157,
+      "step": 10310
+    },
+    {
+      "epoch": 7.710123272319761,
+      "grad_norm": 1.4523862600326538,
+      "learning_rate": 0.0002,
+      "loss": 1.0621,
+      "step": 10320
+    },
+    {
+      "epoch": 7.717594322002241,
+      "grad_norm": 1.401842474937439,
+      "learning_rate": 0.0002,
+      "loss": 1.0413,
+      "step": 10330
+    },
+    {
+      "epoch": 7.725065371684722,
+      "grad_norm": 1.517730474472046,
+      "learning_rate": 0.0002,
+      "loss": 1.0028,
+      "step": 10340
+    },
+    {
+      "epoch": 7.732536421367202,
+      "grad_norm": 1.3876111507415771,
+      "learning_rate": 0.0002,
+      "loss": 1.0061,
+      "step": 10350
+    },
+    {
+      "epoch": 7.740007471049682,
+      "grad_norm": 1.5741353034973145,
+      "learning_rate": 0.0002,
+      "loss": 1.0071,
+      "step": 10360
+    },
+    {
+      "epoch": 7.747478520732163,
+      "grad_norm": 1.3465591669082642,
+      "learning_rate": 0.0002,
+      "loss": 1.0472,
+      "step": 10370
+    },
+    {
+      "epoch": 7.754949570414643,
+      "grad_norm": 1.3611412048339844,
+      "learning_rate": 0.0002,
+      "loss": 0.9961,
+      "step": 10380
+    },
+    {
+      "epoch": 7.7624206200971235,
+      "grad_norm": 1.693565011024475,
+      "learning_rate": 0.0002,
+      "loss": 1.0118,
+      "step": 10390
+    },
+    {
+      "epoch": 7.769891669779604,
+      "grad_norm": 1.4654128551483154,
+      "learning_rate": 0.0002,
+      "loss": 1.0981,
+      "step": 10400
+    },
+    {
+      "epoch": 7.777362719462085,
+      "grad_norm": 1.417768955230713,
+      "learning_rate": 0.0002,
+      "loss": 1.029,
+      "step": 10410
+    },
+    {
+      "epoch": 7.784833769144565,
+      "grad_norm": 1.3143322467803955,
+      "learning_rate": 0.0002,
+      "loss": 1.0218,
+      "step": 10420
+    },
+    {
+      "epoch": 7.792304818827045,
+      "grad_norm": 1.3467497825622559,
+      "learning_rate": 0.0002,
+      "loss": 1.0224,
+      "step": 10430
+    },
+    {
+      "epoch": 7.799775868509526,
+      "grad_norm": 1.223697543144226,
+      "learning_rate": 0.0002,
+      "loss": 1.0555,
+      "step": 10440
+    },
+    {
+      "epoch": 7.807246918192006,
+      "grad_norm": 1.3060917854309082,
+      "learning_rate": 0.0002,
+      "loss": 1.0198,
+      "step": 10450
+    },
+    {
+      "epoch": 7.814717967874486,
+      "grad_norm": 1.5561134815216064,
+      "learning_rate": 0.0002,
+      "loss": 1.0896,
+      "step": 10460
+    },
+    {
+      "epoch": 7.822189017556966,
+      "grad_norm": 1.2789647579193115,
+      "learning_rate": 0.0002,
+      "loss": 1.0981,
+      "step": 10470
+    },
+    {
+      "epoch": 7.829660067239447,
+      "grad_norm": 1.2422796487808228,
+      "learning_rate": 0.0002,
+      "loss": 1.0549,
+      "step": 10480
+    },
+    {
+      "epoch": 7.8371311169219275,
+      "grad_norm": 1.377565622329712,
+      "learning_rate": 0.0002,
+      "loss": 1.0255,
+      "step": 10490
+    },
+    {
+      "epoch": 7.844602166604408,
+      "grad_norm": 1.2221037149429321,
+      "learning_rate": 0.0002,
+      "loss": 1.0864,
+      "step": 10500
+    },
+    {
+      "epoch": 7.852073216286888,
+      "grad_norm": 1.3779186010360718,
+      "learning_rate": 0.0002,
+      "loss": 1.0944,
+      "step": 10510
+    },
+    {
+      "epoch": 7.859544265969369,
+      "grad_norm": 1.3062539100646973,
+      "learning_rate": 0.0002,
+      "loss": 1.0694,
+      "step": 10520
+    },
+    {
+      "epoch": 7.867015315651849,
+      "grad_norm": 1.4066052436828613,
+      "learning_rate": 0.0002,
+      "loss": 1.048,
+      "step": 10530
+    },
+    {
+      "epoch": 7.874486365334329,
+      "grad_norm": 1.6326613426208496,
+      "learning_rate": 0.0002,
+      "loss": 1.0674,
+      "step": 10540
+    },
+    {
+      "epoch": 7.88195741501681,
+      "grad_norm": 1.1732137203216553,
+      "learning_rate": 0.0002,
+      "loss": 1.1237,
+      "step": 10550
+    },
+    {
+      "epoch": 7.88942846469929,
+      "grad_norm": 1.303125023841858,
+      "learning_rate": 0.0002,
+      "loss": 1.0682,
+      "step": 10560
+    },
+    {
+      "epoch": 7.89689951438177,
+      "grad_norm": 1.294990062713623,
+      "learning_rate": 0.0002,
+      "loss": 1.0524,
+      "step": 10570
+    },
+    {
+      "epoch": 7.904370564064251,
+      "grad_norm": 1.4719983339309692,
+      "learning_rate": 0.0002,
+      "loss": 1.0577,
+      "step": 10580
+    },
+    {
+      "epoch": 7.911841613746732,
+      "grad_norm": 1.4117742776870728,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 10590
+    },
+    {
+      "epoch": 7.919312663429212,
+      "grad_norm": 1.384812355041504,
+      "learning_rate": 0.0002,
+      "loss": 1.1129,
+      "step": 10600
+    },
+    {
+      "epoch": 7.926783713111692,
+      "grad_norm": 1.5743740797042847,
+      "learning_rate": 0.0002,
+      "loss": 0.9994,
+      "step": 10610
+    },
+    {
+      "epoch": 7.934254762794173,
+      "grad_norm": 1.2799863815307617,
+      "learning_rate": 0.0002,
+      "loss": 1.0168,
+      "step": 10620
+    },
+    {
+      "epoch": 7.941725812476653,
+      "grad_norm": 1.4822591543197632,
+      "learning_rate": 0.0002,
+      "loss": 1.0893,
+      "step": 10630
+    },
+    {
+      "epoch": 7.949196862159133,
+      "grad_norm": 1.4634777307510376,
+      "learning_rate": 0.0002,
+      "loss": 1.0362,
+      "step": 10640
+    },
+    {
+      "epoch": 7.956667911841613,
+      "grad_norm": 1.5230964422225952,
+      "learning_rate": 0.0002,
+      "loss": 1.0116,
+      "step": 10650
+    },
+    {
+      "epoch": 7.964138961524094,
+      "grad_norm": 1.3622701168060303,
+      "learning_rate": 0.0002,
+      "loss": 1.0582,
+      "step": 10660
+    },
+    {
+      "epoch": 7.971610011206574,
+      "grad_norm": 1.2133928537368774,
+      "learning_rate": 0.0002,
+      "loss": 1.1127,
+      "step": 10670
+    },
+    {
+      "epoch": 7.9790810608890546,
+      "grad_norm": 1.2852206230163574,
+      "learning_rate": 0.0002,
+      "loss": 1.096,
+      "step": 10680
+    },
+    {
+      "epoch": 7.986552110571536,
+      "grad_norm": 1.243310570716858,
+      "learning_rate": 0.0002,
+      "loss": 1.0286,
+      "step": 10690
+    },
+    {
+      "epoch": 7.994023160254016,
+      "grad_norm": 1.459757924079895,
+      "learning_rate": 0.0002,
+      "loss": 1.053,
+      "step": 10700
+    },
+    {
+      "epoch": 7.997011580127008,
+      "eval_loss": 2.2722349166870117,
+      "eval_runtime": 38.9945,
+      "eval_samples_per_second": 13.207,
+      "eval_steps_per_second": 1.667,
+      "step": 10704
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10704,
+  "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": 4.9535715277327565e+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-0.9-num-6326-sd-42/checkpoint-10704/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-0.9-num-6326-sd-42/checkpoint-10704/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..67c7b4ca126d7b712ad1985ef15ffb29ebe76633
--- /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-0.9-num-6326-sd-42/checkpoint-10704/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fc87da605e94ff0ecd8f5b371302a2d5f8727b77984707a2185ddb447fc3796
+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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6faa882079d4ad3a446600d712cd5c40781e2203
--- /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-0.9-num-6326-sd-42/checkpoint-1338/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ea4fc35757651c7c3a6bde3cbe3a29f159b4d024130edde539538989c54f675
+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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..95d92dee815ea20b248982d95f547f4cda7984a8
--- /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-0.9-num-6326-sd-42/checkpoint-1338/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4d3550f9d0c229652556a63f6a1db40a586e6c5c4c6a62b3dda5721a65c6553
+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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..2d9a9abb6f674fd4d3a5911d973b34ab7cb5369a
--- /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-0.9-num-6326-sd-42/checkpoint-1338/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4da1ccefac00163faaeff70f727c5f6340a85466e9e936f079a69b938730bb96
+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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5ec9a23aaf79117a31b48df9967f21e29cdce898
--- /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-0.9-num-6326-sd-42/checkpoint-1338/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:392d8ed6255678b167709c169ac1b7830ba060af79331ac91cf29c25699c5e20
+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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..95d5b113fb7382447effb87ae81d204039519304
--- /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-0.9-num-6326-sd-42/checkpoint-1338/trainer_state.json
@@ -0,0 +1,972 @@
+{
+  "best_metric": 1.8051470518112183,
+  "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-0.9-num-6326-sd-42/checkpoint-1338",
+  "epoch": 0.9996264475158759,
+  "eval_steps": 10,
+  "global_step": 1338,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007471049682480389,
+      "grad_norm": 0.4912872612476349,
+      "learning_rate": 0.0002,
+      "loss": 2.6181,
+      "step": 10
+    },
+    {
+      "epoch": 0.014942099364960777,
+      "grad_norm": 0.4856316149234772,
+      "learning_rate": 0.0002,
+      "loss": 2.2606,
+      "step": 20
+    },
+    {
+      "epoch": 0.022413149047441166,
+      "grad_norm": 0.47683125734329224,
+      "learning_rate": 0.0002,
+      "loss": 2.0957,
+      "step": 30
+    },
+    {
+      "epoch": 0.029884198729921554,
+      "grad_norm": 0.515082597732544,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 40
+    },
+    {
+      "epoch": 0.03735524841240194,
+      "grad_norm": 0.5299215316772461,
+      "learning_rate": 0.0002,
+      "loss": 1.9704,
+      "step": 50
+    },
+    {
+      "epoch": 0.04482629809488233,
+      "grad_norm": 0.4951399862766266,
+      "learning_rate": 0.0002,
+      "loss": 1.9225,
+      "step": 60
+    },
+    {
+      "epoch": 0.05229734777736272,
+      "grad_norm": 0.48079821467399597,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05976839745984311,
+      "grad_norm": 0.49402132630348206,
+      "learning_rate": 0.0002,
+      "loss": 1.9466,
+      "step": 80
+    },
+    {
+      "epoch": 0.0672394471423235,
+      "grad_norm": 0.4778193235397339,
+      "learning_rate": 0.0002,
+      "loss": 1.8691,
+      "step": 90
+    },
+    {
+      "epoch": 0.07471049682480388,
+      "grad_norm": 0.42472657561302185,
+      "learning_rate": 0.0002,
+      "loss": 1.8455,
+      "step": 100
+    },
+    {
+      "epoch": 0.08218154650728428,
+      "grad_norm": 0.4433092474937439,
+      "learning_rate": 0.0002,
+      "loss": 1.8744,
+      "step": 110
+    },
+    {
+      "epoch": 0.08965259618976466,
+      "grad_norm": 0.4472862780094147,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 120
+    },
+    {
+      "epoch": 0.09712364587224505,
+      "grad_norm": 0.42596298456192017,
+      "learning_rate": 0.0002,
+      "loss": 1.9256,
+      "step": 130
+    },
+    {
+      "epoch": 0.10459469555472543,
+      "grad_norm": 0.46645811200141907,
+      "learning_rate": 0.0002,
+      "loss": 1.8015,
+      "step": 140
+    },
+    {
+      "epoch": 0.11206574523720583,
+      "grad_norm": 0.41041234135627747,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 150
+    },
+    {
+      "epoch": 0.11953679491968622,
+      "grad_norm": 0.5329819917678833,
+      "learning_rate": 0.0002,
+      "loss": 1.8276,
+      "step": 160
+    },
+    {
+      "epoch": 0.1270078446021666,
+      "grad_norm": 0.4065922200679779,
+      "learning_rate": 0.0002,
+      "loss": 1.8118,
+      "step": 170
+    },
+    {
+      "epoch": 0.134478894284647,
+      "grad_norm": 0.38406994938850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8559,
+      "step": 180
+    },
+    {
+      "epoch": 0.14194994396712737,
+      "grad_norm": 0.4246881306171417,
+      "learning_rate": 0.0002,
+      "loss": 1.8647,
+      "step": 190
+    },
+    {
+      "epoch": 0.14942099364960776,
+      "grad_norm": 0.35136649012565613,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 200
+    },
+    {
+      "epoch": 0.15689204333208817,
+      "grad_norm": 0.43252742290496826,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.16436309301456856,
+      "grad_norm": 0.39236941933631897,
+      "learning_rate": 0.0002,
+      "loss": 1.7823,
+      "step": 220
+    },
+    {
+      "epoch": 0.17183414269704894,
+      "grad_norm": 0.3748249113559723,
+      "learning_rate": 0.0002,
+      "loss": 1.818,
+      "step": 230
+    },
+    {
+      "epoch": 0.17930519237952933,
+      "grad_norm": 0.6432855725288391,
+      "learning_rate": 0.0002,
+      "loss": 1.866,
+      "step": 240
+    },
+    {
+      "epoch": 0.1867762420620097,
+      "grad_norm": 0.34874802827835083,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 250
+    },
+    {
+      "epoch": 0.1942472917444901,
+      "grad_norm": 0.3721984326839447,
+      "learning_rate": 0.0002,
+      "loss": 1.79,
+      "step": 260
+    },
+    {
+      "epoch": 0.20171834142697048,
+      "grad_norm": 0.4339311420917511,
+      "learning_rate": 0.0002,
+      "loss": 1.8464,
+      "step": 270
+    },
+    {
+      "epoch": 0.20918939110945087,
+      "grad_norm": 0.4018215537071228,
+      "learning_rate": 0.0002,
+      "loss": 1.8665,
+      "step": 280
+    },
+    {
+      "epoch": 0.21666044079193125,
+      "grad_norm": 0.3278839886188507,
+      "learning_rate": 0.0002,
+      "loss": 1.8048,
+      "step": 290
+    },
+    {
+      "epoch": 0.22413149047441167,
+      "grad_norm": 0.36146077513694763,
+      "learning_rate": 0.0002,
+      "loss": 1.7395,
+      "step": 300
+    },
+    {
+      "epoch": 0.23160254015689205,
+      "grad_norm": 0.38175010681152344,
+      "learning_rate": 0.0002,
+      "loss": 1.7916,
+      "step": 310
+    },
+    {
+      "epoch": 0.23907358983937244,
+      "grad_norm": 0.44776618480682373,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 320
+    },
+    {
+      "epoch": 0.24654463952185282,
+      "grad_norm": 0.3933652937412262,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 330
+    },
+    {
+      "epoch": 0.2540156892043332,
+      "grad_norm": 0.3515005111694336,
+      "learning_rate": 0.0002,
+      "loss": 1.8393,
+      "step": 340
+    },
+    {
+      "epoch": 0.2614867388868136,
+      "grad_norm": 0.6683304309844971,
+      "learning_rate": 0.0002,
+      "loss": 1.8653,
+      "step": 350
+    },
+    {
+      "epoch": 0.268957788569294,
+      "grad_norm": 0.37093454599380493,
+      "learning_rate": 0.0002,
+      "loss": 1.8797,
+      "step": 360
+    },
+    {
+      "epoch": 0.2764288382517744,
+      "grad_norm": 0.3450651168823242,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 370
+    },
+    {
+      "epoch": 0.28389988793425475,
+      "grad_norm": 0.5140917301177979,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 380
+    },
+    {
+      "epoch": 0.29137093761673516,
+      "grad_norm": 0.32885563373565674,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 390
+    },
+    {
+      "epoch": 0.2988419872992155,
+      "grad_norm": 0.33962297439575195,
+      "learning_rate": 0.0002,
+      "loss": 1.8174,
+      "step": 400
+    },
+    {
+      "epoch": 0.30631303698169593,
+      "grad_norm": 0.3723141849040985,
+      "learning_rate": 0.0002,
+      "loss": 1.7467,
+      "step": 410
+    },
+    {
+      "epoch": 0.31378408666417634,
+      "grad_norm": 0.37173134088516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8459,
+      "step": 420
+    },
+    {
+      "epoch": 0.3212551363466567,
+      "grad_norm": 0.33736956119537354,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 430
+    },
+    {
+      "epoch": 0.3287261860291371,
+      "grad_norm": 0.3602448105812073,
+      "learning_rate": 0.0002,
+      "loss": 1.8367,
+      "step": 440
+    },
+    {
+      "epoch": 0.33619723571161747,
+      "grad_norm": 0.3569699227809906,
+      "learning_rate": 0.0002,
+      "loss": 1.8058,
+      "step": 450
+    },
+    {
+      "epoch": 0.3436682853940979,
+      "grad_norm": 0.31009167432785034,
+      "learning_rate": 0.0002,
+      "loss": 1.8086,
+      "step": 460
+    },
+    {
+      "epoch": 0.35113933507657824,
+      "grad_norm": 0.5278693437576294,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 470
+    },
+    {
+      "epoch": 0.35861038475905865,
+      "grad_norm": 0.3587537109851837,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 480
+    },
+    {
+      "epoch": 0.366081434441539,
+      "grad_norm": 0.3859670162200928,
+      "learning_rate": 0.0002,
+      "loss": 1.8046,
+      "step": 490
+    },
+    {
+      "epoch": 0.3735524841240194,
+      "grad_norm": 0.395913690328598,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 500
+    },
+    {
+      "epoch": 0.38102353380649984,
+      "grad_norm": 0.35052940249443054,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 510
+    },
+    {
+      "epoch": 0.3884945834889802,
+      "grad_norm": 0.2979494333267212,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 520
+    },
+    {
+      "epoch": 0.3959656331714606,
+      "grad_norm": 0.3062683343887329,
+      "learning_rate": 0.0002,
+      "loss": 1.8641,
+      "step": 530
+    },
+    {
+      "epoch": 0.40343668285394096,
+      "grad_norm": 0.3172847330570221,
+      "learning_rate": 0.0002,
+      "loss": 1.7651,
+      "step": 540
+    },
+    {
+      "epoch": 0.4109077325364214,
+      "grad_norm": 0.360435426235199,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 550
+    },
+    {
+      "epoch": 0.41837878221890173,
+      "grad_norm": 0.3427872359752655,
+      "learning_rate": 0.0002,
+      "loss": 1.9054,
+      "step": 560
+    },
+    {
+      "epoch": 0.42584983190138215,
+      "grad_norm": 0.34036558866500854,
+      "learning_rate": 0.0002,
+      "loss": 1.7562,
+      "step": 570
+    },
+    {
+      "epoch": 0.4333208815838625,
+      "grad_norm": 0.3365345299243927,
+      "learning_rate": 0.0002,
+      "loss": 1.7254,
+      "step": 580
+    },
+    {
+      "epoch": 0.4407919312663429,
+      "grad_norm": 0.35619041323661804,
+      "learning_rate": 0.0002,
+      "loss": 1.8328,
+      "step": 590
+    },
+    {
+      "epoch": 0.44826298094882333,
+      "grad_norm": 0.3569088280200958,
+      "learning_rate": 0.0002,
+      "loss": 1.8114,
+      "step": 600
+    },
+    {
+      "epoch": 0.4557340306313037,
+      "grad_norm": 0.3581278622150421,
+      "learning_rate": 0.0002,
+      "loss": 1.8599,
+      "step": 610
+    },
+    {
+      "epoch": 0.4632050803137841,
+      "grad_norm": 0.43197110295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.7078,
+      "step": 620
+    },
+    {
+      "epoch": 0.47067612999626446,
+      "grad_norm": 0.33966198563575745,
+      "learning_rate": 0.0002,
+      "loss": 1.8257,
+      "step": 630
+    },
+    {
+      "epoch": 0.47814717967874487,
+      "grad_norm": 0.3343866467475891,
+      "learning_rate": 0.0002,
+      "loss": 1.7528,
+      "step": 640
+    },
+    {
+      "epoch": 0.48561822936122523,
+      "grad_norm": 0.33878564834594727,
+      "learning_rate": 0.0002,
+      "loss": 1.8191,
+      "step": 650
+    },
+    {
+      "epoch": 0.49308927904370564,
+      "grad_norm": 0.387195885181427,
+      "learning_rate": 0.0002,
+      "loss": 1.8801,
+      "step": 660
+    },
+    {
+      "epoch": 0.500560328726186,
+      "grad_norm": 0.3755440413951874,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 670
+    },
+    {
+      "epoch": 0.5080313784086664,
+      "grad_norm": 0.3272816836833954,
+      "learning_rate": 0.0002,
+      "loss": 1.8057,
+      "step": 680
+    },
+    {
+      "epoch": 0.5155024280911468,
+      "grad_norm": 0.36063864827156067,
+      "learning_rate": 0.0002,
+      "loss": 1.8156,
+      "step": 690
+    },
+    {
+      "epoch": 0.5229734777736272,
+      "grad_norm": 0.35317373275756836,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 700
+    },
+    {
+      "epoch": 0.5304445274561076,
+      "grad_norm": 0.3561195433139801,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 710
+    },
+    {
+      "epoch": 0.537915577138588,
+      "grad_norm": 0.31124624609947205,
+      "learning_rate": 0.0002,
+      "loss": 1.8149,
+      "step": 720
+    },
+    {
+      "epoch": 0.5453866268210683,
+      "grad_norm": 0.3294544517993927,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 730
+    },
+    {
+      "epoch": 0.5528576765035488,
+      "grad_norm": 0.31933900713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.8027,
+      "step": 740
+    },
+    {
+      "epoch": 0.5603287261860291,
+      "grad_norm": 0.3226020634174347,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 750
+    },
+    {
+      "epoch": 0.5677997758685095,
+      "grad_norm": 0.3147525489330292,
+      "learning_rate": 0.0002,
+      "loss": 1.7862,
+      "step": 760
+    },
+    {
+      "epoch": 0.57527082555099,
+      "grad_norm": 0.32234328985214233,
+      "learning_rate": 0.0002,
+      "loss": 1.9028,
+      "step": 770
+    },
+    {
+      "epoch": 0.5827418752334703,
+      "grad_norm": 0.3258664309978485,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 780
+    },
+    {
+      "epoch": 0.5902129249159507,
+      "grad_norm": 0.3166961967945099,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 790
+    },
+    {
+      "epoch": 0.597683974598431,
+      "grad_norm": 0.35621458292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 800
+    },
+    {
+      "epoch": 0.6051550242809115,
+      "grad_norm": 0.3236999213695526,
+      "learning_rate": 0.0002,
+      "loss": 1.8313,
+      "step": 810
+    },
+    {
+      "epoch": 0.6126260739633919,
+      "grad_norm": 0.2892923653125763,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 820
+    },
+    {
+      "epoch": 0.6200971236458722,
+      "grad_norm": 0.4098321497440338,
+      "learning_rate": 0.0002,
+      "loss": 1.8709,
+      "step": 830
+    },
+    {
+      "epoch": 0.6275681733283527,
+      "grad_norm": 0.3337118923664093,
+      "learning_rate": 0.0002,
+      "loss": 1.7637,
+      "step": 840
+    },
+    {
+      "epoch": 0.635039223010833,
+      "grad_norm": 0.30416029691696167,
+      "learning_rate": 0.0002,
+      "loss": 1.7375,
+      "step": 850
+    },
+    {
+      "epoch": 0.6425102726933134,
+      "grad_norm": 0.3361026346683502,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 860
+    },
+    {
+      "epoch": 0.6499813223757938,
+      "grad_norm": 0.3537365198135376,
+      "learning_rate": 0.0002,
+      "loss": 1.732,
+      "step": 870
+    },
+    {
+      "epoch": 0.6574523720582742,
+      "grad_norm": 0.33854469656944275,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 880
+    },
+    {
+      "epoch": 0.6649234217407546,
+      "grad_norm": 0.3332272469997406,
+      "learning_rate": 0.0002,
+      "loss": 1.7561,
+      "step": 890
+    },
+    {
+      "epoch": 0.6723944714232349,
+      "grad_norm": 0.34954726696014404,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 900
+    },
+    {
+      "epoch": 0.6798655211057153,
+      "grad_norm": 0.2921750247478485,
+      "learning_rate": 0.0002,
+      "loss": 1.7917,
+      "step": 910
+    },
+    {
+      "epoch": 0.6873365707881958,
+      "grad_norm": 0.30508682131767273,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 920
+    },
+    {
+      "epoch": 0.6948076204706761,
+      "grad_norm": 0.32268425822257996,
+      "learning_rate": 0.0002,
+      "loss": 1.8082,
+      "step": 930
+    },
+    {
+      "epoch": 0.7022786701531565,
+      "grad_norm": 0.2844390869140625,
+      "learning_rate": 0.0002,
+      "loss": 1.8283,
+      "step": 940
+    },
+    {
+      "epoch": 0.709749719835637,
+      "grad_norm": 0.31263890862464905,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 950
+    },
+    {
+      "epoch": 0.7172207695181173,
+      "grad_norm": 0.3626808822154999,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 960
+    },
+    {
+      "epoch": 0.7246918192005977,
+      "grad_norm": 0.3322749733924866,
+      "learning_rate": 0.0002,
+      "loss": 1.853,
+      "step": 970
+    },
+    {
+      "epoch": 0.732162868883078,
+      "grad_norm": 0.29177871346473694,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 980
+    },
+    {
+      "epoch": 0.7396339185655585,
+      "grad_norm": 0.35405513644218445,
+      "learning_rate": 0.0002,
+      "loss": 1.8447,
+      "step": 990
+    },
+    {
+      "epoch": 0.7471049682480388,
+      "grad_norm": 0.39318400621414185,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7545760179305192,
+      "grad_norm": 0.29401418566703796,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7620470676129997,
+      "grad_norm": 0.3271748721599579,
+      "learning_rate": 0.0002,
+      "loss": 1.7649,
+      "step": 1020
+    },
+    {
+      "epoch": 0.76951811729548,
+      "grad_norm": 0.30883970856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7769891669779604,
+      "grad_norm": 0.3411838412284851,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7844602166604407,
+      "grad_norm": 0.30608129501342773,
+      "learning_rate": 0.0002,
+      "loss": 1.829,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7919312663429212,
+      "grad_norm": 0.30899080634117126,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7994023160254016,
+      "grad_norm": 0.3160453140735626,
+      "learning_rate": 0.0002,
+      "loss": 1.7625,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8068733657078819,
+      "grad_norm": 0.30947187542915344,
+      "learning_rate": 0.0002,
+      "loss": 1.8452,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8143444153903624,
+      "grad_norm": 0.3103134036064148,
+      "learning_rate": 0.0002,
+      "loss": 1.7418,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8218154650728428,
+      "grad_norm": 0.31771138310432434,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8292865147553231,
+      "grad_norm": 0.5860997438430786,
+      "learning_rate": 0.0002,
+      "loss": 1.7918,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8367575644378035,
+      "grad_norm": 0.3230148255825043,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8442286141202839,
+      "grad_norm": 0.29611510038375854,
+      "learning_rate": 0.0002,
+      "loss": 1.8478,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8516996638027643,
+      "grad_norm": 0.3373654782772064,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8591707134852447,
+      "grad_norm": 0.3474279046058655,
+      "learning_rate": 0.0002,
+      "loss": 1.7997,
+      "step": 1150
+    },
+    {
+      "epoch": 0.866641763167725,
+      "grad_norm": 0.35057875514030457,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8741128128502055,
+      "grad_norm": 0.39537495374679565,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8815838625326858,
+      "grad_norm": 0.3714233636856079,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8890549122151662,
+      "grad_norm": 0.2950296998023987,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8965259618976467,
+      "grad_norm": 0.38182979822158813,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 1200
+    },
+    {
+      "epoch": 0.903997011580127,
+      "grad_norm": 0.27883678674697876,
+      "learning_rate": 0.0002,
+      "loss": 1.827,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9114680612626074,
+      "grad_norm": 0.33874374628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9189391109450877,
+      "grad_norm": 0.3014272153377533,
+      "learning_rate": 0.0002,
+      "loss": 1.7334,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9264101606275682,
+      "grad_norm": 0.3194271922111511,
+      "learning_rate": 0.0002,
+      "loss": 1.8235,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9338812103100486,
+      "grad_norm": 0.3049403429031372,
+      "learning_rate": 0.0002,
+      "loss": 1.7924,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9413522599925289,
+      "grad_norm": 0.30621254444122314,
+      "learning_rate": 0.0002,
+      "loss": 1.7535,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9488233096750094,
+      "grad_norm": 0.28675132989883423,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9562943593574897,
+      "grad_norm": 0.3322032690048218,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9637654090399701,
+      "grad_norm": 0.35408294200897217,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9712364587224505,
+      "grad_norm": 0.36386919021606445,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9787075084049309,
+      "grad_norm": 0.32338324189186096,
+      "learning_rate": 0.0002,
+      "loss": 1.8633,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9861785580874113,
+      "grad_norm": 0.3714013993740082,
+      "learning_rate": 0.0002,
+      "loss": 1.7724,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9936496077698916,
+      "grad_norm": 0.3133082389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.7766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9996264475158759,
+      "eval_loss": 1.8051470518112183,
+      "eval_runtime": 38.6332,
+      "eval_samples_per_second": 13.331,
+      "eval_steps_per_second": 1.682,
+      "step": 1338
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10704,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.194278297711411e+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-0.9-num-6326-sd-42/checkpoint-1338/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-0.9-num-6326-sd-42/checkpoint-1338/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..67c7b4ca126d7b712ad1985ef15ffb29ebe76633
--- /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-0.9-num-6326-sd-42/checkpoint-1338/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fc87da605e94ff0ecd8f5b371302a2d5f8727b77984707a2185ddb447fc3796
+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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f512b3457d4ef0662b10279451f78e55458a70a9
--- /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-0.9-num-6326-sd-42/checkpoint-2677/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8767e0d0604ccff5925fd06b1ebf5a468436644640dabefecdff8b512facf8aa
+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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b07ea83cdced8ab88316d0a2aaf9c0350dab2963
--- /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-0.9-num-6326-sd-42/checkpoint-2677/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7853c5f0c85c20e1969da93901ceeec41a9e7e3c7cdea7b0086247ba5bce952
+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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..159de3c223247df116f6bba8651f19728f842ff1
--- /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-0.9-num-6326-sd-42/checkpoint-2677/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd7a777006d9d84e8e167a92deaa029dbda5c325805a5aee24ceddb0c3465d4a
+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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..15d7adb63848557720d420168a0662fb562a353c
--- /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-0.9-num-6326-sd-42/checkpoint-2677/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5948bcf2b8a11466457f31ca0ea8be19b105a5e0386ed276110793dbcdcaaf44
+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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..b655c58fe7925c016a2d13d8dce2390ae7c6ef34
--- /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-0.9-num-6326-sd-42/checkpoint-2677/trainer_state.json
@@ -0,0 +1,1918 @@
+{
+  "best_metric": 1.8046749830245972,
+  "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-0.9-num-6326-sd-42/checkpoint-2677",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 2677,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007471049682480389,
+      "grad_norm": 0.4912872612476349,
+      "learning_rate": 0.0002,
+      "loss": 2.6181,
+      "step": 10
+    },
+    {
+      "epoch": 0.014942099364960777,
+      "grad_norm": 0.4856316149234772,
+      "learning_rate": 0.0002,
+      "loss": 2.2606,
+      "step": 20
+    },
+    {
+      "epoch": 0.022413149047441166,
+      "grad_norm": 0.47683125734329224,
+      "learning_rate": 0.0002,
+      "loss": 2.0957,
+      "step": 30
+    },
+    {
+      "epoch": 0.029884198729921554,
+      "grad_norm": 0.515082597732544,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 40
+    },
+    {
+      "epoch": 0.03735524841240194,
+      "grad_norm": 0.5299215316772461,
+      "learning_rate": 0.0002,
+      "loss": 1.9704,
+      "step": 50
+    },
+    {
+      "epoch": 0.04482629809488233,
+      "grad_norm": 0.4951399862766266,
+      "learning_rate": 0.0002,
+      "loss": 1.9225,
+      "step": 60
+    },
+    {
+      "epoch": 0.05229734777736272,
+      "grad_norm": 0.48079821467399597,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05976839745984311,
+      "grad_norm": 0.49402132630348206,
+      "learning_rate": 0.0002,
+      "loss": 1.9466,
+      "step": 80
+    },
+    {
+      "epoch": 0.0672394471423235,
+      "grad_norm": 0.4778193235397339,
+      "learning_rate": 0.0002,
+      "loss": 1.8691,
+      "step": 90
+    },
+    {
+      "epoch": 0.07471049682480388,
+      "grad_norm": 0.42472657561302185,
+      "learning_rate": 0.0002,
+      "loss": 1.8455,
+      "step": 100
+    },
+    {
+      "epoch": 0.08218154650728428,
+      "grad_norm": 0.4433092474937439,
+      "learning_rate": 0.0002,
+      "loss": 1.8744,
+      "step": 110
+    },
+    {
+      "epoch": 0.08965259618976466,
+      "grad_norm": 0.4472862780094147,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 120
+    },
+    {
+      "epoch": 0.09712364587224505,
+      "grad_norm": 0.42596298456192017,
+      "learning_rate": 0.0002,
+      "loss": 1.9256,
+      "step": 130
+    },
+    {
+      "epoch": 0.10459469555472543,
+      "grad_norm": 0.46645811200141907,
+      "learning_rate": 0.0002,
+      "loss": 1.8015,
+      "step": 140
+    },
+    {
+      "epoch": 0.11206574523720583,
+      "grad_norm": 0.41041234135627747,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 150
+    },
+    {
+      "epoch": 0.11953679491968622,
+      "grad_norm": 0.5329819917678833,
+      "learning_rate": 0.0002,
+      "loss": 1.8276,
+      "step": 160
+    },
+    {
+      "epoch": 0.1270078446021666,
+      "grad_norm": 0.4065922200679779,
+      "learning_rate": 0.0002,
+      "loss": 1.8118,
+      "step": 170
+    },
+    {
+      "epoch": 0.134478894284647,
+      "grad_norm": 0.38406994938850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8559,
+      "step": 180
+    },
+    {
+      "epoch": 0.14194994396712737,
+      "grad_norm": 0.4246881306171417,
+      "learning_rate": 0.0002,
+      "loss": 1.8647,
+      "step": 190
+    },
+    {
+      "epoch": 0.14942099364960776,
+      "grad_norm": 0.35136649012565613,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 200
+    },
+    {
+      "epoch": 0.15689204333208817,
+      "grad_norm": 0.43252742290496826,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.16436309301456856,
+      "grad_norm": 0.39236941933631897,
+      "learning_rate": 0.0002,
+      "loss": 1.7823,
+      "step": 220
+    },
+    {
+      "epoch": 0.17183414269704894,
+      "grad_norm": 0.3748249113559723,
+      "learning_rate": 0.0002,
+      "loss": 1.818,
+      "step": 230
+    },
+    {
+      "epoch": 0.17930519237952933,
+      "grad_norm": 0.6432855725288391,
+      "learning_rate": 0.0002,
+      "loss": 1.866,
+      "step": 240
+    },
+    {
+      "epoch": 0.1867762420620097,
+      "grad_norm": 0.34874802827835083,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 250
+    },
+    {
+      "epoch": 0.1942472917444901,
+      "grad_norm": 0.3721984326839447,
+      "learning_rate": 0.0002,
+      "loss": 1.79,
+      "step": 260
+    },
+    {
+      "epoch": 0.20171834142697048,
+      "grad_norm": 0.4339311420917511,
+      "learning_rate": 0.0002,
+      "loss": 1.8464,
+      "step": 270
+    },
+    {
+      "epoch": 0.20918939110945087,
+      "grad_norm": 0.4018215537071228,
+      "learning_rate": 0.0002,
+      "loss": 1.8665,
+      "step": 280
+    },
+    {
+      "epoch": 0.21666044079193125,
+      "grad_norm": 0.3278839886188507,
+      "learning_rate": 0.0002,
+      "loss": 1.8048,
+      "step": 290
+    },
+    {
+      "epoch": 0.22413149047441167,
+      "grad_norm": 0.36146077513694763,
+      "learning_rate": 0.0002,
+      "loss": 1.7395,
+      "step": 300
+    },
+    {
+      "epoch": 0.23160254015689205,
+      "grad_norm": 0.38175010681152344,
+      "learning_rate": 0.0002,
+      "loss": 1.7916,
+      "step": 310
+    },
+    {
+      "epoch": 0.23907358983937244,
+      "grad_norm": 0.44776618480682373,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 320
+    },
+    {
+      "epoch": 0.24654463952185282,
+      "grad_norm": 0.3933652937412262,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 330
+    },
+    {
+      "epoch": 0.2540156892043332,
+      "grad_norm": 0.3515005111694336,
+      "learning_rate": 0.0002,
+      "loss": 1.8393,
+      "step": 340
+    },
+    {
+      "epoch": 0.2614867388868136,
+      "grad_norm": 0.6683304309844971,
+      "learning_rate": 0.0002,
+      "loss": 1.8653,
+      "step": 350
+    },
+    {
+      "epoch": 0.268957788569294,
+      "grad_norm": 0.37093454599380493,
+      "learning_rate": 0.0002,
+      "loss": 1.8797,
+      "step": 360
+    },
+    {
+      "epoch": 0.2764288382517744,
+      "grad_norm": 0.3450651168823242,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 370
+    },
+    {
+      "epoch": 0.28389988793425475,
+      "grad_norm": 0.5140917301177979,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 380
+    },
+    {
+      "epoch": 0.29137093761673516,
+      "grad_norm": 0.32885563373565674,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 390
+    },
+    {
+      "epoch": 0.2988419872992155,
+      "grad_norm": 0.33962297439575195,
+      "learning_rate": 0.0002,
+      "loss": 1.8174,
+      "step": 400
+    },
+    {
+      "epoch": 0.30631303698169593,
+      "grad_norm": 0.3723141849040985,
+      "learning_rate": 0.0002,
+      "loss": 1.7467,
+      "step": 410
+    },
+    {
+      "epoch": 0.31378408666417634,
+      "grad_norm": 0.37173134088516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8459,
+      "step": 420
+    },
+    {
+      "epoch": 0.3212551363466567,
+      "grad_norm": 0.33736956119537354,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 430
+    },
+    {
+      "epoch": 0.3287261860291371,
+      "grad_norm": 0.3602448105812073,
+      "learning_rate": 0.0002,
+      "loss": 1.8367,
+      "step": 440
+    },
+    {
+      "epoch": 0.33619723571161747,
+      "grad_norm": 0.3569699227809906,
+      "learning_rate": 0.0002,
+      "loss": 1.8058,
+      "step": 450
+    },
+    {
+      "epoch": 0.3436682853940979,
+      "grad_norm": 0.31009167432785034,
+      "learning_rate": 0.0002,
+      "loss": 1.8086,
+      "step": 460
+    },
+    {
+      "epoch": 0.35113933507657824,
+      "grad_norm": 0.5278693437576294,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 470
+    },
+    {
+      "epoch": 0.35861038475905865,
+      "grad_norm": 0.3587537109851837,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 480
+    },
+    {
+      "epoch": 0.366081434441539,
+      "grad_norm": 0.3859670162200928,
+      "learning_rate": 0.0002,
+      "loss": 1.8046,
+      "step": 490
+    },
+    {
+      "epoch": 0.3735524841240194,
+      "grad_norm": 0.395913690328598,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 500
+    },
+    {
+      "epoch": 0.38102353380649984,
+      "grad_norm": 0.35052940249443054,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 510
+    },
+    {
+      "epoch": 0.3884945834889802,
+      "grad_norm": 0.2979494333267212,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 520
+    },
+    {
+      "epoch": 0.3959656331714606,
+      "grad_norm": 0.3062683343887329,
+      "learning_rate": 0.0002,
+      "loss": 1.8641,
+      "step": 530
+    },
+    {
+      "epoch": 0.40343668285394096,
+      "grad_norm": 0.3172847330570221,
+      "learning_rate": 0.0002,
+      "loss": 1.7651,
+      "step": 540
+    },
+    {
+      "epoch": 0.4109077325364214,
+      "grad_norm": 0.360435426235199,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 550
+    },
+    {
+      "epoch": 0.41837878221890173,
+      "grad_norm": 0.3427872359752655,
+      "learning_rate": 0.0002,
+      "loss": 1.9054,
+      "step": 560
+    },
+    {
+      "epoch": 0.42584983190138215,
+      "grad_norm": 0.34036558866500854,
+      "learning_rate": 0.0002,
+      "loss": 1.7562,
+      "step": 570
+    },
+    {
+      "epoch": 0.4333208815838625,
+      "grad_norm": 0.3365345299243927,
+      "learning_rate": 0.0002,
+      "loss": 1.7254,
+      "step": 580
+    },
+    {
+      "epoch": 0.4407919312663429,
+      "grad_norm": 0.35619041323661804,
+      "learning_rate": 0.0002,
+      "loss": 1.8328,
+      "step": 590
+    },
+    {
+      "epoch": 0.44826298094882333,
+      "grad_norm": 0.3569088280200958,
+      "learning_rate": 0.0002,
+      "loss": 1.8114,
+      "step": 600
+    },
+    {
+      "epoch": 0.4557340306313037,
+      "grad_norm": 0.3581278622150421,
+      "learning_rate": 0.0002,
+      "loss": 1.8599,
+      "step": 610
+    },
+    {
+      "epoch": 0.4632050803137841,
+      "grad_norm": 0.43197110295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.7078,
+      "step": 620
+    },
+    {
+      "epoch": 0.47067612999626446,
+      "grad_norm": 0.33966198563575745,
+      "learning_rate": 0.0002,
+      "loss": 1.8257,
+      "step": 630
+    },
+    {
+      "epoch": 0.47814717967874487,
+      "grad_norm": 0.3343866467475891,
+      "learning_rate": 0.0002,
+      "loss": 1.7528,
+      "step": 640
+    },
+    {
+      "epoch": 0.48561822936122523,
+      "grad_norm": 0.33878564834594727,
+      "learning_rate": 0.0002,
+      "loss": 1.8191,
+      "step": 650
+    },
+    {
+      "epoch": 0.49308927904370564,
+      "grad_norm": 0.387195885181427,
+      "learning_rate": 0.0002,
+      "loss": 1.8801,
+      "step": 660
+    },
+    {
+      "epoch": 0.500560328726186,
+      "grad_norm": 0.3755440413951874,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 670
+    },
+    {
+      "epoch": 0.5080313784086664,
+      "grad_norm": 0.3272816836833954,
+      "learning_rate": 0.0002,
+      "loss": 1.8057,
+      "step": 680
+    },
+    {
+      "epoch": 0.5155024280911468,
+      "grad_norm": 0.36063864827156067,
+      "learning_rate": 0.0002,
+      "loss": 1.8156,
+      "step": 690
+    },
+    {
+      "epoch": 0.5229734777736272,
+      "grad_norm": 0.35317373275756836,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 700
+    },
+    {
+      "epoch": 0.5304445274561076,
+      "grad_norm": 0.3561195433139801,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 710
+    },
+    {
+      "epoch": 0.537915577138588,
+      "grad_norm": 0.31124624609947205,
+      "learning_rate": 0.0002,
+      "loss": 1.8149,
+      "step": 720
+    },
+    {
+      "epoch": 0.5453866268210683,
+      "grad_norm": 0.3294544517993927,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 730
+    },
+    {
+      "epoch": 0.5528576765035488,
+      "grad_norm": 0.31933900713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.8027,
+      "step": 740
+    },
+    {
+      "epoch": 0.5603287261860291,
+      "grad_norm": 0.3226020634174347,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 750
+    },
+    {
+      "epoch": 0.5677997758685095,
+      "grad_norm": 0.3147525489330292,
+      "learning_rate": 0.0002,
+      "loss": 1.7862,
+      "step": 760
+    },
+    {
+      "epoch": 0.57527082555099,
+      "grad_norm": 0.32234328985214233,
+      "learning_rate": 0.0002,
+      "loss": 1.9028,
+      "step": 770
+    },
+    {
+      "epoch": 0.5827418752334703,
+      "grad_norm": 0.3258664309978485,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 780
+    },
+    {
+      "epoch": 0.5902129249159507,
+      "grad_norm": 0.3166961967945099,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 790
+    },
+    {
+      "epoch": 0.597683974598431,
+      "grad_norm": 0.35621458292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 800
+    },
+    {
+      "epoch": 0.6051550242809115,
+      "grad_norm": 0.3236999213695526,
+      "learning_rate": 0.0002,
+      "loss": 1.8313,
+      "step": 810
+    },
+    {
+      "epoch": 0.6126260739633919,
+      "grad_norm": 0.2892923653125763,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 820
+    },
+    {
+      "epoch": 0.6200971236458722,
+      "grad_norm": 0.4098321497440338,
+      "learning_rate": 0.0002,
+      "loss": 1.8709,
+      "step": 830
+    },
+    {
+      "epoch": 0.6275681733283527,
+      "grad_norm": 0.3337118923664093,
+      "learning_rate": 0.0002,
+      "loss": 1.7637,
+      "step": 840
+    },
+    {
+      "epoch": 0.635039223010833,
+      "grad_norm": 0.30416029691696167,
+      "learning_rate": 0.0002,
+      "loss": 1.7375,
+      "step": 850
+    },
+    {
+      "epoch": 0.6425102726933134,
+      "grad_norm": 0.3361026346683502,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 860
+    },
+    {
+      "epoch": 0.6499813223757938,
+      "grad_norm": 0.3537365198135376,
+      "learning_rate": 0.0002,
+      "loss": 1.732,
+      "step": 870
+    },
+    {
+      "epoch": 0.6574523720582742,
+      "grad_norm": 0.33854469656944275,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 880
+    },
+    {
+      "epoch": 0.6649234217407546,
+      "grad_norm": 0.3332272469997406,
+      "learning_rate": 0.0002,
+      "loss": 1.7561,
+      "step": 890
+    },
+    {
+      "epoch": 0.6723944714232349,
+      "grad_norm": 0.34954726696014404,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 900
+    },
+    {
+      "epoch": 0.6798655211057153,
+      "grad_norm": 0.2921750247478485,
+      "learning_rate": 0.0002,
+      "loss": 1.7917,
+      "step": 910
+    },
+    {
+      "epoch": 0.6873365707881958,
+      "grad_norm": 0.30508682131767273,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 920
+    },
+    {
+      "epoch": 0.6948076204706761,
+      "grad_norm": 0.32268425822257996,
+      "learning_rate": 0.0002,
+      "loss": 1.8082,
+      "step": 930
+    },
+    {
+      "epoch": 0.7022786701531565,
+      "grad_norm": 0.2844390869140625,
+      "learning_rate": 0.0002,
+      "loss": 1.8283,
+      "step": 940
+    },
+    {
+      "epoch": 0.709749719835637,
+      "grad_norm": 0.31263890862464905,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 950
+    },
+    {
+      "epoch": 0.7172207695181173,
+      "grad_norm": 0.3626808822154999,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 960
+    },
+    {
+      "epoch": 0.7246918192005977,
+      "grad_norm": 0.3322749733924866,
+      "learning_rate": 0.0002,
+      "loss": 1.853,
+      "step": 970
+    },
+    {
+      "epoch": 0.732162868883078,
+      "grad_norm": 0.29177871346473694,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 980
+    },
+    {
+      "epoch": 0.7396339185655585,
+      "grad_norm": 0.35405513644218445,
+      "learning_rate": 0.0002,
+      "loss": 1.8447,
+      "step": 990
+    },
+    {
+      "epoch": 0.7471049682480388,
+      "grad_norm": 0.39318400621414185,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7545760179305192,
+      "grad_norm": 0.29401418566703796,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7620470676129997,
+      "grad_norm": 0.3271748721599579,
+      "learning_rate": 0.0002,
+      "loss": 1.7649,
+      "step": 1020
+    },
+    {
+      "epoch": 0.76951811729548,
+      "grad_norm": 0.30883970856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7769891669779604,
+      "grad_norm": 0.3411838412284851,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7844602166604407,
+      "grad_norm": 0.30608129501342773,
+      "learning_rate": 0.0002,
+      "loss": 1.829,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7919312663429212,
+      "grad_norm": 0.30899080634117126,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7994023160254016,
+      "grad_norm": 0.3160453140735626,
+      "learning_rate": 0.0002,
+      "loss": 1.7625,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8068733657078819,
+      "grad_norm": 0.30947187542915344,
+      "learning_rate": 0.0002,
+      "loss": 1.8452,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8143444153903624,
+      "grad_norm": 0.3103134036064148,
+      "learning_rate": 0.0002,
+      "loss": 1.7418,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8218154650728428,
+      "grad_norm": 0.31771138310432434,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8292865147553231,
+      "grad_norm": 0.5860997438430786,
+      "learning_rate": 0.0002,
+      "loss": 1.7918,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8367575644378035,
+      "grad_norm": 0.3230148255825043,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8442286141202839,
+      "grad_norm": 0.29611510038375854,
+      "learning_rate": 0.0002,
+      "loss": 1.8478,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8516996638027643,
+      "grad_norm": 0.3373654782772064,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8591707134852447,
+      "grad_norm": 0.3474279046058655,
+      "learning_rate": 0.0002,
+      "loss": 1.7997,
+      "step": 1150
+    },
+    {
+      "epoch": 0.866641763167725,
+      "grad_norm": 0.35057875514030457,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8741128128502055,
+      "grad_norm": 0.39537495374679565,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8815838625326858,
+      "grad_norm": 0.3714233636856079,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8890549122151662,
+      "grad_norm": 0.2950296998023987,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8965259618976467,
+      "grad_norm": 0.38182979822158813,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 1200
+    },
+    {
+      "epoch": 0.903997011580127,
+      "grad_norm": 0.27883678674697876,
+      "learning_rate": 0.0002,
+      "loss": 1.827,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9114680612626074,
+      "grad_norm": 0.33874374628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9189391109450877,
+      "grad_norm": 0.3014272153377533,
+      "learning_rate": 0.0002,
+      "loss": 1.7334,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9264101606275682,
+      "grad_norm": 0.3194271922111511,
+      "learning_rate": 0.0002,
+      "loss": 1.8235,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9338812103100486,
+      "grad_norm": 0.3049403429031372,
+      "learning_rate": 0.0002,
+      "loss": 1.7924,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9413522599925289,
+      "grad_norm": 0.30621254444122314,
+      "learning_rate": 0.0002,
+      "loss": 1.7535,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9488233096750094,
+      "grad_norm": 0.28675132989883423,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9562943593574897,
+      "grad_norm": 0.3322032690048218,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9637654090399701,
+      "grad_norm": 0.35408294200897217,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9712364587224505,
+      "grad_norm": 0.36386919021606445,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9787075084049309,
+      "grad_norm": 0.32338324189186096,
+      "learning_rate": 0.0002,
+      "loss": 1.8633,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9861785580874113,
+      "grad_norm": 0.3714013993740082,
+      "learning_rate": 0.0002,
+      "loss": 1.7724,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9936496077698916,
+      "grad_norm": 0.3133082389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.7766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9996264475158759,
+      "eval_loss": 1.8051470518112183,
+      "eval_runtime": 38.6332,
+      "eval_samples_per_second": 13.331,
+      "eval_steps_per_second": 1.682,
+      "step": 1338
+    },
+    {
+      "epoch": 1.001120657452372,
+      "grad_norm": 0.31595754623413086,
+      "learning_rate": 0.0002,
+      "loss": 1.8035,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0085917071348525,
+      "grad_norm": 0.3095700144767761,
+      "learning_rate": 0.0002,
+      "loss": 1.7486,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0160627568173328,
+      "grad_norm": 0.34677496552467346,
+      "learning_rate": 0.0002,
+      "loss": 1.6981,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0235338064998132,
+      "grad_norm": 0.29108840227127075,
+      "learning_rate": 0.0002,
+      "loss": 1.7377,
+      "step": 1370
+    },
+    {
+      "epoch": 1.0310048561822935,
+      "grad_norm": 0.32356950640678406,
+      "learning_rate": 0.0002,
+      "loss": 1.7194,
+      "step": 1380
+    },
+    {
+      "epoch": 1.038475905864774,
+      "grad_norm": 0.4200669229030609,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0459469555472545,
+      "grad_norm": 0.3283711373806,
+      "learning_rate": 0.0002,
+      "loss": 1.797,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0534180052297348,
+      "grad_norm": 0.32898256182670593,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0608890549122152,
+      "grad_norm": 0.38790300488471985,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0683601045946955,
+      "grad_norm": 0.339800089597702,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1430
+    },
+    {
+      "epoch": 1.075831154277176,
+      "grad_norm": 0.3548751175403595,
+      "learning_rate": 0.0002,
+      "loss": 1.7076,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0833022039596563,
+      "grad_norm": 0.35114359855651855,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0907732536421366,
+      "grad_norm": 0.35226720571517944,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0982443033246172,
+      "grad_norm": 0.33665576577186584,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1057153530070976,
+      "grad_norm": 0.363889217376709,
+      "learning_rate": 0.0002,
+      "loss": 1.6699,
+      "step": 1480
+    },
+    {
+      "epoch": 1.113186402689578,
+      "grad_norm": 0.3826201856136322,
+      "learning_rate": 0.0002,
+      "loss": 1.7933,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1206574523720583,
+      "grad_norm": 0.34058740735054016,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1281285020545386,
+      "grad_norm": 0.3462134301662445,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 1510
+    },
+    {
+      "epoch": 1.135599551737019,
+      "grad_norm": 0.3396756052970886,
+      "learning_rate": 0.0002,
+      "loss": 1.7147,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1430706014194993,
+      "grad_norm": 0.32004743814468384,
+      "learning_rate": 0.0002,
+      "loss": 1.7219,
+      "step": 1530
+    },
+    {
+      "epoch": 1.15054165110198,
+      "grad_norm": 0.3397733271121979,
+      "learning_rate": 0.0002,
+      "loss": 1.743,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1580127007844603,
+      "grad_norm": 0.3783262073993683,
+      "learning_rate": 0.0002,
+      "loss": 1.7333,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1654837504669406,
+      "grad_norm": 0.35121291875839233,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 1560
+    },
+    {
+      "epoch": 1.172954800149421,
+      "grad_norm": 0.35816895961761475,
+      "learning_rate": 0.0002,
+      "loss": 1.678,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1804258498319014,
+      "grad_norm": 0.33843839168548584,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1878968995143817,
+      "grad_norm": 0.3371972143650055,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 1590
+    },
+    {
+      "epoch": 1.195367949196862,
+      "grad_norm": 0.36016878485679626,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2028389988793426,
+      "grad_norm": 0.40879473090171814,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 1610
+    },
+    {
+      "epoch": 1.210310048561823,
+      "grad_norm": 0.3216715455055237,
+      "learning_rate": 0.0002,
+      "loss": 1.6955,
+      "step": 1620
+    },
+    {
+      "epoch": 1.2177810982443034,
+      "grad_norm": 0.4482610821723938,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2252521479267837,
+      "grad_norm": 0.3257700502872467,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1640
+    },
+    {
+      "epoch": 1.232723197609264,
+      "grad_norm": 0.38646459579467773,
+      "learning_rate": 0.0002,
+      "loss": 1.7177,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2401942472917444,
+      "grad_norm": 0.4081360697746277,
+      "learning_rate": 0.0002,
+      "loss": 1.7081,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2476652969742248,
+      "grad_norm": 0.4326848089694977,
+      "learning_rate": 0.0002,
+      "loss": 1.7519,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2551363466567054,
+      "grad_norm": 0.346401572227478,
+      "learning_rate": 0.0002,
+      "loss": 1.6752,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2626073963391857,
+      "grad_norm": 0.34536251425743103,
+      "learning_rate": 0.0002,
+      "loss": 1.7425,
+      "step": 1690
+    },
+    {
+      "epoch": 1.270078446021666,
+      "grad_norm": 0.41359591484069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7061,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2775494957041464,
+      "grad_norm": 0.3530874252319336,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2850205453866268,
+      "grad_norm": 0.3702719211578369,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2924915950691072,
+      "grad_norm": 0.3703329563140869,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2999626447515875,
+      "grad_norm": 0.37919729948043823,
+      "learning_rate": 0.0002,
+      "loss": 1.7221,
+      "step": 1740
+    },
+    {
+      "epoch": 1.307433694434068,
+      "grad_norm": 0.32526856660842896,
+      "learning_rate": 0.0002,
+      "loss": 1.7859,
+      "step": 1750
+    },
+    {
+      "epoch": 1.3149047441165485,
+      "grad_norm": 0.36752620339393616,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3223757937990288,
+      "grad_norm": 0.3398192524909973,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3298468434815092,
+      "grad_norm": 0.37435585260391235,
+      "learning_rate": 0.0002,
+      "loss": 1.7492,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3373178931639895,
+      "grad_norm": 0.35793280601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3447889428464699,
+      "grad_norm": 0.35481882095336914,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3522599925289502,
+      "grad_norm": 0.3786393105983734,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3597310422114308,
+      "grad_norm": 0.33245593309402466,
+      "learning_rate": 0.0002,
+      "loss": 1.7169,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3672020918939112,
+      "grad_norm": 0.35388344526290894,
+      "learning_rate": 0.0002,
+      "loss": 1.7577,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3746731415763915,
+      "grad_norm": 0.3695325553417206,
+      "learning_rate": 0.0002,
+      "loss": 1.6968,
+      "step": 1840
+    },
+    {
+      "epoch": 1.382144191258872,
+      "grad_norm": 0.3683604598045349,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3896152409413522,
+      "grad_norm": 0.3753012418746948,
+      "learning_rate": 0.0002,
+      "loss": 1.7878,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3970862906238326,
+      "grad_norm": 0.3331069350242615,
+      "learning_rate": 0.0002,
+      "loss": 1.6969,
+      "step": 1870
+    },
+    {
+      "epoch": 1.404557340306313,
+      "grad_norm": 0.3877500295639038,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4120283899887935,
+      "grad_norm": 0.33525151014328003,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4194994396712737,
+      "grad_norm": 0.3697299659252167,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4269704893537543,
+      "grad_norm": 0.4029286205768585,
+      "learning_rate": 0.0002,
+      "loss": 1.6956,
+      "step": 1910
+    },
+    {
+      "epoch": 1.4344415390362346,
+      "grad_norm": 0.3596203029155731,
+      "learning_rate": 0.0002,
+      "loss": 1.6897,
+      "step": 1920
+    },
+    {
+      "epoch": 1.441912588718715,
+      "grad_norm": 0.450783908367157,
+      "learning_rate": 0.0002,
+      "loss": 1.7139,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4493836384011953,
+      "grad_norm": 0.3651481866836548,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4568546880836757,
+      "grad_norm": 0.3608424663543701,
+      "learning_rate": 0.0002,
+      "loss": 1.6637,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4643257377661563,
+      "grad_norm": 0.39684420824050903,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4717967874486364,
+      "grad_norm": 0.34618663787841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7514,
+      "step": 1970
+    },
+    {
+      "epoch": 1.479267837131117,
+      "grad_norm": 0.4150386452674866,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4867388868135973,
+      "grad_norm": 0.35500776767730713,
+      "learning_rate": 0.0002,
+      "loss": 1.7021,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4942099364960777,
+      "grad_norm": 0.344144344329834,
+      "learning_rate": 0.0002,
+      "loss": 1.7322,
+      "step": 2000
+    },
+    {
+      "epoch": 1.501680986178558,
+      "grad_norm": 0.3340149223804474,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5091520358610384,
+      "grad_norm": 0.37685006856918335,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 2020
+    },
+    {
+      "epoch": 1.516623085543519,
+      "grad_norm": 0.3699876368045807,
+      "learning_rate": 0.0002,
+      "loss": 1.8299,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5240941352259991,
+      "grad_norm": 0.3370307385921478,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5315651849084797,
+      "grad_norm": 0.37780630588531494,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 2050
+    },
+    {
+      "epoch": 1.53903623459096,
+      "grad_norm": 0.370259165763855,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5465072842734404,
+      "grad_norm": 0.3440011441707611,
+      "learning_rate": 0.0002,
+      "loss": 1.7398,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5539783339559208,
+      "grad_norm": 0.40382063388824463,
+      "learning_rate": 0.0002,
+      "loss": 1.7105,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5614493836384011,
+      "grad_norm": 0.38002029061317444,
+      "learning_rate": 0.0002,
+      "loss": 1.7071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5689204333208817,
+      "grad_norm": 0.3658451437950134,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5763914830033618,
+      "grad_norm": 0.354842871427536,
+      "learning_rate": 0.0002,
+      "loss": 1.7598,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5838625326858424,
+      "grad_norm": 0.34735530614852905,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5913335823683228,
+      "grad_norm": 0.377581924200058,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5988046320508031,
+      "grad_norm": 0.41254034638404846,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6062756817332835,
+      "grad_norm": 0.3630715310573578,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6137467314157639,
+      "grad_norm": 0.36980143189430237,
+      "learning_rate": 0.0002,
+      "loss": 1.7531,
+      "step": 2160
+    },
+    {
+      "epoch": 1.6212177810982444,
+      "grad_norm": 0.3634769320487976,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6286888307807246,
+      "grad_norm": 0.3794139623641968,
+      "learning_rate": 0.0002,
+      "loss": 1.6367,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6361598804632052,
+      "grad_norm": 0.359742134809494,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6436309301456855,
+      "grad_norm": 0.3770543932914734,
+      "learning_rate": 0.0002,
+      "loss": 1.7027,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6511019798281659,
+      "grad_norm": 0.3797036409378052,
+      "learning_rate": 0.0002,
+      "loss": 1.784,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6585730295106462,
+      "grad_norm": 0.35622093081474304,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6660440791931266,
+      "grad_norm": 0.34552520513534546,
+      "learning_rate": 0.0002,
+      "loss": 1.6615,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6735151288756072,
+      "grad_norm": 0.379926860332489,
+      "learning_rate": 0.0002,
+      "loss": 1.7522,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6809861785580873,
+      "grad_norm": 0.37083810567855835,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6884572282405679,
+      "grad_norm": 0.42746543884277344,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6959282779230482,
+      "grad_norm": 0.3372884690761566,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7033993276055286,
+      "grad_norm": 0.35220256447792053,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2280
+    },
+    {
+      "epoch": 1.710870377288009,
+      "grad_norm": 0.3659130930900574,
+      "learning_rate": 0.0002,
+      "loss": 1.7154,
+      "step": 2290
+    },
+    {
+      "epoch": 1.7183414269704893,
+      "grad_norm": 0.37629297375679016,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7258124766529699,
+      "grad_norm": 0.36312398314476013,
+      "learning_rate": 0.0002,
+      "loss": 1.7212,
+      "step": 2310
+    },
+    {
+      "epoch": 1.73328352633545,
+      "grad_norm": 0.467709481716156,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7407545760179306,
+      "grad_norm": 0.38685527443885803,
+      "learning_rate": 0.0002,
+      "loss": 1.696,
+      "step": 2330
+    },
+    {
+      "epoch": 1.748225625700411,
+      "grad_norm": 0.3578338325023651,
+      "learning_rate": 0.0002,
+      "loss": 1.7041,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7556966753828913,
+      "grad_norm": 0.36057502031326294,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7631677250653717,
+      "grad_norm": 0.3615196645259857,
+      "learning_rate": 0.0002,
+      "loss": 1.6853,
+      "step": 2360
+    },
+    {
+      "epoch": 1.770638774747852,
+      "grad_norm": 0.4118947684764862,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7781098244303326,
+      "grad_norm": 0.4067276120185852,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7855808741128127,
+      "grad_norm": 0.3979823887348175,
+      "learning_rate": 0.0002,
+      "loss": 1.712,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7930519237952933,
+      "grad_norm": 0.44045883417129517,
+      "learning_rate": 0.0002,
+      "loss": 1.7644,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8005229734777737,
+      "grad_norm": 0.3998069167137146,
+      "learning_rate": 0.0002,
+      "loss": 1.7251,
+      "step": 2410
+    },
+    {
+      "epoch": 1.807994023160254,
+      "grad_norm": 0.3450094759464264,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8154650728427344,
+      "grad_norm": 0.3759009838104248,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2430
+    },
+    {
+      "epoch": 1.8229361225252148,
+      "grad_norm": 0.34347015619277954,
+      "learning_rate": 0.0002,
+      "loss": 1.7706,
+      "step": 2440
+    },
+    {
+      "epoch": 1.8304071722076953,
+      "grad_norm": 0.3511228859424591,
+      "learning_rate": 0.0002,
+      "loss": 1.7345,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8378782218901755,
+      "grad_norm": 0.36853715777397156,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 2460
+    },
+    {
+      "epoch": 1.845349271572656,
+      "grad_norm": 0.40659376978874207,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8528203212551362,
+      "grad_norm": 0.39621320366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8602913709376168,
+      "grad_norm": 0.3753979504108429,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8677624206200971,
+      "grad_norm": 0.3811938464641571,
+      "learning_rate": 0.0002,
+      "loss": 1.6622,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8752334703025775,
+      "grad_norm": 0.3432596027851105,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 2510
+    },
+    {
+      "epoch": 1.882704519985058,
+      "grad_norm": 0.3670712113380432,
+      "learning_rate": 0.0002,
+      "loss": 1.7488,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8901755696675382,
+      "grad_norm": 0.40907177329063416,
+      "learning_rate": 0.0002,
+      "loss": 1.705,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8976466193500188,
+      "grad_norm": 0.3821999728679657,
+      "learning_rate": 0.0002,
+      "loss": 1.7148,
+      "step": 2540
+    },
+    {
+      "epoch": 1.905117669032499,
+      "grad_norm": 0.36173978447914124,
+      "learning_rate": 0.0002,
+      "loss": 1.7934,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9125887187149795,
+      "grad_norm": 0.38990336656570435,
+      "learning_rate": 0.0002,
+      "loss": 1.6939,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9200597683974598,
+      "grad_norm": 0.35242322087287903,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 2570
+    },
+    {
+      "epoch": 1.9275308180799402,
+      "grad_norm": 0.3506428003311157,
+      "learning_rate": 0.0002,
+      "loss": 1.7268,
+      "step": 2580
+    },
+    {
+      "epoch": 1.9350018677624208,
+      "grad_norm": 0.39540135860443115,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2590
+    },
+    {
+      "epoch": 1.942472917444901,
+      "grad_norm": 0.3444725573062897,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9499439671273815,
+      "grad_norm": 0.3963521718978882,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9574150168098616,
+      "grad_norm": 0.3689815402030945,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9648860664923422,
+      "grad_norm": 0.3482626676559448,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9723571161748226,
+      "grad_norm": 0.35832616686820984,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 2640
+    },
+    {
+      "epoch": 1.979828165857303,
+      "grad_norm": 0.4776208996772766,
+      "learning_rate": 0.0002,
+      "loss": 1.6681,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9872992155397835,
+      "grad_norm": 0.32570165395736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6696,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9947702652222636,
+      "grad_norm": 0.3380725085735321,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8046749830245972,
+      "eval_runtime": 38.5096,
+      "eval_samples_per_second": 13.373,
+      "eval_steps_per_second": 1.688,
+      "step": 2677
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10704,
+  "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": 1.2388556595422822e+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-0.9-num-6326-sd-42/checkpoint-2677/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-0.9-num-6326-sd-42/checkpoint-2677/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..67c7b4ca126d7b712ad1985ef15ffb29ebe76633
--- /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-0.9-num-6326-sd-42/checkpoint-2677/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fc87da605e94ff0ecd8f5b371302a2d5f8727b77984707a2185ddb447fc3796
+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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..4926b7ab82fda49f8d82c31524a0f039d62e0c6e
--- /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-0.9-num-6326-sd-42/checkpoint-4015/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d478f3d289857e8c3bac6e8f78ca8e9abcdca9a67a0f9422f335d9f204ba81c
+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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fcac5e5b1957a6ed6fa07c0ed71cfeff2f15a579
--- /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-0.9-num-6326-sd-42/checkpoint-4015/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d177228f65938693aca16f68ff150166213b5361c99e1eb61010f54f8a11485
+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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6937dae766b596444efa9130eff0caa9fad15f74
--- /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-0.9-num-6326-sd-42/checkpoint-4015/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:815e355ea07f47424238b4562e23ff869072de740aa219e0094905863e498161
+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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..434a5ea562ea258150dbf9fa7244707ac6216167
--- /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-0.9-num-6326-sd-42/checkpoint-4015/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e3229c0eb3893b4ed9dc78c51b78922835046ad4263156ec195b41cd09021c9
+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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..08a91535c7d262f7eac4f74081b715ea9a9c9aa1
--- /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-0.9-num-6326-sd-42/checkpoint-4015/trainer_state.json
@@ -0,0 +1,2864 @@
+{
+  "best_metric": 1.8046749830245972,
+  "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-0.9-num-6326-sd-42/checkpoint-2677",
+  "epoch": 2.999626447515876,
+  "eval_steps": 10,
+  "global_step": 4015,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007471049682480389,
+      "grad_norm": 0.4912872612476349,
+      "learning_rate": 0.0002,
+      "loss": 2.6181,
+      "step": 10
+    },
+    {
+      "epoch": 0.014942099364960777,
+      "grad_norm": 0.4856316149234772,
+      "learning_rate": 0.0002,
+      "loss": 2.2606,
+      "step": 20
+    },
+    {
+      "epoch": 0.022413149047441166,
+      "grad_norm": 0.47683125734329224,
+      "learning_rate": 0.0002,
+      "loss": 2.0957,
+      "step": 30
+    },
+    {
+      "epoch": 0.029884198729921554,
+      "grad_norm": 0.515082597732544,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 40
+    },
+    {
+      "epoch": 0.03735524841240194,
+      "grad_norm": 0.5299215316772461,
+      "learning_rate": 0.0002,
+      "loss": 1.9704,
+      "step": 50
+    },
+    {
+      "epoch": 0.04482629809488233,
+      "grad_norm": 0.4951399862766266,
+      "learning_rate": 0.0002,
+      "loss": 1.9225,
+      "step": 60
+    },
+    {
+      "epoch": 0.05229734777736272,
+      "grad_norm": 0.48079821467399597,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05976839745984311,
+      "grad_norm": 0.49402132630348206,
+      "learning_rate": 0.0002,
+      "loss": 1.9466,
+      "step": 80
+    },
+    {
+      "epoch": 0.0672394471423235,
+      "grad_norm": 0.4778193235397339,
+      "learning_rate": 0.0002,
+      "loss": 1.8691,
+      "step": 90
+    },
+    {
+      "epoch": 0.07471049682480388,
+      "grad_norm": 0.42472657561302185,
+      "learning_rate": 0.0002,
+      "loss": 1.8455,
+      "step": 100
+    },
+    {
+      "epoch": 0.08218154650728428,
+      "grad_norm": 0.4433092474937439,
+      "learning_rate": 0.0002,
+      "loss": 1.8744,
+      "step": 110
+    },
+    {
+      "epoch": 0.08965259618976466,
+      "grad_norm": 0.4472862780094147,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 120
+    },
+    {
+      "epoch": 0.09712364587224505,
+      "grad_norm": 0.42596298456192017,
+      "learning_rate": 0.0002,
+      "loss": 1.9256,
+      "step": 130
+    },
+    {
+      "epoch": 0.10459469555472543,
+      "grad_norm": 0.46645811200141907,
+      "learning_rate": 0.0002,
+      "loss": 1.8015,
+      "step": 140
+    },
+    {
+      "epoch": 0.11206574523720583,
+      "grad_norm": 0.41041234135627747,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 150
+    },
+    {
+      "epoch": 0.11953679491968622,
+      "grad_norm": 0.5329819917678833,
+      "learning_rate": 0.0002,
+      "loss": 1.8276,
+      "step": 160
+    },
+    {
+      "epoch": 0.1270078446021666,
+      "grad_norm": 0.4065922200679779,
+      "learning_rate": 0.0002,
+      "loss": 1.8118,
+      "step": 170
+    },
+    {
+      "epoch": 0.134478894284647,
+      "grad_norm": 0.38406994938850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8559,
+      "step": 180
+    },
+    {
+      "epoch": 0.14194994396712737,
+      "grad_norm": 0.4246881306171417,
+      "learning_rate": 0.0002,
+      "loss": 1.8647,
+      "step": 190
+    },
+    {
+      "epoch": 0.14942099364960776,
+      "grad_norm": 0.35136649012565613,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 200
+    },
+    {
+      "epoch": 0.15689204333208817,
+      "grad_norm": 0.43252742290496826,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.16436309301456856,
+      "grad_norm": 0.39236941933631897,
+      "learning_rate": 0.0002,
+      "loss": 1.7823,
+      "step": 220
+    },
+    {
+      "epoch": 0.17183414269704894,
+      "grad_norm": 0.3748249113559723,
+      "learning_rate": 0.0002,
+      "loss": 1.818,
+      "step": 230
+    },
+    {
+      "epoch": 0.17930519237952933,
+      "grad_norm": 0.6432855725288391,
+      "learning_rate": 0.0002,
+      "loss": 1.866,
+      "step": 240
+    },
+    {
+      "epoch": 0.1867762420620097,
+      "grad_norm": 0.34874802827835083,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 250
+    },
+    {
+      "epoch": 0.1942472917444901,
+      "grad_norm": 0.3721984326839447,
+      "learning_rate": 0.0002,
+      "loss": 1.79,
+      "step": 260
+    },
+    {
+      "epoch": 0.20171834142697048,
+      "grad_norm": 0.4339311420917511,
+      "learning_rate": 0.0002,
+      "loss": 1.8464,
+      "step": 270
+    },
+    {
+      "epoch": 0.20918939110945087,
+      "grad_norm": 0.4018215537071228,
+      "learning_rate": 0.0002,
+      "loss": 1.8665,
+      "step": 280
+    },
+    {
+      "epoch": 0.21666044079193125,
+      "grad_norm": 0.3278839886188507,
+      "learning_rate": 0.0002,
+      "loss": 1.8048,
+      "step": 290
+    },
+    {
+      "epoch": 0.22413149047441167,
+      "grad_norm": 0.36146077513694763,
+      "learning_rate": 0.0002,
+      "loss": 1.7395,
+      "step": 300
+    },
+    {
+      "epoch": 0.23160254015689205,
+      "grad_norm": 0.38175010681152344,
+      "learning_rate": 0.0002,
+      "loss": 1.7916,
+      "step": 310
+    },
+    {
+      "epoch": 0.23907358983937244,
+      "grad_norm": 0.44776618480682373,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 320
+    },
+    {
+      "epoch": 0.24654463952185282,
+      "grad_norm": 0.3933652937412262,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 330
+    },
+    {
+      "epoch": 0.2540156892043332,
+      "grad_norm": 0.3515005111694336,
+      "learning_rate": 0.0002,
+      "loss": 1.8393,
+      "step": 340
+    },
+    {
+      "epoch": 0.2614867388868136,
+      "grad_norm": 0.6683304309844971,
+      "learning_rate": 0.0002,
+      "loss": 1.8653,
+      "step": 350
+    },
+    {
+      "epoch": 0.268957788569294,
+      "grad_norm": 0.37093454599380493,
+      "learning_rate": 0.0002,
+      "loss": 1.8797,
+      "step": 360
+    },
+    {
+      "epoch": 0.2764288382517744,
+      "grad_norm": 0.3450651168823242,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 370
+    },
+    {
+      "epoch": 0.28389988793425475,
+      "grad_norm": 0.5140917301177979,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 380
+    },
+    {
+      "epoch": 0.29137093761673516,
+      "grad_norm": 0.32885563373565674,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 390
+    },
+    {
+      "epoch": 0.2988419872992155,
+      "grad_norm": 0.33962297439575195,
+      "learning_rate": 0.0002,
+      "loss": 1.8174,
+      "step": 400
+    },
+    {
+      "epoch": 0.30631303698169593,
+      "grad_norm": 0.3723141849040985,
+      "learning_rate": 0.0002,
+      "loss": 1.7467,
+      "step": 410
+    },
+    {
+      "epoch": 0.31378408666417634,
+      "grad_norm": 0.37173134088516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8459,
+      "step": 420
+    },
+    {
+      "epoch": 0.3212551363466567,
+      "grad_norm": 0.33736956119537354,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 430
+    },
+    {
+      "epoch": 0.3287261860291371,
+      "grad_norm": 0.3602448105812073,
+      "learning_rate": 0.0002,
+      "loss": 1.8367,
+      "step": 440
+    },
+    {
+      "epoch": 0.33619723571161747,
+      "grad_norm": 0.3569699227809906,
+      "learning_rate": 0.0002,
+      "loss": 1.8058,
+      "step": 450
+    },
+    {
+      "epoch": 0.3436682853940979,
+      "grad_norm": 0.31009167432785034,
+      "learning_rate": 0.0002,
+      "loss": 1.8086,
+      "step": 460
+    },
+    {
+      "epoch": 0.35113933507657824,
+      "grad_norm": 0.5278693437576294,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 470
+    },
+    {
+      "epoch": 0.35861038475905865,
+      "grad_norm": 0.3587537109851837,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 480
+    },
+    {
+      "epoch": 0.366081434441539,
+      "grad_norm": 0.3859670162200928,
+      "learning_rate": 0.0002,
+      "loss": 1.8046,
+      "step": 490
+    },
+    {
+      "epoch": 0.3735524841240194,
+      "grad_norm": 0.395913690328598,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 500
+    },
+    {
+      "epoch": 0.38102353380649984,
+      "grad_norm": 0.35052940249443054,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 510
+    },
+    {
+      "epoch": 0.3884945834889802,
+      "grad_norm": 0.2979494333267212,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 520
+    },
+    {
+      "epoch": 0.3959656331714606,
+      "grad_norm": 0.3062683343887329,
+      "learning_rate": 0.0002,
+      "loss": 1.8641,
+      "step": 530
+    },
+    {
+      "epoch": 0.40343668285394096,
+      "grad_norm": 0.3172847330570221,
+      "learning_rate": 0.0002,
+      "loss": 1.7651,
+      "step": 540
+    },
+    {
+      "epoch": 0.4109077325364214,
+      "grad_norm": 0.360435426235199,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 550
+    },
+    {
+      "epoch": 0.41837878221890173,
+      "grad_norm": 0.3427872359752655,
+      "learning_rate": 0.0002,
+      "loss": 1.9054,
+      "step": 560
+    },
+    {
+      "epoch": 0.42584983190138215,
+      "grad_norm": 0.34036558866500854,
+      "learning_rate": 0.0002,
+      "loss": 1.7562,
+      "step": 570
+    },
+    {
+      "epoch": 0.4333208815838625,
+      "grad_norm": 0.3365345299243927,
+      "learning_rate": 0.0002,
+      "loss": 1.7254,
+      "step": 580
+    },
+    {
+      "epoch": 0.4407919312663429,
+      "grad_norm": 0.35619041323661804,
+      "learning_rate": 0.0002,
+      "loss": 1.8328,
+      "step": 590
+    },
+    {
+      "epoch": 0.44826298094882333,
+      "grad_norm": 0.3569088280200958,
+      "learning_rate": 0.0002,
+      "loss": 1.8114,
+      "step": 600
+    },
+    {
+      "epoch": 0.4557340306313037,
+      "grad_norm": 0.3581278622150421,
+      "learning_rate": 0.0002,
+      "loss": 1.8599,
+      "step": 610
+    },
+    {
+      "epoch": 0.4632050803137841,
+      "grad_norm": 0.43197110295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.7078,
+      "step": 620
+    },
+    {
+      "epoch": 0.47067612999626446,
+      "grad_norm": 0.33966198563575745,
+      "learning_rate": 0.0002,
+      "loss": 1.8257,
+      "step": 630
+    },
+    {
+      "epoch": 0.47814717967874487,
+      "grad_norm": 0.3343866467475891,
+      "learning_rate": 0.0002,
+      "loss": 1.7528,
+      "step": 640
+    },
+    {
+      "epoch": 0.48561822936122523,
+      "grad_norm": 0.33878564834594727,
+      "learning_rate": 0.0002,
+      "loss": 1.8191,
+      "step": 650
+    },
+    {
+      "epoch": 0.49308927904370564,
+      "grad_norm": 0.387195885181427,
+      "learning_rate": 0.0002,
+      "loss": 1.8801,
+      "step": 660
+    },
+    {
+      "epoch": 0.500560328726186,
+      "grad_norm": 0.3755440413951874,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 670
+    },
+    {
+      "epoch": 0.5080313784086664,
+      "grad_norm": 0.3272816836833954,
+      "learning_rate": 0.0002,
+      "loss": 1.8057,
+      "step": 680
+    },
+    {
+      "epoch": 0.5155024280911468,
+      "grad_norm": 0.36063864827156067,
+      "learning_rate": 0.0002,
+      "loss": 1.8156,
+      "step": 690
+    },
+    {
+      "epoch": 0.5229734777736272,
+      "grad_norm": 0.35317373275756836,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 700
+    },
+    {
+      "epoch": 0.5304445274561076,
+      "grad_norm": 0.3561195433139801,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 710
+    },
+    {
+      "epoch": 0.537915577138588,
+      "grad_norm": 0.31124624609947205,
+      "learning_rate": 0.0002,
+      "loss": 1.8149,
+      "step": 720
+    },
+    {
+      "epoch": 0.5453866268210683,
+      "grad_norm": 0.3294544517993927,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 730
+    },
+    {
+      "epoch": 0.5528576765035488,
+      "grad_norm": 0.31933900713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.8027,
+      "step": 740
+    },
+    {
+      "epoch": 0.5603287261860291,
+      "grad_norm": 0.3226020634174347,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 750
+    },
+    {
+      "epoch": 0.5677997758685095,
+      "grad_norm": 0.3147525489330292,
+      "learning_rate": 0.0002,
+      "loss": 1.7862,
+      "step": 760
+    },
+    {
+      "epoch": 0.57527082555099,
+      "grad_norm": 0.32234328985214233,
+      "learning_rate": 0.0002,
+      "loss": 1.9028,
+      "step": 770
+    },
+    {
+      "epoch": 0.5827418752334703,
+      "grad_norm": 0.3258664309978485,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 780
+    },
+    {
+      "epoch": 0.5902129249159507,
+      "grad_norm": 0.3166961967945099,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 790
+    },
+    {
+      "epoch": 0.597683974598431,
+      "grad_norm": 0.35621458292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 800
+    },
+    {
+      "epoch": 0.6051550242809115,
+      "grad_norm": 0.3236999213695526,
+      "learning_rate": 0.0002,
+      "loss": 1.8313,
+      "step": 810
+    },
+    {
+      "epoch": 0.6126260739633919,
+      "grad_norm": 0.2892923653125763,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 820
+    },
+    {
+      "epoch": 0.6200971236458722,
+      "grad_norm": 0.4098321497440338,
+      "learning_rate": 0.0002,
+      "loss": 1.8709,
+      "step": 830
+    },
+    {
+      "epoch": 0.6275681733283527,
+      "grad_norm": 0.3337118923664093,
+      "learning_rate": 0.0002,
+      "loss": 1.7637,
+      "step": 840
+    },
+    {
+      "epoch": 0.635039223010833,
+      "grad_norm": 0.30416029691696167,
+      "learning_rate": 0.0002,
+      "loss": 1.7375,
+      "step": 850
+    },
+    {
+      "epoch": 0.6425102726933134,
+      "grad_norm": 0.3361026346683502,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 860
+    },
+    {
+      "epoch": 0.6499813223757938,
+      "grad_norm": 0.3537365198135376,
+      "learning_rate": 0.0002,
+      "loss": 1.732,
+      "step": 870
+    },
+    {
+      "epoch": 0.6574523720582742,
+      "grad_norm": 0.33854469656944275,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 880
+    },
+    {
+      "epoch": 0.6649234217407546,
+      "grad_norm": 0.3332272469997406,
+      "learning_rate": 0.0002,
+      "loss": 1.7561,
+      "step": 890
+    },
+    {
+      "epoch": 0.6723944714232349,
+      "grad_norm": 0.34954726696014404,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 900
+    },
+    {
+      "epoch": 0.6798655211057153,
+      "grad_norm": 0.2921750247478485,
+      "learning_rate": 0.0002,
+      "loss": 1.7917,
+      "step": 910
+    },
+    {
+      "epoch": 0.6873365707881958,
+      "grad_norm": 0.30508682131767273,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 920
+    },
+    {
+      "epoch": 0.6948076204706761,
+      "grad_norm": 0.32268425822257996,
+      "learning_rate": 0.0002,
+      "loss": 1.8082,
+      "step": 930
+    },
+    {
+      "epoch": 0.7022786701531565,
+      "grad_norm": 0.2844390869140625,
+      "learning_rate": 0.0002,
+      "loss": 1.8283,
+      "step": 940
+    },
+    {
+      "epoch": 0.709749719835637,
+      "grad_norm": 0.31263890862464905,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 950
+    },
+    {
+      "epoch": 0.7172207695181173,
+      "grad_norm": 0.3626808822154999,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 960
+    },
+    {
+      "epoch": 0.7246918192005977,
+      "grad_norm": 0.3322749733924866,
+      "learning_rate": 0.0002,
+      "loss": 1.853,
+      "step": 970
+    },
+    {
+      "epoch": 0.732162868883078,
+      "grad_norm": 0.29177871346473694,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 980
+    },
+    {
+      "epoch": 0.7396339185655585,
+      "grad_norm": 0.35405513644218445,
+      "learning_rate": 0.0002,
+      "loss": 1.8447,
+      "step": 990
+    },
+    {
+      "epoch": 0.7471049682480388,
+      "grad_norm": 0.39318400621414185,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7545760179305192,
+      "grad_norm": 0.29401418566703796,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7620470676129997,
+      "grad_norm": 0.3271748721599579,
+      "learning_rate": 0.0002,
+      "loss": 1.7649,
+      "step": 1020
+    },
+    {
+      "epoch": 0.76951811729548,
+      "grad_norm": 0.30883970856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7769891669779604,
+      "grad_norm": 0.3411838412284851,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7844602166604407,
+      "grad_norm": 0.30608129501342773,
+      "learning_rate": 0.0002,
+      "loss": 1.829,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7919312663429212,
+      "grad_norm": 0.30899080634117126,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7994023160254016,
+      "grad_norm": 0.3160453140735626,
+      "learning_rate": 0.0002,
+      "loss": 1.7625,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8068733657078819,
+      "grad_norm": 0.30947187542915344,
+      "learning_rate": 0.0002,
+      "loss": 1.8452,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8143444153903624,
+      "grad_norm": 0.3103134036064148,
+      "learning_rate": 0.0002,
+      "loss": 1.7418,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8218154650728428,
+      "grad_norm": 0.31771138310432434,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8292865147553231,
+      "grad_norm": 0.5860997438430786,
+      "learning_rate": 0.0002,
+      "loss": 1.7918,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8367575644378035,
+      "grad_norm": 0.3230148255825043,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8442286141202839,
+      "grad_norm": 0.29611510038375854,
+      "learning_rate": 0.0002,
+      "loss": 1.8478,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8516996638027643,
+      "grad_norm": 0.3373654782772064,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8591707134852447,
+      "grad_norm": 0.3474279046058655,
+      "learning_rate": 0.0002,
+      "loss": 1.7997,
+      "step": 1150
+    },
+    {
+      "epoch": 0.866641763167725,
+      "grad_norm": 0.35057875514030457,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8741128128502055,
+      "grad_norm": 0.39537495374679565,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8815838625326858,
+      "grad_norm": 0.3714233636856079,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8890549122151662,
+      "grad_norm": 0.2950296998023987,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8965259618976467,
+      "grad_norm": 0.38182979822158813,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 1200
+    },
+    {
+      "epoch": 0.903997011580127,
+      "grad_norm": 0.27883678674697876,
+      "learning_rate": 0.0002,
+      "loss": 1.827,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9114680612626074,
+      "grad_norm": 0.33874374628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9189391109450877,
+      "grad_norm": 0.3014272153377533,
+      "learning_rate": 0.0002,
+      "loss": 1.7334,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9264101606275682,
+      "grad_norm": 0.3194271922111511,
+      "learning_rate": 0.0002,
+      "loss": 1.8235,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9338812103100486,
+      "grad_norm": 0.3049403429031372,
+      "learning_rate": 0.0002,
+      "loss": 1.7924,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9413522599925289,
+      "grad_norm": 0.30621254444122314,
+      "learning_rate": 0.0002,
+      "loss": 1.7535,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9488233096750094,
+      "grad_norm": 0.28675132989883423,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9562943593574897,
+      "grad_norm": 0.3322032690048218,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9637654090399701,
+      "grad_norm": 0.35408294200897217,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9712364587224505,
+      "grad_norm": 0.36386919021606445,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9787075084049309,
+      "grad_norm": 0.32338324189186096,
+      "learning_rate": 0.0002,
+      "loss": 1.8633,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9861785580874113,
+      "grad_norm": 0.3714013993740082,
+      "learning_rate": 0.0002,
+      "loss": 1.7724,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9936496077698916,
+      "grad_norm": 0.3133082389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.7766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9996264475158759,
+      "eval_loss": 1.8051470518112183,
+      "eval_runtime": 38.6332,
+      "eval_samples_per_second": 13.331,
+      "eval_steps_per_second": 1.682,
+      "step": 1338
+    },
+    {
+      "epoch": 1.001120657452372,
+      "grad_norm": 0.31595754623413086,
+      "learning_rate": 0.0002,
+      "loss": 1.8035,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0085917071348525,
+      "grad_norm": 0.3095700144767761,
+      "learning_rate": 0.0002,
+      "loss": 1.7486,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0160627568173328,
+      "grad_norm": 0.34677496552467346,
+      "learning_rate": 0.0002,
+      "loss": 1.6981,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0235338064998132,
+      "grad_norm": 0.29108840227127075,
+      "learning_rate": 0.0002,
+      "loss": 1.7377,
+      "step": 1370
+    },
+    {
+      "epoch": 1.0310048561822935,
+      "grad_norm": 0.32356950640678406,
+      "learning_rate": 0.0002,
+      "loss": 1.7194,
+      "step": 1380
+    },
+    {
+      "epoch": 1.038475905864774,
+      "grad_norm": 0.4200669229030609,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0459469555472545,
+      "grad_norm": 0.3283711373806,
+      "learning_rate": 0.0002,
+      "loss": 1.797,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0534180052297348,
+      "grad_norm": 0.32898256182670593,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0608890549122152,
+      "grad_norm": 0.38790300488471985,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0683601045946955,
+      "grad_norm": 0.339800089597702,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1430
+    },
+    {
+      "epoch": 1.075831154277176,
+      "grad_norm": 0.3548751175403595,
+      "learning_rate": 0.0002,
+      "loss": 1.7076,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0833022039596563,
+      "grad_norm": 0.35114359855651855,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0907732536421366,
+      "grad_norm": 0.35226720571517944,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0982443033246172,
+      "grad_norm": 0.33665576577186584,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1057153530070976,
+      "grad_norm": 0.363889217376709,
+      "learning_rate": 0.0002,
+      "loss": 1.6699,
+      "step": 1480
+    },
+    {
+      "epoch": 1.113186402689578,
+      "grad_norm": 0.3826201856136322,
+      "learning_rate": 0.0002,
+      "loss": 1.7933,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1206574523720583,
+      "grad_norm": 0.34058740735054016,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1281285020545386,
+      "grad_norm": 0.3462134301662445,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 1510
+    },
+    {
+      "epoch": 1.135599551737019,
+      "grad_norm": 0.3396756052970886,
+      "learning_rate": 0.0002,
+      "loss": 1.7147,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1430706014194993,
+      "grad_norm": 0.32004743814468384,
+      "learning_rate": 0.0002,
+      "loss": 1.7219,
+      "step": 1530
+    },
+    {
+      "epoch": 1.15054165110198,
+      "grad_norm": 0.3397733271121979,
+      "learning_rate": 0.0002,
+      "loss": 1.743,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1580127007844603,
+      "grad_norm": 0.3783262073993683,
+      "learning_rate": 0.0002,
+      "loss": 1.7333,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1654837504669406,
+      "grad_norm": 0.35121291875839233,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 1560
+    },
+    {
+      "epoch": 1.172954800149421,
+      "grad_norm": 0.35816895961761475,
+      "learning_rate": 0.0002,
+      "loss": 1.678,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1804258498319014,
+      "grad_norm": 0.33843839168548584,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1878968995143817,
+      "grad_norm": 0.3371972143650055,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 1590
+    },
+    {
+      "epoch": 1.195367949196862,
+      "grad_norm": 0.36016878485679626,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2028389988793426,
+      "grad_norm": 0.40879473090171814,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 1610
+    },
+    {
+      "epoch": 1.210310048561823,
+      "grad_norm": 0.3216715455055237,
+      "learning_rate": 0.0002,
+      "loss": 1.6955,
+      "step": 1620
+    },
+    {
+      "epoch": 1.2177810982443034,
+      "grad_norm": 0.4482610821723938,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2252521479267837,
+      "grad_norm": 0.3257700502872467,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1640
+    },
+    {
+      "epoch": 1.232723197609264,
+      "grad_norm": 0.38646459579467773,
+      "learning_rate": 0.0002,
+      "loss": 1.7177,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2401942472917444,
+      "grad_norm": 0.4081360697746277,
+      "learning_rate": 0.0002,
+      "loss": 1.7081,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2476652969742248,
+      "grad_norm": 0.4326848089694977,
+      "learning_rate": 0.0002,
+      "loss": 1.7519,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2551363466567054,
+      "grad_norm": 0.346401572227478,
+      "learning_rate": 0.0002,
+      "loss": 1.6752,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2626073963391857,
+      "grad_norm": 0.34536251425743103,
+      "learning_rate": 0.0002,
+      "loss": 1.7425,
+      "step": 1690
+    },
+    {
+      "epoch": 1.270078446021666,
+      "grad_norm": 0.41359591484069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7061,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2775494957041464,
+      "grad_norm": 0.3530874252319336,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2850205453866268,
+      "grad_norm": 0.3702719211578369,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2924915950691072,
+      "grad_norm": 0.3703329563140869,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2999626447515875,
+      "grad_norm": 0.37919729948043823,
+      "learning_rate": 0.0002,
+      "loss": 1.7221,
+      "step": 1740
+    },
+    {
+      "epoch": 1.307433694434068,
+      "grad_norm": 0.32526856660842896,
+      "learning_rate": 0.0002,
+      "loss": 1.7859,
+      "step": 1750
+    },
+    {
+      "epoch": 1.3149047441165485,
+      "grad_norm": 0.36752620339393616,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3223757937990288,
+      "grad_norm": 0.3398192524909973,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3298468434815092,
+      "grad_norm": 0.37435585260391235,
+      "learning_rate": 0.0002,
+      "loss": 1.7492,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3373178931639895,
+      "grad_norm": 0.35793280601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3447889428464699,
+      "grad_norm": 0.35481882095336914,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3522599925289502,
+      "grad_norm": 0.3786393105983734,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3597310422114308,
+      "grad_norm": 0.33245593309402466,
+      "learning_rate": 0.0002,
+      "loss": 1.7169,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3672020918939112,
+      "grad_norm": 0.35388344526290894,
+      "learning_rate": 0.0002,
+      "loss": 1.7577,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3746731415763915,
+      "grad_norm": 0.3695325553417206,
+      "learning_rate": 0.0002,
+      "loss": 1.6968,
+      "step": 1840
+    },
+    {
+      "epoch": 1.382144191258872,
+      "grad_norm": 0.3683604598045349,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3896152409413522,
+      "grad_norm": 0.3753012418746948,
+      "learning_rate": 0.0002,
+      "loss": 1.7878,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3970862906238326,
+      "grad_norm": 0.3331069350242615,
+      "learning_rate": 0.0002,
+      "loss": 1.6969,
+      "step": 1870
+    },
+    {
+      "epoch": 1.404557340306313,
+      "grad_norm": 0.3877500295639038,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4120283899887935,
+      "grad_norm": 0.33525151014328003,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4194994396712737,
+      "grad_norm": 0.3697299659252167,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4269704893537543,
+      "grad_norm": 0.4029286205768585,
+      "learning_rate": 0.0002,
+      "loss": 1.6956,
+      "step": 1910
+    },
+    {
+      "epoch": 1.4344415390362346,
+      "grad_norm": 0.3596203029155731,
+      "learning_rate": 0.0002,
+      "loss": 1.6897,
+      "step": 1920
+    },
+    {
+      "epoch": 1.441912588718715,
+      "grad_norm": 0.450783908367157,
+      "learning_rate": 0.0002,
+      "loss": 1.7139,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4493836384011953,
+      "grad_norm": 0.3651481866836548,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4568546880836757,
+      "grad_norm": 0.3608424663543701,
+      "learning_rate": 0.0002,
+      "loss": 1.6637,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4643257377661563,
+      "grad_norm": 0.39684420824050903,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4717967874486364,
+      "grad_norm": 0.34618663787841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7514,
+      "step": 1970
+    },
+    {
+      "epoch": 1.479267837131117,
+      "grad_norm": 0.4150386452674866,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4867388868135973,
+      "grad_norm": 0.35500776767730713,
+      "learning_rate": 0.0002,
+      "loss": 1.7021,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4942099364960777,
+      "grad_norm": 0.344144344329834,
+      "learning_rate": 0.0002,
+      "loss": 1.7322,
+      "step": 2000
+    },
+    {
+      "epoch": 1.501680986178558,
+      "grad_norm": 0.3340149223804474,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5091520358610384,
+      "grad_norm": 0.37685006856918335,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 2020
+    },
+    {
+      "epoch": 1.516623085543519,
+      "grad_norm": 0.3699876368045807,
+      "learning_rate": 0.0002,
+      "loss": 1.8299,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5240941352259991,
+      "grad_norm": 0.3370307385921478,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5315651849084797,
+      "grad_norm": 0.37780630588531494,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 2050
+    },
+    {
+      "epoch": 1.53903623459096,
+      "grad_norm": 0.370259165763855,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5465072842734404,
+      "grad_norm": 0.3440011441707611,
+      "learning_rate": 0.0002,
+      "loss": 1.7398,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5539783339559208,
+      "grad_norm": 0.40382063388824463,
+      "learning_rate": 0.0002,
+      "loss": 1.7105,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5614493836384011,
+      "grad_norm": 0.38002029061317444,
+      "learning_rate": 0.0002,
+      "loss": 1.7071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5689204333208817,
+      "grad_norm": 0.3658451437950134,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5763914830033618,
+      "grad_norm": 0.354842871427536,
+      "learning_rate": 0.0002,
+      "loss": 1.7598,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5838625326858424,
+      "grad_norm": 0.34735530614852905,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5913335823683228,
+      "grad_norm": 0.377581924200058,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5988046320508031,
+      "grad_norm": 0.41254034638404846,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6062756817332835,
+      "grad_norm": 0.3630715310573578,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6137467314157639,
+      "grad_norm": 0.36980143189430237,
+      "learning_rate": 0.0002,
+      "loss": 1.7531,
+      "step": 2160
+    },
+    {
+      "epoch": 1.6212177810982444,
+      "grad_norm": 0.3634769320487976,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6286888307807246,
+      "grad_norm": 0.3794139623641968,
+      "learning_rate": 0.0002,
+      "loss": 1.6367,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6361598804632052,
+      "grad_norm": 0.359742134809494,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6436309301456855,
+      "grad_norm": 0.3770543932914734,
+      "learning_rate": 0.0002,
+      "loss": 1.7027,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6511019798281659,
+      "grad_norm": 0.3797036409378052,
+      "learning_rate": 0.0002,
+      "loss": 1.784,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6585730295106462,
+      "grad_norm": 0.35622093081474304,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6660440791931266,
+      "grad_norm": 0.34552520513534546,
+      "learning_rate": 0.0002,
+      "loss": 1.6615,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6735151288756072,
+      "grad_norm": 0.379926860332489,
+      "learning_rate": 0.0002,
+      "loss": 1.7522,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6809861785580873,
+      "grad_norm": 0.37083810567855835,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6884572282405679,
+      "grad_norm": 0.42746543884277344,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6959282779230482,
+      "grad_norm": 0.3372884690761566,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7033993276055286,
+      "grad_norm": 0.35220256447792053,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2280
+    },
+    {
+      "epoch": 1.710870377288009,
+      "grad_norm": 0.3659130930900574,
+      "learning_rate": 0.0002,
+      "loss": 1.7154,
+      "step": 2290
+    },
+    {
+      "epoch": 1.7183414269704893,
+      "grad_norm": 0.37629297375679016,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7258124766529699,
+      "grad_norm": 0.36312398314476013,
+      "learning_rate": 0.0002,
+      "loss": 1.7212,
+      "step": 2310
+    },
+    {
+      "epoch": 1.73328352633545,
+      "grad_norm": 0.467709481716156,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7407545760179306,
+      "grad_norm": 0.38685527443885803,
+      "learning_rate": 0.0002,
+      "loss": 1.696,
+      "step": 2330
+    },
+    {
+      "epoch": 1.748225625700411,
+      "grad_norm": 0.3578338325023651,
+      "learning_rate": 0.0002,
+      "loss": 1.7041,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7556966753828913,
+      "grad_norm": 0.36057502031326294,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7631677250653717,
+      "grad_norm": 0.3615196645259857,
+      "learning_rate": 0.0002,
+      "loss": 1.6853,
+      "step": 2360
+    },
+    {
+      "epoch": 1.770638774747852,
+      "grad_norm": 0.4118947684764862,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7781098244303326,
+      "grad_norm": 0.4067276120185852,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7855808741128127,
+      "grad_norm": 0.3979823887348175,
+      "learning_rate": 0.0002,
+      "loss": 1.712,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7930519237952933,
+      "grad_norm": 0.44045883417129517,
+      "learning_rate": 0.0002,
+      "loss": 1.7644,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8005229734777737,
+      "grad_norm": 0.3998069167137146,
+      "learning_rate": 0.0002,
+      "loss": 1.7251,
+      "step": 2410
+    },
+    {
+      "epoch": 1.807994023160254,
+      "grad_norm": 0.3450094759464264,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8154650728427344,
+      "grad_norm": 0.3759009838104248,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2430
+    },
+    {
+      "epoch": 1.8229361225252148,
+      "grad_norm": 0.34347015619277954,
+      "learning_rate": 0.0002,
+      "loss": 1.7706,
+      "step": 2440
+    },
+    {
+      "epoch": 1.8304071722076953,
+      "grad_norm": 0.3511228859424591,
+      "learning_rate": 0.0002,
+      "loss": 1.7345,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8378782218901755,
+      "grad_norm": 0.36853715777397156,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 2460
+    },
+    {
+      "epoch": 1.845349271572656,
+      "grad_norm": 0.40659376978874207,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8528203212551362,
+      "grad_norm": 0.39621320366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8602913709376168,
+      "grad_norm": 0.3753979504108429,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8677624206200971,
+      "grad_norm": 0.3811938464641571,
+      "learning_rate": 0.0002,
+      "loss": 1.6622,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8752334703025775,
+      "grad_norm": 0.3432596027851105,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 2510
+    },
+    {
+      "epoch": 1.882704519985058,
+      "grad_norm": 0.3670712113380432,
+      "learning_rate": 0.0002,
+      "loss": 1.7488,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8901755696675382,
+      "grad_norm": 0.40907177329063416,
+      "learning_rate": 0.0002,
+      "loss": 1.705,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8976466193500188,
+      "grad_norm": 0.3821999728679657,
+      "learning_rate": 0.0002,
+      "loss": 1.7148,
+      "step": 2540
+    },
+    {
+      "epoch": 1.905117669032499,
+      "grad_norm": 0.36173978447914124,
+      "learning_rate": 0.0002,
+      "loss": 1.7934,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9125887187149795,
+      "grad_norm": 0.38990336656570435,
+      "learning_rate": 0.0002,
+      "loss": 1.6939,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9200597683974598,
+      "grad_norm": 0.35242322087287903,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 2570
+    },
+    {
+      "epoch": 1.9275308180799402,
+      "grad_norm": 0.3506428003311157,
+      "learning_rate": 0.0002,
+      "loss": 1.7268,
+      "step": 2580
+    },
+    {
+      "epoch": 1.9350018677624208,
+      "grad_norm": 0.39540135860443115,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2590
+    },
+    {
+      "epoch": 1.942472917444901,
+      "grad_norm": 0.3444725573062897,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9499439671273815,
+      "grad_norm": 0.3963521718978882,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9574150168098616,
+      "grad_norm": 0.3689815402030945,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9648860664923422,
+      "grad_norm": 0.3482626676559448,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9723571161748226,
+      "grad_norm": 0.35832616686820984,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 2640
+    },
+    {
+      "epoch": 1.979828165857303,
+      "grad_norm": 0.4776208996772766,
+      "learning_rate": 0.0002,
+      "loss": 1.6681,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9872992155397835,
+      "grad_norm": 0.32570165395736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6696,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9947702652222636,
+      "grad_norm": 0.3380725085735321,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8046749830245972,
+      "eval_runtime": 38.5096,
+      "eval_samples_per_second": 13.373,
+      "eval_steps_per_second": 1.688,
+      "step": 2677
+    },
+    {
+      "epoch": 2.002241314904744,
+      "grad_norm": 0.36817631125450134,
+      "learning_rate": 0.0002,
+      "loss": 1.7265,
+      "step": 2680
+    },
+    {
+      "epoch": 2.0097123645872244,
+      "grad_norm": 0.4056456685066223,
+      "learning_rate": 0.0002,
+      "loss": 1.548,
+      "step": 2690
+    },
+    {
+      "epoch": 2.017183414269705,
+      "grad_norm": 0.37416863441467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5515,
+      "step": 2700
+    },
+    {
+      "epoch": 2.024654463952185,
+      "grad_norm": 0.4273638427257538,
+      "learning_rate": 0.0002,
+      "loss": 1.5895,
+      "step": 2710
+    },
+    {
+      "epoch": 2.0321255136346656,
+      "grad_norm": 0.36497923731803894,
+      "learning_rate": 0.0002,
+      "loss": 1.5884,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0395965633171462,
+      "grad_norm": 0.5021994113922119,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 2730
+    },
+    {
+      "epoch": 2.0470676129996264,
+      "grad_norm": 0.45896220207214355,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 2740
+    },
+    {
+      "epoch": 2.054538662682107,
+      "grad_norm": 0.3973815143108368,
+      "learning_rate": 0.0002,
+      "loss": 1.6305,
+      "step": 2750
+    },
+    {
+      "epoch": 2.062009712364587,
+      "grad_norm": 0.4521815776824951,
+      "learning_rate": 0.0002,
+      "loss": 1.6301,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0694807620470677,
+      "grad_norm": 0.42775002121925354,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2770
+    },
+    {
+      "epoch": 2.076951811729548,
+      "grad_norm": 0.48158586025238037,
+      "learning_rate": 0.0002,
+      "loss": 1.6491,
+      "step": 2780
+    },
+    {
+      "epoch": 2.0844228614120284,
+      "grad_norm": 0.4612371623516083,
+      "learning_rate": 0.0002,
+      "loss": 1.6301,
+      "step": 2790
+    },
+    {
+      "epoch": 2.091893911094509,
+      "grad_norm": 0.42536866664886475,
+      "learning_rate": 0.0002,
+      "loss": 1.6327,
+      "step": 2800
+    },
+    {
+      "epoch": 2.099364960776989,
+      "grad_norm": 0.48515772819519043,
+      "learning_rate": 0.0002,
+      "loss": 1.651,
+      "step": 2810
+    },
+    {
+      "epoch": 2.1068360104594697,
+      "grad_norm": 0.41418662667274475,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 2820
+    },
+    {
+      "epoch": 2.11430706014195,
+      "grad_norm": 0.4683697819709778,
+      "learning_rate": 0.0002,
+      "loss": 1.6266,
+      "step": 2830
+    },
+    {
+      "epoch": 2.1217781098244304,
+      "grad_norm": 0.4484657049179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6586,
+      "step": 2840
+    },
+    {
+      "epoch": 2.1292491595069105,
+      "grad_norm": 0.6621400713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.6483,
+      "step": 2850
+    },
+    {
+      "epoch": 2.136720209189391,
+      "grad_norm": 0.45074811577796936,
+      "learning_rate": 0.0002,
+      "loss": 1.5755,
+      "step": 2860
+    },
+    {
+      "epoch": 2.1441912588718717,
+      "grad_norm": 0.3513113558292389,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2870
+    },
+    {
+      "epoch": 2.151662308554352,
+      "grad_norm": 0.40411314368247986,
+      "learning_rate": 0.0002,
+      "loss": 1.6081,
+      "step": 2880
+    },
+    {
+      "epoch": 2.1591333582368324,
+      "grad_norm": 0.4121065139770508,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 2890
+    },
+    {
+      "epoch": 2.1666044079193125,
+      "grad_norm": 0.44723689556121826,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 2900
+    },
+    {
+      "epoch": 2.174075457601793,
+      "grad_norm": 0.4226122498512268,
+      "learning_rate": 0.0002,
+      "loss": 1.5699,
+      "step": 2910
+    },
+    {
+      "epoch": 2.1815465072842732,
+      "grad_norm": 0.46617650985717773,
+      "learning_rate": 0.0002,
+      "loss": 1.5652,
+      "step": 2920
+    },
+    {
+      "epoch": 2.189017556966754,
+      "grad_norm": 0.4506422281265259,
+      "learning_rate": 0.0002,
+      "loss": 1.6378,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1964886066492344,
+      "grad_norm": 0.4892672896385193,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 2940
+    },
+    {
+      "epoch": 2.2039596563317145,
+      "grad_norm": 0.44095516204833984,
+      "learning_rate": 0.0002,
+      "loss": 1.6176,
+      "step": 2950
+    },
+    {
+      "epoch": 2.211430706014195,
+      "grad_norm": 0.41522109508514404,
+      "learning_rate": 0.0002,
+      "loss": 1.6058,
+      "step": 2960
+    },
+    {
+      "epoch": 2.2189017556966752,
+      "grad_norm": 0.4860858917236328,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2970
+    },
+    {
+      "epoch": 2.226372805379156,
+      "grad_norm": 0.42662516236305237,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2980
+    },
+    {
+      "epoch": 2.233843855061636,
+      "grad_norm": 0.4390648305416107,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 2990
+    },
+    {
+      "epoch": 2.2413149047441165,
+      "grad_norm": 0.47515565156936646,
+      "learning_rate": 0.0002,
+      "loss": 1.5992,
+      "step": 3000
+    },
+    {
+      "epoch": 2.248785954426597,
+      "grad_norm": 0.4104543924331665,
+      "learning_rate": 0.0002,
+      "loss": 1.5563,
+      "step": 3010
+    },
+    {
+      "epoch": 2.2562570041090773,
+      "grad_norm": 0.4404028654098511,
+      "learning_rate": 0.0002,
+      "loss": 1.6895,
+      "step": 3020
+    },
+    {
+      "epoch": 2.263728053791558,
+      "grad_norm": 0.4717366695404053,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 3030
+    },
+    {
+      "epoch": 2.271199103474038,
+      "grad_norm": 0.48345857858657837,
+      "learning_rate": 0.0002,
+      "loss": 1.7287,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2786701531565186,
+      "grad_norm": 0.5312452912330627,
+      "learning_rate": 0.0002,
+      "loss": 1.681,
+      "step": 3050
+    },
+    {
+      "epoch": 2.2861412028389987,
+      "grad_norm": 0.5073099732398987,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2936122525214793,
+      "grad_norm": 0.5027463436126709,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 3070
+    },
+    {
+      "epoch": 2.30108330220396,
+      "grad_norm": 0.5436304807662964,
+      "learning_rate": 0.0002,
+      "loss": 1.5862,
+      "step": 3080
+    },
+    {
+      "epoch": 2.30855435188644,
+      "grad_norm": 0.4701065123081207,
+      "learning_rate": 0.0002,
+      "loss": 1.5763,
+      "step": 3090
+    },
+    {
+      "epoch": 2.3160254015689206,
+      "grad_norm": 0.46988746523857117,
+      "learning_rate": 0.0002,
+      "loss": 1.6177,
+      "step": 3100
+    },
+    {
+      "epoch": 2.3234964512514007,
+      "grad_norm": 0.45112869143486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6502,
+      "step": 3110
+    },
+    {
+      "epoch": 2.3309675009338813,
+      "grad_norm": 0.5173566937446594,
+      "learning_rate": 0.0002,
+      "loss": 1.6291,
+      "step": 3120
+    },
+    {
+      "epoch": 2.3384385506163614,
+      "grad_norm": 0.40345850586891174,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 3130
+    },
+    {
+      "epoch": 2.345909600298842,
+      "grad_norm": 0.4218924939632416,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 3140
+    },
+    {
+      "epoch": 2.3533806499813226,
+      "grad_norm": 0.41857317090034485,
+      "learning_rate": 0.0002,
+      "loss": 1.6341,
+      "step": 3150
+    },
+    {
+      "epoch": 2.3608516996638027,
+      "grad_norm": 0.4197218418121338,
+      "learning_rate": 0.0002,
+      "loss": 1.6087,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3683227493462833,
+      "grad_norm": 0.4260677397251129,
+      "learning_rate": 0.0002,
+      "loss": 1.6572,
+      "step": 3170
+    },
+    {
+      "epoch": 2.3757937990287634,
+      "grad_norm": 0.4209042191505432,
+      "learning_rate": 0.0002,
+      "loss": 1.6376,
+      "step": 3180
+    },
+    {
+      "epoch": 2.383264848711244,
+      "grad_norm": 0.4092234969139099,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 3190
+    },
+    {
+      "epoch": 2.390735898393724,
+      "grad_norm": 0.4928431510925293,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3982069480762047,
+      "grad_norm": 0.49252402782440186,
+      "learning_rate": 0.0002,
+      "loss": 1.6015,
+      "step": 3210
+    },
+    {
+      "epoch": 2.4056779977586853,
+      "grad_norm": 0.4368397295475006,
+      "learning_rate": 0.0002,
+      "loss": 1.5773,
+      "step": 3220
+    },
+    {
+      "epoch": 2.4131490474411654,
+      "grad_norm": 0.46122390031814575,
+      "learning_rate": 0.0002,
+      "loss": 1.6629,
+      "step": 3230
+    },
+    {
+      "epoch": 2.420620097123646,
+      "grad_norm": 0.4272301197052002,
+      "learning_rate": 0.0002,
+      "loss": 1.6224,
+      "step": 3240
+    },
+    {
+      "epoch": 2.428091146806126,
+      "grad_norm": 0.41480937600135803,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 3250
+    },
+    {
+      "epoch": 2.4355621964886067,
+      "grad_norm": 0.48911941051483154,
+      "learning_rate": 0.0002,
+      "loss": 1.6281,
+      "step": 3260
+    },
+    {
+      "epoch": 2.443033246171087,
+      "grad_norm": 0.4444098472595215,
+      "learning_rate": 0.0002,
+      "loss": 1.6846,
+      "step": 3270
+    },
+    {
+      "epoch": 2.4505042958535674,
+      "grad_norm": 0.5111684799194336,
+      "learning_rate": 0.0002,
+      "loss": 1.6961,
+      "step": 3280
+    },
+    {
+      "epoch": 2.457975345536048,
+      "grad_norm": 0.5058825016021729,
+      "learning_rate": 0.0002,
+      "loss": 1.6152,
+      "step": 3290
+    },
+    {
+      "epoch": 2.465446395218528,
+      "grad_norm": 0.44173210859298706,
+      "learning_rate": 0.0002,
+      "loss": 1.625,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4729174449010087,
+      "grad_norm": 0.4659745991230011,
+      "learning_rate": 0.0002,
+      "loss": 1.6491,
+      "step": 3310
+    },
+    {
+      "epoch": 2.480388494583489,
+      "grad_norm": 0.47237497568130493,
+      "learning_rate": 0.0002,
+      "loss": 1.6114,
+      "step": 3320
+    },
+    {
+      "epoch": 2.4878595442659694,
+      "grad_norm": 0.47303131222724915,
+      "learning_rate": 0.0002,
+      "loss": 1.6193,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4953305939484496,
+      "grad_norm": 0.4522389769554138,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 3340
+    },
+    {
+      "epoch": 2.50280164363093,
+      "grad_norm": 0.4467332363128662,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3350
+    },
+    {
+      "epoch": 2.5102726933134107,
+      "grad_norm": 0.4413762092590332,
+      "learning_rate": 0.0002,
+      "loss": 1.6108,
+      "step": 3360
+    },
+    {
+      "epoch": 2.517743742995891,
+      "grad_norm": 0.495514452457428,
+      "learning_rate": 0.0002,
+      "loss": 1.537,
+      "step": 3370
+    },
+    {
+      "epoch": 2.5252147926783715,
+      "grad_norm": 0.4429773986339569,
+      "learning_rate": 0.0002,
+      "loss": 1.5839,
+      "step": 3380
+    },
+    {
+      "epoch": 2.5326858423608516,
+      "grad_norm": 0.4589079022407532,
+      "learning_rate": 0.0002,
+      "loss": 1.6522,
+      "step": 3390
+    },
+    {
+      "epoch": 2.540156892043332,
+      "grad_norm": 0.4683997333049774,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 3400
+    },
+    {
+      "epoch": 2.5476279417258123,
+      "grad_norm": 0.4651731252670288,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 3410
+    },
+    {
+      "epoch": 2.555098991408293,
+      "grad_norm": 0.45818084478378296,
+      "learning_rate": 0.0002,
+      "loss": 1.5918,
+      "step": 3420
+    },
+    {
+      "epoch": 2.5625700410907735,
+      "grad_norm": 0.45209529995918274,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3430
+    },
+    {
+      "epoch": 2.5700410907732536,
+      "grad_norm": 0.4344733655452728,
+      "learning_rate": 0.0002,
+      "loss": 1.5606,
+      "step": 3440
+    },
+    {
+      "epoch": 2.577512140455734,
+      "grad_norm": 0.47435566782951355,
+      "learning_rate": 0.0002,
+      "loss": 1.6748,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5849831901382143,
+      "grad_norm": 0.43841999769210815,
+      "learning_rate": 0.0002,
+      "loss": 1.6237,
+      "step": 3460
+    },
+    {
+      "epoch": 2.592454239820695,
+      "grad_norm": 0.4323869049549103,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 3470
+    },
+    {
+      "epoch": 2.599925289503175,
+      "grad_norm": 0.44355881214141846,
+      "learning_rate": 0.0002,
+      "loss": 1.5494,
+      "step": 3480
+    },
+    {
+      "epoch": 2.6073963391856556,
+      "grad_norm": 0.45847779512405396,
+      "learning_rate": 0.0002,
+      "loss": 1.665,
+      "step": 3490
+    },
+    {
+      "epoch": 2.614867388868136,
+      "grad_norm": 0.4411061704158783,
+      "learning_rate": 0.0002,
+      "loss": 1.6006,
+      "step": 3500
+    },
+    {
+      "epoch": 2.6223384385506163,
+      "grad_norm": 0.4446796178817749,
+      "learning_rate": 0.0002,
+      "loss": 1.5868,
+      "step": 3510
+    },
+    {
+      "epoch": 2.629809488233097,
+      "grad_norm": 0.41969653964042664,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 3520
+    },
+    {
+      "epoch": 2.637280537915577,
+      "grad_norm": 0.5263747572898865,
+      "learning_rate": 0.0002,
+      "loss": 1.6798,
+      "step": 3530
+    },
+    {
+      "epoch": 2.6447515875980576,
+      "grad_norm": 0.47719451785087585,
+      "learning_rate": 0.0002,
+      "loss": 1.6309,
+      "step": 3540
+    },
+    {
+      "epoch": 2.6522226372805378,
+      "grad_norm": 0.46574118733406067,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 3550
+    },
+    {
+      "epoch": 2.6596936869630183,
+      "grad_norm": 0.46867135167121887,
+      "learning_rate": 0.0002,
+      "loss": 1.618,
+      "step": 3560
+    },
+    {
+      "epoch": 2.667164736645499,
+      "grad_norm": 0.4441198706626892,
+      "learning_rate": 0.0002,
+      "loss": 1.5885,
+      "step": 3570
+    },
+    {
+      "epoch": 2.674635786327979,
+      "grad_norm": 0.4871319830417633,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6821068360104596,
+      "grad_norm": 0.43900373578071594,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6895778856929398,
+      "grad_norm": 0.42509549856185913,
+      "learning_rate": 0.0002,
+      "loss": 1.6071,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6970489353754203,
+      "grad_norm": 0.4691086709499359,
+      "learning_rate": 0.0002,
+      "loss": 1.5651,
+      "step": 3610
+    },
+    {
+      "epoch": 2.7045199850579005,
+      "grad_norm": 0.46318942308425903,
+      "learning_rate": 0.0002,
+      "loss": 1.5491,
+      "step": 3620
+    },
+    {
+      "epoch": 2.711991034740381,
+      "grad_norm": 0.44631096720695496,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 3630
+    },
+    {
+      "epoch": 2.7194620844228616,
+      "grad_norm": 0.42315489053726196,
+      "learning_rate": 0.0002,
+      "loss": 1.6831,
+      "step": 3640
+    },
+    {
+      "epoch": 2.7269331341053418,
+      "grad_norm": 0.4971241056919098,
+      "learning_rate": 0.0002,
+      "loss": 1.6008,
+      "step": 3650
+    },
+    {
+      "epoch": 2.7344041837878224,
+      "grad_norm": 0.4578486382961273,
+      "learning_rate": 0.0002,
+      "loss": 1.6042,
+      "step": 3660
+    },
+    {
+      "epoch": 2.7418752334703025,
+      "grad_norm": 0.46584776043891907,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 3670
+    },
+    {
+      "epoch": 2.749346283152783,
+      "grad_norm": 0.4951731264591217,
+      "learning_rate": 0.0002,
+      "loss": 1.6809,
+      "step": 3680
+    },
+    {
+      "epoch": 2.756817332835263,
+      "grad_norm": 0.4935225546360016,
+      "learning_rate": 0.0002,
+      "loss": 1.6226,
+      "step": 3690
+    },
+    {
+      "epoch": 2.764288382517744,
+      "grad_norm": 0.41805586218833923,
+      "learning_rate": 0.0002,
+      "loss": 1.5878,
+      "step": 3700
+    },
+    {
+      "epoch": 2.7717594322002244,
+      "grad_norm": 0.4417555630207062,
+      "learning_rate": 0.0002,
+      "loss": 1.7173,
+      "step": 3710
+    },
+    {
+      "epoch": 2.7792304818827045,
+      "grad_norm": 0.48229655623435974,
+      "learning_rate": 0.0002,
+      "loss": 1.6398,
+      "step": 3720
+    },
+    {
+      "epoch": 2.786701531565185,
+      "grad_norm": 0.48562315106391907,
+      "learning_rate": 0.0002,
+      "loss": 1.6074,
+      "step": 3730
+    },
+    {
+      "epoch": 2.794172581247665,
+      "grad_norm": 0.4473940432071686,
+      "learning_rate": 0.0002,
+      "loss": 1.607,
+      "step": 3740
+    },
+    {
+      "epoch": 2.801643630930146,
+      "grad_norm": 0.4626813232898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6065,
+      "step": 3750
+    },
+    {
+      "epoch": 2.809114680612626,
+      "grad_norm": 0.4339792728424072,
+      "learning_rate": 0.0002,
+      "loss": 1.6296,
+      "step": 3760
+    },
+    {
+      "epoch": 2.8165857302951065,
+      "grad_norm": 0.5250858068466187,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 3770
+    },
+    {
+      "epoch": 2.824056779977587,
+      "grad_norm": 0.4537523090839386,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 3780
+    },
+    {
+      "epoch": 2.831527829660067,
+      "grad_norm": 0.5646113157272339,
+      "learning_rate": 0.0002,
+      "loss": 1.6535,
+      "step": 3790
+    },
+    {
+      "epoch": 2.8389988793425474,
+      "grad_norm": 0.44243332743644714,
+      "learning_rate": 0.0002,
+      "loss": 1.5712,
+      "step": 3800
+    },
+    {
+      "epoch": 2.846469929025028,
+      "grad_norm": 0.4585791826248169,
+      "learning_rate": 0.0002,
+      "loss": 1.6478,
+      "step": 3810
+    },
+    {
+      "epoch": 2.8539409787075085,
+      "grad_norm": 0.489702045917511,
+      "learning_rate": 0.0002,
+      "loss": 1.6854,
+      "step": 3820
+    },
+    {
+      "epoch": 2.8614120283899886,
+      "grad_norm": 0.502470850944519,
+      "learning_rate": 0.0002,
+      "loss": 1.7066,
+      "step": 3830
+    },
+    {
+      "epoch": 2.8688830780724692,
+      "grad_norm": 0.4395960867404938,
+      "learning_rate": 0.0002,
+      "loss": 1.5785,
+      "step": 3840
+    },
+    {
+      "epoch": 2.87635412775495,
+      "grad_norm": 0.4348670244216919,
+      "learning_rate": 0.0002,
+      "loss": 1.6434,
+      "step": 3850
+    },
+    {
+      "epoch": 2.88382517743743,
+      "grad_norm": 0.48852720856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 3860
+    },
+    {
+      "epoch": 2.89129622711991,
+      "grad_norm": 0.45317450165748596,
+      "learning_rate": 0.0002,
+      "loss": 1.5916,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8987672768023907,
+      "grad_norm": 0.4732758700847626,
+      "learning_rate": 0.0002,
+      "loss": 1.6486,
+      "step": 3880
+    },
+    {
+      "epoch": 2.9062383264848712,
+      "grad_norm": 0.45238012075424194,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 3890
+    },
+    {
+      "epoch": 2.9137093761673514,
+      "grad_norm": 0.48838064074516296,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 3900
+    },
+    {
+      "epoch": 2.921180425849832,
+      "grad_norm": 0.43496349453926086,
+      "learning_rate": 0.0002,
+      "loss": 1.658,
+      "step": 3910
+    },
+    {
+      "epoch": 2.9286514755323125,
+      "grad_norm": 0.47963935136795044,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 3920
+    },
+    {
+      "epoch": 2.9361225252147927,
+      "grad_norm": 0.4544987976551056,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 3930
+    },
+    {
+      "epoch": 2.943593574897273,
+      "grad_norm": 0.4622892141342163,
+      "learning_rate": 0.0002,
+      "loss": 1.6192,
+      "step": 3940
+    },
+    {
+      "epoch": 2.9510646245797534,
+      "grad_norm": 0.47026222944259644,
+      "learning_rate": 0.0002,
+      "loss": 1.6178,
+      "step": 3950
+    },
+    {
+      "epoch": 2.958535674262234,
+      "grad_norm": 0.4549552798271179,
+      "learning_rate": 0.0002,
+      "loss": 1.6612,
+      "step": 3960
+    },
+    {
+      "epoch": 2.966006723944714,
+      "grad_norm": 0.46647515892982483,
+      "learning_rate": 0.0002,
+      "loss": 1.6458,
+      "step": 3970
+    },
+    {
+      "epoch": 2.9734777736271947,
+      "grad_norm": 0.45095112919807434,
+      "learning_rate": 0.0002,
+      "loss": 1.6051,
+      "step": 3980
+    },
+    {
+      "epoch": 2.9809488233096753,
+      "grad_norm": 0.4690017104148865,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 3990
+    },
+    {
+      "epoch": 2.9884198729921554,
+      "grad_norm": 0.4603444039821625,
+      "learning_rate": 0.0002,
+      "loss": 1.6061,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9958909226746355,
+      "grad_norm": 0.4743294417858124,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 4010
+    },
+    {
+      "epoch": 2.999626447515876,
+      "eval_loss": 1.8252571821212769,
+      "eval_runtime": 38.7853,
+      "eval_samples_per_second": 13.278,
+      "eval_steps_per_second": 1.676,
+      "step": 4015
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10704,
+  "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": 1.8582834893134234e+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-0.9-num-6326-sd-42/checkpoint-4015/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-0.9-num-6326-sd-42/checkpoint-4015/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..67c7b4ca126d7b712ad1985ef15ffb29ebe76633
--- /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-0.9-num-6326-sd-42/checkpoint-4015/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fc87da605e94ff0ecd8f5b371302a2d5f8727b77984707a2185ddb447fc3796
+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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d6dd4e080a63c8ce12c05a3186c0385567b67691
--- /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-0.9-num-6326-sd-42/checkpoint-5354/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:023bb3375e82678547f483aeed88017e5e4cb836f95329c3807e2116a94569ef
+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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1e8c7589fe0246488eec927afd0ad93330197dc0
--- /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-0.9-num-6326-sd-42/checkpoint-5354/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6f5d437093ba4fd926c6f2b9af844b0db46d5cffed16d995a7746d6caca17f3
+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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f1f3818fcabdac36ea21169cde6edf57becaba19
--- /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-0.9-num-6326-sd-42/checkpoint-5354/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81343a5862e20c6075e31e0516d064e185be4f7553555b02fb5c78ef8c752e8b
+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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..46ed1c9e06360fb5de0d44678d86c87349076cfb
--- /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-0.9-num-6326-sd-42/checkpoint-5354/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:943c28f99829ff3d008d59ce446faf88e17ac41806214c04b0d404cc44482013
+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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..0fa16c3ae3c5f91f2caed42992c6dc6b9cb0f148
--- /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-0.9-num-6326-sd-42/checkpoint-5354/trainer_state.json
@@ -0,0 +1,3810 @@
+{
+  "best_metric": 1.8046749830245972,
+  "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-0.9-num-6326-sd-42/checkpoint-2677",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 5354,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007471049682480389,
+      "grad_norm": 0.4912872612476349,
+      "learning_rate": 0.0002,
+      "loss": 2.6181,
+      "step": 10
+    },
+    {
+      "epoch": 0.014942099364960777,
+      "grad_norm": 0.4856316149234772,
+      "learning_rate": 0.0002,
+      "loss": 2.2606,
+      "step": 20
+    },
+    {
+      "epoch": 0.022413149047441166,
+      "grad_norm": 0.47683125734329224,
+      "learning_rate": 0.0002,
+      "loss": 2.0957,
+      "step": 30
+    },
+    {
+      "epoch": 0.029884198729921554,
+      "grad_norm": 0.515082597732544,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 40
+    },
+    {
+      "epoch": 0.03735524841240194,
+      "grad_norm": 0.5299215316772461,
+      "learning_rate": 0.0002,
+      "loss": 1.9704,
+      "step": 50
+    },
+    {
+      "epoch": 0.04482629809488233,
+      "grad_norm": 0.4951399862766266,
+      "learning_rate": 0.0002,
+      "loss": 1.9225,
+      "step": 60
+    },
+    {
+      "epoch": 0.05229734777736272,
+      "grad_norm": 0.48079821467399597,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05976839745984311,
+      "grad_norm": 0.49402132630348206,
+      "learning_rate": 0.0002,
+      "loss": 1.9466,
+      "step": 80
+    },
+    {
+      "epoch": 0.0672394471423235,
+      "grad_norm": 0.4778193235397339,
+      "learning_rate": 0.0002,
+      "loss": 1.8691,
+      "step": 90
+    },
+    {
+      "epoch": 0.07471049682480388,
+      "grad_norm": 0.42472657561302185,
+      "learning_rate": 0.0002,
+      "loss": 1.8455,
+      "step": 100
+    },
+    {
+      "epoch": 0.08218154650728428,
+      "grad_norm": 0.4433092474937439,
+      "learning_rate": 0.0002,
+      "loss": 1.8744,
+      "step": 110
+    },
+    {
+      "epoch": 0.08965259618976466,
+      "grad_norm": 0.4472862780094147,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 120
+    },
+    {
+      "epoch": 0.09712364587224505,
+      "grad_norm": 0.42596298456192017,
+      "learning_rate": 0.0002,
+      "loss": 1.9256,
+      "step": 130
+    },
+    {
+      "epoch": 0.10459469555472543,
+      "grad_norm": 0.46645811200141907,
+      "learning_rate": 0.0002,
+      "loss": 1.8015,
+      "step": 140
+    },
+    {
+      "epoch": 0.11206574523720583,
+      "grad_norm": 0.41041234135627747,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 150
+    },
+    {
+      "epoch": 0.11953679491968622,
+      "grad_norm": 0.5329819917678833,
+      "learning_rate": 0.0002,
+      "loss": 1.8276,
+      "step": 160
+    },
+    {
+      "epoch": 0.1270078446021666,
+      "grad_norm": 0.4065922200679779,
+      "learning_rate": 0.0002,
+      "loss": 1.8118,
+      "step": 170
+    },
+    {
+      "epoch": 0.134478894284647,
+      "grad_norm": 0.38406994938850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8559,
+      "step": 180
+    },
+    {
+      "epoch": 0.14194994396712737,
+      "grad_norm": 0.4246881306171417,
+      "learning_rate": 0.0002,
+      "loss": 1.8647,
+      "step": 190
+    },
+    {
+      "epoch": 0.14942099364960776,
+      "grad_norm": 0.35136649012565613,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 200
+    },
+    {
+      "epoch": 0.15689204333208817,
+      "grad_norm": 0.43252742290496826,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.16436309301456856,
+      "grad_norm": 0.39236941933631897,
+      "learning_rate": 0.0002,
+      "loss": 1.7823,
+      "step": 220
+    },
+    {
+      "epoch": 0.17183414269704894,
+      "grad_norm": 0.3748249113559723,
+      "learning_rate": 0.0002,
+      "loss": 1.818,
+      "step": 230
+    },
+    {
+      "epoch": 0.17930519237952933,
+      "grad_norm": 0.6432855725288391,
+      "learning_rate": 0.0002,
+      "loss": 1.866,
+      "step": 240
+    },
+    {
+      "epoch": 0.1867762420620097,
+      "grad_norm": 0.34874802827835083,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 250
+    },
+    {
+      "epoch": 0.1942472917444901,
+      "grad_norm": 0.3721984326839447,
+      "learning_rate": 0.0002,
+      "loss": 1.79,
+      "step": 260
+    },
+    {
+      "epoch": 0.20171834142697048,
+      "grad_norm": 0.4339311420917511,
+      "learning_rate": 0.0002,
+      "loss": 1.8464,
+      "step": 270
+    },
+    {
+      "epoch": 0.20918939110945087,
+      "grad_norm": 0.4018215537071228,
+      "learning_rate": 0.0002,
+      "loss": 1.8665,
+      "step": 280
+    },
+    {
+      "epoch": 0.21666044079193125,
+      "grad_norm": 0.3278839886188507,
+      "learning_rate": 0.0002,
+      "loss": 1.8048,
+      "step": 290
+    },
+    {
+      "epoch": 0.22413149047441167,
+      "grad_norm": 0.36146077513694763,
+      "learning_rate": 0.0002,
+      "loss": 1.7395,
+      "step": 300
+    },
+    {
+      "epoch": 0.23160254015689205,
+      "grad_norm": 0.38175010681152344,
+      "learning_rate": 0.0002,
+      "loss": 1.7916,
+      "step": 310
+    },
+    {
+      "epoch": 0.23907358983937244,
+      "grad_norm": 0.44776618480682373,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 320
+    },
+    {
+      "epoch": 0.24654463952185282,
+      "grad_norm": 0.3933652937412262,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 330
+    },
+    {
+      "epoch": 0.2540156892043332,
+      "grad_norm": 0.3515005111694336,
+      "learning_rate": 0.0002,
+      "loss": 1.8393,
+      "step": 340
+    },
+    {
+      "epoch": 0.2614867388868136,
+      "grad_norm": 0.6683304309844971,
+      "learning_rate": 0.0002,
+      "loss": 1.8653,
+      "step": 350
+    },
+    {
+      "epoch": 0.268957788569294,
+      "grad_norm": 0.37093454599380493,
+      "learning_rate": 0.0002,
+      "loss": 1.8797,
+      "step": 360
+    },
+    {
+      "epoch": 0.2764288382517744,
+      "grad_norm": 0.3450651168823242,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 370
+    },
+    {
+      "epoch": 0.28389988793425475,
+      "grad_norm": 0.5140917301177979,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 380
+    },
+    {
+      "epoch": 0.29137093761673516,
+      "grad_norm": 0.32885563373565674,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 390
+    },
+    {
+      "epoch": 0.2988419872992155,
+      "grad_norm": 0.33962297439575195,
+      "learning_rate": 0.0002,
+      "loss": 1.8174,
+      "step": 400
+    },
+    {
+      "epoch": 0.30631303698169593,
+      "grad_norm": 0.3723141849040985,
+      "learning_rate": 0.0002,
+      "loss": 1.7467,
+      "step": 410
+    },
+    {
+      "epoch": 0.31378408666417634,
+      "grad_norm": 0.37173134088516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8459,
+      "step": 420
+    },
+    {
+      "epoch": 0.3212551363466567,
+      "grad_norm": 0.33736956119537354,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 430
+    },
+    {
+      "epoch": 0.3287261860291371,
+      "grad_norm": 0.3602448105812073,
+      "learning_rate": 0.0002,
+      "loss": 1.8367,
+      "step": 440
+    },
+    {
+      "epoch": 0.33619723571161747,
+      "grad_norm": 0.3569699227809906,
+      "learning_rate": 0.0002,
+      "loss": 1.8058,
+      "step": 450
+    },
+    {
+      "epoch": 0.3436682853940979,
+      "grad_norm": 0.31009167432785034,
+      "learning_rate": 0.0002,
+      "loss": 1.8086,
+      "step": 460
+    },
+    {
+      "epoch": 0.35113933507657824,
+      "grad_norm": 0.5278693437576294,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 470
+    },
+    {
+      "epoch": 0.35861038475905865,
+      "grad_norm": 0.3587537109851837,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 480
+    },
+    {
+      "epoch": 0.366081434441539,
+      "grad_norm": 0.3859670162200928,
+      "learning_rate": 0.0002,
+      "loss": 1.8046,
+      "step": 490
+    },
+    {
+      "epoch": 0.3735524841240194,
+      "grad_norm": 0.395913690328598,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 500
+    },
+    {
+      "epoch": 0.38102353380649984,
+      "grad_norm": 0.35052940249443054,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 510
+    },
+    {
+      "epoch": 0.3884945834889802,
+      "grad_norm": 0.2979494333267212,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 520
+    },
+    {
+      "epoch": 0.3959656331714606,
+      "grad_norm": 0.3062683343887329,
+      "learning_rate": 0.0002,
+      "loss": 1.8641,
+      "step": 530
+    },
+    {
+      "epoch": 0.40343668285394096,
+      "grad_norm": 0.3172847330570221,
+      "learning_rate": 0.0002,
+      "loss": 1.7651,
+      "step": 540
+    },
+    {
+      "epoch": 0.4109077325364214,
+      "grad_norm": 0.360435426235199,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 550
+    },
+    {
+      "epoch": 0.41837878221890173,
+      "grad_norm": 0.3427872359752655,
+      "learning_rate": 0.0002,
+      "loss": 1.9054,
+      "step": 560
+    },
+    {
+      "epoch": 0.42584983190138215,
+      "grad_norm": 0.34036558866500854,
+      "learning_rate": 0.0002,
+      "loss": 1.7562,
+      "step": 570
+    },
+    {
+      "epoch": 0.4333208815838625,
+      "grad_norm": 0.3365345299243927,
+      "learning_rate": 0.0002,
+      "loss": 1.7254,
+      "step": 580
+    },
+    {
+      "epoch": 0.4407919312663429,
+      "grad_norm": 0.35619041323661804,
+      "learning_rate": 0.0002,
+      "loss": 1.8328,
+      "step": 590
+    },
+    {
+      "epoch": 0.44826298094882333,
+      "grad_norm": 0.3569088280200958,
+      "learning_rate": 0.0002,
+      "loss": 1.8114,
+      "step": 600
+    },
+    {
+      "epoch": 0.4557340306313037,
+      "grad_norm": 0.3581278622150421,
+      "learning_rate": 0.0002,
+      "loss": 1.8599,
+      "step": 610
+    },
+    {
+      "epoch": 0.4632050803137841,
+      "grad_norm": 0.43197110295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.7078,
+      "step": 620
+    },
+    {
+      "epoch": 0.47067612999626446,
+      "grad_norm": 0.33966198563575745,
+      "learning_rate": 0.0002,
+      "loss": 1.8257,
+      "step": 630
+    },
+    {
+      "epoch": 0.47814717967874487,
+      "grad_norm": 0.3343866467475891,
+      "learning_rate": 0.0002,
+      "loss": 1.7528,
+      "step": 640
+    },
+    {
+      "epoch": 0.48561822936122523,
+      "grad_norm": 0.33878564834594727,
+      "learning_rate": 0.0002,
+      "loss": 1.8191,
+      "step": 650
+    },
+    {
+      "epoch": 0.49308927904370564,
+      "grad_norm": 0.387195885181427,
+      "learning_rate": 0.0002,
+      "loss": 1.8801,
+      "step": 660
+    },
+    {
+      "epoch": 0.500560328726186,
+      "grad_norm": 0.3755440413951874,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 670
+    },
+    {
+      "epoch": 0.5080313784086664,
+      "grad_norm": 0.3272816836833954,
+      "learning_rate": 0.0002,
+      "loss": 1.8057,
+      "step": 680
+    },
+    {
+      "epoch": 0.5155024280911468,
+      "grad_norm": 0.36063864827156067,
+      "learning_rate": 0.0002,
+      "loss": 1.8156,
+      "step": 690
+    },
+    {
+      "epoch": 0.5229734777736272,
+      "grad_norm": 0.35317373275756836,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 700
+    },
+    {
+      "epoch": 0.5304445274561076,
+      "grad_norm": 0.3561195433139801,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 710
+    },
+    {
+      "epoch": 0.537915577138588,
+      "grad_norm": 0.31124624609947205,
+      "learning_rate": 0.0002,
+      "loss": 1.8149,
+      "step": 720
+    },
+    {
+      "epoch": 0.5453866268210683,
+      "grad_norm": 0.3294544517993927,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 730
+    },
+    {
+      "epoch": 0.5528576765035488,
+      "grad_norm": 0.31933900713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.8027,
+      "step": 740
+    },
+    {
+      "epoch": 0.5603287261860291,
+      "grad_norm": 0.3226020634174347,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 750
+    },
+    {
+      "epoch": 0.5677997758685095,
+      "grad_norm": 0.3147525489330292,
+      "learning_rate": 0.0002,
+      "loss": 1.7862,
+      "step": 760
+    },
+    {
+      "epoch": 0.57527082555099,
+      "grad_norm": 0.32234328985214233,
+      "learning_rate": 0.0002,
+      "loss": 1.9028,
+      "step": 770
+    },
+    {
+      "epoch": 0.5827418752334703,
+      "grad_norm": 0.3258664309978485,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 780
+    },
+    {
+      "epoch": 0.5902129249159507,
+      "grad_norm": 0.3166961967945099,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 790
+    },
+    {
+      "epoch": 0.597683974598431,
+      "grad_norm": 0.35621458292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 800
+    },
+    {
+      "epoch": 0.6051550242809115,
+      "grad_norm": 0.3236999213695526,
+      "learning_rate": 0.0002,
+      "loss": 1.8313,
+      "step": 810
+    },
+    {
+      "epoch": 0.6126260739633919,
+      "grad_norm": 0.2892923653125763,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 820
+    },
+    {
+      "epoch": 0.6200971236458722,
+      "grad_norm": 0.4098321497440338,
+      "learning_rate": 0.0002,
+      "loss": 1.8709,
+      "step": 830
+    },
+    {
+      "epoch": 0.6275681733283527,
+      "grad_norm": 0.3337118923664093,
+      "learning_rate": 0.0002,
+      "loss": 1.7637,
+      "step": 840
+    },
+    {
+      "epoch": 0.635039223010833,
+      "grad_norm": 0.30416029691696167,
+      "learning_rate": 0.0002,
+      "loss": 1.7375,
+      "step": 850
+    },
+    {
+      "epoch": 0.6425102726933134,
+      "grad_norm": 0.3361026346683502,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 860
+    },
+    {
+      "epoch": 0.6499813223757938,
+      "grad_norm": 0.3537365198135376,
+      "learning_rate": 0.0002,
+      "loss": 1.732,
+      "step": 870
+    },
+    {
+      "epoch": 0.6574523720582742,
+      "grad_norm": 0.33854469656944275,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 880
+    },
+    {
+      "epoch": 0.6649234217407546,
+      "grad_norm": 0.3332272469997406,
+      "learning_rate": 0.0002,
+      "loss": 1.7561,
+      "step": 890
+    },
+    {
+      "epoch": 0.6723944714232349,
+      "grad_norm": 0.34954726696014404,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 900
+    },
+    {
+      "epoch": 0.6798655211057153,
+      "grad_norm": 0.2921750247478485,
+      "learning_rate": 0.0002,
+      "loss": 1.7917,
+      "step": 910
+    },
+    {
+      "epoch": 0.6873365707881958,
+      "grad_norm": 0.30508682131767273,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 920
+    },
+    {
+      "epoch": 0.6948076204706761,
+      "grad_norm": 0.32268425822257996,
+      "learning_rate": 0.0002,
+      "loss": 1.8082,
+      "step": 930
+    },
+    {
+      "epoch": 0.7022786701531565,
+      "grad_norm": 0.2844390869140625,
+      "learning_rate": 0.0002,
+      "loss": 1.8283,
+      "step": 940
+    },
+    {
+      "epoch": 0.709749719835637,
+      "grad_norm": 0.31263890862464905,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 950
+    },
+    {
+      "epoch": 0.7172207695181173,
+      "grad_norm": 0.3626808822154999,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 960
+    },
+    {
+      "epoch": 0.7246918192005977,
+      "grad_norm": 0.3322749733924866,
+      "learning_rate": 0.0002,
+      "loss": 1.853,
+      "step": 970
+    },
+    {
+      "epoch": 0.732162868883078,
+      "grad_norm": 0.29177871346473694,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 980
+    },
+    {
+      "epoch": 0.7396339185655585,
+      "grad_norm": 0.35405513644218445,
+      "learning_rate": 0.0002,
+      "loss": 1.8447,
+      "step": 990
+    },
+    {
+      "epoch": 0.7471049682480388,
+      "grad_norm": 0.39318400621414185,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7545760179305192,
+      "grad_norm": 0.29401418566703796,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7620470676129997,
+      "grad_norm": 0.3271748721599579,
+      "learning_rate": 0.0002,
+      "loss": 1.7649,
+      "step": 1020
+    },
+    {
+      "epoch": 0.76951811729548,
+      "grad_norm": 0.30883970856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7769891669779604,
+      "grad_norm": 0.3411838412284851,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7844602166604407,
+      "grad_norm": 0.30608129501342773,
+      "learning_rate": 0.0002,
+      "loss": 1.829,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7919312663429212,
+      "grad_norm": 0.30899080634117126,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7994023160254016,
+      "grad_norm": 0.3160453140735626,
+      "learning_rate": 0.0002,
+      "loss": 1.7625,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8068733657078819,
+      "grad_norm": 0.30947187542915344,
+      "learning_rate": 0.0002,
+      "loss": 1.8452,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8143444153903624,
+      "grad_norm": 0.3103134036064148,
+      "learning_rate": 0.0002,
+      "loss": 1.7418,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8218154650728428,
+      "grad_norm": 0.31771138310432434,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8292865147553231,
+      "grad_norm": 0.5860997438430786,
+      "learning_rate": 0.0002,
+      "loss": 1.7918,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8367575644378035,
+      "grad_norm": 0.3230148255825043,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8442286141202839,
+      "grad_norm": 0.29611510038375854,
+      "learning_rate": 0.0002,
+      "loss": 1.8478,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8516996638027643,
+      "grad_norm": 0.3373654782772064,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8591707134852447,
+      "grad_norm": 0.3474279046058655,
+      "learning_rate": 0.0002,
+      "loss": 1.7997,
+      "step": 1150
+    },
+    {
+      "epoch": 0.866641763167725,
+      "grad_norm": 0.35057875514030457,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8741128128502055,
+      "grad_norm": 0.39537495374679565,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8815838625326858,
+      "grad_norm": 0.3714233636856079,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8890549122151662,
+      "grad_norm": 0.2950296998023987,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8965259618976467,
+      "grad_norm": 0.38182979822158813,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 1200
+    },
+    {
+      "epoch": 0.903997011580127,
+      "grad_norm": 0.27883678674697876,
+      "learning_rate": 0.0002,
+      "loss": 1.827,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9114680612626074,
+      "grad_norm": 0.33874374628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9189391109450877,
+      "grad_norm": 0.3014272153377533,
+      "learning_rate": 0.0002,
+      "loss": 1.7334,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9264101606275682,
+      "grad_norm": 0.3194271922111511,
+      "learning_rate": 0.0002,
+      "loss": 1.8235,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9338812103100486,
+      "grad_norm": 0.3049403429031372,
+      "learning_rate": 0.0002,
+      "loss": 1.7924,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9413522599925289,
+      "grad_norm": 0.30621254444122314,
+      "learning_rate": 0.0002,
+      "loss": 1.7535,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9488233096750094,
+      "grad_norm": 0.28675132989883423,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9562943593574897,
+      "grad_norm": 0.3322032690048218,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9637654090399701,
+      "grad_norm": 0.35408294200897217,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9712364587224505,
+      "grad_norm": 0.36386919021606445,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9787075084049309,
+      "grad_norm": 0.32338324189186096,
+      "learning_rate": 0.0002,
+      "loss": 1.8633,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9861785580874113,
+      "grad_norm": 0.3714013993740082,
+      "learning_rate": 0.0002,
+      "loss": 1.7724,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9936496077698916,
+      "grad_norm": 0.3133082389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.7766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9996264475158759,
+      "eval_loss": 1.8051470518112183,
+      "eval_runtime": 38.6332,
+      "eval_samples_per_second": 13.331,
+      "eval_steps_per_second": 1.682,
+      "step": 1338
+    },
+    {
+      "epoch": 1.001120657452372,
+      "grad_norm": 0.31595754623413086,
+      "learning_rate": 0.0002,
+      "loss": 1.8035,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0085917071348525,
+      "grad_norm": 0.3095700144767761,
+      "learning_rate": 0.0002,
+      "loss": 1.7486,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0160627568173328,
+      "grad_norm": 0.34677496552467346,
+      "learning_rate": 0.0002,
+      "loss": 1.6981,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0235338064998132,
+      "grad_norm": 0.29108840227127075,
+      "learning_rate": 0.0002,
+      "loss": 1.7377,
+      "step": 1370
+    },
+    {
+      "epoch": 1.0310048561822935,
+      "grad_norm": 0.32356950640678406,
+      "learning_rate": 0.0002,
+      "loss": 1.7194,
+      "step": 1380
+    },
+    {
+      "epoch": 1.038475905864774,
+      "grad_norm": 0.4200669229030609,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0459469555472545,
+      "grad_norm": 0.3283711373806,
+      "learning_rate": 0.0002,
+      "loss": 1.797,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0534180052297348,
+      "grad_norm": 0.32898256182670593,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0608890549122152,
+      "grad_norm": 0.38790300488471985,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0683601045946955,
+      "grad_norm": 0.339800089597702,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1430
+    },
+    {
+      "epoch": 1.075831154277176,
+      "grad_norm": 0.3548751175403595,
+      "learning_rate": 0.0002,
+      "loss": 1.7076,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0833022039596563,
+      "grad_norm": 0.35114359855651855,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0907732536421366,
+      "grad_norm": 0.35226720571517944,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0982443033246172,
+      "grad_norm": 0.33665576577186584,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1057153530070976,
+      "grad_norm": 0.363889217376709,
+      "learning_rate": 0.0002,
+      "loss": 1.6699,
+      "step": 1480
+    },
+    {
+      "epoch": 1.113186402689578,
+      "grad_norm": 0.3826201856136322,
+      "learning_rate": 0.0002,
+      "loss": 1.7933,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1206574523720583,
+      "grad_norm": 0.34058740735054016,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1281285020545386,
+      "grad_norm": 0.3462134301662445,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 1510
+    },
+    {
+      "epoch": 1.135599551737019,
+      "grad_norm": 0.3396756052970886,
+      "learning_rate": 0.0002,
+      "loss": 1.7147,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1430706014194993,
+      "grad_norm": 0.32004743814468384,
+      "learning_rate": 0.0002,
+      "loss": 1.7219,
+      "step": 1530
+    },
+    {
+      "epoch": 1.15054165110198,
+      "grad_norm": 0.3397733271121979,
+      "learning_rate": 0.0002,
+      "loss": 1.743,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1580127007844603,
+      "grad_norm": 0.3783262073993683,
+      "learning_rate": 0.0002,
+      "loss": 1.7333,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1654837504669406,
+      "grad_norm": 0.35121291875839233,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 1560
+    },
+    {
+      "epoch": 1.172954800149421,
+      "grad_norm": 0.35816895961761475,
+      "learning_rate": 0.0002,
+      "loss": 1.678,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1804258498319014,
+      "grad_norm": 0.33843839168548584,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1878968995143817,
+      "grad_norm": 0.3371972143650055,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 1590
+    },
+    {
+      "epoch": 1.195367949196862,
+      "grad_norm": 0.36016878485679626,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2028389988793426,
+      "grad_norm": 0.40879473090171814,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 1610
+    },
+    {
+      "epoch": 1.210310048561823,
+      "grad_norm": 0.3216715455055237,
+      "learning_rate": 0.0002,
+      "loss": 1.6955,
+      "step": 1620
+    },
+    {
+      "epoch": 1.2177810982443034,
+      "grad_norm": 0.4482610821723938,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2252521479267837,
+      "grad_norm": 0.3257700502872467,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1640
+    },
+    {
+      "epoch": 1.232723197609264,
+      "grad_norm": 0.38646459579467773,
+      "learning_rate": 0.0002,
+      "loss": 1.7177,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2401942472917444,
+      "grad_norm": 0.4081360697746277,
+      "learning_rate": 0.0002,
+      "loss": 1.7081,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2476652969742248,
+      "grad_norm": 0.4326848089694977,
+      "learning_rate": 0.0002,
+      "loss": 1.7519,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2551363466567054,
+      "grad_norm": 0.346401572227478,
+      "learning_rate": 0.0002,
+      "loss": 1.6752,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2626073963391857,
+      "grad_norm": 0.34536251425743103,
+      "learning_rate": 0.0002,
+      "loss": 1.7425,
+      "step": 1690
+    },
+    {
+      "epoch": 1.270078446021666,
+      "grad_norm": 0.41359591484069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7061,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2775494957041464,
+      "grad_norm": 0.3530874252319336,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2850205453866268,
+      "grad_norm": 0.3702719211578369,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2924915950691072,
+      "grad_norm": 0.3703329563140869,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2999626447515875,
+      "grad_norm": 0.37919729948043823,
+      "learning_rate": 0.0002,
+      "loss": 1.7221,
+      "step": 1740
+    },
+    {
+      "epoch": 1.307433694434068,
+      "grad_norm": 0.32526856660842896,
+      "learning_rate": 0.0002,
+      "loss": 1.7859,
+      "step": 1750
+    },
+    {
+      "epoch": 1.3149047441165485,
+      "grad_norm": 0.36752620339393616,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3223757937990288,
+      "grad_norm": 0.3398192524909973,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3298468434815092,
+      "grad_norm": 0.37435585260391235,
+      "learning_rate": 0.0002,
+      "loss": 1.7492,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3373178931639895,
+      "grad_norm": 0.35793280601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3447889428464699,
+      "grad_norm": 0.35481882095336914,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3522599925289502,
+      "grad_norm": 0.3786393105983734,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3597310422114308,
+      "grad_norm": 0.33245593309402466,
+      "learning_rate": 0.0002,
+      "loss": 1.7169,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3672020918939112,
+      "grad_norm": 0.35388344526290894,
+      "learning_rate": 0.0002,
+      "loss": 1.7577,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3746731415763915,
+      "grad_norm": 0.3695325553417206,
+      "learning_rate": 0.0002,
+      "loss": 1.6968,
+      "step": 1840
+    },
+    {
+      "epoch": 1.382144191258872,
+      "grad_norm": 0.3683604598045349,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3896152409413522,
+      "grad_norm": 0.3753012418746948,
+      "learning_rate": 0.0002,
+      "loss": 1.7878,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3970862906238326,
+      "grad_norm": 0.3331069350242615,
+      "learning_rate": 0.0002,
+      "loss": 1.6969,
+      "step": 1870
+    },
+    {
+      "epoch": 1.404557340306313,
+      "grad_norm": 0.3877500295639038,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4120283899887935,
+      "grad_norm": 0.33525151014328003,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4194994396712737,
+      "grad_norm": 0.3697299659252167,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4269704893537543,
+      "grad_norm": 0.4029286205768585,
+      "learning_rate": 0.0002,
+      "loss": 1.6956,
+      "step": 1910
+    },
+    {
+      "epoch": 1.4344415390362346,
+      "grad_norm": 0.3596203029155731,
+      "learning_rate": 0.0002,
+      "loss": 1.6897,
+      "step": 1920
+    },
+    {
+      "epoch": 1.441912588718715,
+      "grad_norm": 0.450783908367157,
+      "learning_rate": 0.0002,
+      "loss": 1.7139,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4493836384011953,
+      "grad_norm": 0.3651481866836548,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4568546880836757,
+      "grad_norm": 0.3608424663543701,
+      "learning_rate": 0.0002,
+      "loss": 1.6637,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4643257377661563,
+      "grad_norm": 0.39684420824050903,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4717967874486364,
+      "grad_norm": 0.34618663787841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7514,
+      "step": 1970
+    },
+    {
+      "epoch": 1.479267837131117,
+      "grad_norm": 0.4150386452674866,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4867388868135973,
+      "grad_norm": 0.35500776767730713,
+      "learning_rate": 0.0002,
+      "loss": 1.7021,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4942099364960777,
+      "grad_norm": 0.344144344329834,
+      "learning_rate": 0.0002,
+      "loss": 1.7322,
+      "step": 2000
+    },
+    {
+      "epoch": 1.501680986178558,
+      "grad_norm": 0.3340149223804474,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5091520358610384,
+      "grad_norm": 0.37685006856918335,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 2020
+    },
+    {
+      "epoch": 1.516623085543519,
+      "grad_norm": 0.3699876368045807,
+      "learning_rate": 0.0002,
+      "loss": 1.8299,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5240941352259991,
+      "grad_norm": 0.3370307385921478,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5315651849084797,
+      "grad_norm": 0.37780630588531494,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 2050
+    },
+    {
+      "epoch": 1.53903623459096,
+      "grad_norm": 0.370259165763855,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5465072842734404,
+      "grad_norm": 0.3440011441707611,
+      "learning_rate": 0.0002,
+      "loss": 1.7398,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5539783339559208,
+      "grad_norm": 0.40382063388824463,
+      "learning_rate": 0.0002,
+      "loss": 1.7105,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5614493836384011,
+      "grad_norm": 0.38002029061317444,
+      "learning_rate": 0.0002,
+      "loss": 1.7071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5689204333208817,
+      "grad_norm": 0.3658451437950134,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5763914830033618,
+      "grad_norm": 0.354842871427536,
+      "learning_rate": 0.0002,
+      "loss": 1.7598,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5838625326858424,
+      "grad_norm": 0.34735530614852905,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5913335823683228,
+      "grad_norm": 0.377581924200058,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5988046320508031,
+      "grad_norm": 0.41254034638404846,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6062756817332835,
+      "grad_norm": 0.3630715310573578,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6137467314157639,
+      "grad_norm": 0.36980143189430237,
+      "learning_rate": 0.0002,
+      "loss": 1.7531,
+      "step": 2160
+    },
+    {
+      "epoch": 1.6212177810982444,
+      "grad_norm": 0.3634769320487976,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6286888307807246,
+      "grad_norm": 0.3794139623641968,
+      "learning_rate": 0.0002,
+      "loss": 1.6367,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6361598804632052,
+      "grad_norm": 0.359742134809494,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6436309301456855,
+      "grad_norm": 0.3770543932914734,
+      "learning_rate": 0.0002,
+      "loss": 1.7027,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6511019798281659,
+      "grad_norm": 0.3797036409378052,
+      "learning_rate": 0.0002,
+      "loss": 1.784,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6585730295106462,
+      "grad_norm": 0.35622093081474304,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6660440791931266,
+      "grad_norm": 0.34552520513534546,
+      "learning_rate": 0.0002,
+      "loss": 1.6615,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6735151288756072,
+      "grad_norm": 0.379926860332489,
+      "learning_rate": 0.0002,
+      "loss": 1.7522,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6809861785580873,
+      "grad_norm": 0.37083810567855835,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6884572282405679,
+      "grad_norm": 0.42746543884277344,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6959282779230482,
+      "grad_norm": 0.3372884690761566,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7033993276055286,
+      "grad_norm": 0.35220256447792053,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2280
+    },
+    {
+      "epoch": 1.710870377288009,
+      "grad_norm": 0.3659130930900574,
+      "learning_rate": 0.0002,
+      "loss": 1.7154,
+      "step": 2290
+    },
+    {
+      "epoch": 1.7183414269704893,
+      "grad_norm": 0.37629297375679016,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7258124766529699,
+      "grad_norm": 0.36312398314476013,
+      "learning_rate": 0.0002,
+      "loss": 1.7212,
+      "step": 2310
+    },
+    {
+      "epoch": 1.73328352633545,
+      "grad_norm": 0.467709481716156,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7407545760179306,
+      "grad_norm": 0.38685527443885803,
+      "learning_rate": 0.0002,
+      "loss": 1.696,
+      "step": 2330
+    },
+    {
+      "epoch": 1.748225625700411,
+      "grad_norm": 0.3578338325023651,
+      "learning_rate": 0.0002,
+      "loss": 1.7041,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7556966753828913,
+      "grad_norm": 0.36057502031326294,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7631677250653717,
+      "grad_norm": 0.3615196645259857,
+      "learning_rate": 0.0002,
+      "loss": 1.6853,
+      "step": 2360
+    },
+    {
+      "epoch": 1.770638774747852,
+      "grad_norm": 0.4118947684764862,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7781098244303326,
+      "grad_norm": 0.4067276120185852,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7855808741128127,
+      "grad_norm": 0.3979823887348175,
+      "learning_rate": 0.0002,
+      "loss": 1.712,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7930519237952933,
+      "grad_norm": 0.44045883417129517,
+      "learning_rate": 0.0002,
+      "loss": 1.7644,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8005229734777737,
+      "grad_norm": 0.3998069167137146,
+      "learning_rate": 0.0002,
+      "loss": 1.7251,
+      "step": 2410
+    },
+    {
+      "epoch": 1.807994023160254,
+      "grad_norm": 0.3450094759464264,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8154650728427344,
+      "grad_norm": 0.3759009838104248,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2430
+    },
+    {
+      "epoch": 1.8229361225252148,
+      "grad_norm": 0.34347015619277954,
+      "learning_rate": 0.0002,
+      "loss": 1.7706,
+      "step": 2440
+    },
+    {
+      "epoch": 1.8304071722076953,
+      "grad_norm": 0.3511228859424591,
+      "learning_rate": 0.0002,
+      "loss": 1.7345,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8378782218901755,
+      "grad_norm": 0.36853715777397156,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 2460
+    },
+    {
+      "epoch": 1.845349271572656,
+      "grad_norm": 0.40659376978874207,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8528203212551362,
+      "grad_norm": 0.39621320366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8602913709376168,
+      "grad_norm": 0.3753979504108429,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8677624206200971,
+      "grad_norm": 0.3811938464641571,
+      "learning_rate": 0.0002,
+      "loss": 1.6622,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8752334703025775,
+      "grad_norm": 0.3432596027851105,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 2510
+    },
+    {
+      "epoch": 1.882704519985058,
+      "grad_norm": 0.3670712113380432,
+      "learning_rate": 0.0002,
+      "loss": 1.7488,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8901755696675382,
+      "grad_norm": 0.40907177329063416,
+      "learning_rate": 0.0002,
+      "loss": 1.705,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8976466193500188,
+      "grad_norm": 0.3821999728679657,
+      "learning_rate": 0.0002,
+      "loss": 1.7148,
+      "step": 2540
+    },
+    {
+      "epoch": 1.905117669032499,
+      "grad_norm": 0.36173978447914124,
+      "learning_rate": 0.0002,
+      "loss": 1.7934,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9125887187149795,
+      "grad_norm": 0.38990336656570435,
+      "learning_rate": 0.0002,
+      "loss": 1.6939,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9200597683974598,
+      "grad_norm": 0.35242322087287903,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 2570
+    },
+    {
+      "epoch": 1.9275308180799402,
+      "grad_norm": 0.3506428003311157,
+      "learning_rate": 0.0002,
+      "loss": 1.7268,
+      "step": 2580
+    },
+    {
+      "epoch": 1.9350018677624208,
+      "grad_norm": 0.39540135860443115,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2590
+    },
+    {
+      "epoch": 1.942472917444901,
+      "grad_norm": 0.3444725573062897,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9499439671273815,
+      "grad_norm": 0.3963521718978882,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9574150168098616,
+      "grad_norm": 0.3689815402030945,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9648860664923422,
+      "grad_norm": 0.3482626676559448,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9723571161748226,
+      "grad_norm": 0.35832616686820984,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 2640
+    },
+    {
+      "epoch": 1.979828165857303,
+      "grad_norm": 0.4776208996772766,
+      "learning_rate": 0.0002,
+      "loss": 1.6681,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9872992155397835,
+      "grad_norm": 0.32570165395736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6696,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9947702652222636,
+      "grad_norm": 0.3380725085735321,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8046749830245972,
+      "eval_runtime": 38.5096,
+      "eval_samples_per_second": 13.373,
+      "eval_steps_per_second": 1.688,
+      "step": 2677
+    },
+    {
+      "epoch": 2.002241314904744,
+      "grad_norm": 0.36817631125450134,
+      "learning_rate": 0.0002,
+      "loss": 1.7265,
+      "step": 2680
+    },
+    {
+      "epoch": 2.0097123645872244,
+      "grad_norm": 0.4056456685066223,
+      "learning_rate": 0.0002,
+      "loss": 1.548,
+      "step": 2690
+    },
+    {
+      "epoch": 2.017183414269705,
+      "grad_norm": 0.37416863441467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5515,
+      "step": 2700
+    },
+    {
+      "epoch": 2.024654463952185,
+      "grad_norm": 0.4273638427257538,
+      "learning_rate": 0.0002,
+      "loss": 1.5895,
+      "step": 2710
+    },
+    {
+      "epoch": 2.0321255136346656,
+      "grad_norm": 0.36497923731803894,
+      "learning_rate": 0.0002,
+      "loss": 1.5884,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0395965633171462,
+      "grad_norm": 0.5021994113922119,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 2730
+    },
+    {
+      "epoch": 2.0470676129996264,
+      "grad_norm": 0.45896220207214355,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 2740
+    },
+    {
+      "epoch": 2.054538662682107,
+      "grad_norm": 0.3973815143108368,
+      "learning_rate": 0.0002,
+      "loss": 1.6305,
+      "step": 2750
+    },
+    {
+      "epoch": 2.062009712364587,
+      "grad_norm": 0.4521815776824951,
+      "learning_rate": 0.0002,
+      "loss": 1.6301,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0694807620470677,
+      "grad_norm": 0.42775002121925354,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2770
+    },
+    {
+      "epoch": 2.076951811729548,
+      "grad_norm": 0.48158586025238037,
+      "learning_rate": 0.0002,
+      "loss": 1.6491,
+      "step": 2780
+    },
+    {
+      "epoch": 2.0844228614120284,
+      "grad_norm": 0.4612371623516083,
+      "learning_rate": 0.0002,
+      "loss": 1.6301,
+      "step": 2790
+    },
+    {
+      "epoch": 2.091893911094509,
+      "grad_norm": 0.42536866664886475,
+      "learning_rate": 0.0002,
+      "loss": 1.6327,
+      "step": 2800
+    },
+    {
+      "epoch": 2.099364960776989,
+      "grad_norm": 0.48515772819519043,
+      "learning_rate": 0.0002,
+      "loss": 1.651,
+      "step": 2810
+    },
+    {
+      "epoch": 2.1068360104594697,
+      "grad_norm": 0.41418662667274475,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 2820
+    },
+    {
+      "epoch": 2.11430706014195,
+      "grad_norm": 0.4683697819709778,
+      "learning_rate": 0.0002,
+      "loss": 1.6266,
+      "step": 2830
+    },
+    {
+      "epoch": 2.1217781098244304,
+      "grad_norm": 0.4484657049179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6586,
+      "step": 2840
+    },
+    {
+      "epoch": 2.1292491595069105,
+      "grad_norm": 0.6621400713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.6483,
+      "step": 2850
+    },
+    {
+      "epoch": 2.136720209189391,
+      "grad_norm": 0.45074811577796936,
+      "learning_rate": 0.0002,
+      "loss": 1.5755,
+      "step": 2860
+    },
+    {
+      "epoch": 2.1441912588718717,
+      "grad_norm": 0.3513113558292389,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2870
+    },
+    {
+      "epoch": 2.151662308554352,
+      "grad_norm": 0.40411314368247986,
+      "learning_rate": 0.0002,
+      "loss": 1.6081,
+      "step": 2880
+    },
+    {
+      "epoch": 2.1591333582368324,
+      "grad_norm": 0.4121065139770508,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 2890
+    },
+    {
+      "epoch": 2.1666044079193125,
+      "grad_norm": 0.44723689556121826,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 2900
+    },
+    {
+      "epoch": 2.174075457601793,
+      "grad_norm": 0.4226122498512268,
+      "learning_rate": 0.0002,
+      "loss": 1.5699,
+      "step": 2910
+    },
+    {
+      "epoch": 2.1815465072842732,
+      "grad_norm": 0.46617650985717773,
+      "learning_rate": 0.0002,
+      "loss": 1.5652,
+      "step": 2920
+    },
+    {
+      "epoch": 2.189017556966754,
+      "grad_norm": 0.4506422281265259,
+      "learning_rate": 0.0002,
+      "loss": 1.6378,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1964886066492344,
+      "grad_norm": 0.4892672896385193,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 2940
+    },
+    {
+      "epoch": 2.2039596563317145,
+      "grad_norm": 0.44095516204833984,
+      "learning_rate": 0.0002,
+      "loss": 1.6176,
+      "step": 2950
+    },
+    {
+      "epoch": 2.211430706014195,
+      "grad_norm": 0.41522109508514404,
+      "learning_rate": 0.0002,
+      "loss": 1.6058,
+      "step": 2960
+    },
+    {
+      "epoch": 2.2189017556966752,
+      "grad_norm": 0.4860858917236328,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2970
+    },
+    {
+      "epoch": 2.226372805379156,
+      "grad_norm": 0.42662516236305237,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2980
+    },
+    {
+      "epoch": 2.233843855061636,
+      "grad_norm": 0.4390648305416107,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 2990
+    },
+    {
+      "epoch": 2.2413149047441165,
+      "grad_norm": 0.47515565156936646,
+      "learning_rate": 0.0002,
+      "loss": 1.5992,
+      "step": 3000
+    },
+    {
+      "epoch": 2.248785954426597,
+      "grad_norm": 0.4104543924331665,
+      "learning_rate": 0.0002,
+      "loss": 1.5563,
+      "step": 3010
+    },
+    {
+      "epoch": 2.2562570041090773,
+      "grad_norm": 0.4404028654098511,
+      "learning_rate": 0.0002,
+      "loss": 1.6895,
+      "step": 3020
+    },
+    {
+      "epoch": 2.263728053791558,
+      "grad_norm": 0.4717366695404053,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 3030
+    },
+    {
+      "epoch": 2.271199103474038,
+      "grad_norm": 0.48345857858657837,
+      "learning_rate": 0.0002,
+      "loss": 1.7287,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2786701531565186,
+      "grad_norm": 0.5312452912330627,
+      "learning_rate": 0.0002,
+      "loss": 1.681,
+      "step": 3050
+    },
+    {
+      "epoch": 2.2861412028389987,
+      "grad_norm": 0.5073099732398987,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2936122525214793,
+      "grad_norm": 0.5027463436126709,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 3070
+    },
+    {
+      "epoch": 2.30108330220396,
+      "grad_norm": 0.5436304807662964,
+      "learning_rate": 0.0002,
+      "loss": 1.5862,
+      "step": 3080
+    },
+    {
+      "epoch": 2.30855435188644,
+      "grad_norm": 0.4701065123081207,
+      "learning_rate": 0.0002,
+      "loss": 1.5763,
+      "step": 3090
+    },
+    {
+      "epoch": 2.3160254015689206,
+      "grad_norm": 0.46988746523857117,
+      "learning_rate": 0.0002,
+      "loss": 1.6177,
+      "step": 3100
+    },
+    {
+      "epoch": 2.3234964512514007,
+      "grad_norm": 0.45112869143486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6502,
+      "step": 3110
+    },
+    {
+      "epoch": 2.3309675009338813,
+      "grad_norm": 0.5173566937446594,
+      "learning_rate": 0.0002,
+      "loss": 1.6291,
+      "step": 3120
+    },
+    {
+      "epoch": 2.3384385506163614,
+      "grad_norm": 0.40345850586891174,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 3130
+    },
+    {
+      "epoch": 2.345909600298842,
+      "grad_norm": 0.4218924939632416,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 3140
+    },
+    {
+      "epoch": 2.3533806499813226,
+      "grad_norm": 0.41857317090034485,
+      "learning_rate": 0.0002,
+      "loss": 1.6341,
+      "step": 3150
+    },
+    {
+      "epoch": 2.3608516996638027,
+      "grad_norm": 0.4197218418121338,
+      "learning_rate": 0.0002,
+      "loss": 1.6087,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3683227493462833,
+      "grad_norm": 0.4260677397251129,
+      "learning_rate": 0.0002,
+      "loss": 1.6572,
+      "step": 3170
+    },
+    {
+      "epoch": 2.3757937990287634,
+      "grad_norm": 0.4209042191505432,
+      "learning_rate": 0.0002,
+      "loss": 1.6376,
+      "step": 3180
+    },
+    {
+      "epoch": 2.383264848711244,
+      "grad_norm": 0.4092234969139099,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 3190
+    },
+    {
+      "epoch": 2.390735898393724,
+      "grad_norm": 0.4928431510925293,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3982069480762047,
+      "grad_norm": 0.49252402782440186,
+      "learning_rate": 0.0002,
+      "loss": 1.6015,
+      "step": 3210
+    },
+    {
+      "epoch": 2.4056779977586853,
+      "grad_norm": 0.4368397295475006,
+      "learning_rate": 0.0002,
+      "loss": 1.5773,
+      "step": 3220
+    },
+    {
+      "epoch": 2.4131490474411654,
+      "grad_norm": 0.46122390031814575,
+      "learning_rate": 0.0002,
+      "loss": 1.6629,
+      "step": 3230
+    },
+    {
+      "epoch": 2.420620097123646,
+      "grad_norm": 0.4272301197052002,
+      "learning_rate": 0.0002,
+      "loss": 1.6224,
+      "step": 3240
+    },
+    {
+      "epoch": 2.428091146806126,
+      "grad_norm": 0.41480937600135803,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 3250
+    },
+    {
+      "epoch": 2.4355621964886067,
+      "grad_norm": 0.48911941051483154,
+      "learning_rate": 0.0002,
+      "loss": 1.6281,
+      "step": 3260
+    },
+    {
+      "epoch": 2.443033246171087,
+      "grad_norm": 0.4444098472595215,
+      "learning_rate": 0.0002,
+      "loss": 1.6846,
+      "step": 3270
+    },
+    {
+      "epoch": 2.4505042958535674,
+      "grad_norm": 0.5111684799194336,
+      "learning_rate": 0.0002,
+      "loss": 1.6961,
+      "step": 3280
+    },
+    {
+      "epoch": 2.457975345536048,
+      "grad_norm": 0.5058825016021729,
+      "learning_rate": 0.0002,
+      "loss": 1.6152,
+      "step": 3290
+    },
+    {
+      "epoch": 2.465446395218528,
+      "grad_norm": 0.44173210859298706,
+      "learning_rate": 0.0002,
+      "loss": 1.625,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4729174449010087,
+      "grad_norm": 0.4659745991230011,
+      "learning_rate": 0.0002,
+      "loss": 1.6491,
+      "step": 3310
+    },
+    {
+      "epoch": 2.480388494583489,
+      "grad_norm": 0.47237497568130493,
+      "learning_rate": 0.0002,
+      "loss": 1.6114,
+      "step": 3320
+    },
+    {
+      "epoch": 2.4878595442659694,
+      "grad_norm": 0.47303131222724915,
+      "learning_rate": 0.0002,
+      "loss": 1.6193,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4953305939484496,
+      "grad_norm": 0.4522389769554138,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 3340
+    },
+    {
+      "epoch": 2.50280164363093,
+      "grad_norm": 0.4467332363128662,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3350
+    },
+    {
+      "epoch": 2.5102726933134107,
+      "grad_norm": 0.4413762092590332,
+      "learning_rate": 0.0002,
+      "loss": 1.6108,
+      "step": 3360
+    },
+    {
+      "epoch": 2.517743742995891,
+      "grad_norm": 0.495514452457428,
+      "learning_rate": 0.0002,
+      "loss": 1.537,
+      "step": 3370
+    },
+    {
+      "epoch": 2.5252147926783715,
+      "grad_norm": 0.4429773986339569,
+      "learning_rate": 0.0002,
+      "loss": 1.5839,
+      "step": 3380
+    },
+    {
+      "epoch": 2.5326858423608516,
+      "grad_norm": 0.4589079022407532,
+      "learning_rate": 0.0002,
+      "loss": 1.6522,
+      "step": 3390
+    },
+    {
+      "epoch": 2.540156892043332,
+      "grad_norm": 0.4683997333049774,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 3400
+    },
+    {
+      "epoch": 2.5476279417258123,
+      "grad_norm": 0.4651731252670288,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 3410
+    },
+    {
+      "epoch": 2.555098991408293,
+      "grad_norm": 0.45818084478378296,
+      "learning_rate": 0.0002,
+      "loss": 1.5918,
+      "step": 3420
+    },
+    {
+      "epoch": 2.5625700410907735,
+      "grad_norm": 0.45209529995918274,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3430
+    },
+    {
+      "epoch": 2.5700410907732536,
+      "grad_norm": 0.4344733655452728,
+      "learning_rate": 0.0002,
+      "loss": 1.5606,
+      "step": 3440
+    },
+    {
+      "epoch": 2.577512140455734,
+      "grad_norm": 0.47435566782951355,
+      "learning_rate": 0.0002,
+      "loss": 1.6748,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5849831901382143,
+      "grad_norm": 0.43841999769210815,
+      "learning_rate": 0.0002,
+      "loss": 1.6237,
+      "step": 3460
+    },
+    {
+      "epoch": 2.592454239820695,
+      "grad_norm": 0.4323869049549103,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 3470
+    },
+    {
+      "epoch": 2.599925289503175,
+      "grad_norm": 0.44355881214141846,
+      "learning_rate": 0.0002,
+      "loss": 1.5494,
+      "step": 3480
+    },
+    {
+      "epoch": 2.6073963391856556,
+      "grad_norm": 0.45847779512405396,
+      "learning_rate": 0.0002,
+      "loss": 1.665,
+      "step": 3490
+    },
+    {
+      "epoch": 2.614867388868136,
+      "grad_norm": 0.4411061704158783,
+      "learning_rate": 0.0002,
+      "loss": 1.6006,
+      "step": 3500
+    },
+    {
+      "epoch": 2.6223384385506163,
+      "grad_norm": 0.4446796178817749,
+      "learning_rate": 0.0002,
+      "loss": 1.5868,
+      "step": 3510
+    },
+    {
+      "epoch": 2.629809488233097,
+      "grad_norm": 0.41969653964042664,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 3520
+    },
+    {
+      "epoch": 2.637280537915577,
+      "grad_norm": 0.5263747572898865,
+      "learning_rate": 0.0002,
+      "loss": 1.6798,
+      "step": 3530
+    },
+    {
+      "epoch": 2.6447515875980576,
+      "grad_norm": 0.47719451785087585,
+      "learning_rate": 0.0002,
+      "loss": 1.6309,
+      "step": 3540
+    },
+    {
+      "epoch": 2.6522226372805378,
+      "grad_norm": 0.46574118733406067,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 3550
+    },
+    {
+      "epoch": 2.6596936869630183,
+      "grad_norm": 0.46867135167121887,
+      "learning_rate": 0.0002,
+      "loss": 1.618,
+      "step": 3560
+    },
+    {
+      "epoch": 2.667164736645499,
+      "grad_norm": 0.4441198706626892,
+      "learning_rate": 0.0002,
+      "loss": 1.5885,
+      "step": 3570
+    },
+    {
+      "epoch": 2.674635786327979,
+      "grad_norm": 0.4871319830417633,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6821068360104596,
+      "grad_norm": 0.43900373578071594,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6895778856929398,
+      "grad_norm": 0.42509549856185913,
+      "learning_rate": 0.0002,
+      "loss": 1.6071,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6970489353754203,
+      "grad_norm": 0.4691086709499359,
+      "learning_rate": 0.0002,
+      "loss": 1.5651,
+      "step": 3610
+    },
+    {
+      "epoch": 2.7045199850579005,
+      "grad_norm": 0.46318942308425903,
+      "learning_rate": 0.0002,
+      "loss": 1.5491,
+      "step": 3620
+    },
+    {
+      "epoch": 2.711991034740381,
+      "grad_norm": 0.44631096720695496,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 3630
+    },
+    {
+      "epoch": 2.7194620844228616,
+      "grad_norm": 0.42315489053726196,
+      "learning_rate": 0.0002,
+      "loss": 1.6831,
+      "step": 3640
+    },
+    {
+      "epoch": 2.7269331341053418,
+      "grad_norm": 0.4971241056919098,
+      "learning_rate": 0.0002,
+      "loss": 1.6008,
+      "step": 3650
+    },
+    {
+      "epoch": 2.7344041837878224,
+      "grad_norm": 0.4578486382961273,
+      "learning_rate": 0.0002,
+      "loss": 1.6042,
+      "step": 3660
+    },
+    {
+      "epoch": 2.7418752334703025,
+      "grad_norm": 0.46584776043891907,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 3670
+    },
+    {
+      "epoch": 2.749346283152783,
+      "grad_norm": 0.4951731264591217,
+      "learning_rate": 0.0002,
+      "loss": 1.6809,
+      "step": 3680
+    },
+    {
+      "epoch": 2.756817332835263,
+      "grad_norm": 0.4935225546360016,
+      "learning_rate": 0.0002,
+      "loss": 1.6226,
+      "step": 3690
+    },
+    {
+      "epoch": 2.764288382517744,
+      "grad_norm": 0.41805586218833923,
+      "learning_rate": 0.0002,
+      "loss": 1.5878,
+      "step": 3700
+    },
+    {
+      "epoch": 2.7717594322002244,
+      "grad_norm": 0.4417555630207062,
+      "learning_rate": 0.0002,
+      "loss": 1.7173,
+      "step": 3710
+    },
+    {
+      "epoch": 2.7792304818827045,
+      "grad_norm": 0.48229655623435974,
+      "learning_rate": 0.0002,
+      "loss": 1.6398,
+      "step": 3720
+    },
+    {
+      "epoch": 2.786701531565185,
+      "grad_norm": 0.48562315106391907,
+      "learning_rate": 0.0002,
+      "loss": 1.6074,
+      "step": 3730
+    },
+    {
+      "epoch": 2.794172581247665,
+      "grad_norm": 0.4473940432071686,
+      "learning_rate": 0.0002,
+      "loss": 1.607,
+      "step": 3740
+    },
+    {
+      "epoch": 2.801643630930146,
+      "grad_norm": 0.4626813232898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6065,
+      "step": 3750
+    },
+    {
+      "epoch": 2.809114680612626,
+      "grad_norm": 0.4339792728424072,
+      "learning_rate": 0.0002,
+      "loss": 1.6296,
+      "step": 3760
+    },
+    {
+      "epoch": 2.8165857302951065,
+      "grad_norm": 0.5250858068466187,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 3770
+    },
+    {
+      "epoch": 2.824056779977587,
+      "grad_norm": 0.4537523090839386,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 3780
+    },
+    {
+      "epoch": 2.831527829660067,
+      "grad_norm": 0.5646113157272339,
+      "learning_rate": 0.0002,
+      "loss": 1.6535,
+      "step": 3790
+    },
+    {
+      "epoch": 2.8389988793425474,
+      "grad_norm": 0.44243332743644714,
+      "learning_rate": 0.0002,
+      "loss": 1.5712,
+      "step": 3800
+    },
+    {
+      "epoch": 2.846469929025028,
+      "grad_norm": 0.4585791826248169,
+      "learning_rate": 0.0002,
+      "loss": 1.6478,
+      "step": 3810
+    },
+    {
+      "epoch": 2.8539409787075085,
+      "grad_norm": 0.489702045917511,
+      "learning_rate": 0.0002,
+      "loss": 1.6854,
+      "step": 3820
+    },
+    {
+      "epoch": 2.8614120283899886,
+      "grad_norm": 0.502470850944519,
+      "learning_rate": 0.0002,
+      "loss": 1.7066,
+      "step": 3830
+    },
+    {
+      "epoch": 2.8688830780724692,
+      "grad_norm": 0.4395960867404938,
+      "learning_rate": 0.0002,
+      "loss": 1.5785,
+      "step": 3840
+    },
+    {
+      "epoch": 2.87635412775495,
+      "grad_norm": 0.4348670244216919,
+      "learning_rate": 0.0002,
+      "loss": 1.6434,
+      "step": 3850
+    },
+    {
+      "epoch": 2.88382517743743,
+      "grad_norm": 0.48852720856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 3860
+    },
+    {
+      "epoch": 2.89129622711991,
+      "grad_norm": 0.45317450165748596,
+      "learning_rate": 0.0002,
+      "loss": 1.5916,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8987672768023907,
+      "grad_norm": 0.4732758700847626,
+      "learning_rate": 0.0002,
+      "loss": 1.6486,
+      "step": 3880
+    },
+    {
+      "epoch": 2.9062383264848712,
+      "grad_norm": 0.45238012075424194,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 3890
+    },
+    {
+      "epoch": 2.9137093761673514,
+      "grad_norm": 0.48838064074516296,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 3900
+    },
+    {
+      "epoch": 2.921180425849832,
+      "grad_norm": 0.43496349453926086,
+      "learning_rate": 0.0002,
+      "loss": 1.658,
+      "step": 3910
+    },
+    {
+      "epoch": 2.9286514755323125,
+      "grad_norm": 0.47963935136795044,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 3920
+    },
+    {
+      "epoch": 2.9361225252147927,
+      "grad_norm": 0.4544987976551056,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 3930
+    },
+    {
+      "epoch": 2.943593574897273,
+      "grad_norm": 0.4622892141342163,
+      "learning_rate": 0.0002,
+      "loss": 1.6192,
+      "step": 3940
+    },
+    {
+      "epoch": 2.9510646245797534,
+      "grad_norm": 0.47026222944259644,
+      "learning_rate": 0.0002,
+      "loss": 1.6178,
+      "step": 3950
+    },
+    {
+      "epoch": 2.958535674262234,
+      "grad_norm": 0.4549552798271179,
+      "learning_rate": 0.0002,
+      "loss": 1.6612,
+      "step": 3960
+    },
+    {
+      "epoch": 2.966006723944714,
+      "grad_norm": 0.46647515892982483,
+      "learning_rate": 0.0002,
+      "loss": 1.6458,
+      "step": 3970
+    },
+    {
+      "epoch": 2.9734777736271947,
+      "grad_norm": 0.45095112919807434,
+      "learning_rate": 0.0002,
+      "loss": 1.6051,
+      "step": 3980
+    },
+    {
+      "epoch": 2.9809488233096753,
+      "grad_norm": 0.4690017104148865,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 3990
+    },
+    {
+      "epoch": 2.9884198729921554,
+      "grad_norm": 0.4603444039821625,
+      "learning_rate": 0.0002,
+      "loss": 1.6061,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9958909226746355,
+      "grad_norm": 0.4743294417858124,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 4010
+    },
+    {
+      "epoch": 2.999626447515876,
+      "eval_loss": 1.8252571821212769,
+      "eval_runtime": 38.7853,
+      "eval_samples_per_second": 13.278,
+      "eval_steps_per_second": 1.676,
+      "step": 4015
+    },
+    {
+      "epoch": 3.003361972357116,
+      "grad_norm": 0.4919724464416504,
+      "learning_rate": 0.0002,
+      "loss": 1.6512,
+      "step": 4020
+    },
+    {
+      "epoch": 3.0108330220395967,
+      "grad_norm": 0.4747185707092285,
+      "learning_rate": 0.0002,
+      "loss": 1.5354,
+      "step": 4030
+    },
+    {
+      "epoch": 3.018304071722077,
+      "grad_norm": 0.4797595143318176,
+      "learning_rate": 0.0002,
+      "loss": 1.568,
+      "step": 4040
+    },
+    {
+      "epoch": 3.0257751214045574,
+      "grad_norm": 0.5450999140739441,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 4050
+    },
+    {
+      "epoch": 3.0332461710870375,
+      "grad_norm": 0.49058812856674194,
+      "learning_rate": 0.0002,
+      "loss": 1.5065,
+      "step": 4060
+    },
+    {
+      "epoch": 3.040717220769518,
+      "grad_norm": 0.5219563841819763,
+      "learning_rate": 0.0002,
+      "loss": 1.4884,
+      "step": 4070
+    },
+    {
+      "epoch": 3.0481882704519987,
+      "grad_norm": 0.515628457069397,
+      "learning_rate": 0.0002,
+      "loss": 1.4742,
+      "step": 4080
+    },
+    {
+      "epoch": 3.055659320134479,
+      "grad_norm": 0.6145984530448914,
+      "learning_rate": 0.0002,
+      "loss": 1.5313,
+      "step": 4090
+    },
+    {
+      "epoch": 3.0631303698169594,
+      "grad_norm": 0.6067144274711609,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 4100
+    },
+    {
+      "epoch": 3.0706014194994395,
+      "grad_norm": 0.5773133039474487,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 4110
+    },
+    {
+      "epoch": 3.07807246918192,
+      "grad_norm": 0.6894241571426392,
+      "learning_rate": 0.0002,
+      "loss": 1.5374,
+      "step": 4120
+    },
+    {
+      "epoch": 3.0855435188644003,
+      "grad_norm": 0.6422514915466309,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 4130
+    },
+    {
+      "epoch": 3.093014568546881,
+      "grad_norm": 0.6119855046272278,
+      "learning_rate": 0.0002,
+      "loss": 1.4724,
+      "step": 4140
+    },
+    {
+      "epoch": 3.1004856182293614,
+      "grad_norm": 0.5847280025482178,
+      "learning_rate": 0.0002,
+      "loss": 1.5361,
+      "step": 4150
+    },
+    {
+      "epoch": 3.1079566679118416,
+      "grad_norm": 0.5401515960693359,
+      "learning_rate": 0.0002,
+      "loss": 1.5151,
+      "step": 4160
+    },
+    {
+      "epoch": 3.115427717594322,
+      "grad_norm": 0.6501587629318237,
+      "learning_rate": 0.0002,
+      "loss": 1.502,
+      "step": 4170
+    },
+    {
+      "epoch": 3.1228987672768023,
+      "grad_norm": 0.5988039374351501,
+      "learning_rate": 0.0002,
+      "loss": 1.4952,
+      "step": 4180
+    },
+    {
+      "epoch": 3.130369816959283,
+      "grad_norm": 0.4982665181159973,
+      "learning_rate": 0.0002,
+      "loss": 1.5287,
+      "step": 4190
+    },
+    {
+      "epoch": 3.137840866641763,
+      "grad_norm": 0.5548039078712463,
+      "learning_rate": 0.0002,
+      "loss": 1.5078,
+      "step": 4200
+    },
+    {
+      "epoch": 3.1453119163242436,
+      "grad_norm": 0.5920777320861816,
+      "learning_rate": 0.0002,
+      "loss": 1.4904,
+      "step": 4210
+    },
+    {
+      "epoch": 3.152782966006724,
+      "grad_norm": 0.6965190172195435,
+      "learning_rate": 0.0002,
+      "loss": 1.442,
+      "step": 4220
+    },
+    {
+      "epoch": 3.1602540156892043,
+      "grad_norm": 0.5196244716644287,
+      "learning_rate": 0.0002,
+      "loss": 1.557,
+      "step": 4230
+    },
+    {
+      "epoch": 3.167725065371685,
+      "grad_norm": 0.6942682266235352,
+      "learning_rate": 0.0002,
+      "loss": 1.5706,
+      "step": 4240
+    },
+    {
+      "epoch": 3.175196115054165,
+      "grad_norm": 0.5765156149864197,
+      "learning_rate": 0.0002,
+      "loss": 1.5407,
+      "step": 4250
+    },
+    {
+      "epoch": 3.1826671647366456,
+      "grad_norm": 0.5801976919174194,
+      "learning_rate": 0.0002,
+      "loss": 1.4963,
+      "step": 4260
+    },
+    {
+      "epoch": 3.1901382144191257,
+      "grad_norm": 0.6260752081871033,
+      "learning_rate": 0.0002,
+      "loss": 1.4988,
+      "step": 4270
+    },
+    {
+      "epoch": 3.1976092641016063,
+      "grad_norm": 0.6610770225524902,
+      "learning_rate": 0.0002,
+      "loss": 1.5074,
+      "step": 4280
+    },
+    {
+      "epoch": 3.205080313784087,
+      "grad_norm": 0.5762143135070801,
+      "learning_rate": 0.0002,
+      "loss": 1.4657,
+      "step": 4290
+    },
+    {
+      "epoch": 3.212551363466567,
+      "grad_norm": 0.5926990509033203,
+      "learning_rate": 0.0002,
+      "loss": 1.5181,
+      "step": 4300
+    },
+    {
+      "epoch": 3.2200224131490476,
+      "grad_norm": 0.7373854517936707,
+      "learning_rate": 0.0002,
+      "loss": 1.5492,
+      "step": 4310
+    },
+    {
+      "epoch": 3.2274934628315277,
+      "grad_norm": 0.5963311195373535,
+      "learning_rate": 0.0002,
+      "loss": 1.4648,
+      "step": 4320
+    },
+    {
+      "epoch": 3.2349645125140083,
+      "grad_norm": 0.5754616856575012,
+      "learning_rate": 0.0002,
+      "loss": 1.5262,
+      "step": 4330
+    },
+    {
+      "epoch": 3.2424355621964884,
+      "grad_norm": 0.6116095781326294,
+      "learning_rate": 0.0002,
+      "loss": 1.4767,
+      "step": 4340
+    },
+    {
+      "epoch": 3.249906611878969,
+      "grad_norm": 0.6001536846160889,
+      "learning_rate": 0.0002,
+      "loss": 1.5008,
+      "step": 4350
+    },
+    {
+      "epoch": 3.257377661561449,
+      "grad_norm": 0.5270227789878845,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 4360
+    },
+    {
+      "epoch": 3.2648487112439297,
+      "grad_norm": 0.6666602492332458,
+      "learning_rate": 0.0002,
+      "loss": 1.5235,
+      "step": 4370
+    },
+    {
+      "epoch": 3.2723197609264103,
+      "grad_norm": 0.520310640335083,
+      "learning_rate": 0.0002,
+      "loss": 1.5665,
+      "step": 4380
+    },
+    {
+      "epoch": 3.2797908106088904,
+      "grad_norm": 0.5165975093841553,
+      "learning_rate": 0.0002,
+      "loss": 1.542,
+      "step": 4390
+    },
+    {
+      "epoch": 3.287261860291371,
+      "grad_norm": 0.6080228686332703,
+      "learning_rate": 0.0002,
+      "loss": 1.4746,
+      "step": 4400
+    },
+    {
+      "epoch": 3.294732909973851,
+      "grad_norm": 0.670122504234314,
+      "learning_rate": 0.0002,
+      "loss": 1.4901,
+      "step": 4410
+    },
+    {
+      "epoch": 3.3022039596563317,
+      "grad_norm": 0.6019457578659058,
+      "learning_rate": 0.0002,
+      "loss": 1.4677,
+      "step": 4420
+    },
+    {
+      "epoch": 3.309675009338812,
+      "grad_norm": 0.5519300103187561,
+      "learning_rate": 0.0002,
+      "loss": 1.4249,
+      "step": 4430
+    },
+    {
+      "epoch": 3.3171460590212924,
+      "grad_norm": 0.5958521962165833,
+      "learning_rate": 0.0002,
+      "loss": 1.555,
+      "step": 4440
+    },
+    {
+      "epoch": 3.324617108703773,
+      "grad_norm": 0.5552705526351929,
+      "learning_rate": 0.0002,
+      "loss": 1.5067,
+      "step": 4450
+    },
+    {
+      "epoch": 3.332088158386253,
+      "grad_norm": 0.6583784818649292,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 4460
+    },
+    {
+      "epoch": 3.3395592080687337,
+      "grad_norm": 0.5815939903259277,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 4470
+    },
+    {
+      "epoch": 3.347030257751214,
+      "grad_norm": 1.3342205286026,
+      "learning_rate": 0.0002,
+      "loss": 1.5942,
+      "step": 4480
+    },
+    {
+      "epoch": 3.3545013074336945,
+      "grad_norm": 0.6341500878334045,
+      "learning_rate": 0.0002,
+      "loss": 1.484,
+      "step": 4490
+    },
+    {
+      "epoch": 3.3619723571161746,
+      "grad_norm": 0.6384079456329346,
+      "learning_rate": 0.0002,
+      "loss": 1.5219,
+      "step": 4500
+    },
+    {
+      "epoch": 3.369443406798655,
+      "grad_norm": 0.6098346710205078,
+      "learning_rate": 0.0002,
+      "loss": 1.5222,
+      "step": 4510
+    },
+    {
+      "epoch": 3.3769144564811358,
+      "grad_norm": 0.5958296656608582,
+      "learning_rate": 0.0002,
+      "loss": 1.5475,
+      "step": 4520
+    },
+    {
+      "epoch": 3.384385506163616,
+      "grad_norm": 0.6157881617546082,
+      "learning_rate": 0.0002,
+      "loss": 1.5171,
+      "step": 4530
+    },
+    {
+      "epoch": 3.3918565558460965,
+      "grad_norm": 0.5671007037162781,
+      "learning_rate": 0.0002,
+      "loss": 1.569,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3993276055285766,
+      "grad_norm": 0.6203294992446899,
+      "learning_rate": 0.0002,
+      "loss": 1.604,
+      "step": 4550
+    },
+    {
+      "epoch": 3.406798655211057,
+      "grad_norm": 0.6743317246437073,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 4560
+    },
+    {
+      "epoch": 3.4142697048935373,
+      "grad_norm": 0.731765627861023,
+      "learning_rate": 0.0002,
+      "loss": 1.5034,
+      "step": 4570
+    },
+    {
+      "epoch": 3.421740754576018,
+      "grad_norm": 0.6285187602043152,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 4580
+    },
+    {
+      "epoch": 3.4292118042584985,
+      "grad_norm": 0.612680196762085,
+      "learning_rate": 0.0002,
+      "loss": 1.5296,
+      "step": 4590
+    },
+    {
+      "epoch": 3.4366828539409786,
+      "grad_norm": 0.6413681507110596,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 4600
+    },
+    {
+      "epoch": 3.444153903623459,
+      "grad_norm": 0.6240990161895752,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4610
+    },
+    {
+      "epoch": 3.4516249533059393,
+      "grad_norm": 0.5095735192298889,
+      "learning_rate": 0.0002,
+      "loss": 1.5887,
+      "step": 4620
+    },
+    {
+      "epoch": 3.45909600298842,
+      "grad_norm": 0.5699611902236938,
+      "learning_rate": 0.0002,
+      "loss": 1.4906,
+      "step": 4630
+    },
+    {
+      "epoch": 3.4665670526709,
+      "grad_norm": 0.7289775609970093,
+      "learning_rate": 0.0002,
+      "loss": 1.5176,
+      "step": 4640
+    },
+    {
+      "epoch": 3.4740381023533806,
+      "grad_norm": 0.6211609840393066,
+      "learning_rate": 0.0002,
+      "loss": 1.5467,
+      "step": 4650
+    },
+    {
+      "epoch": 3.481509152035861,
+      "grad_norm": 0.5714802145957947,
+      "learning_rate": 0.0002,
+      "loss": 1.533,
+      "step": 4660
+    },
+    {
+      "epoch": 3.4889802017183413,
+      "grad_norm": 0.6287049651145935,
+      "learning_rate": 0.0002,
+      "loss": 1.5096,
+      "step": 4670
+    },
+    {
+      "epoch": 3.496451251400822,
+      "grad_norm": 0.5480595827102661,
+      "learning_rate": 0.0002,
+      "loss": 1.4212,
+      "step": 4680
+    },
+    {
+      "epoch": 3.503922301083302,
+      "grad_norm": 0.5683253407478333,
+      "learning_rate": 0.0002,
+      "loss": 1.4746,
+      "step": 4690
+    },
+    {
+      "epoch": 3.5113933507657826,
+      "grad_norm": 0.601140558719635,
+      "learning_rate": 0.0002,
+      "loss": 1.5012,
+      "step": 4700
+    },
+    {
+      "epoch": 3.5188644004482628,
+      "grad_norm": 0.5344498157501221,
+      "learning_rate": 0.0002,
+      "loss": 1.5383,
+      "step": 4710
+    },
+    {
+      "epoch": 3.5263354501307433,
+      "grad_norm": 0.5739690661430359,
+      "learning_rate": 0.0002,
+      "loss": 1.5428,
+      "step": 4720
+    },
+    {
+      "epoch": 3.533806499813224,
+      "grad_norm": 0.5640085935592651,
+      "learning_rate": 0.0002,
+      "loss": 1.5589,
+      "step": 4730
+    },
+    {
+      "epoch": 3.541277549495704,
+      "grad_norm": 0.5967805981636047,
+      "learning_rate": 0.0002,
+      "loss": 1.487,
+      "step": 4740
+    },
+    {
+      "epoch": 3.5487485991781846,
+      "grad_norm": 0.6138835549354553,
+      "learning_rate": 0.0002,
+      "loss": 1.5461,
+      "step": 4750
+    },
+    {
+      "epoch": 3.5562196488606648,
+      "grad_norm": 0.6779900193214417,
+      "learning_rate": 0.0002,
+      "loss": 1.5502,
+      "step": 4760
+    },
+    {
+      "epoch": 3.5636906985431454,
+      "grad_norm": 0.6122010350227356,
+      "learning_rate": 0.0002,
+      "loss": 1.4917,
+      "step": 4770
+    },
+    {
+      "epoch": 3.5711617482256255,
+      "grad_norm": 0.5685241222381592,
+      "learning_rate": 0.0002,
+      "loss": 1.5405,
+      "step": 4780
+    },
+    {
+      "epoch": 3.578632797908106,
+      "grad_norm": 0.604583203792572,
+      "learning_rate": 0.0002,
+      "loss": 1.5427,
+      "step": 4790
+    },
+    {
+      "epoch": 3.5861038475905866,
+      "grad_norm": 0.651165246963501,
+      "learning_rate": 0.0002,
+      "loss": 1.4514,
+      "step": 4800
+    },
+    {
+      "epoch": 3.593574897273067,
+      "grad_norm": 0.6398511528968811,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 4810
+    },
+    {
+      "epoch": 3.6010459469555474,
+      "grad_norm": 0.6444641351699829,
+      "learning_rate": 0.0002,
+      "loss": 1.4261,
+      "step": 4820
+    },
+    {
+      "epoch": 3.6085169966380275,
+      "grad_norm": 0.6018481850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.5274,
+      "step": 4830
+    },
+    {
+      "epoch": 3.615988046320508,
+      "grad_norm": 0.6025291085243225,
+      "learning_rate": 0.0002,
+      "loss": 1.4647,
+      "step": 4840
+    },
+    {
+      "epoch": 3.623459096002988,
+      "grad_norm": 0.6810156106948853,
+      "learning_rate": 0.0002,
+      "loss": 1.5609,
+      "step": 4850
+    },
+    {
+      "epoch": 3.630930145685469,
+      "grad_norm": 0.6408044695854187,
+      "learning_rate": 0.0002,
+      "loss": 1.5299,
+      "step": 4860
+    },
+    {
+      "epoch": 3.6384011953679494,
+      "grad_norm": 0.5608272552490234,
+      "learning_rate": 0.0002,
+      "loss": 1.5366,
+      "step": 4870
+    },
+    {
+      "epoch": 3.6458722450504295,
+      "grad_norm": 0.6136814951896667,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 4880
+    },
+    {
+      "epoch": 3.65334329473291,
+      "grad_norm": 0.5927900075912476,
+      "learning_rate": 0.0002,
+      "loss": 1.5021,
+      "step": 4890
+    },
+    {
+      "epoch": 3.66081434441539,
+      "grad_norm": 0.5336901545524597,
+      "learning_rate": 0.0002,
+      "loss": 1.6084,
+      "step": 4900
+    },
+    {
+      "epoch": 3.668285394097871,
+      "grad_norm": 0.7823320627212524,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 4910
+    },
+    {
+      "epoch": 3.675756443780351,
+      "grad_norm": 0.6703504323959351,
+      "learning_rate": 0.0002,
+      "loss": 1.4881,
+      "step": 4920
+    },
+    {
+      "epoch": 3.6832274934628315,
+      "grad_norm": 0.6061160564422607,
+      "learning_rate": 0.0002,
+      "loss": 1.5332,
+      "step": 4930
+    },
+    {
+      "epoch": 3.690698543145312,
+      "grad_norm": 0.6237227916717529,
+      "learning_rate": 0.0002,
+      "loss": 1.5405,
+      "step": 4940
+    },
+    {
+      "epoch": 3.6981695928277922,
+      "grad_norm": 0.5985278487205505,
+      "learning_rate": 0.0002,
+      "loss": 1.497,
+      "step": 4950
+    },
+    {
+      "epoch": 3.705640642510273,
+      "grad_norm": 0.6483839750289917,
+      "learning_rate": 0.0002,
+      "loss": 1.5132,
+      "step": 4960
+    },
+    {
+      "epoch": 3.713111692192753,
+      "grad_norm": 0.5788805484771729,
+      "learning_rate": 0.0002,
+      "loss": 1.5338,
+      "step": 4970
+    },
+    {
+      "epoch": 3.7205827418752335,
+      "grad_norm": 0.5609974265098572,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 4980
+    },
+    {
+      "epoch": 3.7280537915577137,
+      "grad_norm": 0.5681300759315491,
+      "learning_rate": 0.0002,
+      "loss": 1.4759,
+      "step": 4990
+    },
+    {
+      "epoch": 3.7355248412401942,
+      "grad_norm": 0.5860186219215393,
+      "learning_rate": 0.0002,
+      "loss": 1.6018,
+      "step": 5000
+    },
+    {
+      "epoch": 3.742995890922675,
+      "grad_norm": 0.5718157291412354,
+      "learning_rate": 0.0002,
+      "loss": 1.58,
+      "step": 5010
+    },
+    {
+      "epoch": 3.750466940605155,
+      "grad_norm": 0.6173721551895142,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 5020
+    },
+    {
+      "epoch": 3.7579379902876355,
+      "grad_norm": 0.629152238368988,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 5030
+    },
+    {
+      "epoch": 3.7654090399701157,
+      "grad_norm": 0.5666284561157227,
+      "learning_rate": 0.0002,
+      "loss": 1.519,
+      "step": 5040
+    },
+    {
+      "epoch": 3.7728800896525962,
+      "grad_norm": 0.6053005456924438,
+      "learning_rate": 0.0002,
+      "loss": 1.5329,
+      "step": 5050
+    },
+    {
+      "epoch": 3.7803511393350764,
+      "grad_norm": 0.5870583057403564,
+      "learning_rate": 0.0002,
+      "loss": 1.5404,
+      "step": 5060
+    },
+    {
+      "epoch": 3.787822189017557,
+      "grad_norm": 0.5422009229660034,
+      "learning_rate": 0.0002,
+      "loss": 1.4444,
+      "step": 5070
+    },
+    {
+      "epoch": 3.7952932387000375,
+      "grad_norm": 0.5396918058395386,
+      "learning_rate": 0.0002,
+      "loss": 1.5308,
+      "step": 5080
+    },
+    {
+      "epoch": 3.8027642883825177,
+      "grad_norm": 0.5544713139533997,
+      "learning_rate": 0.0002,
+      "loss": 1.464,
+      "step": 5090
+    },
+    {
+      "epoch": 3.8102353380649983,
+      "grad_norm": 0.5983749628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5100
+    },
+    {
+      "epoch": 3.8177063877474784,
+      "grad_norm": 0.5702024102210999,
+      "learning_rate": 0.0002,
+      "loss": 1.4972,
+      "step": 5110
+    },
+    {
+      "epoch": 3.825177437429959,
+      "grad_norm": 0.5436882376670837,
+      "learning_rate": 0.0002,
+      "loss": 1.5471,
+      "step": 5120
+    },
+    {
+      "epoch": 3.832648487112439,
+      "grad_norm": 0.5453617572784424,
+      "learning_rate": 0.0002,
+      "loss": 1.5118,
+      "step": 5130
+    },
+    {
+      "epoch": 3.8401195367949197,
+      "grad_norm": 0.6269069314002991,
+      "learning_rate": 0.0002,
+      "loss": 1.5732,
+      "step": 5140
+    },
+    {
+      "epoch": 3.8475905864774003,
+      "grad_norm": 0.6189185380935669,
+      "learning_rate": 0.0002,
+      "loss": 1.4959,
+      "step": 5150
+    },
+    {
+      "epoch": 3.8550616361598804,
+      "grad_norm": 0.6653388142585754,
+      "learning_rate": 0.0002,
+      "loss": 1.4999,
+      "step": 5160
+    },
+    {
+      "epoch": 3.862532685842361,
+      "grad_norm": 0.5771768689155579,
+      "learning_rate": 0.0002,
+      "loss": 1.5075,
+      "step": 5170
+    },
+    {
+      "epoch": 3.870003735524841,
+      "grad_norm": 0.6052790880203247,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5180
+    },
+    {
+      "epoch": 3.8774747852073217,
+      "grad_norm": 0.6572316884994507,
+      "learning_rate": 0.0002,
+      "loss": 1.4987,
+      "step": 5190
+    },
+    {
+      "epoch": 3.884945834889802,
+      "grad_norm": 0.670576810836792,
+      "learning_rate": 0.0002,
+      "loss": 1.5241,
+      "step": 5200
+    },
+    {
+      "epoch": 3.8924168845722824,
+      "grad_norm": 0.5728798508644104,
+      "learning_rate": 0.0002,
+      "loss": 1.4777,
+      "step": 5210
+    },
+    {
+      "epoch": 3.899887934254763,
+      "grad_norm": 0.6340774297714233,
+      "learning_rate": 0.0002,
+      "loss": 1.5351,
+      "step": 5220
+    },
+    {
+      "epoch": 3.907358983937243,
+      "grad_norm": 0.5981315970420837,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 5230
+    },
+    {
+      "epoch": 3.9148300336197237,
+      "grad_norm": 0.6212025880813599,
+      "learning_rate": 0.0002,
+      "loss": 1.4875,
+      "step": 5240
+    },
+    {
+      "epoch": 3.922301083302204,
+      "grad_norm": 0.6202296018600464,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5250
+    },
+    {
+      "epoch": 3.9297721329846844,
+      "grad_norm": 0.6159142255783081,
+      "learning_rate": 0.0002,
+      "loss": 1.5765,
+      "step": 5260
+    },
+    {
+      "epoch": 3.9372431826671646,
+      "grad_norm": 0.6519438624382019,
+      "learning_rate": 0.0002,
+      "loss": 1.4938,
+      "step": 5270
+    },
+    {
+      "epoch": 3.944714232349645,
+      "grad_norm": 0.539813756942749,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 5280
+    },
+    {
+      "epoch": 3.9521852820321257,
+      "grad_norm": 0.6443665027618408,
+      "learning_rate": 0.0002,
+      "loss": 1.5921,
+      "step": 5290
+    },
+    {
+      "epoch": 3.959656331714606,
+      "grad_norm": 0.6635757684707642,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 5300
+    },
+    {
+      "epoch": 3.9671273813970864,
+      "grad_norm": 0.589363157749176,
+      "learning_rate": 0.0002,
+      "loss": 1.5485,
+      "step": 5310
+    },
+    {
+      "epoch": 3.9745984310795666,
+      "grad_norm": 0.5788735747337341,
+      "learning_rate": 0.0002,
+      "loss": 1.5498,
+      "step": 5320
+    },
+    {
+      "epoch": 3.982069480762047,
+      "grad_norm": 0.5976864695549011,
+      "learning_rate": 0.0002,
+      "loss": 1.5607,
+      "step": 5330
+    },
+    {
+      "epoch": 3.9895405304445273,
+      "grad_norm": 0.6624067425727844,
+      "learning_rate": 0.0002,
+      "loss": 1.5302,
+      "step": 5340
+    },
+    {
+      "epoch": 3.997011580127008,
+      "grad_norm": 0.6738956570625305,
+      "learning_rate": 0.0002,
+      "loss": 1.5904,
+      "step": 5350
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.868006944656372,
+      "eval_runtime": 38.5153,
+      "eval_samples_per_second": 13.371,
+      "eval_steps_per_second": 1.688,
+      "step": 5354
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10704,
+  "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": 2.4777113190845645e+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-0.9-num-6326-sd-42/checkpoint-5354/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-0.9-num-6326-sd-42/checkpoint-5354/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..67c7b4ca126d7b712ad1985ef15ffb29ebe76633
--- /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-0.9-num-6326-sd-42/checkpoint-5354/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fc87da605e94ff0ecd8f5b371302a2d5f8727b77984707a2185ddb447fc3796
+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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3b524961a7cacdf0b8dad16e7f55191ca2cbf84b
--- /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-0.9-num-6326-sd-42/checkpoint-6692/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf78460aaba2dfe8b0336cc414547b7bc27589452c9b9f4ca884bdc52f50bc9b
+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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..cea8f8db4feb4d3342d8ba1d777f602eb342a958
--- /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-0.9-num-6326-sd-42/checkpoint-6692/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:273d385eaf26e07c5a4d9fce01ecf1c9e52ba8ae5bf7ab1de30a91ed640e217d
+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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..7dbddeba79e26dee6931b482dbf74d3fb6512ceb
--- /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-0.9-num-6326-sd-42/checkpoint-6692/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c08434e14019b3642e7ee20049bf0db48349e255ee7fadc6ceced85de3293e5f
+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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f570954cd0dd056cda1acf31c975a205e6074667
--- /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-0.9-num-6326-sd-42/checkpoint-6692/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11c060c2039ed35dedb25f4237d17aaa81972fd44cf51b001f2fb57693fb37dd
+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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..0478ceb5cf64acd9d9e34e9143bef5bd316acbc4
--- /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-0.9-num-6326-sd-42/checkpoint-6692/trainer_state.json
@@ -0,0 +1,4756 @@
+{
+  "best_metric": 1.8046749830245972,
+  "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-0.9-num-6326-sd-42/checkpoint-2677",
+  "epoch": 4.999626447515876,
+  "eval_steps": 10,
+  "global_step": 6692,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007471049682480389,
+      "grad_norm": 0.4912872612476349,
+      "learning_rate": 0.0002,
+      "loss": 2.6181,
+      "step": 10
+    },
+    {
+      "epoch": 0.014942099364960777,
+      "grad_norm": 0.4856316149234772,
+      "learning_rate": 0.0002,
+      "loss": 2.2606,
+      "step": 20
+    },
+    {
+      "epoch": 0.022413149047441166,
+      "grad_norm": 0.47683125734329224,
+      "learning_rate": 0.0002,
+      "loss": 2.0957,
+      "step": 30
+    },
+    {
+      "epoch": 0.029884198729921554,
+      "grad_norm": 0.515082597732544,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 40
+    },
+    {
+      "epoch": 0.03735524841240194,
+      "grad_norm": 0.5299215316772461,
+      "learning_rate": 0.0002,
+      "loss": 1.9704,
+      "step": 50
+    },
+    {
+      "epoch": 0.04482629809488233,
+      "grad_norm": 0.4951399862766266,
+      "learning_rate": 0.0002,
+      "loss": 1.9225,
+      "step": 60
+    },
+    {
+      "epoch": 0.05229734777736272,
+      "grad_norm": 0.48079821467399597,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05976839745984311,
+      "grad_norm": 0.49402132630348206,
+      "learning_rate": 0.0002,
+      "loss": 1.9466,
+      "step": 80
+    },
+    {
+      "epoch": 0.0672394471423235,
+      "grad_norm": 0.4778193235397339,
+      "learning_rate": 0.0002,
+      "loss": 1.8691,
+      "step": 90
+    },
+    {
+      "epoch": 0.07471049682480388,
+      "grad_norm": 0.42472657561302185,
+      "learning_rate": 0.0002,
+      "loss": 1.8455,
+      "step": 100
+    },
+    {
+      "epoch": 0.08218154650728428,
+      "grad_norm": 0.4433092474937439,
+      "learning_rate": 0.0002,
+      "loss": 1.8744,
+      "step": 110
+    },
+    {
+      "epoch": 0.08965259618976466,
+      "grad_norm": 0.4472862780094147,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 120
+    },
+    {
+      "epoch": 0.09712364587224505,
+      "grad_norm": 0.42596298456192017,
+      "learning_rate": 0.0002,
+      "loss": 1.9256,
+      "step": 130
+    },
+    {
+      "epoch": 0.10459469555472543,
+      "grad_norm": 0.46645811200141907,
+      "learning_rate": 0.0002,
+      "loss": 1.8015,
+      "step": 140
+    },
+    {
+      "epoch": 0.11206574523720583,
+      "grad_norm": 0.41041234135627747,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 150
+    },
+    {
+      "epoch": 0.11953679491968622,
+      "grad_norm": 0.5329819917678833,
+      "learning_rate": 0.0002,
+      "loss": 1.8276,
+      "step": 160
+    },
+    {
+      "epoch": 0.1270078446021666,
+      "grad_norm": 0.4065922200679779,
+      "learning_rate": 0.0002,
+      "loss": 1.8118,
+      "step": 170
+    },
+    {
+      "epoch": 0.134478894284647,
+      "grad_norm": 0.38406994938850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8559,
+      "step": 180
+    },
+    {
+      "epoch": 0.14194994396712737,
+      "grad_norm": 0.4246881306171417,
+      "learning_rate": 0.0002,
+      "loss": 1.8647,
+      "step": 190
+    },
+    {
+      "epoch": 0.14942099364960776,
+      "grad_norm": 0.35136649012565613,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 200
+    },
+    {
+      "epoch": 0.15689204333208817,
+      "grad_norm": 0.43252742290496826,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.16436309301456856,
+      "grad_norm": 0.39236941933631897,
+      "learning_rate": 0.0002,
+      "loss": 1.7823,
+      "step": 220
+    },
+    {
+      "epoch": 0.17183414269704894,
+      "grad_norm": 0.3748249113559723,
+      "learning_rate": 0.0002,
+      "loss": 1.818,
+      "step": 230
+    },
+    {
+      "epoch": 0.17930519237952933,
+      "grad_norm": 0.6432855725288391,
+      "learning_rate": 0.0002,
+      "loss": 1.866,
+      "step": 240
+    },
+    {
+      "epoch": 0.1867762420620097,
+      "grad_norm": 0.34874802827835083,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 250
+    },
+    {
+      "epoch": 0.1942472917444901,
+      "grad_norm": 0.3721984326839447,
+      "learning_rate": 0.0002,
+      "loss": 1.79,
+      "step": 260
+    },
+    {
+      "epoch": 0.20171834142697048,
+      "grad_norm": 0.4339311420917511,
+      "learning_rate": 0.0002,
+      "loss": 1.8464,
+      "step": 270
+    },
+    {
+      "epoch": 0.20918939110945087,
+      "grad_norm": 0.4018215537071228,
+      "learning_rate": 0.0002,
+      "loss": 1.8665,
+      "step": 280
+    },
+    {
+      "epoch": 0.21666044079193125,
+      "grad_norm": 0.3278839886188507,
+      "learning_rate": 0.0002,
+      "loss": 1.8048,
+      "step": 290
+    },
+    {
+      "epoch": 0.22413149047441167,
+      "grad_norm": 0.36146077513694763,
+      "learning_rate": 0.0002,
+      "loss": 1.7395,
+      "step": 300
+    },
+    {
+      "epoch": 0.23160254015689205,
+      "grad_norm": 0.38175010681152344,
+      "learning_rate": 0.0002,
+      "loss": 1.7916,
+      "step": 310
+    },
+    {
+      "epoch": 0.23907358983937244,
+      "grad_norm": 0.44776618480682373,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 320
+    },
+    {
+      "epoch": 0.24654463952185282,
+      "grad_norm": 0.3933652937412262,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 330
+    },
+    {
+      "epoch": 0.2540156892043332,
+      "grad_norm": 0.3515005111694336,
+      "learning_rate": 0.0002,
+      "loss": 1.8393,
+      "step": 340
+    },
+    {
+      "epoch": 0.2614867388868136,
+      "grad_norm": 0.6683304309844971,
+      "learning_rate": 0.0002,
+      "loss": 1.8653,
+      "step": 350
+    },
+    {
+      "epoch": 0.268957788569294,
+      "grad_norm": 0.37093454599380493,
+      "learning_rate": 0.0002,
+      "loss": 1.8797,
+      "step": 360
+    },
+    {
+      "epoch": 0.2764288382517744,
+      "grad_norm": 0.3450651168823242,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 370
+    },
+    {
+      "epoch": 0.28389988793425475,
+      "grad_norm": 0.5140917301177979,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 380
+    },
+    {
+      "epoch": 0.29137093761673516,
+      "grad_norm": 0.32885563373565674,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 390
+    },
+    {
+      "epoch": 0.2988419872992155,
+      "grad_norm": 0.33962297439575195,
+      "learning_rate": 0.0002,
+      "loss": 1.8174,
+      "step": 400
+    },
+    {
+      "epoch": 0.30631303698169593,
+      "grad_norm": 0.3723141849040985,
+      "learning_rate": 0.0002,
+      "loss": 1.7467,
+      "step": 410
+    },
+    {
+      "epoch": 0.31378408666417634,
+      "grad_norm": 0.37173134088516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8459,
+      "step": 420
+    },
+    {
+      "epoch": 0.3212551363466567,
+      "grad_norm": 0.33736956119537354,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 430
+    },
+    {
+      "epoch": 0.3287261860291371,
+      "grad_norm": 0.3602448105812073,
+      "learning_rate": 0.0002,
+      "loss": 1.8367,
+      "step": 440
+    },
+    {
+      "epoch": 0.33619723571161747,
+      "grad_norm": 0.3569699227809906,
+      "learning_rate": 0.0002,
+      "loss": 1.8058,
+      "step": 450
+    },
+    {
+      "epoch": 0.3436682853940979,
+      "grad_norm": 0.31009167432785034,
+      "learning_rate": 0.0002,
+      "loss": 1.8086,
+      "step": 460
+    },
+    {
+      "epoch": 0.35113933507657824,
+      "grad_norm": 0.5278693437576294,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 470
+    },
+    {
+      "epoch": 0.35861038475905865,
+      "grad_norm": 0.3587537109851837,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 480
+    },
+    {
+      "epoch": 0.366081434441539,
+      "grad_norm": 0.3859670162200928,
+      "learning_rate": 0.0002,
+      "loss": 1.8046,
+      "step": 490
+    },
+    {
+      "epoch": 0.3735524841240194,
+      "grad_norm": 0.395913690328598,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 500
+    },
+    {
+      "epoch": 0.38102353380649984,
+      "grad_norm": 0.35052940249443054,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 510
+    },
+    {
+      "epoch": 0.3884945834889802,
+      "grad_norm": 0.2979494333267212,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 520
+    },
+    {
+      "epoch": 0.3959656331714606,
+      "grad_norm": 0.3062683343887329,
+      "learning_rate": 0.0002,
+      "loss": 1.8641,
+      "step": 530
+    },
+    {
+      "epoch": 0.40343668285394096,
+      "grad_norm": 0.3172847330570221,
+      "learning_rate": 0.0002,
+      "loss": 1.7651,
+      "step": 540
+    },
+    {
+      "epoch": 0.4109077325364214,
+      "grad_norm": 0.360435426235199,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 550
+    },
+    {
+      "epoch": 0.41837878221890173,
+      "grad_norm": 0.3427872359752655,
+      "learning_rate": 0.0002,
+      "loss": 1.9054,
+      "step": 560
+    },
+    {
+      "epoch": 0.42584983190138215,
+      "grad_norm": 0.34036558866500854,
+      "learning_rate": 0.0002,
+      "loss": 1.7562,
+      "step": 570
+    },
+    {
+      "epoch": 0.4333208815838625,
+      "grad_norm": 0.3365345299243927,
+      "learning_rate": 0.0002,
+      "loss": 1.7254,
+      "step": 580
+    },
+    {
+      "epoch": 0.4407919312663429,
+      "grad_norm": 0.35619041323661804,
+      "learning_rate": 0.0002,
+      "loss": 1.8328,
+      "step": 590
+    },
+    {
+      "epoch": 0.44826298094882333,
+      "grad_norm": 0.3569088280200958,
+      "learning_rate": 0.0002,
+      "loss": 1.8114,
+      "step": 600
+    },
+    {
+      "epoch": 0.4557340306313037,
+      "grad_norm": 0.3581278622150421,
+      "learning_rate": 0.0002,
+      "loss": 1.8599,
+      "step": 610
+    },
+    {
+      "epoch": 0.4632050803137841,
+      "grad_norm": 0.43197110295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.7078,
+      "step": 620
+    },
+    {
+      "epoch": 0.47067612999626446,
+      "grad_norm": 0.33966198563575745,
+      "learning_rate": 0.0002,
+      "loss": 1.8257,
+      "step": 630
+    },
+    {
+      "epoch": 0.47814717967874487,
+      "grad_norm": 0.3343866467475891,
+      "learning_rate": 0.0002,
+      "loss": 1.7528,
+      "step": 640
+    },
+    {
+      "epoch": 0.48561822936122523,
+      "grad_norm": 0.33878564834594727,
+      "learning_rate": 0.0002,
+      "loss": 1.8191,
+      "step": 650
+    },
+    {
+      "epoch": 0.49308927904370564,
+      "grad_norm": 0.387195885181427,
+      "learning_rate": 0.0002,
+      "loss": 1.8801,
+      "step": 660
+    },
+    {
+      "epoch": 0.500560328726186,
+      "grad_norm": 0.3755440413951874,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 670
+    },
+    {
+      "epoch": 0.5080313784086664,
+      "grad_norm": 0.3272816836833954,
+      "learning_rate": 0.0002,
+      "loss": 1.8057,
+      "step": 680
+    },
+    {
+      "epoch": 0.5155024280911468,
+      "grad_norm": 0.36063864827156067,
+      "learning_rate": 0.0002,
+      "loss": 1.8156,
+      "step": 690
+    },
+    {
+      "epoch": 0.5229734777736272,
+      "grad_norm": 0.35317373275756836,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 700
+    },
+    {
+      "epoch": 0.5304445274561076,
+      "grad_norm": 0.3561195433139801,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 710
+    },
+    {
+      "epoch": 0.537915577138588,
+      "grad_norm": 0.31124624609947205,
+      "learning_rate": 0.0002,
+      "loss": 1.8149,
+      "step": 720
+    },
+    {
+      "epoch": 0.5453866268210683,
+      "grad_norm": 0.3294544517993927,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 730
+    },
+    {
+      "epoch": 0.5528576765035488,
+      "grad_norm": 0.31933900713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.8027,
+      "step": 740
+    },
+    {
+      "epoch": 0.5603287261860291,
+      "grad_norm": 0.3226020634174347,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 750
+    },
+    {
+      "epoch": 0.5677997758685095,
+      "grad_norm": 0.3147525489330292,
+      "learning_rate": 0.0002,
+      "loss": 1.7862,
+      "step": 760
+    },
+    {
+      "epoch": 0.57527082555099,
+      "grad_norm": 0.32234328985214233,
+      "learning_rate": 0.0002,
+      "loss": 1.9028,
+      "step": 770
+    },
+    {
+      "epoch": 0.5827418752334703,
+      "grad_norm": 0.3258664309978485,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 780
+    },
+    {
+      "epoch": 0.5902129249159507,
+      "grad_norm": 0.3166961967945099,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 790
+    },
+    {
+      "epoch": 0.597683974598431,
+      "grad_norm": 0.35621458292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 800
+    },
+    {
+      "epoch": 0.6051550242809115,
+      "grad_norm": 0.3236999213695526,
+      "learning_rate": 0.0002,
+      "loss": 1.8313,
+      "step": 810
+    },
+    {
+      "epoch": 0.6126260739633919,
+      "grad_norm": 0.2892923653125763,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 820
+    },
+    {
+      "epoch": 0.6200971236458722,
+      "grad_norm": 0.4098321497440338,
+      "learning_rate": 0.0002,
+      "loss": 1.8709,
+      "step": 830
+    },
+    {
+      "epoch": 0.6275681733283527,
+      "grad_norm": 0.3337118923664093,
+      "learning_rate": 0.0002,
+      "loss": 1.7637,
+      "step": 840
+    },
+    {
+      "epoch": 0.635039223010833,
+      "grad_norm": 0.30416029691696167,
+      "learning_rate": 0.0002,
+      "loss": 1.7375,
+      "step": 850
+    },
+    {
+      "epoch": 0.6425102726933134,
+      "grad_norm": 0.3361026346683502,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 860
+    },
+    {
+      "epoch": 0.6499813223757938,
+      "grad_norm": 0.3537365198135376,
+      "learning_rate": 0.0002,
+      "loss": 1.732,
+      "step": 870
+    },
+    {
+      "epoch": 0.6574523720582742,
+      "grad_norm": 0.33854469656944275,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 880
+    },
+    {
+      "epoch": 0.6649234217407546,
+      "grad_norm": 0.3332272469997406,
+      "learning_rate": 0.0002,
+      "loss": 1.7561,
+      "step": 890
+    },
+    {
+      "epoch": 0.6723944714232349,
+      "grad_norm": 0.34954726696014404,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 900
+    },
+    {
+      "epoch": 0.6798655211057153,
+      "grad_norm": 0.2921750247478485,
+      "learning_rate": 0.0002,
+      "loss": 1.7917,
+      "step": 910
+    },
+    {
+      "epoch": 0.6873365707881958,
+      "grad_norm": 0.30508682131767273,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 920
+    },
+    {
+      "epoch": 0.6948076204706761,
+      "grad_norm": 0.32268425822257996,
+      "learning_rate": 0.0002,
+      "loss": 1.8082,
+      "step": 930
+    },
+    {
+      "epoch": 0.7022786701531565,
+      "grad_norm": 0.2844390869140625,
+      "learning_rate": 0.0002,
+      "loss": 1.8283,
+      "step": 940
+    },
+    {
+      "epoch": 0.709749719835637,
+      "grad_norm": 0.31263890862464905,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 950
+    },
+    {
+      "epoch": 0.7172207695181173,
+      "grad_norm": 0.3626808822154999,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 960
+    },
+    {
+      "epoch": 0.7246918192005977,
+      "grad_norm": 0.3322749733924866,
+      "learning_rate": 0.0002,
+      "loss": 1.853,
+      "step": 970
+    },
+    {
+      "epoch": 0.732162868883078,
+      "grad_norm": 0.29177871346473694,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 980
+    },
+    {
+      "epoch": 0.7396339185655585,
+      "grad_norm": 0.35405513644218445,
+      "learning_rate": 0.0002,
+      "loss": 1.8447,
+      "step": 990
+    },
+    {
+      "epoch": 0.7471049682480388,
+      "grad_norm": 0.39318400621414185,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7545760179305192,
+      "grad_norm": 0.29401418566703796,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7620470676129997,
+      "grad_norm": 0.3271748721599579,
+      "learning_rate": 0.0002,
+      "loss": 1.7649,
+      "step": 1020
+    },
+    {
+      "epoch": 0.76951811729548,
+      "grad_norm": 0.30883970856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7769891669779604,
+      "grad_norm": 0.3411838412284851,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7844602166604407,
+      "grad_norm": 0.30608129501342773,
+      "learning_rate": 0.0002,
+      "loss": 1.829,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7919312663429212,
+      "grad_norm": 0.30899080634117126,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7994023160254016,
+      "grad_norm": 0.3160453140735626,
+      "learning_rate": 0.0002,
+      "loss": 1.7625,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8068733657078819,
+      "grad_norm": 0.30947187542915344,
+      "learning_rate": 0.0002,
+      "loss": 1.8452,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8143444153903624,
+      "grad_norm": 0.3103134036064148,
+      "learning_rate": 0.0002,
+      "loss": 1.7418,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8218154650728428,
+      "grad_norm": 0.31771138310432434,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8292865147553231,
+      "grad_norm": 0.5860997438430786,
+      "learning_rate": 0.0002,
+      "loss": 1.7918,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8367575644378035,
+      "grad_norm": 0.3230148255825043,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8442286141202839,
+      "grad_norm": 0.29611510038375854,
+      "learning_rate": 0.0002,
+      "loss": 1.8478,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8516996638027643,
+      "grad_norm": 0.3373654782772064,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8591707134852447,
+      "grad_norm": 0.3474279046058655,
+      "learning_rate": 0.0002,
+      "loss": 1.7997,
+      "step": 1150
+    },
+    {
+      "epoch": 0.866641763167725,
+      "grad_norm": 0.35057875514030457,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8741128128502055,
+      "grad_norm": 0.39537495374679565,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8815838625326858,
+      "grad_norm": 0.3714233636856079,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8890549122151662,
+      "grad_norm": 0.2950296998023987,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8965259618976467,
+      "grad_norm": 0.38182979822158813,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 1200
+    },
+    {
+      "epoch": 0.903997011580127,
+      "grad_norm": 0.27883678674697876,
+      "learning_rate": 0.0002,
+      "loss": 1.827,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9114680612626074,
+      "grad_norm": 0.33874374628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9189391109450877,
+      "grad_norm": 0.3014272153377533,
+      "learning_rate": 0.0002,
+      "loss": 1.7334,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9264101606275682,
+      "grad_norm": 0.3194271922111511,
+      "learning_rate": 0.0002,
+      "loss": 1.8235,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9338812103100486,
+      "grad_norm": 0.3049403429031372,
+      "learning_rate": 0.0002,
+      "loss": 1.7924,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9413522599925289,
+      "grad_norm": 0.30621254444122314,
+      "learning_rate": 0.0002,
+      "loss": 1.7535,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9488233096750094,
+      "grad_norm": 0.28675132989883423,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9562943593574897,
+      "grad_norm": 0.3322032690048218,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9637654090399701,
+      "grad_norm": 0.35408294200897217,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9712364587224505,
+      "grad_norm": 0.36386919021606445,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9787075084049309,
+      "grad_norm": 0.32338324189186096,
+      "learning_rate": 0.0002,
+      "loss": 1.8633,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9861785580874113,
+      "grad_norm": 0.3714013993740082,
+      "learning_rate": 0.0002,
+      "loss": 1.7724,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9936496077698916,
+      "grad_norm": 0.3133082389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.7766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9996264475158759,
+      "eval_loss": 1.8051470518112183,
+      "eval_runtime": 38.6332,
+      "eval_samples_per_second": 13.331,
+      "eval_steps_per_second": 1.682,
+      "step": 1338
+    },
+    {
+      "epoch": 1.001120657452372,
+      "grad_norm": 0.31595754623413086,
+      "learning_rate": 0.0002,
+      "loss": 1.8035,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0085917071348525,
+      "grad_norm": 0.3095700144767761,
+      "learning_rate": 0.0002,
+      "loss": 1.7486,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0160627568173328,
+      "grad_norm": 0.34677496552467346,
+      "learning_rate": 0.0002,
+      "loss": 1.6981,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0235338064998132,
+      "grad_norm": 0.29108840227127075,
+      "learning_rate": 0.0002,
+      "loss": 1.7377,
+      "step": 1370
+    },
+    {
+      "epoch": 1.0310048561822935,
+      "grad_norm": 0.32356950640678406,
+      "learning_rate": 0.0002,
+      "loss": 1.7194,
+      "step": 1380
+    },
+    {
+      "epoch": 1.038475905864774,
+      "grad_norm": 0.4200669229030609,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0459469555472545,
+      "grad_norm": 0.3283711373806,
+      "learning_rate": 0.0002,
+      "loss": 1.797,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0534180052297348,
+      "grad_norm": 0.32898256182670593,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0608890549122152,
+      "grad_norm": 0.38790300488471985,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0683601045946955,
+      "grad_norm": 0.339800089597702,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1430
+    },
+    {
+      "epoch": 1.075831154277176,
+      "grad_norm": 0.3548751175403595,
+      "learning_rate": 0.0002,
+      "loss": 1.7076,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0833022039596563,
+      "grad_norm": 0.35114359855651855,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0907732536421366,
+      "grad_norm": 0.35226720571517944,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0982443033246172,
+      "grad_norm": 0.33665576577186584,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1057153530070976,
+      "grad_norm": 0.363889217376709,
+      "learning_rate": 0.0002,
+      "loss": 1.6699,
+      "step": 1480
+    },
+    {
+      "epoch": 1.113186402689578,
+      "grad_norm": 0.3826201856136322,
+      "learning_rate": 0.0002,
+      "loss": 1.7933,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1206574523720583,
+      "grad_norm": 0.34058740735054016,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1281285020545386,
+      "grad_norm": 0.3462134301662445,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 1510
+    },
+    {
+      "epoch": 1.135599551737019,
+      "grad_norm": 0.3396756052970886,
+      "learning_rate": 0.0002,
+      "loss": 1.7147,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1430706014194993,
+      "grad_norm": 0.32004743814468384,
+      "learning_rate": 0.0002,
+      "loss": 1.7219,
+      "step": 1530
+    },
+    {
+      "epoch": 1.15054165110198,
+      "grad_norm": 0.3397733271121979,
+      "learning_rate": 0.0002,
+      "loss": 1.743,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1580127007844603,
+      "grad_norm": 0.3783262073993683,
+      "learning_rate": 0.0002,
+      "loss": 1.7333,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1654837504669406,
+      "grad_norm": 0.35121291875839233,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 1560
+    },
+    {
+      "epoch": 1.172954800149421,
+      "grad_norm": 0.35816895961761475,
+      "learning_rate": 0.0002,
+      "loss": 1.678,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1804258498319014,
+      "grad_norm": 0.33843839168548584,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1878968995143817,
+      "grad_norm": 0.3371972143650055,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 1590
+    },
+    {
+      "epoch": 1.195367949196862,
+      "grad_norm": 0.36016878485679626,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2028389988793426,
+      "grad_norm": 0.40879473090171814,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 1610
+    },
+    {
+      "epoch": 1.210310048561823,
+      "grad_norm": 0.3216715455055237,
+      "learning_rate": 0.0002,
+      "loss": 1.6955,
+      "step": 1620
+    },
+    {
+      "epoch": 1.2177810982443034,
+      "grad_norm": 0.4482610821723938,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2252521479267837,
+      "grad_norm": 0.3257700502872467,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1640
+    },
+    {
+      "epoch": 1.232723197609264,
+      "grad_norm": 0.38646459579467773,
+      "learning_rate": 0.0002,
+      "loss": 1.7177,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2401942472917444,
+      "grad_norm": 0.4081360697746277,
+      "learning_rate": 0.0002,
+      "loss": 1.7081,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2476652969742248,
+      "grad_norm": 0.4326848089694977,
+      "learning_rate": 0.0002,
+      "loss": 1.7519,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2551363466567054,
+      "grad_norm": 0.346401572227478,
+      "learning_rate": 0.0002,
+      "loss": 1.6752,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2626073963391857,
+      "grad_norm": 0.34536251425743103,
+      "learning_rate": 0.0002,
+      "loss": 1.7425,
+      "step": 1690
+    },
+    {
+      "epoch": 1.270078446021666,
+      "grad_norm": 0.41359591484069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7061,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2775494957041464,
+      "grad_norm": 0.3530874252319336,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2850205453866268,
+      "grad_norm": 0.3702719211578369,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2924915950691072,
+      "grad_norm": 0.3703329563140869,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2999626447515875,
+      "grad_norm": 0.37919729948043823,
+      "learning_rate": 0.0002,
+      "loss": 1.7221,
+      "step": 1740
+    },
+    {
+      "epoch": 1.307433694434068,
+      "grad_norm": 0.32526856660842896,
+      "learning_rate": 0.0002,
+      "loss": 1.7859,
+      "step": 1750
+    },
+    {
+      "epoch": 1.3149047441165485,
+      "grad_norm": 0.36752620339393616,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3223757937990288,
+      "grad_norm": 0.3398192524909973,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3298468434815092,
+      "grad_norm": 0.37435585260391235,
+      "learning_rate": 0.0002,
+      "loss": 1.7492,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3373178931639895,
+      "grad_norm": 0.35793280601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3447889428464699,
+      "grad_norm": 0.35481882095336914,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3522599925289502,
+      "grad_norm": 0.3786393105983734,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3597310422114308,
+      "grad_norm": 0.33245593309402466,
+      "learning_rate": 0.0002,
+      "loss": 1.7169,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3672020918939112,
+      "grad_norm": 0.35388344526290894,
+      "learning_rate": 0.0002,
+      "loss": 1.7577,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3746731415763915,
+      "grad_norm": 0.3695325553417206,
+      "learning_rate": 0.0002,
+      "loss": 1.6968,
+      "step": 1840
+    },
+    {
+      "epoch": 1.382144191258872,
+      "grad_norm": 0.3683604598045349,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3896152409413522,
+      "grad_norm": 0.3753012418746948,
+      "learning_rate": 0.0002,
+      "loss": 1.7878,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3970862906238326,
+      "grad_norm": 0.3331069350242615,
+      "learning_rate": 0.0002,
+      "loss": 1.6969,
+      "step": 1870
+    },
+    {
+      "epoch": 1.404557340306313,
+      "grad_norm": 0.3877500295639038,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4120283899887935,
+      "grad_norm": 0.33525151014328003,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4194994396712737,
+      "grad_norm": 0.3697299659252167,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4269704893537543,
+      "grad_norm": 0.4029286205768585,
+      "learning_rate": 0.0002,
+      "loss": 1.6956,
+      "step": 1910
+    },
+    {
+      "epoch": 1.4344415390362346,
+      "grad_norm": 0.3596203029155731,
+      "learning_rate": 0.0002,
+      "loss": 1.6897,
+      "step": 1920
+    },
+    {
+      "epoch": 1.441912588718715,
+      "grad_norm": 0.450783908367157,
+      "learning_rate": 0.0002,
+      "loss": 1.7139,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4493836384011953,
+      "grad_norm": 0.3651481866836548,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4568546880836757,
+      "grad_norm": 0.3608424663543701,
+      "learning_rate": 0.0002,
+      "loss": 1.6637,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4643257377661563,
+      "grad_norm": 0.39684420824050903,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4717967874486364,
+      "grad_norm": 0.34618663787841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7514,
+      "step": 1970
+    },
+    {
+      "epoch": 1.479267837131117,
+      "grad_norm": 0.4150386452674866,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4867388868135973,
+      "grad_norm": 0.35500776767730713,
+      "learning_rate": 0.0002,
+      "loss": 1.7021,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4942099364960777,
+      "grad_norm": 0.344144344329834,
+      "learning_rate": 0.0002,
+      "loss": 1.7322,
+      "step": 2000
+    },
+    {
+      "epoch": 1.501680986178558,
+      "grad_norm": 0.3340149223804474,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5091520358610384,
+      "grad_norm": 0.37685006856918335,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 2020
+    },
+    {
+      "epoch": 1.516623085543519,
+      "grad_norm": 0.3699876368045807,
+      "learning_rate": 0.0002,
+      "loss": 1.8299,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5240941352259991,
+      "grad_norm": 0.3370307385921478,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5315651849084797,
+      "grad_norm": 0.37780630588531494,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 2050
+    },
+    {
+      "epoch": 1.53903623459096,
+      "grad_norm": 0.370259165763855,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5465072842734404,
+      "grad_norm": 0.3440011441707611,
+      "learning_rate": 0.0002,
+      "loss": 1.7398,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5539783339559208,
+      "grad_norm": 0.40382063388824463,
+      "learning_rate": 0.0002,
+      "loss": 1.7105,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5614493836384011,
+      "grad_norm": 0.38002029061317444,
+      "learning_rate": 0.0002,
+      "loss": 1.7071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5689204333208817,
+      "grad_norm": 0.3658451437950134,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5763914830033618,
+      "grad_norm": 0.354842871427536,
+      "learning_rate": 0.0002,
+      "loss": 1.7598,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5838625326858424,
+      "grad_norm": 0.34735530614852905,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5913335823683228,
+      "grad_norm": 0.377581924200058,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5988046320508031,
+      "grad_norm": 0.41254034638404846,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6062756817332835,
+      "grad_norm": 0.3630715310573578,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6137467314157639,
+      "grad_norm": 0.36980143189430237,
+      "learning_rate": 0.0002,
+      "loss": 1.7531,
+      "step": 2160
+    },
+    {
+      "epoch": 1.6212177810982444,
+      "grad_norm": 0.3634769320487976,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6286888307807246,
+      "grad_norm": 0.3794139623641968,
+      "learning_rate": 0.0002,
+      "loss": 1.6367,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6361598804632052,
+      "grad_norm": 0.359742134809494,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6436309301456855,
+      "grad_norm": 0.3770543932914734,
+      "learning_rate": 0.0002,
+      "loss": 1.7027,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6511019798281659,
+      "grad_norm": 0.3797036409378052,
+      "learning_rate": 0.0002,
+      "loss": 1.784,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6585730295106462,
+      "grad_norm": 0.35622093081474304,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6660440791931266,
+      "grad_norm": 0.34552520513534546,
+      "learning_rate": 0.0002,
+      "loss": 1.6615,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6735151288756072,
+      "grad_norm": 0.379926860332489,
+      "learning_rate": 0.0002,
+      "loss": 1.7522,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6809861785580873,
+      "grad_norm": 0.37083810567855835,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6884572282405679,
+      "grad_norm": 0.42746543884277344,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6959282779230482,
+      "grad_norm": 0.3372884690761566,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7033993276055286,
+      "grad_norm": 0.35220256447792053,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2280
+    },
+    {
+      "epoch": 1.710870377288009,
+      "grad_norm": 0.3659130930900574,
+      "learning_rate": 0.0002,
+      "loss": 1.7154,
+      "step": 2290
+    },
+    {
+      "epoch": 1.7183414269704893,
+      "grad_norm": 0.37629297375679016,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7258124766529699,
+      "grad_norm": 0.36312398314476013,
+      "learning_rate": 0.0002,
+      "loss": 1.7212,
+      "step": 2310
+    },
+    {
+      "epoch": 1.73328352633545,
+      "grad_norm": 0.467709481716156,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7407545760179306,
+      "grad_norm": 0.38685527443885803,
+      "learning_rate": 0.0002,
+      "loss": 1.696,
+      "step": 2330
+    },
+    {
+      "epoch": 1.748225625700411,
+      "grad_norm": 0.3578338325023651,
+      "learning_rate": 0.0002,
+      "loss": 1.7041,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7556966753828913,
+      "grad_norm": 0.36057502031326294,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7631677250653717,
+      "grad_norm": 0.3615196645259857,
+      "learning_rate": 0.0002,
+      "loss": 1.6853,
+      "step": 2360
+    },
+    {
+      "epoch": 1.770638774747852,
+      "grad_norm": 0.4118947684764862,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7781098244303326,
+      "grad_norm": 0.4067276120185852,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7855808741128127,
+      "grad_norm": 0.3979823887348175,
+      "learning_rate": 0.0002,
+      "loss": 1.712,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7930519237952933,
+      "grad_norm": 0.44045883417129517,
+      "learning_rate": 0.0002,
+      "loss": 1.7644,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8005229734777737,
+      "grad_norm": 0.3998069167137146,
+      "learning_rate": 0.0002,
+      "loss": 1.7251,
+      "step": 2410
+    },
+    {
+      "epoch": 1.807994023160254,
+      "grad_norm": 0.3450094759464264,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8154650728427344,
+      "grad_norm": 0.3759009838104248,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2430
+    },
+    {
+      "epoch": 1.8229361225252148,
+      "grad_norm": 0.34347015619277954,
+      "learning_rate": 0.0002,
+      "loss": 1.7706,
+      "step": 2440
+    },
+    {
+      "epoch": 1.8304071722076953,
+      "grad_norm": 0.3511228859424591,
+      "learning_rate": 0.0002,
+      "loss": 1.7345,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8378782218901755,
+      "grad_norm": 0.36853715777397156,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 2460
+    },
+    {
+      "epoch": 1.845349271572656,
+      "grad_norm": 0.40659376978874207,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8528203212551362,
+      "grad_norm": 0.39621320366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8602913709376168,
+      "grad_norm": 0.3753979504108429,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8677624206200971,
+      "grad_norm": 0.3811938464641571,
+      "learning_rate": 0.0002,
+      "loss": 1.6622,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8752334703025775,
+      "grad_norm": 0.3432596027851105,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 2510
+    },
+    {
+      "epoch": 1.882704519985058,
+      "grad_norm": 0.3670712113380432,
+      "learning_rate": 0.0002,
+      "loss": 1.7488,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8901755696675382,
+      "grad_norm": 0.40907177329063416,
+      "learning_rate": 0.0002,
+      "loss": 1.705,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8976466193500188,
+      "grad_norm": 0.3821999728679657,
+      "learning_rate": 0.0002,
+      "loss": 1.7148,
+      "step": 2540
+    },
+    {
+      "epoch": 1.905117669032499,
+      "grad_norm": 0.36173978447914124,
+      "learning_rate": 0.0002,
+      "loss": 1.7934,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9125887187149795,
+      "grad_norm": 0.38990336656570435,
+      "learning_rate": 0.0002,
+      "loss": 1.6939,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9200597683974598,
+      "grad_norm": 0.35242322087287903,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 2570
+    },
+    {
+      "epoch": 1.9275308180799402,
+      "grad_norm": 0.3506428003311157,
+      "learning_rate": 0.0002,
+      "loss": 1.7268,
+      "step": 2580
+    },
+    {
+      "epoch": 1.9350018677624208,
+      "grad_norm": 0.39540135860443115,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2590
+    },
+    {
+      "epoch": 1.942472917444901,
+      "grad_norm": 0.3444725573062897,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9499439671273815,
+      "grad_norm": 0.3963521718978882,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9574150168098616,
+      "grad_norm": 0.3689815402030945,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9648860664923422,
+      "grad_norm": 0.3482626676559448,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9723571161748226,
+      "grad_norm": 0.35832616686820984,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 2640
+    },
+    {
+      "epoch": 1.979828165857303,
+      "grad_norm": 0.4776208996772766,
+      "learning_rate": 0.0002,
+      "loss": 1.6681,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9872992155397835,
+      "grad_norm": 0.32570165395736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6696,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9947702652222636,
+      "grad_norm": 0.3380725085735321,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8046749830245972,
+      "eval_runtime": 38.5096,
+      "eval_samples_per_second": 13.373,
+      "eval_steps_per_second": 1.688,
+      "step": 2677
+    },
+    {
+      "epoch": 2.002241314904744,
+      "grad_norm": 0.36817631125450134,
+      "learning_rate": 0.0002,
+      "loss": 1.7265,
+      "step": 2680
+    },
+    {
+      "epoch": 2.0097123645872244,
+      "grad_norm": 0.4056456685066223,
+      "learning_rate": 0.0002,
+      "loss": 1.548,
+      "step": 2690
+    },
+    {
+      "epoch": 2.017183414269705,
+      "grad_norm": 0.37416863441467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5515,
+      "step": 2700
+    },
+    {
+      "epoch": 2.024654463952185,
+      "grad_norm": 0.4273638427257538,
+      "learning_rate": 0.0002,
+      "loss": 1.5895,
+      "step": 2710
+    },
+    {
+      "epoch": 2.0321255136346656,
+      "grad_norm": 0.36497923731803894,
+      "learning_rate": 0.0002,
+      "loss": 1.5884,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0395965633171462,
+      "grad_norm": 0.5021994113922119,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 2730
+    },
+    {
+      "epoch": 2.0470676129996264,
+      "grad_norm": 0.45896220207214355,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 2740
+    },
+    {
+      "epoch": 2.054538662682107,
+      "grad_norm": 0.3973815143108368,
+      "learning_rate": 0.0002,
+      "loss": 1.6305,
+      "step": 2750
+    },
+    {
+      "epoch": 2.062009712364587,
+      "grad_norm": 0.4521815776824951,
+      "learning_rate": 0.0002,
+      "loss": 1.6301,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0694807620470677,
+      "grad_norm": 0.42775002121925354,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2770
+    },
+    {
+      "epoch": 2.076951811729548,
+      "grad_norm": 0.48158586025238037,
+      "learning_rate": 0.0002,
+      "loss": 1.6491,
+      "step": 2780
+    },
+    {
+      "epoch": 2.0844228614120284,
+      "grad_norm": 0.4612371623516083,
+      "learning_rate": 0.0002,
+      "loss": 1.6301,
+      "step": 2790
+    },
+    {
+      "epoch": 2.091893911094509,
+      "grad_norm": 0.42536866664886475,
+      "learning_rate": 0.0002,
+      "loss": 1.6327,
+      "step": 2800
+    },
+    {
+      "epoch": 2.099364960776989,
+      "grad_norm": 0.48515772819519043,
+      "learning_rate": 0.0002,
+      "loss": 1.651,
+      "step": 2810
+    },
+    {
+      "epoch": 2.1068360104594697,
+      "grad_norm": 0.41418662667274475,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 2820
+    },
+    {
+      "epoch": 2.11430706014195,
+      "grad_norm": 0.4683697819709778,
+      "learning_rate": 0.0002,
+      "loss": 1.6266,
+      "step": 2830
+    },
+    {
+      "epoch": 2.1217781098244304,
+      "grad_norm": 0.4484657049179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6586,
+      "step": 2840
+    },
+    {
+      "epoch": 2.1292491595069105,
+      "grad_norm": 0.6621400713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.6483,
+      "step": 2850
+    },
+    {
+      "epoch": 2.136720209189391,
+      "grad_norm": 0.45074811577796936,
+      "learning_rate": 0.0002,
+      "loss": 1.5755,
+      "step": 2860
+    },
+    {
+      "epoch": 2.1441912588718717,
+      "grad_norm": 0.3513113558292389,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2870
+    },
+    {
+      "epoch": 2.151662308554352,
+      "grad_norm": 0.40411314368247986,
+      "learning_rate": 0.0002,
+      "loss": 1.6081,
+      "step": 2880
+    },
+    {
+      "epoch": 2.1591333582368324,
+      "grad_norm": 0.4121065139770508,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 2890
+    },
+    {
+      "epoch": 2.1666044079193125,
+      "grad_norm": 0.44723689556121826,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 2900
+    },
+    {
+      "epoch": 2.174075457601793,
+      "grad_norm": 0.4226122498512268,
+      "learning_rate": 0.0002,
+      "loss": 1.5699,
+      "step": 2910
+    },
+    {
+      "epoch": 2.1815465072842732,
+      "grad_norm": 0.46617650985717773,
+      "learning_rate": 0.0002,
+      "loss": 1.5652,
+      "step": 2920
+    },
+    {
+      "epoch": 2.189017556966754,
+      "grad_norm": 0.4506422281265259,
+      "learning_rate": 0.0002,
+      "loss": 1.6378,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1964886066492344,
+      "grad_norm": 0.4892672896385193,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 2940
+    },
+    {
+      "epoch": 2.2039596563317145,
+      "grad_norm": 0.44095516204833984,
+      "learning_rate": 0.0002,
+      "loss": 1.6176,
+      "step": 2950
+    },
+    {
+      "epoch": 2.211430706014195,
+      "grad_norm": 0.41522109508514404,
+      "learning_rate": 0.0002,
+      "loss": 1.6058,
+      "step": 2960
+    },
+    {
+      "epoch": 2.2189017556966752,
+      "grad_norm": 0.4860858917236328,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2970
+    },
+    {
+      "epoch": 2.226372805379156,
+      "grad_norm": 0.42662516236305237,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2980
+    },
+    {
+      "epoch": 2.233843855061636,
+      "grad_norm": 0.4390648305416107,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 2990
+    },
+    {
+      "epoch": 2.2413149047441165,
+      "grad_norm": 0.47515565156936646,
+      "learning_rate": 0.0002,
+      "loss": 1.5992,
+      "step": 3000
+    },
+    {
+      "epoch": 2.248785954426597,
+      "grad_norm": 0.4104543924331665,
+      "learning_rate": 0.0002,
+      "loss": 1.5563,
+      "step": 3010
+    },
+    {
+      "epoch": 2.2562570041090773,
+      "grad_norm": 0.4404028654098511,
+      "learning_rate": 0.0002,
+      "loss": 1.6895,
+      "step": 3020
+    },
+    {
+      "epoch": 2.263728053791558,
+      "grad_norm": 0.4717366695404053,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 3030
+    },
+    {
+      "epoch": 2.271199103474038,
+      "grad_norm": 0.48345857858657837,
+      "learning_rate": 0.0002,
+      "loss": 1.7287,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2786701531565186,
+      "grad_norm": 0.5312452912330627,
+      "learning_rate": 0.0002,
+      "loss": 1.681,
+      "step": 3050
+    },
+    {
+      "epoch": 2.2861412028389987,
+      "grad_norm": 0.5073099732398987,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2936122525214793,
+      "grad_norm": 0.5027463436126709,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 3070
+    },
+    {
+      "epoch": 2.30108330220396,
+      "grad_norm": 0.5436304807662964,
+      "learning_rate": 0.0002,
+      "loss": 1.5862,
+      "step": 3080
+    },
+    {
+      "epoch": 2.30855435188644,
+      "grad_norm": 0.4701065123081207,
+      "learning_rate": 0.0002,
+      "loss": 1.5763,
+      "step": 3090
+    },
+    {
+      "epoch": 2.3160254015689206,
+      "grad_norm": 0.46988746523857117,
+      "learning_rate": 0.0002,
+      "loss": 1.6177,
+      "step": 3100
+    },
+    {
+      "epoch": 2.3234964512514007,
+      "grad_norm": 0.45112869143486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6502,
+      "step": 3110
+    },
+    {
+      "epoch": 2.3309675009338813,
+      "grad_norm": 0.5173566937446594,
+      "learning_rate": 0.0002,
+      "loss": 1.6291,
+      "step": 3120
+    },
+    {
+      "epoch": 2.3384385506163614,
+      "grad_norm": 0.40345850586891174,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 3130
+    },
+    {
+      "epoch": 2.345909600298842,
+      "grad_norm": 0.4218924939632416,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 3140
+    },
+    {
+      "epoch": 2.3533806499813226,
+      "grad_norm": 0.41857317090034485,
+      "learning_rate": 0.0002,
+      "loss": 1.6341,
+      "step": 3150
+    },
+    {
+      "epoch": 2.3608516996638027,
+      "grad_norm": 0.4197218418121338,
+      "learning_rate": 0.0002,
+      "loss": 1.6087,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3683227493462833,
+      "grad_norm": 0.4260677397251129,
+      "learning_rate": 0.0002,
+      "loss": 1.6572,
+      "step": 3170
+    },
+    {
+      "epoch": 2.3757937990287634,
+      "grad_norm": 0.4209042191505432,
+      "learning_rate": 0.0002,
+      "loss": 1.6376,
+      "step": 3180
+    },
+    {
+      "epoch": 2.383264848711244,
+      "grad_norm": 0.4092234969139099,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 3190
+    },
+    {
+      "epoch": 2.390735898393724,
+      "grad_norm": 0.4928431510925293,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3982069480762047,
+      "grad_norm": 0.49252402782440186,
+      "learning_rate": 0.0002,
+      "loss": 1.6015,
+      "step": 3210
+    },
+    {
+      "epoch": 2.4056779977586853,
+      "grad_norm": 0.4368397295475006,
+      "learning_rate": 0.0002,
+      "loss": 1.5773,
+      "step": 3220
+    },
+    {
+      "epoch": 2.4131490474411654,
+      "grad_norm": 0.46122390031814575,
+      "learning_rate": 0.0002,
+      "loss": 1.6629,
+      "step": 3230
+    },
+    {
+      "epoch": 2.420620097123646,
+      "grad_norm": 0.4272301197052002,
+      "learning_rate": 0.0002,
+      "loss": 1.6224,
+      "step": 3240
+    },
+    {
+      "epoch": 2.428091146806126,
+      "grad_norm": 0.41480937600135803,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 3250
+    },
+    {
+      "epoch": 2.4355621964886067,
+      "grad_norm": 0.48911941051483154,
+      "learning_rate": 0.0002,
+      "loss": 1.6281,
+      "step": 3260
+    },
+    {
+      "epoch": 2.443033246171087,
+      "grad_norm": 0.4444098472595215,
+      "learning_rate": 0.0002,
+      "loss": 1.6846,
+      "step": 3270
+    },
+    {
+      "epoch": 2.4505042958535674,
+      "grad_norm": 0.5111684799194336,
+      "learning_rate": 0.0002,
+      "loss": 1.6961,
+      "step": 3280
+    },
+    {
+      "epoch": 2.457975345536048,
+      "grad_norm": 0.5058825016021729,
+      "learning_rate": 0.0002,
+      "loss": 1.6152,
+      "step": 3290
+    },
+    {
+      "epoch": 2.465446395218528,
+      "grad_norm": 0.44173210859298706,
+      "learning_rate": 0.0002,
+      "loss": 1.625,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4729174449010087,
+      "grad_norm": 0.4659745991230011,
+      "learning_rate": 0.0002,
+      "loss": 1.6491,
+      "step": 3310
+    },
+    {
+      "epoch": 2.480388494583489,
+      "grad_norm": 0.47237497568130493,
+      "learning_rate": 0.0002,
+      "loss": 1.6114,
+      "step": 3320
+    },
+    {
+      "epoch": 2.4878595442659694,
+      "grad_norm": 0.47303131222724915,
+      "learning_rate": 0.0002,
+      "loss": 1.6193,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4953305939484496,
+      "grad_norm": 0.4522389769554138,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 3340
+    },
+    {
+      "epoch": 2.50280164363093,
+      "grad_norm": 0.4467332363128662,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3350
+    },
+    {
+      "epoch": 2.5102726933134107,
+      "grad_norm": 0.4413762092590332,
+      "learning_rate": 0.0002,
+      "loss": 1.6108,
+      "step": 3360
+    },
+    {
+      "epoch": 2.517743742995891,
+      "grad_norm": 0.495514452457428,
+      "learning_rate": 0.0002,
+      "loss": 1.537,
+      "step": 3370
+    },
+    {
+      "epoch": 2.5252147926783715,
+      "grad_norm": 0.4429773986339569,
+      "learning_rate": 0.0002,
+      "loss": 1.5839,
+      "step": 3380
+    },
+    {
+      "epoch": 2.5326858423608516,
+      "grad_norm": 0.4589079022407532,
+      "learning_rate": 0.0002,
+      "loss": 1.6522,
+      "step": 3390
+    },
+    {
+      "epoch": 2.540156892043332,
+      "grad_norm": 0.4683997333049774,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 3400
+    },
+    {
+      "epoch": 2.5476279417258123,
+      "grad_norm": 0.4651731252670288,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 3410
+    },
+    {
+      "epoch": 2.555098991408293,
+      "grad_norm": 0.45818084478378296,
+      "learning_rate": 0.0002,
+      "loss": 1.5918,
+      "step": 3420
+    },
+    {
+      "epoch": 2.5625700410907735,
+      "grad_norm": 0.45209529995918274,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3430
+    },
+    {
+      "epoch": 2.5700410907732536,
+      "grad_norm": 0.4344733655452728,
+      "learning_rate": 0.0002,
+      "loss": 1.5606,
+      "step": 3440
+    },
+    {
+      "epoch": 2.577512140455734,
+      "grad_norm": 0.47435566782951355,
+      "learning_rate": 0.0002,
+      "loss": 1.6748,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5849831901382143,
+      "grad_norm": 0.43841999769210815,
+      "learning_rate": 0.0002,
+      "loss": 1.6237,
+      "step": 3460
+    },
+    {
+      "epoch": 2.592454239820695,
+      "grad_norm": 0.4323869049549103,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 3470
+    },
+    {
+      "epoch": 2.599925289503175,
+      "grad_norm": 0.44355881214141846,
+      "learning_rate": 0.0002,
+      "loss": 1.5494,
+      "step": 3480
+    },
+    {
+      "epoch": 2.6073963391856556,
+      "grad_norm": 0.45847779512405396,
+      "learning_rate": 0.0002,
+      "loss": 1.665,
+      "step": 3490
+    },
+    {
+      "epoch": 2.614867388868136,
+      "grad_norm": 0.4411061704158783,
+      "learning_rate": 0.0002,
+      "loss": 1.6006,
+      "step": 3500
+    },
+    {
+      "epoch": 2.6223384385506163,
+      "grad_norm": 0.4446796178817749,
+      "learning_rate": 0.0002,
+      "loss": 1.5868,
+      "step": 3510
+    },
+    {
+      "epoch": 2.629809488233097,
+      "grad_norm": 0.41969653964042664,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 3520
+    },
+    {
+      "epoch": 2.637280537915577,
+      "grad_norm": 0.5263747572898865,
+      "learning_rate": 0.0002,
+      "loss": 1.6798,
+      "step": 3530
+    },
+    {
+      "epoch": 2.6447515875980576,
+      "grad_norm": 0.47719451785087585,
+      "learning_rate": 0.0002,
+      "loss": 1.6309,
+      "step": 3540
+    },
+    {
+      "epoch": 2.6522226372805378,
+      "grad_norm": 0.46574118733406067,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 3550
+    },
+    {
+      "epoch": 2.6596936869630183,
+      "grad_norm": 0.46867135167121887,
+      "learning_rate": 0.0002,
+      "loss": 1.618,
+      "step": 3560
+    },
+    {
+      "epoch": 2.667164736645499,
+      "grad_norm": 0.4441198706626892,
+      "learning_rate": 0.0002,
+      "loss": 1.5885,
+      "step": 3570
+    },
+    {
+      "epoch": 2.674635786327979,
+      "grad_norm": 0.4871319830417633,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6821068360104596,
+      "grad_norm": 0.43900373578071594,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6895778856929398,
+      "grad_norm": 0.42509549856185913,
+      "learning_rate": 0.0002,
+      "loss": 1.6071,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6970489353754203,
+      "grad_norm": 0.4691086709499359,
+      "learning_rate": 0.0002,
+      "loss": 1.5651,
+      "step": 3610
+    },
+    {
+      "epoch": 2.7045199850579005,
+      "grad_norm": 0.46318942308425903,
+      "learning_rate": 0.0002,
+      "loss": 1.5491,
+      "step": 3620
+    },
+    {
+      "epoch": 2.711991034740381,
+      "grad_norm": 0.44631096720695496,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 3630
+    },
+    {
+      "epoch": 2.7194620844228616,
+      "grad_norm": 0.42315489053726196,
+      "learning_rate": 0.0002,
+      "loss": 1.6831,
+      "step": 3640
+    },
+    {
+      "epoch": 2.7269331341053418,
+      "grad_norm": 0.4971241056919098,
+      "learning_rate": 0.0002,
+      "loss": 1.6008,
+      "step": 3650
+    },
+    {
+      "epoch": 2.7344041837878224,
+      "grad_norm": 0.4578486382961273,
+      "learning_rate": 0.0002,
+      "loss": 1.6042,
+      "step": 3660
+    },
+    {
+      "epoch": 2.7418752334703025,
+      "grad_norm": 0.46584776043891907,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 3670
+    },
+    {
+      "epoch": 2.749346283152783,
+      "grad_norm": 0.4951731264591217,
+      "learning_rate": 0.0002,
+      "loss": 1.6809,
+      "step": 3680
+    },
+    {
+      "epoch": 2.756817332835263,
+      "grad_norm": 0.4935225546360016,
+      "learning_rate": 0.0002,
+      "loss": 1.6226,
+      "step": 3690
+    },
+    {
+      "epoch": 2.764288382517744,
+      "grad_norm": 0.41805586218833923,
+      "learning_rate": 0.0002,
+      "loss": 1.5878,
+      "step": 3700
+    },
+    {
+      "epoch": 2.7717594322002244,
+      "grad_norm": 0.4417555630207062,
+      "learning_rate": 0.0002,
+      "loss": 1.7173,
+      "step": 3710
+    },
+    {
+      "epoch": 2.7792304818827045,
+      "grad_norm": 0.48229655623435974,
+      "learning_rate": 0.0002,
+      "loss": 1.6398,
+      "step": 3720
+    },
+    {
+      "epoch": 2.786701531565185,
+      "grad_norm": 0.48562315106391907,
+      "learning_rate": 0.0002,
+      "loss": 1.6074,
+      "step": 3730
+    },
+    {
+      "epoch": 2.794172581247665,
+      "grad_norm": 0.4473940432071686,
+      "learning_rate": 0.0002,
+      "loss": 1.607,
+      "step": 3740
+    },
+    {
+      "epoch": 2.801643630930146,
+      "grad_norm": 0.4626813232898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6065,
+      "step": 3750
+    },
+    {
+      "epoch": 2.809114680612626,
+      "grad_norm": 0.4339792728424072,
+      "learning_rate": 0.0002,
+      "loss": 1.6296,
+      "step": 3760
+    },
+    {
+      "epoch": 2.8165857302951065,
+      "grad_norm": 0.5250858068466187,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 3770
+    },
+    {
+      "epoch": 2.824056779977587,
+      "grad_norm": 0.4537523090839386,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 3780
+    },
+    {
+      "epoch": 2.831527829660067,
+      "grad_norm": 0.5646113157272339,
+      "learning_rate": 0.0002,
+      "loss": 1.6535,
+      "step": 3790
+    },
+    {
+      "epoch": 2.8389988793425474,
+      "grad_norm": 0.44243332743644714,
+      "learning_rate": 0.0002,
+      "loss": 1.5712,
+      "step": 3800
+    },
+    {
+      "epoch": 2.846469929025028,
+      "grad_norm": 0.4585791826248169,
+      "learning_rate": 0.0002,
+      "loss": 1.6478,
+      "step": 3810
+    },
+    {
+      "epoch": 2.8539409787075085,
+      "grad_norm": 0.489702045917511,
+      "learning_rate": 0.0002,
+      "loss": 1.6854,
+      "step": 3820
+    },
+    {
+      "epoch": 2.8614120283899886,
+      "grad_norm": 0.502470850944519,
+      "learning_rate": 0.0002,
+      "loss": 1.7066,
+      "step": 3830
+    },
+    {
+      "epoch": 2.8688830780724692,
+      "grad_norm": 0.4395960867404938,
+      "learning_rate": 0.0002,
+      "loss": 1.5785,
+      "step": 3840
+    },
+    {
+      "epoch": 2.87635412775495,
+      "grad_norm": 0.4348670244216919,
+      "learning_rate": 0.0002,
+      "loss": 1.6434,
+      "step": 3850
+    },
+    {
+      "epoch": 2.88382517743743,
+      "grad_norm": 0.48852720856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 3860
+    },
+    {
+      "epoch": 2.89129622711991,
+      "grad_norm": 0.45317450165748596,
+      "learning_rate": 0.0002,
+      "loss": 1.5916,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8987672768023907,
+      "grad_norm": 0.4732758700847626,
+      "learning_rate": 0.0002,
+      "loss": 1.6486,
+      "step": 3880
+    },
+    {
+      "epoch": 2.9062383264848712,
+      "grad_norm": 0.45238012075424194,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 3890
+    },
+    {
+      "epoch": 2.9137093761673514,
+      "grad_norm": 0.48838064074516296,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 3900
+    },
+    {
+      "epoch": 2.921180425849832,
+      "grad_norm": 0.43496349453926086,
+      "learning_rate": 0.0002,
+      "loss": 1.658,
+      "step": 3910
+    },
+    {
+      "epoch": 2.9286514755323125,
+      "grad_norm": 0.47963935136795044,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 3920
+    },
+    {
+      "epoch": 2.9361225252147927,
+      "grad_norm": 0.4544987976551056,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 3930
+    },
+    {
+      "epoch": 2.943593574897273,
+      "grad_norm": 0.4622892141342163,
+      "learning_rate": 0.0002,
+      "loss": 1.6192,
+      "step": 3940
+    },
+    {
+      "epoch": 2.9510646245797534,
+      "grad_norm": 0.47026222944259644,
+      "learning_rate": 0.0002,
+      "loss": 1.6178,
+      "step": 3950
+    },
+    {
+      "epoch": 2.958535674262234,
+      "grad_norm": 0.4549552798271179,
+      "learning_rate": 0.0002,
+      "loss": 1.6612,
+      "step": 3960
+    },
+    {
+      "epoch": 2.966006723944714,
+      "grad_norm": 0.46647515892982483,
+      "learning_rate": 0.0002,
+      "loss": 1.6458,
+      "step": 3970
+    },
+    {
+      "epoch": 2.9734777736271947,
+      "grad_norm": 0.45095112919807434,
+      "learning_rate": 0.0002,
+      "loss": 1.6051,
+      "step": 3980
+    },
+    {
+      "epoch": 2.9809488233096753,
+      "grad_norm": 0.4690017104148865,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 3990
+    },
+    {
+      "epoch": 2.9884198729921554,
+      "grad_norm": 0.4603444039821625,
+      "learning_rate": 0.0002,
+      "loss": 1.6061,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9958909226746355,
+      "grad_norm": 0.4743294417858124,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 4010
+    },
+    {
+      "epoch": 2.999626447515876,
+      "eval_loss": 1.8252571821212769,
+      "eval_runtime": 38.7853,
+      "eval_samples_per_second": 13.278,
+      "eval_steps_per_second": 1.676,
+      "step": 4015
+    },
+    {
+      "epoch": 3.003361972357116,
+      "grad_norm": 0.4919724464416504,
+      "learning_rate": 0.0002,
+      "loss": 1.6512,
+      "step": 4020
+    },
+    {
+      "epoch": 3.0108330220395967,
+      "grad_norm": 0.4747185707092285,
+      "learning_rate": 0.0002,
+      "loss": 1.5354,
+      "step": 4030
+    },
+    {
+      "epoch": 3.018304071722077,
+      "grad_norm": 0.4797595143318176,
+      "learning_rate": 0.0002,
+      "loss": 1.568,
+      "step": 4040
+    },
+    {
+      "epoch": 3.0257751214045574,
+      "grad_norm": 0.5450999140739441,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 4050
+    },
+    {
+      "epoch": 3.0332461710870375,
+      "grad_norm": 0.49058812856674194,
+      "learning_rate": 0.0002,
+      "loss": 1.5065,
+      "step": 4060
+    },
+    {
+      "epoch": 3.040717220769518,
+      "grad_norm": 0.5219563841819763,
+      "learning_rate": 0.0002,
+      "loss": 1.4884,
+      "step": 4070
+    },
+    {
+      "epoch": 3.0481882704519987,
+      "grad_norm": 0.515628457069397,
+      "learning_rate": 0.0002,
+      "loss": 1.4742,
+      "step": 4080
+    },
+    {
+      "epoch": 3.055659320134479,
+      "grad_norm": 0.6145984530448914,
+      "learning_rate": 0.0002,
+      "loss": 1.5313,
+      "step": 4090
+    },
+    {
+      "epoch": 3.0631303698169594,
+      "grad_norm": 0.6067144274711609,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 4100
+    },
+    {
+      "epoch": 3.0706014194994395,
+      "grad_norm": 0.5773133039474487,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 4110
+    },
+    {
+      "epoch": 3.07807246918192,
+      "grad_norm": 0.6894241571426392,
+      "learning_rate": 0.0002,
+      "loss": 1.5374,
+      "step": 4120
+    },
+    {
+      "epoch": 3.0855435188644003,
+      "grad_norm": 0.6422514915466309,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 4130
+    },
+    {
+      "epoch": 3.093014568546881,
+      "grad_norm": 0.6119855046272278,
+      "learning_rate": 0.0002,
+      "loss": 1.4724,
+      "step": 4140
+    },
+    {
+      "epoch": 3.1004856182293614,
+      "grad_norm": 0.5847280025482178,
+      "learning_rate": 0.0002,
+      "loss": 1.5361,
+      "step": 4150
+    },
+    {
+      "epoch": 3.1079566679118416,
+      "grad_norm": 0.5401515960693359,
+      "learning_rate": 0.0002,
+      "loss": 1.5151,
+      "step": 4160
+    },
+    {
+      "epoch": 3.115427717594322,
+      "grad_norm": 0.6501587629318237,
+      "learning_rate": 0.0002,
+      "loss": 1.502,
+      "step": 4170
+    },
+    {
+      "epoch": 3.1228987672768023,
+      "grad_norm": 0.5988039374351501,
+      "learning_rate": 0.0002,
+      "loss": 1.4952,
+      "step": 4180
+    },
+    {
+      "epoch": 3.130369816959283,
+      "grad_norm": 0.4982665181159973,
+      "learning_rate": 0.0002,
+      "loss": 1.5287,
+      "step": 4190
+    },
+    {
+      "epoch": 3.137840866641763,
+      "grad_norm": 0.5548039078712463,
+      "learning_rate": 0.0002,
+      "loss": 1.5078,
+      "step": 4200
+    },
+    {
+      "epoch": 3.1453119163242436,
+      "grad_norm": 0.5920777320861816,
+      "learning_rate": 0.0002,
+      "loss": 1.4904,
+      "step": 4210
+    },
+    {
+      "epoch": 3.152782966006724,
+      "grad_norm": 0.6965190172195435,
+      "learning_rate": 0.0002,
+      "loss": 1.442,
+      "step": 4220
+    },
+    {
+      "epoch": 3.1602540156892043,
+      "grad_norm": 0.5196244716644287,
+      "learning_rate": 0.0002,
+      "loss": 1.557,
+      "step": 4230
+    },
+    {
+      "epoch": 3.167725065371685,
+      "grad_norm": 0.6942682266235352,
+      "learning_rate": 0.0002,
+      "loss": 1.5706,
+      "step": 4240
+    },
+    {
+      "epoch": 3.175196115054165,
+      "grad_norm": 0.5765156149864197,
+      "learning_rate": 0.0002,
+      "loss": 1.5407,
+      "step": 4250
+    },
+    {
+      "epoch": 3.1826671647366456,
+      "grad_norm": 0.5801976919174194,
+      "learning_rate": 0.0002,
+      "loss": 1.4963,
+      "step": 4260
+    },
+    {
+      "epoch": 3.1901382144191257,
+      "grad_norm": 0.6260752081871033,
+      "learning_rate": 0.0002,
+      "loss": 1.4988,
+      "step": 4270
+    },
+    {
+      "epoch": 3.1976092641016063,
+      "grad_norm": 0.6610770225524902,
+      "learning_rate": 0.0002,
+      "loss": 1.5074,
+      "step": 4280
+    },
+    {
+      "epoch": 3.205080313784087,
+      "grad_norm": 0.5762143135070801,
+      "learning_rate": 0.0002,
+      "loss": 1.4657,
+      "step": 4290
+    },
+    {
+      "epoch": 3.212551363466567,
+      "grad_norm": 0.5926990509033203,
+      "learning_rate": 0.0002,
+      "loss": 1.5181,
+      "step": 4300
+    },
+    {
+      "epoch": 3.2200224131490476,
+      "grad_norm": 0.7373854517936707,
+      "learning_rate": 0.0002,
+      "loss": 1.5492,
+      "step": 4310
+    },
+    {
+      "epoch": 3.2274934628315277,
+      "grad_norm": 0.5963311195373535,
+      "learning_rate": 0.0002,
+      "loss": 1.4648,
+      "step": 4320
+    },
+    {
+      "epoch": 3.2349645125140083,
+      "grad_norm": 0.5754616856575012,
+      "learning_rate": 0.0002,
+      "loss": 1.5262,
+      "step": 4330
+    },
+    {
+      "epoch": 3.2424355621964884,
+      "grad_norm": 0.6116095781326294,
+      "learning_rate": 0.0002,
+      "loss": 1.4767,
+      "step": 4340
+    },
+    {
+      "epoch": 3.249906611878969,
+      "grad_norm": 0.6001536846160889,
+      "learning_rate": 0.0002,
+      "loss": 1.5008,
+      "step": 4350
+    },
+    {
+      "epoch": 3.257377661561449,
+      "grad_norm": 0.5270227789878845,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 4360
+    },
+    {
+      "epoch": 3.2648487112439297,
+      "grad_norm": 0.6666602492332458,
+      "learning_rate": 0.0002,
+      "loss": 1.5235,
+      "step": 4370
+    },
+    {
+      "epoch": 3.2723197609264103,
+      "grad_norm": 0.520310640335083,
+      "learning_rate": 0.0002,
+      "loss": 1.5665,
+      "step": 4380
+    },
+    {
+      "epoch": 3.2797908106088904,
+      "grad_norm": 0.5165975093841553,
+      "learning_rate": 0.0002,
+      "loss": 1.542,
+      "step": 4390
+    },
+    {
+      "epoch": 3.287261860291371,
+      "grad_norm": 0.6080228686332703,
+      "learning_rate": 0.0002,
+      "loss": 1.4746,
+      "step": 4400
+    },
+    {
+      "epoch": 3.294732909973851,
+      "grad_norm": 0.670122504234314,
+      "learning_rate": 0.0002,
+      "loss": 1.4901,
+      "step": 4410
+    },
+    {
+      "epoch": 3.3022039596563317,
+      "grad_norm": 0.6019457578659058,
+      "learning_rate": 0.0002,
+      "loss": 1.4677,
+      "step": 4420
+    },
+    {
+      "epoch": 3.309675009338812,
+      "grad_norm": 0.5519300103187561,
+      "learning_rate": 0.0002,
+      "loss": 1.4249,
+      "step": 4430
+    },
+    {
+      "epoch": 3.3171460590212924,
+      "grad_norm": 0.5958521962165833,
+      "learning_rate": 0.0002,
+      "loss": 1.555,
+      "step": 4440
+    },
+    {
+      "epoch": 3.324617108703773,
+      "grad_norm": 0.5552705526351929,
+      "learning_rate": 0.0002,
+      "loss": 1.5067,
+      "step": 4450
+    },
+    {
+      "epoch": 3.332088158386253,
+      "grad_norm": 0.6583784818649292,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 4460
+    },
+    {
+      "epoch": 3.3395592080687337,
+      "grad_norm": 0.5815939903259277,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 4470
+    },
+    {
+      "epoch": 3.347030257751214,
+      "grad_norm": 1.3342205286026,
+      "learning_rate": 0.0002,
+      "loss": 1.5942,
+      "step": 4480
+    },
+    {
+      "epoch": 3.3545013074336945,
+      "grad_norm": 0.6341500878334045,
+      "learning_rate": 0.0002,
+      "loss": 1.484,
+      "step": 4490
+    },
+    {
+      "epoch": 3.3619723571161746,
+      "grad_norm": 0.6384079456329346,
+      "learning_rate": 0.0002,
+      "loss": 1.5219,
+      "step": 4500
+    },
+    {
+      "epoch": 3.369443406798655,
+      "grad_norm": 0.6098346710205078,
+      "learning_rate": 0.0002,
+      "loss": 1.5222,
+      "step": 4510
+    },
+    {
+      "epoch": 3.3769144564811358,
+      "grad_norm": 0.5958296656608582,
+      "learning_rate": 0.0002,
+      "loss": 1.5475,
+      "step": 4520
+    },
+    {
+      "epoch": 3.384385506163616,
+      "grad_norm": 0.6157881617546082,
+      "learning_rate": 0.0002,
+      "loss": 1.5171,
+      "step": 4530
+    },
+    {
+      "epoch": 3.3918565558460965,
+      "grad_norm": 0.5671007037162781,
+      "learning_rate": 0.0002,
+      "loss": 1.569,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3993276055285766,
+      "grad_norm": 0.6203294992446899,
+      "learning_rate": 0.0002,
+      "loss": 1.604,
+      "step": 4550
+    },
+    {
+      "epoch": 3.406798655211057,
+      "grad_norm": 0.6743317246437073,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 4560
+    },
+    {
+      "epoch": 3.4142697048935373,
+      "grad_norm": 0.731765627861023,
+      "learning_rate": 0.0002,
+      "loss": 1.5034,
+      "step": 4570
+    },
+    {
+      "epoch": 3.421740754576018,
+      "grad_norm": 0.6285187602043152,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 4580
+    },
+    {
+      "epoch": 3.4292118042584985,
+      "grad_norm": 0.612680196762085,
+      "learning_rate": 0.0002,
+      "loss": 1.5296,
+      "step": 4590
+    },
+    {
+      "epoch": 3.4366828539409786,
+      "grad_norm": 0.6413681507110596,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 4600
+    },
+    {
+      "epoch": 3.444153903623459,
+      "grad_norm": 0.6240990161895752,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4610
+    },
+    {
+      "epoch": 3.4516249533059393,
+      "grad_norm": 0.5095735192298889,
+      "learning_rate": 0.0002,
+      "loss": 1.5887,
+      "step": 4620
+    },
+    {
+      "epoch": 3.45909600298842,
+      "grad_norm": 0.5699611902236938,
+      "learning_rate": 0.0002,
+      "loss": 1.4906,
+      "step": 4630
+    },
+    {
+      "epoch": 3.4665670526709,
+      "grad_norm": 0.7289775609970093,
+      "learning_rate": 0.0002,
+      "loss": 1.5176,
+      "step": 4640
+    },
+    {
+      "epoch": 3.4740381023533806,
+      "grad_norm": 0.6211609840393066,
+      "learning_rate": 0.0002,
+      "loss": 1.5467,
+      "step": 4650
+    },
+    {
+      "epoch": 3.481509152035861,
+      "grad_norm": 0.5714802145957947,
+      "learning_rate": 0.0002,
+      "loss": 1.533,
+      "step": 4660
+    },
+    {
+      "epoch": 3.4889802017183413,
+      "grad_norm": 0.6287049651145935,
+      "learning_rate": 0.0002,
+      "loss": 1.5096,
+      "step": 4670
+    },
+    {
+      "epoch": 3.496451251400822,
+      "grad_norm": 0.5480595827102661,
+      "learning_rate": 0.0002,
+      "loss": 1.4212,
+      "step": 4680
+    },
+    {
+      "epoch": 3.503922301083302,
+      "grad_norm": 0.5683253407478333,
+      "learning_rate": 0.0002,
+      "loss": 1.4746,
+      "step": 4690
+    },
+    {
+      "epoch": 3.5113933507657826,
+      "grad_norm": 0.601140558719635,
+      "learning_rate": 0.0002,
+      "loss": 1.5012,
+      "step": 4700
+    },
+    {
+      "epoch": 3.5188644004482628,
+      "grad_norm": 0.5344498157501221,
+      "learning_rate": 0.0002,
+      "loss": 1.5383,
+      "step": 4710
+    },
+    {
+      "epoch": 3.5263354501307433,
+      "grad_norm": 0.5739690661430359,
+      "learning_rate": 0.0002,
+      "loss": 1.5428,
+      "step": 4720
+    },
+    {
+      "epoch": 3.533806499813224,
+      "grad_norm": 0.5640085935592651,
+      "learning_rate": 0.0002,
+      "loss": 1.5589,
+      "step": 4730
+    },
+    {
+      "epoch": 3.541277549495704,
+      "grad_norm": 0.5967805981636047,
+      "learning_rate": 0.0002,
+      "loss": 1.487,
+      "step": 4740
+    },
+    {
+      "epoch": 3.5487485991781846,
+      "grad_norm": 0.6138835549354553,
+      "learning_rate": 0.0002,
+      "loss": 1.5461,
+      "step": 4750
+    },
+    {
+      "epoch": 3.5562196488606648,
+      "grad_norm": 0.6779900193214417,
+      "learning_rate": 0.0002,
+      "loss": 1.5502,
+      "step": 4760
+    },
+    {
+      "epoch": 3.5636906985431454,
+      "grad_norm": 0.6122010350227356,
+      "learning_rate": 0.0002,
+      "loss": 1.4917,
+      "step": 4770
+    },
+    {
+      "epoch": 3.5711617482256255,
+      "grad_norm": 0.5685241222381592,
+      "learning_rate": 0.0002,
+      "loss": 1.5405,
+      "step": 4780
+    },
+    {
+      "epoch": 3.578632797908106,
+      "grad_norm": 0.604583203792572,
+      "learning_rate": 0.0002,
+      "loss": 1.5427,
+      "step": 4790
+    },
+    {
+      "epoch": 3.5861038475905866,
+      "grad_norm": 0.651165246963501,
+      "learning_rate": 0.0002,
+      "loss": 1.4514,
+      "step": 4800
+    },
+    {
+      "epoch": 3.593574897273067,
+      "grad_norm": 0.6398511528968811,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 4810
+    },
+    {
+      "epoch": 3.6010459469555474,
+      "grad_norm": 0.6444641351699829,
+      "learning_rate": 0.0002,
+      "loss": 1.4261,
+      "step": 4820
+    },
+    {
+      "epoch": 3.6085169966380275,
+      "grad_norm": 0.6018481850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.5274,
+      "step": 4830
+    },
+    {
+      "epoch": 3.615988046320508,
+      "grad_norm": 0.6025291085243225,
+      "learning_rate": 0.0002,
+      "loss": 1.4647,
+      "step": 4840
+    },
+    {
+      "epoch": 3.623459096002988,
+      "grad_norm": 0.6810156106948853,
+      "learning_rate": 0.0002,
+      "loss": 1.5609,
+      "step": 4850
+    },
+    {
+      "epoch": 3.630930145685469,
+      "grad_norm": 0.6408044695854187,
+      "learning_rate": 0.0002,
+      "loss": 1.5299,
+      "step": 4860
+    },
+    {
+      "epoch": 3.6384011953679494,
+      "grad_norm": 0.5608272552490234,
+      "learning_rate": 0.0002,
+      "loss": 1.5366,
+      "step": 4870
+    },
+    {
+      "epoch": 3.6458722450504295,
+      "grad_norm": 0.6136814951896667,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 4880
+    },
+    {
+      "epoch": 3.65334329473291,
+      "grad_norm": 0.5927900075912476,
+      "learning_rate": 0.0002,
+      "loss": 1.5021,
+      "step": 4890
+    },
+    {
+      "epoch": 3.66081434441539,
+      "grad_norm": 0.5336901545524597,
+      "learning_rate": 0.0002,
+      "loss": 1.6084,
+      "step": 4900
+    },
+    {
+      "epoch": 3.668285394097871,
+      "grad_norm": 0.7823320627212524,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 4910
+    },
+    {
+      "epoch": 3.675756443780351,
+      "grad_norm": 0.6703504323959351,
+      "learning_rate": 0.0002,
+      "loss": 1.4881,
+      "step": 4920
+    },
+    {
+      "epoch": 3.6832274934628315,
+      "grad_norm": 0.6061160564422607,
+      "learning_rate": 0.0002,
+      "loss": 1.5332,
+      "step": 4930
+    },
+    {
+      "epoch": 3.690698543145312,
+      "grad_norm": 0.6237227916717529,
+      "learning_rate": 0.0002,
+      "loss": 1.5405,
+      "step": 4940
+    },
+    {
+      "epoch": 3.6981695928277922,
+      "grad_norm": 0.5985278487205505,
+      "learning_rate": 0.0002,
+      "loss": 1.497,
+      "step": 4950
+    },
+    {
+      "epoch": 3.705640642510273,
+      "grad_norm": 0.6483839750289917,
+      "learning_rate": 0.0002,
+      "loss": 1.5132,
+      "step": 4960
+    },
+    {
+      "epoch": 3.713111692192753,
+      "grad_norm": 0.5788805484771729,
+      "learning_rate": 0.0002,
+      "loss": 1.5338,
+      "step": 4970
+    },
+    {
+      "epoch": 3.7205827418752335,
+      "grad_norm": 0.5609974265098572,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 4980
+    },
+    {
+      "epoch": 3.7280537915577137,
+      "grad_norm": 0.5681300759315491,
+      "learning_rate": 0.0002,
+      "loss": 1.4759,
+      "step": 4990
+    },
+    {
+      "epoch": 3.7355248412401942,
+      "grad_norm": 0.5860186219215393,
+      "learning_rate": 0.0002,
+      "loss": 1.6018,
+      "step": 5000
+    },
+    {
+      "epoch": 3.742995890922675,
+      "grad_norm": 0.5718157291412354,
+      "learning_rate": 0.0002,
+      "loss": 1.58,
+      "step": 5010
+    },
+    {
+      "epoch": 3.750466940605155,
+      "grad_norm": 0.6173721551895142,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 5020
+    },
+    {
+      "epoch": 3.7579379902876355,
+      "grad_norm": 0.629152238368988,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 5030
+    },
+    {
+      "epoch": 3.7654090399701157,
+      "grad_norm": 0.5666284561157227,
+      "learning_rate": 0.0002,
+      "loss": 1.519,
+      "step": 5040
+    },
+    {
+      "epoch": 3.7728800896525962,
+      "grad_norm": 0.6053005456924438,
+      "learning_rate": 0.0002,
+      "loss": 1.5329,
+      "step": 5050
+    },
+    {
+      "epoch": 3.7803511393350764,
+      "grad_norm": 0.5870583057403564,
+      "learning_rate": 0.0002,
+      "loss": 1.5404,
+      "step": 5060
+    },
+    {
+      "epoch": 3.787822189017557,
+      "grad_norm": 0.5422009229660034,
+      "learning_rate": 0.0002,
+      "loss": 1.4444,
+      "step": 5070
+    },
+    {
+      "epoch": 3.7952932387000375,
+      "grad_norm": 0.5396918058395386,
+      "learning_rate": 0.0002,
+      "loss": 1.5308,
+      "step": 5080
+    },
+    {
+      "epoch": 3.8027642883825177,
+      "grad_norm": 0.5544713139533997,
+      "learning_rate": 0.0002,
+      "loss": 1.464,
+      "step": 5090
+    },
+    {
+      "epoch": 3.8102353380649983,
+      "grad_norm": 0.5983749628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5100
+    },
+    {
+      "epoch": 3.8177063877474784,
+      "grad_norm": 0.5702024102210999,
+      "learning_rate": 0.0002,
+      "loss": 1.4972,
+      "step": 5110
+    },
+    {
+      "epoch": 3.825177437429959,
+      "grad_norm": 0.5436882376670837,
+      "learning_rate": 0.0002,
+      "loss": 1.5471,
+      "step": 5120
+    },
+    {
+      "epoch": 3.832648487112439,
+      "grad_norm": 0.5453617572784424,
+      "learning_rate": 0.0002,
+      "loss": 1.5118,
+      "step": 5130
+    },
+    {
+      "epoch": 3.8401195367949197,
+      "grad_norm": 0.6269069314002991,
+      "learning_rate": 0.0002,
+      "loss": 1.5732,
+      "step": 5140
+    },
+    {
+      "epoch": 3.8475905864774003,
+      "grad_norm": 0.6189185380935669,
+      "learning_rate": 0.0002,
+      "loss": 1.4959,
+      "step": 5150
+    },
+    {
+      "epoch": 3.8550616361598804,
+      "grad_norm": 0.6653388142585754,
+      "learning_rate": 0.0002,
+      "loss": 1.4999,
+      "step": 5160
+    },
+    {
+      "epoch": 3.862532685842361,
+      "grad_norm": 0.5771768689155579,
+      "learning_rate": 0.0002,
+      "loss": 1.5075,
+      "step": 5170
+    },
+    {
+      "epoch": 3.870003735524841,
+      "grad_norm": 0.6052790880203247,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5180
+    },
+    {
+      "epoch": 3.8774747852073217,
+      "grad_norm": 0.6572316884994507,
+      "learning_rate": 0.0002,
+      "loss": 1.4987,
+      "step": 5190
+    },
+    {
+      "epoch": 3.884945834889802,
+      "grad_norm": 0.670576810836792,
+      "learning_rate": 0.0002,
+      "loss": 1.5241,
+      "step": 5200
+    },
+    {
+      "epoch": 3.8924168845722824,
+      "grad_norm": 0.5728798508644104,
+      "learning_rate": 0.0002,
+      "loss": 1.4777,
+      "step": 5210
+    },
+    {
+      "epoch": 3.899887934254763,
+      "grad_norm": 0.6340774297714233,
+      "learning_rate": 0.0002,
+      "loss": 1.5351,
+      "step": 5220
+    },
+    {
+      "epoch": 3.907358983937243,
+      "grad_norm": 0.5981315970420837,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 5230
+    },
+    {
+      "epoch": 3.9148300336197237,
+      "grad_norm": 0.6212025880813599,
+      "learning_rate": 0.0002,
+      "loss": 1.4875,
+      "step": 5240
+    },
+    {
+      "epoch": 3.922301083302204,
+      "grad_norm": 0.6202296018600464,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5250
+    },
+    {
+      "epoch": 3.9297721329846844,
+      "grad_norm": 0.6159142255783081,
+      "learning_rate": 0.0002,
+      "loss": 1.5765,
+      "step": 5260
+    },
+    {
+      "epoch": 3.9372431826671646,
+      "grad_norm": 0.6519438624382019,
+      "learning_rate": 0.0002,
+      "loss": 1.4938,
+      "step": 5270
+    },
+    {
+      "epoch": 3.944714232349645,
+      "grad_norm": 0.539813756942749,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 5280
+    },
+    {
+      "epoch": 3.9521852820321257,
+      "grad_norm": 0.6443665027618408,
+      "learning_rate": 0.0002,
+      "loss": 1.5921,
+      "step": 5290
+    },
+    {
+      "epoch": 3.959656331714606,
+      "grad_norm": 0.6635757684707642,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 5300
+    },
+    {
+      "epoch": 3.9671273813970864,
+      "grad_norm": 0.589363157749176,
+      "learning_rate": 0.0002,
+      "loss": 1.5485,
+      "step": 5310
+    },
+    {
+      "epoch": 3.9745984310795666,
+      "grad_norm": 0.5788735747337341,
+      "learning_rate": 0.0002,
+      "loss": 1.5498,
+      "step": 5320
+    },
+    {
+      "epoch": 3.982069480762047,
+      "grad_norm": 0.5976864695549011,
+      "learning_rate": 0.0002,
+      "loss": 1.5607,
+      "step": 5330
+    },
+    {
+      "epoch": 3.9895405304445273,
+      "grad_norm": 0.6624067425727844,
+      "learning_rate": 0.0002,
+      "loss": 1.5302,
+      "step": 5340
+    },
+    {
+      "epoch": 3.997011580127008,
+      "grad_norm": 0.6738956570625305,
+      "learning_rate": 0.0002,
+      "loss": 1.5904,
+      "step": 5350
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.868006944656372,
+      "eval_runtime": 38.5153,
+      "eval_samples_per_second": 13.371,
+      "eval_steps_per_second": 1.688,
+      "step": 5354
+    },
+    {
+      "epoch": 4.004482629809488,
+      "grad_norm": 0.6023468971252441,
+      "learning_rate": 0.0002,
+      "loss": 1.4535,
+      "step": 5360
+    },
+    {
+      "epoch": 4.011953679491969,
+      "grad_norm": 0.8589285612106323,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 5370
+    },
+    {
+      "epoch": 4.019424729174449,
+      "grad_norm": 0.7477491497993469,
+      "learning_rate": 0.0002,
+      "loss": 1.3952,
+      "step": 5380
+    },
+    {
+      "epoch": 4.02689577885693,
+      "grad_norm": 0.7601922154426575,
+      "learning_rate": 0.0002,
+      "loss": 1.3745,
+      "step": 5390
+    },
+    {
+      "epoch": 4.03436682853941,
+      "grad_norm": 0.8115614056587219,
+      "learning_rate": 0.0002,
+      "loss": 1.4133,
+      "step": 5400
+    },
+    {
+      "epoch": 4.04183787822189,
+      "grad_norm": 0.669925332069397,
+      "learning_rate": 0.0002,
+      "loss": 1.3748,
+      "step": 5410
+    },
+    {
+      "epoch": 4.04930892790437,
+      "grad_norm": 0.8091904520988464,
+      "learning_rate": 0.0002,
+      "loss": 1.2835,
+      "step": 5420
+    },
+    {
+      "epoch": 4.056779977586851,
+      "grad_norm": 0.709405779838562,
+      "learning_rate": 0.0002,
+      "loss": 1.3615,
+      "step": 5430
+    },
+    {
+      "epoch": 4.064251027269331,
+      "grad_norm": 1.0006179809570312,
+      "learning_rate": 0.0002,
+      "loss": 1.3558,
+      "step": 5440
+    },
+    {
+      "epoch": 4.071722076951811,
+      "grad_norm": 0.7017965912818909,
+      "learning_rate": 0.0002,
+      "loss": 1.3491,
+      "step": 5450
+    },
+    {
+      "epoch": 4.0791931266342925,
+      "grad_norm": 0.8991572260856628,
+      "learning_rate": 0.0002,
+      "loss": 1.3642,
+      "step": 5460
+    },
+    {
+      "epoch": 4.086664176316773,
+      "grad_norm": 0.9064797759056091,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 5470
+    },
+    {
+      "epoch": 4.094135225999253,
+      "grad_norm": 0.7981749176979065,
+      "learning_rate": 0.0002,
+      "loss": 1.3425,
+      "step": 5480
+    },
+    {
+      "epoch": 4.101606275681733,
+      "grad_norm": 0.7280883193016052,
+      "learning_rate": 0.0002,
+      "loss": 1.3826,
+      "step": 5490
+    },
+    {
+      "epoch": 4.109077325364214,
+      "grad_norm": 0.7419600486755371,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 5500
+    },
+    {
+      "epoch": 4.116548375046694,
+      "grad_norm": 0.8019949197769165,
+      "learning_rate": 0.0002,
+      "loss": 1.3199,
+      "step": 5510
+    },
+    {
+      "epoch": 4.124019424729174,
+      "grad_norm": 0.7501229047775269,
+      "learning_rate": 0.0002,
+      "loss": 1.3133,
+      "step": 5520
+    },
+    {
+      "epoch": 4.131490474411655,
+      "grad_norm": 0.8166249990463257,
+      "learning_rate": 0.0002,
+      "loss": 1.4432,
+      "step": 5530
+    },
+    {
+      "epoch": 4.138961524094135,
+      "grad_norm": 0.9728496074676514,
+      "learning_rate": 0.0002,
+      "loss": 1.3901,
+      "step": 5540
+    },
+    {
+      "epoch": 4.1464325737766154,
+      "grad_norm": 0.7590922117233276,
+      "learning_rate": 0.0002,
+      "loss": 1.3538,
+      "step": 5550
+    },
+    {
+      "epoch": 4.153903623459096,
+      "grad_norm": 0.7759010791778564,
+      "learning_rate": 0.0002,
+      "loss": 1.4368,
+      "step": 5560
+    },
+    {
+      "epoch": 4.161374673141577,
+      "grad_norm": 0.9057986736297607,
+      "learning_rate": 0.0002,
+      "loss": 1.3635,
+      "step": 5570
+    },
+    {
+      "epoch": 4.168845722824057,
+      "grad_norm": 0.8853937983512878,
+      "learning_rate": 0.0002,
+      "loss": 1.4152,
+      "step": 5580
+    },
+    {
+      "epoch": 4.176316772506537,
+      "grad_norm": 0.7070684432983398,
+      "learning_rate": 0.0002,
+      "loss": 1.3633,
+      "step": 5590
+    },
+    {
+      "epoch": 4.183787822189018,
+      "grad_norm": 0.7649410963058472,
+      "learning_rate": 0.0002,
+      "loss": 1.3218,
+      "step": 5600
+    },
+    {
+      "epoch": 4.191258871871498,
+      "grad_norm": 1.2048029899597168,
+      "learning_rate": 0.0002,
+      "loss": 1.3857,
+      "step": 5610
+    },
+    {
+      "epoch": 4.198729921553978,
+      "grad_norm": 0.7986605763435364,
+      "learning_rate": 0.0002,
+      "loss": 1.3629,
+      "step": 5620
+    },
+    {
+      "epoch": 4.206200971236458,
+      "grad_norm": 0.8151885867118835,
+      "learning_rate": 0.0002,
+      "loss": 1.3995,
+      "step": 5630
+    },
+    {
+      "epoch": 4.213672020918939,
+      "grad_norm": 0.7719064354896545,
+      "learning_rate": 0.0002,
+      "loss": 1.3782,
+      "step": 5640
+    },
+    {
+      "epoch": 4.2211430706014195,
+      "grad_norm": 0.8422448039054871,
+      "learning_rate": 0.0002,
+      "loss": 1.3852,
+      "step": 5650
+    },
+    {
+      "epoch": 4.2286141202839,
+      "grad_norm": 0.7017164826393127,
+      "learning_rate": 0.0002,
+      "loss": 1.3321,
+      "step": 5660
+    },
+    {
+      "epoch": 4.236085169966381,
+      "grad_norm": 0.8559677600860596,
+      "learning_rate": 0.0002,
+      "loss": 1.4105,
+      "step": 5670
+    },
+    {
+      "epoch": 4.243556219648861,
+      "grad_norm": 0.8216157555580139,
+      "learning_rate": 0.0002,
+      "loss": 1.3701,
+      "step": 5680
+    },
+    {
+      "epoch": 4.251027269331341,
+      "grad_norm": 0.7681755423545837,
+      "learning_rate": 0.0002,
+      "loss": 1.3565,
+      "step": 5690
+    },
+    {
+      "epoch": 4.258498319013821,
+      "grad_norm": 0.811665952205658,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 5700
+    },
+    {
+      "epoch": 4.265969368696302,
+      "grad_norm": 0.7242204546928406,
+      "learning_rate": 0.0002,
+      "loss": 1.4161,
+      "step": 5710
+    },
+    {
+      "epoch": 4.273440418378782,
+      "grad_norm": 0.7570181488990784,
+      "learning_rate": 0.0002,
+      "loss": 1.2958,
+      "step": 5720
+    },
+    {
+      "epoch": 4.280911468061262,
+      "grad_norm": 0.8951969146728516,
+      "learning_rate": 0.0002,
+      "loss": 1.4265,
+      "step": 5730
+    },
+    {
+      "epoch": 4.288382517743743,
+      "grad_norm": 0.7222902178764343,
+      "learning_rate": 0.0002,
+      "loss": 1.3895,
+      "step": 5740
+    },
+    {
+      "epoch": 4.2958535674262235,
+      "grad_norm": 0.8508469462394714,
+      "learning_rate": 0.0002,
+      "loss": 1.4155,
+      "step": 5750
+    },
+    {
+      "epoch": 4.303324617108704,
+      "grad_norm": 0.7215430736541748,
+      "learning_rate": 0.0002,
+      "loss": 1.365,
+      "step": 5760
+    },
+    {
+      "epoch": 4.310795666791184,
+      "grad_norm": 0.8774884939193726,
+      "learning_rate": 0.0002,
+      "loss": 1.4472,
+      "step": 5770
+    },
+    {
+      "epoch": 4.318266716473665,
+      "grad_norm": 0.8354552984237671,
+      "learning_rate": 0.0002,
+      "loss": 1.427,
+      "step": 5780
+    },
+    {
+      "epoch": 4.325737766156145,
+      "grad_norm": 0.6938814520835876,
+      "learning_rate": 0.0002,
+      "loss": 1.3222,
+      "step": 5790
+    },
+    {
+      "epoch": 4.333208815838625,
+      "grad_norm": 0.78675377368927,
+      "learning_rate": 0.0002,
+      "loss": 1.3589,
+      "step": 5800
+    },
+    {
+      "epoch": 4.340679865521106,
+      "grad_norm": 0.7147697806358337,
+      "learning_rate": 0.0002,
+      "loss": 1.3662,
+      "step": 5810
+    },
+    {
+      "epoch": 4.348150915203586,
+      "grad_norm": 0.7693623304367065,
+      "learning_rate": 0.0002,
+      "loss": 1.3597,
+      "step": 5820
+    },
+    {
+      "epoch": 4.355621964886066,
+      "grad_norm": 0.856517493724823,
+      "learning_rate": 0.0002,
+      "loss": 1.2944,
+      "step": 5830
+    },
+    {
+      "epoch": 4.3630930145685465,
+      "grad_norm": 0.7200973033905029,
+      "learning_rate": 0.0002,
+      "loss": 1.4307,
+      "step": 5840
+    },
+    {
+      "epoch": 4.3705640642510275,
+      "grad_norm": 0.743281364440918,
+      "learning_rate": 0.0002,
+      "loss": 1.442,
+      "step": 5850
+    },
+    {
+      "epoch": 4.378035113933508,
+      "grad_norm": 0.7627727389335632,
+      "learning_rate": 0.0002,
+      "loss": 1.3999,
+      "step": 5860
+    },
+    {
+      "epoch": 4.385506163615988,
+      "grad_norm": 0.7238836884498596,
+      "learning_rate": 0.0002,
+      "loss": 1.4082,
+      "step": 5870
+    },
+    {
+      "epoch": 4.392977213298469,
+      "grad_norm": 0.7253410816192627,
+      "learning_rate": 0.0002,
+      "loss": 1.4292,
+      "step": 5880
+    },
+    {
+      "epoch": 4.400448262980949,
+      "grad_norm": 0.8232238292694092,
+      "learning_rate": 0.0002,
+      "loss": 1.3774,
+      "step": 5890
+    },
+    {
+      "epoch": 4.407919312663429,
+      "grad_norm": 0.8778504729270935,
+      "learning_rate": 0.0002,
+      "loss": 1.3757,
+      "step": 5900
+    },
+    {
+      "epoch": 4.415390362345909,
+      "grad_norm": 0.7639474868774414,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 5910
+    },
+    {
+      "epoch": 4.42286141202839,
+      "grad_norm": 0.7666519284248352,
+      "learning_rate": 0.0002,
+      "loss": 1.3862,
+      "step": 5920
+    },
+    {
+      "epoch": 4.43033246171087,
+      "grad_norm": 0.867132842540741,
+      "learning_rate": 0.0002,
+      "loss": 1.4168,
+      "step": 5930
+    },
+    {
+      "epoch": 4.4378035113933505,
+      "grad_norm": 0.7571166753768921,
+      "learning_rate": 0.0002,
+      "loss": 1.4772,
+      "step": 5940
+    },
+    {
+      "epoch": 4.4452745610758315,
+      "grad_norm": 0.7911370992660522,
+      "learning_rate": 0.0002,
+      "loss": 1.4401,
+      "step": 5950
+    },
+    {
+      "epoch": 4.452745610758312,
+      "grad_norm": 0.8844250440597534,
+      "learning_rate": 0.0002,
+      "loss": 1.4516,
+      "step": 5960
+    },
+    {
+      "epoch": 4.460216660440792,
+      "grad_norm": 0.7336231470108032,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 5970
+    },
+    {
+      "epoch": 4.467687710123272,
+      "grad_norm": 0.8162738084793091,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 5980
+    },
+    {
+      "epoch": 4.475158759805753,
+      "grad_norm": 0.7413017153739929,
+      "learning_rate": 0.0002,
+      "loss": 1.393,
+      "step": 5990
+    },
+    {
+      "epoch": 4.482629809488233,
+      "grad_norm": 0.7215432524681091,
+      "learning_rate": 0.0002,
+      "loss": 1.3712,
+      "step": 6000
+    },
+    {
+      "epoch": 4.490100859170713,
+      "grad_norm": 0.8943389058113098,
+      "learning_rate": 0.0002,
+      "loss": 1.3521,
+      "step": 6010
+    },
+    {
+      "epoch": 4.497571908853194,
+      "grad_norm": 0.7850823998451233,
+      "learning_rate": 0.0002,
+      "loss": 1.4172,
+      "step": 6020
+    },
+    {
+      "epoch": 4.505042958535674,
+      "grad_norm": 0.8117504119873047,
+      "learning_rate": 0.0002,
+      "loss": 1.3582,
+      "step": 6030
+    },
+    {
+      "epoch": 4.5125140082181545,
+      "grad_norm": 0.8381605744361877,
+      "learning_rate": 0.0002,
+      "loss": 1.4272,
+      "step": 6040
+    },
+    {
+      "epoch": 4.519985057900635,
+      "grad_norm": 0.7964059710502625,
+      "learning_rate": 0.0002,
+      "loss": 1.3829,
+      "step": 6050
+    },
+    {
+      "epoch": 4.527456107583116,
+      "grad_norm": 0.7935128211975098,
+      "learning_rate": 0.0002,
+      "loss": 1.3555,
+      "step": 6060
+    },
+    {
+      "epoch": 4.534927157265596,
+      "grad_norm": 0.8725124597549438,
+      "learning_rate": 0.0002,
+      "loss": 1.3994,
+      "step": 6070
+    },
+    {
+      "epoch": 4.542398206948076,
+      "grad_norm": 0.880325198173523,
+      "learning_rate": 0.0002,
+      "loss": 1.3923,
+      "step": 6080
+    },
+    {
+      "epoch": 4.549869256630557,
+      "grad_norm": 0.7220637202262878,
+      "learning_rate": 0.0002,
+      "loss": 1.4459,
+      "step": 6090
+    },
+    {
+      "epoch": 4.557340306313037,
+      "grad_norm": 0.6908547878265381,
+      "learning_rate": 0.0002,
+      "loss": 1.3281,
+      "step": 6100
+    },
+    {
+      "epoch": 4.564811355995517,
+      "grad_norm": 0.797931969165802,
+      "learning_rate": 0.0002,
+      "loss": 1.437,
+      "step": 6110
+    },
+    {
+      "epoch": 4.572282405677997,
+      "grad_norm": 0.7056134343147278,
+      "learning_rate": 0.0002,
+      "loss": 1.4023,
+      "step": 6120
+    },
+    {
+      "epoch": 4.579753455360478,
+      "grad_norm": 0.7850478887557983,
+      "learning_rate": 0.0002,
+      "loss": 1.3814,
+      "step": 6130
+    },
+    {
+      "epoch": 4.5872245050429585,
+      "grad_norm": 0.8112621307373047,
+      "learning_rate": 0.0002,
+      "loss": 1.3579,
+      "step": 6140
+    },
+    {
+      "epoch": 4.594695554725439,
+      "grad_norm": 0.7040849328041077,
+      "learning_rate": 0.0002,
+      "loss": 1.3523,
+      "step": 6150
+    },
+    {
+      "epoch": 4.60216660440792,
+      "grad_norm": 0.7214553952217102,
+      "learning_rate": 0.0002,
+      "loss": 1.3526,
+      "step": 6160
+    },
+    {
+      "epoch": 4.6096376540904,
+      "grad_norm": 0.8616511821746826,
+      "learning_rate": 0.0002,
+      "loss": 1.3932,
+      "step": 6170
+    },
+    {
+      "epoch": 4.61710870377288,
+      "grad_norm": 0.8374658226966858,
+      "learning_rate": 0.0002,
+      "loss": 1.4622,
+      "step": 6180
+    },
+    {
+      "epoch": 4.62457975345536,
+      "grad_norm": 0.6761606931686401,
+      "learning_rate": 0.0002,
+      "loss": 1.3703,
+      "step": 6190
+    },
+    {
+      "epoch": 4.632050803137841,
+      "grad_norm": 0.768028199672699,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 6200
+    },
+    {
+      "epoch": 4.639521852820321,
+      "grad_norm": 0.9372717142105103,
+      "learning_rate": 0.0002,
+      "loss": 1.3772,
+      "step": 6210
+    },
+    {
+      "epoch": 4.646992902502801,
+      "grad_norm": 0.7906546592712402,
+      "learning_rate": 0.0002,
+      "loss": 1.4098,
+      "step": 6220
+    },
+    {
+      "epoch": 4.654463952185282,
+      "grad_norm": 0.7376723289489746,
+      "learning_rate": 0.0002,
+      "loss": 1.3962,
+      "step": 6230
+    },
+    {
+      "epoch": 4.6619350018677626,
+      "grad_norm": 0.8972630500793457,
+      "learning_rate": 0.0002,
+      "loss": 1.4529,
+      "step": 6240
+    },
+    {
+      "epoch": 4.669406051550243,
+      "grad_norm": 0.8261756300926208,
+      "learning_rate": 0.0002,
+      "loss": 1.4668,
+      "step": 6250
+    },
+    {
+      "epoch": 4.676877101232723,
+      "grad_norm": 0.7512393593788147,
+      "learning_rate": 0.0002,
+      "loss": 1.3267,
+      "step": 6260
+    },
+    {
+      "epoch": 4.684348150915204,
+      "grad_norm": 0.7132362127304077,
+      "learning_rate": 0.0002,
+      "loss": 1.4278,
+      "step": 6270
+    },
+    {
+      "epoch": 4.691819200597684,
+      "grad_norm": 0.7690575122833252,
+      "learning_rate": 0.0002,
+      "loss": 1.4299,
+      "step": 6280
+    },
+    {
+      "epoch": 4.699290250280164,
+      "grad_norm": 0.9886258840560913,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 6290
+    },
+    {
+      "epoch": 4.706761299962645,
+      "grad_norm": 0.9502435922622681,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 6300
+    },
+    {
+      "epoch": 4.714232349645125,
+      "grad_norm": 0.702255129814148,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 6310
+    },
+    {
+      "epoch": 4.721703399327605,
+      "grad_norm": 0.7713103890419006,
+      "learning_rate": 0.0002,
+      "loss": 1.4447,
+      "step": 6320
+    },
+    {
+      "epoch": 4.7291744490100855,
+      "grad_norm": 0.7778580784797668,
+      "learning_rate": 0.0002,
+      "loss": 1.4392,
+      "step": 6330
+    },
+    {
+      "epoch": 4.736645498692567,
+      "grad_norm": 0.7275111079216003,
+      "learning_rate": 0.0002,
+      "loss": 1.4169,
+      "step": 6340
+    },
+    {
+      "epoch": 4.744116548375047,
+      "grad_norm": 0.7728744149208069,
+      "learning_rate": 0.0002,
+      "loss": 1.4429,
+      "step": 6350
+    },
+    {
+      "epoch": 4.751587598057527,
+      "grad_norm": 0.9724260568618774,
+      "learning_rate": 0.0002,
+      "loss": 1.3756,
+      "step": 6360
+    },
+    {
+      "epoch": 4.759058647740007,
+      "grad_norm": 0.7505622506141663,
+      "learning_rate": 0.0002,
+      "loss": 1.3358,
+      "step": 6370
+    },
+    {
+      "epoch": 4.766529697422488,
+      "grad_norm": 0.7994682788848877,
+      "learning_rate": 0.0002,
+      "loss": 1.379,
+      "step": 6380
+    },
+    {
+      "epoch": 4.774000747104968,
+      "grad_norm": 0.8432038426399231,
+      "learning_rate": 0.0002,
+      "loss": 1.4275,
+      "step": 6390
+    },
+    {
+      "epoch": 4.781471796787448,
+      "grad_norm": 0.7436022758483887,
+      "learning_rate": 0.0002,
+      "loss": 1.4606,
+      "step": 6400
+    },
+    {
+      "epoch": 4.788942846469929,
+      "grad_norm": 0.7709194421768188,
+      "learning_rate": 0.0002,
+      "loss": 1.3461,
+      "step": 6410
+    },
+    {
+      "epoch": 4.796413896152409,
+      "grad_norm": 0.8798436522483826,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 6420
+    },
+    {
+      "epoch": 4.80388494583489,
+      "grad_norm": 0.790189266204834,
+      "learning_rate": 0.0002,
+      "loss": 1.3761,
+      "step": 6430
+    },
+    {
+      "epoch": 4.811355995517371,
+      "grad_norm": 0.6824303865432739,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 6440
+    },
+    {
+      "epoch": 4.818827045199851,
+      "grad_norm": 0.7501044869422913,
+      "learning_rate": 0.0002,
+      "loss": 1.3877,
+      "step": 6450
+    },
+    {
+      "epoch": 4.826298094882331,
+      "grad_norm": 0.8840398192405701,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 6460
+    },
+    {
+      "epoch": 4.833769144564811,
+      "grad_norm": 0.7812688946723938,
+      "learning_rate": 0.0002,
+      "loss": 1.4412,
+      "step": 6470
+    },
+    {
+      "epoch": 4.841240194247292,
+      "grad_norm": 0.7429926991462708,
+      "learning_rate": 0.0002,
+      "loss": 1.4299,
+      "step": 6480
+    },
+    {
+      "epoch": 4.848711243929772,
+      "grad_norm": 0.7778021693229675,
+      "learning_rate": 0.0002,
+      "loss": 1.5062,
+      "step": 6490
+    },
+    {
+      "epoch": 4.856182293612252,
+      "grad_norm": 0.8270702362060547,
+      "learning_rate": 0.0002,
+      "loss": 1.4589,
+      "step": 6500
+    },
+    {
+      "epoch": 4.863653343294732,
+      "grad_norm": 0.6960513591766357,
+      "learning_rate": 0.0002,
+      "loss": 1.4091,
+      "step": 6510
+    },
+    {
+      "epoch": 4.8711243929772134,
+      "grad_norm": 0.7728942632675171,
+      "learning_rate": 0.0002,
+      "loss": 1.376,
+      "step": 6520
+    },
+    {
+      "epoch": 4.878595442659694,
+      "grad_norm": 0.7377303838729858,
+      "learning_rate": 0.0002,
+      "loss": 1.4852,
+      "step": 6530
+    },
+    {
+      "epoch": 4.886066492342174,
+      "grad_norm": 0.7257253527641296,
+      "learning_rate": 0.0002,
+      "loss": 1.3846,
+      "step": 6540
+    },
+    {
+      "epoch": 4.893537542024655,
+      "grad_norm": 0.7875821590423584,
+      "learning_rate": 0.0002,
+      "loss": 1.4166,
+      "step": 6550
+    },
+    {
+      "epoch": 4.901008591707135,
+      "grad_norm": 0.8346304297447205,
+      "learning_rate": 0.0002,
+      "loss": 1.357,
+      "step": 6560
+    },
+    {
+      "epoch": 4.908479641389615,
+      "grad_norm": 0.7710739374160767,
+      "learning_rate": 0.0002,
+      "loss": 1.4522,
+      "step": 6570
+    },
+    {
+      "epoch": 4.915950691072096,
+      "grad_norm": 0.7015138268470764,
+      "learning_rate": 0.0002,
+      "loss": 1.4465,
+      "step": 6580
+    },
+    {
+      "epoch": 4.923421740754576,
+      "grad_norm": 0.8707432150840759,
+      "learning_rate": 0.0002,
+      "loss": 1.435,
+      "step": 6590
+    },
+    {
+      "epoch": 4.930892790437056,
+      "grad_norm": 0.786601185798645,
+      "learning_rate": 0.0002,
+      "loss": 1.2968,
+      "step": 6600
+    },
+    {
+      "epoch": 4.938363840119536,
+      "grad_norm": 0.978519082069397,
+      "learning_rate": 0.0002,
+      "loss": 1.4385,
+      "step": 6610
+    },
+    {
+      "epoch": 4.9458348898020175,
+      "grad_norm": 0.8102927207946777,
+      "learning_rate": 0.0002,
+      "loss": 1.3997,
+      "step": 6620
+    },
+    {
+      "epoch": 4.953305939484498,
+      "grad_norm": 0.7628704309463501,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 6630
+    },
+    {
+      "epoch": 4.960776989166978,
+      "grad_norm": 0.8053455352783203,
+      "learning_rate": 0.0002,
+      "loss": 1.3774,
+      "step": 6640
+    },
+    {
+      "epoch": 4.968248038849458,
+      "grad_norm": 0.8680412173271179,
+      "learning_rate": 0.0002,
+      "loss": 1.5092,
+      "step": 6650
+    },
+    {
+      "epoch": 4.975719088531939,
+      "grad_norm": 0.7415758371353149,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 6660
+    },
+    {
+      "epoch": 4.983190138214419,
+      "grad_norm": 0.7730312347412109,
+      "learning_rate": 0.0002,
+      "loss": 1.3793,
+      "step": 6670
+    },
+    {
+      "epoch": 4.990661187896899,
+      "grad_norm": 0.7924041152000427,
+      "learning_rate": 0.0002,
+      "loss": 1.4863,
+      "step": 6680
+    },
+    {
+      "epoch": 4.99813223757938,
+      "grad_norm": 0.8677893877029419,
+      "learning_rate": 0.0002,
+      "loss": 1.4137,
+      "step": 6690
+    },
+    {
+      "epoch": 4.999626447515876,
+      "eval_loss": 1.9444633722305298,
+      "eval_runtime": 39.3488,
+      "eval_samples_per_second": 13.088,
+      "eval_steps_per_second": 1.652,
+      "step": 6692
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10704,
+  "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": 3.0971391488557056e+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-0.9-num-6326-sd-42/checkpoint-6692/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-0.9-num-6326-sd-42/checkpoint-6692/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..67c7b4ca126d7b712ad1985ef15ffb29ebe76633
--- /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-0.9-num-6326-sd-42/checkpoint-6692/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fc87da605e94ff0ecd8f5b371302a2d5f8727b77984707a2185ddb447fc3796
+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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6f7272d48984d2d8648f89d62a64d97cf42adcc4
--- /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-0.9-num-6326-sd-42/checkpoint-8031/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46904eb2dc787478b572840fe84051b93ba02c40147865a235946aefa6c1711a
+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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2febd43ac8c979c08c9655a9f74869e9a06d6e6a
--- /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-0.9-num-6326-sd-42/checkpoint-8031/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c690d8d86a430610e78803a274c3ac920a0a69e8dfb445585a26730a9bd6a18a
+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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..4c9f197d87560ca90f917f637cb1077d6b4503e5
--- /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-0.9-num-6326-sd-42/checkpoint-8031/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:042d0f7c0322075840dbdbfc8564dfcb7d871dae8201d1a5e5ea29425c005636
+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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..9b8422814b4a9fcc34918e1564540e393428e2a0
--- /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-0.9-num-6326-sd-42/checkpoint-8031/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc0b46f21e02353093efbfa2699d5185e23bdb76aaa14d3a0676bc7227f85624
+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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..146718c3cbc7117cf6f0e7a0ba86f7ce72aa9dc6
--- /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-0.9-num-6326-sd-42/checkpoint-8031/trainer_state.json
@@ -0,0 +1,5702 @@
+{
+  "best_metric": 1.8046749830245972,
+  "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-0.9-num-6326-sd-42/checkpoint-2677",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 8031,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007471049682480389,
+      "grad_norm": 0.4912872612476349,
+      "learning_rate": 0.0002,
+      "loss": 2.6181,
+      "step": 10
+    },
+    {
+      "epoch": 0.014942099364960777,
+      "grad_norm": 0.4856316149234772,
+      "learning_rate": 0.0002,
+      "loss": 2.2606,
+      "step": 20
+    },
+    {
+      "epoch": 0.022413149047441166,
+      "grad_norm": 0.47683125734329224,
+      "learning_rate": 0.0002,
+      "loss": 2.0957,
+      "step": 30
+    },
+    {
+      "epoch": 0.029884198729921554,
+      "grad_norm": 0.515082597732544,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 40
+    },
+    {
+      "epoch": 0.03735524841240194,
+      "grad_norm": 0.5299215316772461,
+      "learning_rate": 0.0002,
+      "loss": 1.9704,
+      "step": 50
+    },
+    {
+      "epoch": 0.04482629809488233,
+      "grad_norm": 0.4951399862766266,
+      "learning_rate": 0.0002,
+      "loss": 1.9225,
+      "step": 60
+    },
+    {
+      "epoch": 0.05229734777736272,
+      "grad_norm": 0.48079821467399597,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05976839745984311,
+      "grad_norm": 0.49402132630348206,
+      "learning_rate": 0.0002,
+      "loss": 1.9466,
+      "step": 80
+    },
+    {
+      "epoch": 0.0672394471423235,
+      "grad_norm": 0.4778193235397339,
+      "learning_rate": 0.0002,
+      "loss": 1.8691,
+      "step": 90
+    },
+    {
+      "epoch": 0.07471049682480388,
+      "grad_norm": 0.42472657561302185,
+      "learning_rate": 0.0002,
+      "loss": 1.8455,
+      "step": 100
+    },
+    {
+      "epoch": 0.08218154650728428,
+      "grad_norm": 0.4433092474937439,
+      "learning_rate": 0.0002,
+      "loss": 1.8744,
+      "step": 110
+    },
+    {
+      "epoch": 0.08965259618976466,
+      "grad_norm": 0.4472862780094147,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 120
+    },
+    {
+      "epoch": 0.09712364587224505,
+      "grad_norm": 0.42596298456192017,
+      "learning_rate": 0.0002,
+      "loss": 1.9256,
+      "step": 130
+    },
+    {
+      "epoch": 0.10459469555472543,
+      "grad_norm": 0.46645811200141907,
+      "learning_rate": 0.0002,
+      "loss": 1.8015,
+      "step": 140
+    },
+    {
+      "epoch": 0.11206574523720583,
+      "grad_norm": 0.41041234135627747,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 150
+    },
+    {
+      "epoch": 0.11953679491968622,
+      "grad_norm": 0.5329819917678833,
+      "learning_rate": 0.0002,
+      "loss": 1.8276,
+      "step": 160
+    },
+    {
+      "epoch": 0.1270078446021666,
+      "grad_norm": 0.4065922200679779,
+      "learning_rate": 0.0002,
+      "loss": 1.8118,
+      "step": 170
+    },
+    {
+      "epoch": 0.134478894284647,
+      "grad_norm": 0.38406994938850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8559,
+      "step": 180
+    },
+    {
+      "epoch": 0.14194994396712737,
+      "grad_norm": 0.4246881306171417,
+      "learning_rate": 0.0002,
+      "loss": 1.8647,
+      "step": 190
+    },
+    {
+      "epoch": 0.14942099364960776,
+      "grad_norm": 0.35136649012565613,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 200
+    },
+    {
+      "epoch": 0.15689204333208817,
+      "grad_norm": 0.43252742290496826,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.16436309301456856,
+      "grad_norm": 0.39236941933631897,
+      "learning_rate": 0.0002,
+      "loss": 1.7823,
+      "step": 220
+    },
+    {
+      "epoch": 0.17183414269704894,
+      "grad_norm": 0.3748249113559723,
+      "learning_rate": 0.0002,
+      "loss": 1.818,
+      "step": 230
+    },
+    {
+      "epoch": 0.17930519237952933,
+      "grad_norm": 0.6432855725288391,
+      "learning_rate": 0.0002,
+      "loss": 1.866,
+      "step": 240
+    },
+    {
+      "epoch": 0.1867762420620097,
+      "grad_norm": 0.34874802827835083,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 250
+    },
+    {
+      "epoch": 0.1942472917444901,
+      "grad_norm": 0.3721984326839447,
+      "learning_rate": 0.0002,
+      "loss": 1.79,
+      "step": 260
+    },
+    {
+      "epoch": 0.20171834142697048,
+      "grad_norm": 0.4339311420917511,
+      "learning_rate": 0.0002,
+      "loss": 1.8464,
+      "step": 270
+    },
+    {
+      "epoch": 0.20918939110945087,
+      "grad_norm": 0.4018215537071228,
+      "learning_rate": 0.0002,
+      "loss": 1.8665,
+      "step": 280
+    },
+    {
+      "epoch": 0.21666044079193125,
+      "grad_norm": 0.3278839886188507,
+      "learning_rate": 0.0002,
+      "loss": 1.8048,
+      "step": 290
+    },
+    {
+      "epoch": 0.22413149047441167,
+      "grad_norm": 0.36146077513694763,
+      "learning_rate": 0.0002,
+      "loss": 1.7395,
+      "step": 300
+    },
+    {
+      "epoch": 0.23160254015689205,
+      "grad_norm": 0.38175010681152344,
+      "learning_rate": 0.0002,
+      "loss": 1.7916,
+      "step": 310
+    },
+    {
+      "epoch": 0.23907358983937244,
+      "grad_norm": 0.44776618480682373,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 320
+    },
+    {
+      "epoch": 0.24654463952185282,
+      "grad_norm": 0.3933652937412262,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 330
+    },
+    {
+      "epoch": 0.2540156892043332,
+      "grad_norm": 0.3515005111694336,
+      "learning_rate": 0.0002,
+      "loss": 1.8393,
+      "step": 340
+    },
+    {
+      "epoch": 0.2614867388868136,
+      "grad_norm": 0.6683304309844971,
+      "learning_rate": 0.0002,
+      "loss": 1.8653,
+      "step": 350
+    },
+    {
+      "epoch": 0.268957788569294,
+      "grad_norm": 0.37093454599380493,
+      "learning_rate": 0.0002,
+      "loss": 1.8797,
+      "step": 360
+    },
+    {
+      "epoch": 0.2764288382517744,
+      "grad_norm": 0.3450651168823242,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 370
+    },
+    {
+      "epoch": 0.28389988793425475,
+      "grad_norm": 0.5140917301177979,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 380
+    },
+    {
+      "epoch": 0.29137093761673516,
+      "grad_norm": 0.32885563373565674,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 390
+    },
+    {
+      "epoch": 0.2988419872992155,
+      "grad_norm": 0.33962297439575195,
+      "learning_rate": 0.0002,
+      "loss": 1.8174,
+      "step": 400
+    },
+    {
+      "epoch": 0.30631303698169593,
+      "grad_norm": 0.3723141849040985,
+      "learning_rate": 0.0002,
+      "loss": 1.7467,
+      "step": 410
+    },
+    {
+      "epoch": 0.31378408666417634,
+      "grad_norm": 0.37173134088516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8459,
+      "step": 420
+    },
+    {
+      "epoch": 0.3212551363466567,
+      "grad_norm": 0.33736956119537354,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 430
+    },
+    {
+      "epoch": 0.3287261860291371,
+      "grad_norm": 0.3602448105812073,
+      "learning_rate": 0.0002,
+      "loss": 1.8367,
+      "step": 440
+    },
+    {
+      "epoch": 0.33619723571161747,
+      "grad_norm": 0.3569699227809906,
+      "learning_rate": 0.0002,
+      "loss": 1.8058,
+      "step": 450
+    },
+    {
+      "epoch": 0.3436682853940979,
+      "grad_norm": 0.31009167432785034,
+      "learning_rate": 0.0002,
+      "loss": 1.8086,
+      "step": 460
+    },
+    {
+      "epoch": 0.35113933507657824,
+      "grad_norm": 0.5278693437576294,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 470
+    },
+    {
+      "epoch": 0.35861038475905865,
+      "grad_norm": 0.3587537109851837,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 480
+    },
+    {
+      "epoch": 0.366081434441539,
+      "grad_norm": 0.3859670162200928,
+      "learning_rate": 0.0002,
+      "loss": 1.8046,
+      "step": 490
+    },
+    {
+      "epoch": 0.3735524841240194,
+      "grad_norm": 0.395913690328598,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 500
+    },
+    {
+      "epoch": 0.38102353380649984,
+      "grad_norm": 0.35052940249443054,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 510
+    },
+    {
+      "epoch": 0.3884945834889802,
+      "grad_norm": 0.2979494333267212,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 520
+    },
+    {
+      "epoch": 0.3959656331714606,
+      "grad_norm": 0.3062683343887329,
+      "learning_rate": 0.0002,
+      "loss": 1.8641,
+      "step": 530
+    },
+    {
+      "epoch": 0.40343668285394096,
+      "grad_norm": 0.3172847330570221,
+      "learning_rate": 0.0002,
+      "loss": 1.7651,
+      "step": 540
+    },
+    {
+      "epoch": 0.4109077325364214,
+      "grad_norm": 0.360435426235199,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 550
+    },
+    {
+      "epoch": 0.41837878221890173,
+      "grad_norm": 0.3427872359752655,
+      "learning_rate": 0.0002,
+      "loss": 1.9054,
+      "step": 560
+    },
+    {
+      "epoch": 0.42584983190138215,
+      "grad_norm": 0.34036558866500854,
+      "learning_rate": 0.0002,
+      "loss": 1.7562,
+      "step": 570
+    },
+    {
+      "epoch": 0.4333208815838625,
+      "grad_norm": 0.3365345299243927,
+      "learning_rate": 0.0002,
+      "loss": 1.7254,
+      "step": 580
+    },
+    {
+      "epoch": 0.4407919312663429,
+      "grad_norm": 0.35619041323661804,
+      "learning_rate": 0.0002,
+      "loss": 1.8328,
+      "step": 590
+    },
+    {
+      "epoch": 0.44826298094882333,
+      "grad_norm": 0.3569088280200958,
+      "learning_rate": 0.0002,
+      "loss": 1.8114,
+      "step": 600
+    },
+    {
+      "epoch": 0.4557340306313037,
+      "grad_norm": 0.3581278622150421,
+      "learning_rate": 0.0002,
+      "loss": 1.8599,
+      "step": 610
+    },
+    {
+      "epoch": 0.4632050803137841,
+      "grad_norm": 0.43197110295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.7078,
+      "step": 620
+    },
+    {
+      "epoch": 0.47067612999626446,
+      "grad_norm": 0.33966198563575745,
+      "learning_rate": 0.0002,
+      "loss": 1.8257,
+      "step": 630
+    },
+    {
+      "epoch": 0.47814717967874487,
+      "grad_norm": 0.3343866467475891,
+      "learning_rate": 0.0002,
+      "loss": 1.7528,
+      "step": 640
+    },
+    {
+      "epoch": 0.48561822936122523,
+      "grad_norm": 0.33878564834594727,
+      "learning_rate": 0.0002,
+      "loss": 1.8191,
+      "step": 650
+    },
+    {
+      "epoch": 0.49308927904370564,
+      "grad_norm": 0.387195885181427,
+      "learning_rate": 0.0002,
+      "loss": 1.8801,
+      "step": 660
+    },
+    {
+      "epoch": 0.500560328726186,
+      "grad_norm": 0.3755440413951874,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 670
+    },
+    {
+      "epoch": 0.5080313784086664,
+      "grad_norm": 0.3272816836833954,
+      "learning_rate": 0.0002,
+      "loss": 1.8057,
+      "step": 680
+    },
+    {
+      "epoch": 0.5155024280911468,
+      "grad_norm": 0.36063864827156067,
+      "learning_rate": 0.0002,
+      "loss": 1.8156,
+      "step": 690
+    },
+    {
+      "epoch": 0.5229734777736272,
+      "grad_norm": 0.35317373275756836,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 700
+    },
+    {
+      "epoch": 0.5304445274561076,
+      "grad_norm": 0.3561195433139801,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 710
+    },
+    {
+      "epoch": 0.537915577138588,
+      "grad_norm": 0.31124624609947205,
+      "learning_rate": 0.0002,
+      "loss": 1.8149,
+      "step": 720
+    },
+    {
+      "epoch": 0.5453866268210683,
+      "grad_norm": 0.3294544517993927,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 730
+    },
+    {
+      "epoch": 0.5528576765035488,
+      "grad_norm": 0.31933900713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.8027,
+      "step": 740
+    },
+    {
+      "epoch": 0.5603287261860291,
+      "grad_norm": 0.3226020634174347,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 750
+    },
+    {
+      "epoch": 0.5677997758685095,
+      "grad_norm": 0.3147525489330292,
+      "learning_rate": 0.0002,
+      "loss": 1.7862,
+      "step": 760
+    },
+    {
+      "epoch": 0.57527082555099,
+      "grad_norm": 0.32234328985214233,
+      "learning_rate": 0.0002,
+      "loss": 1.9028,
+      "step": 770
+    },
+    {
+      "epoch": 0.5827418752334703,
+      "grad_norm": 0.3258664309978485,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 780
+    },
+    {
+      "epoch": 0.5902129249159507,
+      "grad_norm": 0.3166961967945099,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 790
+    },
+    {
+      "epoch": 0.597683974598431,
+      "grad_norm": 0.35621458292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 800
+    },
+    {
+      "epoch": 0.6051550242809115,
+      "grad_norm": 0.3236999213695526,
+      "learning_rate": 0.0002,
+      "loss": 1.8313,
+      "step": 810
+    },
+    {
+      "epoch": 0.6126260739633919,
+      "grad_norm": 0.2892923653125763,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 820
+    },
+    {
+      "epoch": 0.6200971236458722,
+      "grad_norm": 0.4098321497440338,
+      "learning_rate": 0.0002,
+      "loss": 1.8709,
+      "step": 830
+    },
+    {
+      "epoch": 0.6275681733283527,
+      "grad_norm": 0.3337118923664093,
+      "learning_rate": 0.0002,
+      "loss": 1.7637,
+      "step": 840
+    },
+    {
+      "epoch": 0.635039223010833,
+      "grad_norm": 0.30416029691696167,
+      "learning_rate": 0.0002,
+      "loss": 1.7375,
+      "step": 850
+    },
+    {
+      "epoch": 0.6425102726933134,
+      "grad_norm": 0.3361026346683502,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 860
+    },
+    {
+      "epoch": 0.6499813223757938,
+      "grad_norm": 0.3537365198135376,
+      "learning_rate": 0.0002,
+      "loss": 1.732,
+      "step": 870
+    },
+    {
+      "epoch": 0.6574523720582742,
+      "grad_norm": 0.33854469656944275,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 880
+    },
+    {
+      "epoch": 0.6649234217407546,
+      "grad_norm": 0.3332272469997406,
+      "learning_rate": 0.0002,
+      "loss": 1.7561,
+      "step": 890
+    },
+    {
+      "epoch": 0.6723944714232349,
+      "grad_norm": 0.34954726696014404,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 900
+    },
+    {
+      "epoch": 0.6798655211057153,
+      "grad_norm": 0.2921750247478485,
+      "learning_rate": 0.0002,
+      "loss": 1.7917,
+      "step": 910
+    },
+    {
+      "epoch": 0.6873365707881958,
+      "grad_norm": 0.30508682131767273,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 920
+    },
+    {
+      "epoch": 0.6948076204706761,
+      "grad_norm": 0.32268425822257996,
+      "learning_rate": 0.0002,
+      "loss": 1.8082,
+      "step": 930
+    },
+    {
+      "epoch": 0.7022786701531565,
+      "grad_norm": 0.2844390869140625,
+      "learning_rate": 0.0002,
+      "loss": 1.8283,
+      "step": 940
+    },
+    {
+      "epoch": 0.709749719835637,
+      "grad_norm": 0.31263890862464905,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 950
+    },
+    {
+      "epoch": 0.7172207695181173,
+      "grad_norm": 0.3626808822154999,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 960
+    },
+    {
+      "epoch": 0.7246918192005977,
+      "grad_norm": 0.3322749733924866,
+      "learning_rate": 0.0002,
+      "loss": 1.853,
+      "step": 970
+    },
+    {
+      "epoch": 0.732162868883078,
+      "grad_norm": 0.29177871346473694,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 980
+    },
+    {
+      "epoch": 0.7396339185655585,
+      "grad_norm": 0.35405513644218445,
+      "learning_rate": 0.0002,
+      "loss": 1.8447,
+      "step": 990
+    },
+    {
+      "epoch": 0.7471049682480388,
+      "grad_norm": 0.39318400621414185,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7545760179305192,
+      "grad_norm": 0.29401418566703796,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7620470676129997,
+      "grad_norm": 0.3271748721599579,
+      "learning_rate": 0.0002,
+      "loss": 1.7649,
+      "step": 1020
+    },
+    {
+      "epoch": 0.76951811729548,
+      "grad_norm": 0.30883970856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7769891669779604,
+      "grad_norm": 0.3411838412284851,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7844602166604407,
+      "grad_norm": 0.30608129501342773,
+      "learning_rate": 0.0002,
+      "loss": 1.829,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7919312663429212,
+      "grad_norm": 0.30899080634117126,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7994023160254016,
+      "grad_norm": 0.3160453140735626,
+      "learning_rate": 0.0002,
+      "loss": 1.7625,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8068733657078819,
+      "grad_norm": 0.30947187542915344,
+      "learning_rate": 0.0002,
+      "loss": 1.8452,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8143444153903624,
+      "grad_norm": 0.3103134036064148,
+      "learning_rate": 0.0002,
+      "loss": 1.7418,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8218154650728428,
+      "grad_norm": 0.31771138310432434,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8292865147553231,
+      "grad_norm": 0.5860997438430786,
+      "learning_rate": 0.0002,
+      "loss": 1.7918,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8367575644378035,
+      "grad_norm": 0.3230148255825043,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8442286141202839,
+      "grad_norm": 0.29611510038375854,
+      "learning_rate": 0.0002,
+      "loss": 1.8478,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8516996638027643,
+      "grad_norm": 0.3373654782772064,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8591707134852447,
+      "grad_norm": 0.3474279046058655,
+      "learning_rate": 0.0002,
+      "loss": 1.7997,
+      "step": 1150
+    },
+    {
+      "epoch": 0.866641763167725,
+      "grad_norm": 0.35057875514030457,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8741128128502055,
+      "grad_norm": 0.39537495374679565,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8815838625326858,
+      "grad_norm": 0.3714233636856079,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8890549122151662,
+      "grad_norm": 0.2950296998023987,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8965259618976467,
+      "grad_norm": 0.38182979822158813,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 1200
+    },
+    {
+      "epoch": 0.903997011580127,
+      "grad_norm": 0.27883678674697876,
+      "learning_rate": 0.0002,
+      "loss": 1.827,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9114680612626074,
+      "grad_norm": 0.33874374628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9189391109450877,
+      "grad_norm": 0.3014272153377533,
+      "learning_rate": 0.0002,
+      "loss": 1.7334,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9264101606275682,
+      "grad_norm": 0.3194271922111511,
+      "learning_rate": 0.0002,
+      "loss": 1.8235,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9338812103100486,
+      "grad_norm": 0.3049403429031372,
+      "learning_rate": 0.0002,
+      "loss": 1.7924,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9413522599925289,
+      "grad_norm": 0.30621254444122314,
+      "learning_rate": 0.0002,
+      "loss": 1.7535,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9488233096750094,
+      "grad_norm": 0.28675132989883423,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9562943593574897,
+      "grad_norm": 0.3322032690048218,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9637654090399701,
+      "grad_norm": 0.35408294200897217,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9712364587224505,
+      "grad_norm": 0.36386919021606445,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9787075084049309,
+      "grad_norm": 0.32338324189186096,
+      "learning_rate": 0.0002,
+      "loss": 1.8633,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9861785580874113,
+      "grad_norm": 0.3714013993740082,
+      "learning_rate": 0.0002,
+      "loss": 1.7724,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9936496077698916,
+      "grad_norm": 0.3133082389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.7766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9996264475158759,
+      "eval_loss": 1.8051470518112183,
+      "eval_runtime": 38.6332,
+      "eval_samples_per_second": 13.331,
+      "eval_steps_per_second": 1.682,
+      "step": 1338
+    },
+    {
+      "epoch": 1.001120657452372,
+      "grad_norm": 0.31595754623413086,
+      "learning_rate": 0.0002,
+      "loss": 1.8035,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0085917071348525,
+      "grad_norm": 0.3095700144767761,
+      "learning_rate": 0.0002,
+      "loss": 1.7486,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0160627568173328,
+      "grad_norm": 0.34677496552467346,
+      "learning_rate": 0.0002,
+      "loss": 1.6981,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0235338064998132,
+      "grad_norm": 0.29108840227127075,
+      "learning_rate": 0.0002,
+      "loss": 1.7377,
+      "step": 1370
+    },
+    {
+      "epoch": 1.0310048561822935,
+      "grad_norm": 0.32356950640678406,
+      "learning_rate": 0.0002,
+      "loss": 1.7194,
+      "step": 1380
+    },
+    {
+      "epoch": 1.038475905864774,
+      "grad_norm": 0.4200669229030609,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0459469555472545,
+      "grad_norm": 0.3283711373806,
+      "learning_rate": 0.0002,
+      "loss": 1.797,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0534180052297348,
+      "grad_norm": 0.32898256182670593,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0608890549122152,
+      "grad_norm": 0.38790300488471985,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0683601045946955,
+      "grad_norm": 0.339800089597702,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1430
+    },
+    {
+      "epoch": 1.075831154277176,
+      "grad_norm": 0.3548751175403595,
+      "learning_rate": 0.0002,
+      "loss": 1.7076,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0833022039596563,
+      "grad_norm": 0.35114359855651855,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0907732536421366,
+      "grad_norm": 0.35226720571517944,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0982443033246172,
+      "grad_norm": 0.33665576577186584,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1057153530070976,
+      "grad_norm": 0.363889217376709,
+      "learning_rate": 0.0002,
+      "loss": 1.6699,
+      "step": 1480
+    },
+    {
+      "epoch": 1.113186402689578,
+      "grad_norm": 0.3826201856136322,
+      "learning_rate": 0.0002,
+      "loss": 1.7933,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1206574523720583,
+      "grad_norm": 0.34058740735054016,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1281285020545386,
+      "grad_norm": 0.3462134301662445,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 1510
+    },
+    {
+      "epoch": 1.135599551737019,
+      "grad_norm": 0.3396756052970886,
+      "learning_rate": 0.0002,
+      "loss": 1.7147,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1430706014194993,
+      "grad_norm": 0.32004743814468384,
+      "learning_rate": 0.0002,
+      "loss": 1.7219,
+      "step": 1530
+    },
+    {
+      "epoch": 1.15054165110198,
+      "grad_norm": 0.3397733271121979,
+      "learning_rate": 0.0002,
+      "loss": 1.743,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1580127007844603,
+      "grad_norm": 0.3783262073993683,
+      "learning_rate": 0.0002,
+      "loss": 1.7333,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1654837504669406,
+      "grad_norm": 0.35121291875839233,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 1560
+    },
+    {
+      "epoch": 1.172954800149421,
+      "grad_norm": 0.35816895961761475,
+      "learning_rate": 0.0002,
+      "loss": 1.678,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1804258498319014,
+      "grad_norm": 0.33843839168548584,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1878968995143817,
+      "grad_norm": 0.3371972143650055,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 1590
+    },
+    {
+      "epoch": 1.195367949196862,
+      "grad_norm": 0.36016878485679626,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2028389988793426,
+      "grad_norm": 0.40879473090171814,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 1610
+    },
+    {
+      "epoch": 1.210310048561823,
+      "grad_norm": 0.3216715455055237,
+      "learning_rate": 0.0002,
+      "loss": 1.6955,
+      "step": 1620
+    },
+    {
+      "epoch": 1.2177810982443034,
+      "grad_norm": 0.4482610821723938,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2252521479267837,
+      "grad_norm": 0.3257700502872467,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1640
+    },
+    {
+      "epoch": 1.232723197609264,
+      "grad_norm": 0.38646459579467773,
+      "learning_rate": 0.0002,
+      "loss": 1.7177,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2401942472917444,
+      "grad_norm": 0.4081360697746277,
+      "learning_rate": 0.0002,
+      "loss": 1.7081,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2476652969742248,
+      "grad_norm": 0.4326848089694977,
+      "learning_rate": 0.0002,
+      "loss": 1.7519,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2551363466567054,
+      "grad_norm": 0.346401572227478,
+      "learning_rate": 0.0002,
+      "loss": 1.6752,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2626073963391857,
+      "grad_norm": 0.34536251425743103,
+      "learning_rate": 0.0002,
+      "loss": 1.7425,
+      "step": 1690
+    },
+    {
+      "epoch": 1.270078446021666,
+      "grad_norm": 0.41359591484069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7061,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2775494957041464,
+      "grad_norm": 0.3530874252319336,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2850205453866268,
+      "grad_norm": 0.3702719211578369,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2924915950691072,
+      "grad_norm": 0.3703329563140869,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2999626447515875,
+      "grad_norm": 0.37919729948043823,
+      "learning_rate": 0.0002,
+      "loss": 1.7221,
+      "step": 1740
+    },
+    {
+      "epoch": 1.307433694434068,
+      "grad_norm": 0.32526856660842896,
+      "learning_rate": 0.0002,
+      "loss": 1.7859,
+      "step": 1750
+    },
+    {
+      "epoch": 1.3149047441165485,
+      "grad_norm": 0.36752620339393616,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3223757937990288,
+      "grad_norm": 0.3398192524909973,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3298468434815092,
+      "grad_norm": 0.37435585260391235,
+      "learning_rate": 0.0002,
+      "loss": 1.7492,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3373178931639895,
+      "grad_norm": 0.35793280601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3447889428464699,
+      "grad_norm": 0.35481882095336914,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3522599925289502,
+      "grad_norm": 0.3786393105983734,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3597310422114308,
+      "grad_norm": 0.33245593309402466,
+      "learning_rate": 0.0002,
+      "loss": 1.7169,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3672020918939112,
+      "grad_norm": 0.35388344526290894,
+      "learning_rate": 0.0002,
+      "loss": 1.7577,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3746731415763915,
+      "grad_norm": 0.3695325553417206,
+      "learning_rate": 0.0002,
+      "loss": 1.6968,
+      "step": 1840
+    },
+    {
+      "epoch": 1.382144191258872,
+      "grad_norm": 0.3683604598045349,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3896152409413522,
+      "grad_norm": 0.3753012418746948,
+      "learning_rate": 0.0002,
+      "loss": 1.7878,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3970862906238326,
+      "grad_norm": 0.3331069350242615,
+      "learning_rate": 0.0002,
+      "loss": 1.6969,
+      "step": 1870
+    },
+    {
+      "epoch": 1.404557340306313,
+      "grad_norm": 0.3877500295639038,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4120283899887935,
+      "grad_norm": 0.33525151014328003,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4194994396712737,
+      "grad_norm": 0.3697299659252167,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4269704893537543,
+      "grad_norm": 0.4029286205768585,
+      "learning_rate": 0.0002,
+      "loss": 1.6956,
+      "step": 1910
+    },
+    {
+      "epoch": 1.4344415390362346,
+      "grad_norm": 0.3596203029155731,
+      "learning_rate": 0.0002,
+      "loss": 1.6897,
+      "step": 1920
+    },
+    {
+      "epoch": 1.441912588718715,
+      "grad_norm": 0.450783908367157,
+      "learning_rate": 0.0002,
+      "loss": 1.7139,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4493836384011953,
+      "grad_norm": 0.3651481866836548,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4568546880836757,
+      "grad_norm": 0.3608424663543701,
+      "learning_rate": 0.0002,
+      "loss": 1.6637,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4643257377661563,
+      "grad_norm": 0.39684420824050903,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4717967874486364,
+      "grad_norm": 0.34618663787841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7514,
+      "step": 1970
+    },
+    {
+      "epoch": 1.479267837131117,
+      "grad_norm": 0.4150386452674866,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4867388868135973,
+      "grad_norm": 0.35500776767730713,
+      "learning_rate": 0.0002,
+      "loss": 1.7021,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4942099364960777,
+      "grad_norm": 0.344144344329834,
+      "learning_rate": 0.0002,
+      "loss": 1.7322,
+      "step": 2000
+    },
+    {
+      "epoch": 1.501680986178558,
+      "grad_norm": 0.3340149223804474,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5091520358610384,
+      "grad_norm": 0.37685006856918335,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 2020
+    },
+    {
+      "epoch": 1.516623085543519,
+      "grad_norm": 0.3699876368045807,
+      "learning_rate": 0.0002,
+      "loss": 1.8299,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5240941352259991,
+      "grad_norm": 0.3370307385921478,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5315651849084797,
+      "grad_norm": 0.37780630588531494,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 2050
+    },
+    {
+      "epoch": 1.53903623459096,
+      "grad_norm": 0.370259165763855,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5465072842734404,
+      "grad_norm": 0.3440011441707611,
+      "learning_rate": 0.0002,
+      "loss": 1.7398,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5539783339559208,
+      "grad_norm": 0.40382063388824463,
+      "learning_rate": 0.0002,
+      "loss": 1.7105,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5614493836384011,
+      "grad_norm": 0.38002029061317444,
+      "learning_rate": 0.0002,
+      "loss": 1.7071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5689204333208817,
+      "grad_norm": 0.3658451437950134,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5763914830033618,
+      "grad_norm": 0.354842871427536,
+      "learning_rate": 0.0002,
+      "loss": 1.7598,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5838625326858424,
+      "grad_norm": 0.34735530614852905,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5913335823683228,
+      "grad_norm": 0.377581924200058,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5988046320508031,
+      "grad_norm": 0.41254034638404846,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6062756817332835,
+      "grad_norm": 0.3630715310573578,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6137467314157639,
+      "grad_norm": 0.36980143189430237,
+      "learning_rate": 0.0002,
+      "loss": 1.7531,
+      "step": 2160
+    },
+    {
+      "epoch": 1.6212177810982444,
+      "grad_norm": 0.3634769320487976,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6286888307807246,
+      "grad_norm": 0.3794139623641968,
+      "learning_rate": 0.0002,
+      "loss": 1.6367,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6361598804632052,
+      "grad_norm": 0.359742134809494,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6436309301456855,
+      "grad_norm": 0.3770543932914734,
+      "learning_rate": 0.0002,
+      "loss": 1.7027,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6511019798281659,
+      "grad_norm": 0.3797036409378052,
+      "learning_rate": 0.0002,
+      "loss": 1.784,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6585730295106462,
+      "grad_norm": 0.35622093081474304,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6660440791931266,
+      "grad_norm": 0.34552520513534546,
+      "learning_rate": 0.0002,
+      "loss": 1.6615,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6735151288756072,
+      "grad_norm": 0.379926860332489,
+      "learning_rate": 0.0002,
+      "loss": 1.7522,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6809861785580873,
+      "grad_norm": 0.37083810567855835,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6884572282405679,
+      "grad_norm": 0.42746543884277344,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6959282779230482,
+      "grad_norm": 0.3372884690761566,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7033993276055286,
+      "grad_norm": 0.35220256447792053,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2280
+    },
+    {
+      "epoch": 1.710870377288009,
+      "grad_norm": 0.3659130930900574,
+      "learning_rate": 0.0002,
+      "loss": 1.7154,
+      "step": 2290
+    },
+    {
+      "epoch": 1.7183414269704893,
+      "grad_norm": 0.37629297375679016,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7258124766529699,
+      "grad_norm": 0.36312398314476013,
+      "learning_rate": 0.0002,
+      "loss": 1.7212,
+      "step": 2310
+    },
+    {
+      "epoch": 1.73328352633545,
+      "grad_norm": 0.467709481716156,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7407545760179306,
+      "grad_norm": 0.38685527443885803,
+      "learning_rate": 0.0002,
+      "loss": 1.696,
+      "step": 2330
+    },
+    {
+      "epoch": 1.748225625700411,
+      "grad_norm": 0.3578338325023651,
+      "learning_rate": 0.0002,
+      "loss": 1.7041,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7556966753828913,
+      "grad_norm": 0.36057502031326294,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7631677250653717,
+      "grad_norm": 0.3615196645259857,
+      "learning_rate": 0.0002,
+      "loss": 1.6853,
+      "step": 2360
+    },
+    {
+      "epoch": 1.770638774747852,
+      "grad_norm": 0.4118947684764862,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7781098244303326,
+      "grad_norm": 0.4067276120185852,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7855808741128127,
+      "grad_norm": 0.3979823887348175,
+      "learning_rate": 0.0002,
+      "loss": 1.712,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7930519237952933,
+      "grad_norm": 0.44045883417129517,
+      "learning_rate": 0.0002,
+      "loss": 1.7644,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8005229734777737,
+      "grad_norm": 0.3998069167137146,
+      "learning_rate": 0.0002,
+      "loss": 1.7251,
+      "step": 2410
+    },
+    {
+      "epoch": 1.807994023160254,
+      "grad_norm": 0.3450094759464264,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8154650728427344,
+      "grad_norm": 0.3759009838104248,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2430
+    },
+    {
+      "epoch": 1.8229361225252148,
+      "grad_norm": 0.34347015619277954,
+      "learning_rate": 0.0002,
+      "loss": 1.7706,
+      "step": 2440
+    },
+    {
+      "epoch": 1.8304071722076953,
+      "grad_norm": 0.3511228859424591,
+      "learning_rate": 0.0002,
+      "loss": 1.7345,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8378782218901755,
+      "grad_norm": 0.36853715777397156,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 2460
+    },
+    {
+      "epoch": 1.845349271572656,
+      "grad_norm": 0.40659376978874207,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8528203212551362,
+      "grad_norm": 0.39621320366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8602913709376168,
+      "grad_norm": 0.3753979504108429,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8677624206200971,
+      "grad_norm": 0.3811938464641571,
+      "learning_rate": 0.0002,
+      "loss": 1.6622,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8752334703025775,
+      "grad_norm": 0.3432596027851105,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 2510
+    },
+    {
+      "epoch": 1.882704519985058,
+      "grad_norm": 0.3670712113380432,
+      "learning_rate": 0.0002,
+      "loss": 1.7488,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8901755696675382,
+      "grad_norm": 0.40907177329063416,
+      "learning_rate": 0.0002,
+      "loss": 1.705,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8976466193500188,
+      "grad_norm": 0.3821999728679657,
+      "learning_rate": 0.0002,
+      "loss": 1.7148,
+      "step": 2540
+    },
+    {
+      "epoch": 1.905117669032499,
+      "grad_norm": 0.36173978447914124,
+      "learning_rate": 0.0002,
+      "loss": 1.7934,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9125887187149795,
+      "grad_norm": 0.38990336656570435,
+      "learning_rate": 0.0002,
+      "loss": 1.6939,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9200597683974598,
+      "grad_norm": 0.35242322087287903,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 2570
+    },
+    {
+      "epoch": 1.9275308180799402,
+      "grad_norm": 0.3506428003311157,
+      "learning_rate": 0.0002,
+      "loss": 1.7268,
+      "step": 2580
+    },
+    {
+      "epoch": 1.9350018677624208,
+      "grad_norm": 0.39540135860443115,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2590
+    },
+    {
+      "epoch": 1.942472917444901,
+      "grad_norm": 0.3444725573062897,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9499439671273815,
+      "grad_norm": 0.3963521718978882,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9574150168098616,
+      "grad_norm": 0.3689815402030945,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9648860664923422,
+      "grad_norm": 0.3482626676559448,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9723571161748226,
+      "grad_norm": 0.35832616686820984,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 2640
+    },
+    {
+      "epoch": 1.979828165857303,
+      "grad_norm": 0.4776208996772766,
+      "learning_rate": 0.0002,
+      "loss": 1.6681,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9872992155397835,
+      "grad_norm": 0.32570165395736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6696,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9947702652222636,
+      "grad_norm": 0.3380725085735321,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8046749830245972,
+      "eval_runtime": 38.5096,
+      "eval_samples_per_second": 13.373,
+      "eval_steps_per_second": 1.688,
+      "step": 2677
+    },
+    {
+      "epoch": 2.002241314904744,
+      "grad_norm": 0.36817631125450134,
+      "learning_rate": 0.0002,
+      "loss": 1.7265,
+      "step": 2680
+    },
+    {
+      "epoch": 2.0097123645872244,
+      "grad_norm": 0.4056456685066223,
+      "learning_rate": 0.0002,
+      "loss": 1.548,
+      "step": 2690
+    },
+    {
+      "epoch": 2.017183414269705,
+      "grad_norm": 0.37416863441467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5515,
+      "step": 2700
+    },
+    {
+      "epoch": 2.024654463952185,
+      "grad_norm": 0.4273638427257538,
+      "learning_rate": 0.0002,
+      "loss": 1.5895,
+      "step": 2710
+    },
+    {
+      "epoch": 2.0321255136346656,
+      "grad_norm": 0.36497923731803894,
+      "learning_rate": 0.0002,
+      "loss": 1.5884,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0395965633171462,
+      "grad_norm": 0.5021994113922119,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 2730
+    },
+    {
+      "epoch": 2.0470676129996264,
+      "grad_norm": 0.45896220207214355,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 2740
+    },
+    {
+      "epoch": 2.054538662682107,
+      "grad_norm": 0.3973815143108368,
+      "learning_rate": 0.0002,
+      "loss": 1.6305,
+      "step": 2750
+    },
+    {
+      "epoch": 2.062009712364587,
+      "grad_norm": 0.4521815776824951,
+      "learning_rate": 0.0002,
+      "loss": 1.6301,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0694807620470677,
+      "grad_norm": 0.42775002121925354,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2770
+    },
+    {
+      "epoch": 2.076951811729548,
+      "grad_norm": 0.48158586025238037,
+      "learning_rate": 0.0002,
+      "loss": 1.6491,
+      "step": 2780
+    },
+    {
+      "epoch": 2.0844228614120284,
+      "grad_norm": 0.4612371623516083,
+      "learning_rate": 0.0002,
+      "loss": 1.6301,
+      "step": 2790
+    },
+    {
+      "epoch": 2.091893911094509,
+      "grad_norm": 0.42536866664886475,
+      "learning_rate": 0.0002,
+      "loss": 1.6327,
+      "step": 2800
+    },
+    {
+      "epoch": 2.099364960776989,
+      "grad_norm": 0.48515772819519043,
+      "learning_rate": 0.0002,
+      "loss": 1.651,
+      "step": 2810
+    },
+    {
+      "epoch": 2.1068360104594697,
+      "grad_norm": 0.41418662667274475,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 2820
+    },
+    {
+      "epoch": 2.11430706014195,
+      "grad_norm": 0.4683697819709778,
+      "learning_rate": 0.0002,
+      "loss": 1.6266,
+      "step": 2830
+    },
+    {
+      "epoch": 2.1217781098244304,
+      "grad_norm": 0.4484657049179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6586,
+      "step": 2840
+    },
+    {
+      "epoch": 2.1292491595069105,
+      "grad_norm": 0.6621400713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.6483,
+      "step": 2850
+    },
+    {
+      "epoch": 2.136720209189391,
+      "grad_norm": 0.45074811577796936,
+      "learning_rate": 0.0002,
+      "loss": 1.5755,
+      "step": 2860
+    },
+    {
+      "epoch": 2.1441912588718717,
+      "grad_norm": 0.3513113558292389,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2870
+    },
+    {
+      "epoch": 2.151662308554352,
+      "grad_norm": 0.40411314368247986,
+      "learning_rate": 0.0002,
+      "loss": 1.6081,
+      "step": 2880
+    },
+    {
+      "epoch": 2.1591333582368324,
+      "grad_norm": 0.4121065139770508,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 2890
+    },
+    {
+      "epoch": 2.1666044079193125,
+      "grad_norm": 0.44723689556121826,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 2900
+    },
+    {
+      "epoch": 2.174075457601793,
+      "grad_norm": 0.4226122498512268,
+      "learning_rate": 0.0002,
+      "loss": 1.5699,
+      "step": 2910
+    },
+    {
+      "epoch": 2.1815465072842732,
+      "grad_norm": 0.46617650985717773,
+      "learning_rate": 0.0002,
+      "loss": 1.5652,
+      "step": 2920
+    },
+    {
+      "epoch": 2.189017556966754,
+      "grad_norm": 0.4506422281265259,
+      "learning_rate": 0.0002,
+      "loss": 1.6378,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1964886066492344,
+      "grad_norm": 0.4892672896385193,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 2940
+    },
+    {
+      "epoch": 2.2039596563317145,
+      "grad_norm": 0.44095516204833984,
+      "learning_rate": 0.0002,
+      "loss": 1.6176,
+      "step": 2950
+    },
+    {
+      "epoch": 2.211430706014195,
+      "grad_norm": 0.41522109508514404,
+      "learning_rate": 0.0002,
+      "loss": 1.6058,
+      "step": 2960
+    },
+    {
+      "epoch": 2.2189017556966752,
+      "grad_norm": 0.4860858917236328,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2970
+    },
+    {
+      "epoch": 2.226372805379156,
+      "grad_norm": 0.42662516236305237,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2980
+    },
+    {
+      "epoch": 2.233843855061636,
+      "grad_norm": 0.4390648305416107,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 2990
+    },
+    {
+      "epoch": 2.2413149047441165,
+      "grad_norm": 0.47515565156936646,
+      "learning_rate": 0.0002,
+      "loss": 1.5992,
+      "step": 3000
+    },
+    {
+      "epoch": 2.248785954426597,
+      "grad_norm": 0.4104543924331665,
+      "learning_rate": 0.0002,
+      "loss": 1.5563,
+      "step": 3010
+    },
+    {
+      "epoch": 2.2562570041090773,
+      "grad_norm": 0.4404028654098511,
+      "learning_rate": 0.0002,
+      "loss": 1.6895,
+      "step": 3020
+    },
+    {
+      "epoch": 2.263728053791558,
+      "grad_norm": 0.4717366695404053,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 3030
+    },
+    {
+      "epoch": 2.271199103474038,
+      "grad_norm": 0.48345857858657837,
+      "learning_rate": 0.0002,
+      "loss": 1.7287,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2786701531565186,
+      "grad_norm": 0.5312452912330627,
+      "learning_rate": 0.0002,
+      "loss": 1.681,
+      "step": 3050
+    },
+    {
+      "epoch": 2.2861412028389987,
+      "grad_norm": 0.5073099732398987,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2936122525214793,
+      "grad_norm": 0.5027463436126709,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 3070
+    },
+    {
+      "epoch": 2.30108330220396,
+      "grad_norm": 0.5436304807662964,
+      "learning_rate": 0.0002,
+      "loss": 1.5862,
+      "step": 3080
+    },
+    {
+      "epoch": 2.30855435188644,
+      "grad_norm": 0.4701065123081207,
+      "learning_rate": 0.0002,
+      "loss": 1.5763,
+      "step": 3090
+    },
+    {
+      "epoch": 2.3160254015689206,
+      "grad_norm": 0.46988746523857117,
+      "learning_rate": 0.0002,
+      "loss": 1.6177,
+      "step": 3100
+    },
+    {
+      "epoch": 2.3234964512514007,
+      "grad_norm": 0.45112869143486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6502,
+      "step": 3110
+    },
+    {
+      "epoch": 2.3309675009338813,
+      "grad_norm": 0.5173566937446594,
+      "learning_rate": 0.0002,
+      "loss": 1.6291,
+      "step": 3120
+    },
+    {
+      "epoch": 2.3384385506163614,
+      "grad_norm": 0.40345850586891174,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 3130
+    },
+    {
+      "epoch": 2.345909600298842,
+      "grad_norm": 0.4218924939632416,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 3140
+    },
+    {
+      "epoch": 2.3533806499813226,
+      "grad_norm": 0.41857317090034485,
+      "learning_rate": 0.0002,
+      "loss": 1.6341,
+      "step": 3150
+    },
+    {
+      "epoch": 2.3608516996638027,
+      "grad_norm": 0.4197218418121338,
+      "learning_rate": 0.0002,
+      "loss": 1.6087,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3683227493462833,
+      "grad_norm": 0.4260677397251129,
+      "learning_rate": 0.0002,
+      "loss": 1.6572,
+      "step": 3170
+    },
+    {
+      "epoch": 2.3757937990287634,
+      "grad_norm": 0.4209042191505432,
+      "learning_rate": 0.0002,
+      "loss": 1.6376,
+      "step": 3180
+    },
+    {
+      "epoch": 2.383264848711244,
+      "grad_norm": 0.4092234969139099,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 3190
+    },
+    {
+      "epoch": 2.390735898393724,
+      "grad_norm": 0.4928431510925293,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3982069480762047,
+      "grad_norm": 0.49252402782440186,
+      "learning_rate": 0.0002,
+      "loss": 1.6015,
+      "step": 3210
+    },
+    {
+      "epoch": 2.4056779977586853,
+      "grad_norm": 0.4368397295475006,
+      "learning_rate": 0.0002,
+      "loss": 1.5773,
+      "step": 3220
+    },
+    {
+      "epoch": 2.4131490474411654,
+      "grad_norm": 0.46122390031814575,
+      "learning_rate": 0.0002,
+      "loss": 1.6629,
+      "step": 3230
+    },
+    {
+      "epoch": 2.420620097123646,
+      "grad_norm": 0.4272301197052002,
+      "learning_rate": 0.0002,
+      "loss": 1.6224,
+      "step": 3240
+    },
+    {
+      "epoch": 2.428091146806126,
+      "grad_norm": 0.41480937600135803,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 3250
+    },
+    {
+      "epoch": 2.4355621964886067,
+      "grad_norm": 0.48911941051483154,
+      "learning_rate": 0.0002,
+      "loss": 1.6281,
+      "step": 3260
+    },
+    {
+      "epoch": 2.443033246171087,
+      "grad_norm": 0.4444098472595215,
+      "learning_rate": 0.0002,
+      "loss": 1.6846,
+      "step": 3270
+    },
+    {
+      "epoch": 2.4505042958535674,
+      "grad_norm": 0.5111684799194336,
+      "learning_rate": 0.0002,
+      "loss": 1.6961,
+      "step": 3280
+    },
+    {
+      "epoch": 2.457975345536048,
+      "grad_norm": 0.5058825016021729,
+      "learning_rate": 0.0002,
+      "loss": 1.6152,
+      "step": 3290
+    },
+    {
+      "epoch": 2.465446395218528,
+      "grad_norm": 0.44173210859298706,
+      "learning_rate": 0.0002,
+      "loss": 1.625,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4729174449010087,
+      "grad_norm": 0.4659745991230011,
+      "learning_rate": 0.0002,
+      "loss": 1.6491,
+      "step": 3310
+    },
+    {
+      "epoch": 2.480388494583489,
+      "grad_norm": 0.47237497568130493,
+      "learning_rate": 0.0002,
+      "loss": 1.6114,
+      "step": 3320
+    },
+    {
+      "epoch": 2.4878595442659694,
+      "grad_norm": 0.47303131222724915,
+      "learning_rate": 0.0002,
+      "loss": 1.6193,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4953305939484496,
+      "grad_norm": 0.4522389769554138,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 3340
+    },
+    {
+      "epoch": 2.50280164363093,
+      "grad_norm": 0.4467332363128662,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3350
+    },
+    {
+      "epoch": 2.5102726933134107,
+      "grad_norm": 0.4413762092590332,
+      "learning_rate": 0.0002,
+      "loss": 1.6108,
+      "step": 3360
+    },
+    {
+      "epoch": 2.517743742995891,
+      "grad_norm": 0.495514452457428,
+      "learning_rate": 0.0002,
+      "loss": 1.537,
+      "step": 3370
+    },
+    {
+      "epoch": 2.5252147926783715,
+      "grad_norm": 0.4429773986339569,
+      "learning_rate": 0.0002,
+      "loss": 1.5839,
+      "step": 3380
+    },
+    {
+      "epoch": 2.5326858423608516,
+      "grad_norm": 0.4589079022407532,
+      "learning_rate": 0.0002,
+      "loss": 1.6522,
+      "step": 3390
+    },
+    {
+      "epoch": 2.540156892043332,
+      "grad_norm": 0.4683997333049774,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 3400
+    },
+    {
+      "epoch": 2.5476279417258123,
+      "grad_norm": 0.4651731252670288,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 3410
+    },
+    {
+      "epoch": 2.555098991408293,
+      "grad_norm": 0.45818084478378296,
+      "learning_rate": 0.0002,
+      "loss": 1.5918,
+      "step": 3420
+    },
+    {
+      "epoch": 2.5625700410907735,
+      "grad_norm": 0.45209529995918274,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3430
+    },
+    {
+      "epoch": 2.5700410907732536,
+      "grad_norm": 0.4344733655452728,
+      "learning_rate": 0.0002,
+      "loss": 1.5606,
+      "step": 3440
+    },
+    {
+      "epoch": 2.577512140455734,
+      "grad_norm": 0.47435566782951355,
+      "learning_rate": 0.0002,
+      "loss": 1.6748,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5849831901382143,
+      "grad_norm": 0.43841999769210815,
+      "learning_rate": 0.0002,
+      "loss": 1.6237,
+      "step": 3460
+    },
+    {
+      "epoch": 2.592454239820695,
+      "grad_norm": 0.4323869049549103,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 3470
+    },
+    {
+      "epoch": 2.599925289503175,
+      "grad_norm": 0.44355881214141846,
+      "learning_rate": 0.0002,
+      "loss": 1.5494,
+      "step": 3480
+    },
+    {
+      "epoch": 2.6073963391856556,
+      "grad_norm": 0.45847779512405396,
+      "learning_rate": 0.0002,
+      "loss": 1.665,
+      "step": 3490
+    },
+    {
+      "epoch": 2.614867388868136,
+      "grad_norm": 0.4411061704158783,
+      "learning_rate": 0.0002,
+      "loss": 1.6006,
+      "step": 3500
+    },
+    {
+      "epoch": 2.6223384385506163,
+      "grad_norm": 0.4446796178817749,
+      "learning_rate": 0.0002,
+      "loss": 1.5868,
+      "step": 3510
+    },
+    {
+      "epoch": 2.629809488233097,
+      "grad_norm": 0.41969653964042664,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 3520
+    },
+    {
+      "epoch": 2.637280537915577,
+      "grad_norm": 0.5263747572898865,
+      "learning_rate": 0.0002,
+      "loss": 1.6798,
+      "step": 3530
+    },
+    {
+      "epoch": 2.6447515875980576,
+      "grad_norm": 0.47719451785087585,
+      "learning_rate": 0.0002,
+      "loss": 1.6309,
+      "step": 3540
+    },
+    {
+      "epoch": 2.6522226372805378,
+      "grad_norm": 0.46574118733406067,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 3550
+    },
+    {
+      "epoch": 2.6596936869630183,
+      "grad_norm": 0.46867135167121887,
+      "learning_rate": 0.0002,
+      "loss": 1.618,
+      "step": 3560
+    },
+    {
+      "epoch": 2.667164736645499,
+      "grad_norm": 0.4441198706626892,
+      "learning_rate": 0.0002,
+      "loss": 1.5885,
+      "step": 3570
+    },
+    {
+      "epoch": 2.674635786327979,
+      "grad_norm": 0.4871319830417633,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6821068360104596,
+      "grad_norm": 0.43900373578071594,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6895778856929398,
+      "grad_norm": 0.42509549856185913,
+      "learning_rate": 0.0002,
+      "loss": 1.6071,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6970489353754203,
+      "grad_norm": 0.4691086709499359,
+      "learning_rate": 0.0002,
+      "loss": 1.5651,
+      "step": 3610
+    },
+    {
+      "epoch": 2.7045199850579005,
+      "grad_norm": 0.46318942308425903,
+      "learning_rate": 0.0002,
+      "loss": 1.5491,
+      "step": 3620
+    },
+    {
+      "epoch": 2.711991034740381,
+      "grad_norm": 0.44631096720695496,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 3630
+    },
+    {
+      "epoch": 2.7194620844228616,
+      "grad_norm": 0.42315489053726196,
+      "learning_rate": 0.0002,
+      "loss": 1.6831,
+      "step": 3640
+    },
+    {
+      "epoch": 2.7269331341053418,
+      "grad_norm": 0.4971241056919098,
+      "learning_rate": 0.0002,
+      "loss": 1.6008,
+      "step": 3650
+    },
+    {
+      "epoch": 2.7344041837878224,
+      "grad_norm": 0.4578486382961273,
+      "learning_rate": 0.0002,
+      "loss": 1.6042,
+      "step": 3660
+    },
+    {
+      "epoch": 2.7418752334703025,
+      "grad_norm": 0.46584776043891907,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 3670
+    },
+    {
+      "epoch": 2.749346283152783,
+      "grad_norm": 0.4951731264591217,
+      "learning_rate": 0.0002,
+      "loss": 1.6809,
+      "step": 3680
+    },
+    {
+      "epoch": 2.756817332835263,
+      "grad_norm": 0.4935225546360016,
+      "learning_rate": 0.0002,
+      "loss": 1.6226,
+      "step": 3690
+    },
+    {
+      "epoch": 2.764288382517744,
+      "grad_norm": 0.41805586218833923,
+      "learning_rate": 0.0002,
+      "loss": 1.5878,
+      "step": 3700
+    },
+    {
+      "epoch": 2.7717594322002244,
+      "grad_norm": 0.4417555630207062,
+      "learning_rate": 0.0002,
+      "loss": 1.7173,
+      "step": 3710
+    },
+    {
+      "epoch": 2.7792304818827045,
+      "grad_norm": 0.48229655623435974,
+      "learning_rate": 0.0002,
+      "loss": 1.6398,
+      "step": 3720
+    },
+    {
+      "epoch": 2.786701531565185,
+      "grad_norm": 0.48562315106391907,
+      "learning_rate": 0.0002,
+      "loss": 1.6074,
+      "step": 3730
+    },
+    {
+      "epoch": 2.794172581247665,
+      "grad_norm": 0.4473940432071686,
+      "learning_rate": 0.0002,
+      "loss": 1.607,
+      "step": 3740
+    },
+    {
+      "epoch": 2.801643630930146,
+      "grad_norm": 0.4626813232898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6065,
+      "step": 3750
+    },
+    {
+      "epoch": 2.809114680612626,
+      "grad_norm": 0.4339792728424072,
+      "learning_rate": 0.0002,
+      "loss": 1.6296,
+      "step": 3760
+    },
+    {
+      "epoch": 2.8165857302951065,
+      "grad_norm": 0.5250858068466187,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 3770
+    },
+    {
+      "epoch": 2.824056779977587,
+      "grad_norm": 0.4537523090839386,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 3780
+    },
+    {
+      "epoch": 2.831527829660067,
+      "grad_norm": 0.5646113157272339,
+      "learning_rate": 0.0002,
+      "loss": 1.6535,
+      "step": 3790
+    },
+    {
+      "epoch": 2.8389988793425474,
+      "grad_norm": 0.44243332743644714,
+      "learning_rate": 0.0002,
+      "loss": 1.5712,
+      "step": 3800
+    },
+    {
+      "epoch": 2.846469929025028,
+      "grad_norm": 0.4585791826248169,
+      "learning_rate": 0.0002,
+      "loss": 1.6478,
+      "step": 3810
+    },
+    {
+      "epoch": 2.8539409787075085,
+      "grad_norm": 0.489702045917511,
+      "learning_rate": 0.0002,
+      "loss": 1.6854,
+      "step": 3820
+    },
+    {
+      "epoch": 2.8614120283899886,
+      "grad_norm": 0.502470850944519,
+      "learning_rate": 0.0002,
+      "loss": 1.7066,
+      "step": 3830
+    },
+    {
+      "epoch": 2.8688830780724692,
+      "grad_norm": 0.4395960867404938,
+      "learning_rate": 0.0002,
+      "loss": 1.5785,
+      "step": 3840
+    },
+    {
+      "epoch": 2.87635412775495,
+      "grad_norm": 0.4348670244216919,
+      "learning_rate": 0.0002,
+      "loss": 1.6434,
+      "step": 3850
+    },
+    {
+      "epoch": 2.88382517743743,
+      "grad_norm": 0.48852720856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 3860
+    },
+    {
+      "epoch": 2.89129622711991,
+      "grad_norm": 0.45317450165748596,
+      "learning_rate": 0.0002,
+      "loss": 1.5916,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8987672768023907,
+      "grad_norm": 0.4732758700847626,
+      "learning_rate": 0.0002,
+      "loss": 1.6486,
+      "step": 3880
+    },
+    {
+      "epoch": 2.9062383264848712,
+      "grad_norm": 0.45238012075424194,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 3890
+    },
+    {
+      "epoch": 2.9137093761673514,
+      "grad_norm": 0.48838064074516296,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 3900
+    },
+    {
+      "epoch": 2.921180425849832,
+      "grad_norm": 0.43496349453926086,
+      "learning_rate": 0.0002,
+      "loss": 1.658,
+      "step": 3910
+    },
+    {
+      "epoch": 2.9286514755323125,
+      "grad_norm": 0.47963935136795044,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 3920
+    },
+    {
+      "epoch": 2.9361225252147927,
+      "grad_norm": 0.4544987976551056,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 3930
+    },
+    {
+      "epoch": 2.943593574897273,
+      "grad_norm": 0.4622892141342163,
+      "learning_rate": 0.0002,
+      "loss": 1.6192,
+      "step": 3940
+    },
+    {
+      "epoch": 2.9510646245797534,
+      "grad_norm": 0.47026222944259644,
+      "learning_rate": 0.0002,
+      "loss": 1.6178,
+      "step": 3950
+    },
+    {
+      "epoch": 2.958535674262234,
+      "grad_norm": 0.4549552798271179,
+      "learning_rate": 0.0002,
+      "loss": 1.6612,
+      "step": 3960
+    },
+    {
+      "epoch": 2.966006723944714,
+      "grad_norm": 0.46647515892982483,
+      "learning_rate": 0.0002,
+      "loss": 1.6458,
+      "step": 3970
+    },
+    {
+      "epoch": 2.9734777736271947,
+      "grad_norm": 0.45095112919807434,
+      "learning_rate": 0.0002,
+      "loss": 1.6051,
+      "step": 3980
+    },
+    {
+      "epoch": 2.9809488233096753,
+      "grad_norm": 0.4690017104148865,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 3990
+    },
+    {
+      "epoch": 2.9884198729921554,
+      "grad_norm": 0.4603444039821625,
+      "learning_rate": 0.0002,
+      "loss": 1.6061,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9958909226746355,
+      "grad_norm": 0.4743294417858124,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 4010
+    },
+    {
+      "epoch": 2.999626447515876,
+      "eval_loss": 1.8252571821212769,
+      "eval_runtime": 38.7853,
+      "eval_samples_per_second": 13.278,
+      "eval_steps_per_second": 1.676,
+      "step": 4015
+    },
+    {
+      "epoch": 3.003361972357116,
+      "grad_norm": 0.4919724464416504,
+      "learning_rate": 0.0002,
+      "loss": 1.6512,
+      "step": 4020
+    },
+    {
+      "epoch": 3.0108330220395967,
+      "grad_norm": 0.4747185707092285,
+      "learning_rate": 0.0002,
+      "loss": 1.5354,
+      "step": 4030
+    },
+    {
+      "epoch": 3.018304071722077,
+      "grad_norm": 0.4797595143318176,
+      "learning_rate": 0.0002,
+      "loss": 1.568,
+      "step": 4040
+    },
+    {
+      "epoch": 3.0257751214045574,
+      "grad_norm": 0.5450999140739441,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 4050
+    },
+    {
+      "epoch": 3.0332461710870375,
+      "grad_norm": 0.49058812856674194,
+      "learning_rate": 0.0002,
+      "loss": 1.5065,
+      "step": 4060
+    },
+    {
+      "epoch": 3.040717220769518,
+      "grad_norm": 0.5219563841819763,
+      "learning_rate": 0.0002,
+      "loss": 1.4884,
+      "step": 4070
+    },
+    {
+      "epoch": 3.0481882704519987,
+      "grad_norm": 0.515628457069397,
+      "learning_rate": 0.0002,
+      "loss": 1.4742,
+      "step": 4080
+    },
+    {
+      "epoch": 3.055659320134479,
+      "grad_norm": 0.6145984530448914,
+      "learning_rate": 0.0002,
+      "loss": 1.5313,
+      "step": 4090
+    },
+    {
+      "epoch": 3.0631303698169594,
+      "grad_norm": 0.6067144274711609,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 4100
+    },
+    {
+      "epoch": 3.0706014194994395,
+      "grad_norm": 0.5773133039474487,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 4110
+    },
+    {
+      "epoch": 3.07807246918192,
+      "grad_norm": 0.6894241571426392,
+      "learning_rate": 0.0002,
+      "loss": 1.5374,
+      "step": 4120
+    },
+    {
+      "epoch": 3.0855435188644003,
+      "grad_norm": 0.6422514915466309,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 4130
+    },
+    {
+      "epoch": 3.093014568546881,
+      "grad_norm": 0.6119855046272278,
+      "learning_rate": 0.0002,
+      "loss": 1.4724,
+      "step": 4140
+    },
+    {
+      "epoch": 3.1004856182293614,
+      "grad_norm": 0.5847280025482178,
+      "learning_rate": 0.0002,
+      "loss": 1.5361,
+      "step": 4150
+    },
+    {
+      "epoch": 3.1079566679118416,
+      "grad_norm": 0.5401515960693359,
+      "learning_rate": 0.0002,
+      "loss": 1.5151,
+      "step": 4160
+    },
+    {
+      "epoch": 3.115427717594322,
+      "grad_norm": 0.6501587629318237,
+      "learning_rate": 0.0002,
+      "loss": 1.502,
+      "step": 4170
+    },
+    {
+      "epoch": 3.1228987672768023,
+      "grad_norm": 0.5988039374351501,
+      "learning_rate": 0.0002,
+      "loss": 1.4952,
+      "step": 4180
+    },
+    {
+      "epoch": 3.130369816959283,
+      "grad_norm": 0.4982665181159973,
+      "learning_rate": 0.0002,
+      "loss": 1.5287,
+      "step": 4190
+    },
+    {
+      "epoch": 3.137840866641763,
+      "grad_norm": 0.5548039078712463,
+      "learning_rate": 0.0002,
+      "loss": 1.5078,
+      "step": 4200
+    },
+    {
+      "epoch": 3.1453119163242436,
+      "grad_norm": 0.5920777320861816,
+      "learning_rate": 0.0002,
+      "loss": 1.4904,
+      "step": 4210
+    },
+    {
+      "epoch": 3.152782966006724,
+      "grad_norm": 0.6965190172195435,
+      "learning_rate": 0.0002,
+      "loss": 1.442,
+      "step": 4220
+    },
+    {
+      "epoch": 3.1602540156892043,
+      "grad_norm": 0.5196244716644287,
+      "learning_rate": 0.0002,
+      "loss": 1.557,
+      "step": 4230
+    },
+    {
+      "epoch": 3.167725065371685,
+      "grad_norm": 0.6942682266235352,
+      "learning_rate": 0.0002,
+      "loss": 1.5706,
+      "step": 4240
+    },
+    {
+      "epoch": 3.175196115054165,
+      "grad_norm": 0.5765156149864197,
+      "learning_rate": 0.0002,
+      "loss": 1.5407,
+      "step": 4250
+    },
+    {
+      "epoch": 3.1826671647366456,
+      "grad_norm": 0.5801976919174194,
+      "learning_rate": 0.0002,
+      "loss": 1.4963,
+      "step": 4260
+    },
+    {
+      "epoch": 3.1901382144191257,
+      "grad_norm": 0.6260752081871033,
+      "learning_rate": 0.0002,
+      "loss": 1.4988,
+      "step": 4270
+    },
+    {
+      "epoch": 3.1976092641016063,
+      "grad_norm": 0.6610770225524902,
+      "learning_rate": 0.0002,
+      "loss": 1.5074,
+      "step": 4280
+    },
+    {
+      "epoch": 3.205080313784087,
+      "grad_norm": 0.5762143135070801,
+      "learning_rate": 0.0002,
+      "loss": 1.4657,
+      "step": 4290
+    },
+    {
+      "epoch": 3.212551363466567,
+      "grad_norm": 0.5926990509033203,
+      "learning_rate": 0.0002,
+      "loss": 1.5181,
+      "step": 4300
+    },
+    {
+      "epoch": 3.2200224131490476,
+      "grad_norm": 0.7373854517936707,
+      "learning_rate": 0.0002,
+      "loss": 1.5492,
+      "step": 4310
+    },
+    {
+      "epoch": 3.2274934628315277,
+      "grad_norm": 0.5963311195373535,
+      "learning_rate": 0.0002,
+      "loss": 1.4648,
+      "step": 4320
+    },
+    {
+      "epoch": 3.2349645125140083,
+      "grad_norm": 0.5754616856575012,
+      "learning_rate": 0.0002,
+      "loss": 1.5262,
+      "step": 4330
+    },
+    {
+      "epoch": 3.2424355621964884,
+      "grad_norm": 0.6116095781326294,
+      "learning_rate": 0.0002,
+      "loss": 1.4767,
+      "step": 4340
+    },
+    {
+      "epoch": 3.249906611878969,
+      "grad_norm": 0.6001536846160889,
+      "learning_rate": 0.0002,
+      "loss": 1.5008,
+      "step": 4350
+    },
+    {
+      "epoch": 3.257377661561449,
+      "grad_norm": 0.5270227789878845,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 4360
+    },
+    {
+      "epoch": 3.2648487112439297,
+      "grad_norm": 0.6666602492332458,
+      "learning_rate": 0.0002,
+      "loss": 1.5235,
+      "step": 4370
+    },
+    {
+      "epoch": 3.2723197609264103,
+      "grad_norm": 0.520310640335083,
+      "learning_rate": 0.0002,
+      "loss": 1.5665,
+      "step": 4380
+    },
+    {
+      "epoch": 3.2797908106088904,
+      "grad_norm": 0.5165975093841553,
+      "learning_rate": 0.0002,
+      "loss": 1.542,
+      "step": 4390
+    },
+    {
+      "epoch": 3.287261860291371,
+      "grad_norm": 0.6080228686332703,
+      "learning_rate": 0.0002,
+      "loss": 1.4746,
+      "step": 4400
+    },
+    {
+      "epoch": 3.294732909973851,
+      "grad_norm": 0.670122504234314,
+      "learning_rate": 0.0002,
+      "loss": 1.4901,
+      "step": 4410
+    },
+    {
+      "epoch": 3.3022039596563317,
+      "grad_norm": 0.6019457578659058,
+      "learning_rate": 0.0002,
+      "loss": 1.4677,
+      "step": 4420
+    },
+    {
+      "epoch": 3.309675009338812,
+      "grad_norm": 0.5519300103187561,
+      "learning_rate": 0.0002,
+      "loss": 1.4249,
+      "step": 4430
+    },
+    {
+      "epoch": 3.3171460590212924,
+      "grad_norm": 0.5958521962165833,
+      "learning_rate": 0.0002,
+      "loss": 1.555,
+      "step": 4440
+    },
+    {
+      "epoch": 3.324617108703773,
+      "grad_norm": 0.5552705526351929,
+      "learning_rate": 0.0002,
+      "loss": 1.5067,
+      "step": 4450
+    },
+    {
+      "epoch": 3.332088158386253,
+      "grad_norm": 0.6583784818649292,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 4460
+    },
+    {
+      "epoch": 3.3395592080687337,
+      "grad_norm": 0.5815939903259277,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 4470
+    },
+    {
+      "epoch": 3.347030257751214,
+      "grad_norm": 1.3342205286026,
+      "learning_rate": 0.0002,
+      "loss": 1.5942,
+      "step": 4480
+    },
+    {
+      "epoch": 3.3545013074336945,
+      "grad_norm": 0.6341500878334045,
+      "learning_rate": 0.0002,
+      "loss": 1.484,
+      "step": 4490
+    },
+    {
+      "epoch": 3.3619723571161746,
+      "grad_norm": 0.6384079456329346,
+      "learning_rate": 0.0002,
+      "loss": 1.5219,
+      "step": 4500
+    },
+    {
+      "epoch": 3.369443406798655,
+      "grad_norm": 0.6098346710205078,
+      "learning_rate": 0.0002,
+      "loss": 1.5222,
+      "step": 4510
+    },
+    {
+      "epoch": 3.3769144564811358,
+      "grad_norm": 0.5958296656608582,
+      "learning_rate": 0.0002,
+      "loss": 1.5475,
+      "step": 4520
+    },
+    {
+      "epoch": 3.384385506163616,
+      "grad_norm": 0.6157881617546082,
+      "learning_rate": 0.0002,
+      "loss": 1.5171,
+      "step": 4530
+    },
+    {
+      "epoch": 3.3918565558460965,
+      "grad_norm": 0.5671007037162781,
+      "learning_rate": 0.0002,
+      "loss": 1.569,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3993276055285766,
+      "grad_norm": 0.6203294992446899,
+      "learning_rate": 0.0002,
+      "loss": 1.604,
+      "step": 4550
+    },
+    {
+      "epoch": 3.406798655211057,
+      "grad_norm": 0.6743317246437073,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 4560
+    },
+    {
+      "epoch": 3.4142697048935373,
+      "grad_norm": 0.731765627861023,
+      "learning_rate": 0.0002,
+      "loss": 1.5034,
+      "step": 4570
+    },
+    {
+      "epoch": 3.421740754576018,
+      "grad_norm": 0.6285187602043152,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 4580
+    },
+    {
+      "epoch": 3.4292118042584985,
+      "grad_norm": 0.612680196762085,
+      "learning_rate": 0.0002,
+      "loss": 1.5296,
+      "step": 4590
+    },
+    {
+      "epoch": 3.4366828539409786,
+      "grad_norm": 0.6413681507110596,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 4600
+    },
+    {
+      "epoch": 3.444153903623459,
+      "grad_norm": 0.6240990161895752,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4610
+    },
+    {
+      "epoch": 3.4516249533059393,
+      "grad_norm": 0.5095735192298889,
+      "learning_rate": 0.0002,
+      "loss": 1.5887,
+      "step": 4620
+    },
+    {
+      "epoch": 3.45909600298842,
+      "grad_norm": 0.5699611902236938,
+      "learning_rate": 0.0002,
+      "loss": 1.4906,
+      "step": 4630
+    },
+    {
+      "epoch": 3.4665670526709,
+      "grad_norm": 0.7289775609970093,
+      "learning_rate": 0.0002,
+      "loss": 1.5176,
+      "step": 4640
+    },
+    {
+      "epoch": 3.4740381023533806,
+      "grad_norm": 0.6211609840393066,
+      "learning_rate": 0.0002,
+      "loss": 1.5467,
+      "step": 4650
+    },
+    {
+      "epoch": 3.481509152035861,
+      "grad_norm": 0.5714802145957947,
+      "learning_rate": 0.0002,
+      "loss": 1.533,
+      "step": 4660
+    },
+    {
+      "epoch": 3.4889802017183413,
+      "grad_norm": 0.6287049651145935,
+      "learning_rate": 0.0002,
+      "loss": 1.5096,
+      "step": 4670
+    },
+    {
+      "epoch": 3.496451251400822,
+      "grad_norm": 0.5480595827102661,
+      "learning_rate": 0.0002,
+      "loss": 1.4212,
+      "step": 4680
+    },
+    {
+      "epoch": 3.503922301083302,
+      "grad_norm": 0.5683253407478333,
+      "learning_rate": 0.0002,
+      "loss": 1.4746,
+      "step": 4690
+    },
+    {
+      "epoch": 3.5113933507657826,
+      "grad_norm": 0.601140558719635,
+      "learning_rate": 0.0002,
+      "loss": 1.5012,
+      "step": 4700
+    },
+    {
+      "epoch": 3.5188644004482628,
+      "grad_norm": 0.5344498157501221,
+      "learning_rate": 0.0002,
+      "loss": 1.5383,
+      "step": 4710
+    },
+    {
+      "epoch": 3.5263354501307433,
+      "grad_norm": 0.5739690661430359,
+      "learning_rate": 0.0002,
+      "loss": 1.5428,
+      "step": 4720
+    },
+    {
+      "epoch": 3.533806499813224,
+      "grad_norm": 0.5640085935592651,
+      "learning_rate": 0.0002,
+      "loss": 1.5589,
+      "step": 4730
+    },
+    {
+      "epoch": 3.541277549495704,
+      "grad_norm": 0.5967805981636047,
+      "learning_rate": 0.0002,
+      "loss": 1.487,
+      "step": 4740
+    },
+    {
+      "epoch": 3.5487485991781846,
+      "grad_norm": 0.6138835549354553,
+      "learning_rate": 0.0002,
+      "loss": 1.5461,
+      "step": 4750
+    },
+    {
+      "epoch": 3.5562196488606648,
+      "grad_norm": 0.6779900193214417,
+      "learning_rate": 0.0002,
+      "loss": 1.5502,
+      "step": 4760
+    },
+    {
+      "epoch": 3.5636906985431454,
+      "grad_norm": 0.6122010350227356,
+      "learning_rate": 0.0002,
+      "loss": 1.4917,
+      "step": 4770
+    },
+    {
+      "epoch": 3.5711617482256255,
+      "grad_norm": 0.5685241222381592,
+      "learning_rate": 0.0002,
+      "loss": 1.5405,
+      "step": 4780
+    },
+    {
+      "epoch": 3.578632797908106,
+      "grad_norm": 0.604583203792572,
+      "learning_rate": 0.0002,
+      "loss": 1.5427,
+      "step": 4790
+    },
+    {
+      "epoch": 3.5861038475905866,
+      "grad_norm": 0.651165246963501,
+      "learning_rate": 0.0002,
+      "loss": 1.4514,
+      "step": 4800
+    },
+    {
+      "epoch": 3.593574897273067,
+      "grad_norm": 0.6398511528968811,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 4810
+    },
+    {
+      "epoch": 3.6010459469555474,
+      "grad_norm": 0.6444641351699829,
+      "learning_rate": 0.0002,
+      "loss": 1.4261,
+      "step": 4820
+    },
+    {
+      "epoch": 3.6085169966380275,
+      "grad_norm": 0.6018481850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.5274,
+      "step": 4830
+    },
+    {
+      "epoch": 3.615988046320508,
+      "grad_norm": 0.6025291085243225,
+      "learning_rate": 0.0002,
+      "loss": 1.4647,
+      "step": 4840
+    },
+    {
+      "epoch": 3.623459096002988,
+      "grad_norm": 0.6810156106948853,
+      "learning_rate": 0.0002,
+      "loss": 1.5609,
+      "step": 4850
+    },
+    {
+      "epoch": 3.630930145685469,
+      "grad_norm": 0.6408044695854187,
+      "learning_rate": 0.0002,
+      "loss": 1.5299,
+      "step": 4860
+    },
+    {
+      "epoch": 3.6384011953679494,
+      "grad_norm": 0.5608272552490234,
+      "learning_rate": 0.0002,
+      "loss": 1.5366,
+      "step": 4870
+    },
+    {
+      "epoch": 3.6458722450504295,
+      "grad_norm": 0.6136814951896667,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 4880
+    },
+    {
+      "epoch": 3.65334329473291,
+      "grad_norm": 0.5927900075912476,
+      "learning_rate": 0.0002,
+      "loss": 1.5021,
+      "step": 4890
+    },
+    {
+      "epoch": 3.66081434441539,
+      "grad_norm": 0.5336901545524597,
+      "learning_rate": 0.0002,
+      "loss": 1.6084,
+      "step": 4900
+    },
+    {
+      "epoch": 3.668285394097871,
+      "grad_norm": 0.7823320627212524,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 4910
+    },
+    {
+      "epoch": 3.675756443780351,
+      "grad_norm": 0.6703504323959351,
+      "learning_rate": 0.0002,
+      "loss": 1.4881,
+      "step": 4920
+    },
+    {
+      "epoch": 3.6832274934628315,
+      "grad_norm": 0.6061160564422607,
+      "learning_rate": 0.0002,
+      "loss": 1.5332,
+      "step": 4930
+    },
+    {
+      "epoch": 3.690698543145312,
+      "grad_norm": 0.6237227916717529,
+      "learning_rate": 0.0002,
+      "loss": 1.5405,
+      "step": 4940
+    },
+    {
+      "epoch": 3.6981695928277922,
+      "grad_norm": 0.5985278487205505,
+      "learning_rate": 0.0002,
+      "loss": 1.497,
+      "step": 4950
+    },
+    {
+      "epoch": 3.705640642510273,
+      "grad_norm": 0.6483839750289917,
+      "learning_rate": 0.0002,
+      "loss": 1.5132,
+      "step": 4960
+    },
+    {
+      "epoch": 3.713111692192753,
+      "grad_norm": 0.5788805484771729,
+      "learning_rate": 0.0002,
+      "loss": 1.5338,
+      "step": 4970
+    },
+    {
+      "epoch": 3.7205827418752335,
+      "grad_norm": 0.5609974265098572,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 4980
+    },
+    {
+      "epoch": 3.7280537915577137,
+      "grad_norm": 0.5681300759315491,
+      "learning_rate": 0.0002,
+      "loss": 1.4759,
+      "step": 4990
+    },
+    {
+      "epoch": 3.7355248412401942,
+      "grad_norm": 0.5860186219215393,
+      "learning_rate": 0.0002,
+      "loss": 1.6018,
+      "step": 5000
+    },
+    {
+      "epoch": 3.742995890922675,
+      "grad_norm": 0.5718157291412354,
+      "learning_rate": 0.0002,
+      "loss": 1.58,
+      "step": 5010
+    },
+    {
+      "epoch": 3.750466940605155,
+      "grad_norm": 0.6173721551895142,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 5020
+    },
+    {
+      "epoch": 3.7579379902876355,
+      "grad_norm": 0.629152238368988,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 5030
+    },
+    {
+      "epoch": 3.7654090399701157,
+      "grad_norm": 0.5666284561157227,
+      "learning_rate": 0.0002,
+      "loss": 1.519,
+      "step": 5040
+    },
+    {
+      "epoch": 3.7728800896525962,
+      "grad_norm": 0.6053005456924438,
+      "learning_rate": 0.0002,
+      "loss": 1.5329,
+      "step": 5050
+    },
+    {
+      "epoch": 3.7803511393350764,
+      "grad_norm": 0.5870583057403564,
+      "learning_rate": 0.0002,
+      "loss": 1.5404,
+      "step": 5060
+    },
+    {
+      "epoch": 3.787822189017557,
+      "grad_norm": 0.5422009229660034,
+      "learning_rate": 0.0002,
+      "loss": 1.4444,
+      "step": 5070
+    },
+    {
+      "epoch": 3.7952932387000375,
+      "grad_norm": 0.5396918058395386,
+      "learning_rate": 0.0002,
+      "loss": 1.5308,
+      "step": 5080
+    },
+    {
+      "epoch": 3.8027642883825177,
+      "grad_norm": 0.5544713139533997,
+      "learning_rate": 0.0002,
+      "loss": 1.464,
+      "step": 5090
+    },
+    {
+      "epoch": 3.8102353380649983,
+      "grad_norm": 0.5983749628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5100
+    },
+    {
+      "epoch": 3.8177063877474784,
+      "grad_norm": 0.5702024102210999,
+      "learning_rate": 0.0002,
+      "loss": 1.4972,
+      "step": 5110
+    },
+    {
+      "epoch": 3.825177437429959,
+      "grad_norm": 0.5436882376670837,
+      "learning_rate": 0.0002,
+      "loss": 1.5471,
+      "step": 5120
+    },
+    {
+      "epoch": 3.832648487112439,
+      "grad_norm": 0.5453617572784424,
+      "learning_rate": 0.0002,
+      "loss": 1.5118,
+      "step": 5130
+    },
+    {
+      "epoch": 3.8401195367949197,
+      "grad_norm": 0.6269069314002991,
+      "learning_rate": 0.0002,
+      "loss": 1.5732,
+      "step": 5140
+    },
+    {
+      "epoch": 3.8475905864774003,
+      "grad_norm": 0.6189185380935669,
+      "learning_rate": 0.0002,
+      "loss": 1.4959,
+      "step": 5150
+    },
+    {
+      "epoch": 3.8550616361598804,
+      "grad_norm": 0.6653388142585754,
+      "learning_rate": 0.0002,
+      "loss": 1.4999,
+      "step": 5160
+    },
+    {
+      "epoch": 3.862532685842361,
+      "grad_norm": 0.5771768689155579,
+      "learning_rate": 0.0002,
+      "loss": 1.5075,
+      "step": 5170
+    },
+    {
+      "epoch": 3.870003735524841,
+      "grad_norm": 0.6052790880203247,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5180
+    },
+    {
+      "epoch": 3.8774747852073217,
+      "grad_norm": 0.6572316884994507,
+      "learning_rate": 0.0002,
+      "loss": 1.4987,
+      "step": 5190
+    },
+    {
+      "epoch": 3.884945834889802,
+      "grad_norm": 0.670576810836792,
+      "learning_rate": 0.0002,
+      "loss": 1.5241,
+      "step": 5200
+    },
+    {
+      "epoch": 3.8924168845722824,
+      "grad_norm": 0.5728798508644104,
+      "learning_rate": 0.0002,
+      "loss": 1.4777,
+      "step": 5210
+    },
+    {
+      "epoch": 3.899887934254763,
+      "grad_norm": 0.6340774297714233,
+      "learning_rate": 0.0002,
+      "loss": 1.5351,
+      "step": 5220
+    },
+    {
+      "epoch": 3.907358983937243,
+      "grad_norm": 0.5981315970420837,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 5230
+    },
+    {
+      "epoch": 3.9148300336197237,
+      "grad_norm": 0.6212025880813599,
+      "learning_rate": 0.0002,
+      "loss": 1.4875,
+      "step": 5240
+    },
+    {
+      "epoch": 3.922301083302204,
+      "grad_norm": 0.6202296018600464,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5250
+    },
+    {
+      "epoch": 3.9297721329846844,
+      "grad_norm": 0.6159142255783081,
+      "learning_rate": 0.0002,
+      "loss": 1.5765,
+      "step": 5260
+    },
+    {
+      "epoch": 3.9372431826671646,
+      "grad_norm": 0.6519438624382019,
+      "learning_rate": 0.0002,
+      "loss": 1.4938,
+      "step": 5270
+    },
+    {
+      "epoch": 3.944714232349645,
+      "grad_norm": 0.539813756942749,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 5280
+    },
+    {
+      "epoch": 3.9521852820321257,
+      "grad_norm": 0.6443665027618408,
+      "learning_rate": 0.0002,
+      "loss": 1.5921,
+      "step": 5290
+    },
+    {
+      "epoch": 3.959656331714606,
+      "grad_norm": 0.6635757684707642,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 5300
+    },
+    {
+      "epoch": 3.9671273813970864,
+      "grad_norm": 0.589363157749176,
+      "learning_rate": 0.0002,
+      "loss": 1.5485,
+      "step": 5310
+    },
+    {
+      "epoch": 3.9745984310795666,
+      "grad_norm": 0.5788735747337341,
+      "learning_rate": 0.0002,
+      "loss": 1.5498,
+      "step": 5320
+    },
+    {
+      "epoch": 3.982069480762047,
+      "grad_norm": 0.5976864695549011,
+      "learning_rate": 0.0002,
+      "loss": 1.5607,
+      "step": 5330
+    },
+    {
+      "epoch": 3.9895405304445273,
+      "grad_norm": 0.6624067425727844,
+      "learning_rate": 0.0002,
+      "loss": 1.5302,
+      "step": 5340
+    },
+    {
+      "epoch": 3.997011580127008,
+      "grad_norm": 0.6738956570625305,
+      "learning_rate": 0.0002,
+      "loss": 1.5904,
+      "step": 5350
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.868006944656372,
+      "eval_runtime": 38.5153,
+      "eval_samples_per_second": 13.371,
+      "eval_steps_per_second": 1.688,
+      "step": 5354
+    },
+    {
+      "epoch": 4.004482629809488,
+      "grad_norm": 0.6023468971252441,
+      "learning_rate": 0.0002,
+      "loss": 1.4535,
+      "step": 5360
+    },
+    {
+      "epoch": 4.011953679491969,
+      "grad_norm": 0.8589285612106323,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 5370
+    },
+    {
+      "epoch": 4.019424729174449,
+      "grad_norm": 0.7477491497993469,
+      "learning_rate": 0.0002,
+      "loss": 1.3952,
+      "step": 5380
+    },
+    {
+      "epoch": 4.02689577885693,
+      "grad_norm": 0.7601922154426575,
+      "learning_rate": 0.0002,
+      "loss": 1.3745,
+      "step": 5390
+    },
+    {
+      "epoch": 4.03436682853941,
+      "grad_norm": 0.8115614056587219,
+      "learning_rate": 0.0002,
+      "loss": 1.4133,
+      "step": 5400
+    },
+    {
+      "epoch": 4.04183787822189,
+      "grad_norm": 0.669925332069397,
+      "learning_rate": 0.0002,
+      "loss": 1.3748,
+      "step": 5410
+    },
+    {
+      "epoch": 4.04930892790437,
+      "grad_norm": 0.8091904520988464,
+      "learning_rate": 0.0002,
+      "loss": 1.2835,
+      "step": 5420
+    },
+    {
+      "epoch": 4.056779977586851,
+      "grad_norm": 0.709405779838562,
+      "learning_rate": 0.0002,
+      "loss": 1.3615,
+      "step": 5430
+    },
+    {
+      "epoch": 4.064251027269331,
+      "grad_norm": 1.0006179809570312,
+      "learning_rate": 0.0002,
+      "loss": 1.3558,
+      "step": 5440
+    },
+    {
+      "epoch": 4.071722076951811,
+      "grad_norm": 0.7017965912818909,
+      "learning_rate": 0.0002,
+      "loss": 1.3491,
+      "step": 5450
+    },
+    {
+      "epoch": 4.0791931266342925,
+      "grad_norm": 0.8991572260856628,
+      "learning_rate": 0.0002,
+      "loss": 1.3642,
+      "step": 5460
+    },
+    {
+      "epoch": 4.086664176316773,
+      "grad_norm": 0.9064797759056091,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 5470
+    },
+    {
+      "epoch": 4.094135225999253,
+      "grad_norm": 0.7981749176979065,
+      "learning_rate": 0.0002,
+      "loss": 1.3425,
+      "step": 5480
+    },
+    {
+      "epoch": 4.101606275681733,
+      "grad_norm": 0.7280883193016052,
+      "learning_rate": 0.0002,
+      "loss": 1.3826,
+      "step": 5490
+    },
+    {
+      "epoch": 4.109077325364214,
+      "grad_norm": 0.7419600486755371,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 5500
+    },
+    {
+      "epoch": 4.116548375046694,
+      "grad_norm": 0.8019949197769165,
+      "learning_rate": 0.0002,
+      "loss": 1.3199,
+      "step": 5510
+    },
+    {
+      "epoch": 4.124019424729174,
+      "grad_norm": 0.7501229047775269,
+      "learning_rate": 0.0002,
+      "loss": 1.3133,
+      "step": 5520
+    },
+    {
+      "epoch": 4.131490474411655,
+      "grad_norm": 0.8166249990463257,
+      "learning_rate": 0.0002,
+      "loss": 1.4432,
+      "step": 5530
+    },
+    {
+      "epoch": 4.138961524094135,
+      "grad_norm": 0.9728496074676514,
+      "learning_rate": 0.0002,
+      "loss": 1.3901,
+      "step": 5540
+    },
+    {
+      "epoch": 4.1464325737766154,
+      "grad_norm": 0.7590922117233276,
+      "learning_rate": 0.0002,
+      "loss": 1.3538,
+      "step": 5550
+    },
+    {
+      "epoch": 4.153903623459096,
+      "grad_norm": 0.7759010791778564,
+      "learning_rate": 0.0002,
+      "loss": 1.4368,
+      "step": 5560
+    },
+    {
+      "epoch": 4.161374673141577,
+      "grad_norm": 0.9057986736297607,
+      "learning_rate": 0.0002,
+      "loss": 1.3635,
+      "step": 5570
+    },
+    {
+      "epoch": 4.168845722824057,
+      "grad_norm": 0.8853937983512878,
+      "learning_rate": 0.0002,
+      "loss": 1.4152,
+      "step": 5580
+    },
+    {
+      "epoch": 4.176316772506537,
+      "grad_norm": 0.7070684432983398,
+      "learning_rate": 0.0002,
+      "loss": 1.3633,
+      "step": 5590
+    },
+    {
+      "epoch": 4.183787822189018,
+      "grad_norm": 0.7649410963058472,
+      "learning_rate": 0.0002,
+      "loss": 1.3218,
+      "step": 5600
+    },
+    {
+      "epoch": 4.191258871871498,
+      "grad_norm": 1.2048029899597168,
+      "learning_rate": 0.0002,
+      "loss": 1.3857,
+      "step": 5610
+    },
+    {
+      "epoch": 4.198729921553978,
+      "grad_norm": 0.7986605763435364,
+      "learning_rate": 0.0002,
+      "loss": 1.3629,
+      "step": 5620
+    },
+    {
+      "epoch": 4.206200971236458,
+      "grad_norm": 0.8151885867118835,
+      "learning_rate": 0.0002,
+      "loss": 1.3995,
+      "step": 5630
+    },
+    {
+      "epoch": 4.213672020918939,
+      "grad_norm": 0.7719064354896545,
+      "learning_rate": 0.0002,
+      "loss": 1.3782,
+      "step": 5640
+    },
+    {
+      "epoch": 4.2211430706014195,
+      "grad_norm": 0.8422448039054871,
+      "learning_rate": 0.0002,
+      "loss": 1.3852,
+      "step": 5650
+    },
+    {
+      "epoch": 4.2286141202839,
+      "grad_norm": 0.7017164826393127,
+      "learning_rate": 0.0002,
+      "loss": 1.3321,
+      "step": 5660
+    },
+    {
+      "epoch": 4.236085169966381,
+      "grad_norm": 0.8559677600860596,
+      "learning_rate": 0.0002,
+      "loss": 1.4105,
+      "step": 5670
+    },
+    {
+      "epoch": 4.243556219648861,
+      "grad_norm": 0.8216157555580139,
+      "learning_rate": 0.0002,
+      "loss": 1.3701,
+      "step": 5680
+    },
+    {
+      "epoch": 4.251027269331341,
+      "grad_norm": 0.7681755423545837,
+      "learning_rate": 0.0002,
+      "loss": 1.3565,
+      "step": 5690
+    },
+    {
+      "epoch": 4.258498319013821,
+      "grad_norm": 0.811665952205658,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 5700
+    },
+    {
+      "epoch": 4.265969368696302,
+      "grad_norm": 0.7242204546928406,
+      "learning_rate": 0.0002,
+      "loss": 1.4161,
+      "step": 5710
+    },
+    {
+      "epoch": 4.273440418378782,
+      "grad_norm": 0.7570181488990784,
+      "learning_rate": 0.0002,
+      "loss": 1.2958,
+      "step": 5720
+    },
+    {
+      "epoch": 4.280911468061262,
+      "grad_norm": 0.8951969146728516,
+      "learning_rate": 0.0002,
+      "loss": 1.4265,
+      "step": 5730
+    },
+    {
+      "epoch": 4.288382517743743,
+      "grad_norm": 0.7222902178764343,
+      "learning_rate": 0.0002,
+      "loss": 1.3895,
+      "step": 5740
+    },
+    {
+      "epoch": 4.2958535674262235,
+      "grad_norm": 0.8508469462394714,
+      "learning_rate": 0.0002,
+      "loss": 1.4155,
+      "step": 5750
+    },
+    {
+      "epoch": 4.303324617108704,
+      "grad_norm": 0.7215430736541748,
+      "learning_rate": 0.0002,
+      "loss": 1.365,
+      "step": 5760
+    },
+    {
+      "epoch": 4.310795666791184,
+      "grad_norm": 0.8774884939193726,
+      "learning_rate": 0.0002,
+      "loss": 1.4472,
+      "step": 5770
+    },
+    {
+      "epoch": 4.318266716473665,
+      "grad_norm": 0.8354552984237671,
+      "learning_rate": 0.0002,
+      "loss": 1.427,
+      "step": 5780
+    },
+    {
+      "epoch": 4.325737766156145,
+      "grad_norm": 0.6938814520835876,
+      "learning_rate": 0.0002,
+      "loss": 1.3222,
+      "step": 5790
+    },
+    {
+      "epoch": 4.333208815838625,
+      "grad_norm": 0.78675377368927,
+      "learning_rate": 0.0002,
+      "loss": 1.3589,
+      "step": 5800
+    },
+    {
+      "epoch": 4.340679865521106,
+      "grad_norm": 0.7147697806358337,
+      "learning_rate": 0.0002,
+      "loss": 1.3662,
+      "step": 5810
+    },
+    {
+      "epoch": 4.348150915203586,
+      "grad_norm": 0.7693623304367065,
+      "learning_rate": 0.0002,
+      "loss": 1.3597,
+      "step": 5820
+    },
+    {
+      "epoch": 4.355621964886066,
+      "grad_norm": 0.856517493724823,
+      "learning_rate": 0.0002,
+      "loss": 1.2944,
+      "step": 5830
+    },
+    {
+      "epoch": 4.3630930145685465,
+      "grad_norm": 0.7200973033905029,
+      "learning_rate": 0.0002,
+      "loss": 1.4307,
+      "step": 5840
+    },
+    {
+      "epoch": 4.3705640642510275,
+      "grad_norm": 0.743281364440918,
+      "learning_rate": 0.0002,
+      "loss": 1.442,
+      "step": 5850
+    },
+    {
+      "epoch": 4.378035113933508,
+      "grad_norm": 0.7627727389335632,
+      "learning_rate": 0.0002,
+      "loss": 1.3999,
+      "step": 5860
+    },
+    {
+      "epoch": 4.385506163615988,
+      "grad_norm": 0.7238836884498596,
+      "learning_rate": 0.0002,
+      "loss": 1.4082,
+      "step": 5870
+    },
+    {
+      "epoch": 4.392977213298469,
+      "grad_norm": 0.7253410816192627,
+      "learning_rate": 0.0002,
+      "loss": 1.4292,
+      "step": 5880
+    },
+    {
+      "epoch": 4.400448262980949,
+      "grad_norm": 0.8232238292694092,
+      "learning_rate": 0.0002,
+      "loss": 1.3774,
+      "step": 5890
+    },
+    {
+      "epoch": 4.407919312663429,
+      "grad_norm": 0.8778504729270935,
+      "learning_rate": 0.0002,
+      "loss": 1.3757,
+      "step": 5900
+    },
+    {
+      "epoch": 4.415390362345909,
+      "grad_norm": 0.7639474868774414,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 5910
+    },
+    {
+      "epoch": 4.42286141202839,
+      "grad_norm": 0.7666519284248352,
+      "learning_rate": 0.0002,
+      "loss": 1.3862,
+      "step": 5920
+    },
+    {
+      "epoch": 4.43033246171087,
+      "grad_norm": 0.867132842540741,
+      "learning_rate": 0.0002,
+      "loss": 1.4168,
+      "step": 5930
+    },
+    {
+      "epoch": 4.4378035113933505,
+      "grad_norm": 0.7571166753768921,
+      "learning_rate": 0.0002,
+      "loss": 1.4772,
+      "step": 5940
+    },
+    {
+      "epoch": 4.4452745610758315,
+      "grad_norm": 0.7911370992660522,
+      "learning_rate": 0.0002,
+      "loss": 1.4401,
+      "step": 5950
+    },
+    {
+      "epoch": 4.452745610758312,
+      "grad_norm": 0.8844250440597534,
+      "learning_rate": 0.0002,
+      "loss": 1.4516,
+      "step": 5960
+    },
+    {
+      "epoch": 4.460216660440792,
+      "grad_norm": 0.7336231470108032,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 5970
+    },
+    {
+      "epoch": 4.467687710123272,
+      "grad_norm": 0.8162738084793091,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 5980
+    },
+    {
+      "epoch": 4.475158759805753,
+      "grad_norm": 0.7413017153739929,
+      "learning_rate": 0.0002,
+      "loss": 1.393,
+      "step": 5990
+    },
+    {
+      "epoch": 4.482629809488233,
+      "grad_norm": 0.7215432524681091,
+      "learning_rate": 0.0002,
+      "loss": 1.3712,
+      "step": 6000
+    },
+    {
+      "epoch": 4.490100859170713,
+      "grad_norm": 0.8943389058113098,
+      "learning_rate": 0.0002,
+      "loss": 1.3521,
+      "step": 6010
+    },
+    {
+      "epoch": 4.497571908853194,
+      "grad_norm": 0.7850823998451233,
+      "learning_rate": 0.0002,
+      "loss": 1.4172,
+      "step": 6020
+    },
+    {
+      "epoch": 4.505042958535674,
+      "grad_norm": 0.8117504119873047,
+      "learning_rate": 0.0002,
+      "loss": 1.3582,
+      "step": 6030
+    },
+    {
+      "epoch": 4.5125140082181545,
+      "grad_norm": 0.8381605744361877,
+      "learning_rate": 0.0002,
+      "loss": 1.4272,
+      "step": 6040
+    },
+    {
+      "epoch": 4.519985057900635,
+      "grad_norm": 0.7964059710502625,
+      "learning_rate": 0.0002,
+      "loss": 1.3829,
+      "step": 6050
+    },
+    {
+      "epoch": 4.527456107583116,
+      "grad_norm": 0.7935128211975098,
+      "learning_rate": 0.0002,
+      "loss": 1.3555,
+      "step": 6060
+    },
+    {
+      "epoch": 4.534927157265596,
+      "grad_norm": 0.8725124597549438,
+      "learning_rate": 0.0002,
+      "loss": 1.3994,
+      "step": 6070
+    },
+    {
+      "epoch": 4.542398206948076,
+      "grad_norm": 0.880325198173523,
+      "learning_rate": 0.0002,
+      "loss": 1.3923,
+      "step": 6080
+    },
+    {
+      "epoch": 4.549869256630557,
+      "grad_norm": 0.7220637202262878,
+      "learning_rate": 0.0002,
+      "loss": 1.4459,
+      "step": 6090
+    },
+    {
+      "epoch": 4.557340306313037,
+      "grad_norm": 0.6908547878265381,
+      "learning_rate": 0.0002,
+      "loss": 1.3281,
+      "step": 6100
+    },
+    {
+      "epoch": 4.564811355995517,
+      "grad_norm": 0.797931969165802,
+      "learning_rate": 0.0002,
+      "loss": 1.437,
+      "step": 6110
+    },
+    {
+      "epoch": 4.572282405677997,
+      "grad_norm": 0.7056134343147278,
+      "learning_rate": 0.0002,
+      "loss": 1.4023,
+      "step": 6120
+    },
+    {
+      "epoch": 4.579753455360478,
+      "grad_norm": 0.7850478887557983,
+      "learning_rate": 0.0002,
+      "loss": 1.3814,
+      "step": 6130
+    },
+    {
+      "epoch": 4.5872245050429585,
+      "grad_norm": 0.8112621307373047,
+      "learning_rate": 0.0002,
+      "loss": 1.3579,
+      "step": 6140
+    },
+    {
+      "epoch": 4.594695554725439,
+      "grad_norm": 0.7040849328041077,
+      "learning_rate": 0.0002,
+      "loss": 1.3523,
+      "step": 6150
+    },
+    {
+      "epoch": 4.60216660440792,
+      "grad_norm": 0.7214553952217102,
+      "learning_rate": 0.0002,
+      "loss": 1.3526,
+      "step": 6160
+    },
+    {
+      "epoch": 4.6096376540904,
+      "grad_norm": 0.8616511821746826,
+      "learning_rate": 0.0002,
+      "loss": 1.3932,
+      "step": 6170
+    },
+    {
+      "epoch": 4.61710870377288,
+      "grad_norm": 0.8374658226966858,
+      "learning_rate": 0.0002,
+      "loss": 1.4622,
+      "step": 6180
+    },
+    {
+      "epoch": 4.62457975345536,
+      "grad_norm": 0.6761606931686401,
+      "learning_rate": 0.0002,
+      "loss": 1.3703,
+      "step": 6190
+    },
+    {
+      "epoch": 4.632050803137841,
+      "grad_norm": 0.768028199672699,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 6200
+    },
+    {
+      "epoch": 4.639521852820321,
+      "grad_norm": 0.9372717142105103,
+      "learning_rate": 0.0002,
+      "loss": 1.3772,
+      "step": 6210
+    },
+    {
+      "epoch": 4.646992902502801,
+      "grad_norm": 0.7906546592712402,
+      "learning_rate": 0.0002,
+      "loss": 1.4098,
+      "step": 6220
+    },
+    {
+      "epoch": 4.654463952185282,
+      "grad_norm": 0.7376723289489746,
+      "learning_rate": 0.0002,
+      "loss": 1.3962,
+      "step": 6230
+    },
+    {
+      "epoch": 4.6619350018677626,
+      "grad_norm": 0.8972630500793457,
+      "learning_rate": 0.0002,
+      "loss": 1.4529,
+      "step": 6240
+    },
+    {
+      "epoch": 4.669406051550243,
+      "grad_norm": 0.8261756300926208,
+      "learning_rate": 0.0002,
+      "loss": 1.4668,
+      "step": 6250
+    },
+    {
+      "epoch": 4.676877101232723,
+      "grad_norm": 0.7512393593788147,
+      "learning_rate": 0.0002,
+      "loss": 1.3267,
+      "step": 6260
+    },
+    {
+      "epoch": 4.684348150915204,
+      "grad_norm": 0.7132362127304077,
+      "learning_rate": 0.0002,
+      "loss": 1.4278,
+      "step": 6270
+    },
+    {
+      "epoch": 4.691819200597684,
+      "grad_norm": 0.7690575122833252,
+      "learning_rate": 0.0002,
+      "loss": 1.4299,
+      "step": 6280
+    },
+    {
+      "epoch": 4.699290250280164,
+      "grad_norm": 0.9886258840560913,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 6290
+    },
+    {
+      "epoch": 4.706761299962645,
+      "grad_norm": 0.9502435922622681,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 6300
+    },
+    {
+      "epoch": 4.714232349645125,
+      "grad_norm": 0.702255129814148,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 6310
+    },
+    {
+      "epoch": 4.721703399327605,
+      "grad_norm": 0.7713103890419006,
+      "learning_rate": 0.0002,
+      "loss": 1.4447,
+      "step": 6320
+    },
+    {
+      "epoch": 4.7291744490100855,
+      "grad_norm": 0.7778580784797668,
+      "learning_rate": 0.0002,
+      "loss": 1.4392,
+      "step": 6330
+    },
+    {
+      "epoch": 4.736645498692567,
+      "grad_norm": 0.7275111079216003,
+      "learning_rate": 0.0002,
+      "loss": 1.4169,
+      "step": 6340
+    },
+    {
+      "epoch": 4.744116548375047,
+      "grad_norm": 0.7728744149208069,
+      "learning_rate": 0.0002,
+      "loss": 1.4429,
+      "step": 6350
+    },
+    {
+      "epoch": 4.751587598057527,
+      "grad_norm": 0.9724260568618774,
+      "learning_rate": 0.0002,
+      "loss": 1.3756,
+      "step": 6360
+    },
+    {
+      "epoch": 4.759058647740007,
+      "grad_norm": 0.7505622506141663,
+      "learning_rate": 0.0002,
+      "loss": 1.3358,
+      "step": 6370
+    },
+    {
+      "epoch": 4.766529697422488,
+      "grad_norm": 0.7994682788848877,
+      "learning_rate": 0.0002,
+      "loss": 1.379,
+      "step": 6380
+    },
+    {
+      "epoch": 4.774000747104968,
+      "grad_norm": 0.8432038426399231,
+      "learning_rate": 0.0002,
+      "loss": 1.4275,
+      "step": 6390
+    },
+    {
+      "epoch": 4.781471796787448,
+      "grad_norm": 0.7436022758483887,
+      "learning_rate": 0.0002,
+      "loss": 1.4606,
+      "step": 6400
+    },
+    {
+      "epoch": 4.788942846469929,
+      "grad_norm": 0.7709194421768188,
+      "learning_rate": 0.0002,
+      "loss": 1.3461,
+      "step": 6410
+    },
+    {
+      "epoch": 4.796413896152409,
+      "grad_norm": 0.8798436522483826,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 6420
+    },
+    {
+      "epoch": 4.80388494583489,
+      "grad_norm": 0.790189266204834,
+      "learning_rate": 0.0002,
+      "loss": 1.3761,
+      "step": 6430
+    },
+    {
+      "epoch": 4.811355995517371,
+      "grad_norm": 0.6824303865432739,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 6440
+    },
+    {
+      "epoch": 4.818827045199851,
+      "grad_norm": 0.7501044869422913,
+      "learning_rate": 0.0002,
+      "loss": 1.3877,
+      "step": 6450
+    },
+    {
+      "epoch": 4.826298094882331,
+      "grad_norm": 0.8840398192405701,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 6460
+    },
+    {
+      "epoch": 4.833769144564811,
+      "grad_norm": 0.7812688946723938,
+      "learning_rate": 0.0002,
+      "loss": 1.4412,
+      "step": 6470
+    },
+    {
+      "epoch": 4.841240194247292,
+      "grad_norm": 0.7429926991462708,
+      "learning_rate": 0.0002,
+      "loss": 1.4299,
+      "step": 6480
+    },
+    {
+      "epoch": 4.848711243929772,
+      "grad_norm": 0.7778021693229675,
+      "learning_rate": 0.0002,
+      "loss": 1.5062,
+      "step": 6490
+    },
+    {
+      "epoch": 4.856182293612252,
+      "grad_norm": 0.8270702362060547,
+      "learning_rate": 0.0002,
+      "loss": 1.4589,
+      "step": 6500
+    },
+    {
+      "epoch": 4.863653343294732,
+      "grad_norm": 0.6960513591766357,
+      "learning_rate": 0.0002,
+      "loss": 1.4091,
+      "step": 6510
+    },
+    {
+      "epoch": 4.8711243929772134,
+      "grad_norm": 0.7728942632675171,
+      "learning_rate": 0.0002,
+      "loss": 1.376,
+      "step": 6520
+    },
+    {
+      "epoch": 4.878595442659694,
+      "grad_norm": 0.7377303838729858,
+      "learning_rate": 0.0002,
+      "loss": 1.4852,
+      "step": 6530
+    },
+    {
+      "epoch": 4.886066492342174,
+      "grad_norm": 0.7257253527641296,
+      "learning_rate": 0.0002,
+      "loss": 1.3846,
+      "step": 6540
+    },
+    {
+      "epoch": 4.893537542024655,
+      "grad_norm": 0.7875821590423584,
+      "learning_rate": 0.0002,
+      "loss": 1.4166,
+      "step": 6550
+    },
+    {
+      "epoch": 4.901008591707135,
+      "grad_norm": 0.8346304297447205,
+      "learning_rate": 0.0002,
+      "loss": 1.357,
+      "step": 6560
+    },
+    {
+      "epoch": 4.908479641389615,
+      "grad_norm": 0.7710739374160767,
+      "learning_rate": 0.0002,
+      "loss": 1.4522,
+      "step": 6570
+    },
+    {
+      "epoch": 4.915950691072096,
+      "grad_norm": 0.7015138268470764,
+      "learning_rate": 0.0002,
+      "loss": 1.4465,
+      "step": 6580
+    },
+    {
+      "epoch": 4.923421740754576,
+      "grad_norm": 0.8707432150840759,
+      "learning_rate": 0.0002,
+      "loss": 1.435,
+      "step": 6590
+    },
+    {
+      "epoch": 4.930892790437056,
+      "grad_norm": 0.786601185798645,
+      "learning_rate": 0.0002,
+      "loss": 1.2968,
+      "step": 6600
+    },
+    {
+      "epoch": 4.938363840119536,
+      "grad_norm": 0.978519082069397,
+      "learning_rate": 0.0002,
+      "loss": 1.4385,
+      "step": 6610
+    },
+    {
+      "epoch": 4.9458348898020175,
+      "grad_norm": 0.8102927207946777,
+      "learning_rate": 0.0002,
+      "loss": 1.3997,
+      "step": 6620
+    },
+    {
+      "epoch": 4.953305939484498,
+      "grad_norm": 0.7628704309463501,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 6630
+    },
+    {
+      "epoch": 4.960776989166978,
+      "grad_norm": 0.8053455352783203,
+      "learning_rate": 0.0002,
+      "loss": 1.3774,
+      "step": 6640
+    },
+    {
+      "epoch": 4.968248038849458,
+      "grad_norm": 0.8680412173271179,
+      "learning_rate": 0.0002,
+      "loss": 1.5092,
+      "step": 6650
+    },
+    {
+      "epoch": 4.975719088531939,
+      "grad_norm": 0.7415758371353149,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 6660
+    },
+    {
+      "epoch": 4.983190138214419,
+      "grad_norm": 0.7730312347412109,
+      "learning_rate": 0.0002,
+      "loss": 1.3793,
+      "step": 6670
+    },
+    {
+      "epoch": 4.990661187896899,
+      "grad_norm": 0.7924041152000427,
+      "learning_rate": 0.0002,
+      "loss": 1.4863,
+      "step": 6680
+    },
+    {
+      "epoch": 4.99813223757938,
+      "grad_norm": 0.8677893877029419,
+      "learning_rate": 0.0002,
+      "loss": 1.4137,
+      "step": 6690
+    },
+    {
+      "epoch": 4.999626447515876,
+      "eval_loss": 1.9444633722305298,
+      "eval_runtime": 39.3488,
+      "eval_samples_per_second": 13.088,
+      "eval_steps_per_second": 1.652,
+      "step": 6692
+    },
+    {
+      "epoch": 5.00560328726186,
+      "grad_norm": 0.7102245092391968,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 6700
+    },
+    {
+      "epoch": 5.0130743369443405,
+      "grad_norm": 1.0425463914871216,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 6710
+    },
+    {
+      "epoch": 5.0205453866268215,
+      "grad_norm": 0.9320756793022156,
+      "learning_rate": 0.0002,
+      "loss": 1.181,
+      "step": 6720
+    },
+    {
+      "epoch": 5.028016436309302,
+      "grad_norm": 0.8797217607498169,
+      "learning_rate": 0.0002,
+      "loss": 1.1786,
+      "step": 6730
+    },
+    {
+      "epoch": 5.035487485991782,
+      "grad_norm": 2.135707139968872,
+      "learning_rate": 0.0002,
+      "loss": 1.2097,
+      "step": 6740
+    },
+    {
+      "epoch": 5.042958535674262,
+      "grad_norm": 0.8747734427452087,
+      "learning_rate": 0.0002,
+      "loss": 1.1761,
+      "step": 6750
+    },
+    {
+      "epoch": 5.050429585356743,
+      "grad_norm": 0.9981076717376709,
+      "learning_rate": 0.0002,
+      "loss": 1.1675,
+      "step": 6760
+    },
+    {
+      "epoch": 5.057900635039223,
+      "grad_norm": 0.985078752040863,
+      "learning_rate": 0.0002,
+      "loss": 1.1976,
+      "step": 6770
+    },
+    {
+      "epoch": 5.065371684721703,
+      "grad_norm": 1.0974019765853882,
+      "learning_rate": 0.0002,
+      "loss": 1.2688,
+      "step": 6780
+    },
+    {
+      "epoch": 5.072842734404184,
+      "grad_norm": 0.9823219180107117,
+      "learning_rate": 0.0002,
+      "loss": 1.1982,
+      "step": 6790
+    },
+    {
+      "epoch": 5.080313784086664,
+      "grad_norm": 1.122605562210083,
+      "learning_rate": 0.0002,
+      "loss": 1.2586,
+      "step": 6800
+    },
+    {
+      "epoch": 5.0877848337691445,
+      "grad_norm": 0.8556802272796631,
+      "learning_rate": 0.0002,
+      "loss": 1.2069,
+      "step": 6810
+    },
+    {
+      "epoch": 5.095255883451625,
+      "grad_norm": 1.1699262857437134,
+      "learning_rate": 0.0002,
+      "loss": 1.1908,
+      "step": 6820
+    },
+    {
+      "epoch": 5.102726933134106,
+      "grad_norm": 1.0440590381622314,
+      "learning_rate": 0.0002,
+      "loss": 1.1869,
+      "step": 6830
+    },
+    {
+      "epoch": 5.110197982816586,
+      "grad_norm": 1.0445300340652466,
+      "learning_rate": 0.0002,
+      "loss": 1.1655,
+      "step": 6840
+    },
+    {
+      "epoch": 5.117669032499066,
+      "grad_norm": 0.8289563059806824,
+      "learning_rate": 0.0002,
+      "loss": 1.2392,
+      "step": 6850
+    },
+    {
+      "epoch": 5.125140082181547,
+      "grad_norm": 1.1051193475723267,
+      "learning_rate": 0.0002,
+      "loss": 1.1687,
+      "step": 6860
+    },
+    {
+      "epoch": 5.132611131864027,
+      "grad_norm": 0.9345614910125732,
+      "learning_rate": 0.0002,
+      "loss": 1.2737,
+      "step": 6870
+    },
+    {
+      "epoch": 5.140082181546507,
+      "grad_norm": 1.1222996711730957,
+      "learning_rate": 0.0002,
+      "loss": 1.3021,
+      "step": 6880
+    },
+    {
+      "epoch": 5.147553231228987,
+      "grad_norm": 0.9405338764190674,
+      "learning_rate": 0.0002,
+      "loss": 1.2408,
+      "step": 6890
+    },
+    {
+      "epoch": 5.155024280911468,
+      "grad_norm": 1.0935171842575073,
+      "learning_rate": 0.0002,
+      "loss": 1.2367,
+      "step": 6900
+    },
+    {
+      "epoch": 5.1624953305939485,
+      "grad_norm": 1.0438612699508667,
+      "learning_rate": 0.0002,
+      "loss": 1.2458,
+      "step": 6910
+    },
+    {
+      "epoch": 5.169966380276429,
+      "grad_norm": 1.1189004182815552,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 6920
+    },
+    {
+      "epoch": 5.17743742995891,
+      "grad_norm": 1.0533215999603271,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 6930
+    },
+    {
+      "epoch": 5.18490847964139,
+      "grad_norm": 0.9779648780822754,
+      "learning_rate": 0.0002,
+      "loss": 1.2974,
+      "step": 6940
+    },
+    {
+      "epoch": 5.19237952932387,
+      "grad_norm": 0.8920868635177612,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 6950
+    },
+    {
+      "epoch": 5.19985057900635,
+      "grad_norm": 0.8374548554420471,
+      "learning_rate": 0.0002,
+      "loss": 1.283,
+      "step": 6960
+    },
+    {
+      "epoch": 5.207321628688831,
+      "grad_norm": 1.0490682125091553,
+      "learning_rate": 0.0002,
+      "loss": 1.2775,
+      "step": 6970
+    },
+    {
+      "epoch": 5.214792678371311,
+      "grad_norm": 0.9658287167549133,
+      "learning_rate": 0.0002,
+      "loss": 1.1826,
+      "step": 6980
+    },
+    {
+      "epoch": 5.222263728053791,
+      "grad_norm": 0.9652056097984314,
+      "learning_rate": 0.0002,
+      "loss": 1.2647,
+      "step": 6990
+    },
+    {
+      "epoch": 5.229734777736272,
+      "grad_norm": 0.9141794443130493,
+      "learning_rate": 0.0002,
+      "loss": 1.3023,
+      "step": 7000
+    },
+    {
+      "epoch": 5.2372058274187525,
+      "grad_norm": 0.9831376671791077,
+      "learning_rate": 0.0002,
+      "loss": 1.2456,
+      "step": 7010
+    },
+    {
+      "epoch": 5.244676877101233,
+      "grad_norm": 1.0198718309402466,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 7020
+    },
+    {
+      "epoch": 5.252147926783713,
+      "grad_norm": 0.9647888541221619,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 7030
+    },
+    {
+      "epoch": 5.259618976466194,
+      "grad_norm": 1.3941649198532104,
+      "learning_rate": 0.0002,
+      "loss": 1.2106,
+      "step": 7040
+    },
+    {
+      "epoch": 5.267090026148674,
+      "grad_norm": 1.0305466651916504,
+      "learning_rate": 0.0002,
+      "loss": 1.2885,
+      "step": 7050
+    },
+    {
+      "epoch": 5.274561075831154,
+      "grad_norm": 0.9577859044075012,
+      "learning_rate": 0.0002,
+      "loss": 1.2362,
+      "step": 7060
+    },
+    {
+      "epoch": 5.282032125513634,
+      "grad_norm": 1.149092197418213,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 7070
+    },
+    {
+      "epoch": 5.289503175196115,
+      "grad_norm": 1.2582733631134033,
+      "learning_rate": 0.0002,
+      "loss": 1.2986,
+      "step": 7080
+    },
+    {
+      "epoch": 5.296974224878595,
+      "grad_norm": 1.1777442693710327,
+      "learning_rate": 0.0002,
+      "loss": 1.2307,
+      "step": 7090
+    },
+    {
+      "epoch": 5.3044452745610755,
+      "grad_norm": 1.0076404809951782,
+      "learning_rate": 0.0002,
+      "loss": 1.24,
+      "step": 7100
+    },
+    {
+      "epoch": 5.3119163242435565,
+      "grad_norm": 0.9037365913391113,
+      "learning_rate": 0.0002,
+      "loss": 1.1407,
+      "step": 7110
+    },
+    {
+      "epoch": 5.319387373926037,
+      "grad_norm": 0.9428724646568298,
+      "learning_rate": 0.0002,
+      "loss": 1.238,
+      "step": 7120
+    },
+    {
+      "epoch": 5.326858423608517,
+      "grad_norm": 0.9935154318809509,
+      "learning_rate": 0.0002,
+      "loss": 1.2571,
+      "step": 7130
+    },
+    {
+      "epoch": 5.334329473290998,
+      "grad_norm": 1.087500810623169,
+      "learning_rate": 0.0002,
+      "loss": 1.2833,
+      "step": 7140
+    },
+    {
+      "epoch": 5.341800522973478,
+      "grad_norm": 0.8543072938919067,
+      "learning_rate": 0.0002,
+      "loss": 1.2304,
+      "step": 7150
+    },
+    {
+      "epoch": 5.349271572655958,
+      "grad_norm": 0.9323700070381165,
+      "learning_rate": 0.0002,
+      "loss": 1.2755,
+      "step": 7160
+    },
+    {
+      "epoch": 5.356742622338438,
+      "grad_norm": 1.0037827491760254,
+      "learning_rate": 0.0002,
+      "loss": 1.2769,
+      "step": 7170
+    },
+    {
+      "epoch": 5.364213672020919,
+      "grad_norm": 0.8746469616889954,
+      "learning_rate": 0.0002,
+      "loss": 1.3204,
+      "step": 7180
+    },
+    {
+      "epoch": 5.371684721703399,
+      "grad_norm": 0.9516328573226929,
+      "learning_rate": 0.0002,
+      "loss": 1.2759,
+      "step": 7190
+    },
+    {
+      "epoch": 5.3791557713858795,
+      "grad_norm": 0.9395177364349365,
+      "learning_rate": 0.0002,
+      "loss": 1.2428,
+      "step": 7200
+    },
+    {
+      "epoch": 5.38662682106836,
+      "grad_norm": 1.000369906425476,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 7210
+    },
+    {
+      "epoch": 5.394097870750841,
+      "grad_norm": 1.0845502614974976,
+      "learning_rate": 0.0002,
+      "loss": 1.2337,
+      "step": 7220
+    },
+    {
+      "epoch": 5.401568920433321,
+      "grad_norm": 0.8975145220756531,
+      "learning_rate": 0.0002,
+      "loss": 1.2776,
+      "step": 7230
+    },
+    {
+      "epoch": 5.409039970115801,
+      "grad_norm": 1.040077805519104,
+      "learning_rate": 0.0002,
+      "loss": 1.2306,
+      "step": 7240
+    },
+    {
+      "epoch": 5.416511019798282,
+      "grad_norm": 1.0729942321777344,
+      "learning_rate": 0.0002,
+      "loss": 1.2277,
+      "step": 7250
+    },
+    {
+      "epoch": 5.423982069480762,
+      "grad_norm": 0.8322232961654663,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 7260
+    },
+    {
+      "epoch": 5.431453119163242,
+      "grad_norm": 1.0654641389846802,
+      "learning_rate": 0.0002,
+      "loss": 1.3036,
+      "step": 7270
+    },
+    {
+      "epoch": 5.438924168845723,
+      "grad_norm": 1.0445852279663086,
+      "learning_rate": 0.0002,
+      "loss": 1.268,
+      "step": 7280
+    },
+    {
+      "epoch": 5.446395218528203,
+      "grad_norm": 1.0762956142425537,
+      "learning_rate": 0.0002,
+      "loss": 1.2743,
+      "step": 7290
+    },
+    {
+      "epoch": 5.4538662682106835,
+      "grad_norm": 0.9721953868865967,
+      "learning_rate": 0.0002,
+      "loss": 1.2887,
+      "step": 7300
+    },
+    {
+      "epoch": 5.461337317893164,
+      "grad_norm": 0.9238539338111877,
+      "learning_rate": 0.0002,
+      "loss": 1.2833,
+      "step": 7310
+    },
+    {
+      "epoch": 5.468808367575645,
+      "grad_norm": 0.9912874102592468,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 7320
+    },
+    {
+      "epoch": 5.476279417258125,
+      "grad_norm": 1.0727077722549438,
+      "learning_rate": 0.0002,
+      "loss": 1.2557,
+      "step": 7330
+    },
+    {
+      "epoch": 5.483750466940605,
+      "grad_norm": 0.8633865118026733,
+      "learning_rate": 0.0002,
+      "loss": 1.3471,
+      "step": 7340
+    },
+    {
+      "epoch": 5.491221516623085,
+      "grad_norm": 0.9396262764930725,
+      "learning_rate": 0.0002,
+      "loss": 1.3155,
+      "step": 7350
+    },
+    {
+      "epoch": 5.498692566305566,
+      "grad_norm": 1.0253715515136719,
+      "learning_rate": 0.0002,
+      "loss": 1.3146,
+      "step": 7360
+    },
+    {
+      "epoch": 5.506163615988046,
+      "grad_norm": 1.006047010421753,
+      "learning_rate": 0.0002,
+      "loss": 1.3156,
+      "step": 7370
+    },
+    {
+      "epoch": 5.513634665670526,
+      "grad_norm": 0.9781233072280884,
+      "learning_rate": 0.0002,
+      "loss": 1.3107,
+      "step": 7380
+    },
+    {
+      "epoch": 5.521105715353007,
+      "grad_norm": 0.9945126175880432,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 7390
+    },
+    {
+      "epoch": 5.528576765035488,
+      "grad_norm": 0.9081175327301025,
+      "learning_rate": 0.0002,
+      "loss": 1.1936,
+      "step": 7400
+    },
+    {
+      "epoch": 5.536047814717968,
+      "grad_norm": 1.2215938568115234,
+      "learning_rate": 0.0002,
+      "loss": 1.2651,
+      "step": 7410
+    },
+    {
+      "epoch": 5.543518864400449,
+      "grad_norm": 1.0724077224731445,
+      "learning_rate": 0.0002,
+      "loss": 1.2484,
+      "step": 7420
+    },
+    {
+      "epoch": 5.550989914082929,
+      "grad_norm": 1.106955885887146,
+      "learning_rate": 0.0002,
+      "loss": 1.3083,
+      "step": 7430
+    },
+    {
+      "epoch": 5.558460963765409,
+      "grad_norm": 1.0657650232315063,
+      "learning_rate": 0.0002,
+      "loss": 1.2125,
+      "step": 7440
+    },
+    {
+      "epoch": 5.565932013447889,
+      "grad_norm": 0.9725455641746521,
+      "learning_rate": 0.0002,
+      "loss": 1.2576,
+      "step": 7450
+    },
+    {
+      "epoch": 5.57340306313037,
+      "grad_norm": 0.8604224324226379,
+      "learning_rate": 0.0002,
+      "loss": 1.3297,
+      "step": 7460
+    },
+    {
+      "epoch": 5.58087411281285,
+      "grad_norm": 0.9913371205329895,
+      "learning_rate": 0.0002,
+      "loss": 1.3084,
+      "step": 7470
+    },
+    {
+      "epoch": 5.58834516249533,
+      "grad_norm": 1.012073040008545,
+      "learning_rate": 0.0002,
+      "loss": 1.3371,
+      "step": 7480
+    },
+    {
+      "epoch": 5.5958162121778106,
+      "grad_norm": 1.1003159284591675,
+      "learning_rate": 0.0002,
+      "loss": 1.2526,
+      "step": 7490
+    },
+    {
+      "epoch": 5.603287261860292,
+      "grad_norm": 0.9104593992233276,
+      "learning_rate": 0.0002,
+      "loss": 1.2577,
+      "step": 7500
+    },
+    {
+      "epoch": 5.610758311542772,
+      "grad_norm": 0.9480831623077393,
+      "learning_rate": 0.0002,
+      "loss": 1.2578,
+      "step": 7510
+    },
+    {
+      "epoch": 5.618229361225252,
+      "grad_norm": 1.0826456546783447,
+      "learning_rate": 0.0002,
+      "loss": 1.3056,
+      "step": 7520
+    },
+    {
+      "epoch": 5.625700410907733,
+      "grad_norm": 0.8286259174346924,
+      "learning_rate": 0.0002,
+      "loss": 1.2931,
+      "step": 7530
+    },
+    {
+      "epoch": 5.633171460590213,
+      "grad_norm": 0.9145061373710632,
+      "learning_rate": 0.0002,
+      "loss": 1.2918,
+      "step": 7540
+    },
+    {
+      "epoch": 5.640642510272693,
+      "grad_norm": 0.9363601803779602,
+      "learning_rate": 0.0002,
+      "loss": 1.1736,
+      "step": 7550
+    },
+    {
+      "epoch": 5.648113559955174,
+      "grad_norm": 0.9553244709968567,
+      "learning_rate": 0.0002,
+      "loss": 1.2265,
+      "step": 7560
+    },
+    {
+      "epoch": 5.655584609637654,
+      "grad_norm": 1.0343557596206665,
+      "learning_rate": 0.0002,
+      "loss": 1.2356,
+      "step": 7570
+    },
+    {
+      "epoch": 5.663055659320134,
+      "grad_norm": 0.8734238743782043,
+      "learning_rate": 0.0002,
+      "loss": 1.3171,
+      "step": 7580
+    },
+    {
+      "epoch": 5.670526709002615,
+      "grad_norm": 1.0230586528778076,
+      "learning_rate": 0.0002,
+      "loss": 1.2785,
+      "step": 7590
+    },
+    {
+      "epoch": 5.677997758685096,
+      "grad_norm": 1.0063409805297852,
+      "learning_rate": 0.0002,
+      "loss": 1.2936,
+      "step": 7600
+    },
+    {
+      "epoch": 5.685468808367576,
+      "grad_norm": 1.0104626417160034,
+      "learning_rate": 0.0002,
+      "loss": 1.2396,
+      "step": 7610
+    },
+    {
+      "epoch": 5.692939858050056,
+      "grad_norm": 0.9528168439865112,
+      "learning_rate": 0.0002,
+      "loss": 1.2581,
+      "step": 7620
+    },
+    {
+      "epoch": 5.700410907732536,
+      "grad_norm": 0.9799878597259521,
+      "learning_rate": 0.0002,
+      "loss": 1.3116,
+      "step": 7630
+    },
+    {
+      "epoch": 5.707881957415017,
+      "grad_norm": 0.969351589679718,
+      "learning_rate": 0.0002,
+      "loss": 1.2632,
+      "step": 7640
+    },
+    {
+      "epoch": 5.715353007097497,
+      "grad_norm": 1.3037652969360352,
+      "learning_rate": 0.0002,
+      "loss": 1.3055,
+      "step": 7650
+    },
+    {
+      "epoch": 5.722824056779977,
+      "grad_norm": 1.0640486478805542,
+      "learning_rate": 0.0002,
+      "loss": 1.3126,
+      "step": 7660
+    },
+    {
+      "epoch": 5.730295106462458,
+      "grad_norm": 1.0416420698165894,
+      "learning_rate": 0.0002,
+      "loss": 1.3325,
+      "step": 7670
+    },
+    {
+      "epoch": 5.7377661561449385,
+      "grad_norm": 0.8893619775772095,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 7680
+    },
+    {
+      "epoch": 5.745237205827419,
+      "grad_norm": 0.8512844443321228,
+      "learning_rate": 0.0002,
+      "loss": 1.319,
+      "step": 7690
+    },
+    {
+      "epoch": 5.7527082555099,
+      "grad_norm": 0.9955748319625854,
+      "learning_rate": 0.0002,
+      "loss": 1.3328,
+      "step": 7700
+    },
+    {
+      "epoch": 5.76017930519238,
+      "grad_norm": 1.0409910678863525,
+      "learning_rate": 0.0002,
+      "loss": 1.294,
+      "step": 7710
+    },
+    {
+      "epoch": 5.76765035487486,
+      "grad_norm": 1.010097861289978,
+      "learning_rate": 0.0002,
+      "loss": 1.3518,
+      "step": 7720
+    },
+    {
+      "epoch": 5.77512140455734,
+      "grad_norm": 0.8974892497062683,
+      "learning_rate": 0.0002,
+      "loss": 1.2106,
+      "step": 7730
+    },
+    {
+      "epoch": 5.782592454239821,
+      "grad_norm": 0.972835123538971,
+      "learning_rate": 0.0002,
+      "loss": 1.2743,
+      "step": 7740
+    },
+    {
+      "epoch": 5.790063503922301,
+      "grad_norm": 0.9607440829277039,
+      "learning_rate": 0.0002,
+      "loss": 1.3549,
+      "step": 7750
+    },
+    {
+      "epoch": 5.797534553604781,
+      "grad_norm": 0.9426500797271729,
+      "learning_rate": 0.0002,
+      "loss": 1.29,
+      "step": 7760
+    },
+    {
+      "epoch": 5.8050056032872615,
+      "grad_norm": 0.8745320439338684,
+      "learning_rate": 0.0002,
+      "loss": 1.274,
+      "step": 7770
+    },
+    {
+      "epoch": 5.8124766529697425,
+      "grad_norm": 1.0117204189300537,
+      "learning_rate": 0.0002,
+      "loss": 1.3009,
+      "step": 7780
+    },
+    {
+      "epoch": 5.819947702652223,
+      "grad_norm": 1.0387755632400513,
+      "learning_rate": 0.0002,
+      "loss": 1.3135,
+      "step": 7790
+    },
+    {
+      "epoch": 5.827418752334703,
+      "grad_norm": 1.0709784030914307,
+      "learning_rate": 0.0002,
+      "loss": 1.2709,
+      "step": 7800
+    },
+    {
+      "epoch": 5.834889802017184,
+      "grad_norm": 0.9512667655944824,
+      "learning_rate": 0.0002,
+      "loss": 1.225,
+      "step": 7810
+    },
+    {
+      "epoch": 5.842360851699664,
+      "grad_norm": 1.021094560623169,
+      "learning_rate": 0.0002,
+      "loss": 1.3284,
+      "step": 7820
+    },
+    {
+      "epoch": 5.849831901382144,
+      "grad_norm": 1.117491364479065,
+      "learning_rate": 0.0002,
+      "loss": 1.2794,
+      "step": 7830
+    },
+    {
+      "epoch": 5.857302951064625,
+      "grad_norm": 0.9252554178237915,
+      "learning_rate": 0.0002,
+      "loss": 1.3646,
+      "step": 7840
+    },
+    {
+      "epoch": 5.864774000747105,
+      "grad_norm": 1.1416207551956177,
+      "learning_rate": 0.0002,
+      "loss": 1.2976,
+      "step": 7850
+    },
+    {
+      "epoch": 5.872245050429585,
+      "grad_norm": 1.1219907999038696,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 7860
+    },
+    {
+      "epoch": 5.8797161001120655,
+      "grad_norm": 0.8300467729568481,
+      "learning_rate": 0.0002,
+      "loss": 1.2334,
+      "step": 7870
+    },
+    {
+      "epoch": 5.8871871497945465,
+      "grad_norm": 1.00551438331604,
+      "learning_rate": 0.0002,
+      "loss": 1.3132,
+      "step": 7880
+    },
+    {
+      "epoch": 5.894658199477027,
+      "grad_norm": 0.8981153964996338,
+      "learning_rate": 0.0002,
+      "loss": 1.2609,
+      "step": 7890
+    },
+    {
+      "epoch": 5.902129249159507,
+      "grad_norm": 1.0247976779937744,
+      "learning_rate": 0.0002,
+      "loss": 1.2817,
+      "step": 7900
+    },
+    {
+      "epoch": 5.909600298841987,
+      "grad_norm": 1.0820319652557373,
+      "learning_rate": 0.0002,
+      "loss": 1.2866,
+      "step": 7910
+    },
+    {
+      "epoch": 5.917071348524468,
+      "grad_norm": 0.952675461769104,
+      "learning_rate": 0.0002,
+      "loss": 1.2941,
+      "step": 7920
+    },
+    {
+      "epoch": 5.924542398206948,
+      "grad_norm": 0.8666740655899048,
+      "learning_rate": 0.0002,
+      "loss": 1.307,
+      "step": 7930
+    },
+    {
+      "epoch": 5.932013447889428,
+      "grad_norm": 0.8640421032905579,
+      "learning_rate": 0.0002,
+      "loss": 1.2752,
+      "step": 7940
+    },
+    {
+      "epoch": 5.939484497571909,
+      "grad_norm": 1.2343276739120483,
+      "learning_rate": 0.0002,
+      "loss": 1.2386,
+      "step": 7950
+    },
+    {
+      "epoch": 5.946955547254389,
+      "grad_norm": 0.958046555519104,
+      "learning_rate": 0.0002,
+      "loss": 1.2333,
+      "step": 7960
+    },
+    {
+      "epoch": 5.9544265969368695,
+      "grad_norm": 1.0538510084152222,
+      "learning_rate": 0.0002,
+      "loss": 1.2352,
+      "step": 7970
+    },
+    {
+      "epoch": 5.9618976466193505,
+      "grad_norm": 1.2681571245193481,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 7980
+    },
+    {
+      "epoch": 5.969368696301831,
+      "grad_norm": 0.8171183466911316,
+      "learning_rate": 0.0002,
+      "loss": 1.2514,
+      "step": 7990
+    },
+    {
+      "epoch": 5.976839745984311,
+      "grad_norm": 0.9109523892402649,
+      "learning_rate": 0.0002,
+      "loss": 1.3412,
+      "step": 8000
+    },
+    {
+      "epoch": 5.984310795666791,
+      "grad_norm": 1.0040639638900757,
+      "learning_rate": 0.0002,
+      "loss": 1.3497,
+      "step": 8010
+    },
+    {
+      "epoch": 5.991781845349272,
+      "grad_norm": 0.9596554040908813,
+      "learning_rate": 0.0002,
+      "loss": 1.3299,
+      "step": 8020
+    },
+    {
+      "epoch": 5.999252895031752,
+      "grad_norm": 0.9782963991165161,
+      "learning_rate": 0.0002,
+      "loss": 1.3109,
+      "step": 8030
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.0417845249176025,
+      "eval_runtime": 38.8465,
+      "eval_samples_per_second": 13.257,
+      "eval_steps_per_second": 1.673,
+      "step": 8031
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10704,
+  "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": 3.716566978626847e+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-0.9-num-6326-sd-42/checkpoint-8031/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-0.9-num-6326-sd-42/checkpoint-8031/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..67c7b4ca126d7b712ad1985ef15ffb29ebe76633
--- /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-0.9-num-6326-sd-42/checkpoint-8031/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fc87da605e94ff0ecd8f5b371302a2d5f8727b77984707a2185ddb447fc3796
+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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5accf51f7e52c9bee3ed368a11822fe2a33af2c0
--- /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-0.9-num-6326-sd-42/checkpoint-9369/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:426bc36655480ed54250101ac6f4186df4dc96d04f6982887be431fc47bdad37
+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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..efc57b90381ca916655b2cd9864166283f838984
--- /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-0.9-num-6326-sd-42/checkpoint-9369/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72e12464e1a6b9f8107b802c37b1dc5216408641b8e1fbe26005f72261a45e9e
+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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..72ca13584784595405f5a7f35258a0b2246b4293
--- /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-0.9-num-6326-sd-42/checkpoint-9369/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2040d668c6e113af8075201ef00445e6e60836ece991e36e11c7cba911d1a2e9
+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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a03b1d1b1241d394e4954b9fcd6b8c4f978efd7e
--- /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-0.9-num-6326-sd-42/checkpoint-9369/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e89ee177280e72c7dd395ddce5b22981f15c113da83fc9225f2ba2292930a05
+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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..82a9934215d2273dccdea055783314001f0ff7b6
--- /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-0.9-num-6326-sd-42/checkpoint-9369/trainer_state.json
@@ -0,0 +1,6641 @@
+{
+  "best_metric": 1.8046749830245972,
+  "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-0.9-num-6326-sd-42/checkpoint-2677",
+  "epoch": 6.999626447515876,
+  "eval_steps": 10,
+  "global_step": 9369,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007471049682480389,
+      "grad_norm": 0.4912872612476349,
+      "learning_rate": 0.0002,
+      "loss": 2.6181,
+      "step": 10
+    },
+    {
+      "epoch": 0.014942099364960777,
+      "grad_norm": 0.4856316149234772,
+      "learning_rate": 0.0002,
+      "loss": 2.2606,
+      "step": 20
+    },
+    {
+      "epoch": 0.022413149047441166,
+      "grad_norm": 0.47683125734329224,
+      "learning_rate": 0.0002,
+      "loss": 2.0957,
+      "step": 30
+    },
+    {
+      "epoch": 0.029884198729921554,
+      "grad_norm": 0.515082597732544,
+      "learning_rate": 0.0002,
+      "loss": 1.8908,
+      "step": 40
+    },
+    {
+      "epoch": 0.03735524841240194,
+      "grad_norm": 0.5299215316772461,
+      "learning_rate": 0.0002,
+      "loss": 1.9704,
+      "step": 50
+    },
+    {
+      "epoch": 0.04482629809488233,
+      "grad_norm": 0.4951399862766266,
+      "learning_rate": 0.0002,
+      "loss": 1.9225,
+      "step": 60
+    },
+    {
+      "epoch": 0.05229734777736272,
+      "grad_norm": 0.48079821467399597,
+      "learning_rate": 0.0002,
+      "loss": 1.9742,
+      "step": 70
+    },
+    {
+      "epoch": 0.05976839745984311,
+      "grad_norm": 0.49402132630348206,
+      "learning_rate": 0.0002,
+      "loss": 1.9466,
+      "step": 80
+    },
+    {
+      "epoch": 0.0672394471423235,
+      "grad_norm": 0.4778193235397339,
+      "learning_rate": 0.0002,
+      "loss": 1.8691,
+      "step": 90
+    },
+    {
+      "epoch": 0.07471049682480388,
+      "grad_norm": 0.42472657561302185,
+      "learning_rate": 0.0002,
+      "loss": 1.8455,
+      "step": 100
+    },
+    {
+      "epoch": 0.08218154650728428,
+      "grad_norm": 0.4433092474937439,
+      "learning_rate": 0.0002,
+      "loss": 1.8744,
+      "step": 110
+    },
+    {
+      "epoch": 0.08965259618976466,
+      "grad_norm": 0.4472862780094147,
+      "learning_rate": 0.0002,
+      "loss": 1.865,
+      "step": 120
+    },
+    {
+      "epoch": 0.09712364587224505,
+      "grad_norm": 0.42596298456192017,
+      "learning_rate": 0.0002,
+      "loss": 1.9256,
+      "step": 130
+    },
+    {
+      "epoch": 0.10459469555472543,
+      "grad_norm": 0.46645811200141907,
+      "learning_rate": 0.0002,
+      "loss": 1.8015,
+      "step": 140
+    },
+    {
+      "epoch": 0.11206574523720583,
+      "grad_norm": 0.41041234135627747,
+      "learning_rate": 0.0002,
+      "loss": 1.8307,
+      "step": 150
+    },
+    {
+      "epoch": 0.11953679491968622,
+      "grad_norm": 0.5329819917678833,
+      "learning_rate": 0.0002,
+      "loss": 1.8276,
+      "step": 160
+    },
+    {
+      "epoch": 0.1270078446021666,
+      "grad_norm": 0.4065922200679779,
+      "learning_rate": 0.0002,
+      "loss": 1.8118,
+      "step": 170
+    },
+    {
+      "epoch": 0.134478894284647,
+      "grad_norm": 0.38406994938850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8559,
+      "step": 180
+    },
+    {
+      "epoch": 0.14194994396712737,
+      "grad_norm": 0.4246881306171417,
+      "learning_rate": 0.0002,
+      "loss": 1.8647,
+      "step": 190
+    },
+    {
+      "epoch": 0.14942099364960776,
+      "grad_norm": 0.35136649012565613,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 200
+    },
+    {
+      "epoch": 0.15689204333208817,
+      "grad_norm": 0.43252742290496826,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.16436309301456856,
+      "grad_norm": 0.39236941933631897,
+      "learning_rate": 0.0002,
+      "loss": 1.7823,
+      "step": 220
+    },
+    {
+      "epoch": 0.17183414269704894,
+      "grad_norm": 0.3748249113559723,
+      "learning_rate": 0.0002,
+      "loss": 1.818,
+      "step": 230
+    },
+    {
+      "epoch": 0.17930519237952933,
+      "grad_norm": 0.6432855725288391,
+      "learning_rate": 0.0002,
+      "loss": 1.866,
+      "step": 240
+    },
+    {
+      "epoch": 0.1867762420620097,
+      "grad_norm": 0.34874802827835083,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 250
+    },
+    {
+      "epoch": 0.1942472917444901,
+      "grad_norm": 0.3721984326839447,
+      "learning_rate": 0.0002,
+      "loss": 1.79,
+      "step": 260
+    },
+    {
+      "epoch": 0.20171834142697048,
+      "grad_norm": 0.4339311420917511,
+      "learning_rate": 0.0002,
+      "loss": 1.8464,
+      "step": 270
+    },
+    {
+      "epoch": 0.20918939110945087,
+      "grad_norm": 0.4018215537071228,
+      "learning_rate": 0.0002,
+      "loss": 1.8665,
+      "step": 280
+    },
+    {
+      "epoch": 0.21666044079193125,
+      "grad_norm": 0.3278839886188507,
+      "learning_rate": 0.0002,
+      "loss": 1.8048,
+      "step": 290
+    },
+    {
+      "epoch": 0.22413149047441167,
+      "grad_norm": 0.36146077513694763,
+      "learning_rate": 0.0002,
+      "loss": 1.7395,
+      "step": 300
+    },
+    {
+      "epoch": 0.23160254015689205,
+      "grad_norm": 0.38175010681152344,
+      "learning_rate": 0.0002,
+      "loss": 1.7916,
+      "step": 310
+    },
+    {
+      "epoch": 0.23907358983937244,
+      "grad_norm": 0.44776618480682373,
+      "learning_rate": 0.0002,
+      "loss": 1.8593,
+      "step": 320
+    },
+    {
+      "epoch": 0.24654463952185282,
+      "grad_norm": 0.3933652937412262,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 330
+    },
+    {
+      "epoch": 0.2540156892043332,
+      "grad_norm": 0.3515005111694336,
+      "learning_rate": 0.0002,
+      "loss": 1.8393,
+      "step": 340
+    },
+    {
+      "epoch": 0.2614867388868136,
+      "grad_norm": 0.6683304309844971,
+      "learning_rate": 0.0002,
+      "loss": 1.8653,
+      "step": 350
+    },
+    {
+      "epoch": 0.268957788569294,
+      "grad_norm": 0.37093454599380493,
+      "learning_rate": 0.0002,
+      "loss": 1.8797,
+      "step": 360
+    },
+    {
+      "epoch": 0.2764288382517744,
+      "grad_norm": 0.3450651168823242,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 370
+    },
+    {
+      "epoch": 0.28389988793425475,
+      "grad_norm": 0.5140917301177979,
+      "learning_rate": 0.0002,
+      "loss": 1.7435,
+      "step": 380
+    },
+    {
+      "epoch": 0.29137093761673516,
+      "grad_norm": 0.32885563373565674,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 390
+    },
+    {
+      "epoch": 0.2988419872992155,
+      "grad_norm": 0.33962297439575195,
+      "learning_rate": 0.0002,
+      "loss": 1.8174,
+      "step": 400
+    },
+    {
+      "epoch": 0.30631303698169593,
+      "grad_norm": 0.3723141849040985,
+      "learning_rate": 0.0002,
+      "loss": 1.7467,
+      "step": 410
+    },
+    {
+      "epoch": 0.31378408666417634,
+      "grad_norm": 0.37173134088516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8459,
+      "step": 420
+    },
+    {
+      "epoch": 0.3212551363466567,
+      "grad_norm": 0.33736956119537354,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 430
+    },
+    {
+      "epoch": 0.3287261860291371,
+      "grad_norm": 0.3602448105812073,
+      "learning_rate": 0.0002,
+      "loss": 1.8367,
+      "step": 440
+    },
+    {
+      "epoch": 0.33619723571161747,
+      "grad_norm": 0.3569699227809906,
+      "learning_rate": 0.0002,
+      "loss": 1.8058,
+      "step": 450
+    },
+    {
+      "epoch": 0.3436682853940979,
+      "grad_norm": 0.31009167432785034,
+      "learning_rate": 0.0002,
+      "loss": 1.8086,
+      "step": 460
+    },
+    {
+      "epoch": 0.35113933507657824,
+      "grad_norm": 0.5278693437576294,
+      "learning_rate": 0.0002,
+      "loss": 1.8876,
+      "step": 470
+    },
+    {
+      "epoch": 0.35861038475905865,
+      "grad_norm": 0.3587537109851837,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 480
+    },
+    {
+      "epoch": 0.366081434441539,
+      "grad_norm": 0.3859670162200928,
+      "learning_rate": 0.0002,
+      "loss": 1.8046,
+      "step": 490
+    },
+    {
+      "epoch": 0.3735524841240194,
+      "grad_norm": 0.395913690328598,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 500
+    },
+    {
+      "epoch": 0.38102353380649984,
+      "grad_norm": 0.35052940249443054,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 510
+    },
+    {
+      "epoch": 0.3884945834889802,
+      "grad_norm": 0.2979494333267212,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 520
+    },
+    {
+      "epoch": 0.3959656331714606,
+      "grad_norm": 0.3062683343887329,
+      "learning_rate": 0.0002,
+      "loss": 1.8641,
+      "step": 530
+    },
+    {
+      "epoch": 0.40343668285394096,
+      "grad_norm": 0.3172847330570221,
+      "learning_rate": 0.0002,
+      "loss": 1.7651,
+      "step": 540
+    },
+    {
+      "epoch": 0.4109077325364214,
+      "grad_norm": 0.360435426235199,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 550
+    },
+    {
+      "epoch": 0.41837878221890173,
+      "grad_norm": 0.3427872359752655,
+      "learning_rate": 0.0002,
+      "loss": 1.9054,
+      "step": 560
+    },
+    {
+      "epoch": 0.42584983190138215,
+      "grad_norm": 0.34036558866500854,
+      "learning_rate": 0.0002,
+      "loss": 1.7562,
+      "step": 570
+    },
+    {
+      "epoch": 0.4333208815838625,
+      "grad_norm": 0.3365345299243927,
+      "learning_rate": 0.0002,
+      "loss": 1.7254,
+      "step": 580
+    },
+    {
+      "epoch": 0.4407919312663429,
+      "grad_norm": 0.35619041323661804,
+      "learning_rate": 0.0002,
+      "loss": 1.8328,
+      "step": 590
+    },
+    {
+      "epoch": 0.44826298094882333,
+      "grad_norm": 0.3569088280200958,
+      "learning_rate": 0.0002,
+      "loss": 1.8114,
+      "step": 600
+    },
+    {
+      "epoch": 0.4557340306313037,
+      "grad_norm": 0.3581278622150421,
+      "learning_rate": 0.0002,
+      "loss": 1.8599,
+      "step": 610
+    },
+    {
+      "epoch": 0.4632050803137841,
+      "grad_norm": 0.43197110295295715,
+      "learning_rate": 0.0002,
+      "loss": 1.7078,
+      "step": 620
+    },
+    {
+      "epoch": 0.47067612999626446,
+      "grad_norm": 0.33966198563575745,
+      "learning_rate": 0.0002,
+      "loss": 1.8257,
+      "step": 630
+    },
+    {
+      "epoch": 0.47814717967874487,
+      "grad_norm": 0.3343866467475891,
+      "learning_rate": 0.0002,
+      "loss": 1.7528,
+      "step": 640
+    },
+    {
+      "epoch": 0.48561822936122523,
+      "grad_norm": 0.33878564834594727,
+      "learning_rate": 0.0002,
+      "loss": 1.8191,
+      "step": 650
+    },
+    {
+      "epoch": 0.49308927904370564,
+      "grad_norm": 0.387195885181427,
+      "learning_rate": 0.0002,
+      "loss": 1.8801,
+      "step": 660
+    },
+    {
+      "epoch": 0.500560328726186,
+      "grad_norm": 0.3755440413951874,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 670
+    },
+    {
+      "epoch": 0.5080313784086664,
+      "grad_norm": 0.3272816836833954,
+      "learning_rate": 0.0002,
+      "loss": 1.8057,
+      "step": 680
+    },
+    {
+      "epoch": 0.5155024280911468,
+      "grad_norm": 0.36063864827156067,
+      "learning_rate": 0.0002,
+      "loss": 1.8156,
+      "step": 690
+    },
+    {
+      "epoch": 0.5229734777736272,
+      "grad_norm": 0.35317373275756836,
+      "learning_rate": 0.0002,
+      "loss": 1.8397,
+      "step": 700
+    },
+    {
+      "epoch": 0.5304445274561076,
+      "grad_norm": 0.3561195433139801,
+      "learning_rate": 0.0002,
+      "loss": 1.7603,
+      "step": 710
+    },
+    {
+      "epoch": 0.537915577138588,
+      "grad_norm": 0.31124624609947205,
+      "learning_rate": 0.0002,
+      "loss": 1.8149,
+      "step": 720
+    },
+    {
+      "epoch": 0.5453866268210683,
+      "grad_norm": 0.3294544517993927,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 730
+    },
+    {
+      "epoch": 0.5528576765035488,
+      "grad_norm": 0.31933900713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.8027,
+      "step": 740
+    },
+    {
+      "epoch": 0.5603287261860291,
+      "grad_norm": 0.3226020634174347,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 750
+    },
+    {
+      "epoch": 0.5677997758685095,
+      "grad_norm": 0.3147525489330292,
+      "learning_rate": 0.0002,
+      "loss": 1.7862,
+      "step": 760
+    },
+    {
+      "epoch": 0.57527082555099,
+      "grad_norm": 0.32234328985214233,
+      "learning_rate": 0.0002,
+      "loss": 1.9028,
+      "step": 770
+    },
+    {
+      "epoch": 0.5827418752334703,
+      "grad_norm": 0.3258664309978485,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 780
+    },
+    {
+      "epoch": 0.5902129249159507,
+      "grad_norm": 0.3166961967945099,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 790
+    },
+    {
+      "epoch": 0.597683974598431,
+      "grad_norm": 0.35621458292007446,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 800
+    },
+    {
+      "epoch": 0.6051550242809115,
+      "grad_norm": 0.3236999213695526,
+      "learning_rate": 0.0002,
+      "loss": 1.8313,
+      "step": 810
+    },
+    {
+      "epoch": 0.6126260739633919,
+      "grad_norm": 0.2892923653125763,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 820
+    },
+    {
+      "epoch": 0.6200971236458722,
+      "grad_norm": 0.4098321497440338,
+      "learning_rate": 0.0002,
+      "loss": 1.8709,
+      "step": 830
+    },
+    {
+      "epoch": 0.6275681733283527,
+      "grad_norm": 0.3337118923664093,
+      "learning_rate": 0.0002,
+      "loss": 1.7637,
+      "step": 840
+    },
+    {
+      "epoch": 0.635039223010833,
+      "grad_norm": 0.30416029691696167,
+      "learning_rate": 0.0002,
+      "loss": 1.7375,
+      "step": 850
+    },
+    {
+      "epoch": 0.6425102726933134,
+      "grad_norm": 0.3361026346683502,
+      "learning_rate": 0.0002,
+      "loss": 1.7419,
+      "step": 860
+    },
+    {
+      "epoch": 0.6499813223757938,
+      "grad_norm": 0.3537365198135376,
+      "learning_rate": 0.0002,
+      "loss": 1.732,
+      "step": 870
+    },
+    {
+      "epoch": 0.6574523720582742,
+      "grad_norm": 0.33854469656944275,
+      "learning_rate": 0.0002,
+      "loss": 1.7825,
+      "step": 880
+    },
+    {
+      "epoch": 0.6649234217407546,
+      "grad_norm": 0.3332272469997406,
+      "learning_rate": 0.0002,
+      "loss": 1.7561,
+      "step": 890
+    },
+    {
+      "epoch": 0.6723944714232349,
+      "grad_norm": 0.34954726696014404,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 900
+    },
+    {
+      "epoch": 0.6798655211057153,
+      "grad_norm": 0.2921750247478485,
+      "learning_rate": 0.0002,
+      "loss": 1.7917,
+      "step": 910
+    },
+    {
+      "epoch": 0.6873365707881958,
+      "grad_norm": 0.30508682131767273,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 920
+    },
+    {
+      "epoch": 0.6948076204706761,
+      "grad_norm": 0.32268425822257996,
+      "learning_rate": 0.0002,
+      "loss": 1.8082,
+      "step": 930
+    },
+    {
+      "epoch": 0.7022786701531565,
+      "grad_norm": 0.2844390869140625,
+      "learning_rate": 0.0002,
+      "loss": 1.8283,
+      "step": 940
+    },
+    {
+      "epoch": 0.709749719835637,
+      "grad_norm": 0.31263890862464905,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 950
+    },
+    {
+      "epoch": 0.7172207695181173,
+      "grad_norm": 0.3626808822154999,
+      "learning_rate": 0.0002,
+      "loss": 1.8081,
+      "step": 960
+    },
+    {
+      "epoch": 0.7246918192005977,
+      "grad_norm": 0.3322749733924866,
+      "learning_rate": 0.0002,
+      "loss": 1.853,
+      "step": 970
+    },
+    {
+      "epoch": 0.732162868883078,
+      "grad_norm": 0.29177871346473694,
+      "learning_rate": 0.0002,
+      "loss": 1.7912,
+      "step": 980
+    },
+    {
+      "epoch": 0.7396339185655585,
+      "grad_norm": 0.35405513644218445,
+      "learning_rate": 0.0002,
+      "loss": 1.8447,
+      "step": 990
+    },
+    {
+      "epoch": 0.7471049682480388,
+      "grad_norm": 0.39318400621414185,
+      "learning_rate": 0.0002,
+      "loss": 1.7008,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7545760179305192,
+      "grad_norm": 0.29401418566703796,
+      "learning_rate": 0.0002,
+      "loss": 1.7803,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7620470676129997,
+      "grad_norm": 0.3271748721599579,
+      "learning_rate": 0.0002,
+      "loss": 1.7649,
+      "step": 1020
+    },
+    {
+      "epoch": 0.76951811729548,
+      "grad_norm": 0.30883970856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7769891669779604,
+      "grad_norm": 0.3411838412284851,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7844602166604407,
+      "grad_norm": 0.30608129501342773,
+      "learning_rate": 0.0002,
+      "loss": 1.829,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7919312663429212,
+      "grad_norm": 0.30899080634117126,
+      "learning_rate": 0.0002,
+      "loss": 1.7815,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7994023160254016,
+      "grad_norm": 0.3160453140735626,
+      "learning_rate": 0.0002,
+      "loss": 1.7625,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8068733657078819,
+      "grad_norm": 0.30947187542915344,
+      "learning_rate": 0.0002,
+      "loss": 1.8452,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8143444153903624,
+      "grad_norm": 0.3103134036064148,
+      "learning_rate": 0.0002,
+      "loss": 1.7418,
+      "step": 1090
+    },
+    {
+      "epoch": 0.8218154650728428,
+      "grad_norm": 0.31771138310432434,
+      "learning_rate": 0.0002,
+      "loss": 1.842,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8292865147553231,
+      "grad_norm": 0.5860997438430786,
+      "learning_rate": 0.0002,
+      "loss": 1.7918,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8367575644378035,
+      "grad_norm": 0.3230148255825043,
+      "learning_rate": 0.0002,
+      "loss": 1.8443,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8442286141202839,
+      "grad_norm": 0.29611510038375854,
+      "learning_rate": 0.0002,
+      "loss": 1.8478,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8516996638027643,
+      "grad_norm": 0.3373654782772064,
+      "learning_rate": 0.0002,
+      "loss": 1.7673,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8591707134852447,
+      "grad_norm": 0.3474279046058655,
+      "learning_rate": 0.0002,
+      "loss": 1.7997,
+      "step": 1150
+    },
+    {
+      "epoch": 0.866641763167725,
+      "grad_norm": 0.35057875514030457,
+      "learning_rate": 0.0002,
+      "loss": 1.75,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8741128128502055,
+      "grad_norm": 0.39537495374679565,
+      "learning_rate": 0.0002,
+      "loss": 1.8273,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8815838625326858,
+      "grad_norm": 0.3714233636856079,
+      "learning_rate": 0.0002,
+      "loss": 1.7682,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8890549122151662,
+      "grad_norm": 0.2950296998023987,
+      "learning_rate": 0.0002,
+      "loss": 1.7549,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8965259618976467,
+      "grad_norm": 0.38182979822158813,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 1200
+    },
+    {
+      "epoch": 0.903997011580127,
+      "grad_norm": 0.27883678674697876,
+      "learning_rate": 0.0002,
+      "loss": 1.827,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9114680612626074,
+      "grad_norm": 0.33874374628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9189391109450877,
+      "grad_norm": 0.3014272153377533,
+      "learning_rate": 0.0002,
+      "loss": 1.7334,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9264101606275682,
+      "grad_norm": 0.3194271922111511,
+      "learning_rate": 0.0002,
+      "loss": 1.8235,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9338812103100486,
+      "grad_norm": 0.3049403429031372,
+      "learning_rate": 0.0002,
+      "loss": 1.7924,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9413522599925289,
+      "grad_norm": 0.30621254444122314,
+      "learning_rate": 0.0002,
+      "loss": 1.7535,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9488233096750094,
+      "grad_norm": 0.28675132989883423,
+      "learning_rate": 0.0002,
+      "loss": 1.8287,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9562943593574897,
+      "grad_norm": 0.3322032690048218,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9637654090399701,
+      "grad_norm": 0.35408294200897217,
+      "learning_rate": 0.0002,
+      "loss": 1.8054,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9712364587224505,
+      "grad_norm": 0.36386919021606445,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9787075084049309,
+      "grad_norm": 0.32338324189186096,
+      "learning_rate": 0.0002,
+      "loss": 1.8633,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9861785580874113,
+      "grad_norm": 0.3714013993740082,
+      "learning_rate": 0.0002,
+      "loss": 1.7724,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9936496077698916,
+      "grad_norm": 0.3133082389831543,
+      "learning_rate": 0.0002,
+      "loss": 1.7766,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9996264475158759,
+      "eval_loss": 1.8051470518112183,
+      "eval_runtime": 38.6332,
+      "eval_samples_per_second": 13.331,
+      "eval_steps_per_second": 1.682,
+      "step": 1338
+    },
+    {
+      "epoch": 1.001120657452372,
+      "grad_norm": 0.31595754623413086,
+      "learning_rate": 0.0002,
+      "loss": 1.8035,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0085917071348525,
+      "grad_norm": 0.3095700144767761,
+      "learning_rate": 0.0002,
+      "loss": 1.7486,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0160627568173328,
+      "grad_norm": 0.34677496552467346,
+      "learning_rate": 0.0002,
+      "loss": 1.6981,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0235338064998132,
+      "grad_norm": 0.29108840227127075,
+      "learning_rate": 0.0002,
+      "loss": 1.7377,
+      "step": 1370
+    },
+    {
+      "epoch": 1.0310048561822935,
+      "grad_norm": 0.32356950640678406,
+      "learning_rate": 0.0002,
+      "loss": 1.7194,
+      "step": 1380
+    },
+    {
+      "epoch": 1.038475905864774,
+      "grad_norm": 0.4200669229030609,
+      "learning_rate": 0.0002,
+      "loss": 1.7593,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0459469555472545,
+      "grad_norm": 0.3283711373806,
+      "learning_rate": 0.0002,
+      "loss": 1.797,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0534180052297348,
+      "grad_norm": 0.32898256182670593,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1410
+    },
+    {
+      "epoch": 1.0608890549122152,
+      "grad_norm": 0.38790300488471985,
+      "learning_rate": 0.0002,
+      "loss": 1.7559,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0683601045946955,
+      "grad_norm": 0.339800089597702,
+      "learning_rate": 0.0002,
+      "loss": 1.6922,
+      "step": 1430
+    },
+    {
+      "epoch": 1.075831154277176,
+      "grad_norm": 0.3548751175403595,
+      "learning_rate": 0.0002,
+      "loss": 1.7076,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0833022039596563,
+      "grad_norm": 0.35114359855651855,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1450
+    },
+    {
+      "epoch": 1.0907732536421366,
+      "grad_norm": 0.35226720571517944,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0982443033246172,
+      "grad_norm": 0.33665576577186584,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 1470
+    },
+    {
+      "epoch": 1.1057153530070976,
+      "grad_norm": 0.363889217376709,
+      "learning_rate": 0.0002,
+      "loss": 1.6699,
+      "step": 1480
+    },
+    {
+      "epoch": 1.113186402689578,
+      "grad_norm": 0.3826201856136322,
+      "learning_rate": 0.0002,
+      "loss": 1.7933,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1206574523720583,
+      "grad_norm": 0.34058740735054016,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1281285020545386,
+      "grad_norm": 0.3462134301662445,
+      "learning_rate": 0.0002,
+      "loss": 1.6375,
+      "step": 1510
+    },
+    {
+      "epoch": 1.135599551737019,
+      "grad_norm": 0.3396756052970886,
+      "learning_rate": 0.0002,
+      "loss": 1.7147,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1430706014194993,
+      "grad_norm": 0.32004743814468384,
+      "learning_rate": 0.0002,
+      "loss": 1.7219,
+      "step": 1530
+    },
+    {
+      "epoch": 1.15054165110198,
+      "grad_norm": 0.3397733271121979,
+      "learning_rate": 0.0002,
+      "loss": 1.743,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1580127007844603,
+      "grad_norm": 0.3783262073993683,
+      "learning_rate": 0.0002,
+      "loss": 1.7333,
+      "step": 1550
+    },
+    {
+      "epoch": 1.1654837504669406,
+      "grad_norm": 0.35121291875839233,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 1560
+    },
+    {
+      "epoch": 1.172954800149421,
+      "grad_norm": 0.35816895961761475,
+      "learning_rate": 0.0002,
+      "loss": 1.678,
+      "step": 1570
+    },
+    {
+      "epoch": 1.1804258498319014,
+      "grad_norm": 0.33843839168548584,
+      "learning_rate": 0.0002,
+      "loss": 1.7143,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1878968995143817,
+      "grad_norm": 0.3371972143650055,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 1590
+    },
+    {
+      "epoch": 1.195367949196862,
+      "grad_norm": 0.36016878485679626,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2028389988793426,
+      "grad_norm": 0.40879473090171814,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 1610
+    },
+    {
+      "epoch": 1.210310048561823,
+      "grad_norm": 0.3216715455055237,
+      "learning_rate": 0.0002,
+      "loss": 1.6955,
+      "step": 1620
+    },
+    {
+      "epoch": 1.2177810982443034,
+      "grad_norm": 0.4482610821723938,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2252521479267837,
+      "grad_norm": 0.3257700502872467,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 1640
+    },
+    {
+      "epoch": 1.232723197609264,
+      "grad_norm": 0.38646459579467773,
+      "learning_rate": 0.0002,
+      "loss": 1.7177,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2401942472917444,
+      "grad_norm": 0.4081360697746277,
+      "learning_rate": 0.0002,
+      "loss": 1.7081,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2476652969742248,
+      "grad_norm": 0.4326848089694977,
+      "learning_rate": 0.0002,
+      "loss": 1.7519,
+      "step": 1670
+    },
+    {
+      "epoch": 1.2551363466567054,
+      "grad_norm": 0.346401572227478,
+      "learning_rate": 0.0002,
+      "loss": 1.6752,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2626073963391857,
+      "grad_norm": 0.34536251425743103,
+      "learning_rate": 0.0002,
+      "loss": 1.7425,
+      "step": 1690
+    },
+    {
+      "epoch": 1.270078446021666,
+      "grad_norm": 0.41359591484069824,
+      "learning_rate": 0.0002,
+      "loss": 1.7061,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2775494957041464,
+      "grad_norm": 0.3530874252319336,
+      "learning_rate": 0.0002,
+      "loss": 1.7906,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2850205453866268,
+      "grad_norm": 0.3702719211578369,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2924915950691072,
+      "grad_norm": 0.3703329563140869,
+      "learning_rate": 0.0002,
+      "loss": 1.766,
+      "step": 1730
+    },
+    {
+      "epoch": 1.2999626447515875,
+      "grad_norm": 0.37919729948043823,
+      "learning_rate": 0.0002,
+      "loss": 1.7221,
+      "step": 1740
+    },
+    {
+      "epoch": 1.307433694434068,
+      "grad_norm": 0.32526856660842896,
+      "learning_rate": 0.0002,
+      "loss": 1.7859,
+      "step": 1750
+    },
+    {
+      "epoch": 1.3149047441165485,
+      "grad_norm": 0.36752620339393616,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3223757937990288,
+      "grad_norm": 0.3398192524909973,
+      "learning_rate": 0.0002,
+      "loss": 1.7335,
+      "step": 1770
+    },
+    {
+      "epoch": 1.3298468434815092,
+      "grad_norm": 0.37435585260391235,
+      "learning_rate": 0.0002,
+      "loss": 1.7492,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3373178931639895,
+      "grad_norm": 0.35793280601501465,
+      "learning_rate": 0.0002,
+      "loss": 1.7393,
+      "step": 1790
+    },
+    {
+      "epoch": 1.3447889428464699,
+      "grad_norm": 0.35481882095336914,
+      "learning_rate": 0.0002,
+      "loss": 1.7266,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3522599925289502,
+      "grad_norm": 0.3786393105983734,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3597310422114308,
+      "grad_norm": 0.33245593309402466,
+      "learning_rate": 0.0002,
+      "loss": 1.7169,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3672020918939112,
+      "grad_norm": 0.35388344526290894,
+      "learning_rate": 0.0002,
+      "loss": 1.7577,
+      "step": 1830
+    },
+    {
+      "epoch": 1.3746731415763915,
+      "grad_norm": 0.3695325553417206,
+      "learning_rate": 0.0002,
+      "loss": 1.6968,
+      "step": 1840
+    },
+    {
+      "epoch": 1.382144191258872,
+      "grad_norm": 0.3683604598045349,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3896152409413522,
+      "grad_norm": 0.3753012418746948,
+      "learning_rate": 0.0002,
+      "loss": 1.7878,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3970862906238326,
+      "grad_norm": 0.3331069350242615,
+      "learning_rate": 0.0002,
+      "loss": 1.6969,
+      "step": 1870
+    },
+    {
+      "epoch": 1.404557340306313,
+      "grad_norm": 0.3877500295639038,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4120283899887935,
+      "grad_norm": 0.33525151014328003,
+      "learning_rate": 0.0002,
+      "loss": 1.7586,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4194994396712737,
+      "grad_norm": 0.3697299659252167,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4269704893537543,
+      "grad_norm": 0.4029286205768585,
+      "learning_rate": 0.0002,
+      "loss": 1.6956,
+      "step": 1910
+    },
+    {
+      "epoch": 1.4344415390362346,
+      "grad_norm": 0.3596203029155731,
+      "learning_rate": 0.0002,
+      "loss": 1.6897,
+      "step": 1920
+    },
+    {
+      "epoch": 1.441912588718715,
+      "grad_norm": 0.450783908367157,
+      "learning_rate": 0.0002,
+      "loss": 1.7139,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4493836384011953,
+      "grad_norm": 0.3651481866836548,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4568546880836757,
+      "grad_norm": 0.3608424663543701,
+      "learning_rate": 0.0002,
+      "loss": 1.6637,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4643257377661563,
+      "grad_norm": 0.39684420824050903,
+      "learning_rate": 0.0002,
+      "loss": 1.8285,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4717967874486364,
+      "grad_norm": 0.34618663787841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7514,
+      "step": 1970
+    },
+    {
+      "epoch": 1.479267837131117,
+      "grad_norm": 0.4150386452674866,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4867388868135973,
+      "grad_norm": 0.35500776767730713,
+      "learning_rate": 0.0002,
+      "loss": 1.7021,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4942099364960777,
+      "grad_norm": 0.344144344329834,
+      "learning_rate": 0.0002,
+      "loss": 1.7322,
+      "step": 2000
+    },
+    {
+      "epoch": 1.501680986178558,
+      "grad_norm": 0.3340149223804474,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2010
+    },
+    {
+      "epoch": 1.5091520358610384,
+      "grad_norm": 0.37685006856918335,
+      "learning_rate": 0.0002,
+      "loss": 1.7508,
+      "step": 2020
+    },
+    {
+      "epoch": 1.516623085543519,
+      "grad_norm": 0.3699876368045807,
+      "learning_rate": 0.0002,
+      "loss": 1.8299,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5240941352259991,
+      "grad_norm": 0.3370307385921478,
+      "learning_rate": 0.0002,
+      "loss": 1.7357,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5315651849084797,
+      "grad_norm": 0.37780630588531494,
+      "learning_rate": 0.0002,
+      "loss": 1.8044,
+      "step": 2050
+    },
+    {
+      "epoch": 1.53903623459096,
+      "grad_norm": 0.370259165763855,
+      "learning_rate": 0.0002,
+      "loss": 1.7408,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5465072842734404,
+      "grad_norm": 0.3440011441707611,
+      "learning_rate": 0.0002,
+      "loss": 1.7398,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5539783339559208,
+      "grad_norm": 0.40382063388824463,
+      "learning_rate": 0.0002,
+      "loss": 1.7105,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5614493836384011,
+      "grad_norm": 0.38002029061317444,
+      "learning_rate": 0.0002,
+      "loss": 1.7071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5689204333208817,
+      "grad_norm": 0.3658451437950134,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5763914830033618,
+      "grad_norm": 0.354842871427536,
+      "learning_rate": 0.0002,
+      "loss": 1.7598,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5838625326858424,
+      "grad_norm": 0.34735530614852905,
+      "learning_rate": 0.0002,
+      "loss": 1.6898,
+      "step": 2120
+    },
+    {
+      "epoch": 1.5913335823683228,
+      "grad_norm": 0.377581924200058,
+      "learning_rate": 0.0002,
+      "loss": 1.7363,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5988046320508031,
+      "grad_norm": 0.41254034638404846,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6062756817332835,
+      "grad_norm": 0.3630715310573578,
+      "learning_rate": 0.0002,
+      "loss": 1.6782,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6137467314157639,
+      "grad_norm": 0.36980143189430237,
+      "learning_rate": 0.0002,
+      "loss": 1.7531,
+      "step": 2160
+    },
+    {
+      "epoch": 1.6212177810982444,
+      "grad_norm": 0.3634769320487976,
+      "learning_rate": 0.0002,
+      "loss": 1.6847,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6286888307807246,
+      "grad_norm": 0.3794139623641968,
+      "learning_rate": 0.0002,
+      "loss": 1.6367,
+      "step": 2180
+    },
+    {
+      "epoch": 1.6361598804632052,
+      "grad_norm": 0.359742134809494,
+      "learning_rate": 0.0002,
+      "loss": 1.7064,
+      "step": 2190
+    },
+    {
+      "epoch": 1.6436309301456855,
+      "grad_norm": 0.3770543932914734,
+      "learning_rate": 0.0002,
+      "loss": 1.7027,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6511019798281659,
+      "grad_norm": 0.3797036409378052,
+      "learning_rate": 0.0002,
+      "loss": 1.784,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6585730295106462,
+      "grad_norm": 0.35622093081474304,
+      "learning_rate": 0.0002,
+      "loss": 1.7875,
+      "step": 2220
+    },
+    {
+      "epoch": 1.6660440791931266,
+      "grad_norm": 0.34552520513534546,
+      "learning_rate": 0.0002,
+      "loss": 1.6615,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6735151288756072,
+      "grad_norm": 0.379926860332489,
+      "learning_rate": 0.0002,
+      "loss": 1.7522,
+      "step": 2240
+    },
+    {
+      "epoch": 1.6809861785580873,
+      "grad_norm": 0.37083810567855835,
+      "learning_rate": 0.0002,
+      "loss": 1.7953,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6884572282405679,
+      "grad_norm": 0.42746543884277344,
+      "learning_rate": 0.0002,
+      "loss": 1.7485,
+      "step": 2260
+    },
+    {
+      "epoch": 1.6959282779230482,
+      "grad_norm": 0.3372884690761566,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7033993276055286,
+      "grad_norm": 0.35220256447792053,
+      "learning_rate": 0.0002,
+      "loss": 1.7604,
+      "step": 2280
+    },
+    {
+      "epoch": 1.710870377288009,
+      "grad_norm": 0.3659130930900574,
+      "learning_rate": 0.0002,
+      "loss": 1.7154,
+      "step": 2290
+    },
+    {
+      "epoch": 1.7183414269704893,
+      "grad_norm": 0.37629297375679016,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2300
+    },
+    {
+      "epoch": 1.7258124766529699,
+      "grad_norm": 0.36312398314476013,
+      "learning_rate": 0.0002,
+      "loss": 1.7212,
+      "step": 2310
+    },
+    {
+      "epoch": 1.73328352633545,
+      "grad_norm": 0.467709481716156,
+      "learning_rate": 0.0002,
+      "loss": 1.7903,
+      "step": 2320
+    },
+    {
+      "epoch": 1.7407545760179306,
+      "grad_norm": 0.38685527443885803,
+      "learning_rate": 0.0002,
+      "loss": 1.696,
+      "step": 2330
+    },
+    {
+      "epoch": 1.748225625700411,
+      "grad_norm": 0.3578338325023651,
+      "learning_rate": 0.0002,
+      "loss": 1.7041,
+      "step": 2340
+    },
+    {
+      "epoch": 1.7556966753828913,
+      "grad_norm": 0.36057502031326294,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7631677250653717,
+      "grad_norm": 0.3615196645259857,
+      "learning_rate": 0.0002,
+      "loss": 1.6853,
+      "step": 2360
+    },
+    {
+      "epoch": 1.770638774747852,
+      "grad_norm": 0.4118947684764862,
+      "learning_rate": 0.0002,
+      "loss": 1.7612,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7781098244303326,
+      "grad_norm": 0.4067276120185852,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2380
+    },
+    {
+      "epoch": 1.7855808741128127,
+      "grad_norm": 0.3979823887348175,
+      "learning_rate": 0.0002,
+      "loss": 1.712,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7930519237952933,
+      "grad_norm": 0.44045883417129517,
+      "learning_rate": 0.0002,
+      "loss": 1.7644,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8005229734777737,
+      "grad_norm": 0.3998069167137146,
+      "learning_rate": 0.0002,
+      "loss": 1.7251,
+      "step": 2410
+    },
+    {
+      "epoch": 1.807994023160254,
+      "grad_norm": 0.3450094759464264,
+      "learning_rate": 0.0002,
+      "loss": 1.7354,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8154650728427344,
+      "grad_norm": 0.3759009838104248,
+      "learning_rate": 0.0002,
+      "loss": 1.6998,
+      "step": 2430
+    },
+    {
+      "epoch": 1.8229361225252148,
+      "grad_norm": 0.34347015619277954,
+      "learning_rate": 0.0002,
+      "loss": 1.7706,
+      "step": 2440
+    },
+    {
+      "epoch": 1.8304071722076953,
+      "grad_norm": 0.3511228859424591,
+      "learning_rate": 0.0002,
+      "loss": 1.7345,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8378782218901755,
+      "grad_norm": 0.36853715777397156,
+      "learning_rate": 0.0002,
+      "loss": 1.6909,
+      "step": 2460
+    },
+    {
+      "epoch": 1.845349271572656,
+      "grad_norm": 0.40659376978874207,
+      "learning_rate": 0.0002,
+      "loss": 1.6931,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8528203212551362,
+      "grad_norm": 0.39621320366859436,
+      "learning_rate": 0.0002,
+      "loss": 1.7626,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8602913709376168,
+      "grad_norm": 0.3753979504108429,
+      "learning_rate": 0.0002,
+      "loss": 1.7427,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8677624206200971,
+      "grad_norm": 0.3811938464641571,
+      "learning_rate": 0.0002,
+      "loss": 1.6622,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8752334703025775,
+      "grad_norm": 0.3432596027851105,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 2510
+    },
+    {
+      "epoch": 1.882704519985058,
+      "grad_norm": 0.3670712113380432,
+      "learning_rate": 0.0002,
+      "loss": 1.7488,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8901755696675382,
+      "grad_norm": 0.40907177329063416,
+      "learning_rate": 0.0002,
+      "loss": 1.705,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8976466193500188,
+      "grad_norm": 0.3821999728679657,
+      "learning_rate": 0.0002,
+      "loss": 1.7148,
+      "step": 2540
+    },
+    {
+      "epoch": 1.905117669032499,
+      "grad_norm": 0.36173978447914124,
+      "learning_rate": 0.0002,
+      "loss": 1.7934,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9125887187149795,
+      "grad_norm": 0.38990336656570435,
+      "learning_rate": 0.0002,
+      "loss": 1.6939,
+      "step": 2560
+    },
+    {
+      "epoch": 1.9200597683974598,
+      "grad_norm": 0.35242322087287903,
+      "learning_rate": 0.0002,
+      "loss": 1.6893,
+      "step": 2570
+    },
+    {
+      "epoch": 1.9275308180799402,
+      "grad_norm": 0.3506428003311157,
+      "learning_rate": 0.0002,
+      "loss": 1.7268,
+      "step": 2580
+    },
+    {
+      "epoch": 1.9350018677624208,
+      "grad_norm": 0.39540135860443115,
+      "learning_rate": 0.0002,
+      "loss": 1.6953,
+      "step": 2590
+    },
+    {
+      "epoch": 1.942472917444901,
+      "grad_norm": 0.3444725573062897,
+      "learning_rate": 0.0002,
+      "loss": 1.6511,
+      "step": 2600
+    },
+    {
+      "epoch": 1.9499439671273815,
+      "grad_norm": 0.3963521718978882,
+      "learning_rate": 0.0002,
+      "loss": 1.7259,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9574150168098616,
+      "grad_norm": 0.3689815402030945,
+      "learning_rate": 0.0002,
+      "loss": 1.6946,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9648860664923422,
+      "grad_norm": 0.3482626676559448,
+      "learning_rate": 0.0002,
+      "loss": 1.7384,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9723571161748226,
+      "grad_norm": 0.35832616686820984,
+      "learning_rate": 0.0002,
+      "loss": 1.7048,
+      "step": 2640
+    },
+    {
+      "epoch": 1.979828165857303,
+      "grad_norm": 0.4776208996772766,
+      "learning_rate": 0.0002,
+      "loss": 1.6681,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9872992155397835,
+      "grad_norm": 0.32570165395736694,
+      "learning_rate": 0.0002,
+      "loss": 1.6696,
+      "step": 2660
+    },
+    {
+      "epoch": 1.9947702652222636,
+      "grad_norm": 0.3380725085735321,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2670
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8046749830245972,
+      "eval_runtime": 38.5096,
+      "eval_samples_per_second": 13.373,
+      "eval_steps_per_second": 1.688,
+      "step": 2677
+    },
+    {
+      "epoch": 2.002241314904744,
+      "grad_norm": 0.36817631125450134,
+      "learning_rate": 0.0002,
+      "loss": 1.7265,
+      "step": 2680
+    },
+    {
+      "epoch": 2.0097123645872244,
+      "grad_norm": 0.4056456685066223,
+      "learning_rate": 0.0002,
+      "loss": 1.548,
+      "step": 2690
+    },
+    {
+      "epoch": 2.017183414269705,
+      "grad_norm": 0.37416863441467285,
+      "learning_rate": 0.0002,
+      "loss": 1.5515,
+      "step": 2700
+    },
+    {
+      "epoch": 2.024654463952185,
+      "grad_norm": 0.4273638427257538,
+      "learning_rate": 0.0002,
+      "loss": 1.5895,
+      "step": 2710
+    },
+    {
+      "epoch": 2.0321255136346656,
+      "grad_norm": 0.36497923731803894,
+      "learning_rate": 0.0002,
+      "loss": 1.5884,
+      "step": 2720
+    },
+    {
+      "epoch": 2.0395965633171462,
+      "grad_norm": 0.5021994113922119,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 2730
+    },
+    {
+      "epoch": 2.0470676129996264,
+      "grad_norm": 0.45896220207214355,
+      "learning_rate": 0.0002,
+      "loss": 1.6655,
+      "step": 2740
+    },
+    {
+      "epoch": 2.054538662682107,
+      "grad_norm": 0.3973815143108368,
+      "learning_rate": 0.0002,
+      "loss": 1.6305,
+      "step": 2750
+    },
+    {
+      "epoch": 2.062009712364587,
+      "grad_norm": 0.4521815776824951,
+      "learning_rate": 0.0002,
+      "loss": 1.6301,
+      "step": 2760
+    },
+    {
+      "epoch": 2.0694807620470677,
+      "grad_norm": 0.42775002121925354,
+      "learning_rate": 0.0002,
+      "loss": 1.6189,
+      "step": 2770
+    },
+    {
+      "epoch": 2.076951811729548,
+      "grad_norm": 0.48158586025238037,
+      "learning_rate": 0.0002,
+      "loss": 1.6491,
+      "step": 2780
+    },
+    {
+      "epoch": 2.0844228614120284,
+      "grad_norm": 0.4612371623516083,
+      "learning_rate": 0.0002,
+      "loss": 1.6301,
+      "step": 2790
+    },
+    {
+      "epoch": 2.091893911094509,
+      "grad_norm": 0.42536866664886475,
+      "learning_rate": 0.0002,
+      "loss": 1.6327,
+      "step": 2800
+    },
+    {
+      "epoch": 2.099364960776989,
+      "grad_norm": 0.48515772819519043,
+      "learning_rate": 0.0002,
+      "loss": 1.651,
+      "step": 2810
+    },
+    {
+      "epoch": 2.1068360104594697,
+      "grad_norm": 0.41418662667274475,
+      "learning_rate": 0.0002,
+      "loss": 1.6829,
+      "step": 2820
+    },
+    {
+      "epoch": 2.11430706014195,
+      "grad_norm": 0.4683697819709778,
+      "learning_rate": 0.0002,
+      "loss": 1.6266,
+      "step": 2830
+    },
+    {
+      "epoch": 2.1217781098244304,
+      "grad_norm": 0.4484657049179077,
+      "learning_rate": 0.0002,
+      "loss": 1.6586,
+      "step": 2840
+    },
+    {
+      "epoch": 2.1292491595069105,
+      "grad_norm": 0.6621400713920593,
+      "learning_rate": 0.0002,
+      "loss": 1.6483,
+      "step": 2850
+    },
+    {
+      "epoch": 2.136720209189391,
+      "grad_norm": 0.45074811577796936,
+      "learning_rate": 0.0002,
+      "loss": 1.5755,
+      "step": 2860
+    },
+    {
+      "epoch": 2.1441912588718717,
+      "grad_norm": 0.3513113558292389,
+      "learning_rate": 0.0002,
+      "loss": 1.6456,
+      "step": 2870
+    },
+    {
+      "epoch": 2.151662308554352,
+      "grad_norm": 0.40411314368247986,
+      "learning_rate": 0.0002,
+      "loss": 1.6081,
+      "step": 2880
+    },
+    {
+      "epoch": 2.1591333582368324,
+      "grad_norm": 0.4121065139770508,
+      "learning_rate": 0.0002,
+      "loss": 1.6323,
+      "step": 2890
+    },
+    {
+      "epoch": 2.1666044079193125,
+      "grad_norm": 0.44723689556121826,
+      "learning_rate": 0.0002,
+      "loss": 1.6324,
+      "step": 2900
+    },
+    {
+      "epoch": 2.174075457601793,
+      "grad_norm": 0.4226122498512268,
+      "learning_rate": 0.0002,
+      "loss": 1.5699,
+      "step": 2910
+    },
+    {
+      "epoch": 2.1815465072842732,
+      "grad_norm": 0.46617650985717773,
+      "learning_rate": 0.0002,
+      "loss": 1.5652,
+      "step": 2920
+    },
+    {
+      "epoch": 2.189017556966754,
+      "grad_norm": 0.4506422281265259,
+      "learning_rate": 0.0002,
+      "loss": 1.6378,
+      "step": 2930
+    },
+    {
+      "epoch": 2.1964886066492344,
+      "grad_norm": 0.4892672896385193,
+      "learning_rate": 0.0002,
+      "loss": 1.6112,
+      "step": 2940
+    },
+    {
+      "epoch": 2.2039596563317145,
+      "grad_norm": 0.44095516204833984,
+      "learning_rate": 0.0002,
+      "loss": 1.6176,
+      "step": 2950
+    },
+    {
+      "epoch": 2.211430706014195,
+      "grad_norm": 0.41522109508514404,
+      "learning_rate": 0.0002,
+      "loss": 1.6058,
+      "step": 2960
+    },
+    {
+      "epoch": 2.2189017556966752,
+      "grad_norm": 0.4860858917236328,
+      "learning_rate": 0.0002,
+      "loss": 1.5964,
+      "step": 2970
+    },
+    {
+      "epoch": 2.226372805379156,
+      "grad_norm": 0.42662516236305237,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2980
+    },
+    {
+      "epoch": 2.233843855061636,
+      "grad_norm": 0.4390648305416107,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 2990
+    },
+    {
+      "epoch": 2.2413149047441165,
+      "grad_norm": 0.47515565156936646,
+      "learning_rate": 0.0002,
+      "loss": 1.5992,
+      "step": 3000
+    },
+    {
+      "epoch": 2.248785954426597,
+      "grad_norm": 0.4104543924331665,
+      "learning_rate": 0.0002,
+      "loss": 1.5563,
+      "step": 3010
+    },
+    {
+      "epoch": 2.2562570041090773,
+      "grad_norm": 0.4404028654098511,
+      "learning_rate": 0.0002,
+      "loss": 1.6895,
+      "step": 3020
+    },
+    {
+      "epoch": 2.263728053791558,
+      "grad_norm": 0.4717366695404053,
+      "learning_rate": 0.0002,
+      "loss": 1.6088,
+      "step": 3030
+    },
+    {
+      "epoch": 2.271199103474038,
+      "grad_norm": 0.48345857858657837,
+      "learning_rate": 0.0002,
+      "loss": 1.7287,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2786701531565186,
+      "grad_norm": 0.5312452912330627,
+      "learning_rate": 0.0002,
+      "loss": 1.681,
+      "step": 3050
+    },
+    {
+      "epoch": 2.2861412028389987,
+      "grad_norm": 0.5073099732398987,
+      "learning_rate": 0.0002,
+      "loss": 1.5901,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2936122525214793,
+      "grad_norm": 0.5027463436126709,
+      "learning_rate": 0.0002,
+      "loss": 1.6914,
+      "step": 3070
+    },
+    {
+      "epoch": 2.30108330220396,
+      "grad_norm": 0.5436304807662964,
+      "learning_rate": 0.0002,
+      "loss": 1.5862,
+      "step": 3080
+    },
+    {
+      "epoch": 2.30855435188644,
+      "grad_norm": 0.4701065123081207,
+      "learning_rate": 0.0002,
+      "loss": 1.5763,
+      "step": 3090
+    },
+    {
+      "epoch": 2.3160254015689206,
+      "grad_norm": 0.46988746523857117,
+      "learning_rate": 0.0002,
+      "loss": 1.6177,
+      "step": 3100
+    },
+    {
+      "epoch": 2.3234964512514007,
+      "grad_norm": 0.45112869143486023,
+      "learning_rate": 0.0002,
+      "loss": 1.6502,
+      "step": 3110
+    },
+    {
+      "epoch": 2.3309675009338813,
+      "grad_norm": 0.5173566937446594,
+      "learning_rate": 0.0002,
+      "loss": 1.6291,
+      "step": 3120
+    },
+    {
+      "epoch": 2.3384385506163614,
+      "grad_norm": 0.40345850586891174,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 3130
+    },
+    {
+      "epoch": 2.345909600298842,
+      "grad_norm": 0.4218924939632416,
+      "learning_rate": 0.0002,
+      "loss": 1.621,
+      "step": 3140
+    },
+    {
+      "epoch": 2.3533806499813226,
+      "grad_norm": 0.41857317090034485,
+      "learning_rate": 0.0002,
+      "loss": 1.6341,
+      "step": 3150
+    },
+    {
+      "epoch": 2.3608516996638027,
+      "grad_norm": 0.4197218418121338,
+      "learning_rate": 0.0002,
+      "loss": 1.6087,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3683227493462833,
+      "grad_norm": 0.4260677397251129,
+      "learning_rate": 0.0002,
+      "loss": 1.6572,
+      "step": 3170
+    },
+    {
+      "epoch": 2.3757937990287634,
+      "grad_norm": 0.4209042191505432,
+      "learning_rate": 0.0002,
+      "loss": 1.6376,
+      "step": 3180
+    },
+    {
+      "epoch": 2.383264848711244,
+      "grad_norm": 0.4092234969139099,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 3190
+    },
+    {
+      "epoch": 2.390735898393724,
+      "grad_norm": 0.4928431510925293,
+      "learning_rate": 0.0002,
+      "loss": 1.6339,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3982069480762047,
+      "grad_norm": 0.49252402782440186,
+      "learning_rate": 0.0002,
+      "loss": 1.6015,
+      "step": 3210
+    },
+    {
+      "epoch": 2.4056779977586853,
+      "grad_norm": 0.4368397295475006,
+      "learning_rate": 0.0002,
+      "loss": 1.5773,
+      "step": 3220
+    },
+    {
+      "epoch": 2.4131490474411654,
+      "grad_norm": 0.46122390031814575,
+      "learning_rate": 0.0002,
+      "loss": 1.6629,
+      "step": 3230
+    },
+    {
+      "epoch": 2.420620097123646,
+      "grad_norm": 0.4272301197052002,
+      "learning_rate": 0.0002,
+      "loss": 1.6224,
+      "step": 3240
+    },
+    {
+      "epoch": 2.428091146806126,
+      "grad_norm": 0.41480937600135803,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 3250
+    },
+    {
+      "epoch": 2.4355621964886067,
+      "grad_norm": 0.48911941051483154,
+      "learning_rate": 0.0002,
+      "loss": 1.6281,
+      "step": 3260
+    },
+    {
+      "epoch": 2.443033246171087,
+      "grad_norm": 0.4444098472595215,
+      "learning_rate": 0.0002,
+      "loss": 1.6846,
+      "step": 3270
+    },
+    {
+      "epoch": 2.4505042958535674,
+      "grad_norm": 0.5111684799194336,
+      "learning_rate": 0.0002,
+      "loss": 1.6961,
+      "step": 3280
+    },
+    {
+      "epoch": 2.457975345536048,
+      "grad_norm": 0.5058825016021729,
+      "learning_rate": 0.0002,
+      "loss": 1.6152,
+      "step": 3290
+    },
+    {
+      "epoch": 2.465446395218528,
+      "grad_norm": 0.44173210859298706,
+      "learning_rate": 0.0002,
+      "loss": 1.625,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4729174449010087,
+      "grad_norm": 0.4659745991230011,
+      "learning_rate": 0.0002,
+      "loss": 1.6491,
+      "step": 3310
+    },
+    {
+      "epoch": 2.480388494583489,
+      "grad_norm": 0.47237497568130493,
+      "learning_rate": 0.0002,
+      "loss": 1.6114,
+      "step": 3320
+    },
+    {
+      "epoch": 2.4878595442659694,
+      "grad_norm": 0.47303131222724915,
+      "learning_rate": 0.0002,
+      "loss": 1.6193,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4953305939484496,
+      "grad_norm": 0.4522389769554138,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 3340
+    },
+    {
+      "epoch": 2.50280164363093,
+      "grad_norm": 0.4467332363128662,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3350
+    },
+    {
+      "epoch": 2.5102726933134107,
+      "grad_norm": 0.4413762092590332,
+      "learning_rate": 0.0002,
+      "loss": 1.6108,
+      "step": 3360
+    },
+    {
+      "epoch": 2.517743742995891,
+      "grad_norm": 0.495514452457428,
+      "learning_rate": 0.0002,
+      "loss": 1.537,
+      "step": 3370
+    },
+    {
+      "epoch": 2.5252147926783715,
+      "grad_norm": 0.4429773986339569,
+      "learning_rate": 0.0002,
+      "loss": 1.5839,
+      "step": 3380
+    },
+    {
+      "epoch": 2.5326858423608516,
+      "grad_norm": 0.4589079022407532,
+      "learning_rate": 0.0002,
+      "loss": 1.6522,
+      "step": 3390
+    },
+    {
+      "epoch": 2.540156892043332,
+      "grad_norm": 0.4683997333049774,
+      "learning_rate": 0.0002,
+      "loss": 1.6529,
+      "step": 3400
+    },
+    {
+      "epoch": 2.5476279417258123,
+      "grad_norm": 0.4651731252670288,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 3410
+    },
+    {
+      "epoch": 2.555098991408293,
+      "grad_norm": 0.45818084478378296,
+      "learning_rate": 0.0002,
+      "loss": 1.5918,
+      "step": 3420
+    },
+    {
+      "epoch": 2.5625700410907735,
+      "grad_norm": 0.45209529995918274,
+      "learning_rate": 0.0002,
+      "loss": 1.6326,
+      "step": 3430
+    },
+    {
+      "epoch": 2.5700410907732536,
+      "grad_norm": 0.4344733655452728,
+      "learning_rate": 0.0002,
+      "loss": 1.5606,
+      "step": 3440
+    },
+    {
+      "epoch": 2.577512140455734,
+      "grad_norm": 0.47435566782951355,
+      "learning_rate": 0.0002,
+      "loss": 1.6748,
+      "step": 3450
+    },
+    {
+      "epoch": 2.5849831901382143,
+      "grad_norm": 0.43841999769210815,
+      "learning_rate": 0.0002,
+      "loss": 1.6237,
+      "step": 3460
+    },
+    {
+      "epoch": 2.592454239820695,
+      "grad_norm": 0.4323869049549103,
+      "learning_rate": 0.0002,
+      "loss": 1.7207,
+      "step": 3470
+    },
+    {
+      "epoch": 2.599925289503175,
+      "grad_norm": 0.44355881214141846,
+      "learning_rate": 0.0002,
+      "loss": 1.5494,
+      "step": 3480
+    },
+    {
+      "epoch": 2.6073963391856556,
+      "grad_norm": 0.45847779512405396,
+      "learning_rate": 0.0002,
+      "loss": 1.665,
+      "step": 3490
+    },
+    {
+      "epoch": 2.614867388868136,
+      "grad_norm": 0.4411061704158783,
+      "learning_rate": 0.0002,
+      "loss": 1.6006,
+      "step": 3500
+    },
+    {
+      "epoch": 2.6223384385506163,
+      "grad_norm": 0.4446796178817749,
+      "learning_rate": 0.0002,
+      "loss": 1.5868,
+      "step": 3510
+    },
+    {
+      "epoch": 2.629809488233097,
+      "grad_norm": 0.41969653964042664,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 3520
+    },
+    {
+      "epoch": 2.637280537915577,
+      "grad_norm": 0.5263747572898865,
+      "learning_rate": 0.0002,
+      "loss": 1.6798,
+      "step": 3530
+    },
+    {
+      "epoch": 2.6447515875980576,
+      "grad_norm": 0.47719451785087585,
+      "learning_rate": 0.0002,
+      "loss": 1.6309,
+      "step": 3540
+    },
+    {
+      "epoch": 2.6522226372805378,
+      "grad_norm": 0.46574118733406067,
+      "learning_rate": 0.0002,
+      "loss": 1.7024,
+      "step": 3550
+    },
+    {
+      "epoch": 2.6596936869630183,
+      "grad_norm": 0.46867135167121887,
+      "learning_rate": 0.0002,
+      "loss": 1.618,
+      "step": 3560
+    },
+    {
+      "epoch": 2.667164736645499,
+      "grad_norm": 0.4441198706626892,
+      "learning_rate": 0.0002,
+      "loss": 1.5885,
+      "step": 3570
+    },
+    {
+      "epoch": 2.674635786327979,
+      "grad_norm": 0.4871319830417633,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6821068360104596,
+      "grad_norm": 0.43900373578071594,
+      "learning_rate": 0.0002,
+      "loss": 1.6575,
+      "step": 3590
+    },
+    {
+      "epoch": 2.6895778856929398,
+      "grad_norm": 0.42509549856185913,
+      "learning_rate": 0.0002,
+      "loss": 1.6071,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6970489353754203,
+      "grad_norm": 0.4691086709499359,
+      "learning_rate": 0.0002,
+      "loss": 1.5651,
+      "step": 3610
+    },
+    {
+      "epoch": 2.7045199850579005,
+      "grad_norm": 0.46318942308425903,
+      "learning_rate": 0.0002,
+      "loss": 1.5491,
+      "step": 3620
+    },
+    {
+      "epoch": 2.711991034740381,
+      "grad_norm": 0.44631096720695496,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 3630
+    },
+    {
+      "epoch": 2.7194620844228616,
+      "grad_norm": 0.42315489053726196,
+      "learning_rate": 0.0002,
+      "loss": 1.6831,
+      "step": 3640
+    },
+    {
+      "epoch": 2.7269331341053418,
+      "grad_norm": 0.4971241056919098,
+      "learning_rate": 0.0002,
+      "loss": 1.6008,
+      "step": 3650
+    },
+    {
+      "epoch": 2.7344041837878224,
+      "grad_norm": 0.4578486382961273,
+      "learning_rate": 0.0002,
+      "loss": 1.6042,
+      "step": 3660
+    },
+    {
+      "epoch": 2.7418752334703025,
+      "grad_norm": 0.46584776043891907,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 3670
+    },
+    {
+      "epoch": 2.749346283152783,
+      "grad_norm": 0.4951731264591217,
+      "learning_rate": 0.0002,
+      "loss": 1.6809,
+      "step": 3680
+    },
+    {
+      "epoch": 2.756817332835263,
+      "grad_norm": 0.4935225546360016,
+      "learning_rate": 0.0002,
+      "loss": 1.6226,
+      "step": 3690
+    },
+    {
+      "epoch": 2.764288382517744,
+      "grad_norm": 0.41805586218833923,
+      "learning_rate": 0.0002,
+      "loss": 1.5878,
+      "step": 3700
+    },
+    {
+      "epoch": 2.7717594322002244,
+      "grad_norm": 0.4417555630207062,
+      "learning_rate": 0.0002,
+      "loss": 1.7173,
+      "step": 3710
+    },
+    {
+      "epoch": 2.7792304818827045,
+      "grad_norm": 0.48229655623435974,
+      "learning_rate": 0.0002,
+      "loss": 1.6398,
+      "step": 3720
+    },
+    {
+      "epoch": 2.786701531565185,
+      "grad_norm": 0.48562315106391907,
+      "learning_rate": 0.0002,
+      "loss": 1.6074,
+      "step": 3730
+    },
+    {
+      "epoch": 2.794172581247665,
+      "grad_norm": 0.4473940432071686,
+      "learning_rate": 0.0002,
+      "loss": 1.607,
+      "step": 3740
+    },
+    {
+      "epoch": 2.801643630930146,
+      "grad_norm": 0.4626813232898712,
+      "learning_rate": 0.0002,
+      "loss": 1.6065,
+      "step": 3750
+    },
+    {
+      "epoch": 2.809114680612626,
+      "grad_norm": 0.4339792728424072,
+      "learning_rate": 0.0002,
+      "loss": 1.6296,
+      "step": 3760
+    },
+    {
+      "epoch": 2.8165857302951065,
+      "grad_norm": 0.5250858068466187,
+      "learning_rate": 0.0002,
+      "loss": 1.6815,
+      "step": 3770
+    },
+    {
+      "epoch": 2.824056779977587,
+      "grad_norm": 0.4537523090839386,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 3780
+    },
+    {
+      "epoch": 2.831527829660067,
+      "grad_norm": 0.5646113157272339,
+      "learning_rate": 0.0002,
+      "loss": 1.6535,
+      "step": 3790
+    },
+    {
+      "epoch": 2.8389988793425474,
+      "grad_norm": 0.44243332743644714,
+      "learning_rate": 0.0002,
+      "loss": 1.5712,
+      "step": 3800
+    },
+    {
+      "epoch": 2.846469929025028,
+      "grad_norm": 0.4585791826248169,
+      "learning_rate": 0.0002,
+      "loss": 1.6478,
+      "step": 3810
+    },
+    {
+      "epoch": 2.8539409787075085,
+      "grad_norm": 0.489702045917511,
+      "learning_rate": 0.0002,
+      "loss": 1.6854,
+      "step": 3820
+    },
+    {
+      "epoch": 2.8614120283899886,
+      "grad_norm": 0.502470850944519,
+      "learning_rate": 0.0002,
+      "loss": 1.7066,
+      "step": 3830
+    },
+    {
+      "epoch": 2.8688830780724692,
+      "grad_norm": 0.4395960867404938,
+      "learning_rate": 0.0002,
+      "loss": 1.5785,
+      "step": 3840
+    },
+    {
+      "epoch": 2.87635412775495,
+      "grad_norm": 0.4348670244216919,
+      "learning_rate": 0.0002,
+      "loss": 1.6434,
+      "step": 3850
+    },
+    {
+      "epoch": 2.88382517743743,
+      "grad_norm": 0.48852720856666565,
+      "learning_rate": 0.0002,
+      "loss": 1.6163,
+      "step": 3860
+    },
+    {
+      "epoch": 2.89129622711991,
+      "grad_norm": 0.45317450165748596,
+      "learning_rate": 0.0002,
+      "loss": 1.5916,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8987672768023907,
+      "grad_norm": 0.4732758700847626,
+      "learning_rate": 0.0002,
+      "loss": 1.6486,
+      "step": 3880
+    },
+    {
+      "epoch": 2.9062383264848712,
+      "grad_norm": 0.45238012075424194,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 3890
+    },
+    {
+      "epoch": 2.9137093761673514,
+      "grad_norm": 0.48838064074516296,
+      "learning_rate": 0.0002,
+      "loss": 1.6228,
+      "step": 3900
+    },
+    {
+      "epoch": 2.921180425849832,
+      "grad_norm": 0.43496349453926086,
+      "learning_rate": 0.0002,
+      "loss": 1.658,
+      "step": 3910
+    },
+    {
+      "epoch": 2.9286514755323125,
+      "grad_norm": 0.47963935136795044,
+      "learning_rate": 0.0002,
+      "loss": 1.7063,
+      "step": 3920
+    },
+    {
+      "epoch": 2.9361225252147927,
+      "grad_norm": 0.4544987976551056,
+      "learning_rate": 0.0002,
+      "loss": 1.6553,
+      "step": 3930
+    },
+    {
+      "epoch": 2.943593574897273,
+      "grad_norm": 0.4622892141342163,
+      "learning_rate": 0.0002,
+      "loss": 1.6192,
+      "step": 3940
+    },
+    {
+      "epoch": 2.9510646245797534,
+      "grad_norm": 0.47026222944259644,
+      "learning_rate": 0.0002,
+      "loss": 1.6178,
+      "step": 3950
+    },
+    {
+      "epoch": 2.958535674262234,
+      "grad_norm": 0.4549552798271179,
+      "learning_rate": 0.0002,
+      "loss": 1.6612,
+      "step": 3960
+    },
+    {
+      "epoch": 2.966006723944714,
+      "grad_norm": 0.46647515892982483,
+      "learning_rate": 0.0002,
+      "loss": 1.6458,
+      "step": 3970
+    },
+    {
+      "epoch": 2.9734777736271947,
+      "grad_norm": 0.45095112919807434,
+      "learning_rate": 0.0002,
+      "loss": 1.6051,
+      "step": 3980
+    },
+    {
+      "epoch": 2.9809488233096753,
+      "grad_norm": 0.4690017104148865,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 3990
+    },
+    {
+      "epoch": 2.9884198729921554,
+      "grad_norm": 0.4603444039821625,
+      "learning_rate": 0.0002,
+      "loss": 1.6061,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9958909226746355,
+      "grad_norm": 0.4743294417858124,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 4010
+    },
+    {
+      "epoch": 2.999626447515876,
+      "eval_loss": 1.8252571821212769,
+      "eval_runtime": 38.7853,
+      "eval_samples_per_second": 13.278,
+      "eval_steps_per_second": 1.676,
+      "step": 4015
+    },
+    {
+      "epoch": 3.003361972357116,
+      "grad_norm": 0.4919724464416504,
+      "learning_rate": 0.0002,
+      "loss": 1.6512,
+      "step": 4020
+    },
+    {
+      "epoch": 3.0108330220395967,
+      "grad_norm": 0.4747185707092285,
+      "learning_rate": 0.0002,
+      "loss": 1.5354,
+      "step": 4030
+    },
+    {
+      "epoch": 3.018304071722077,
+      "grad_norm": 0.4797595143318176,
+      "learning_rate": 0.0002,
+      "loss": 1.568,
+      "step": 4040
+    },
+    {
+      "epoch": 3.0257751214045574,
+      "grad_norm": 0.5450999140739441,
+      "learning_rate": 0.0002,
+      "loss": 1.5194,
+      "step": 4050
+    },
+    {
+      "epoch": 3.0332461710870375,
+      "grad_norm": 0.49058812856674194,
+      "learning_rate": 0.0002,
+      "loss": 1.5065,
+      "step": 4060
+    },
+    {
+      "epoch": 3.040717220769518,
+      "grad_norm": 0.5219563841819763,
+      "learning_rate": 0.0002,
+      "loss": 1.4884,
+      "step": 4070
+    },
+    {
+      "epoch": 3.0481882704519987,
+      "grad_norm": 0.515628457069397,
+      "learning_rate": 0.0002,
+      "loss": 1.4742,
+      "step": 4080
+    },
+    {
+      "epoch": 3.055659320134479,
+      "grad_norm": 0.6145984530448914,
+      "learning_rate": 0.0002,
+      "loss": 1.5313,
+      "step": 4090
+    },
+    {
+      "epoch": 3.0631303698169594,
+      "grad_norm": 0.6067144274711609,
+      "learning_rate": 0.0002,
+      "loss": 1.4989,
+      "step": 4100
+    },
+    {
+      "epoch": 3.0706014194994395,
+      "grad_norm": 0.5773133039474487,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 4110
+    },
+    {
+      "epoch": 3.07807246918192,
+      "grad_norm": 0.6894241571426392,
+      "learning_rate": 0.0002,
+      "loss": 1.5374,
+      "step": 4120
+    },
+    {
+      "epoch": 3.0855435188644003,
+      "grad_norm": 0.6422514915466309,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 4130
+    },
+    {
+      "epoch": 3.093014568546881,
+      "grad_norm": 0.6119855046272278,
+      "learning_rate": 0.0002,
+      "loss": 1.4724,
+      "step": 4140
+    },
+    {
+      "epoch": 3.1004856182293614,
+      "grad_norm": 0.5847280025482178,
+      "learning_rate": 0.0002,
+      "loss": 1.5361,
+      "step": 4150
+    },
+    {
+      "epoch": 3.1079566679118416,
+      "grad_norm": 0.5401515960693359,
+      "learning_rate": 0.0002,
+      "loss": 1.5151,
+      "step": 4160
+    },
+    {
+      "epoch": 3.115427717594322,
+      "grad_norm": 0.6501587629318237,
+      "learning_rate": 0.0002,
+      "loss": 1.502,
+      "step": 4170
+    },
+    {
+      "epoch": 3.1228987672768023,
+      "grad_norm": 0.5988039374351501,
+      "learning_rate": 0.0002,
+      "loss": 1.4952,
+      "step": 4180
+    },
+    {
+      "epoch": 3.130369816959283,
+      "grad_norm": 0.4982665181159973,
+      "learning_rate": 0.0002,
+      "loss": 1.5287,
+      "step": 4190
+    },
+    {
+      "epoch": 3.137840866641763,
+      "grad_norm": 0.5548039078712463,
+      "learning_rate": 0.0002,
+      "loss": 1.5078,
+      "step": 4200
+    },
+    {
+      "epoch": 3.1453119163242436,
+      "grad_norm": 0.5920777320861816,
+      "learning_rate": 0.0002,
+      "loss": 1.4904,
+      "step": 4210
+    },
+    {
+      "epoch": 3.152782966006724,
+      "grad_norm": 0.6965190172195435,
+      "learning_rate": 0.0002,
+      "loss": 1.442,
+      "step": 4220
+    },
+    {
+      "epoch": 3.1602540156892043,
+      "grad_norm": 0.5196244716644287,
+      "learning_rate": 0.0002,
+      "loss": 1.557,
+      "step": 4230
+    },
+    {
+      "epoch": 3.167725065371685,
+      "grad_norm": 0.6942682266235352,
+      "learning_rate": 0.0002,
+      "loss": 1.5706,
+      "step": 4240
+    },
+    {
+      "epoch": 3.175196115054165,
+      "grad_norm": 0.5765156149864197,
+      "learning_rate": 0.0002,
+      "loss": 1.5407,
+      "step": 4250
+    },
+    {
+      "epoch": 3.1826671647366456,
+      "grad_norm": 0.5801976919174194,
+      "learning_rate": 0.0002,
+      "loss": 1.4963,
+      "step": 4260
+    },
+    {
+      "epoch": 3.1901382144191257,
+      "grad_norm": 0.6260752081871033,
+      "learning_rate": 0.0002,
+      "loss": 1.4988,
+      "step": 4270
+    },
+    {
+      "epoch": 3.1976092641016063,
+      "grad_norm": 0.6610770225524902,
+      "learning_rate": 0.0002,
+      "loss": 1.5074,
+      "step": 4280
+    },
+    {
+      "epoch": 3.205080313784087,
+      "grad_norm": 0.5762143135070801,
+      "learning_rate": 0.0002,
+      "loss": 1.4657,
+      "step": 4290
+    },
+    {
+      "epoch": 3.212551363466567,
+      "grad_norm": 0.5926990509033203,
+      "learning_rate": 0.0002,
+      "loss": 1.5181,
+      "step": 4300
+    },
+    {
+      "epoch": 3.2200224131490476,
+      "grad_norm": 0.7373854517936707,
+      "learning_rate": 0.0002,
+      "loss": 1.5492,
+      "step": 4310
+    },
+    {
+      "epoch": 3.2274934628315277,
+      "grad_norm": 0.5963311195373535,
+      "learning_rate": 0.0002,
+      "loss": 1.4648,
+      "step": 4320
+    },
+    {
+      "epoch": 3.2349645125140083,
+      "grad_norm": 0.5754616856575012,
+      "learning_rate": 0.0002,
+      "loss": 1.5262,
+      "step": 4330
+    },
+    {
+      "epoch": 3.2424355621964884,
+      "grad_norm": 0.6116095781326294,
+      "learning_rate": 0.0002,
+      "loss": 1.4767,
+      "step": 4340
+    },
+    {
+      "epoch": 3.249906611878969,
+      "grad_norm": 0.6001536846160889,
+      "learning_rate": 0.0002,
+      "loss": 1.5008,
+      "step": 4350
+    },
+    {
+      "epoch": 3.257377661561449,
+      "grad_norm": 0.5270227789878845,
+      "learning_rate": 0.0002,
+      "loss": 1.5738,
+      "step": 4360
+    },
+    {
+      "epoch": 3.2648487112439297,
+      "grad_norm": 0.6666602492332458,
+      "learning_rate": 0.0002,
+      "loss": 1.5235,
+      "step": 4370
+    },
+    {
+      "epoch": 3.2723197609264103,
+      "grad_norm": 0.520310640335083,
+      "learning_rate": 0.0002,
+      "loss": 1.5665,
+      "step": 4380
+    },
+    {
+      "epoch": 3.2797908106088904,
+      "grad_norm": 0.5165975093841553,
+      "learning_rate": 0.0002,
+      "loss": 1.542,
+      "step": 4390
+    },
+    {
+      "epoch": 3.287261860291371,
+      "grad_norm": 0.6080228686332703,
+      "learning_rate": 0.0002,
+      "loss": 1.4746,
+      "step": 4400
+    },
+    {
+      "epoch": 3.294732909973851,
+      "grad_norm": 0.670122504234314,
+      "learning_rate": 0.0002,
+      "loss": 1.4901,
+      "step": 4410
+    },
+    {
+      "epoch": 3.3022039596563317,
+      "grad_norm": 0.6019457578659058,
+      "learning_rate": 0.0002,
+      "loss": 1.4677,
+      "step": 4420
+    },
+    {
+      "epoch": 3.309675009338812,
+      "grad_norm": 0.5519300103187561,
+      "learning_rate": 0.0002,
+      "loss": 1.4249,
+      "step": 4430
+    },
+    {
+      "epoch": 3.3171460590212924,
+      "grad_norm": 0.5958521962165833,
+      "learning_rate": 0.0002,
+      "loss": 1.555,
+      "step": 4440
+    },
+    {
+      "epoch": 3.324617108703773,
+      "grad_norm": 0.5552705526351929,
+      "learning_rate": 0.0002,
+      "loss": 1.5067,
+      "step": 4450
+    },
+    {
+      "epoch": 3.332088158386253,
+      "grad_norm": 0.6583784818649292,
+      "learning_rate": 0.0002,
+      "loss": 1.5926,
+      "step": 4460
+    },
+    {
+      "epoch": 3.3395592080687337,
+      "grad_norm": 0.5815939903259277,
+      "learning_rate": 0.0002,
+      "loss": 1.4206,
+      "step": 4470
+    },
+    {
+      "epoch": 3.347030257751214,
+      "grad_norm": 1.3342205286026,
+      "learning_rate": 0.0002,
+      "loss": 1.5942,
+      "step": 4480
+    },
+    {
+      "epoch": 3.3545013074336945,
+      "grad_norm": 0.6341500878334045,
+      "learning_rate": 0.0002,
+      "loss": 1.484,
+      "step": 4490
+    },
+    {
+      "epoch": 3.3619723571161746,
+      "grad_norm": 0.6384079456329346,
+      "learning_rate": 0.0002,
+      "loss": 1.5219,
+      "step": 4500
+    },
+    {
+      "epoch": 3.369443406798655,
+      "grad_norm": 0.6098346710205078,
+      "learning_rate": 0.0002,
+      "loss": 1.5222,
+      "step": 4510
+    },
+    {
+      "epoch": 3.3769144564811358,
+      "grad_norm": 0.5958296656608582,
+      "learning_rate": 0.0002,
+      "loss": 1.5475,
+      "step": 4520
+    },
+    {
+      "epoch": 3.384385506163616,
+      "grad_norm": 0.6157881617546082,
+      "learning_rate": 0.0002,
+      "loss": 1.5171,
+      "step": 4530
+    },
+    {
+      "epoch": 3.3918565558460965,
+      "grad_norm": 0.5671007037162781,
+      "learning_rate": 0.0002,
+      "loss": 1.569,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3993276055285766,
+      "grad_norm": 0.6203294992446899,
+      "learning_rate": 0.0002,
+      "loss": 1.604,
+      "step": 4550
+    },
+    {
+      "epoch": 3.406798655211057,
+      "grad_norm": 0.6743317246437073,
+      "learning_rate": 0.0002,
+      "loss": 1.5364,
+      "step": 4560
+    },
+    {
+      "epoch": 3.4142697048935373,
+      "grad_norm": 0.731765627861023,
+      "learning_rate": 0.0002,
+      "loss": 1.5034,
+      "step": 4570
+    },
+    {
+      "epoch": 3.421740754576018,
+      "grad_norm": 0.6285187602043152,
+      "learning_rate": 0.0002,
+      "loss": 1.4585,
+      "step": 4580
+    },
+    {
+      "epoch": 3.4292118042584985,
+      "grad_norm": 0.612680196762085,
+      "learning_rate": 0.0002,
+      "loss": 1.5296,
+      "step": 4590
+    },
+    {
+      "epoch": 3.4366828539409786,
+      "grad_norm": 0.6413681507110596,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 4600
+    },
+    {
+      "epoch": 3.444153903623459,
+      "grad_norm": 0.6240990161895752,
+      "learning_rate": 0.0002,
+      "loss": 1.5026,
+      "step": 4610
+    },
+    {
+      "epoch": 3.4516249533059393,
+      "grad_norm": 0.5095735192298889,
+      "learning_rate": 0.0002,
+      "loss": 1.5887,
+      "step": 4620
+    },
+    {
+      "epoch": 3.45909600298842,
+      "grad_norm": 0.5699611902236938,
+      "learning_rate": 0.0002,
+      "loss": 1.4906,
+      "step": 4630
+    },
+    {
+      "epoch": 3.4665670526709,
+      "grad_norm": 0.7289775609970093,
+      "learning_rate": 0.0002,
+      "loss": 1.5176,
+      "step": 4640
+    },
+    {
+      "epoch": 3.4740381023533806,
+      "grad_norm": 0.6211609840393066,
+      "learning_rate": 0.0002,
+      "loss": 1.5467,
+      "step": 4650
+    },
+    {
+      "epoch": 3.481509152035861,
+      "grad_norm": 0.5714802145957947,
+      "learning_rate": 0.0002,
+      "loss": 1.533,
+      "step": 4660
+    },
+    {
+      "epoch": 3.4889802017183413,
+      "grad_norm": 0.6287049651145935,
+      "learning_rate": 0.0002,
+      "loss": 1.5096,
+      "step": 4670
+    },
+    {
+      "epoch": 3.496451251400822,
+      "grad_norm": 0.5480595827102661,
+      "learning_rate": 0.0002,
+      "loss": 1.4212,
+      "step": 4680
+    },
+    {
+      "epoch": 3.503922301083302,
+      "grad_norm": 0.5683253407478333,
+      "learning_rate": 0.0002,
+      "loss": 1.4746,
+      "step": 4690
+    },
+    {
+      "epoch": 3.5113933507657826,
+      "grad_norm": 0.601140558719635,
+      "learning_rate": 0.0002,
+      "loss": 1.5012,
+      "step": 4700
+    },
+    {
+      "epoch": 3.5188644004482628,
+      "grad_norm": 0.5344498157501221,
+      "learning_rate": 0.0002,
+      "loss": 1.5383,
+      "step": 4710
+    },
+    {
+      "epoch": 3.5263354501307433,
+      "grad_norm": 0.5739690661430359,
+      "learning_rate": 0.0002,
+      "loss": 1.5428,
+      "step": 4720
+    },
+    {
+      "epoch": 3.533806499813224,
+      "grad_norm": 0.5640085935592651,
+      "learning_rate": 0.0002,
+      "loss": 1.5589,
+      "step": 4730
+    },
+    {
+      "epoch": 3.541277549495704,
+      "grad_norm": 0.5967805981636047,
+      "learning_rate": 0.0002,
+      "loss": 1.487,
+      "step": 4740
+    },
+    {
+      "epoch": 3.5487485991781846,
+      "grad_norm": 0.6138835549354553,
+      "learning_rate": 0.0002,
+      "loss": 1.5461,
+      "step": 4750
+    },
+    {
+      "epoch": 3.5562196488606648,
+      "grad_norm": 0.6779900193214417,
+      "learning_rate": 0.0002,
+      "loss": 1.5502,
+      "step": 4760
+    },
+    {
+      "epoch": 3.5636906985431454,
+      "grad_norm": 0.6122010350227356,
+      "learning_rate": 0.0002,
+      "loss": 1.4917,
+      "step": 4770
+    },
+    {
+      "epoch": 3.5711617482256255,
+      "grad_norm": 0.5685241222381592,
+      "learning_rate": 0.0002,
+      "loss": 1.5405,
+      "step": 4780
+    },
+    {
+      "epoch": 3.578632797908106,
+      "grad_norm": 0.604583203792572,
+      "learning_rate": 0.0002,
+      "loss": 1.5427,
+      "step": 4790
+    },
+    {
+      "epoch": 3.5861038475905866,
+      "grad_norm": 0.651165246963501,
+      "learning_rate": 0.0002,
+      "loss": 1.4514,
+      "step": 4800
+    },
+    {
+      "epoch": 3.593574897273067,
+      "grad_norm": 0.6398511528968811,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 4810
+    },
+    {
+      "epoch": 3.6010459469555474,
+      "grad_norm": 0.6444641351699829,
+      "learning_rate": 0.0002,
+      "loss": 1.4261,
+      "step": 4820
+    },
+    {
+      "epoch": 3.6085169966380275,
+      "grad_norm": 0.6018481850624084,
+      "learning_rate": 0.0002,
+      "loss": 1.5274,
+      "step": 4830
+    },
+    {
+      "epoch": 3.615988046320508,
+      "grad_norm": 0.6025291085243225,
+      "learning_rate": 0.0002,
+      "loss": 1.4647,
+      "step": 4840
+    },
+    {
+      "epoch": 3.623459096002988,
+      "grad_norm": 0.6810156106948853,
+      "learning_rate": 0.0002,
+      "loss": 1.5609,
+      "step": 4850
+    },
+    {
+      "epoch": 3.630930145685469,
+      "grad_norm": 0.6408044695854187,
+      "learning_rate": 0.0002,
+      "loss": 1.5299,
+      "step": 4860
+    },
+    {
+      "epoch": 3.6384011953679494,
+      "grad_norm": 0.5608272552490234,
+      "learning_rate": 0.0002,
+      "loss": 1.5366,
+      "step": 4870
+    },
+    {
+      "epoch": 3.6458722450504295,
+      "grad_norm": 0.6136814951896667,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 4880
+    },
+    {
+      "epoch": 3.65334329473291,
+      "grad_norm": 0.5927900075912476,
+      "learning_rate": 0.0002,
+      "loss": 1.5021,
+      "step": 4890
+    },
+    {
+      "epoch": 3.66081434441539,
+      "grad_norm": 0.5336901545524597,
+      "learning_rate": 0.0002,
+      "loss": 1.6084,
+      "step": 4900
+    },
+    {
+      "epoch": 3.668285394097871,
+      "grad_norm": 0.7823320627212524,
+      "learning_rate": 0.0002,
+      "loss": 1.5701,
+      "step": 4910
+    },
+    {
+      "epoch": 3.675756443780351,
+      "grad_norm": 0.6703504323959351,
+      "learning_rate": 0.0002,
+      "loss": 1.4881,
+      "step": 4920
+    },
+    {
+      "epoch": 3.6832274934628315,
+      "grad_norm": 0.6061160564422607,
+      "learning_rate": 0.0002,
+      "loss": 1.5332,
+      "step": 4930
+    },
+    {
+      "epoch": 3.690698543145312,
+      "grad_norm": 0.6237227916717529,
+      "learning_rate": 0.0002,
+      "loss": 1.5405,
+      "step": 4940
+    },
+    {
+      "epoch": 3.6981695928277922,
+      "grad_norm": 0.5985278487205505,
+      "learning_rate": 0.0002,
+      "loss": 1.497,
+      "step": 4950
+    },
+    {
+      "epoch": 3.705640642510273,
+      "grad_norm": 0.6483839750289917,
+      "learning_rate": 0.0002,
+      "loss": 1.5132,
+      "step": 4960
+    },
+    {
+      "epoch": 3.713111692192753,
+      "grad_norm": 0.5788805484771729,
+      "learning_rate": 0.0002,
+      "loss": 1.5338,
+      "step": 4970
+    },
+    {
+      "epoch": 3.7205827418752335,
+      "grad_norm": 0.5609974265098572,
+      "learning_rate": 0.0002,
+      "loss": 1.5258,
+      "step": 4980
+    },
+    {
+      "epoch": 3.7280537915577137,
+      "grad_norm": 0.5681300759315491,
+      "learning_rate": 0.0002,
+      "loss": 1.4759,
+      "step": 4990
+    },
+    {
+      "epoch": 3.7355248412401942,
+      "grad_norm": 0.5860186219215393,
+      "learning_rate": 0.0002,
+      "loss": 1.6018,
+      "step": 5000
+    },
+    {
+      "epoch": 3.742995890922675,
+      "grad_norm": 0.5718157291412354,
+      "learning_rate": 0.0002,
+      "loss": 1.58,
+      "step": 5010
+    },
+    {
+      "epoch": 3.750466940605155,
+      "grad_norm": 0.6173721551895142,
+      "learning_rate": 0.0002,
+      "loss": 1.5834,
+      "step": 5020
+    },
+    {
+      "epoch": 3.7579379902876355,
+      "grad_norm": 0.629152238368988,
+      "learning_rate": 0.0002,
+      "loss": 1.5617,
+      "step": 5030
+    },
+    {
+      "epoch": 3.7654090399701157,
+      "grad_norm": 0.5666284561157227,
+      "learning_rate": 0.0002,
+      "loss": 1.519,
+      "step": 5040
+    },
+    {
+      "epoch": 3.7728800896525962,
+      "grad_norm": 0.6053005456924438,
+      "learning_rate": 0.0002,
+      "loss": 1.5329,
+      "step": 5050
+    },
+    {
+      "epoch": 3.7803511393350764,
+      "grad_norm": 0.5870583057403564,
+      "learning_rate": 0.0002,
+      "loss": 1.5404,
+      "step": 5060
+    },
+    {
+      "epoch": 3.787822189017557,
+      "grad_norm": 0.5422009229660034,
+      "learning_rate": 0.0002,
+      "loss": 1.4444,
+      "step": 5070
+    },
+    {
+      "epoch": 3.7952932387000375,
+      "grad_norm": 0.5396918058395386,
+      "learning_rate": 0.0002,
+      "loss": 1.5308,
+      "step": 5080
+    },
+    {
+      "epoch": 3.8027642883825177,
+      "grad_norm": 0.5544713139533997,
+      "learning_rate": 0.0002,
+      "loss": 1.464,
+      "step": 5090
+    },
+    {
+      "epoch": 3.8102353380649983,
+      "grad_norm": 0.5983749628067017,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5100
+    },
+    {
+      "epoch": 3.8177063877474784,
+      "grad_norm": 0.5702024102210999,
+      "learning_rate": 0.0002,
+      "loss": 1.4972,
+      "step": 5110
+    },
+    {
+      "epoch": 3.825177437429959,
+      "grad_norm": 0.5436882376670837,
+      "learning_rate": 0.0002,
+      "loss": 1.5471,
+      "step": 5120
+    },
+    {
+      "epoch": 3.832648487112439,
+      "grad_norm": 0.5453617572784424,
+      "learning_rate": 0.0002,
+      "loss": 1.5118,
+      "step": 5130
+    },
+    {
+      "epoch": 3.8401195367949197,
+      "grad_norm": 0.6269069314002991,
+      "learning_rate": 0.0002,
+      "loss": 1.5732,
+      "step": 5140
+    },
+    {
+      "epoch": 3.8475905864774003,
+      "grad_norm": 0.6189185380935669,
+      "learning_rate": 0.0002,
+      "loss": 1.4959,
+      "step": 5150
+    },
+    {
+      "epoch": 3.8550616361598804,
+      "grad_norm": 0.6653388142585754,
+      "learning_rate": 0.0002,
+      "loss": 1.4999,
+      "step": 5160
+    },
+    {
+      "epoch": 3.862532685842361,
+      "grad_norm": 0.5771768689155579,
+      "learning_rate": 0.0002,
+      "loss": 1.5075,
+      "step": 5170
+    },
+    {
+      "epoch": 3.870003735524841,
+      "grad_norm": 0.6052790880203247,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5180
+    },
+    {
+      "epoch": 3.8774747852073217,
+      "grad_norm": 0.6572316884994507,
+      "learning_rate": 0.0002,
+      "loss": 1.4987,
+      "step": 5190
+    },
+    {
+      "epoch": 3.884945834889802,
+      "grad_norm": 0.670576810836792,
+      "learning_rate": 0.0002,
+      "loss": 1.5241,
+      "step": 5200
+    },
+    {
+      "epoch": 3.8924168845722824,
+      "grad_norm": 0.5728798508644104,
+      "learning_rate": 0.0002,
+      "loss": 1.4777,
+      "step": 5210
+    },
+    {
+      "epoch": 3.899887934254763,
+      "grad_norm": 0.6340774297714233,
+      "learning_rate": 0.0002,
+      "loss": 1.5351,
+      "step": 5220
+    },
+    {
+      "epoch": 3.907358983937243,
+      "grad_norm": 0.5981315970420837,
+      "learning_rate": 0.0002,
+      "loss": 1.5081,
+      "step": 5230
+    },
+    {
+      "epoch": 3.9148300336197237,
+      "grad_norm": 0.6212025880813599,
+      "learning_rate": 0.0002,
+      "loss": 1.4875,
+      "step": 5240
+    },
+    {
+      "epoch": 3.922301083302204,
+      "grad_norm": 0.6202296018600464,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5250
+    },
+    {
+      "epoch": 3.9297721329846844,
+      "grad_norm": 0.6159142255783081,
+      "learning_rate": 0.0002,
+      "loss": 1.5765,
+      "step": 5260
+    },
+    {
+      "epoch": 3.9372431826671646,
+      "grad_norm": 0.6519438624382019,
+      "learning_rate": 0.0002,
+      "loss": 1.4938,
+      "step": 5270
+    },
+    {
+      "epoch": 3.944714232349645,
+      "grad_norm": 0.539813756942749,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 5280
+    },
+    {
+      "epoch": 3.9521852820321257,
+      "grad_norm": 0.6443665027618408,
+      "learning_rate": 0.0002,
+      "loss": 1.5921,
+      "step": 5290
+    },
+    {
+      "epoch": 3.959656331714606,
+      "grad_norm": 0.6635757684707642,
+      "learning_rate": 0.0002,
+      "loss": 1.5153,
+      "step": 5300
+    },
+    {
+      "epoch": 3.9671273813970864,
+      "grad_norm": 0.589363157749176,
+      "learning_rate": 0.0002,
+      "loss": 1.5485,
+      "step": 5310
+    },
+    {
+      "epoch": 3.9745984310795666,
+      "grad_norm": 0.5788735747337341,
+      "learning_rate": 0.0002,
+      "loss": 1.5498,
+      "step": 5320
+    },
+    {
+      "epoch": 3.982069480762047,
+      "grad_norm": 0.5976864695549011,
+      "learning_rate": 0.0002,
+      "loss": 1.5607,
+      "step": 5330
+    },
+    {
+      "epoch": 3.9895405304445273,
+      "grad_norm": 0.6624067425727844,
+      "learning_rate": 0.0002,
+      "loss": 1.5302,
+      "step": 5340
+    },
+    {
+      "epoch": 3.997011580127008,
+      "grad_norm": 0.6738956570625305,
+      "learning_rate": 0.0002,
+      "loss": 1.5904,
+      "step": 5350
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.868006944656372,
+      "eval_runtime": 38.5153,
+      "eval_samples_per_second": 13.371,
+      "eval_steps_per_second": 1.688,
+      "step": 5354
+    },
+    {
+      "epoch": 4.004482629809488,
+      "grad_norm": 0.6023468971252441,
+      "learning_rate": 0.0002,
+      "loss": 1.4535,
+      "step": 5360
+    },
+    {
+      "epoch": 4.011953679491969,
+      "grad_norm": 0.8589285612106323,
+      "learning_rate": 0.0002,
+      "loss": 1.3987,
+      "step": 5370
+    },
+    {
+      "epoch": 4.019424729174449,
+      "grad_norm": 0.7477491497993469,
+      "learning_rate": 0.0002,
+      "loss": 1.3952,
+      "step": 5380
+    },
+    {
+      "epoch": 4.02689577885693,
+      "grad_norm": 0.7601922154426575,
+      "learning_rate": 0.0002,
+      "loss": 1.3745,
+      "step": 5390
+    },
+    {
+      "epoch": 4.03436682853941,
+      "grad_norm": 0.8115614056587219,
+      "learning_rate": 0.0002,
+      "loss": 1.4133,
+      "step": 5400
+    },
+    {
+      "epoch": 4.04183787822189,
+      "grad_norm": 0.669925332069397,
+      "learning_rate": 0.0002,
+      "loss": 1.3748,
+      "step": 5410
+    },
+    {
+      "epoch": 4.04930892790437,
+      "grad_norm": 0.8091904520988464,
+      "learning_rate": 0.0002,
+      "loss": 1.2835,
+      "step": 5420
+    },
+    {
+      "epoch": 4.056779977586851,
+      "grad_norm": 0.709405779838562,
+      "learning_rate": 0.0002,
+      "loss": 1.3615,
+      "step": 5430
+    },
+    {
+      "epoch": 4.064251027269331,
+      "grad_norm": 1.0006179809570312,
+      "learning_rate": 0.0002,
+      "loss": 1.3558,
+      "step": 5440
+    },
+    {
+      "epoch": 4.071722076951811,
+      "grad_norm": 0.7017965912818909,
+      "learning_rate": 0.0002,
+      "loss": 1.3491,
+      "step": 5450
+    },
+    {
+      "epoch": 4.0791931266342925,
+      "grad_norm": 0.8991572260856628,
+      "learning_rate": 0.0002,
+      "loss": 1.3642,
+      "step": 5460
+    },
+    {
+      "epoch": 4.086664176316773,
+      "grad_norm": 0.9064797759056091,
+      "learning_rate": 0.0002,
+      "loss": 1.392,
+      "step": 5470
+    },
+    {
+      "epoch": 4.094135225999253,
+      "grad_norm": 0.7981749176979065,
+      "learning_rate": 0.0002,
+      "loss": 1.3425,
+      "step": 5480
+    },
+    {
+      "epoch": 4.101606275681733,
+      "grad_norm": 0.7280883193016052,
+      "learning_rate": 0.0002,
+      "loss": 1.3826,
+      "step": 5490
+    },
+    {
+      "epoch": 4.109077325364214,
+      "grad_norm": 0.7419600486755371,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 5500
+    },
+    {
+      "epoch": 4.116548375046694,
+      "grad_norm": 0.8019949197769165,
+      "learning_rate": 0.0002,
+      "loss": 1.3199,
+      "step": 5510
+    },
+    {
+      "epoch": 4.124019424729174,
+      "grad_norm": 0.7501229047775269,
+      "learning_rate": 0.0002,
+      "loss": 1.3133,
+      "step": 5520
+    },
+    {
+      "epoch": 4.131490474411655,
+      "grad_norm": 0.8166249990463257,
+      "learning_rate": 0.0002,
+      "loss": 1.4432,
+      "step": 5530
+    },
+    {
+      "epoch": 4.138961524094135,
+      "grad_norm": 0.9728496074676514,
+      "learning_rate": 0.0002,
+      "loss": 1.3901,
+      "step": 5540
+    },
+    {
+      "epoch": 4.1464325737766154,
+      "grad_norm": 0.7590922117233276,
+      "learning_rate": 0.0002,
+      "loss": 1.3538,
+      "step": 5550
+    },
+    {
+      "epoch": 4.153903623459096,
+      "grad_norm": 0.7759010791778564,
+      "learning_rate": 0.0002,
+      "loss": 1.4368,
+      "step": 5560
+    },
+    {
+      "epoch": 4.161374673141577,
+      "grad_norm": 0.9057986736297607,
+      "learning_rate": 0.0002,
+      "loss": 1.3635,
+      "step": 5570
+    },
+    {
+      "epoch": 4.168845722824057,
+      "grad_norm": 0.8853937983512878,
+      "learning_rate": 0.0002,
+      "loss": 1.4152,
+      "step": 5580
+    },
+    {
+      "epoch": 4.176316772506537,
+      "grad_norm": 0.7070684432983398,
+      "learning_rate": 0.0002,
+      "loss": 1.3633,
+      "step": 5590
+    },
+    {
+      "epoch": 4.183787822189018,
+      "grad_norm": 0.7649410963058472,
+      "learning_rate": 0.0002,
+      "loss": 1.3218,
+      "step": 5600
+    },
+    {
+      "epoch": 4.191258871871498,
+      "grad_norm": 1.2048029899597168,
+      "learning_rate": 0.0002,
+      "loss": 1.3857,
+      "step": 5610
+    },
+    {
+      "epoch": 4.198729921553978,
+      "grad_norm": 0.7986605763435364,
+      "learning_rate": 0.0002,
+      "loss": 1.3629,
+      "step": 5620
+    },
+    {
+      "epoch": 4.206200971236458,
+      "grad_norm": 0.8151885867118835,
+      "learning_rate": 0.0002,
+      "loss": 1.3995,
+      "step": 5630
+    },
+    {
+      "epoch": 4.213672020918939,
+      "grad_norm": 0.7719064354896545,
+      "learning_rate": 0.0002,
+      "loss": 1.3782,
+      "step": 5640
+    },
+    {
+      "epoch": 4.2211430706014195,
+      "grad_norm": 0.8422448039054871,
+      "learning_rate": 0.0002,
+      "loss": 1.3852,
+      "step": 5650
+    },
+    {
+      "epoch": 4.2286141202839,
+      "grad_norm": 0.7017164826393127,
+      "learning_rate": 0.0002,
+      "loss": 1.3321,
+      "step": 5660
+    },
+    {
+      "epoch": 4.236085169966381,
+      "grad_norm": 0.8559677600860596,
+      "learning_rate": 0.0002,
+      "loss": 1.4105,
+      "step": 5670
+    },
+    {
+      "epoch": 4.243556219648861,
+      "grad_norm": 0.8216157555580139,
+      "learning_rate": 0.0002,
+      "loss": 1.3701,
+      "step": 5680
+    },
+    {
+      "epoch": 4.251027269331341,
+      "grad_norm": 0.7681755423545837,
+      "learning_rate": 0.0002,
+      "loss": 1.3565,
+      "step": 5690
+    },
+    {
+      "epoch": 4.258498319013821,
+      "grad_norm": 0.811665952205658,
+      "learning_rate": 0.0002,
+      "loss": 1.3806,
+      "step": 5700
+    },
+    {
+      "epoch": 4.265969368696302,
+      "grad_norm": 0.7242204546928406,
+      "learning_rate": 0.0002,
+      "loss": 1.4161,
+      "step": 5710
+    },
+    {
+      "epoch": 4.273440418378782,
+      "grad_norm": 0.7570181488990784,
+      "learning_rate": 0.0002,
+      "loss": 1.2958,
+      "step": 5720
+    },
+    {
+      "epoch": 4.280911468061262,
+      "grad_norm": 0.8951969146728516,
+      "learning_rate": 0.0002,
+      "loss": 1.4265,
+      "step": 5730
+    },
+    {
+      "epoch": 4.288382517743743,
+      "grad_norm": 0.7222902178764343,
+      "learning_rate": 0.0002,
+      "loss": 1.3895,
+      "step": 5740
+    },
+    {
+      "epoch": 4.2958535674262235,
+      "grad_norm": 0.8508469462394714,
+      "learning_rate": 0.0002,
+      "loss": 1.4155,
+      "step": 5750
+    },
+    {
+      "epoch": 4.303324617108704,
+      "grad_norm": 0.7215430736541748,
+      "learning_rate": 0.0002,
+      "loss": 1.365,
+      "step": 5760
+    },
+    {
+      "epoch": 4.310795666791184,
+      "grad_norm": 0.8774884939193726,
+      "learning_rate": 0.0002,
+      "loss": 1.4472,
+      "step": 5770
+    },
+    {
+      "epoch": 4.318266716473665,
+      "grad_norm": 0.8354552984237671,
+      "learning_rate": 0.0002,
+      "loss": 1.427,
+      "step": 5780
+    },
+    {
+      "epoch": 4.325737766156145,
+      "grad_norm": 0.6938814520835876,
+      "learning_rate": 0.0002,
+      "loss": 1.3222,
+      "step": 5790
+    },
+    {
+      "epoch": 4.333208815838625,
+      "grad_norm": 0.78675377368927,
+      "learning_rate": 0.0002,
+      "loss": 1.3589,
+      "step": 5800
+    },
+    {
+      "epoch": 4.340679865521106,
+      "grad_norm": 0.7147697806358337,
+      "learning_rate": 0.0002,
+      "loss": 1.3662,
+      "step": 5810
+    },
+    {
+      "epoch": 4.348150915203586,
+      "grad_norm": 0.7693623304367065,
+      "learning_rate": 0.0002,
+      "loss": 1.3597,
+      "step": 5820
+    },
+    {
+      "epoch": 4.355621964886066,
+      "grad_norm": 0.856517493724823,
+      "learning_rate": 0.0002,
+      "loss": 1.2944,
+      "step": 5830
+    },
+    {
+      "epoch": 4.3630930145685465,
+      "grad_norm": 0.7200973033905029,
+      "learning_rate": 0.0002,
+      "loss": 1.4307,
+      "step": 5840
+    },
+    {
+      "epoch": 4.3705640642510275,
+      "grad_norm": 0.743281364440918,
+      "learning_rate": 0.0002,
+      "loss": 1.442,
+      "step": 5850
+    },
+    {
+      "epoch": 4.378035113933508,
+      "grad_norm": 0.7627727389335632,
+      "learning_rate": 0.0002,
+      "loss": 1.3999,
+      "step": 5860
+    },
+    {
+      "epoch": 4.385506163615988,
+      "grad_norm": 0.7238836884498596,
+      "learning_rate": 0.0002,
+      "loss": 1.4082,
+      "step": 5870
+    },
+    {
+      "epoch": 4.392977213298469,
+      "grad_norm": 0.7253410816192627,
+      "learning_rate": 0.0002,
+      "loss": 1.4292,
+      "step": 5880
+    },
+    {
+      "epoch": 4.400448262980949,
+      "grad_norm": 0.8232238292694092,
+      "learning_rate": 0.0002,
+      "loss": 1.3774,
+      "step": 5890
+    },
+    {
+      "epoch": 4.407919312663429,
+      "grad_norm": 0.8778504729270935,
+      "learning_rate": 0.0002,
+      "loss": 1.3757,
+      "step": 5900
+    },
+    {
+      "epoch": 4.415390362345909,
+      "grad_norm": 0.7639474868774414,
+      "learning_rate": 0.0002,
+      "loss": 1.387,
+      "step": 5910
+    },
+    {
+      "epoch": 4.42286141202839,
+      "grad_norm": 0.7666519284248352,
+      "learning_rate": 0.0002,
+      "loss": 1.3862,
+      "step": 5920
+    },
+    {
+      "epoch": 4.43033246171087,
+      "grad_norm": 0.867132842540741,
+      "learning_rate": 0.0002,
+      "loss": 1.4168,
+      "step": 5930
+    },
+    {
+      "epoch": 4.4378035113933505,
+      "grad_norm": 0.7571166753768921,
+      "learning_rate": 0.0002,
+      "loss": 1.4772,
+      "step": 5940
+    },
+    {
+      "epoch": 4.4452745610758315,
+      "grad_norm": 0.7911370992660522,
+      "learning_rate": 0.0002,
+      "loss": 1.4401,
+      "step": 5950
+    },
+    {
+      "epoch": 4.452745610758312,
+      "grad_norm": 0.8844250440597534,
+      "learning_rate": 0.0002,
+      "loss": 1.4516,
+      "step": 5960
+    },
+    {
+      "epoch": 4.460216660440792,
+      "grad_norm": 0.7336231470108032,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 5970
+    },
+    {
+      "epoch": 4.467687710123272,
+      "grad_norm": 0.8162738084793091,
+      "learning_rate": 0.0002,
+      "loss": 1.3891,
+      "step": 5980
+    },
+    {
+      "epoch": 4.475158759805753,
+      "grad_norm": 0.7413017153739929,
+      "learning_rate": 0.0002,
+      "loss": 1.393,
+      "step": 5990
+    },
+    {
+      "epoch": 4.482629809488233,
+      "grad_norm": 0.7215432524681091,
+      "learning_rate": 0.0002,
+      "loss": 1.3712,
+      "step": 6000
+    },
+    {
+      "epoch": 4.490100859170713,
+      "grad_norm": 0.8943389058113098,
+      "learning_rate": 0.0002,
+      "loss": 1.3521,
+      "step": 6010
+    },
+    {
+      "epoch": 4.497571908853194,
+      "grad_norm": 0.7850823998451233,
+      "learning_rate": 0.0002,
+      "loss": 1.4172,
+      "step": 6020
+    },
+    {
+      "epoch": 4.505042958535674,
+      "grad_norm": 0.8117504119873047,
+      "learning_rate": 0.0002,
+      "loss": 1.3582,
+      "step": 6030
+    },
+    {
+      "epoch": 4.5125140082181545,
+      "grad_norm": 0.8381605744361877,
+      "learning_rate": 0.0002,
+      "loss": 1.4272,
+      "step": 6040
+    },
+    {
+      "epoch": 4.519985057900635,
+      "grad_norm": 0.7964059710502625,
+      "learning_rate": 0.0002,
+      "loss": 1.3829,
+      "step": 6050
+    },
+    {
+      "epoch": 4.527456107583116,
+      "grad_norm": 0.7935128211975098,
+      "learning_rate": 0.0002,
+      "loss": 1.3555,
+      "step": 6060
+    },
+    {
+      "epoch": 4.534927157265596,
+      "grad_norm": 0.8725124597549438,
+      "learning_rate": 0.0002,
+      "loss": 1.3994,
+      "step": 6070
+    },
+    {
+      "epoch": 4.542398206948076,
+      "grad_norm": 0.880325198173523,
+      "learning_rate": 0.0002,
+      "loss": 1.3923,
+      "step": 6080
+    },
+    {
+      "epoch": 4.549869256630557,
+      "grad_norm": 0.7220637202262878,
+      "learning_rate": 0.0002,
+      "loss": 1.4459,
+      "step": 6090
+    },
+    {
+      "epoch": 4.557340306313037,
+      "grad_norm": 0.6908547878265381,
+      "learning_rate": 0.0002,
+      "loss": 1.3281,
+      "step": 6100
+    },
+    {
+      "epoch": 4.564811355995517,
+      "grad_norm": 0.797931969165802,
+      "learning_rate": 0.0002,
+      "loss": 1.437,
+      "step": 6110
+    },
+    {
+      "epoch": 4.572282405677997,
+      "grad_norm": 0.7056134343147278,
+      "learning_rate": 0.0002,
+      "loss": 1.4023,
+      "step": 6120
+    },
+    {
+      "epoch": 4.579753455360478,
+      "grad_norm": 0.7850478887557983,
+      "learning_rate": 0.0002,
+      "loss": 1.3814,
+      "step": 6130
+    },
+    {
+      "epoch": 4.5872245050429585,
+      "grad_norm": 0.8112621307373047,
+      "learning_rate": 0.0002,
+      "loss": 1.3579,
+      "step": 6140
+    },
+    {
+      "epoch": 4.594695554725439,
+      "grad_norm": 0.7040849328041077,
+      "learning_rate": 0.0002,
+      "loss": 1.3523,
+      "step": 6150
+    },
+    {
+      "epoch": 4.60216660440792,
+      "grad_norm": 0.7214553952217102,
+      "learning_rate": 0.0002,
+      "loss": 1.3526,
+      "step": 6160
+    },
+    {
+      "epoch": 4.6096376540904,
+      "grad_norm": 0.8616511821746826,
+      "learning_rate": 0.0002,
+      "loss": 1.3932,
+      "step": 6170
+    },
+    {
+      "epoch": 4.61710870377288,
+      "grad_norm": 0.8374658226966858,
+      "learning_rate": 0.0002,
+      "loss": 1.4622,
+      "step": 6180
+    },
+    {
+      "epoch": 4.62457975345536,
+      "grad_norm": 0.6761606931686401,
+      "learning_rate": 0.0002,
+      "loss": 1.3703,
+      "step": 6190
+    },
+    {
+      "epoch": 4.632050803137841,
+      "grad_norm": 0.768028199672699,
+      "learning_rate": 0.0002,
+      "loss": 1.3977,
+      "step": 6200
+    },
+    {
+      "epoch": 4.639521852820321,
+      "grad_norm": 0.9372717142105103,
+      "learning_rate": 0.0002,
+      "loss": 1.3772,
+      "step": 6210
+    },
+    {
+      "epoch": 4.646992902502801,
+      "grad_norm": 0.7906546592712402,
+      "learning_rate": 0.0002,
+      "loss": 1.4098,
+      "step": 6220
+    },
+    {
+      "epoch": 4.654463952185282,
+      "grad_norm": 0.7376723289489746,
+      "learning_rate": 0.0002,
+      "loss": 1.3962,
+      "step": 6230
+    },
+    {
+      "epoch": 4.6619350018677626,
+      "grad_norm": 0.8972630500793457,
+      "learning_rate": 0.0002,
+      "loss": 1.4529,
+      "step": 6240
+    },
+    {
+      "epoch": 4.669406051550243,
+      "grad_norm": 0.8261756300926208,
+      "learning_rate": 0.0002,
+      "loss": 1.4668,
+      "step": 6250
+    },
+    {
+      "epoch": 4.676877101232723,
+      "grad_norm": 0.7512393593788147,
+      "learning_rate": 0.0002,
+      "loss": 1.3267,
+      "step": 6260
+    },
+    {
+      "epoch": 4.684348150915204,
+      "grad_norm": 0.7132362127304077,
+      "learning_rate": 0.0002,
+      "loss": 1.4278,
+      "step": 6270
+    },
+    {
+      "epoch": 4.691819200597684,
+      "grad_norm": 0.7690575122833252,
+      "learning_rate": 0.0002,
+      "loss": 1.4299,
+      "step": 6280
+    },
+    {
+      "epoch": 4.699290250280164,
+      "grad_norm": 0.9886258840560913,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 6290
+    },
+    {
+      "epoch": 4.706761299962645,
+      "grad_norm": 0.9502435922622681,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 6300
+    },
+    {
+      "epoch": 4.714232349645125,
+      "grad_norm": 0.702255129814148,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 6310
+    },
+    {
+      "epoch": 4.721703399327605,
+      "grad_norm": 0.7713103890419006,
+      "learning_rate": 0.0002,
+      "loss": 1.4447,
+      "step": 6320
+    },
+    {
+      "epoch": 4.7291744490100855,
+      "grad_norm": 0.7778580784797668,
+      "learning_rate": 0.0002,
+      "loss": 1.4392,
+      "step": 6330
+    },
+    {
+      "epoch": 4.736645498692567,
+      "grad_norm": 0.7275111079216003,
+      "learning_rate": 0.0002,
+      "loss": 1.4169,
+      "step": 6340
+    },
+    {
+      "epoch": 4.744116548375047,
+      "grad_norm": 0.7728744149208069,
+      "learning_rate": 0.0002,
+      "loss": 1.4429,
+      "step": 6350
+    },
+    {
+      "epoch": 4.751587598057527,
+      "grad_norm": 0.9724260568618774,
+      "learning_rate": 0.0002,
+      "loss": 1.3756,
+      "step": 6360
+    },
+    {
+      "epoch": 4.759058647740007,
+      "grad_norm": 0.7505622506141663,
+      "learning_rate": 0.0002,
+      "loss": 1.3358,
+      "step": 6370
+    },
+    {
+      "epoch": 4.766529697422488,
+      "grad_norm": 0.7994682788848877,
+      "learning_rate": 0.0002,
+      "loss": 1.379,
+      "step": 6380
+    },
+    {
+      "epoch": 4.774000747104968,
+      "grad_norm": 0.8432038426399231,
+      "learning_rate": 0.0002,
+      "loss": 1.4275,
+      "step": 6390
+    },
+    {
+      "epoch": 4.781471796787448,
+      "grad_norm": 0.7436022758483887,
+      "learning_rate": 0.0002,
+      "loss": 1.4606,
+      "step": 6400
+    },
+    {
+      "epoch": 4.788942846469929,
+      "grad_norm": 0.7709194421768188,
+      "learning_rate": 0.0002,
+      "loss": 1.3461,
+      "step": 6410
+    },
+    {
+      "epoch": 4.796413896152409,
+      "grad_norm": 0.8798436522483826,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 6420
+    },
+    {
+      "epoch": 4.80388494583489,
+      "grad_norm": 0.790189266204834,
+      "learning_rate": 0.0002,
+      "loss": 1.3761,
+      "step": 6430
+    },
+    {
+      "epoch": 4.811355995517371,
+      "grad_norm": 0.6824303865432739,
+      "learning_rate": 0.0002,
+      "loss": 1.4109,
+      "step": 6440
+    },
+    {
+      "epoch": 4.818827045199851,
+      "grad_norm": 0.7501044869422913,
+      "learning_rate": 0.0002,
+      "loss": 1.3877,
+      "step": 6450
+    },
+    {
+      "epoch": 4.826298094882331,
+      "grad_norm": 0.8840398192405701,
+      "learning_rate": 0.0002,
+      "loss": 1.4458,
+      "step": 6460
+    },
+    {
+      "epoch": 4.833769144564811,
+      "grad_norm": 0.7812688946723938,
+      "learning_rate": 0.0002,
+      "loss": 1.4412,
+      "step": 6470
+    },
+    {
+      "epoch": 4.841240194247292,
+      "grad_norm": 0.7429926991462708,
+      "learning_rate": 0.0002,
+      "loss": 1.4299,
+      "step": 6480
+    },
+    {
+      "epoch": 4.848711243929772,
+      "grad_norm": 0.7778021693229675,
+      "learning_rate": 0.0002,
+      "loss": 1.5062,
+      "step": 6490
+    },
+    {
+      "epoch": 4.856182293612252,
+      "grad_norm": 0.8270702362060547,
+      "learning_rate": 0.0002,
+      "loss": 1.4589,
+      "step": 6500
+    },
+    {
+      "epoch": 4.863653343294732,
+      "grad_norm": 0.6960513591766357,
+      "learning_rate": 0.0002,
+      "loss": 1.4091,
+      "step": 6510
+    },
+    {
+      "epoch": 4.8711243929772134,
+      "grad_norm": 0.7728942632675171,
+      "learning_rate": 0.0002,
+      "loss": 1.376,
+      "step": 6520
+    },
+    {
+      "epoch": 4.878595442659694,
+      "grad_norm": 0.7377303838729858,
+      "learning_rate": 0.0002,
+      "loss": 1.4852,
+      "step": 6530
+    },
+    {
+      "epoch": 4.886066492342174,
+      "grad_norm": 0.7257253527641296,
+      "learning_rate": 0.0002,
+      "loss": 1.3846,
+      "step": 6540
+    },
+    {
+      "epoch": 4.893537542024655,
+      "grad_norm": 0.7875821590423584,
+      "learning_rate": 0.0002,
+      "loss": 1.4166,
+      "step": 6550
+    },
+    {
+      "epoch": 4.901008591707135,
+      "grad_norm": 0.8346304297447205,
+      "learning_rate": 0.0002,
+      "loss": 1.357,
+      "step": 6560
+    },
+    {
+      "epoch": 4.908479641389615,
+      "grad_norm": 0.7710739374160767,
+      "learning_rate": 0.0002,
+      "loss": 1.4522,
+      "step": 6570
+    },
+    {
+      "epoch": 4.915950691072096,
+      "grad_norm": 0.7015138268470764,
+      "learning_rate": 0.0002,
+      "loss": 1.4465,
+      "step": 6580
+    },
+    {
+      "epoch": 4.923421740754576,
+      "grad_norm": 0.8707432150840759,
+      "learning_rate": 0.0002,
+      "loss": 1.435,
+      "step": 6590
+    },
+    {
+      "epoch": 4.930892790437056,
+      "grad_norm": 0.786601185798645,
+      "learning_rate": 0.0002,
+      "loss": 1.2968,
+      "step": 6600
+    },
+    {
+      "epoch": 4.938363840119536,
+      "grad_norm": 0.978519082069397,
+      "learning_rate": 0.0002,
+      "loss": 1.4385,
+      "step": 6610
+    },
+    {
+      "epoch": 4.9458348898020175,
+      "grad_norm": 0.8102927207946777,
+      "learning_rate": 0.0002,
+      "loss": 1.3997,
+      "step": 6620
+    },
+    {
+      "epoch": 4.953305939484498,
+      "grad_norm": 0.7628704309463501,
+      "learning_rate": 0.0002,
+      "loss": 1.4859,
+      "step": 6630
+    },
+    {
+      "epoch": 4.960776989166978,
+      "grad_norm": 0.8053455352783203,
+      "learning_rate": 0.0002,
+      "loss": 1.3774,
+      "step": 6640
+    },
+    {
+      "epoch": 4.968248038849458,
+      "grad_norm": 0.8680412173271179,
+      "learning_rate": 0.0002,
+      "loss": 1.5092,
+      "step": 6650
+    },
+    {
+      "epoch": 4.975719088531939,
+      "grad_norm": 0.7415758371353149,
+      "learning_rate": 0.0002,
+      "loss": 1.3978,
+      "step": 6660
+    },
+    {
+      "epoch": 4.983190138214419,
+      "grad_norm": 0.7730312347412109,
+      "learning_rate": 0.0002,
+      "loss": 1.3793,
+      "step": 6670
+    },
+    {
+      "epoch": 4.990661187896899,
+      "grad_norm": 0.7924041152000427,
+      "learning_rate": 0.0002,
+      "loss": 1.4863,
+      "step": 6680
+    },
+    {
+      "epoch": 4.99813223757938,
+      "grad_norm": 0.8677893877029419,
+      "learning_rate": 0.0002,
+      "loss": 1.4137,
+      "step": 6690
+    },
+    {
+      "epoch": 4.999626447515876,
+      "eval_loss": 1.9444633722305298,
+      "eval_runtime": 39.3488,
+      "eval_samples_per_second": 13.088,
+      "eval_steps_per_second": 1.652,
+      "step": 6692
+    },
+    {
+      "epoch": 5.00560328726186,
+      "grad_norm": 0.7102245092391968,
+      "learning_rate": 0.0002,
+      "loss": 1.3076,
+      "step": 6700
+    },
+    {
+      "epoch": 5.0130743369443405,
+      "grad_norm": 1.0425463914871216,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 6710
+    },
+    {
+      "epoch": 5.0205453866268215,
+      "grad_norm": 0.9320756793022156,
+      "learning_rate": 0.0002,
+      "loss": 1.181,
+      "step": 6720
+    },
+    {
+      "epoch": 5.028016436309302,
+      "grad_norm": 0.8797217607498169,
+      "learning_rate": 0.0002,
+      "loss": 1.1786,
+      "step": 6730
+    },
+    {
+      "epoch": 5.035487485991782,
+      "grad_norm": 2.135707139968872,
+      "learning_rate": 0.0002,
+      "loss": 1.2097,
+      "step": 6740
+    },
+    {
+      "epoch": 5.042958535674262,
+      "grad_norm": 0.8747734427452087,
+      "learning_rate": 0.0002,
+      "loss": 1.1761,
+      "step": 6750
+    },
+    {
+      "epoch": 5.050429585356743,
+      "grad_norm": 0.9981076717376709,
+      "learning_rate": 0.0002,
+      "loss": 1.1675,
+      "step": 6760
+    },
+    {
+      "epoch": 5.057900635039223,
+      "grad_norm": 0.985078752040863,
+      "learning_rate": 0.0002,
+      "loss": 1.1976,
+      "step": 6770
+    },
+    {
+      "epoch": 5.065371684721703,
+      "grad_norm": 1.0974019765853882,
+      "learning_rate": 0.0002,
+      "loss": 1.2688,
+      "step": 6780
+    },
+    {
+      "epoch": 5.072842734404184,
+      "grad_norm": 0.9823219180107117,
+      "learning_rate": 0.0002,
+      "loss": 1.1982,
+      "step": 6790
+    },
+    {
+      "epoch": 5.080313784086664,
+      "grad_norm": 1.122605562210083,
+      "learning_rate": 0.0002,
+      "loss": 1.2586,
+      "step": 6800
+    },
+    {
+      "epoch": 5.0877848337691445,
+      "grad_norm": 0.8556802272796631,
+      "learning_rate": 0.0002,
+      "loss": 1.2069,
+      "step": 6810
+    },
+    {
+      "epoch": 5.095255883451625,
+      "grad_norm": 1.1699262857437134,
+      "learning_rate": 0.0002,
+      "loss": 1.1908,
+      "step": 6820
+    },
+    {
+      "epoch": 5.102726933134106,
+      "grad_norm": 1.0440590381622314,
+      "learning_rate": 0.0002,
+      "loss": 1.1869,
+      "step": 6830
+    },
+    {
+      "epoch": 5.110197982816586,
+      "grad_norm": 1.0445300340652466,
+      "learning_rate": 0.0002,
+      "loss": 1.1655,
+      "step": 6840
+    },
+    {
+      "epoch": 5.117669032499066,
+      "grad_norm": 0.8289563059806824,
+      "learning_rate": 0.0002,
+      "loss": 1.2392,
+      "step": 6850
+    },
+    {
+      "epoch": 5.125140082181547,
+      "grad_norm": 1.1051193475723267,
+      "learning_rate": 0.0002,
+      "loss": 1.1687,
+      "step": 6860
+    },
+    {
+      "epoch": 5.132611131864027,
+      "grad_norm": 0.9345614910125732,
+      "learning_rate": 0.0002,
+      "loss": 1.2737,
+      "step": 6870
+    },
+    {
+      "epoch": 5.140082181546507,
+      "grad_norm": 1.1222996711730957,
+      "learning_rate": 0.0002,
+      "loss": 1.3021,
+      "step": 6880
+    },
+    {
+      "epoch": 5.147553231228987,
+      "grad_norm": 0.9405338764190674,
+      "learning_rate": 0.0002,
+      "loss": 1.2408,
+      "step": 6890
+    },
+    {
+      "epoch": 5.155024280911468,
+      "grad_norm": 1.0935171842575073,
+      "learning_rate": 0.0002,
+      "loss": 1.2367,
+      "step": 6900
+    },
+    {
+      "epoch": 5.1624953305939485,
+      "grad_norm": 1.0438612699508667,
+      "learning_rate": 0.0002,
+      "loss": 1.2458,
+      "step": 6910
+    },
+    {
+      "epoch": 5.169966380276429,
+      "grad_norm": 1.1189004182815552,
+      "learning_rate": 0.0002,
+      "loss": 1.2562,
+      "step": 6920
+    },
+    {
+      "epoch": 5.17743742995891,
+      "grad_norm": 1.0533215999603271,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 6930
+    },
+    {
+      "epoch": 5.18490847964139,
+      "grad_norm": 0.9779648780822754,
+      "learning_rate": 0.0002,
+      "loss": 1.2974,
+      "step": 6940
+    },
+    {
+      "epoch": 5.19237952932387,
+      "grad_norm": 0.8920868635177612,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 6950
+    },
+    {
+      "epoch": 5.19985057900635,
+      "grad_norm": 0.8374548554420471,
+      "learning_rate": 0.0002,
+      "loss": 1.283,
+      "step": 6960
+    },
+    {
+      "epoch": 5.207321628688831,
+      "grad_norm": 1.0490682125091553,
+      "learning_rate": 0.0002,
+      "loss": 1.2775,
+      "step": 6970
+    },
+    {
+      "epoch": 5.214792678371311,
+      "grad_norm": 0.9658287167549133,
+      "learning_rate": 0.0002,
+      "loss": 1.1826,
+      "step": 6980
+    },
+    {
+      "epoch": 5.222263728053791,
+      "grad_norm": 0.9652056097984314,
+      "learning_rate": 0.0002,
+      "loss": 1.2647,
+      "step": 6990
+    },
+    {
+      "epoch": 5.229734777736272,
+      "grad_norm": 0.9141794443130493,
+      "learning_rate": 0.0002,
+      "loss": 1.3023,
+      "step": 7000
+    },
+    {
+      "epoch": 5.2372058274187525,
+      "grad_norm": 0.9831376671791077,
+      "learning_rate": 0.0002,
+      "loss": 1.2456,
+      "step": 7010
+    },
+    {
+      "epoch": 5.244676877101233,
+      "grad_norm": 1.0198718309402466,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 7020
+    },
+    {
+      "epoch": 5.252147926783713,
+      "grad_norm": 0.9647888541221619,
+      "learning_rate": 0.0002,
+      "loss": 1.2643,
+      "step": 7030
+    },
+    {
+      "epoch": 5.259618976466194,
+      "grad_norm": 1.3941649198532104,
+      "learning_rate": 0.0002,
+      "loss": 1.2106,
+      "step": 7040
+    },
+    {
+      "epoch": 5.267090026148674,
+      "grad_norm": 1.0305466651916504,
+      "learning_rate": 0.0002,
+      "loss": 1.2885,
+      "step": 7050
+    },
+    {
+      "epoch": 5.274561075831154,
+      "grad_norm": 0.9577859044075012,
+      "learning_rate": 0.0002,
+      "loss": 1.2362,
+      "step": 7060
+    },
+    {
+      "epoch": 5.282032125513634,
+      "grad_norm": 1.149092197418213,
+      "learning_rate": 0.0002,
+      "loss": 1.2231,
+      "step": 7070
+    },
+    {
+      "epoch": 5.289503175196115,
+      "grad_norm": 1.2582733631134033,
+      "learning_rate": 0.0002,
+      "loss": 1.2986,
+      "step": 7080
+    },
+    {
+      "epoch": 5.296974224878595,
+      "grad_norm": 1.1777442693710327,
+      "learning_rate": 0.0002,
+      "loss": 1.2307,
+      "step": 7090
+    },
+    {
+      "epoch": 5.3044452745610755,
+      "grad_norm": 1.0076404809951782,
+      "learning_rate": 0.0002,
+      "loss": 1.24,
+      "step": 7100
+    },
+    {
+      "epoch": 5.3119163242435565,
+      "grad_norm": 0.9037365913391113,
+      "learning_rate": 0.0002,
+      "loss": 1.1407,
+      "step": 7110
+    },
+    {
+      "epoch": 5.319387373926037,
+      "grad_norm": 0.9428724646568298,
+      "learning_rate": 0.0002,
+      "loss": 1.238,
+      "step": 7120
+    },
+    {
+      "epoch": 5.326858423608517,
+      "grad_norm": 0.9935154318809509,
+      "learning_rate": 0.0002,
+      "loss": 1.2571,
+      "step": 7130
+    },
+    {
+      "epoch": 5.334329473290998,
+      "grad_norm": 1.087500810623169,
+      "learning_rate": 0.0002,
+      "loss": 1.2833,
+      "step": 7140
+    },
+    {
+      "epoch": 5.341800522973478,
+      "grad_norm": 0.8543072938919067,
+      "learning_rate": 0.0002,
+      "loss": 1.2304,
+      "step": 7150
+    },
+    {
+      "epoch": 5.349271572655958,
+      "grad_norm": 0.9323700070381165,
+      "learning_rate": 0.0002,
+      "loss": 1.2755,
+      "step": 7160
+    },
+    {
+      "epoch": 5.356742622338438,
+      "grad_norm": 1.0037827491760254,
+      "learning_rate": 0.0002,
+      "loss": 1.2769,
+      "step": 7170
+    },
+    {
+      "epoch": 5.364213672020919,
+      "grad_norm": 0.8746469616889954,
+      "learning_rate": 0.0002,
+      "loss": 1.3204,
+      "step": 7180
+    },
+    {
+      "epoch": 5.371684721703399,
+      "grad_norm": 0.9516328573226929,
+      "learning_rate": 0.0002,
+      "loss": 1.2759,
+      "step": 7190
+    },
+    {
+      "epoch": 5.3791557713858795,
+      "grad_norm": 0.9395177364349365,
+      "learning_rate": 0.0002,
+      "loss": 1.2428,
+      "step": 7200
+    },
+    {
+      "epoch": 5.38662682106836,
+      "grad_norm": 1.000369906425476,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 7210
+    },
+    {
+      "epoch": 5.394097870750841,
+      "grad_norm": 1.0845502614974976,
+      "learning_rate": 0.0002,
+      "loss": 1.2337,
+      "step": 7220
+    },
+    {
+      "epoch": 5.401568920433321,
+      "grad_norm": 0.8975145220756531,
+      "learning_rate": 0.0002,
+      "loss": 1.2776,
+      "step": 7230
+    },
+    {
+      "epoch": 5.409039970115801,
+      "grad_norm": 1.040077805519104,
+      "learning_rate": 0.0002,
+      "loss": 1.2306,
+      "step": 7240
+    },
+    {
+      "epoch": 5.416511019798282,
+      "grad_norm": 1.0729942321777344,
+      "learning_rate": 0.0002,
+      "loss": 1.2277,
+      "step": 7250
+    },
+    {
+      "epoch": 5.423982069480762,
+      "grad_norm": 0.8322232961654663,
+      "learning_rate": 0.0002,
+      "loss": 1.2714,
+      "step": 7260
+    },
+    {
+      "epoch": 5.431453119163242,
+      "grad_norm": 1.0654641389846802,
+      "learning_rate": 0.0002,
+      "loss": 1.3036,
+      "step": 7270
+    },
+    {
+      "epoch": 5.438924168845723,
+      "grad_norm": 1.0445852279663086,
+      "learning_rate": 0.0002,
+      "loss": 1.268,
+      "step": 7280
+    },
+    {
+      "epoch": 5.446395218528203,
+      "grad_norm": 1.0762956142425537,
+      "learning_rate": 0.0002,
+      "loss": 1.2743,
+      "step": 7290
+    },
+    {
+      "epoch": 5.4538662682106835,
+      "grad_norm": 0.9721953868865967,
+      "learning_rate": 0.0002,
+      "loss": 1.2887,
+      "step": 7300
+    },
+    {
+      "epoch": 5.461337317893164,
+      "grad_norm": 0.9238539338111877,
+      "learning_rate": 0.0002,
+      "loss": 1.2833,
+      "step": 7310
+    },
+    {
+      "epoch": 5.468808367575645,
+      "grad_norm": 0.9912874102592468,
+      "learning_rate": 0.0002,
+      "loss": 1.255,
+      "step": 7320
+    },
+    {
+      "epoch": 5.476279417258125,
+      "grad_norm": 1.0727077722549438,
+      "learning_rate": 0.0002,
+      "loss": 1.2557,
+      "step": 7330
+    },
+    {
+      "epoch": 5.483750466940605,
+      "grad_norm": 0.8633865118026733,
+      "learning_rate": 0.0002,
+      "loss": 1.3471,
+      "step": 7340
+    },
+    {
+      "epoch": 5.491221516623085,
+      "grad_norm": 0.9396262764930725,
+      "learning_rate": 0.0002,
+      "loss": 1.3155,
+      "step": 7350
+    },
+    {
+      "epoch": 5.498692566305566,
+      "grad_norm": 1.0253715515136719,
+      "learning_rate": 0.0002,
+      "loss": 1.3146,
+      "step": 7360
+    },
+    {
+      "epoch": 5.506163615988046,
+      "grad_norm": 1.006047010421753,
+      "learning_rate": 0.0002,
+      "loss": 1.3156,
+      "step": 7370
+    },
+    {
+      "epoch": 5.513634665670526,
+      "grad_norm": 0.9781233072280884,
+      "learning_rate": 0.0002,
+      "loss": 1.3107,
+      "step": 7380
+    },
+    {
+      "epoch": 5.521105715353007,
+      "grad_norm": 0.9945126175880432,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 7390
+    },
+    {
+      "epoch": 5.528576765035488,
+      "grad_norm": 0.9081175327301025,
+      "learning_rate": 0.0002,
+      "loss": 1.1936,
+      "step": 7400
+    },
+    {
+      "epoch": 5.536047814717968,
+      "grad_norm": 1.2215938568115234,
+      "learning_rate": 0.0002,
+      "loss": 1.2651,
+      "step": 7410
+    },
+    {
+      "epoch": 5.543518864400449,
+      "grad_norm": 1.0724077224731445,
+      "learning_rate": 0.0002,
+      "loss": 1.2484,
+      "step": 7420
+    },
+    {
+      "epoch": 5.550989914082929,
+      "grad_norm": 1.106955885887146,
+      "learning_rate": 0.0002,
+      "loss": 1.3083,
+      "step": 7430
+    },
+    {
+      "epoch": 5.558460963765409,
+      "grad_norm": 1.0657650232315063,
+      "learning_rate": 0.0002,
+      "loss": 1.2125,
+      "step": 7440
+    },
+    {
+      "epoch": 5.565932013447889,
+      "grad_norm": 0.9725455641746521,
+      "learning_rate": 0.0002,
+      "loss": 1.2576,
+      "step": 7450
+    },
+    {
+      "epoch": 5.57340306313037,
+      "grad_norm": 0.8604224324226379,
+      "learning_rate": 0.0002,
+      "loss": 1.3297,
+      "step": 7460
+    },
+    {
+      "epoch": 5.58087411281285,
+      "grad_norm": 0.9913371205329895,
+      "learning_rate": 0.0002,
+      "loss": 1.3084,
+      "step": 7470
+    },
+    {
+      "epoch": 5.58834516249533,
+      "grad_norm": 1.012073040008545,
+      "learning_rate": 0.0002,
+      "loss": 1.3371,
+      "step": 7480
+    },
+    {
+      "epoch": 5.5958162121778106,
+      "grad_norm": 1.1003159284591675,
+      "learning_rate": 0.0002,
+      "loss": 1.2526,
+      "step": 7490
+    },
+    {
+      "epoch": 5.603287261860292,
+      "grad_norm": 0.9104593992233276,
+      "learning_rate": 0.0002,
+      "loss": 1.2577,
+      "step": 7500
+    },
+    {
+      "epoch": 5.610758311542772,
+      "grad_norm": 0.9480831623077393,
+      "learning_rate": 0.0002,
+      "loss": 1.2578,
+      "step": 7510
+    },
+    {
+      "epoch": 5.618229361225252,
+      "grad_norm": 1.0826456546783447,
+      "learning_rate": 0.0002,
+      "loss": 1.3056,
+      "step": 7520
+    },
+    {
+      "epoch": 5.625700410907733,
+      "grad_norm": 0.8286259174346924,
+      "learning_rate": 0.0002,
+      "loss": 1.2931,
+      "step": 7530
+    },
+    {
+      "epoch": 5.633171460590213,
+      "grad_norm": 0.9145061373710632,
+      "learning_rate": 0.0002,
+      "loss": 1.2918,
+      "step": 7540
+    },
+    {
+      "epoch": 5.640642510272693,
+      "grad_norm": 0.9363601803779602,
+      "learning_rate": 0.0002,
+      "loss": 1.1736,
+      "step": 7550
+    },
+    {
+      "epoch": 5.648113559955174,
+      "grad_norm": 0.9553244709968567,
+      "learning_rate": 0.0002,
+      "loss": 1.2265,
+      "step": 7560
+    },
+    {
+      "epoch": 5.655584609637654,
+      "grad_norm": 1.0343557596206665,
+      "learning_rate": 0.0002,
+      "loss": 1.2356,
+      "step": 7570
+    },
+    {
+      "epoch": 5.663055659320134,
+      "grad_norm": 0.8734238743782043,
+      "learning_rate": 0.0002,
+      "loss": 1.3171,
+      "step": 7580
+    },
+    {
+      "epoch": 5.670526709002615,
+      "grad_norm": 1.0230586528778076,
+      "learning_rate": 0.0002,
+      "loss": 1.2785,
+      "step": 7590
+    },
+    {
+      "epoch": 5.677997758685096,
+      "grad_norm": 1.0063409805297852,
+      "learning_rate": 0.0002,
+      "loss": 1.2936,
+      "step": 7600
+    },
+    {
+      "epoch": 5.685468808367576,
+      "grad_norm": 1.0104626417160034,
+      "learning_rate": 0.0002,
+      "loss": 1.2396,
+      "step": 7610
+    },
+    {
+      "epoch": 5.692939858050056,
+      "grad_norm": 0.9528168439865112,
+      "learning_rate": 0.0002,
+      "loss": 1.2581,
+      "step": 7620
+    },
+    {
+      "epoch": 5.700410907732536,
+      "grad_norm": 0.9799878597259521,
+      "learning_rate": 0.0002,
+      "loss": 1.3116,
+      "step": 7630
+    },
+    {
+      "epoch": 5.707881957415017,
+      "grad_norm": 0.969351589679718,
+      "learning_rate": 0.0002,
+      "loss": 1.2632,
+      "step": 7640
+    },
+    {
+      "epoch": 5.715353007097497,
+      "grad_norm": 1.3037652969360352,
+      "learning_rate": 0.0002,
+      "loss": 1.3055,
+      "step": 7650
+    },
+    {
+      "epoch": 5.722824056779977,
+      "grad_norm": 1.0640486478805542,
+      "learning_rate": 0.0002,
+      "loss": 1.3126,
+      "step": 7660
+    },
+    {
+      "epoch": 5.730295106462458,
+      "grad_norm": 1.0416420698165894,
+      "learning_rate": 0.0002,
+      "loss": 1.3325,
+      "step": 7670
+    },
+    {
+      "epoch": 5.7377661561449385,
+      "grad_norm": 0.8893619775772095,
+      "learning_rate": 0.0002,
+      "loss": 1.25,
+      "step": 7680
+    },
+    {
+      "epoch": 5.745237205827419,
+      "grad_norm": 0.8512844443321228,
+      "learning_rate": 0.0002,
+      "loss": 1.319,
+      "step": 7690
+    },
+    {
+      "epoch": 5.7527082555099,
+      "grad_norm": 0.9955748319625854,
+      "learning_rate": 0.0002,
+      "loss": 1.3328,
+      "step": 7700
+    },
+    {
+      "epoch": 5.76017930519238,
+      "grad_norm": 1.0409910678863525,
+      "learning_rate": 0.0002,
+      "loss": 1.294,
+      "step": 7710
+    },
+    {
+      "epoch": 5.76765035487486,
+      "grad_norm": 1.010097861289978,
+      "learning_rate": 0.0002,
+      "loss": 1.3518,
+      "step": 7720
+    },
+    {
+      "epoch": 5.77512140455734,
+      "grad_norm": 0.8974892497062683,
+      "learning_rate": 0.0002,
+      "loss": 1.2106,
+      "step": 7730
+    },
+    {
+      "epoch": 5.782592454239821,
+      "grad_norm": 0.972835123538971,
+      "learning_rate": 0.0002,
+      "loss": 1.2743,
+      "step": 7740
+    },
+    {
+      "epoch": 5.790063503922301,
+      "grad_norm": 0.9607440829277039,
+      "learning_rate": 0.0002,
+      "loss": 1.3549,
+      "step": 7750
+    },
+    {
+      "epoch": 5.797534553604781,
+      "grad_norm": 0.9426500797271729,
+      "learning_rate": 0.0002,
+      "loss": 1.29,
+      "step": 7760
+    },
+    {
+      "epoch": 5.8050056032872615,
+      "grad_norm": 0.8745320439338684,
+      "learning_rate": 0.0002,
+      "loss": 1.274,
+      "step": 7770
+    },
+    {
+      "epoch": 5.8124766529697425,
+      "grad_norm": 1.0117204189300537,
+      "learning_rate": 0.0002,
+      "loss": 1.3009,
+      "step": 7780
+    },
+    {
+      "epoch": 5.819947702652223,
+      "grad_norm": 1.0387755632400513,
+      "learning_rate": 0.0002,
+      "loss": 1.3135,
+      "step": 7790
+    },
+    {
+      "epoch": 5.827418752334703,
+      "grad_norm": 1.0709784030914307,
+      "learning_rate": 0.0002,
+      "loss": 1.2709,
+      "step": 7800
+    },
+    {
+      "epoch": 5.834889802017184,
+      "grad_norm": 0.9512667655944824,
+      "learning_rate": 0.0002,
+      "loss": 1.225,
+      "step": 7810
+    },
+    {
+      "epoch": 5.842360851699664,
+      "grad_norm": 1.021094560623169,
+      "learning_rate": 0.0002,
+      "loss": 1.3284,
+      "step": 7820
+    },
+    {
+      "epoch": 5.849831901382144,
+      "grad_norm": 1.117491364479065,
+      "learning_rate": 0.0002,
+      "loss": 1.2794,
+      "step": 7830
+    },
+    {
+      "epoch": 5.857302951064625,
+      "grad_norm": 0.9252554178237915,
+      "learning_rate": 0.0002,
+      "loss": 1.3646,
+      "step": 7840
+    },
+    {
+      "epoch": 5.864774000747105,
+      "grad_norm": 1.1416207551956177,
+      "learning_rate": 0.0002,
+      "loss": 1.2976,
+      "step": 7850
+    },
+    {
+      "epoch": 5.872245050429585,
+      "grad_norm": 1.1219907999038696,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 7860
+    },
+    {
+      "epoch": 5.8797161001120655,
+      "grad_norm": 0.8300467729568481,
+      "learning_rate": 0.0002,
+      "loss": 1.2334,
+      "step": 7870
+    },
+    {
+      "epoch": 5.8871871497945465,
+      "grad_norm": 1.00551438331604,
+      "learning_rate": 0.0002,
+      "loss": 1.3132,
+      "step": 7880
+    },
+    {
+      "epoch": 5.894658199477027,
+      "grad_norm": 0.8981153964996338,
+      "learning_rate": 0.0002,
+      "loss": 1.2609,
+      "step": 7890
+    },
+    {
+      "epoch": 5.902129249159507,
+      "grad_norm": 1.0247976779937744,
+      "learning_rate": 0.0002,
+      "loss": 1.2817,
+      "step": 7900
+    },
+    {
+      "epoch": 5.909600298841987,
+      "grad_norm": 1.0820319652557373,
+      "learning_rate": 0.0002,
+      "loss": 1.2866,
+      "step": 7910
+    },
+    {
+      "epoch": 5.917071348524468,
+      "grad_norm": 0.952675461769104,
+      "learning_rate": 0.0002,
+      "loss": 1.2941,
+      "step": 7920
+    },
+    {
+      "epoch": 5.924542398206948,
+      "grad_norm": 0.8666740655899048,
+      "learning_rate": 0.0002,
+      "loss": 1.307,
+      "step": 7930
+    },
+    {
+      "epoch": 5.932013447889428,
+      "grad_norm": 0.8640421032905579,
+      "learning_rate": 0.0002,
+      "loss": 1.2752,
+      "step": 7940
+    },
+    {
+      "epoch": 5.939484497571909,
+      "grad_norm": 1.2343276739120483,
+      "learning_rate": 0.0002,
+      "loss": 1.2386,
+      "step": 7950
+    },
+    {
+      "epoch": 5.946955547254389,
+      "grad_norm": 0.958046555519104,
+      "learning_rate": 0.0002,
+      "loss": 1.2333,
+      "step": 7960
+    },
+    {
+      "epoch": 5.9544265969368695,
+      "grad_norm": 1.0538510084152222,
+      "learning_rate": 0.0002,
+      "loss": 1.2352,
+      "step": 7970
+    },
+    {
+      "epoch": 5.9618976466193505,
+      "grad_norm": 1.2681571245193481,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 7980
+    },
+    {
+      "epoch": 5.969368696301831,
+      "grad_norm": 0.8171183466911316,
+      "learning_rate": 0.0002,
+      "loss": 1.2514,
+      "step": 7990
+    },
+    {
+      "epoch": 5.976839745984311,
+      "grad_norm": 0.9109523892402649,
+      "learning_rate": 0.0002,
+      "loss": 1.3412,
+      "step": 8000
+    },
+    {
+      "epoch": 5.984310795666791,
+      "grad_norm": 1.0040639638900757,
+      "learning_rate": 0.0002,
+      "loss": 1.3497,
+      "step": 8010
+    },
+    {
+      "epoch": 5.991781845349272,
+      "grad_norm": 0.9596554040908813,
+      "learning_rate": 0.0002,
+      "loss": 1.3299,
+      "step": 8020
+    },
+    {
+      "epoch": 5.999252895031752,
+      "grad_norm": 0.9782963991165161,
+      "learning_rate": 0.0002,
+      "loss": 1.3109,
+      "step": 8030
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.0417845249176025,
+      "eval_runtime": 38.8465,
+      "eval_samples_per_second": 13.257,
+      "eval_steps_per_second": 1.673,
+      "step": 8031
+    },
+    {
+      "epoch": 6.006723944714232,
+      "grad_norm": 1.380823016166687,
+      "learning_rate": 0.0002,
+      "loss": 1.0886,
+      "step": 8040
+    },
+    {
+      "epoch": 6.014194994396712,
+      "grad_norm": 1.067636251449585,
+      "learning_rate": 0.0002,
+      "loss": 1.0413,
+      "step": 8050
+    },
+    {
+      "epoch": 6.021666044079193,
+      "grad_norm": 1.363402009010315,
+      "learning_rate": 0.0002,
+      "loss": 1.0686,
+      "step": 8060
+    },
+    {
+      "epoch": 6.0291370937616735,
+      "grad_norm": 0.9901054501533508,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 8070
+    },
+    {
+      "epoch": 6.036608143444154,
+      "grad_norm": 1.1545379161834717,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 8080
+    },
+    {
+      "epoch": 6.044079193126635,
+      "grad_norm": 1.2259265184402466,
+      "learning_rate": 0.0002,
+      "loss": 1.0644,
+      "step": 8090
+    },
+    {
+      "epoch": 6.051550242809115,
+      "grad_norm": 1.1237425804138184,
+      "learning_rate": 0.0002,
+      "loss": 1.1273,
+      "step": 8100
+    },
+    {
+      "epoch": 6.059021292491595,
+      "grad_norm": 1.2805622816085815,
+      "learning_rate": 0.0002,
+      "loss": 1.1001,
+      "step": 8110
+    },
+    {
+      "epoch": 6.066492342174075,
+      "grad_norm": 1.2270452976226807,
+      "learning_rate": 0.0002,
+      "loss": 1.0731,
+      "step": 8120
+    },
+    {
+      "epoch": 6.073963391856556,
+      "grad_norm": 1.1924101114273071,
+      "learning_rate": 0.0002,
+      "loss": 1.0692,
+      "step": 8130
+    },
+    {
+      "epoch": 6.081434441539036,
+      "grad_norm": 1.2543894052505493,
+      "learning_rate": 0.0002,
+      "loss": 1.1698,
+      "step": 8140
+    },
+    {
+      "epoch": 6.088905491221516,
+      "grad_norm": 1.1821149587631226,
+      "learning_rate": 0.0002,
+      "loss": 1.069,
+      "step": 8150
+    },
+    {
+      "epoch": 6.096376540903997,
+      "grad_norm": 1.2202836275100708,
+      "learning_rate": 0.0002,
+      "loss": 1.109,
+      "step": 8160
+    },
+    {
+      "epoch": 6.1038475905864775,
+      "grad_norm": 1.0576019287109375,
+      "learning_rate": 0.0002,
+      "loss": 1.136,
+      "step": 8170
+    },
+    {
+      "epoch": 6.111318640268958,
+      "grad_norm": 1.31708824634552,
+      "learning_rate": 0.0002,
+      "loss": 1.1395,
+      "step": 8180
+    },
+    {
+      "epoch": 6.118789689951438,
+      "grad_norm": 1.0479495525360107,
+      "learning_rate": 0.0002,
+      "loss": 1.0887,
+      "step": 8190
+    },
+    {
+      "epoch": 6.126260739633919,
+      "grad_norm": 1.285003423690796,
+      "learning_rate": 0.0002,
+      "loss": 1.0764,
+      "step": 8200
+    },
+    {
+      "epoch": 6.133731789316399,
+      "grad_norm": 1.0989165306091309,
+      "learning_rate": 0.0002,
+      "loss": 1.0642,
+      "step": 8210
+    },
+    {
+      "epoch": 6.141202838998879,
+      "grad_norm": 1.1659013032913208,
+      "learning_rate": 0.0002,
+      "loss": 1.0981,
+      "step": 8220
+    },
+    {
+      "epoch": 6.14867388868136,
+      "grad_norm": 1.2796376943588257,
+      "learning_rate": 0.0002,
+      "loss": 1.1138,
+      "step": 8230
+    },
+    {
+      "epoch": 6.15614493836384,
+      "grad_norm": 1.060564637184143,
+      "learning_rate": 0.0002,
+      "loss": 1.1116,
+      "step": 8240
+    },
+    {
+      "epoch": 6.16361598804632,
+      "grad_norm": 1.3884605169296265,
+      "learning_rate": 0.0002,
+      "loss": 1.1493,
+      "step": 8250
+    },
+    {
+      "epoch": 6.1710870377288005,
+      "grad_norm": 1.1570569276809692,
+      "learning_rate": 0.0002,
+      "loss": 1.0504,
+      "step": 8260
+    },
+    {
+      "epoch": 6.1785580874112815,
+      "grad_norm": 1.4136502742767334,
+      "learning_rate": 0.0002,
+      "loss": 1.0386,
+      "step": 8270
+    },
+    {
+      "epoch": 6.186029137093762,
+      "grad_norm": 1.3396095037460327,
+      "learning_rate": 0.0002,
+      "loss": 1.0882,
+      "step": 8280
+    },
+    {
+      "epoch": 6.193500186776242,
+      "grad_norm": 1.2549997568130493,
+      "learning_rate": 0.0002,
+      "loss": 1.133,
+      "step": 8290
+    },
+    {
+      "epoch": 6.200971236458723,
+      "grad_norm": 1.3629751205444336,
+      "learning_rate": 0.0002,
+      "loss": 1.0626,
+      "step": 8300
+    },
+    {
+      "epoch": 6.208442286141203,
+      "grad_norm": 1.1029163599014282,
+      "learning_rate": 0.0002,
+      "loss": 1.1343,
+      "step": 8310
+    },
+    {
+      "epoch": 6.215913335823683,
+      "grad_norm": 1.1992450952529907,
+      "learning_rate": 0.0002,
+      "loss": 1.0895,
+      "step": 8320
+    },
+    {
+      "epoch": 6.223384385506163,
+      "grad_norm": 1.3317986726760864,
+      "learning_rate": 0.0002,
+      "loss": 1.1417,
+      "step": 8330
+    },
+    {
+      "epoch": 6.230855435188644,
+      "grad_norm": 1.0538336038589478,
+      "learning_rate": 0.0002,
+      "loss": 1.0958,
+      "step": 8340
+    },
+    {
+      "epoch": 6.238326484871124,
+      "grad_norm": 1.1767704486846924,
+      "learning_rate": 0.0002,
+      "loss": 1.1557,
+      "step": 8350
+    },
+    {
+      "epoch": 6.2457975345536045,
+      "grad_norm": 1.1213016510009766,
+      "learning_rate": 0.0002,
+      "loss": 1.1038,
+      "step": 8360
+    },
+    {
+      "epoch": 6.253268584236086,
+      "grad_norm": 1.1895716190338135,
+      "learning_rate": 0.0002,
+      "loss": 1.1241,
+      "step": 8370
+    },
+    {
+      "epoch": 6.260739633918566,
+      "grad_norm": 1.1078153848648071,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 8380
+    },
+    {
+      "epoch": 6.268210683601046,
+      "grad_norm": 1.1662801504135132,
+      "learning_rate": 0.0002,
+      "loss": 1.1124,
+      "step": 8390
+    },
+    {
+      "epoch": 6.275681733283526,
+      "grad_norm": 1.2071197032928467,
+      "learning_rate": 0.0002,
+      "loss": 1.125,
+      "step": 8400
+    },
+    {
+      "epoch": 6.283152782966007,
+      "grad_norm": 1.2653778791427612,
+      "learning_rate": 0.0002,
+      "loss": 1.0625,
+      "step": 8410
+    },
+    {
+      "epoch": 6.290623832648487,
+      "grad_norm": 1.6128872632980347,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 8420
+    },
+    {
+      "epoch": 6.298094882330967,
+      "grad_norm": 1.4993070363998413,
+      "learning_rate": 0.0002,
+      "loss": 1.1212,
+      "step": 8430
+    },
+    {
+      "epoch": 6.305565932013448,
+      "grad_norm": 1.16339910030365,
+      "learning_rate": 0.0002,
+      "loss": 1.1516,
+      "step": 8440
+    },
+    {
+      "epoch": 6.313036981695928,
+      "grad_norm": 1.256822943687439,
+      "learning_rate": 0.0002,
+      "loss": 1.0662,
+      "step": 8450
+    },
+    {
+      "epoch": 6.3205080313784086,
+      "grad_norm": 1.1352964639663696,
+      "learning_rate": 0.0002,
+      "loss": 1.1566,
+      "step": 8460
+    },
+    {
+      "epoch": 6.327979081060889,
+      "grad_norm": 1.0061070919036865,
+      "learning_rate": 0.0002,
+      "loss": 1.1297,
+      "step": 8470
+    },
+    {
+      "epoch": 6.33545013074337,
+      "grad_norm": 1.1901768445968628,
+      "learning_rate": 0.0002,
+      "loss": 1.0967,
+      "step": 8480
+    },
+    {
+      "epoch": 6.34292118042585,
+      "grad_norm": 1.2715139389038086,
+      "learning_rate": 0.0002,
+      "loss": 1.1463,
+      "step": 8490
+    },
+    {
+      "epoch": 6.35039223010833,
+      "grad_norm": 1.1583346128463745,
+      "learning_rate": 0.0002,
+      "loss": 1.2143,
+      "step": 8500
+    },
+    {
+      "epoch": 6.357863279790811,
+      "grad_norm": 1.1427477598190308,
+      "learning_rate": 0.0002,
+      "loss": 1.1072,
+      "step": 8510
+    },
+    {
+      "epoch": 6.365334329473291,
+      "grad_norm": 1.1952263116836548,
+      "learning_rate": 0.0002,
+      "loss": 1.1119,
+      "step": 8520
+    },
+    {
+      "epoch": 6.372805379155771,
+      "grad_norm": 1.0599623918533325,
+      "learning_rate": 0.0002,
+      "loss": 1.0797,
+      "step": 8530
+    },
+    {
+      "epoch": 6.380276428838251,
+      "grad_norm": 1.3511574268341064,
+      "learning_rate": 0.0002,
+      "loss": 1.1091,
+      "step": 8540
+    },
+    {
+      "epoch": 6.387747478520732,
+      "grad_norm": 1.171126127243042,
+      "learning_rate": 0.0002,
+      "loss": 1.1272,
+      "step": 8550
+    },
+    {
+      "epoch": 6.395218528203213,
+      "grad_norm": 1.285474419593811,
+      "learning_rate": 0.0002,
+      "loss": 1.1615,
+      "step": 8560
+    },
+    {
+      "epoch": 6.402689577885693,
+      "grad_norm": 0.9751279950141907,
+      "learning_rate": 0.0002,
+      "loss": 1.1505,
+      "step": 8570
+    },
+    {
+      "epoch": 6.410160627568174,
+      "grad_norm": 1.2194149494171143,
+      "learning_rate": 0.0002,
+      "loss": 1.1502,
+      "step": 8580
+    },
+    {
+      "epoch": 6.417631677250654,
+      "grad_norm": 1.255888819694519,
+      "learning_rate": 0.0002,
+      "loss": 1.138,
+      "step": 8590
+    },
+    {
+      "epoch": 6.425102726933134,
+      "grad_norm": 1.1636122465133667,
+      "learning_rate": 0.0002,
+      "loss": 1.1308,
+      "step": 8600
+    },
+    {
+      "epoch": 6.432573776615614,
+      "grad_norm": 1.0769859552383423,
+      "learning_rate": 0.0002,
+      "loss": 1.1398,
+      "step": 8610
+    },
+    {
+      "epoch": 6.440044826298095,
+      "grad_norm": 1.151778221130371,
+      "learning_rate": 0.0002,
+      "loss": 1.1183,
+      "step": 8620
+    },
+    {
+      "epoch": 6.447515875980575,
+      "grad_norm": 1.2749944925308228,
+      "learning_rate": 0.0002,
+      "loss": 1.0706,
+      "step": 8630
+    },
+    {
+      "epoch": 6.454986925663055,
+      "grad_norm": 1.1925828456878662,
+      "learning_rate": 0.0002,
+      "loss": 1.1011,
+      "step": 8640
+    },
+    {
+      "epoch": 6.4624579753455365,
+      "grad_norm": 1.166107416152954,
+      "learning_rate": 0.0002,
+      "loss": 1.1581,
+      "step": 8650
+    },
+    {
+      "epoch": 6.469929025028017,
+      "grad_norm": 1.0372248888015747,
+      "learning_rate": 0.0002,
+      "loss": 1.105,
+      "step": 8660
+    },
+    {
+      "epoch": 6.477400074710497,
+      "grad_norm": 1.26933753490448,
+      "learning_rate": 0.0002,
+      "loss": 1.1546,
+      "step": 8670
+    },
+    {
+      "epoch": 6.484871124392977,
+      "grad_norm": 1.2154223918914795,
+      "learning_rate": 0.0002,
+      "loss": 1.2362,
+      "step": 8680
+    },
+    {
+      "epoch": 6.492342174075458,
+      "grad_norm": 1.09475839138031,
+      "learning_rate": 0.0002,
+      "loss": 1.1096,
+      "step": 8690
+    },
+    {
+      "epoch": 6.499813223757938,
+      "grad_norm": 1.0763037204742432,
+      "learning_rate": 0.0002,
+      "loss": 1.1168,
+      "step": 8700
+    },
+    {
+      "epoch": 6.507284273440418,
+      "grad_norm": 1.1882896423339844,
+      "learning_rate": 0.0002,
+      "loss": 1.1993,
+      "step": 8710
+    },
+    {
+      "epoch": 6.514755323122898,
+      "grad_norm": 1.1662089824676514,
+      "learning_rate": 0.0002,
+      "loss": 1.1498,
+      "step": 8720
+    },
+    {
+      "epoch": 6.522226372805379,
+      "grad_norm": 1.3259495496749878,
+      "learning_rate": 0.0002,
+      "loss": 1.2008,
+      "step": 8730
+    },
+    {
+      "epoch": 6.5296974224878594,
+      "grad_norm": 1.0858017206192017,
+      "learning_rate": 0.0002,
+      "loss": 1.1289,
+      "step": 8740
+    },
+    {
+      "epoch": 6.53716847217034,
+      "grad_norm": 1.240337610244751,
+      "learning_rate": 0.0002,
+      "loss": 1.1335,
+      "step": 8750
+    },
+    {
+      "epoch": 6.544639521852821,
+      "grad_norm": 1.1381462812423706,
+      "learning_rate": 0.0002,
+      "loss": 1.1479,
+      "step": 8760
+    },
+    {
+      "epoch": 6.552110571535301,
+      "grad_norm": 1.2220063209533691,
+      "learning_rate": 0.0002,
+      "loss": 1.0991,
+      "step": 8770
+    },
+    {
+      "epoch": 6.559581621217781,
+      "grad_norm": 1.1553083658218384,
+      "learning_rate": 0.0002,
+      "loss": 1.159,
+      "step": 8780
+    },
+    {
+      "epoch": 6.567052670900262,
+      "grad_norm": 1.1383219957351685,
+      "learning_rate": 0.0002,
+      "loss": 1.0996,
+      "step": 8790
+    },
+    {
+      "epoch": 6.574523720582742,
+      "grad_norm": 1.0379676818847656,
+      "learning_rate": 0.0002,
+      "loss": 1.1355,
+      "step": 8800
+    },
+    {
+      "epoch": 6.581994770265222,
+      "grad_norm": 1.376488447189331,
+      "learning_rate": 0.0002,
+      "loss": 1.1704,
+      "step": 8810
+    },
+    {
+      "epoch": 6.589465819947702,
+      "grad_norm": 1.1586211919784546,
+      "learning_rate": 0.0002,
+      "loss": 1.1265,
+      "step": 8820
+    },
+    {
+      "epoch": 6.596936869630183,
+      "grad_norm": 1.28152334690094,
+      "learning_rate": 0.0002,
+      "loss": 1.1904,
+      "step": 8830
+    },
+    {
+      "epoch": 6.6044079193126635,
+      "grad_norm": 1.2656810283660889,
+      "learning_rate": 0.0002,
+      "loss": 1.1646,
+      "step": 8840
+    },
+    {
+      "epoch": 6.611878968995144,
+      "grad_norm": 1.0636502504348755,
+      "learning_rate": 0.0002,
+      "loss": 1.1865,
+      "step": 8850
+    },
+    {
+      "epoch": 6.619350018677624,
+      "grad_norm": 1.273239254951477,
+      "learning_rate": 0.0002,
+      "loss": 1.125,
+      "step": 8860
+    },
+    {
+      "epoch": 6.626821068360105,
+      "grad_norm": 1.1055482625961304,
+      "learning_rate": 0.0002,
+      "loss": 1.1443,
+      "step": 8870
+    },
+    {
+      "epoch": 6.634292118042585,
+      "grad_norm": 1.1934176683425903,
+      "learning_rate": 0.0002,
+      "loss": 1.0877,
+      "step": 8880
+    },
+    {
+      "epoch": 6.641763167725065,
+      "grad_norm": 1.2248114347457886,
+      "learning_rate": 0.0002,
+      "loss": 1.194,
+      "step": 8890
+    },
+    {
+      "epoch": 6.649234217407546,
+      "grad_norm": 1.1950982809066772,
+      "learning_rate": 0.0002,
+      "loss": 1.1609,
+      "step": 8900
+    },
+    {
+      "epoch": 6.656705267090026,
+      "grad_norm": 1.0821784734725952,
+      "learning_rate": 0.0002,
+      "loss": 1.169,
+      "step": 8910
+    },
+    {
+      "epoch": 6.664176316772506,
+      "grad_norm": 1.0062463283538818,
+      "learning_rate": 0.0002,
+      "loss": 1.1337,
+      "step": 8920
+    },
+    {
+      "epoch": 6.671647366454987,
+      "grad_norm": 1.2373089790344238,
+      "learning_rate": 0.0002,
+      "loss": 1.1403,
+      "step": 8930
+    },
+    {
+      "epoch": 6.6791184161374675,
+      "grad_norm": 1.1821746826171875,
+      "learning_rate": 0.0002,
+      "loss": 1.2051,
+      "step": 8940
+    },
+    {
+      "epoch": 6.686589465819948,
+      "grad_norm": 1.2350659370422363,
+      "learning_rate": 0.0002,
+      "loss": 1.1214,
+      "step": 8950
+    },
+    {
+      "epoch": 6.694060515502428,
+      "grad_norm": 1.1012883186340332,
+      "learning_rate": 0.0002,
+      "loss": 1.225,
+      "step": 8960
+    },
+    {
+      "epoch": 6.701531565184909,
+      "grad_norm": 1.2008943557739258,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 8970
+    },
+    {
+      "epoch": 6.709002614867389,
+      "grad_norm": 1.2355504035949707,
+      "learning_rate": 0.0002,
+      "loss": 1.1769,
+      "step": 8980
+    },
+    {
+      "epoch": 6.716473664549869,
+      "grad_norm": 1.2367502450942993,
+      "learning_rate": 0.0002,
+      "loss": 1.1323,
+      "step": 8990
+    },
+    {
+      "epoch": 6.723944714232349,
+      "grad_norm": 1.1075866222381592,
+      "learning_rate": 0.0002,
+      "loss": 1.1235,
+      "step": 9000
+    },
+    {
+      "epoch": 6.73141576391483,
+      "grad_norm": 1.246480941772461,
+      "learning_rate": 0.0002,
+      "loss": 1.1239,
+      "step": 9010
+    },
+    {
+      "epoch": 6.73888681359731,
+      "grad_norm": 1.1252824068069458,
+      "learning_rate": 0.0002,
+      "loss": 1.2154,
+      "step": 9020
+    },
+    {
+      "epoch": 6.7463578632797905,
+      "grad_norm": 1.0706887245178223,
+      "learning_rate": 0.0002,
+      "loss": 1.1762,
+      "step": 9030
+    },
+    {
+      "epoch": 6.7538289129622715,
+      "grad_norm": 1.0874755382537842,
+      "learning_rate": 0.0002,
+      "loss": 1.1961,
+      "step": 9040
+    },
+    {
+      "epoch": 6.761299962644752,
+      "grad_norm": 1.121434211730957,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 9050
+    },
+    {
+      "epoch": 6.768771012327232,
+      "grad_norm": 1.1517996788024902,
+      "learning_rate": 0.0002,
+      "loss": 1.2018,
+      "step": 9060
+    },
+    {
+      "epoch": 6.776242062009713,
+      "grad_norm": 1.2484540939331055,
+      "learning_rate": 0.0002,
+      "loss": 1.1593,
+      "step": 9070
+    },
+    {
+      "epoch": 6.783713111692193,
+      "grad_norm": 1.023059368133545,
+      "learning_rate": 0.0002,
+      "loss": 1.13,
+      "step": 9080
+    },
+    {
+      "epoch": 6.791184161374673,
+      "grad_norm": 1.1334631443023682,
+      "learning_rate": 0.0002,
+      "loss": 1.1929,
+      "step": 9090
+    },
+    {
+      "epoch": 6.798655211057153,
+      "grad_norm": 1.2991816997528076,
+      "learning_rate": 0.0002,
+      "loss": 1.18,
+      "step": 9100
+    },
+    {
+      "epoch": 6.806126260739634,
+      "grad_norm": 1.4147199392318726,
+      "learning_rate": 0.0002,
+      "loss": 1.2398,
+      "step": 9110
+    },
+    {
+      "epoch": 6.813597310422114,
+      "grad_norm": 1.1353832483291626,
+      "learning_rate": 0.0002,
+      "loss": 1.0958,
+      "step": 9120
+    },
+    {
+      "epoch": 6.8210683601045945,
+      "grad_norm": 1.0332539081573486,
+      "learning_rate": 0.0002,
+      "loss": 1.1379,
+      "step": 9130
+    },
+    {
+      "epoch": 6.828539409787075,
+      "grad_norm": 1.2208142280578613,
+      "learning_rate": 0.0002,
+      "loss": 1.1652,
+      "step": 9140
+    },
+    {
+      "epoch": 6.836010459469556,
+      "grad_norm": 1.3033398389816284,
+      "learning_rate": 0.0002,
+      "loss": 1.1463,
+      "step": 9150
+    },
+    {
+      "epoch": 6.843481509152036,
+      "grad_norm": 1.2676737308502197,
+      "learning_rate": 0.0002,
+      "loss": 1.1834,
+      "step": 9160
+    },
+    {
+      "epoch": 6.850952558834516,
+      "grad_norm": 1.1668603420257568,
+      "learning_rate": 0.0002,
+      "loss": 1.1786,
+      "step": 9170
+    },
+    {
+      "epoch": 6.858423608516997,
+      "grad_norm": 1.1994788646697998,
+      "learning_rate": 0.0002,
+      "loss": 1.1801,
+      "step": 9180
+    },
+    {
+      "epoch": 6.865894658199477,
+      "grad_norm": 1.231873869895935,
+      "learning_rate": 0.0002,
+      "loss": 1.2131,
+      "step": 9190
+    },
+    {
+      "epoch": 6.873365707881957,
+      "grad_norm": 0.9981484413146973,
+      "learning_rate": 0.0002,
+      "loss": 1.2109,
+      "step": 9200
+    },
+    {
+      "epoch": 6.880836757564438,
+      "grad_norm": 1.2799428701400757,
+      "learning_rate": 0.0002,
+      "loss": 1.1084,
+      "step": 9210
+    },
+    {
+      "epoch": 6.888307807246918,
+      "grad_norm": 1.2042057514190674,
+      "learning_rate": 0.0002,
+      "loss": 1.2004,
+      "step": 9220
+    },
+    {
+      "epoch": 6.8957788569293985,
+      "grad_norm": 1.070420265197754,
+      "learning_rate": 0.0002,
+      "loss": 1.1567,
+      "step": 9230
+    },
+    {
+      "epoch": 6.903249906611879,
+      "grad_norm": 1.327160358428955,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 9240
+    },
+    {
+      "epoch": 6.91072095629436,
+      "grad_norm": 1.1109007596969604,
+      "learning_rate": 0.0002,
+      "loss": 1.1945,
+      "step": 9250
+    },
+    {
+      "epoch": 6.91819200597684,
+      "grad_norm": 1.1669930219650269,
+      "learning_rate": 0.0002,
+      "loss": 1.1701,
+      "step": 9260
+    },
+    {
+      "epoch": 6.92566305565932,
+      "grad_norm": 1.034532904624939,
+      "learning_rate": 0.0002,
+      "loss": 1.1854,
+      "step": 9270
+    },
+    {
+      "epoch": 6.9331341053418,
+      "grad_norm": 1.1035540103912354,
+      "learning_rate": 0.0002,
+      "loss": 1.1712,
+      "step": 9280
+    },
+    {
+      "epoch": 6.940605155024281,
+      "grad_norm": 1.366254448890686,
+      "learning_rate": 0.0002,
+      "loss": 1.1767,
+      "step": 9290
+    },
+    {
+      "epoch": 6.948076204706761,
+      "grad_norm": 1.094214677810669,
+      "learning_rate": 0.0002,
+      "loss": 1.1591,
+      "step": 9300
+    },
+    {
+      "epoch": 6.955547254389241,
+      "grad_norm": 1.131238579750061,
+      "learning_rate": 0.0002,
+      "loss": 1.18,
+      "step": 9310
+    },
+    {
+      "epoch": 6.963018304071722,
+      "grad_norm": 1.202369213104248,
+      "learning_rate": 0.0002,
+      "loss": 1.2513,
+      "step": 9320
+    },
+    {
+      "epoch": 6.9704893537542025,
+      "grad_norm": 1.1067225933074951,
+      "learning_rate": 0.0002,
+      "loss": 1.1922,
+      "step": 9330
+    },
+    {
+      "epoch": 6.977960403436683,
+      "grad_norm": 1.0258643627166748,
+      "learning_rate": 0.0002,
+      "loss": 1.1965,
+      "step": 9340
+    },
+    {
+      "epoch": 6.985431453119164,
+      "grad_norm": 1.3311655521392822,
+      "learning_rate": 0.0002,
+      "loss": 1.2053,
+      "step": 9350
+    },
+    {
+      "epoch": 6.992902502801644,
+      "grad_norm": 1.1245559453964233,
+      "learning_rate": 0.0002,
+      "loss": 1.1778,
+      "step": 9360
+    },
+    {
+      "epoch": 6.999626447515876,
+      "eval_loss": 2.128103017807007,
+      "eval_runtime": 39.1339,
+      "eval_samples_per_second": 13.16,
+      "eval_steps_per_second": 1.661,
+      "step": 9369
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10704,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.335994808397988e+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-0.9-num-6326-sd-42/checkpoint-9369/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-0.9-num-6326-sd-42/checkpoint-9369/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..67c7b4ca126d7b712ad1985ef15ffb29ebe76633
--- /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-0.9-num-6326-sd-42/checkpoint-9369/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fc87da605e94ff0ecd8f5b371302a2d5f8727b77984707a2185ddb447fc3796
+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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-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-0.9-num-6326-sd-42/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..67c7b4ca126d7b712ad1985ef15ffb29ebe76633
--- /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-0.9-num-6326-sd-42/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fc87da605e94ff0ecd8f5b371302a2d5f8727b77984707a2185ddb447fc3796
+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-0.9-num-6326-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-0.9-num-6326-sd-42/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..8494c24717c1201780732d97f3f6759d9500a9ae
--- /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-0.9-num-6326-sd-42/training_log.jsonl
@@ -0,0 +1,8 @@
+{"epoch": 0.9996264475158759, "step": 1338, "epoch_duration": 1438.9762296676636, "total_accumulated_duration": 1438.9762296676636, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7672.4541015625}, "peak_memory_usage": {"GPU_0": 9688.99365234375}, "avg_memory_reserved": {"GPU_0": 10406.0}, "peak_memory_reserved": {"GPU_0": 10406.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.6181, "grad_norm": 0.4912872612476349, "learning_rate": 0.0002, "epoch": 0.007471049682480389, "step": 10}, {"loss": 2.2606, "grad_norm": 0.4856316149234772, "learning_rate": 0.0002, "epoch": 0.014942099364960777, "step": 20}, {"loss": 2.0957, "grad_norm": 0.47683125734329224, "learning_rate": 0.0002, "epoch": 0.022413149047441166, "step": 30}, {"loss": 1.8908, "grad_norm": 0.515082597732544, "learning_rate": 0.0002, "epoch": 0.029884198729921554, "step": 40}, {"loss": 1.9704, "grad_norm": 0.5299215316772461, "learning_rate": 0.0002, "epoch": 0.03735524841240194, "step": 50}, {"loss": 1.9225, "grad_norm": 0.4951399862766266, "learning_rate": 0.0002, "epoch": 0.04482629809488233, "step": 60}, {"loss": 1.9742, "grad_norm": 0.48079821467399597, "learning_rate": 0.0002, "epoch": 0.05229734777736272, "step": 70}, {"loss": 1.9466, "grad_norm": 0.49402132630348206, "learning_rate": 0.0002, "epoch": 0.05976839745984311, "step": 80}, {"loss": 1.8691, "grad_norm": 0.4778193235397339, "learning_rate": 0.0002, "epoch": 0.0672394471423235, "step": 90}, {"loss": 1.8455, "grad_norm": 0.42472657561302185, "learning_rate": 0.0002, "epoch": 0.07471049682480388, "step": 100}, {"loss": 1.8744, "grad_norm": 0.4433092474937439, "learning_rate": 0.0002, "epoch": 0.08218154650728428, "step": 110}, {"loss": 1.865, "grad_norm": 0.4472862780094147, "learning_rate": 0.0002, "epoch": 0.08965259618976466, "step": 120}, {"loss": 1.9256, "grad_norm": 0.42596298456192017, "learning_rate": 0.0002, "epoch": 0.09712364587224505, "step": 130}, {"loss": 1.8015, "grad_norm": 0.46645811200141907, "learning_rate": 0.0002, "epoch": 0.10459469555472543, "step": 140}, {"loss": 1.8307, "grad_norm": 0.41041234135627747, "learning_rate": 0.0002, "epoch": 0.11206574523720583, "step": 150}, {"loss": 1.8276, "grad_norm": 0.5329819917678833, "learning_rate": 0.0002, "epoch": 0.11953679491968622, "step": 160}, {"loss": 1.8118, "grad_norm": 0.4065922200679779, "learning_rate": 0.0002, "epoch": 0.1270078446021666, "step": 170}, {"loss": 1.8559, "grad_norm": 0.38406994938850403, "learning_rate": 0.0002, "epoch": 0.134478894284647, "step": 180}, {"loss": 1.8647, "grad_norm": 0.4246881306171417, "learning_rate": 0.0002, "epoch": 0.14194994396712737, "step": 190}, {"loss": 1.8054, "grad_norm": 0.35136649012565613, "learning_rate": 0.0002, "epoch": 0.14942099364960776, "step": 200}, {"loss": 1.802, "grad_norm": 0.43252742290496826, "learning_rate": 0.0002, "epoch": 0.15689204333208817, "step": 210}, {"loss": 1.7823, "grad_norm": 0.39236941933631897, "learning_rate": 0.0002, "epoch": 0.16436309301456856, "step": 220}, {"loss": 1.818, "grad_norm": 0.3748249113559723, "learning_rate": 0.0002, "epoch": 0.17183414269704894, "step": 230}, {"loss": 1.866, "grad_norm": 0.6432855725288391, "learning_rate": 0.0002, "epoch": 0.17930519237952933, "step": 240}, {"loss": 1.8397, "grad_norm": 0.34874802827835083, "learning_rate": 0.0002, "epoch": 0.1867762420620097, "step": 250}, {"loss": 1.79, "grad_norm": 0.3721984326839447, "learning_rate": 0.0002, "epoch": 0.1942472917444901, "step": 260}, {"loss": 1.8464, "grad_norm": 0.4339311420917511, "learning_rate": 0.0002, "epoch": 0.20171834142697048, "step": 270}, {"loss": 1.8665, "grad_norm": 0.4018215537071228, "learning_rate": 0.0002, "epoch": 0.20918939110945087, "step": 280}, {"loss": 1.8048, "grad_norm": 0.3278839886188507, "learning_rate": 0.0002, "epoch": 0.21666044079193125, "step": 290}, {"loss": 1.7395, "grad_norm": 0.36146077513694763, "learning_rate": 0.0002, "epoch": 0.22413149047441167, "step": 300}, {"loss": 1.7916, "grad_norm": 0.38175010681152344, "learning_rate": 0.0002, "epoch": 0.23160254015689205, "step": 310}, {"loss": 1.8593, "grad_norm": 0.44776618480682373, "learning_rate": 0.0002, "epoch": 0.23907358983937244, "step": 320}, {"loss": 1.7824, "grad_norm": 0.3933652937412262, "learning_rate": 0.0002, "epoch": 0.24654463952185282, "step": 330}, {"loss": 1.8393, "grad_norm": 0.3515005111694336, "learning_rate": 0.0002, "epoch": 0.2540156892043332, "step": 340}, {"loss": 1.8653, "grad_norm": 0.6683304309844971, "learning_rate": 0.0002, "epoch": 0.2614867388868136, "step": 350}, {"loss": 1.8797, "grad_norm": 0.37093454599380493, "learning_rate": 0.0002, "epoch": 0.268957788569294, "step": 360}, {"loss": 1.8251, "grad_norm": 0.3450651168823242, "learning_rate": 0.0002, "epoch": 0.2764288382517744, "step": 370}, {"loss": 1.7435, "grad_norm": 0.5140917301177979, "learning_rate": 0.0002, "epoch": 0.28389988793425475, "step": 380}, {"loss": 1.8026, "grad_norm": 0.32885563373565674, "learning_rate": 0.0002, "epoch": 0.29137093761673516, "step": 390}, {"loss": 1.8174, "grad_norm": 0.33962297439575195, "learning_rate": 0.0002, "epoch": 0.2988419872992155, "step": 400}, {"loss": 1.7467, "grad_norm": 0.3723141849040985, "learning_rate": 0.0002, "epoch": 0.30631303698169593, "step": 410}, {"loss": 1.8459, "grad_norm": 0.37173134088516235, "learning_rate": 0.0002, "epoch": 0.31378408666417634, "step": 420}, {"loss": 1.8876, "grad_norm": 0.33736956119537354, "learning_rate": 0.0002, "epoch": 0.3212551363466567, "step": 430}, {"loss": 1.8367, "grad_norm": 0.3602448105812073, "learning_rate": 0.0002, "epoch": 0.3287261860291371, "step": 440}, {"loss": 1.8058, "grad_norm": 0.3569699227809906, "learning_rate": 0.0002, "epoch": 0.33619723571161747, "step": 450}, {"loss": 1.8086, "grad_norm": 0.31009167432785034, "learning_rate": 0.0002, "epoch": 0.3436682853940979, "step": 460}, {"loss": 1.8876, "grad_norm": 0.5278693437576294, "learning_rate": 0.0002, "epoch": 0.35113933507657824, "step": 470}, {"loss": 1.8534, "grad_norm": 0.3587537109851837, "learning_rate": 0.0002, "epoch": 0.35861038475905865, "step": 480}, {"loss": 1.8046, "grad_norm": 0.3859670162200928, "learning_rate": 0.0002, "epoch": 0.366081434441539, "step": 490}, {"loss": 1.8287, "grad_norm": 0.395913690328598, "learning_rate": 0.0002, "epoch": 0.3735524841240194, "step": 500}, {"loss": 1.7619, "grad_norm": 0.35052940249443054, "learning_rate": 0.0002, "epoch": 0.38102353380649984, "step": 510}, {"loss": 1.7824, "grad_norm": 0.2979494333267212, "learning_rate": 0.0002, "epoch": 0.3884945834889802, "step": 520}, {"loss": 1.8641, "grad_norm": 0.3062683343887329, "learning_rate": 0.0002, "epoch": 0.3959656331714606, "step": 530}, {"loss": 1.7651, "grad_norm": 0.3172847330570221, "learning_rate": 0.0002, "epoch": 0.40343668285394096, "step": 540}, {"loss": 1.806, "grad_norm": 0.360435426235199, "learning_rate": 0.0002, "epoch": 0.4109077325364214, "step": 550}, {"loss": 1.9054, "grad_norm": 0.3427872359752655, "learning_rate": 0.0002, "epoch": 0.41837878221890173, "step": 560}, {"loss": 1.7562, "grad_norm": 0.34036558866500854, "learning_rate": 0.0002, "epoch": 0.42584983190138215, "step": 570}, {"loss": 1.7254, "grad_norm": 0.3365345299243927, "learning_rate": 0.0002, "epoch": 0.4333208815838625, "step": 580}, {"loss": 1.8328, "grad_norm": 0.35619041323661804, "learning_rate": 0.0002, "epoch": 0.4407919312663429, "step": 590}, {"loss": 1.8114, "grad_norm": 0.3569088280200958, "learning_rate": 0.0002, "epoch": 0.44826298094882333, "step": 600}, {"loss": 1.8599, "grad_norm": 0.3581278622150421, "learning_rate": 0.0002, "epoch": 0.4557340306313037, "step": 610}, {"loss": 1.7078, "grad_norm": 0.43197110295295715, "learning_rate": 0.0002, "epoch": 0.4632050803137841, "step": 620}, {"loss": 1.8257, "grad_norm": 0.33966198563575745, "learning_rate": 0.0002, "epoch": 0.47067612999626446, "step": 630}, {"loss": 1.7528, "grad_norm": 0.3343866467475891, "learning_rate": 0.0002, "epoch": 0.47814717967874487, "step": 640}, {"loss": 1.8191, "grad_norm": 0.33878564834594727, "learning_rate": 0.0002, "epoch": 0.48561822936122523, "step": 650}, {"loss": 1.8801, "grad_norm": 0.387195885181427, "learning_rate": 0.0002, "epoch": 0.49308927904370564, "step": 660}, {"loss": 1.7559, "grad_norm": 0.3755440413951874, "learning_rate": 0.0002, "epoch": 0.500560328726186, "step": 670}, {"loss": 1.8057, "grad_norm": 0.3272816836833954, "learning_rate": 0.0002, "epoch": 0.5080313784086664, "step": 680}, {"loss": 1.8156, "grad_norm": 0.36063864827156067, "learning_rate": 0.0002, "epoch": 0.5155024280911468, "step": 690}, {"loss": 1.8397, "grad_norm": 0.35317373275756836, "learning_rate": 0.0002, "epoch": 0.5229734777736272, "step": 700}, {"loss": 1.7603, "grad_norm": 0.3561195433139801, "learning_rate": 0.0002, "epoch": 0.5304445274561076, "step": 710}, {"loss": 1.8149, "grad_norm": 0.31124624609947205, "learning_rate": 0.0002, "epoch": 0.537915577138588, "step": 720}, {"loss": 1.7434, "grad_norm": 0.3294544517993927, "learning_rate": 0.0002, "epoch": 0.5453866268210683, "step": 730}, {"loss": 1.8027, "grad_norm": 0.31933900713920593, "learning_rate": 0.0002, "epoch": 0.5528576765035488, "step": 740}, {"loss": 1.7601, "grad_norm": 0.3226020634174347, "learning_rate": 0.0002, "epoch": 0.5603287261860291, "step": 750}, {"loss": 1.7862, "grad_norm": 0.3147525489330292, "learning_rate": 0.0002, "epoch": 0.5677997758685095, "step": 760}, {"loss": 1.9028, "grad_norm": 0.32234328985214233, "learning_rate": 0.0002, "epoch": 0.57527082555099, "step": 770}, {"loss": 1.7623, "grad_norm": 0.3258664309978485, "learning_rate": 0.0002, "epoch": 0.5827418752334703, "step": 780}, {"loss": 1.7384, "grad_norm": 0.3166961967945099, "learning_rate": 0.0002, "epoch": 0.5902129249159507, "step": 790}, {"loss": 1.8799, "grad_norm": 0.35621458292007446, "learning_rate": 0.0002, "epoch": 0.597683974598431, "step": 800}, {"loss": 1.8313, "grad_norm": 0.3236999213695526, "learning_rate": 0.0002, "epoch": 0.6051550242809115, "step": 810}, {"loss": 1.7132, "grad_norm": 0.2892923653125763, "learning_rate": 0.0002, "epoch": 0.6126260739633919, "step": 820}, {"loss": 1.8709, "grad_norm": 0.4098321497440338, "learning_rate": 0.0002, "epoch": 0.6200971236458722, "step": 830}, {"loss": 1.7637, "grad_norm": 0.3337118923664093, "learning_rate": 0.0002, "epoch": 0.6275681733283527, "step": 840}, {"loss": 1.7375, "grad_norm": 0.30416029691696167, "learning_rate": 0.0002, "epoch": 0.635039223010833, "step": 850}, {"loss": 1.7419, "grad_norm": 0.3361026346683502, "learning_rate": 0.0002, "epoch": 0.6425102726933134, "step": 860}, {"loss": 1.732, "grad_norm": 0.3537365198135376, "learning_rate": 0.0002, "epoch": 0.6499813223757938, "step": 870}, {"loss": 1.7825, "grad_norm": 0.33854469656944275, "learning_rate": 0.0002, "epoch": 0.6574523720582742, "step": 880}, {"loss": 1.7561, "grad_norm": 0.3332272469997406, "learning_rate": 0.0002, "epoch": 0.6649234217407546, "step": 890}, {"loss": 1.7247, "grad_norm": 0.34954726696014404, "learning_rate": 0.0002, "epoch": 0.6723944714232349, "step": 900}, {"loss": 1.7917, "grad_norm": 0.2921750247478485, "learning_rate": 0.0002, "epoch": 0.6798655211057153, "step": 910}, {"loss": 1.7807, "grad_norm": 0.30508682131767273, "learning_rate": 0.0002, "epoch": 0.6873365707881958, "step": 920}, {"loss": 1.8082, "grad_norm": 0.32268425822257996, "learning_rate": 0.0002, "epoch": 0.6948076204706761, "step": 930}, {"loss": 1.8283, "grad_norm": 0.2844390869140625, "learning_rate": 0.0002, "epoch": 0.7022786701531565, "step": 940}, {"loss": 1.7363, "grad_norm": 0.31263890862464905, "learning_rate": 0.0002, "epoch": 0.709749719835637, "step": 950}, {"loss": 1.8081, "grad_norm": 0.3626808822154999, "learning_rate": 0.0002, "epoch": 0.7172207695181173, "step": 960}, {"loss": 1.853, "grad_norm": 0.3322749733924866, "learning_rate": 0.0002, "epoch": 0.7246918192005977, "step": 970}, {"loss": 1.7912, "grad_norm": 0.29177871346473694, "learning_rate": 0.0002, "epoch": 0.732162868883078, "step": 980}, {"loss": 1.8447, "grad_norm": 0.35405513644218445, "learning_rate": 0.0002, "epoch": 0.7396339185655585, "step": 990}, {"loss": 1.7008, "grad_norm": 0.39318400621414185, "learning_rate": 0.0002, "epoch": 0.7471049682480388, "step": 1000}, {"loss": 1.7803, "grad_norm": 0.29401418566703796, "learning_rate": 0.0002, "epoch": 0.7545760179305192, "step": 1010}, {"loss": 1.7649, "grad_norm": 0.3271748721599579, "learning_rate": 0.0002, "epoch": 0.7620470676129997, "step": 1020}, {"loss": 1.7266, "grad_norm": 0.30883970856666565, "learning_rate": 0.0002, "epoch": 0.76951811729548, "step": 1030}, {"loss": 1.7722, "grad_norm": 0.3411838412284851, "learning_rate": 0.0002, "epoch": 0.7769891669779604, "step": 1040}, {"loss": 1.829, "grad_norm": 0.30608129501342773, "learning_rate": 0.0002, "epoch": 0.7844602166604407, "step": 1050}, {"loss": 1.7815, "grad_norm": 0.30899080634117126, "learning_rate": 0.0002, "epoch": 0.7919312663429212, "step": 1060}, {"loss": 1.7625, "grad_norm": 0.3160453140735626, "learning_rate": 0.0002, "epoch": 0.7994023160254016, "step": 1070}, {"loss": 1.8452, "grad_norm": 0.30947187542915344, "learning_rate": 0.0002, "epoch": 0.8068733657078819, "step": 1080}, {"loss": 1.7418, "grad_norm": 0.3103134036064148, "learning_rate": 0.0002, "epoch": 0.8143444153903624, "step": 1090}, {"loss": 1.842, "grad_norm": 0.31771138310432434, "learning_rate": 0.0002, "epoch": 0.8218154650728428, "step": 1100}, {"loss": 1.7918, "grad_norm": 0.5860997438430786, "learning_rate": 0.0002, "epoch": 0.8292865147553231, "step": 1110}, {"loss": 1.8443, "grad_norm": 0.3230148255825043, "learning_rate": 0.0002, "epoch": 0.8367575644378035, "step": 1120}, {"loss": 1.8478, "grad_norm": 0.29611510038375854, "learning_rate": 0.0002, "epoch": 0.8442286141202839, "step": 1130}, {"loss": 1.7673, "grad_norm": 0.3373654782772064, "learning_rate": 0.0002, "epoch": 0.8516996638027643, "step": 1140}, {"loss": 1.7997, "grad_norm": 0.3474279046058655, "learning_rate": 0.0002, "epoch": 0.8591707134852447, "step": 1150}, {"loss": 1.75, "grad_norm": 0.35057875514030457, "learning_rate": 0.0002, "epoch": 0.866641763167725, "step": 1160}, {"loss": 1.8273, "grad_norm": 0.39537495374679565, "learning_rate": 0.0002, "epoch": 0.8741128128502055, "step": 1170}, {"loss": 1.7682, "grad_norm": 0.3714233636856079, "learning_rate": 0.0002, "epoch": 0.8815838625326858, "step": 1180}, {"loss": 1.7549, "grad_norm": 0.2950296998023987, "learning_rate": 0.0002, "epoch": 0.8890549122151662, "step": 1190}, {"loss": 1.7612, "grad_norm": 0.38182979822158813, "learning_rate": 0.0002, "epoch": 0.8965259618976467, "step": 1200}, {"loss": 1.827, "grad_norm": 0.27883678674697876, "learning_rate": 0.0002, "epoch": 0.903997011580127, "step": 1210}, {"loss": 1.7623, "grad_norm": 0.33874374628067017, "learning_rate": 0.0002, "epoch": 0.9114680612626074, "step": 1220}, {"loss": 1.7334, "grad_norm": 0.3014272153377533, "learning_rate": 0.0002, "epoch": 0.9189391109450877, "step": 1230}, {"loss": 1.8235, "grad_norm": 0.3194271922111511, "learning_rate": 0.0002, "epoch": 0.9264101606275682, "step": 1240}, {"loss": 1.7924, "grad_norm": 0.3049403429031372, "learning_rate": 0.0002, "epoch": 0.9338812103100486, "step": 1250}, {"loss": 1.7535, "grad_norm": 0.30621254444122314, "learning_rate": 0.0002, "epoch": 0.9413522599925289, "step": 1260}, {"loss": 1.8287, "grad_norm": 0.28675132989883423, "learning_rate": 0.0002, "epoch": 0.9488233096750094, "step": 1270}, {"loss": 1.7586, "grad_norm": 0.3322032690048218, "learning_rate": 0.0002, "epoch": 0.9562943593574897, "step": 1280}, {"loss": 1.8054, "grad_norm": 0.35408294200897217, "learning_rate": 0.0002, "epoch": 0.9637654090399701, "step": 1290}, {"loss": 1.7343, "grad_norm": 0.36386919021606445, "learning_rate": 0.0002, "epoch": 0.9712364587224505, "step": 1300}, {"loss": 1.8633, "grad_norm": 0.32338324189186096, "learning_rate": 0.0002, "epoch": 0.9787075084049309, "step": 1310}, {"loss": 1.7724, "grad_norm": 0.3714013993740082, "learning_rate": 0.0002, "epoch": 0.9861785580874113, "step": 1320}, {"loss": 1.7766, "grad_norm": 0.3133082389831543, "learning_rate": 0.0002, "epoch": 0.9936496077698916, "step": 1330}]}
+{"epoch": 2.0, "step": 2677, "epoch_duration": 1435.5856530666351, "total_accumulated_duration": 2874.5618827342987, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7568.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-42/checkpoint-1338", "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": 2.6181, "grad_norm": 0.4912872612476349, "learning_rate": 0.0002, "epoch": 0.007471049682480389, "step": 10}, {"loss": 2.2606, "grad_norm": 0.4856316149234772, "learning_rate": 0.0002, "epoch": 0.014942099364960777, "step": 20}, {"loss": 2.0957, "grad_norm": 0.47683125734329224, "learning_rate": 0.0002, "epoch": 0.022413149047441166, "step": 30}, {"loss": 1.8908, "grad_norm": 0.515082597732544, "learning_rate": 0.0002, "epoch": 0.029884198729921554, "step": 40}, {"loss": 1.9704, "grad_norm": 0.5299215316772461, "learning_rate": 0.0002, "epoch": 0.03735524841240194, "step": 50}, {"loss": 1.9225, "grad_norm": 0.4951399862766266, "learning_rate": 0.0002, "epoch": 0.04482629809488233, "step": 60}, {"loss": 1.9742, "grad_norm": 0.48079821467399597, "learning_rate": 0.0002, "epoch": 0.05229734777736272, "step": 70}, {"loss": 1.9466, "grad_norm": 0.49402132630348206, "learning_rate": 0.0002, "epoch": 0.05976839745984311, "step": 80}, {"loss": 1.8691, "grad_norm": 0.4778193235397339, "learning_rate": 0.0002, "epoch": 0.0672394471423235, "step": 90}, {"loss": 1.8455, "grad_norm": 0.42472657561302185, "learning_rate": 0.0002, "epoch": 0.07471049682480388, "step": 100}, {"loss": 1.8744, "grad_norm": 0.4433092474937439, "learning_rate": 0.0002, "epoch": 0.08218154650728428, "step": 110}, {"loss": 1.865, "grad_norm": 0.4472862780094147, "learning_rate": 0.0002, "epoch": 0.08965259618976466, "step": 120}, {"loss": 1.9256, "grad_norm": 0.42596298456192017, "learning_rate": 0.0002, "epoch": 0.09712364587224505, "step": 130}, {"loss": 1.8015, "grad_norm": 0.46645811200141907, "learning_rate": 0.0002, "epoch": 0.10459469555472543, "step": 140}, {"loss": 1.8307, "grad_norm": 0.41041234135627747, "learning_rate": 0.0002, "epoch": 0.11206574523720583, "step": 150}, {"loss": 1.8276, "grad_norm": 0.5329819917678833, "learning_rate": 0.0002, "epoch": 0.11953679491968622, "step": 160}, {"loss": 1.8118, "grad_norm": 0.4065922200679779, "learning_rate": 0.0002, "epoch": 0.1270078446021666, "step": 170}, {"loss": 1.8559, "grad_norm": 0.38406994938850403, "learning_rate": 0.0002, "epoch": 0.134478894284647, "step": 180}, {"loss": 1.8647, "grad_norm": 0.4246881306171417, "learning_rate": 0.0002, "epoch": 0.14194994396712737, "step": 190}, {"loss": 1.8054, "grad_norm": 0.35136649012565613, "learning_rate": 0.0002, "epoch": 0.14942099364960776, "step": 200}, {"loss": 1.802, "grad_norm": 0.43252742290496826, "learning_rate": 0.0002, "epoch": 0.15689204333208817, "step": 210}, {"loss": 1.7823, "grad_norm": 0.39236941933631897, "learning_rate": 0.0002, "epoch": 0.16436309301456856, "step": 220}, {"loss": 1.818, "grad_norm": 0.3748249113559723, "learning_rate": 0.0002, "epoch": 0.17183414269704894, "step": 230}, {"loss": 1.866, "grad_norm": 0.6432855725288391, "learning_rate": 0.0002, "epoch": 0.17930519237952933, "step": 240}, {"loss": 1.8397, "grad_norm": 0.34874802827835083, "learning_rate": 0.0002, "epoch": 0.1867762420620097, "step": 250}, {"loss": 1.79, "grad_norm": 0.3721984326839447, "learning_rate": 0.0002, "epoch": 0.1942472917444901, "step": 260}, {"loss": 1.8464, "grad_norm": 0.4339311420917511, "learning_rate": 0.0002, "epoch": 0.20171834142697048, "step": 270}, {"loss": 1.8665, "grad_norm": 0.4018215537071228, "learning_rate": 0.0002, "epoch": 0.20918939110945087, "step": 280}, {"loss": 1.8048, "grad_norm": 0.3278839886188507, "learning_rate": 0.0002, "epoch": 0.21666044079193125, "step": 290}, {"loss": 1.7395, "grad_norm": 0.36146077513694763, "learning_rate": 0.0002, "epoch": 0.22413149047441167, "step": 300}, {"loss": 1.7916, "grad_norm": 0.38175010681152344, "learning_rate": 0.0002, "epoch": 0.23160254015689205, "step": 310}, {"loss": 1.8593, "grad_norm": 0.44776618480682373, "learning_rate": 0.0002, "epoch": 0.23907358983937244, "step": 320}, {"loss": 1.7824, "grad_norm": 0.3933652937412262, "learning_rate": 0.0002, "epoch": 0.24654463952185282, "step": 330}, {"loss": 1.8393, "grad_norm": 0.3515005111694336, "learning_rate": 0.0002, "epoch": 0.2540156892043332, "step": 340}, {"loss": 1.8653, "grad_norm": 0.6683304309844971, "learning_rate": 0.0002, "epoch": 0.2614867388868136, "step": 350}, {"loss": 1.8797, "grad_norm": 0.37093454599380493, "learning_rate": 0.0002, "epoch": 0.268957788569294, "step": 360}, {"loss": 1.8251, "grad_norm": 0.3450651168823242, "learning_rate": 0.0002, "epoch": 0.2764288382517744, "step": 370}, {"loss": 1.7435, "grad_norm": 0.5140917301177979, "learning_rate": 0.0002, "epoch": 0.28389988793425475, "step": 380}, {"loss": 1.8026, "grad_norm": 0.32885563373565674, "learning_rate": 0.0002, "epoch": 0.29137093761673516, "step": 390}, {"loss": 1.8174, "grad_norm": 0.33962297439575195, "learning_rate": 0.0002, "epoch": 0.2988419872992155, "step": 400}, {"loss": 1.7467, "grad_norm": 0.3723141849040985, "learning_rate": 0.0002, "epoch": 0.30631303698169593, "step": 410}, {"loss": 1.8459, "grad_norm": 0.37173134088516235, "learning_rate": 0.0002, "epoch": 0.31378408666417634, "step": 420}, {"loss": 1.8876, "grad_norm": 0.33736956119537354, "learning_rate": 0.0002, "epoch": 0.3212551363466567, "step": 430}, {"loss": 1.8367, "grad_norm": 0.3602448105812073, "learning_rate": 0.0002, "epoch": 0.3287261860291371, "step": 440}, {"loss": 1.8058, "grad_norm": 0.3569699227809906, "learning_rate": 0.0002, "epoch": 0.33619723571161747, "step": 450}, {"loss": 1.8086, "grad_norm": 0.31009167432785034, "learning_rate": 0.0002, "epoch": 0.3436682853940979, "step": 460}, {"loss": 1.8876, "grad_norm": 0.5278693437576294, "learning_rate": 0.0002, "epoch": 0.35113933507657824, "step": 470}, {"loss": 1.8534, "grad_norm": 0.3587537109851837, "learning_rate": 0.0002, "epoch": 0.35861038475905865, "step": 480}, {"loss": 1.8046, "grad_norm": 0.3859670162200928, "learning_rate": 0.0002, "epoch": 0.366081434441539, "step": 490}, {"loss": 1.8287, "grad_norm": 0.395913690328598, "learning_rate": 0.0002, "epoch": 0.3735524841240194, "step": 500}, {"loss": 1.7619, "grad_norm": 0.35052940249443054, "learning_rate": 0.0002, "epoch": 0.38102353380649984, "step": 510}, {"loss": 1.7824, "grad_norm": 0.2979494333267212, "learning_rate": 0.0002, "epoch": 0.3884945834889802, "step": 520}, {"loss": 1.8641, "grad_norm": 0.3062683343887329, "learning_rate": 0.0002, "epoch": 0.3959656331714606, "step": 530}, {"loss": 1.7651, "grad_norm": 0.3172847330570221, "learning_rate": 0.0002, "epoch": 0.40343668285394096, "step": 540}, {"loss": 1.806, "grad_norm": 0.360435426235199, "learning_rate": 0.0002, "epoch": 0.4109077325364214, "step": 550}, {"loss": 1.9054, "grad_norm": 0.3427872359752655, "learning_rate": 0.0002, "epoch": 0.41837878221890173, "step": 560}, {"loss": 1.7562, "grad_norm": 0.34036558866500854, "learning_rate": 0.0002, "epoch": 0.42584983190138215, "step": 570}, {"loss": 1.7254, "grad_norm": 0.3365345299243927, "learning_rate": 0.0002, "epoch": 0.4333208815838625, "step": 580}, {"loss": 1.8328, "grad_norm": 0.35619041323661804, "learning_rate": 0.0002, "epoch": 0.4407919312663429, "step": 590}, {"loss": 1.8114, "grad_norm": 0.3569088280200958, "learning_rate": 0.0002, "epoch": 0.44826298094882333, "step": 600}, {"loss": 1.8599, "grad_norm": 0.3581278622150421, "learning_rate": 0.0002, "epoch": 0.4557340306313037, "step": 610}, {"loss": 1.7078, "grad_norm": 0.43197110295295715, "learning_rate": 0.0002, "epoch": 0.4632050803137841, "step": 620}, {"loss": 1.8257, "grad_norm": 0.33966198563575745, "learning_rate": 0.0002, "epoch": 0.47067612999626446, "step": 630}, {"loss": 1.7528, "grad_norm": 0.3343866467475891, "learning_rate": 0.0002, "epoch": 0.47814717967874487, "step": 640}, {"loss": 1.8191, "grad_norm": 0.33878564834594727, "learning_rate": 0.0002, "epoch": 0.48561822936122523, "step": 650}, {"loss": 1.8801, "grad_norm": 0.387195885181427, "learning_rate": 0.0002, "epoch": 0.49308927904370564, "step": 660}, {"loss": 1.7559, "grad_norm": 0.3755440413951874, "learning_rate": 0.0002, "epoch": 0.500560328726186, "step": 670}, {"loss": 1.8057, "grad_norm": 0.3272816836833954, "learning_rate": 0.0002, "epoch": 0.5080313784086664, "step": 680}, {"loss": 1.8156, "grad_norm": 0.36063864827156067, "learning_rate": 0.0002, "epoch": 0.5155024280911468, "step": 690}, {"loss": 1.8397, "grad_norm": 0.35317373275756836, "learning_rate": 0.0002, "epoch": 0.5229734777736272, "step": 700}, {"loss": 1.7603, "grad_norm": 0.3561195433139801, "learning_rate": 0.0002, "epoch": 0.5304445274561076, "step": 710}, {"loss": 1.8149, "grad_norm": 0.31124624609947205, "learning_rate": 0.0002, "epoch": 0.537915577138588, "step": 720}, {"loss": 1.7434, "grad_norm": 0.3294544517993927, "learning_rate": 0.0002, "epoch": 0.5453866268210683, "step": 730}, {"loss": 1.8027, "grad_norm": 0.31933900713920593, "learning_rate": 0.0002, "epoch": 0.5528576765035488, "step": 740}, {"loss": 1.7601, "grad_norm": 0.3226020634174347, "learning_rate": 0.0002, "epoch": 0.5603287261860291, "step": 750}, {"loss": 1.7862, "grad_norm": 0.3147525489330292, "learning_rate": 0.0002, "epoch": 0.5677997758685095, "step": 760}, {"loss": 1.9028, "grad_norm": 0.32234328985214233, "learning_rate": 0.0002, "epoch": 0.57527082555099, "step": 770}, {"loss": 1.7623, "grad_norm": 0.3258664309978485, "learning_rate": 0.0002, "epoch": 0.5827418752334703, "step": 780}, {"loss": 1.7384, "grad_norm": 0.3166961967945099, "learning_rate": 0.0002, "epoch": 0.5902129249159507, "step": 790}, {"loss": 1.8799, "grad_norm": 0.35621458292007446, "learning_rate": 0.0002, "epoch": 0.597683974598431, "step": 800}, {"loss": 1.8313, "grad_norm": 0.3236999213695526, "learning_rate": 0.0002, "epoch": 0.6051550242809115, "step": 810}, {"loss": 1.7132, "grad_norm": 0.2892923653125763, "learning_rate": 0.0002, "epoch": 0.6126260739633919, "step": 820}, {"loss": 1.8709, "grad_norm": 0.4098321497440338, "learning_rate": 0.0002, "epoch": 0.6200971236458722, "step": 830}, {"loss": 1.7637, "grad_norm": 0.3337118923664093, "learning_rate": 0.0002, "epoch": 0.6275681733283527, "step": 840}, {"loss": 1.7375, "grad_norm": 0.30416029691696167, "learning_rate": 0.0002, "epoch": 0.635039223010833, "step": 850}, {"loss": 1.7419, "grad_norm": 0.3361026346683502, "learning_rate": 0.0002, "epoch": 0.6425102726933134, "step": 860}, {"loss": 1.732, "grad_norm": 0.3537365198135376, "learning_rate": 0.0002, "epoch": 0.6499813223757938, "step": 870}, {"loss": 1.7825, "grad_norm": 0.33854469656944275, "learning_rate": 0.0002, "epoch": 0.6574523720582742, "step": 880}, {"loss": 1.7561, "grad_norm": 0.3332272469997406, "learning_rate": 0.0002, "epoch": 0.6649234217407546, "step": 890}, {"loss": 1.7247, "grad_norm": 0.34954726696014404, "learning_rate": 0.0002, "epoch": 0.6723944714232349, "step": 900}, {"loss": 1.7917, "grad_norm": 0.2921750247478485, "learning_rate": 0.0002, "epoch": 0.6798655211057153, "step": 910}, {"loss": 1.7807, "grad_norm": 0.30508682131767273, "learning_rate": 0.0002, "epoch": 0.6873365707881958, "step": 920}, {"loss": 1.8082, "grad_norm": 0.32268425822257996, "learning_rate": 0.0002, "epoch": 0.6948076204706761, "step": 930}, {"loss": 1.8283, "grad_norm": 0.2844390869140625, "learning_rate": 0.0002, "epoch": 0.7022786701531565, "step": 940}, {"loss": 1.7363, "grad_norm": 0.31263890862464905, "learning_rate": 0.0002, "epoch": 0.709749719835637, "step": 950}, {"loss": 1.8081, "grad_norm": 0.3626808822154999, "learning_rate": 0.0002, "epoch": 0.7172207695181173, "step": 960}, {"loss": 1.853, "grad_norm": 0.3322749733924866, "learning_rate": 0.0002, "epoch": 0.7246918192005977, "step": 970}, {"loss": 1.7912, "grad_norm": 0.29177871346473694, "learning_rate": 0.0002, "epoch": 0.732162868883078, "step": 980}, {"loss": 1.8447, "grad_norm": 0.35405513644218445, "learning_rate": 0.0002, "epoch": 0.7396339185655585, "step": 990}, {"loss": 1.7008, "grad_norm": 0.39318400621414185, "learning_rate": 0.0002, "epoch": 0.7471049682480388, "step": 1000}, {"loss": 1.7803, "grad_norm": 0.29401418566703796, "learning_rate": 0.0002, "epoch": 0.7545760179305192, "step": 1010}, {"loss": 1.7649, "grad_norm": 0.3271748721599579, "learning_rate": 0.0002, "epoch": 0.7620470676129997, "step": 1020}, {"loss": 1.7266, "grad_norm": 0.30883970856666565, "learning_rate": 0.0002, "epoch": 0.76951811729548, "step": 1030}, {"loss": 1.7722, "grad_norm": 0.3411838412284851, "learning_rate": 0.0002, "epoch": 0.7769891669779604, "step": 1040}, {"loss": 1.829, "grad_norm": 0.30608129501342773, "learning_rate": 0.0002, "epoch": 0.7844602166604407, "step": 1050}, {"loss": 1.7815, "grad_norm": 0.30899080634117126, "learning_rate": 0.0002, "epoch": 0.7919312663429212, "step": 1060}, {"loss": 1.7625, "grad_norm": 0.3160453140735626, "learning_rate": 0.0002, "epoch": 0.7994023160254016, "step": 1070}, {"loss": 1.8452, "grad_norm": 0.30947187542915344, "learning_rate": 0.0002, "epoch": 0.8068733657078819, "step": 1080}, {"loss": 1.7418, "grad_norm": 0.3103134036064148, "learning_rate": 0.0002, "epoch": 0.8143444153903624, "step": 1090}, {"loss": 1.842, "grad_norm": 0.31771138310432434, "learning_rate": 0.0002, "epoch": 0.8218154650728428, "step": 1100}, {"loss": 1.7918, "grad_norm": 0.5860997438430786, "learning_rate": 0.0002, "epoch": 0.8292865147553231, "step": 1110}, {"loss": 1.8443, "grad_norm": 0.3230148255825043, "learning_rate": 0.0002, "epoch": 0.8367575644378035, "step": 1120}, {"loss": 1.8478, "grad_norm": 0.29611510038375854, "learning_rate": 0.0002, "epoch": 0.8442286141202839, "step": 1130}, {"loss": 1.7673, "grad_norm": 0.3373654782772064, "learning_rate": 0.0002, "epoch": 0.8516996638027643, "step": 1140}, {"loss": 1.7997, "grad_norm": 0.3474279046058655, "learning_rate": 0.0002, "epoch": 0.8591707134852447, "step": 1150}, {"loss": 1.75, "grad_norm": 0.35057875514030457, "learning_rate": 0.0002, "epoch": 0.866641763167725, "step": 1160}, {"loss": 1.8273, "grad_norm": 0.39537495374679565, "learning_rate": 0.0002, "epoch": 0.8741128128502055, "step": 1170}, {"loss": 1.7682, "grad_norm": 0.3714233636856079, "learning_rate": 0.0002, "epoch": 0.8815838625326858, "step": 1180}, {"loss": 1.7549, "grad_norm": 0.2950296998023987, "learning_rate": 0.0002, "epoch": 0.8890549122151662, "step": 1190}, {"loss": 1.7612, "grad_norm": 0.38182979822158813, "learning_rate": 0.0002, "epoch": 0.8965259618976467, "step": 1200}, {"loss": 1.827, "grad_norm": 0.27883678674697876, "learning_rate": 0.0002, "epoch": 0.903997011580127, "step": 1210}, {"loss": 1.7623, "grad_norm": 0.33874374628067017, "learning_rate": 0.0002, "epoch": 0.9114680612626074, "step": 1220}, {"loss": 1.7334, "grad_norm": 0.3014272153377533, "learning_rate": 0.0002, "epoch": 0.9189391109450877, "step": 1230}, {"loss": 1.8235, "grad_norm": 0.3194271922111511, "learning_rate": 0.0002, "epoch": 0.9264101606275682, "step": 1240}, {"loss": 1.7924, "grad_norm": 0.3049403429031372, "learning_rate": 0.0002, "epoch": 0.9338812103100486, "step": 1250}, {"loss": 1.7535, "grad_norm": 0.30621254444122314, "learning_rate": 0.0002, "epoch": 0.9413522599925289, "step": 1260}, {"loss": 1.8287, "grad_norm": 0.28675132989883423, "learning_rate": 0.0002, "epoch": 0.9488233096750094, "step": 1270}, {"loss": 1.7586, "grad_norm": 0.3322032690048218, "learning_rate": 0.0002, "epoch": 0.9562943593574897, "step": 1280}, {"loss": 1.8054, "grad_norm": 0.35408294200897217, "learning_rate": 0.0002, "epoch": 0.9637654090399701, "step": 1290}, {"loss": 1.7343, "grad_norm": 0.36386919021606445, "learning_rate": 0.0002, "epoch": 0.9712364587224505, "step": 1300}, {"loss": 1.8633, "grad_norm": 0.32338324189186096, "learning_rate": 0.0002, "epoch": 0.9787075084049309, "step": 1310}, {"loss": 1.7724, "grad_norm": 0.3714013993740082, "learning_rate": 0.0002, "epoch": 0.9861785580874113, "step": 1320}, {"loss": 1.7766, "grad_norm": 0.3133082389831543, "learning_rate": 0.0002, "epoch": 0.9936496077698916, "step": 1330}, {"eval_loss": 1.8051470518112183, "eval_runtime": 38.6332, "eval_samples_per_second": 13.331, "eval_steps_per_second": 1.682, "epoch": 0.9996264475158759, "step": 1338}, {"loss": 1.8035, "grad_norm": 0.31595754623413086, "learning_rate": 0.0002, "epoch": 1.001120657452372, "step": 1340}, {"loss": 1.7486, "grad_norm": 0.3095700144767761, "learning_rate": 0.0002, "epoch": 1.0085917071348525, "step": 1350}, {"loss": 1.6981, "grad_norm": 0.34677496552467346, "learning_rate": 0.0002, "epoch": 1.0160627568173328, "step": 1360}, {"loss": 1.7377, "grad_norm": 0.29108840227127075, "learning_rate": 0.0002, "epoch": 1.0235338064998132, "step": 1370}, {"loss": 1.7194, "grad_norm": 0.32356950640678406, "learning_rate": 0.0002, "epoch": 1.0310048561822935, "step": 1380}, {"loss": 1.7593, "grad_norm": 0.4200669229030609, "learning_rate": 0.0002, "epoch": 1.038475905864774, "step": 1390}, {"loss": 1.797, "grad_norm": 0.3283711373806, "learning_rate": 0.0002, "epoch": 1.0459469555472545, "step": 1400}, {"loss": 1.7163, "grad_norm": 0.32898256182670593, "learning_rate": 0.0002, "epoch": 1.0534180052297348, "step": 1410}, {"loss": 1.7559, "grad_norm": 0.38790300488471985, "learning_rate": 0.0002, "epoch": 1.0608890549122152, "step": 1420}, {"loss": 1.6922, "grad_norm": 0.339800089597702, "learning_rate": 0.0002, "epoch": 1.0683601045946955, "step": 1430}, {"loss": 1.7076, "grad_norm": 0.3548751175403595, "learning_rate": 0.0002, "epoch": 1.075831154277176, "step": 1440}, {"loss": 1.6985, "grad_norm": 0.35114359855651855, "learning_rate": 0.0002, "epoch": 1.0833022039596563, "step": 1450}, {"loss": 1.7217, "grad_norm": 0.35226720571517944, "learning_rate": 0.0002, "epoch": 1.0907732536421366, "step": 1460}, {"loss": 1.6822, "grad_norm": 0.33665576577186584, "learning_rate": 0.0002, "epoch": 1.0982443033246172, "step": 1470}, {"loss": 1.6699, "grad_norm": 0.363889217376709, "learning_rate": 0.0002, "epoch": 1.1057153530070976, "step": 1480}, {"loss": 1.7933, "grad_norm": 0.3826201856136322, "learning_rate": 0.0002, "epoch": 1.113186402689578, "step": 1490}, {"loss": 1.7022, "grad_norm": 0.34058740735054016, "learning_rate": 0.0002, "epoch": 1.1206574523720583, "step": 1500}, {"loss": 1.6375, "grad_norm": 0.3462134301662445, "learning_rate": 0.0002, "epoch": 1.1281285020545386, "step": 1510}, {"loss": 1.7147, "grad_norm": 0.3396756052970886, "learning_rate": 0.0002, "epoch": 1.135599551737019, "step": 1520}, {"loss": 1.7219, "grad_norm": 0.32004743814468384, "learning_rate": 0.0002, "epoch": 1.1430706014194993, "step": 1530}, {"loss": 1.743, "grad_norm": 0.3397733271121979, "learning_rate": 0.0002, "epoch": 1.15054165110198, "step": 1540}, {"loss": 1.7333, "grad_norm": 0.3783262073993683, "learning_rate": 0.0002, "epoch": 1.1580127007844603, "step": 1550}, {"loss": 1.6075, "grad_norm": 0.35121291875839233, "learning_rate": 0.0002, "epoch": 1.1654837504669406, "step": 1560}, {"loss": 1.678, "grad_norm": 0.35816895961761475, "learning_rate": 0.0002, "epoch": 1.172954800149421, "step": 1570}, {"loss": 1.7143, "grad_norm": 0.33843839168548584, "learning_rate": 0.0002, "epoch": 1.1804258498319014, "step": 1580}, {"loss": 1.7434, "grad_norm": 0.3371972143650055, "learning_rate": 0.0002, "epoch": 1.1878968995143817, "step": 1590}, {"loss": 1.7671, "grad_norm": 0.36016878485679626, "learning_rate": 0.0002, "epoch": 1.195367949196862, "step": 1600}, {"loss": 1.6914, "grad_norm": 0.40879473090171814, "learning_rate": 0.0002, "epoch": 1.2028389988793426, "step": 1610}, {"loss": 1.6955, "grad_norm": 0.3216715455055237, "learning_rate": 0.0002, "epoch": 1.210310048561823, "step": 1620}, {"loss": 1.632, "grad_norm": 0.4482610821723938, "learning_rate": 0.0002, "epoch": 1.2177810982443034, "step": 1630}, {"loss": 1.6999, "grad_norm": 0.3257700502872467, "learning_rate": 0.0002, "epoch": 1.2252521479267837, "step": 1640}, {"loss": 1.7177, "grad_norm": 0.38646459579467773, "learning_rate": 0.0002, "epoch": 1.232723197609264, "step": 1650}, {"loss": 1.7081, "grad_norm": 0.4081360697746277, "learning_rate": 0.0002, "epoch": 1.2401942472917444, "step": 1660}, {"loss": 1.7519, "grad_norm": 0.4326848089694977, "learning_rate": 0.0002, "epoch": 1.2476652969742248, "step": 1670}, {"loss": 1.6752, "grad_norm": 0.346401572227478, "learning_rate": 0.0002, "epoch": 1.2551363466567054, "step": 1680}, {"loss": 1.7425, "grad_norm": 0.34536251425743103, "learning_rate": 0.0002, "epoch": 1.2626073963391857, "step": 1690}, {"loss": 1.7061, "grad_norm": 0.41359591484069824, "learning_rate": 0.0002, "epoch": 1.270078446021666, "step": 1700}, {"loss": 1.7906, "grad_norm": 0.3530874252319336, "learning_rate": 0.0002, "epoch": 1.2775494957041464, "step": 1710}, {"loss": 1.7357, "grad_norm": 0.3702719211578369, "learning_rate": 0.0002, "epoch": 1.2850205453866268, "step": 1720}, {"loss": 1.766, "grad_norm": 0.3703329563140869, "learning_rate": 0.0002, "epoch": 1.2924915950691072, "step": 1730}, {"loss": 1.7221, "grad_norm": 0.37919729948043823, "learning_rate": 0.0002, "epoch": 1.2999626447515875, "step": 1740}, {"loss": 1.7859, "grad_norm": 0.32526856660842896, "learning_rate": 0.0002, "epoch": 1.307433694434068, "step": 1750}, {"loss": 1.7117, "grad_norm": 0.36752620339393616, "learning_rate": 0.0002, "epoch": 1.3149047441165485, "step": 1760}, {"loss": 1.7335, "grad_norm": 0.3398192524909973, "learning_rate": 0.0002, "epoch": 1.3223757937990288, "step": 1770}, {"loss": 1.7492, "grad_norm": 0.37435585260391235, "learning_rate": 0.0002, "epoch": 1.3298468434815092, "step": 1780}, {"loss": 1.7393, "grad_norm": 0.35793280601501465, "learning_rate": 0.0002, "epoch": 1.3373178931639895, "step": 1790}, {"loss": 1.7266, "grad_norm": 0.35481882095336914, "learning_rate": 0.0002, "epoch": 1.3447889428464699, "step": 1800}, {"loss": 1.7456, "grad_norm": 0.3786393105983734, "learning_rate": 0.0002, "epoch": 1.3522599925289502, "step": 1810}, {"loss": 1.7169, "grad_norm": 0.33245593309402466, "learning_rate": 0.0002, "epoch": 1.3597310422114308, "step": 1820}, {"loss": 1.7577, "grad_norm": 0.35388344526290894, "learning_rate": 0.0002, "epoch": 1.3672020918939112, "step": 1830}, {"loss": 1.6968, "grad_norm": 0.3695325553417206, "learning_rate": 0.0002, "epoch": 1.3746731415763915, "step": 1840}, {"loss": 1.7086, "grad_norm": 0.3683604598045349, "learning_rate": 0.0002, "epoch": 1.382144191258872, "step": 1850}, {"loss": 1.7878, "grad_norm": 0.3753012418746948, "learning_rate": 0.0002, "epoch": 1.3896152409413522, "step": 1860}, {"loss": 1.6969, "grad_norm": 0.3331069350242615, "learning_rate": 0.0002, "epoch": 1.3970862906238326, "step": 1870}, {"loss": 1.6644, "grad_norm": 0.3877500295639038, "learning_rate": 0.0002, "epoch": 1.404557340306313, "step": 1880}, {"loss": 1.7586, "grad_norm": 0.33525151014328003, "learning_rate": 0.0002, "epoch": 1.4120283899887935, "step": 1890}, {"loss": 1.7031, "grad_norm": 0.3697299659252167, "learning_rate": 0.0002, "epoch": 1.4194994396712737, "step": 1900}, {"loss": 1.6956, "grad_norm": 0.4029286205768585, "learning_rate": 0.0002, "epoch": 1.4269704893537543, "step": 1910}, {"loss": 1.6897, "grad_norm": 0.3596203029155731, "learning_rate": 0.0002, "epoch": 1.4344415390362346, "step": 1920}, {"loss": 1.7139, "grad_norm": 0.450783908367157, "learning_rate": 0.0002, "epoch": 1.441912588718715, "step": 1930}, {"loss": 1.7243, "grad_norm": 0.3651481866836548, "learning_rate": 0.0002, "epoch": 1.4493836384011953, "step": 1940}, {"loss": 1.6637, "grad_norm": 0.3608424663543701, "learning_rate": 0.0002, "epoch": 1.4568546880836757, "step": 1950}, {"loss": 1.8285, "grad_norm": 0.39684420824050903, "learning_rate": 0.0002, "epoch": 1.4643257377661563, "step": 1960}, {"loss": 1.7514, "grad_norm": 0.34618663787841797, "learning_rate": 0.0002, "epoch": 1.4717967874486364, "step": 1970}, {"loss": 1.6655, "grad_norm": 0.4150386452674866, "learning_rate": 0.0002, "epoch": 1.479267837131117, "step": 1980}, {"loss": 1.7021, "grad_norm": 0.35500776767730713, "learning_rate": 0.0002, "epoch": 1.4867388868135973, "step": 1990}, {"loss": 1.7322, "grad_norm": 0.344144344329834, "learning_rate": 0.0002, "epoch": 1.4942099364960777, "step": 2000}, {"loss": 1.6998, "grad_norm": 0.3340149223804474, "learning_rate": 0.0002, "epoch": 1.501680986178558, "step": 2010}, {"loss": 1.7508, "grad_norm": 0.37685006856918335, "learning_rate": 0.0002, "epoch": 1.5091520358610384, "step": 2020}, {"loss": 1.8299, "grad_norm": 0.3699876368045807, "learning_rate": 0.0002, "epoch": 1.516623085543519, "step": 2030}, {"loss": 1.7357, "grad_norm": 0.3370307385921478, "learning_rate": 0.0002, "epoch": 1.5240941352259991, "step": 2040}, {"loss": 1.8044, "grad_norm": 0.37780630588531494, "learning_rate": 0.0002, "epoch": 1.5315651849084797, "step": 2050}, {"loss": 1.7408, "grad_norm": 0.370259165763855, "learning_rate": 0.0002, "epoch": 1.53903623459096, "step": 2060}, {"loss": 1.7398, "grad_norm": 0.3440011441707611, "learning_rate": 0.0002, "epoch": 1.5465072842734404, "step": 2070}, {"loss": 1.7105, "grad_norm": 0.40382063388824463, "learning_rate": 0.0002, "epoch": 1.5539783339559208, "step": 2080}, {"loss": 1.7071, "grad_norm": 0.38002029061317444, "learning_rate": 0.0002, "epoch": 1.5614493836384011, "step": 2090}, {"loss": 1.6815, "grad_norm": 0.3658451437950134, "learning_rate": 0.0002, "epoch": 1.5689204333208817, "step": 2100}, {"loss": 1.7598, "grad_norm": 0.354842871427536, "learning_rate": 0.0002, "epoch": 1.5763914830033618, "step": 2110}, {"loss": 1.6898, "grad_norm": 0.34735530614852905, "learning_rate": 0.0002, "epoch": 1.5838625326858424, "step": 2120}, {"loss": 1.7363, "grad_norm": 0.377581924200058, "learning_rate": 0.0002, "epoch": 1.5913335823683228, "step": 2130}, {"loss": 1.7789, "grad_norm": 0.41254034638404846, "learning_rate": 0.0002, "epoch": 1.5988046320508031, "step": 2140}, {"loss": 1.6782, "grad_norm": 0.3630715310573578, "learning_rate": 0.0002, "epoch": 1.6062756817332835, "step": 2150}, {"loss": 1.7531, "grad_norm": 0.36980143189430237, "learning_rate": 0.0002, "epoch": 1.6137467314157639, "step": 2160}, {"loss": 1.6847, "grad_norm": 0.3634769320487976, "learning_rate": 0.0002, "epoch": 1.6212177810982444, "step": 2170}, {"loss": 1.6367, "grad_norm": 0.3794139623641968, "learning_rate": 0.0002, "epoch": 1.6286888307807246, "step": 2180}, {"loss": 1.7064, "grad_norm": 0.359742134809494, "learning_rate": 0.0002, "epoch": 1.6361598804632052, "step": 2190}, {"loss": 1.7027, "grad_norm": 0.3770543932914734, "learning_rate": 0.0002, "epoch": 1.6436309301456855, "step": 2200}, {"loss": 1.784, "grad_norm": 0.3797036409378052, "learning_rate": 0.0002, "epoch": 1.6511019798281659, "step": 2210}, {"loss": 1.7875, "grad_norm": 0.35622093081474304, "learning_rate": 0.0002, "epoch": 1.6585730295106462, "step": 2220}, {"loss": 1.6615, "grad_norm": 0.34552520513534546, "learning_rate": 0.0002, "epoch": 1.6660440791931266, "step": 2230}, {"loss": 1.7522, "grad_norm": 0.379926860332489, "learning_rate": 0.0002, "epoch": 1.6735151288756072, "step": 2240}, {"loss": 1.7953, "grad_norm": 0.37083810567855835, "learning_rate": 0.0002, "epoch": 1.6809861785580873, "step": 2250}, {"loss": 1.7485, "grad_norm": 0.42746543884277344, "learning_rate": 0.0002, "epoch": 1.6884572282405679, "step": 2260}, {"loss": 1.776, "grad_norm": 0.3372884690761566, "learning_rate": 0.0002, "epoch": 1.6959282779230482, "step": 2270}, {"loss": 1.7604, "grad_norm": 0.35220256447792053, "learning_rate": 0.0002, "epoch": 1.7033993276055286, "step": 2280}, {"loss": 1.7154, "grad_norm": 0.3659130930900574, "learning_rate": 0.0002, "epoch": 1.710870377288009, "step": 2290}, {"loss": 1.6953, "grad_norm": 0.37629297375679016, "learning_rate": 0.0002, "epoch": 1.7183414269704893, "step": 2300}, {"loss": 1.7212, "grad_norm": 0.36312398314476013, "learning_rate": 0.0002, "epoch": 1.7258124766529699, "step": 2310}, {"loss": 1.7903, "grad_norm": 0.467709481716156, "learning_rate": 0.0002, "epoch": 1.73328352633545, "step": 2320}, {"loss": 1.696, "grad_norm": 0.38685527443885803, "learning_rate": 0.0002, "epoch": 1.7407545760179306, "step": 2330}, {"loss": 1.7041, "grad_norm": 0.3578338325023651, "learning_rate": 0.0002, "epoch": 1.748225625700411, "step": 2340}, {"loss": 1.6456, "grad_norm": 0.36057502031326294, "learning_rate": 0.0002, "epoch": 1.7556966753828913, "step": 2350}, {"loss": 1.6853, "grad_norm": 0.3615196645259857, "learning_rate": 0.0002, "epoch": 1.7631677250653717, "step": 2360}, {"loss": 1.7612, "grad_norm": 0.4118947684764862, "learning_rate": 0.0002, "epoch": 1.770638774747852, "step": 2370}, {"loss": 1.6946, "grad_norm": 0.4067276120185852, "learning_rate": 0.0002, "epoch": 1.7781098244303326, "step": 2380}, {"loss": 1.712, "grad_norm": 0.3979823887348175, "learning_rate": 0.0002, "epoch": 1.7855808741128127, "step": 2390}, {"loss": 1.7644, "grad_norm": 0.44045883417129517, "learning_rate": 0.0002, "epoch": 1.7930519237952933, "step": 2400}, {"loss": 1.7251, "grad_norm": 0.3998069167137146, "learning_rate": 0.0002, "epoch": 1.8005229734777737, "step": 2410}, {"loss": 1.7354, "grad_norm": 0.3450094759464264, "learning_rate": 0.0002, "epoch": 1.807994023160254, "step": 2420}, {"loss": 1.6998, "grad_norm": 0.3759009838104248, "learning_rate": 0.0002, "epoch": 1.8154650728427344, "step": 2430}, {"loss": 1.7706, "grad_norm": 0.34347015619277954, "learning_rate": 0.0002, "epoch": 1.8229361225252148, "step": 2440}, {"loss": 1.7345, "grad_norm": 0.3511228859424591, "learning_rate": 0.0002, "epoch": 1.8304071722076953, "step": 2450}, {"loss": 1.6909, "grad_norm": 0.36853715777397156, "learning_rate": 0.0002, "epoch": 1.8378782218901755, "step": 2460}, {"loss": 1.6931, "grad_norm": 0.40659376978874207, "learning_rate": 0.0002, "epoch": 1.845349271572656, "step": 2470}, {"loss": 1.7626, "grad_norm": 0.39621320366859436, "learning_rate": 0.0002, "epoch": 1.8528203212551362, "step": 2480}, {"loss": 1.7427, "grad_norm": 0.3753979504108429, "learning_rate": 0.0002, "epoch": 1.8602913709376168, "step": 2490}, {"loss": 1.6622, "grad_norm": 0.3811938464641571, "learning_rate": 0.0002, "epoch": 1.8677624206200971, "step": 2500}, {"loss": 1.7718, "grad_norm": 0.3432596027851105, "learning_rate": 0.0002, "epoch": 1.8752334703025775, "step": 2510}, {"loss": 1.7488, "grad_norm": 0.3670712113380432, "learning_rate": 0.0002, "epoch": 1.882704519985058, "step": 2520}, {"loss": 1.705, "grad_norm": 0.40907177329063416, "learning_rate": 0.0002, "epoch": 1.8901755696675382, "step": 2530}, {"loss": 1.7148, "grad_norm": 0.3821999728679657, "learning_rate": 0.0002, "epoch": 1.8976466193500188, "step": 2540}, {"loss": 1.7934, "grad_norm": 0.36173978447914124, "learning_rate": 0.0002, "epoch": 1.905117669032499, "step": 2550}, {"loss": 1.6939, "grad_norm": 0.38990336656570435, "learning_rate": 0.0002, "epoch": 1.9125887187149795, "step": 2560}, {"loss": 1.6893, "grad_norm": 0.35242322087287903, "learning_rate": 0.0002, "epoch": 1.9200597683974598, "step": 2570}, {"loss": 1.7268, "grad_norm": 0.3506428003311157, "learning_rate": 0.0002, "epoch": 1.9275308180799402, "step": 2580}, {"loss": 1.6953, "grad_norm": 0.39540135860443115, "learning_rate": 0.0002, "epoch": 1.9350018677624208, "step": 2590}, {"loss": 1.6511, "grad_norm": 0.3444725573062897, "learning_rate": 0.0002, "epoch": 1.942472917444901, "step": 2600}, {"loss": 1.7259, "grad_norm": 0.3963521718978882, "learning_rate": 0.0002, "epoch": 1.9499439671273815, "step": 2610}, {"loss": 1.6946, "grad_norm": 0.3689815402030945, "learning_rate": 0.0002, "epoch": 1.9574150168098616, "step": 2620}, {"loss": 1.7384, "grad_norm": 0.3482626676559448, "learning_rate": 0.0002, "epoch": 1.9648860664923422, "step": 2630}, {"loss": 1.7048, "grad_norm": 0.35832616686820984, "learning_rate": 0.0002, "epoch": 1.9723571161748226, "step": 2640}, {"loss": 1.6681, "grad_norm": 0.4776208996772766, "learning_rate": 0.0002, "epoch": 1.979828165857303, "step": 2650}, {"loss": 1.6696, "grad_norm": 0.32570165395736694, "learning_rate": 0.0002, "epoch": 1.9872992155397835, "step": 2660}, {"loss": 1.7232, "grad_norm": 0.3380725085735321, "learning_rate": 0.0002, "epoch": 1.9947702652222636, "step": 2670}]}
+{"epoch": 2.999626447515876, "step": 4015, "epoch_duration": 1438.387151002884, "total_accumulated_duration": 4312.949033737183, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7672.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-42/checkpoint-2677", "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": 2.6181, "grad_norm": 0.4912872612476349, "learning_rate": 0.0002, "epoch": 0.007471049682480389, "step": 10}, {"loss": 2.2606, "grad_norm": 0.4856316149234772, "learning_rate": 0.0002, "epoch": 0.014942099364960777, "step": 20}, {"loss": 2.0957, "grad_norm": 0.47683125734329224, "learning_rate": 0.0002, "epoch": 0.022413149047441166, "step": 30}, {"loss": 1.8908, "grad_norm": 0.515082597732544, "learning_rate": 0.0002, "epoch": 0.029884198729921554, "step": 40}, {"loss": 1.9704, "grad_norm": 0.5299215316772461, "learning_rate": 0.0002, "epoch": 0.03735524841240194, "step": 50}, {"loss": 1.9225, "grad_norm": 0.4951399862766266, "learning_rate": 0.0002, "epoch": 0.04482629809488233, "step": 60}, {"loss": 1.9742, "grad_norm": 0.48079821467399597, "learning_rate": 0.0002, "epoch": 0.05229734777736272, "step": 70}, {"loss": 1.9466, "grad_norm": 0.49402132630348206, "learning_rate": 0.0002, "epoch": 0.05976839745984311, "step": 80}, {"loss": 1.8691, "grad_norm": 0.4778193235397339, "learning_rate": 0.0002, "epoch": 0.0672394471423235, "step": 90}, {"loss": 1.8455, "grad_norm": 0.42472657561302185, "learning_rate": 0.0002, "epoch": 0.07471049682480388, "step": 100}, {"loss": 1.8744, "grad_norm": 0.4433092474937439, "learning_rate": 0.0002, "epoch": 0.08218154650728428, "step": 110}, {"loss": 1.865, "grad_norm": 0.4472862780094147, "learning_rate": 0.0002, "epoch": 0.08965259618976466, "step": 120}, {"loss": 1.9256, "grad_norm": 0.42596298456192017, "learning_rate": 0.0002, "epoch": 0.09712364587224505, "step": 130}, {"loss": 1.8015, "grad_norm": 0.46645811200141907, "learning_rate": 0.0002, "epoch": 0.10459469555472543, "step": 140}, {"loss": 1.8307, "grad_norm": 0.41041234135627747, "learning_rate": 0.0002, "epoch": 0.11206574523720583, "step": 150}, {"loss": 1.8276, "grad_norm": 0.5329819917678833, "learning_rate": 0.0002, "epoch": 0.11953679491968622, "step": 160}, {"loss": 1.8118, "grad_norm": 0.4065922200679779, "learning_rate": 0.0002, "epoch": 0.1270078446021666, "step": 170}, {"loss": 1.8559, "grad_norm": 0.38406994938850403, "learning_rate": 0.0002, "epoch": 0.134478894284647, "step": 180}, {"loss": 1.8647, "grad_norm": 0.4246881306171417, "learning_rate": 0.0002, "epoch": 0.14194994396712737, "step": 190}, {"loss": 1.8054, "grad_norm": 0.35136649012565613, "learning_rate": 0.0002, "epoch": 0.14942099364960776, "step": 200}, {"loss": 1.802, "grad_norm": 0.43252742290496826, "learning_rate": 0.0002, "epoch": 0.15689204333208817, "step": 210}, {"loss": 1.7823, "grad_norm": 0.39236941933631897, "learning_rate": 0.0002, "epoch": 0.16436309301456856, "step": 220}, {"loss": 1.818, "grad_norm": 0.3748249113559723, "learning_rate": 0.0002, "epoch": 0.17183414269704894, "step": 230}, {"loss": 1.866, "grad_norm": 0.6432855725288391, "learning_rate": 0.0002, "epoch": 0.17930519237952933, "step": 240}, {"loss": 1.8397, "grad_norm": 0.34874802827835083, "learning_rate": 0.0002, "epoch": 0.1867762420620097, "step": 250}, {"loss": 1.79, "grad_norm": 0.3721984326839447, "learning_rate": 0.0002, "epoch": 0.1942472917444901, "step": 260}, {"loss": 1.8464, "grad_norm": 0.4339311420917511, "learning_rate": 0.0002, "epoch": 0.20171834142697048, "step": 270}, {"loss": 1.8665, "grad_norm": 0.4018215537071228, "learning_rate": 0.0002, "epoch": 0.20918939110945087, "step": 280}, {"loss": 1.8048, "grad_norm": 0.3278839886188507, "learning_rate": 0.0002, "epoch": 0.21666044079193125, "step": 290}, {"loss": 1.7395, "grad_norm": 0.36146077513694763, "learning_rate": 0.0002, "epoch": 0.22413149047441167, "step": 300}, {"loss": 1.7916, "grad_norm": 0.38175010681152344, "learning_rate": 0.0002, "epoch": 0.23160254015689205, "step": 310}, {"loss": 1.8593, "grad_norm": 0.44776618480682373, "learning_rate": 0.0002, "epoch": 0.23907358983937244, "step": 320}, {"loss": 1.7824, "grad_norm": 0.3933652937412262, "learning_rate": 0.0002, "epoch": 0.24654463952185282, "step": 330}, {"loss": 1.8393, "grad_norm": 0.3515005111694336, "learning_rate": 0.0002, "epoch": 0.2540156892043332, "step": 340}, {"loss": 1.8653, "grad_norm": 0.6683304309844971, "learning_rate": 0.0002, "epoch": 0.2614867388868136, "step": 350}, {"loss": 1.8797, "grad_norm": 0.37093454599380493, "learning_rate": 0.0002, "epoch": 0.268957788569294, "step": 360}, {"loss": 1.8251, "grad_norm": 0.3450651168823242, "learning_rate": 0.0002, "epoch": 0.2764288382517744, "step": 370}, {"loss": 1.7435, "grad_norm": 0.5140917301177979, "learning_rate": 0.0002, "epoch": 0.28389988793425475, "step": 380}, {"loss": 1.8026, "grad_norm": 0.32885563373565674, "learning_rate": 0.0002, "epoch": 0.29137093761673516, "step": 390}, {"loss": 1.8174, "grad_norm": 0.33962297439575195, "learning_rate": 0.0002, "epoch": 0.2988419872992155, "step": 400}, {"loss": 1.7467, "grad_norm": 0.3723141849040985, "learning_rate": 0.0002, "epoch": 0.30631303698169593, "step": 410}, {"loss": 1.8459, "grad_norm": 0.37173134088516235, "learning_rate": 0.0002, "epoch": 0.31378408666417634, "step": 420}, {"loss": 1.8876, "grad_norm": 0.33736956119537354, "learning_rate": 0.0002, "epoch": 0.3212551363466567, "step": 430}, {"loss": 1.8367, "grad_norm": 0.3602448105812073, "learning_rate": 0.0002, "epoch": 0.3287261860291371, "step": 440}, {"loss": 1.8058, "grad_norm": 0.3569699227809906, "learning_rate": 0.0002, "epoch": 0.33619723571161747, "step": 450}, {"loss": 1.8086, "grad_norm": 0.31009167432785034, "learning_rate": 0.0002, "epoch": 0.3436682853940979, "step": 460}, {"loss": 1.8876, "grad_norm": 0.5278693437576294, "learning_rate": 0.0002, "epoch": 0.35113933507657824, "step": 470}, {"loss": 1.8534, "grad_norm": 0.3587537109851837, "learning_rate": 0.0002, "epoch": 0.35861038475905865, "step": 480}, {"loss": 1.8046, "grad_norm": 0.3859670162200928, "learning_rate": 0.0002, "epoch": 0.366081434441539, "step": 490}, {"loss": 1.8287, "grad_norm": 0.395913690328598, "learning_rate": 0.0002, "epoch": 0.3735524841240194, "step": 500}, {"loss": 1.7619, "grad_norm": 0.35052940249443054, "learning_rate": 0.0002, "epoch": 0.38102353380649984, "step": 510}, {"loss": 1.7824, "grad_norm": 0.2979494333267212, "learning_rate": 0.0002, "epoch": 0.3884945834889802, "step": 520}, {"loss": 1.8641, "grad_norm": 0.3062683343887329, "learning_rate": 0.0002, "epoch": 0.3959656331714606, "step": 530}, {"loss": 1.7651, "grad_norm": 0.3172847330570221, "learning_rate": 0.0002, "epoch": 0.40343668285394096, "step": 540}, {"loss": 1.806, "grad_norm": 0.360435426235199, "learning_rate": 0.0002, "epoch": 0.4109077325364214, "step": 550}, {"loss": 1.9054, "grad_norm": 0.3427872359752655, "learning_rate": 0.0002, "epoch": 0.41837878221890173, "step": 560}, {"loss": 1.7562, "grad_norm": 0.34036558866500854, "learning_rate": 0.0002, "epoch": 0.42584983190138215, "step": 570}, {"loss": 1.7254, "grad_norm": 0.3365345299243927, "learning_rate": 0.0002, "epoch": 0.4333208815838625, "step": 580}, {"loss": 1.8328, "grad_norm": 0.35619041323661804, "learning_rate": 0.0002, "epoch": 0.4407919312663429, "step": 590}, {"loss": 1.8114, "grad_norm": 0.3569088280200958, "learning_rate": 0.0002, "epoch": 0.44826298094882333, "step": 600}, {"loss": 1.8599, "grad_norm": 0.3581278622150421, "learning_rate": 0.0002, "epoch": 0.4557340306313037, "step": 610}, {"loss": 1.7078, "grad_norm": 0.43197110295295715, "learning_rate": 0.0002, "epoch": 0.4632050803137841, "step": 620}, {"loss": 1.8257, "grad_norm": 0.33966198563575745, "learning_rate": 0.0002, "epoch": 0.47067612999626446, "step": 630}, {"loss": 1.7528, "grad_norm": 0.3343866467475891, "learning_rate": 0.0002, "epoch": 0.47814717967874487, "step": 640}, {"loss": 1.8191, "grad_norm": 0.33878564834594727, "learning_rate": 0.0002, "epoch": 0.48561822936122523, "step": 650}, {"loss": 1.8801, "grad_norm": 0.387195885181427, "learning_rate": 0.0002, "epoch": 0.49308927904370564, "step": 660}, {"loss": 1.7559, "grad_norm": 0.3755440413951874, "learning_rate": 0.0002, "epoch": 0.500560328726186, "step": 670}, {"loss": 1.8057, "grad_norm": 0.3272816836833954, "learning_rate": 0.0002, "epoch": 0.5080313784086664, "step": 680}, {"loss": 1.8156, "grad_norm": 0.36063864827156067, "learning_rate": 0.0002, "epoch": 0.5155024280911468, "step": 690}, {"loss": 1.8397, "grad_norm": 0.35317373275756836, "learning_rate": 0.0002, "epoch": 0.5229734777736272, "step": 700}, {"loss": 1.7603, "grad_norm": 0.3561195433139801, "learning_rate": 0.0002, "epoch": 0.5304445274561076, "step": 710}, {"loss": 1.8149, "grad_norm": 0.31124624609947205, "learning_rate": 0.0002, "epoch": 0.537915577138588, "step": 720}, {"loss": 1.7434, "grad_norm": 0.3294544517993927, "learning_rate": 0.0002, "epoch": 0.5453866268210683, "step": 730}, {"loss": 1.8027, "grad_norm": 0.31933900713920593, "learning_rate": 0.0002, "epoch": 0.5528576765035488, "step": 740}, {"loss": 1.7601, "grad_norm": 0.3226020634174347, "learning_rate": 0.0002, "epoch": 0.5603287261860291, "step": 750}, {"loss": 1.7862, "grad_norm": 0.3147525489330292, "learning_rate": 0.0002, "epoch": 0.5677997758685095, "step": 760}, {"loss": 1.9028, "grad_norm": 0.32234328985214233, "learning_rate": 0.0002, "epoch": 0.57527082555099, "step": 770}, {"loss": 1.7623, "grad_norm": 0.3258664309978485, "learning_rate": 0.0002, "epoch": 0.5827418752334703, "step": 780}, {"loss": 1.7384, "grad_norm": 0.3166961967945099, "learning_rate": 0.0002, "epoch": 0.5902129249159507, "step": 790}, {"loss": 1.8799, "grad_norm": 0.35621458292007446, "learning_rate": 0.0002, "epoch": 0.597683974598431, "step": 800}, {"loss": 1.8313, "grad_norm": 0.3236999213695526, "learning_rate": 0.0002, "epoch": 0.6051550242809115, "step": 810}, {"loss": 1.7132, "grad_norm": 0.2892923653125763, "learning_rate": 0.0002, "epoch": 0.6126260739633919, "step": 820}, {"loss": 1.8709, "grad_norm": 0.4098321497440338, "learning_rate": 0.0002, "epoch": 0.6200971236458722, "step": 830}, {"loss": 1.7637, "grad_norm": 0.3337118923664093, "learning_rate": 0.0002, "epoch": 0.6275681733283527, "step": 840}, {"loss": 1.7375, "grad_norm": 0.30416029691696167, "learning_rate": 0.0002, "epoch": 0.635039223010833, "step": 850}, {"loss": 1.7419, "grad_norm": 0.3361026346683502, "learning_rate": 0.0002, "epoch": 0.6425102726933134, "step": 860}, {"loss": 1.732, "grad_norm": 0.3537365198135376, "learning_rate": 0.0002, "epoch": 0.6499813223757938, "step": 870}, {"loss": 1.7825, "grad_norm": 0.33854469656944275, "learning_rate": 0.0002, "epoch": 0.6574523720582742, "step": 880}, {"loss": 1.7561, "grad_norm": 0.3332272469997406, "learning_rate": 0.0002, "epoch": 0.6649234217407546, "step": 890}, {"loss": 1.7247, "grad_norm": 0.34954726696014404, "learning_rate": 0.0002, "epoch": 0.6723944714232349, "step": 900}, {"loss": 1.7917, "grad_norm": 0.2921750247478485, "learning_rate": 0.0002, "epoch": 0.6798655211057153, "step": 910}, {"loss": 1.7807, "grad_norm": 0.30508682131767273, "learning_rate": 0.0002, "epoch": 0.6873365707881958, "step": 920}, {"loss": 1.8082, "grad_norm": 0.32268425822257996, "learning_rate": 0.0002, "epoch": 0.6948076204706761, "step": 930}, {"loss": 1.8283, "grad_norm": 0.2844390869140625, "learning_rate": 0.0002, "epoch": 0.7022786701531565, "step": 940}, {"loss": 1.7363, "grad_norm": 0.31263890862464905, "learning_rate": 0.0002, "epoch": 0.709749719835637, "step": 950}, {"loss": 1.8081, "grad_norm": 0.3626808822154999, "learning_rate": 0.0002, "epoch": 0.7172207695181173, "step": 960}, {"loss": 1.853, "grad_norm": 0.3322749733924866, "learning_rate": 0.0002, "epoch": 0.7246918192005977, "step": 970}, {"loss": 1.7912, "grad_norm": 0.29177871346473694, "learning_rate": 0.0002, "epoch": 0.732162868883078, "step": 980}, {"loss": 1.8447, "grad_norm": 0.35405513644218445, "learning_rate": 0.0002, "epoch": 0.7396339185655585, "step": 990}, {"loss": 1.7008, "grad_norm": 0.39318400621414185, "learning_rate": 0.0002, "epoch": 0.7471049682480388, "step": 1000}, {"loss": 1.7803, "grad_norm": 0.29401418566703796, "learning_rate": 0.0002, "epoch": 0.7545760179305192, "step": 1010}, {"loss": 1.7649, "grad_norm": 0.3271748721599579, "learning_rate": 0.0002, "epoch": 0.7620470676129997, "step": 1020}, {"loss": 1.7266, "grad_norm": 0.30883970856666565, "learning_rate": 0.0002, "epoch": 0.76951811729548, "step": 1030}, {"loss": 1.7722, "grad_norm": 0.3411838412284851, "learning_rate": 0.0002, "epoch": 0.7769891669779604, "step": 1040}, {"loss": 1.829, "grad_norm": 0.30608129501342773, "learning_rate": 0.0002, "epoch": 0.7844602166604407, "step": 1050}, {"loss": 1.7815, "grad_norm": 0.30899080634117126, "learning_rate": 0.0002, "epoch": 0.7919312663429212, "step": 1060}, {"loss": 1.7625, "grad_norm": 0.3160453140735626, "learning_rate": 0.0002, "epoch": 0.7994023160254016, "step": 1070}, {"loss": 1.8452, "grad_norm": 0.30947187542915344, "learning_rate": 0.0002, "epoch": 0.8068733657078819, "step": 1080}, {"loss": 1.7418, "grad_norm": 0.3103134036064148, "learning_rate": 0.0002, "epoch": 0.8143444153903624, "step": 1090}, {"loss": 1.842, "grad_norm": 0.31771138310432434, "learning_rate": 0.0002, "epoch": 0.8218154650728428, "step": 1100}, {"loss": 1.7918, "grad_norm": 0.5860997438430786, "learning_rate": 0.0002, "epoch": 0.8292865147553231, "step": 1110}, {"loss": 1.8443, "grad_norm": 0.3230148255825043, "learning_rate": 0.0002, "epoch": 0.8367575644378035, "step": 1120}, {"loss": 1.8478, "grad_norm": 0.29611510038375854, "learning_rate": 0.0002, "epoch": 0.8442286141202839, "step": 1130}, {"loss": 1.7673, "grad_norm": 0.3373654782772064, "learning_rate": 0.0002, "epoch": 0.8516996638027643, "step": 1140}, {"loss": 1.7997, "grad_norm": 0.3474279046058655, "learning_rate": 0.0002, "epoch": 0.8591707134852447, "step": 1150}, {"loss": 1.75, "grad_norm": 0.35057875514030457, "learning_rate": 0.0002, "epoch": 0.866641763167725, "step": 1160}, {"loss": 1.8273, "grad_norm": 0.39537495374679565, "learning_rate": 0.0002, "epoch": 0.8741128128502055, "step": 1170}, {"loss": 1.7682, "grad_norm": 0.3714233636856079, "learning_rate": 0.0002, "epoch": 0.8815838625326858, "step": 1180}, {"loss": 1.7549, "grad_norm": 0.2950296998023987, "learning_rate": 0.0002, "epoch": 0.8890549122151662, "step": 1190}, {"loss": 1.7612, "grad_norm": 0.38182979822158813, "learning_rate": 0.0002, "epoch": 0.8965259618976467, "step": 1200}, {"loss": 1.827, "grad_norm": 0.27883678674697876, "learning_rate": 0.0002, "epoch": 0.903997011580127, "step": 1210}, {"loss": 1.7623, "grad_norm": 0.33874374628067017, "learning_rate": 0.0002, "epoch": 0.9114680612626074, "step": 1220}, {"loss": 1.7334, "grad_norm": 0.3014272153377533, "learning_rate": 0.0002, "epoch": 0.9189391109450877, "step": 1230}, {"loss": 1.8235, "grad_norm": 0.3194271922111511, "learning_rate": 0.0002, "epoch": 0.9264101606275682, "step": 1240}, {"loss": 1.7924, "grad_norm": 0.3049403429031372, "learning_rate": 0.0002, "epoch": 0.9338812103100486, "step": 1250}, {"loss": 1.7535, "grad_norm": 0.30621254444122314, "learning_rate": 0.0002, "epoch": 0.9413522599925289, "step": 1260}, {"loss": 1.8287, "grad_norm": 0.28675132989883423, "learning_rate": 0.0002, "epoch": 0.9488233096750094, "step": 1270}, {"loss": 1.7586, "grad_norm": 0.3322032690048218, "learning_rate": 0.0002, "epoch": 0.9562943593574897, "step": 1280}, {"loss": 1.8054, "grad_norm": 0.35408294200897217, "learning_rate": 0.0002, "epoch": 0.9637654090399701, "step": 1290}, {"loss": 1.7343, "grad_norm": 0.36386919021606445, "learning_rate": 0.0002, "epoch": 0.9712364587224505, "step": 1300}, {"loss": 1.8633, "grad_norm": 0.32338324189186096, "learning_rate": 0.0002, "epoch": 0.9787075084049309, "step": 1310}, {"loss": 1.7724, "grad_norm": 0.3714013993740082, "learning_rate": 0.0002, "epoch": 0.9861785580874113, "step": 1320}, {"loss": 1.7766, "grad_norm": 0.3133082389831543, "learning_rate": 0.0002, "epoch": 0.9936496077698916, "step": 1330}, {"eval_loss": 1.8051470518112183, "eval_runtime": 38.6332, "eval_samples_per_second": 13.331, "eval_steps_per_second": 1.682, "epoch": 0.9996264475158759, "step": 1338}, {"loss": 1.8035, "grad_norm": 0.31595754623413086, "learning_rate": 0.0002, "epoch": 1.001120657452372, "step": 1340}, {"loss": 1.7486, "grad_norm": 0.3095700144767761, "learning_rate": 0.0002, "epoch": 1.0085917071348525, "step": 1350}, {"loss": 1.6981, "grad_norm": 0.34677496552467346, "learning_rate": 0.0002, "epoch": 1.0160627568173328, "step": 1360}, {"loss": 1.7377, "grad_norm": 0.29108840227127075, "learning_rate": 0.0002, "epoch": 1.0235338064998132, "step": 1370}, {"loss": 1.7194, "grad_norm": 0.32356950640678406, "learning_rate": 0.0002, "epoch": 1.0310048561822935, "step": 1380}, {"loss": 1.7593, "grad_norm": 0.4200669229030609, "learning_rate": 0.0002, "epoch": 1.038475905864774, "step": 1390}, {"loss": 1.797, "grad_norm": 0.3283711373806, "learning_rate": 0.0002, "epoch": 1.0459469555472545, "step": 1400}, {"loss": 1.7163, "grad_norm": 0.32898256182670593, "learning_rate": 0.0002, "epoch": 1.0534180052297348, "step": 1410}, {"loss": 1.7559, "grad_norm": 0.38790300488471985, "learning_rate": 0.0002, "epoch": 1.0608890549122152, "step": 1420}, {"loss": 1.6922, "grad_norm": 0.339800089597702, "learning_rate": 0.0002, "epoch": 1.0683601045946955, "step": 1430}, {"loss": 1.7076, "grad_norm": 0.3548751175403595, "learning_rate": 0.0002, "epoch": 1.075831154277176, "step": 1440}, {"loss": 1.6985, "grad_norm": 0.35114359855651855, "learning_rate": 0.0002, "epoch": 1.0833022039596563, "step": 1450}, {"loss": 1.7217, "grad_norm": 0.35226720571517944, "learning_rate": 0.0002, "epoch": 1.0907732536421366, "step": 1460}, {"loss": 1.6822, "grad_norm": 0.33665576577186584, "learning_rate": 0.0002, "epoch": 1.0982443033246172, "step": 1470}, {"loss": 1.6699, "grad_norm": 0.363889217376709, "learning_rate": 0.0002, "epoch": 1.1057153530070976, "step": 1480}, {"loss": 1.7933, "grad_norm": 0.3826201856136322, "learning_rate": 0.0002, "epoch": 1.113186402689578, "step": 1490}, {"loss": 1.7022, "grad_norm": 0.34058740735054016, "learning_rate": 0.0002, "epoch": 1.1206574523720583, "step": 1500}, {"loss": 1.6375, "grad_norm": 0.3462134301662445, "learning_rate": 0.0002, "epoch": 1.1281285020545386, "step": 1510}, {"loss": 1.7147, "grad_norm": 0.3396756052970886, "learning_rate": 0.0002, "epoch": 1.135599551737019, "step": 1520}, {"loss": 1.7219, "grad_norm": 0.32004743814468384, "learning_rate": 0.0002, "epoch": 1.1430706014194993, "step": 1530}, {"loss": 1.743, "grad_norm": 0.3397733271121979, "learning_rate": 0.0002, "epoch": 1.15054165110198, "step": 1540}, {"loss": 1.7333, "grad_norm": 0.3783262073993683, "learning_rate": 0.0002, "epoch": 1.1580127007844603, "step": 1550}, {"loss": 1.6075, "grad_norm": 0.35121291875839233, "learning_rate": 0.0002, "epoch": 1.1654837504669406, "step": 1560}, {"loss": 1.678, "grad_norm": 0.35816895961761475, "learning_rate": 0.0002, "epoch": 1.172954800149421, "step": 1570}, {"loss": 1.7143, "grad_norm": 0.33843839168548584, "learning_rate": 0.0002, "epoch": 1.1804258498319014, "step": 1580}, {"loss": 1.7434, "grad_norm": 0.3371972143650055, "learning_rate": 0.0002, "epoch": 1.1878968995143817, "step": 1590}, {"loss": 1.7671, "grad_norm": 0.36016878485679626, "learning_rate": 0.0002, "epoch": 1.195367949196862, "step": 1600}, {"loss": 1.6914, "grad_norm": 0.40879473090171814, "learning_rate": 0.0002, "epoch": 1.2028389988793426, "step": 1610}, {"loss": 1.6955, "grad_norm": 0.3216715455055237, "learning_rate": 0.0002, "epoch": 1.210310048561823, "step": 1620}, {"loss": 1.632, "grad_norm": 0.4482610821723938, "learning_rate": 0.0002, "epoch": 1.2177810982443034, "step": 1630}, {"loss": 1.6999, "grad_norm": 0.3257700502872467, "learning_rate": 0.0002, "epoch": 1.2252521479267837, "step": 1640}, {"loss": 1.7177, "grad_norm": 0.38646459579467773, "learning_rate": 0.0002, "epoch": 1.232723197609264, "step": 1650}, {"loss": 1.7081, "grad_norm": 0.4081360697746277, "learning_rate": 0.0002, "epoch": 1.2401942472917444, "step": 1660}, {"loss": 1.7519, "grad_norm": 0.4326848089694977, "learning_rate": 0.0002, "epoch": 1.2476652969742248, "step": 1670}, {"loss": 1.6752, "grad_norm": 0.346401572227478, "learning_rate": 0.0002, "epoch": 1.2551363466567054, "step": 1680}, {"loss": 1.7425, "grad_norm": 0.34536251425743103, "learning_rate": 0.0002, "epoch": 1.2626073963391857, "step": 1690}, {"loss": 1.7061, "grad_norm": 0.41359591484069824, "learning_rate": 0.0002, "epoch": 1.270078446021666, "step": 1700}, {"loss": 1.7906, "grad_norm": 0.3530874252319336, "learning_rate": 0.0002, "epoch": 1.2775494957041464, "step": 1710}, {"loss": 1.7357, "grad_norm": 0.3702719211578369, "learning_rate": 0.0002, "epoch": 1.2850205453866268, "step": 1720}, {"loss": 1.766, "grad_norm": 0.3703329563140869, "learning_rate": 0.0002, "epoch": 1.2924915950691072, "step": 1730}, {"loss": 1.7221, "grad_norm": 0.37919729948043823, "learning_rate": 0.0002, "epoch": 1.2999626447515875, "step": 1740}, {"loss": 1.7859, "grad_norm": 0.32526856660842896, "learning_rate": 0.0002, "epoch": 1.307433694434068, "step": 1750}, {"loss": 1.7117, "grad_norm": 0.36752620339393616, "learning_rate": 0.0002, "epoch": 1.3149047441165485, "step": 1760}, {"loss": 1.7335, "grad_norm": 0.3398192524909973, "learning_rate": 0.0002, "epoch": 1.3223757937990288, "step": 1770}, {"loss": 1.7492, "grad_norm": 0.37435585260391235, "learning_rate": 0.0002, "epoch": 1.3298468434815092, "step": 1780}, {"loss": 1.7393, "grad_norm": 0.35793280601501465, "learning_rate": 0.0002, "epoch": 1.3373178931639895, "step": 1790}, {"loss": 1.7266, "grad_norm": 0.35481882095336914, "learning_rate": 0.0002, "epoch": 1.3447889428464699, "step": 1800}, {"loss": 1.7456, "grad_norm": 0.3786393105983734, "learning_rate": 0.0002, "epoch": 1.3522599925289502, "step": 1810}, {"loss": 1.7169, "grad_norm": 0.33245593309402466, "learning_rate": 0.0002, "epoch": 1.3597310422114308, "step": 1820}, {"loss": 1.7577, "grad_norm": 0.35388344526290894, "learning_rate": 0.0002, "epoch": 1.3672020918939112, "step": 1830}, {"loss": 1.6968, "grad_norm": 0.3695325553417206, "learning_rate": 0.0002, "epoch": 1.3746731415763915, "step": 1840}, {"loss": 1.7086, "grad_norm": 0.3683604598045349, "learning_rate": 0.0002, "epoch": 1.382144191258872, "step": 1850}, {"loss": 1.7878, "grad_norm": 0.3753012418746948, "learning_rate": 0.0002, "epoch": 1.3896152409413522, "step": 1860}, {"loss": 1.6969, "grad_norm": 0.3331069350242615, "learning_rate": 0.0002, "epoch": 1.3970862906238326, "step": 1870}, {"loss": 1.6644, "grad_norm": 0.3877500295639038, "learning_rate": 0.0002, "epoch": 1.404557340306313, "step": 1880}, {"loss": 1.7586, "grad_norm": 0.33525151014328003, "learning_rate": 0.0002, "epoch": 1.4120283899887935, "step": 1890}, {"loss": 1.7031, "grad_norm": 0.3697299659252167, "learning_rate": 0.0002, "epoch": 1.4194994396712737, "step": 1900}, {"loss": 1.6956, "grad_norm": 0.4029286205768585, "learning_rate": 0.0002, "epoch": 1.4269704893537543, "step": 1910}, {"loss": 1.6897, "grad_norm": 0.3596203029155731, "learning_rate": 0.0002, "epoch": 1.4344415390362346, "step": 1920}, {"loss": 1.7139, "grad_norm": 0.450783908367157, "learning_rate": 0.0002, "epoch": 1.441912588718715, "step": 1930}, {"loss": 1.7243, "grad_norm": 0.3651481866836548, "learning_rate": 0.0002, "epoch": 1.4493836384011953, "step": 1940}, {"loss": 1.6637, "grad_norm": 0.3608424663543701, "learning_rate": 0.0002, "epoch": 1.4568546880836757, "step": 1950}, {"loss": 1.8285, "grad_norm": 0.39684420824050903, "learning_rate": 0.0002, "epoch": 1.4643257377661563, "step": 1960}, {"loss": 1.7514, "grad_norm": 0.34618663787841797, "learning_rate": 0.0002, "epoch": 1.4717967874486364, "step": 1970}, {"loss": 1.6655, "grad_norm": 0.4150386452674866, "learning_rate": 0.0002, "epoch": 1.479267837131117, "step": 1980}, {"loss": 1.7021, "grad_norm": 0.35500776767730713, "learning_rate": 0.0002, "epoch": 1.4867388868135973, "step": 1990}, {"loss": 1.7322, "grad_norm": 0.344144344329834, "learning_rate": 0.0002, "epoch": 1.4942099364960777, "step": 2000}, {"loss": 1.6998, "grad_norm": 0.3340149223804474, "learning_rate": 0.0002, "epoch": 1.501680986178558, "step": 2010}, {"loss": 1.7508, "grad_norm": 0.37685006856918335, "learning_rate": 0.0002, "epoch": 1.5091520358610384, "step": 2020}, {"loss": 1.8299, "grad_norm": 0.3699876368045807, "learning_rate": 0.0002, "epoch": 1.516623085543519, "step": 2030}, {"loss": 1.7357, "grad_norm": 0.3370307385921478, "learning_rate": 0.0002, "epoch": 1.5240941352259991, "step": 2040}, {"loss": 1.8044, "grad_norm": 0.37780630588531494, "learning_rate": 0.0002, "epoch": 1.5315651849084797, "step": 2050}, {"loss": 1.7408, "grad_norm": 0.370259165763855, "learning_rate": 0.0002, "epoch": 1.53903623459096, "step": 2060}, {"loss": 1.7398, "grad_norm": 0.3440011441707611, "learning_rate": 0.0002, "epoch": 1.5465072842734404, "step": 2070}, {"loss": 1.7105, "grad_norm": 0.40382063388824463, "learning_rate": 0.0002, "epoch": 1.5539783339559208, "step": 2080}, {"loss": 1.7071, "grad_norm": 0.38002029061317444, "learning_rate": 0.0002, "epoch": 1.5614493836384011, "step": 2090}, {"loss": 1.6815, "grad_norm": 0.3658451437950134, "learning_rate": 0.0002, "epoch": 1.5689204333208817, "step": 2100}, {"loss": 1.7598, "grad_norm": 0.354842871427536, "learning_rate": 0.0002, "epoch": 1.5763914830033618, "step": 2110}, {"loss": 1.6898, "grad_norm": 0.34735530614852905, "learning_rate": 0.0002, "epoch": 1.5838625326858424, "step": 2120}, {"loss": 1.7363, "grad_norm": 0.377581924200058, "learning_rate": 0.0002, "epoch": 1.5913335823683228, "step": 2130}, {"loss": 1.7789, "grad_norm": 0.41254034638404846, "learning_rate": 0.0002, "epoch": 1.5988046320508031, "step": 2140}, {"loss": 1.6782, "grad_norm": 0.3630715310573578, "learning_rate": 0.0002, "epoch": 1.6062756817332835, "step": 2150}, {"loss": 1.7531, "grad_norm": 0.36980143189430237, "learning_rate": 0.0002, "epoch": 1.6137467314157639, "step": 2160}, {"loss": 1.6847, "grad_norm": 0.3634769320487976, "learning_rate": 0.0002, "epoch": 1.6212177810982444, "step": 2170}, {"loss": 1.6367, "grad_norm": 0.3794139623641968, "learning_rate": 0.0002, "epoch": 1.6286888307807246, "step": 2180}, {"loss": 1.7064, "grad_norm": 0.359742134809494, "learning_rate": 0.0002, "epoch": 1.6361598804632052, "step": 2190}, {"loss": 1.7027, "grad_norm": 0.3770543932914734, "learning_rate": 0.0002, "epoch": 1.6436309301456855, "step": 2200}, {"loss": 1.784, "grad_norm": 0.3797036409378052, "learning_rate": 0.0002, "epoch": 1.6511019798281659, "step": 2210}, {"loss": 1.7875, "grad_norm": 0.35622093081474304, "learning_rate": 0.0002, "epoch": 1.6585730295106462, "step": 2220}, {"loss": 1.6615, "grad_norm": 0.34552520513534546, "learning_rate": 0.0002, "epoch": 1.6660440791931266, "step": 2230}, {"loss": 1.7522, "grad_norm": 0.379926860332489, "learning_rate": 0.0002, "epoch": 1.6735151288756072, "step": 2240}, {"loss": 1.7953, "grad_norm": 0.37083810567855835, "learning_rate": 0.0002, "epoch": 1.6809861785580873, "step": 2250}, {"loss": 1.7485, "grad_norm": 0.42746543884277344, "learning_rate": 0.0002, "epoch": 1.6884572282405679, "step": 2260}, {"loss": 1.776, "grad_norm": 0.3372884690761566, "learning_rate": 0.0002, "epoch": 1.6959282779230482, "step": 2270}, {"loss": 1.7604, "grad_norm": 0.35220256447792053, "learning_rate": 0.0002, "epoch": 1.7033993276055286, "step": 2280}, {"loss": 1.7154, "grad_norm": 0.3659130930900574, "learning_rate": 0.0002, "epoch": 1.710870377288009, "step": 2290}, {"loss": 1.6953, "grad_norm": 0.37629297375679016, "learning_rate": 0.0002, "epoch": 1.7183414269704893, "step": 2300}, {"loss": 1.7212, "grad_norm": 0.36312398314476013, "learning_rate": 0.0002, "epoch": 1.7258124766529699, "step": 2310}, {"loss": 1.7903, "grad_norm": 0.467709481716156, "learning_rate": 0.0002, "epoch": 1.73328352633545, "step": 2320}, {"loss": 1.696, "grad_norm": 0.38685527443885803, "learning_rate": 0.0002, "epoch": 1.7407545760179306, "step": 2330}, {"loss": 1.7041, "grad_norm": 0.3578338325023651, "learning_rate": 0.0002, "epoch": 1.748225625700411, "step": 2340}, {"loss": 1.6456, "grad_norm": 0.36057502031326294, "learning_rate": 0.0002, "epoch": 1.7556966753828913, "step": 2350}, {"loss": 1.6853, "grad_norm": 0.3615196645259857, "learning_rate": 0.0002, "epoch": 1.7631677250653717, "step": 2360}, {"loss": 1.7612, "grad_norm": 0.4118947684764862, "learning_rate": 0.0002, "epoch": 1.770638774747852, "step": 2370}, {"loss": 1.6946, "grad_norm": 0.4067276120185852, "learning_rate": 0.0002, "epoch": 1.7781098244303326, "step": 2380}, {"loss": 1.712, "grad_norm": 0.3979823887348175, "learning_rate": 0.0002, "epoch": 1.7855808741128127, "step": 2390}, {"loss": 1.7644, "grad_norm": 0.44045883417129517, "learning_rate": 0.0002, "epoch": 1.7930519237952933, "step": 2400}, {"loss": 1.7251, "grad_norm": 0.3998069167137146, "learning_rate": 0.0002, "epoch": 1.8005229734777737, "step": 2410}, {"loss": 1.7354, "grad_norm": 0.3450094759464264, "learning_rate": 0.0002, "epoch": 1.807994023160254, "step": 2420}, {"loss": 1.6998, "grad_norm": 0.3759009838104248, "learning_rate": 0.0002, "epoch": 1.8154650728427344, "step": 2430}, {"loss": 1.7706, "grad_norm": 0.34347015619277954, "learning_rate": 0.0002, "epoch": 1.8229361225252148, "step": 2440}, {"loss": 1.7345, "grad_norm": 0.3511228859424591, "learning_rate": 0.0002, "epoch": 1.8304071722076953, "step": 2450}, {"loss": 1.6909, "grad_norm": 0.36853715777397156, "learning_rate": 0.0002, "epoch": 1.8378782218901755, "step": 2460}, {"loss": 1.6931, "grad_norm": 0.40659376978874207, "learning_rate": 0.0002, "epoch": 1.845349271572656, "step": 2470}, {"loss": 1.7626, "grad_norm": 0.39621320366859436, "learning_rate": 0.0002, "epoch": 1.8528203212551362, "step": 2480}, {"loss": 1.7427, "grad_norm": 0.3753979504108429, "learning_rate": 0.0002, "epoch": 1.8602913709376168, "step": 2490}, {"loss": 1.6622, "grad_norm": 0.3811938464641571, "learning_rate": 0.0002, "epoch": 1.8677624206200971, "step": 2500}, {"loss": 1.7718, "grad_norm": 0.3432596027851105, "learning_rate": 0.0002, "epoch": 1.8752334703025775, "step": 2510}, {"loss": 1.7488, "grad_norm": 0.3670712113380432, "learning_rate": 0.0002, "epoch": 1.882704519985058, "step": 2520}, {"loss": 1.705, "grad_norm": 0.40907177329063416, "learning_rate": 0.0002, "epoch": 1.8901755696675382, "step": 2530}, {"loss": 1.7148, "grad_norm": 0.3821999728679657, "learning_rate": 0.0002, "epoch": 1.8976466193500188, "step": 2540}, {"loss": 1.7934, "grad_norm": 0.36173978447914124, "learning_rate": 0.0002, "epoch": 1.905117669032499, "step": 2550}, {"loss": 1.6939, "grad_norm": 0.38990336656570435, "learning_rate": 0.0002, "epoch": 1.9125887187149795, "step": 2560}, {"loss": 1.6893, "grad_norm": 0.35242322087287903, "learning_rate": 0.0002, "epoch": 1.9200597683974598, "step": 2570}, {"loss": 1.7268, "grad_norm": 0.3506428003311157, "learning_rate": 0.0002, "epoch": 1.9275308180799402, "step": 2580}, {"loss": 1.6953, "grad_norm": 0.39540135860443115, "learning_rate": 0.0002, "epoch": 1.9350018677624208, "step": 2590}, {"loss": 1.6511, "grad_norm": 0.3444725573062897, "learning_rate": 0.0002, "epoch": 1.942472917444901, "step": 2600}, {"loss": 1.7259, "grad_norm": 0.3963521718978882, "learning_rate": 0.0002, "epoch": 1.9499439671273815, "step": 2610}, {"loss": 1.6946, "grad_norm": 0.3689815402030945, "learning_rate": 0.0002, "epoch": 1.9574150168098616, "step": 2620}, {"loss": 1.7384, "grad_norm": 0.3482626676559448, "learning_rate": 0.0002, "epoch": 1.9648860664923422, "step": 2630}, {"loss": 1.7048, "grad_norm": 0.35832616686820984, "learning_rate": 0.0002, "epoch": 1.9723571161748226, "step": 2640}, {"loss": 1.6681, "grad_norm": 0.4776208996772766, "learning_rate": 0.0002, "epoch": 1.979828165857303, "step": 2650}, {"loss": 1.6696, "grad_norm": 0.32570165395736694, "learning_rate": 0.0002, "epoch": 1.9872992155397835, "step": 2660}, {"loss": 1.7232, "grad_norm": 0.3380725085735321, "learning_rate": 0.0002, "epoch": 1.9947702652222636, "step": 2670}, {"eval_loss": 1.8046749830245972, "eval_runtime": 38.5096, "eval_samples_per_second": 13.373, "eval_steps_per_second": 1.688, "epoch": 2.0, "step": 2677}, {"loss": 1.7265, "grad_norm": 0.36817631125450134, "learning_rate": 0.0002, "epoch": 2.002241314904744, "step": 2680}, {"loss": 1.548, "grad_norm": 0.4056456685066223, "learning_rate": 0.0002, "epoch": 2.0097123645872244, "step": 2690}, {"loss": 1.5515, "grad_norm": 0.37416863441467285, "learning_rate": 0.0002, "epoch": 2.017183414269705, "step": 2700}, {"loss": 1.5895, "grad_norm": 0.4273638427257538, "learning_rate": 0.0002, "epoch": 2.024654463952185, "step": 2710}, {"loss": 1.5884, "grad_norm": 0.36497923731803894, "learning_rate": 0.0002, "epoch": 2.0321255136346656, "step": 2720}, {"loss": 1.6999, "grad_norm": 0.5021994113922119, "learning_rate": 0.0002, "epoch": 2.0395965633171462, "step": 2730}, {"loss": 1.6655, "grad_norm": 0.45896220207214355, "learning_rate": 0.0002, "epoch": 2.0470676129996264, "step": 2740}, {"loss": 1.6305, "grad_norm": 0.3973815143108368, "learning_rate": 0.0002, "epoch": 2.054538662682107, "step": 2750}, {"loss": 1.6301, "grad_norm": 0.4521815776824951, "learning_rate": 0.0002, "epoch": 2.062009712364587, "step": 2760}, {"loss": 1.6189, "grad_norm": 0.42775002121925354, "learning_rate": 0.0002, "epoch": 2.0694807620470677, "step": 2770}, {"loss": 1.6491, "grad_norm": 0.48158586025238037, "learning_rate": 0.0002, "epoch": 2.076951811729548, "step": 2780}, {"loss": 1.6301, "grad_norm": 0.4612371623516083, "learning_rate": 0.0002, "epoch": 2.0844228614120284, "step": 2790}, {"loss": 1.6327, "grad_norm": 0.42536866664886475, "learning_rate": 0.0002, "epoch": 2.091893911094509, "step": 2800}, {"loss": 1.651, "grad_norm": 0.48515772819519043, "learning_rate": 0.0002, "epoch": 2.099364960776989, "step": 2810}, {"loss": 1.6829, "grad_norm": 0.41418662667274475, "learning_rate": 0.0002, "epoch": 2.1068360104594697, "step": 2820}, {"loss": 1.6266, "grad_norm": 0.4683697819709778, "learning_rate": 0.0002, "epoch": 2.11430706014195, "step": 2830}, {"loss": 1.6586, "grad_norm": 0.4484657049179077, "learning_rate": 0.0002, "epoch": 2.1217781098244304, "step": 2840}, {"loss": 1.6483, "grad_norm": 0.6621400713920593, "learning_rate": 0.0002, "epoch": 2.1292491595069105, "step": 2850}, {"loss": 1.5755, "grad_norm": 0.45074811577796936, "learning_rate": 0.0002, "epoch": 2.136720209189391, "step": 2860}, {"loss": 1.6456, "grad_norm": 0.3513113558292389, "learning_rate": 0.0002, "epoch": 2.1441912588718717, "step": 2870}, {"loss": 1.6081, "grad_norm": 0.40411314368247986, "learning_rate": 0.0002, "epoch": 2.151662308554352, "step": 2880}, {"loss": 1.6323, "grad_norm": 0.4121065139770508, "learning_rate": 0.0002, "epoch": 2.1591333582368324, "step": 2890}, {"loss": 1.6324, "grad_norm": 0.44723689556121826, "learning_rate": 0.0002, "epoch": 2.1666044079193125, "step": 2900}, {"loss": 1.5699, "grad_norm": 0.4226122498512268, "learning_rate": 0.0002, "epoch": 2.174075457601793, "step": 2910}, {"loss": 1.5652, "grad_norm": 0.46617650985717773, "learning_rate": 0.0002, "epoch": 2.1815465072842732, "step": 2920}, {"loss": 1.6378, "grad_norm": 0.4506422281265259, "learning_rate": 0.0002, "epoch": 2.189017556966754, "step": 2930}, {"loss": 1.6112, "grad_norm": 0.4892672896385193, "learning_rate": 0.0002, "epoch": 2.1964886066492344, "step": 2940}, {"loss": 1.6176, "grad_norm": 0.44095516204833984, "learning_rate": 0.0002, "epoch": 2.2039596563317145, "step": 2950}, {"loss": 1.6058, "grad_norm": 0.41522109508514404, "learning_rate": 0.0002, "epoch": 2.211430706014195, "step": 2960}, {"loss": 1.5964, "grad_norm": 0.4860858917236328, "learning_rate": 0.0002, "epoch": 2.2189017556966752, "step": 2970}, {"loss": 1.6427, "grad_norm": 0.42662516236305237, "learning_rate": 0.0002, "epoch": 2.226372805379156, "step": 2980}, {"loss": 1.6313, "grad_norm": 0.4390648305416107, "learning_rate": 0.0002, "epoch": 2.233843855061636, "step": 2990}, {"loss": 1.5992, "grad_norm": 0.47515565156936646, "learning_rate": 0.0002, "epoch": 2.2413149047441165, "step": 3000}, {"loss": 1.5563, "grad_norm": 0.4104543924331665, "learning_rate": 0.0002, "epoch": 2.248785954426597, "step": 3010}, {"loss": 1.6895, "grad_norm": 0.4404028654098511, "learning_rate": 0.0002, "epoch": 2.2562570041090773, "step": 3020}, {"loss": 1.6088, "grad_norm": 0.4717366695404053, "learning_rate": 0.0002, "epoch": 2.263728053791558, "step": 3030}, {"loss": 1.7287, "grad_norm": 0.48345857858657837, "learning_rate": 0.0002, "epoch": 2.271199103474038, "step": 3040}, {"loss": 1.681, "grad_norm": 0.5312452912330627, "learning_rate": 0.0002, "epoch": 2.2786701531565186, "step": 3050}, {"loss": 1.5901, "grad_norm": 0.5073099732398987, "learning_rate": 0.0002, "epoch": 2.2861412028389987, "step": 3060}, {"loss": 1.6914, "grad_norm": 0.5027463436126709, "learning_rate": 0.0002, "epoch": 2.2936122525214793, "step": 3070}, {"loss": 1.5862, "grad_norm": 0.5436304807662964, "learning_rate": 0.0002, "epoch": 2.30108330220396, "step": 3080}, {"loss": 1.5763, "grad_norm": 0.4701065123081207, "learning_rate": 0.0002, "epoch": 2.30855435188644, "step": 3090}, {"loss": 1.6177, "grad_norm": 0.46988746523857117, "learning_rate": 0.0002, "epoch": 2.3160254015689206, "step": 3100}, {"loss": 1.6502, "grad_norm": 0.45112869143486023, "learning_rate": 0.0002, "epoch": 2.3234964512514007, "step": 3110}, {"loss": 1.6291, "grad_norm": 0.5173566937446594, "learning_rate": 0.0002, "epoch": 2.3309675009338813, "step": 3120}, {"loss": 1.6743, "grad_norm": 0.40345850586891174, "learning_rate": 0.0002, "epoch": 2.3384385506163614, "step": 3130}, {"loss": 1.621, "grad_norm": 0.4218924939632416, "learning_rate": 0.0002, "epoch": 2.345909600298842, "step": 3140}, {"loss": 1.6341, "grad_norm": 0.41857317090034485, "learning_rate": 0.0002, "epoch": 2.3533806499813226, "step": 3150}, {"loss": 1.6087, "grad_norm": 0.4197218418121338, "learning_rate": 0.0002, "epoch": 2.3608516996638027, "step": 3160}, {"loss": 1.6572, "grad_norm": 0.4260677397251129, "learning_rate": 0.0002, "epoch": 2.3683227493462833, "step": 3170}, {"loss": 1.6376, "grad_norm": 0.4209042191505432, "learning_rate": 0.0002, "epoch": 2.3757937990287634, "step": 3180}, {"loss": 1.634, "grad_norm": 0.4092234969139099, "learning_rate": 0.0002, "epoch": 2.383264848711244, "step": 3190}, {"loss": 1.6339, "grad_norm": 0.4928431510925293, "learning_rate": 0.0002, "epoch": 2.390735898393724, "step": 3200}, {"loss": 1.6015, "grad_norm": 0.49252402782440186, "learning_rate": 0.0002, "epoch": 2.3982069480762047, "step": 3210}, {"loss": 1.5773, "grad_norm": 0.4368397295475006, "learning_rate": 0.0002, "epoch": 2.4056779977586853, "step": 3220}, {"loss": 1.6629, "grad_norm": 0.46122390031814575, "learning_rate": 0.0002, "epoch": 2.4131490474411654, "step": 3230}, {"loss": 1.6224, "grad_norm": 0.4272301197052002, "learning_rate": 0.0002, "epoch": 2.420620097123646, "step": 3240}, {"loss": 1.5961, "grad_norm": 0.41480937600135803, "learning_rate": 0.0002, "epoch": 2.428091146806126, "step": 3250}, {"loss": 1.6281, "grad_norm": 0.48911941051483154, "learning_rate": 0.0002, "epoch": 2.4355621964886067, "step": 3260}, {"loss": 1.6846, "grad_norm": 0.4444098472595215, "learning_rate": 0.0002, "epoch": 2.443033246171087, "step": 3270}, {"loss": 1.6961, "grad_norm": 0.5111684799194336, "learning_rate": 0.0002, "epoch": 2.4505042958535674, "step": 3280}, {"loss": 1.6152, "grad_norm": 0.5058825016021729, "learning_rate": 0.0002, "epoch": 2.457975345536048, "step": 3290}, {"loss": 1.625, "grad_norm": 0.44173210859298706, "learning_rate": 0.0002, "epoch": 2.465446395218528, "step": 3300}, {"loss": 1.6491, "grad_norm": 0.4659745991230011, "learning_rate": 0.0002, "epoch": 2.4729174449010087, "step": 3310}, {"loss": 1.6114, "grad_norm": 0.47237497568130493, "learning_rate": 0.0002, "epoch": 2.480388494583489, "step": 3320}, {"loss": 1.6193, "grad_norm": 0.47303131222724915, "learning_rate": 0.0002, "epoch": 2.4878595442659694, "step": 3330}, {"loss": 1.7256, "grad_norm": 0.4522389769554138, "learning_rate": 0.0002, "epoch": 2.4953305939484496, "step": 3340}, {"loss": 1.6834, "grad_norm": 0.4467332363128662, "learning_rate": 0.0002, "epoch": 2.50280164363093, "step": 3350}, {"loss": 1.6108, "grad_norm": 0.4413762092590332, "learning_rate": 0.0002, "epoch": 2.5102726933134107, "step": 3360}, {"loss": 1.537, "grad_norm": 0.495514452457428, "learning_rate": 0.0002, "epoch": 2.517743742995891, "step": 3370}, {"loss": 1.5839, "grad_norm": 0.4429773986339569, "learning_rate": 0.0002, "epoch": 2.5252147926783715, "step": 3380}, {"loss": 1.6522, "grad_norm": 0.4589079022407532, "learning_rate": 0.0002, "epoch": 2.5326858423608516, "step": 3390}, {"loss": 1.6529, "grad_norm": 0.4683997333049774, "learning_rate": 0.0002, "epoch": 2.540156892043332, "step": 3400}, {"loss": 1.6745, "grad_norm": 0.4651731252670288, "learning_rate": 0.0002, "epoch": 2.5476279417258123, "step": 3410}, {"loss": 1.5918, "grad_norm": 0.45818084478378296, "learning_rate": 0.0002, "epoch": 2.555098991408293, "step": 3420}, {"loss": 1.6326, "grad_norm": 0.45209529995918274, "learning_rate": 0.0002, "epoch": 2.5625700410907735, "step": 3430}, {"loss": 1.5606, "grad_norm": 0.4344733655452728, "learning_rate": 0.0002, "epoch": 2.5700410907732536, "step": 3440}, {"loss": 1.6748, "grad_norm": 0.47435566782951355, "learning_rate": 0.0002, "epoch": 2.577512140455734, "step": 3450}, {"loss": 1.6237, "grad_norm": 0.43841999769210815, "learning_rate": 0.0002, "epoch": 2.5849831901382143, "step": 3460}, {"loss": 1.7207, "grad_norm": 0.4323869049549103, "learning_rate": 0.0002, "epoch": 2.592454239820695, "step": 3470}, {"loss": 1.5494, "grad_norm": 0.44355881214141846, "learning_rate": 0.0002, "epoch": 2.599925289503175, "step": 3480}, {"loss": 1.665, "grad_norm": 0.45847779512405396, "learning_rate": 0.0002, "epoch": 2.6073963391856556, "step": 3490}, {"loss": 1.6006, "grad_norm": 0.4411061704158783, "learning_rate": 0.0002, "epoch": 2.614867388868136, "step": 3500}, {"loss": 1.5868, "grad_norm": 0.4446796178817749, "learning_rate": 0.0002, "epoch": 2.6223384385506163, "step": 3510}, {"loss": 1.5946, "grad_norm": 0.41969653964042664, "learning_rate": 0.0002, "epoch": 2.629809488233097, "step": 3520}, {"loss": 1.6798, "grad_norm": 0.5263747572898865, "learning_rate": 0.0002, "epoch": 2.637280537915577, "step": 3530}, {"loss": 1.6309, "grad_norm": 0.47719451785087585, "learning_rate": 0.0002, "epoch": 2.6447515875980576, "step": 3540}, {"loss": 1.7024, "grad_norm": 0.46574118733406067, "learning_rate": 0.0002, "epoch": 2.6522226372805378, "step": 3550}, {"loss": 1.618, "grad_norm": 0.46867135167121887, "learning_rate": 0.0002, "epoch": 2.6596936869630183, "step": 3560}, {"loss": 1.5885, "grad_norm": 0.4441198706626892, "learning_rate": 0.0002, "epoch": 2.667164736645499, "step": 3570}, {"loss": 1.6426, "grad_norm": 0.4871319830417633, "learning_rate": 0.0002, "epoch": 2.674635786327979, "step": 3580}, {"loss": 1.6575, "grad_norm": 0.43900373578071594, "learning_rate": 0.0002, "epoch": 2.6821068360104596, "step": 3590}, {"loss": 1.6071, "grad_norm": 0.42509549856185913, "learning_rate": 0.0002, "epoch": 2.6895778856929398, "step": 3600}, {"loss": 1.5651, "grad_norm": 0.4691086709499359, "learning_rate": 0.0002, "epoch": 2.6970489353754203, "step": 3610}, {"loss": 1.5491, "grad_norm": 0.46318942308425903, "learning_rate": 0.0002, "epoch": 2.7045199850579005, "step": 3620}, {"loss": 1.5422, "grad_norm": 0.44631096720695496, "learning_rate": 0.0002, "epoch": 2.711991034740381, "step": 3630}, {"loss": 1.6831, "grad_norm": 0.42315489053726196, "learning_rate": 0.0002, "epoch": 2.7194620844228616, "step": 3640}, {"loss": 1.6008, "grad_norm": 0.4971241056919098, "learning_rate": 0.0002, "epoch": 2.7269331341053418, "step": 3650}, {"loss": 1.6042, "grad_norm": 0.4578486382961273, "learning_rate": 0.0002, "epoch": 2.7344041837878224, "step": 3660}, {"loss": 1.6076, "grad_norm": 0.46584776043891907, "learning_rate": 0.0002, "epoch": 2.7418752334703025, "step": 3670}, {"loss": 1.6809, "grad_norm": 0.4951731264591217, "learning_rate": 0.0002, "epoch": 2.749346283152783, "step": 3680}, {"loss": 1.6226, "grad_norm": 0.4935225546360016, "learning_rate": 0.0002, "epoch": 2.756817332835263, "step": 3690}, {"loss": 1.5878, "grad_norm": 0.41805586218833923, "learning_rate": 0.0002, "epoch": 2.764288382517744, "step": 3700}, {"loss": 1.7173, "grad_norm": 0.4417555630207062, "learning_rate": 0.0002, "epoch": 2.7717594322002244, "step": 3710}, {"loss": 1.6398, "grad_norm": 0.48229655623435974, "learning_rate": 0.0002, "epoch": 2.7792304818827045, "step": 3720}, {"loss": 1.6074, "grad_norm": 0.48562315106391907, "learning_rate": 0.0002, "epoch": 2.786701531565185, "step": 3730}, {"loss": 1.607, "grad_norm": 0.4473940432071686, "learning_rate": 0.0002, "epoch": 2.794172581247665, "step": 3740}, {"loss": 1.6065, "grad_norm": 0.4626813232898712, "learning_rate": 0.0002, "epoch": 2.801643630930146, "step": 3750}, {"loss": 1.6296, "grad_norm": 0.4339792728424072, "learning_rate": 0.0002, "epoch": 2.809114680612626, "step": 3760}, {"loss": 1.6815, "grad_norm": 0.5250858068466187, "learning_rate": 0.0002, "epoch": 2.8165857302951065, "step": 3770}, {"loss": 1.6644, "grad_norm": 0.4537523090839386, "learning_rate": 0.0002, "epoch": 2.824056779977587, "step": 3780}, {"loss": 1.6535, "grad_norm": 0.5646113157272339, "learning_rate": 0.0002, "epoch": 2.831527829660067, "step": 3790}, {"loss": 1.5712, "grad_norm": 0.44243332743644714, "learning_rate": 0.0002, "epoch": 2.8389988793425474, "step": 3800}, {"loss": 1.6478, "grad_norm": 0.4585791826248169, "learning_rate": 0.0002, "epoch": 2.846469929025028, "step": 3810}, {"loss": 1.6854, "grad_norm": 0.489702045917511, "learning_rate": 0.0002, "epoch": 2.8539409787075085, "step": 3820}, {"loss": 1.7066, "grad_norm": 0.502470850944519, "learning_rate": 0.0002, "epoch": 2.8614120283899886, "step": 3830}, {"loss": 1.5785, "grad_norm": 0.4395960867404938, "learning_rate": 0.0002, "epoch": 2.8688830780724692, "step": 3840}, {"loss": 1.6434, "grad_norm": 0.4348670244216919, "learning_rate": 0.0002, "epoch": 2.87635412775495, "step": 3850}, {"loss": 1.6163, "grad_norm": 0.48852720856666565, "learning_rate": 0.0002, "epoch": 2.88382517743743, "step": 3860}, {"loss": 1.5916, "grad_norm": 0.45317450165748596, "learning_rate": 0.0002, "epoch": 2.89129622711991, "step": 3870}, {"loss": 1.6486, "grad_norm": 0.4732758700847626, "learning_rate": 0.0002, "epoch": 2.8987672768023907, "step": 3880}, {"loss": 1.6758, "grad_norm": 0.45238012075424194, "learning_rate": 0.0002, "epoch": 2.9062383264848712, "step": 3890}, {"loss": 1.6228, "grad_norm": 0.48838064074516296, "learning_rate": 0.0002, "epoch": 2.9137093761673514, "step": 3900}, {"loss": 1.658, "grad_norm": 0.43496349453926086, "learning_rate": 0.0002, "epoch": 2.921180425849832, "step": 3910}, {"loss": 1.7063, "grad_norm": 0.47963935136795044, "learning_rate": 0.0002, "epoch": 2.9286514755323125, "step": 3920}, {"loss": 1.6553, "grad_norm": 0.4544987976551056, "learning_rate": 0.0002, "epoch": 2.9361225252147927, "step": 3930}, {"loss": 1.6192, "grad_norm": 0.4622892141342163, "learning_rate": 0.0002, "epoch": 2.943593574897273, "step": 3940}, {"loss": 1.6178, "grad_norm": 0.47026222944259644, "learning_rate": 0.0002, "epoch": 2.9510646245797534, "step": 3950}, {"loss": 1.6612, "grad_norm": 0.4549552798271179, "learning_rate": 0.0002, "epoch": 2.958535674262234, "step": 3960}, {"loss": 1.6458, "grad_norm": 0.46647515892982483, "learning_rate": 0.0002, "epoch": 2.966006723944714, "step": 3970}, {"loss": 1.6051, "grad_norm": 0.45095112919807434, "learning_rate": 0.0002, "epoch": 2.9734777736271947, "step": 3980}, {"loss": 1.6471, "grad_norm": 0.4690017104148865, "learning_rate": 0.0002, "epoch": 2.9809488233096753, "step": 3990}, {"loss": 1.6061, "grad_norm": 0.4603444039821625, "learning_rate": 0.0002, "epoch": 2.9884198729921554, "step": 4000}, {"loss": 1.6431, "grad_norm": 0.4743294417858124, "learning_rate": 0.0002, "epoch": 2.9958909226746355, "step": 4010}]}
+{"epoch": 4.0, "step": 5354, "epoch_duration": 1510.5811264514923, "total_accumulated_duration": 5823.530160188675, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7568.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-42/checkpoint-2677", "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": 2.6181, "grad_norm": 0.4912872612476349, "learning_rate": 0.0002, "epoch": 0.007471049682480389, "step": 10}, {"loss": 2.2606, "grad_norm": 0.4856316149234772, "learning_rate": 0.0002, "epoch": 0.014942099364960777, "step": 20}, {"loss": 2.0957, "grad_norm": 0.47683125734329224, "learning_rate": 0.0002, "epoch": 0.022413149047441166, "step": 30}, {"loss": 1.8908, "grad_norm": 0.515082597732544, "learning_rate": 0.0002, "epoch": 0.029884198729921554, "step": 40}, {"loss": 1.9704, "grad_norm": 0.5299215316772461, "learning_rate": 0.0002, "epoch": 0.03735524841240194, "step": 50}, {"loss": 1.9225, "grad_norm": 0.4951399862766266, "learning_rate": 0.0002, "epoch": 0.04482629809488233, "step": 60}, {"loss": 1.9742, "grad_norm": 0.48079821467399597, "learning_rate": 0.0002, "epoch": 0.05229734777736272, "step": 70}, {"loss": 1.9466, "grad_norm": 0.49402132630348206, "learning_rate": 0.0002, "epoch": 0.05976839745984311, "step": 80}, {"loss": 1.8691, "grad_norm": 0.4778193235397339, "learning_rate": 0.0002, "epoch": 0.0672394471423235, "step": 90}, {"loss": 1.8455, "grad_norm": 0.42472657561302185, "learning_rate": 0.0002, "epoch": 0.07471049682480388, "step": 100}, {"loss": 1.8744, "grad_norm": 0.4433092474937439, "learning_rate": 0.0002, "epoch": 0.08218154650728428, "step": 110}, {"loss": 1.865, "grad_norm": 0.4472862780094147, "learning_rate": 0.0002, "epoch": 0.08965259618976466, "step": 120}, {"loss": 1.9256, "grad_norm": 0.42596298456192017, "learning_rate": 0.0002, "epoch": 0.09712364587224505, "step": 130}, {"loss": 1.8015, "grad_norm": 0.46645811200141907, "learning_rate": 0.0002, "epoch": 0.10459469555472543, "step": 140}, {"loss": 1.8307, "grad_norm": 0.41041234135627747, "learning_rate": 0.0002, "epoch": 0.11206574523720583, "step": 150}, {"loss": 1.8276, "grad_norm": 0.5329819917678833, "learning_rate": 0.0002, "epoch": 0.11953679491968622, "step": 160}, {"loss": 1.8118, "grad_norm": 0.4065922200679779, "learning_rate": 0.0002, "epoch": 0.1270078446021666, "step": 170}, {"loss": 1.8559, "grad_norm": 0.38406994938850403, "learning_rate": 0.0002, "epoch": 0.134478894284647, "step": 180}, {"loss": 1.8647, "grad_norm": 0.4246881306171417, "learning_rate": 0.0002, "epoch": 0.14194994396712737, "step": 190}, {"loss": 1.8054, "grad_norm": 0.35136649012565613, "learning_rate": 0.0002, "epoch": 0.14942099364960776, "step": 200}, {"loss": 1.802, "grad_norm": 0.43252742290496826, "learning_rate": 0.0002, "epoch": 0.15689204333208817, "step": 210}, {"loss": 1.7823, "grad_norm": 0.39236941933631897, "learning_rate": 0.0002, "epoch": 0.16436309301456856, "step": 220}, {"loss": 1.818, "grad_norm": 0.3748249113559723, "learning_rate": 0.0002, "epoch": 0.17183414269704894, "step": 230}, {"loss": 1.866, "grad_norm": 0.6432855725288391, "learning_rate": 0.0002, "epoch": 0.17930519237952933, "step": 240}, {"loss": 1.8397, "grad_norm": 0.34874802827835083, "learning_rate": 0.0002, "epoch": 0.1867762420620097, "step": 250}, {"loss": 1.79, "grad_norm": 0.3721984326839447, "learning_rate": 0.0002, "epoch": 0.1942472917444901, "step": 260}, {"loss": 1.8464, "grad_norm": 0.4339311420917511, "learning_rate": 0.0002, "epoch": 0.20171834142697048, "step": 270}, {"loss": 1.8665, "grad_norm": 0.4018215537071228, "learning_rate": 0.0002, "epoch": 0.20918939110945087, "step": 280}, {"loss": 1.8048, "grad_norm": 0.3278839886188507, "learning_rate": 0.0002, "epoch": 0.21666044079193125, "step": 290}, {"loss": 1.7395, "grad_norm": 0.36146077513694763, "learning_rate": 0.0002, "epoch": 0.22413149047441167, "step": 300}, {"loss": 1.7916, "grad_norm": 0.38175010681152344, "learning_rate": 0.0002, "epoch": 0.23160254015689205, "step": 310}, {"loss": 1.8593, "grad_norm": 0.44776618480682373, "learning_rate": 0.0002, "epoch": 0.23907358983937244, "step": 320}, {"loss": 1.7824, "grad_norm": 0.3933652937412262, "learning_rate": 0.0002, "epoch": 0.24654463952185282, "step": 330}, {"loss": 1.8393, "grad_norm": 0.3515005111694336, "learning_rate": 0.0002, "epoch": 0.2540156892043332, "step": 340}, {"loss": 1.8653, "grad_norm": 0.6683304309844971, "learning_rate": 0.0002, "epoch": 0.2614867388868136, "step": 350}, {"loss": 1.8797, "grad_norm": 0.37093454599380493, "learning_rate": 0.0002, "epoch": 0.268957788569294, "step": 360}, {"loss": 1.8251, "grad_norm": 0.3450651168823242, "learning_rate": 0.0002, "epoch": 0.2764288382517744, "step": 370}, {"loss": 1.7435, "grad_norm": 0.5140917301177979, "learning_rate": 0.0002, "epoch": 0.28389988793425475, "step": 380}, {"loss": 1.8026, "grad_norm": 0.32885563373565674, "learning_rate": 0.0002, "epoch": 0.29137093761673516, "step": 390}, {"loss": 1.8174, "grad_norm": 0.33962297439575195, "learning_rate": 0.0002, "epoch": 0.2988419872992155, "step": 400}, {"loss": 1.7467, "grad_norm": 0.3723141849040985, "learning_rate": 0.0002, "epoch": 0.30631303698169593, "step": 410}, {"loss": 1.8459, "grad_norm": 0.37173134088516235, "learning_rate": 0.0002, "epoch": 0.31378408666417634, "step": 420}, {"loss": 1.8876, "grad_norm": 0.33736956119537354, "learning_rate": 0.0002, "epoch": 0.3212551363466567, "step": 430}, {"loss": 1.8367, "grad_norm": 0.3602448105812073, "learning_rate": 0.0002, "epoch": 0.3287261860291371, "step": 440}, {"loss": 1.8058, "grad_norm": 0.3569699227809906, "learning_rate": 0.0002, "epoch": 0.33619723571161747, "step": 450}, {"loss": 1.8086, "grad_norm": 0.31009167432785034, "learning_rate": 0.0002, "epoch": 0.3436682853940979, "step": 460}, {"loss": 1.8876, "grad_norm": 0.5278693437576294, "learning_rate": 0.0002, "epoch": 0.35113933507657824, "step": 470}, {"loss": 1.8534, "grad_norm": 0.3587537109851837, "learning_rate": 0.0002, "epoch": 0.35861038475905865, "step": 480}, {"loss": 1.8046, "grad_norm": 0.3859670162200928, "learning_rate": 0.0002, "epoch": 0.366081434441539, "step": 490}, {"loss": 1.8287, "grad_norm": 0.395913690328598, "learning_rate": 0.0002, "epoch": 0.3735524841240194, "step": 500}, {"loss": 1.7619, "grad_norm": 0.35052940249443054, "learning_rate": 0.0002, "epoch": 0.38102353380649984, "step": 510}, {"loss": 1.7824, "grad_norm": 0.2979494333267212, "learning_rate": 0.0002, "epoch": 0.3884945834889802, "step": 520}, {"loss": 1.8641, "grad_norm": 0.3062683343887329, "learning_rate": 0.0002, "epoch": 0.3959656331714606, "step": 530}, {"loss": 1.7651, "grad_norm": 0.3172847330570221, "learning_rate": 0.0002, "epoch": 0.40343668285394096, "step": 540}, {"loss": 1.806, "grad_norm": 0.360435426235199, "learning_rate": 0.0002, "epoch": 0.4109077325364214, "step": 550}, {"loss": 1.9054, "grad_norm": 0.3427872359752655, "learning_rate": 0.0002, "epoch": 0.41837878221890173, "step": 560}, {"loss": 1.7562, "grad_norm": 0.34036558866500854, "learning_rate": 0.0002, "epoch": 0.42584983190138215, "step": 570}, {"loss": 1.7254, "grad_norm": 0.3365345299243927, "learning_rate": 0.0002, "epoch": 0.4333208815838625, "step": 580}, {"loss": 1.8328, "grad_norm": 0.35619041323661804, "learning_rate": 0.0002, "epoch": 0.4407919312663429, "step": 590}, {"loss": 1.8114, "grad_norm": 0.3569088280200958, "learning_rate": 0.0002, "epoch": 0.44826298094882333, "step": 600}, {"loss": 1.8599, "grad_norm": 0.3581278622150421, "learning_rate": 0.0002, "epoch": 0.4557340306313037, "step": 610}, {"loss": 1.7078, "grad_norm": 0.43197110295295715, "learning_rate": 0.0002, "epoch": 0.4632050803137841, "step": 620}, {"loss": 1.8257, "grad_norm": 0.33966198563575745, "learning_rate": 0.0002, "epoch": 0.47067612999626446, "step": 630}, {"loss": 1.7528, "grad_norm": 0.3343866467475891, "learning_rate": 0.0002, "epoch": 0.47814717967874487, "step": 640}, {"loss": 1.8191, "grad_norm": 0.33878564834594727, "learning_rate": 0.0002, "epoch": 0.48561822936122523, "step": 650}, {"loss": 1.8801, "grad_norm": 0.387195885181427, "learning_rate": 0.0002, "epoch": 0.49308927904370564, "step": 660}, {"loss": 1.7559, "grad_norm": 0.3755440413951874, "learning_rate": 0.0002, "epoch": 0.500560328726186, "step": 670}, {"loss": 1.8057, "grad_norm": 0.3272816836833954, "learning_rate": 0.0002, "epoch": 0.5080313784086664, "step": 680}, {"loss": 1.8156, "grad_norm": 0.36063864827156067, "learning_rate": 0.0002, "epoch": 0.5155024280911468, "step": 690}, {"loss": 1.8397, "grad_norm": 0.35317373275756836, "learning_rate": 0.0002, "epoch": 0.5229734777736272, "step": 700}, {"loss": 1.7603, "grad_norm": 0.3561195433139801, "learning_rate": 0.0002, "epoch": 0.5304445274561076, "step": 710}, {"loss": 1.8149, "grad_norm": 0.31124624609947205, "learning_rate": 0.0002, "epoch": 0.537915577138588, "step": 720}, {"loss": 1.7434, "grad_norm": 0.3294544517993927, "learning_rate": 0.0002, "epoch": 0.5453866268210683, "step": 730}, {"loss": 1.8027, "grad_norm": 0.31933900713920593, "learning_rate": 0.0002, "epoch": 0.5528576765035488, "step": 740}, {"loss": 1.7601, "grad_norm": 0.3226020634174347, "learning_rate": 0.0002, "epoch": 0.5603287261860291, "step": 750}, {"loss": 1.7862, "grad_norm": 0.3147525489330292, "learning_rate": 0.0002, "epoch": 0.5677997758685095, "step": 760}, {"loss": 1.9028, "grad_norm": 0.32234328985214233, "learning_rate": 0.0002, "epoch": 0.57527082555099, "step": 770}, {"loss": 1.7623, "grad_norm": 0.3258664309978485, "learning_rate": 0.0002, "epoch": 0.5827418752334703, "step": 780}, {"loss": 1.7384, "grad_norm": 0.3166961967945099, "learning_rate": 0.0002, "epoch": 0.5902129249159507, "step": 790}, {"loss": 1.8799, "grad_norm": 0.35621458292007446, "learning_rate": 0.0002, "epoch": 0.597683974598431, "step": 800}, {"loss": 1.8313, "grad_norm": 0.3236999213695526, "learning_rate": 0.0002, "epoch": 0.6051550242809115, "step": 810}, {"loss": 1.7132, "grad_norm": 0.2892923653125763, "learning_rate": 0.0002, "epoch": 0.6126260739633919, "step": 820}, {"loss": 1.8709, "grad_norm": 0.4098321497440338, "learning_rate": 0.0002, "epoch": 0.6200971236458722, "step": 830}, {"loss": 1.7637, "grad_norm": 0.3337118923664093, "learning_rate": 0.0002, "epoch": 0.6275681733283527, "step": 840}, {"loss": 1.7375, "grad_norm": 0.30416029691696167, "learning_rate": 0.0002, "epoch": 0.635039223010833, "step": 850}, {"loss": 1.7419, "grad_norm": 0.3361026346683502, "learning_rate": 0.0002, "epoch": 0.6425102726933134, "step": 860}, {"loss": 1.732, "grad_norm": 0.3537365198135376, "learning_rate": 0.0002, "epoch": 0.6499813223757938, "step": 870}, {"loss": 1.7825, "grad_norm": 0.33854469656944275, "learning_rate": 0.0002, "epoch": 0.6574523720582742, "step": 880}, {"loss": 1.7561, "grad_norm": 0.3332272469997406, "learning_rate": 0.0002, "epoch": 0.6649234217407546, "step": 890}, {"loss": 1.7247, "grad_norm": 0.34954726696014404, "learning_rate": 0.0002, "epoch": 0.6723944714232349, "step": 900}, {"loss": 1.7917, "grad_norm": 0.2921750247478485, "learning_rate": 0.0002, "epoch": 0.6798655211057153, "step": 910}, {"loss": 1.7807, "grad_norm": 0.30508682131767273, "learning_rate": 0.0002, "epoch": 0.6873365707881958, "step": 920}, {"loss": 1.8082, "grad_norm": 0.32268425822257996, "learning_rate": 0.0002, "epoch": 0.6948076204706761, "step": 930}, {"loss": 1.8283, "grad_norm": 0.2844390869140625, "learning_rate": 0.0002, "epoch": 0.7022786701531565, "step": 940}, {"loss": 1.7363, "grad_norm": 0.31263890862464905, "learning_rate": 0.0002, "epoch": 0.709749719835637, "step": 950}, {"loss": 1.8081, "grad_norm": 0.3626808822154999, "learning_rate": 0.0002, "epoch": 0.7172207695181173, "step": 960}, {"loss": 1.853, "grad_norm": 0.3322749733924866, "learning_rate": 0.0002, "epoch": 0.7246918192005977, "step": 970}, {"loss": 1.7912, "grad_norm": 0.29177871346473694, "learning_rate": 0.0002, "epoch": 0.732162868883078, "step": 980}, {"loss": 1.8447, "grad_norm": 0.35405513644218445, "learning_rate": 0.0002, "epoch": 0.7396339185655585, "step": 990}, {"loss": 1.7008, "grad_norm": 0.39318400621414185, "learning_rate": 0.0002, "epoch": 0.7471049682480388, "step": 1000}, {"loss": 1.7803, "grad_norm": 0.29401418566703796, "learning_rate": 0.0002, "epoch": 0.7545760179305192, "step": 1010}, {"loss": 1.7649, "grad_norm": 0.3271748721599579, "learning_rate": 0.0002, "epoch": 0.7620470676129997, "step": 1020}, {"loss": 1.7266, "grad_norm": 0.30883970856666565, "learning_rate": 0.0002, "epoch": 0.76951811729548, "step": 1030}, {"loss": 1.7722, "grad_norm": 0.3411838412284851, "learning_rate": 0.0002, "epoch": 0.7769891669779604, "step": 1040}, {"loss": 1.829, "grad_norm": 0.30608129501342773, "learning_rate": 0.0002, "epoch": 0.7844602166604407, "step": 1050}, {"loss": 1.7815, "grad_norm": 0.30899080634117126, "learning_rate": 0.0002, "epoch": 0.7919312663429212, "step": 1060}, {"loss": 1.7625, "grad_norm": 0.3160453140735626, "learning_rate": 0.0002, "epoch": 0.7994023160254016, "step": 1070}, {"loss": 1.8452, "grad_norm": 0.30947187542915344, "learning_rate": 0.0002, "epoch": 0.8068733657078819, "step": 1080}, {"loss": 1.7418, "grad_norm": 0.3103134036064148, "learning_rate": 0.0002, "epoch": 0.8143444153903624, "step": 1090}, {"loss": 1.842, "grad_norm": 0.31771138310432434, "learning_rate": 0.0002, "epoch": 0.8218154650728428, "step": 1100}, {"loss": 1.7918, "grad_norm": 0.5860997438430786, "learning_rate": 0.0002, "epoch": 0.8292865147553231, "step": 1110}, {"loss": 1.8443, "grad_norm": 0.3230148255825043, "learning_rate": 0.0002, "epoch": 0.8367575644378035, "step": 1120}, {"loss": 1.8478, "grad_norm": 0.29611510038375854, "learning_rate": 0.0002, "epoch": 0.8442286141202839, "step": 1130}, {"loss": 1.7673, "grad_norm": 0.3373654782772064, "learning_rate": 0.0002, "epoch": 0.8516996638027643, "step": 1140}, {"loss": 1.7997, "grad_norm": 0.3474279046058655, "learning_rate": 0.0002, "epoch": 0.8591707134852447, "step": 1150}, {"loss": 1.75, "grad_norm": 0.35057875514030457, "learning_rate": 0.0002, "epoch": 0.866641763167725, "step": 1160}, {"loss": 1.8273, "grad_norm": 0.39537495374679565, "learning_rate": 0.0002, "epoch": 0.8741128128502055, "step": 1170}, {"loss": 1.7682, "grad_norm": 0.3714233636856079, "learning_rate": 0.0002, "epoch": 0.8815838625326858, "step": 1180}, {"loss": 1.7549, "grad_norm": 0.2950296998023987, "learning_rate": 0.0002, "epoch": 0.8890549122151662, "step": 1190}, {"loss": 1.7612, "grad_norm": 0.38182979822158813, "learning_rate": 0.0002, "epoch": 0.8965259618976467, "step": 1200}, {"loss": 1.827, "grad_norm": 0.27883678674697876, "learning_rate": 0.0002, "epoch": 0.903997011580127, "step": 1210}, {"loss": 1.7623, "grad_norm": 0.33874374628067017, "learning_rate": 0.0002, "epoch": 0.9114680612626074, "step": 1220}, {"loss": 1.7334, "grad_norm": 0.3014272153377533, "learning_rate": 0.0002, "epoch": 0.9189391109450877, "step": 1230}, {"loss": 1.8235, "grad_norm": 0.3194271922111511, "learning_rate": 0.0002, "epoch": 0.9264101606275682, "step": 1240}, {"loss": 1.7924, "grad_norm": 0.3049403429031372, "learning_rate": 0.0002, "epoch": 0.9338812103100486, "step": 1250}, {"loss": 1.7535, "grad_norm": 0.30621254444122314, "learning_rate": 0.0002, "epoch": 0.9413522599925289, "step": 1260}, {"loss": 1.8287, "grad_norm": 0.28675132989883423, "learning_rate": 0.0002, "epoch": 0.9488233096750094, "step": 1270}, {"loss": 1.7586, "grad_norm": 0.3322032690048218, "learning_rate": 0.0002, "epoch": 0.9562943593574897, "step": 1280}, {"loss": 1.8054, "grad_norm": 0.35408294200897217, "learning_rate": 0.0002, "epoch": 0.9637654090399701, "step": 1290}, {"loss": 1.7343, "grad_norm": 0.36386919021606445, "learning_rate": 0.0002, "epoch": 0.9712364587224505, "step": 1300}, {"loss": 1.8633, "grad_norm": 0.32338324189186096, "learning_rate": 0.0002, "epoch": 0.9787075084049309, "step": 1310}, {"loss": 1.7724, "grad_norm": 0.3714013993740082, "learning_rate": 0.0002, "epoch": 0.9861785580874113, "step": 1320}, {"loss": 1.7766, "grad_norm": 0.3133082389831543, "learning_rate": 0.0002, "epoch": 0.9936496077698916, "step": 1330}, {"eval_loss": 1.8051470518112183, "eval_runtime": 38.6332, "eval_samples_per_second": 13.331, "eval_steps_per_second": 1.682, "epoch": 0.9996264475158759, "step": 1338}, {"loss": 1.8035, "grad_norm": 0.31595754623413086, "learning_rate": 0.0002, "epoch": 1.001120657452372, "step": 1340}, {"loss": 1.7486, "grad_norm": 0.3095700144767761, "learning_rate": 0.0002, "epoch": 1.0085917071348525, "step": 1350}, {"loss": 1.6981, "grad_norm": 0.34677496552467346, "learning_rate": 0.0002, "epoch": 1.0160627568173328, "step": 1360}, {"loss": 1.7377, "grad_norm": 0.29108840227127075, "learning_rate": 0.0002, "epoch": 1.0235338064998132, "step": 1370}, {"loss": 1.7194, "grad_norm": 0.32356950640678406, "learning_rate": 0.0002, "epoch": 1.0310048561822935, "step": 1380}, {"loss": 1.7593, "grad_norm": 0.4200669229030609, "learning_rate": 0.0002, "epoch": 1.038475905864774, "step": 1390}, {"loss": 1.797, "grad_norm": 0.3283711373806, "learning_rate": 0.0002, "epoch": 1.0459469555472545, "step": 1400}, {"loss": 1.7163, "grad_norm": 0.32898256182670593, "learning_rate": 0.0002, "epoch": 1.0534180052297348, "step": 1410}, {"loss": 1.7559, "grad_norm": 0.38790300488471985, "learning_rate": 0.0002, "epoch": 1.0608890549122152, "step": 1420}, {"loss": 1.6922, "grad_norm": 0.339800089597702, "learning_rate": 0.0002, "epoch": 1.0683601045946955, "step": 1430}, {"loss": 1.7076, "grad_norm": 0.3548751175403595, "learning_rate": 0.0002, "epoch": 1.075831154277176, "step": 1440}, {"loss": 1.6985, "grad_norm": 0.35114359855651855, "learning_rate": 0.0002, "epoch": 1.0833022039596563, "step": 1450}, {"loss": 1.7217, "grad_norm": 0.35226720571517944, "learning_rate": 0.0002, "epoch": 1.0907732536421366, "step": 1460}, {"loss": 1.6822, "grad_norm": 0.33665576577186584, "learning_rate": 0.0002, "epoch": 1.0982443033246172, "step": 1470}, {"loss": 1.6699, "grad_norm": 0.363889217376709, "learning_rate": 0.0002, "epoch": 1.1057153530070976, "step": 1480}, {"loss": 1.7933, "grad_norm": 0.3826201856136322, "learning_rate": 0.0002, "epoch": 1.113186402689578, "step": 1490}, {"loss": 1.7022, "grad_norm": 0.34058740735054016, "learning_rate": 0.0002, "epoch": 1.1206574523720583, "step": 1500}, {"loss": 1.6375, "grad_norm": 0.3462134301662445, "learning_rate": 0.0002, "epoch": 1.1281285020545386, "step": 1510}, {"loss": 1.7147, "grad_norm": 0.3396756052970886, "learning_rate": 0.0002, "epoch": 1.135599551737019, "step": 1520}, {"loss": 1.7219, "grad_norm": 0.32004743814468384, "learning_rate": 0.0002, "epoch": 1.1430706014194993, "step": 1530}, {"loss": 1.743, "grad_norm": 0.3397733271121979, "learning_rate": 0.0002, "epoch": 1.15054165110198, "step": 1540}, {"loss": 1.7333, "grad_norm": 0.3783262073993683, "learning_rate": 0.0002, "epoch": 1.1580127007844603, "step": 1550}, {"loss": 1.6075, "grad_norm": 0.35121291875839233, "learning_rate": 0.0002, "epoch": 1.1654837504669406, "step": 1560}, {"loss": 1.678, "grad_norm": 0.35816895961761475, "learning_rate": 0.0002, "epoch": 1.172954800149421, "step": 1570}, {"loss": 1.7143, "grad_norm": 0.33843839168548584, "learning_rate": 0.0002, "epoch": 1.1804258498319014, "step": 1580}, {"loss": 1.7434, "grad_norm": 0.3371972143650055, "learning_rate": 0.0002, "epoch": 1.1878968995143817, "step": 1590}, {"loss": 1.7671, "grad_norm": 0.36016878485679626, "learning_rate": 0.0002, "epoch": 1.195367949196862, "step": 1600}, {"loss": 1.6914, "grad_norm": 0.40879473090171814, "learning_rate": 0.0002, "epoch": 1.2028389988793426, "step": 1610}, {"loss": 1.6955, "grad_norm": 0.3216715455055237, "learning_rate": 0.0002, "epoch": 1.210310048561823, "step": 1620}, {"loss": 1.632, "grad_norm": 0.4482610821723938, "learning_rate": 0.0002, "epoch": 1.2177810982443034, "step": 1630}, {"loss": 1.6999, "grad_norm": 0.3257700502872467, "learning_rate": 0.0002, "epoch": 1.2252521479267837, "step": 1640}, {"loss": 1.7177, "grad_norm": 0.38646459579467773, "learning_rate": 0.0002, "epoch": 1.232723197609264, "step": 1650}, {"loss": 1.7081, "grad_norm": 0.4081360697746277, "learning_rate": 0.0002, "epoch": 1.2401942472917444, "step": 1660}, {"loss": 1.7519, "grad_norm": 0.4326848089694977, "learning_rate": 0.0002, "epoch": 1.2476652969742248, "step": 1670}, {"loss": 1.6752, "grad_norm": 0.346401572227478, "learning_rate": 0.0002, "epoch": 1.2551363466567054, "step": 1680}, {"loss": 1.7425, "grad_norm": 0.34536251425743103, "learning_rate": 0.0002, "epoch": 1.2626073963391857, "step": 1690}, {"loss": 1.7061, "grad_norm": 0.41359591484069824, "learning_rate": 0.0002, "epoch": 1.270078446021666, "step": 1700}, {"loss": 1.7906, "grad_norm": 0.3530874252319336, "learning_rate": 0.0002, "epoch": 1.2775494957041464, "step": 1710}, {"loss": 1.7357, "grad_norm": 0.3702719211578369, "learning_rate": 0.0002, "epoch": 1.2850205453866268, "step": 1720}, {"loss": 1.766, "grad_norm": 0.3703329563140869, "learning_rate": 0.0002, "epoch": 1.2924915950691072, "step": 1730}, {"loss": 1.7221, "grad_norm": 0.37919729948043823, "learning_rate": 0.0002, "epoch": 1.2999626447515875, "step": 1740}, {"loss": 1.7859, "grad_norm": 0.32526856660842896, "learning_rate": 0.0002, "epoch": 1.307433694434068, "step": 1750}, {"loss": 1.7117, "grad_norm": 0.36752620339393616, "learning_rate": 0.0002, "epoch": 1.3149047441165485, "step": 1760}, {"loss": 1.7335, "grad_norm": 0.3398192524909973, "learning_rate": 0.0002, "epoch": 1.3223757937990288, "step": 1770}, {"loss": 1.7492, "grad_norm": 0.37435585260391235, "learning_rate": 0.0002, "epoch": 1.3298468434815092, "step": 1780}, {"loss": 1.7393, "grad_norm": 0.35793280601501465, "learning_rate": 0.0002, "epoch": 1.3373178931639895, "step": 1790}, {"loss": 1.7266, "grad_norm": 0.35481882095336914, "learning_rate": 0.0002, "epoch": 1.3447889428464699, "step": 1800}, {"loss": 1.7456, "grad_norm": 0.3786393105983734, "learning_rate": 0.0002, "epoch": 1.3522599925289502, "step": 1810}, {"loss": 1.7169, "grad_norm": 0.33245593309402466, "learning_rate": 0.0002, "epoch": 1.3597310422114308, "step": 1820}, {"loss": 1.7577, "grad_norm": 0.35388344526290894, "learning_rate": 0.0002, "epoch": 1.3672020918939112, "step": 1830}, {"loss": 1.6968, "grad_norm": 0.3695325553417206, "learning_rate": 0.0002, "epoch": 1.3746731415763915, "step": 1840}, {"loss": 1.7086, "grad_norm": 0.3683604598045349, "learning_rate": 0.0002, "epoch": 1.382144191258872, "step": 1850}, {"loss": 1.7878, "grad_norm": 0.3753012418746948, "learning_rate": 0.0002, "epoch": 1.3896152409413522, "step": 1860}, {"loss": 1.6969, "grad_norm": 0.3331069350242615, "learning_rate": 0.0002, "epoch": 1.3970862906238326, "step": 1870}, {"loss": 1.6644, "grad_norm": 0.3877500295639038, "learning_rate": 0.0002, "epoch": 1.404557340306313, "step": 1880}, {"loss": 1.7586, "grad_norm": 0.33525151014328003, "learning_rate": 0.0002, "epoch": 1.4120283899887935, "step": 1890}, {"loss": 1.7031, "grad_norm": 0.3697299659252167, "learning_rate": 0.0002, "epoch": 1.4194994396712737, "step": 1900}, {"loss": 1.6956, "grad_norm": 0.4029286205768585, "learning_rate": 0.0002, "epoch": 1.4269704893537543, "step": 1910}, {"loss": 1.6897, "grad_norm": 0.3596203029155731, "learning_rate": 0.0002, "epoch": 1.4344415390362346, "step": 1920}, {"loss": 1.7139, "grad_norm": 0.450783908367157, "learning_rate": 0.0002, "epoch": 1.441912588718715, "step": 1930}, {"loss": 1.7243, "grad_norm": 0.3651481866836548, "learning_rate": 0.0002, "epoch": 1.4493836384011953, "step": 1940}, {"loss": 1.6637, "grad_norm": 0.3608424663543701, "learning_rate": 0.0002, "epoch": 1.4568546880836757, "step": 1950}, {"loss": 1.8285, "grad_norm": 0.39684420824050903, "learning_rate": 0.0002, "epoch": 1.4643257377661563, "step": 1960}, {"loss": 1.7514, "grad_norm": 0.34618663787841797, "learning_rate": 0.0002, "epoch": 1.4717967874486364, "step": 1970}, {"loss": 1.6655, "grad_norm": 0.4150386452674866, "learning_rate": 0.0002, "epoch": 1.479267837131117, "step": 1980}, {"loss": 1.7021, "grad_norm": 0.35500776767730713, "learning_rate": 0.0002, "epoch": 1.4867388868135973, "step": 1990}, {"loss": 1.7322, "grad_norm": 0.344144344329834, "learning_rate": 0.0002, "epoch": 1.4942099364960777, "step": 2000}, {"loss": 1.6998, "grad_norm": 0.3340149223804474, "learning_rate": 0.0002, "epoch": 1.501680986178558, "step": 2010}, {"loss": 1.7508, "grad_norm": 0.37685006856918335, "learning_rate": 0.0002, "epoch": 1.5091520358610384, "step": 2020}, {"loss": 1.8299, "grad_norm": 0.3699876368045807, "learning_rate": 0.0002, "epoch": 1.516623085543519, "step": 2030}, {"loss": 1.7357, "grad_norm": 0.3370307385921478, "learning_rate": 0.0002, "epoch": 1.5240941352259991, "step": 2040}, {"loss": 1.8044, "grad_norm": 0.37780630588531494, "learning_rate": 0.0002, "epoch": 1.5315651849084797, "step": 2050}, {"loss": 1.7408, "grad_norm": 0.370259165763855, "learning_rate": 0.0002, "epoch": 1.53903623459096, "step": 2060}, {"loss": 1.7398, "grad_norm": 0.3440011441707611, "learning_rate": 0.0002, "epoch": 1.5465072842734404, "step": 2070}, {"loss": 1.7105, "grad_norm": 0.40382063388824463, "learning_rate": 0.0002, "epoch": 1.5539783339559208, "step": 2080}, {"loss": 1.7071, "grad_norm": 0.38002029061317444, "learning_rate": 0.0002, "epoch": 1.5614493836384011, "step": 2090}, {"loss": 1.6815, "grad_norm": 0.3658451437950134, "learning_rate": 0.0002, "epoch": 1.5689204333208817, "step": 2100}, {"loss": 1.7598, "grad_norm": 0.354842871427536, "learning_rate": 0.0002, "epoch": 1.5763914830033618, "step": 2110}, {"loss": 1.6898, "grad_norm": 0.34735530614852905, "learning_rate": 0.0002, "epoch": 1.5838625326858424, "step": 2120}, {"loss": 1.7363, "grad_norm": 0.377581924200058, "learning_rate": 0.0002, "epoch": 1.5913335823683228, "step": 2130}, {"loss": 1.7789, "grad_norm": 0.41254034638404846, "learning_rate": 0.0002, "epoch": 1.5988046320508031, "step": 2140}, {"loss": 1.6782, "grad_norm": 0.3630715310573578, "learning_rate": 0.0002, "epoch": 1.6062756817332835, "step": 2150}, {"loss": 1.7531, "grad_norm": 0.36980143189430237, "learning_rate": 0.0002, "epoch": 1.6137467314157639, "step": 2160}, {"loss": 1.6847, "grad_norm": 0.3634769320487976, "learning_rate": 0.0002, "epoch": 1.6212177810982444, "step": 2170}, {"loss": 1.6367, "grad_norm": 0.3794139623641968, "learning_rate": 0.0002, "epoch": 1.6286888307807246, "step": 2180}, {"loss": 1.7064, "grad_norm": 0.359742134809494, "learning_rate": 0.0002, "epoch": 1.6361598804632052, "step": 2190}, {"loss": 1.7027, "grad_norm": 0.3770543932914734, "learning_rate": 0.0002, "epoch": 1.6436309301456855, "step": 2200}, {"loss": 1.784, "grad_norm": 0.3797036409378052, "learning_rate": 0.0002, "epoch": 1.6511019798281659, "step": 2210}, {"loss": 1.7875, "grad_norm": 0.35622093081474304, "learning_rate": 0.0002, "epoch": 1.6585730295106462, "step": 2220}, {"loss": 1.6615, "grad_norm": 0.34552520513534546, "learning_rate": 0.0002, "epoch": 1.6660440791931266, "step": 2230}, {"loss": 1.7522, "grad_norm": 0.379926860332489, "learning_rate": 0.0002, "epoch": 1.6735151288756072, "step": 2240}, {"loss": 1.7953, "grad_norm": 0.37083810567855835, "learning_rate": 0.0002, "epoch": 1.6809861785580873, "step": 2250}, {"loss": 1.7485, "grad_norm": 0.42746543884277344, "learning_rate": 0.0002, "epoch": 1.6884572282405679, "step": 2260}, {"loss": 1.776, "grad_norm": 0.3372884690761566, "learning_rate": 0.0002, "epoch": 1.6959282779230482, "step": 2270}, {"loss": 1.7604, "grad_norm": 0.35220256447792053, "learning_rate": 0.0002, "epoch": 1.7033993276055286, "step": 2280}, {"loss": 1.7154, "grad_norm": 0.3659130930900574, "learning_rate": 0.0002, "epoch": 1.710870377288009, "step": 2290}, {"loss": 1.6953, "grad_norm": 0.37629297375679016, "learning_rate": 0.0002, "epoch": 1.7183414269704893, "step": 2300}, {"loss": 1.7212, "grad_norm": 0.36312398314476013, "learning_rate": 0.0002, "epoch": 1.7258124766529699, "step": 2310}, {"loss": 1.7903, "grad_norm": 0.467709481716156, "learning_rate": 0.0002, "epoch": 1.73328352633545, "step": 2320}, {"loss": 1.696, "grad_norm": 0.38685527443885803, "learning_rate": 0.0002, "epoch": 1.7407545760179306, "step": 2330}, {"loss": 1.7041, "grad_norm": 0.3578338325023651, "learning_rate": 0.0002, "epoch": 1.748225625700411, "step": 2340}, {"loss": 1.6456, "grad_norm": 0.36057502031326294, "learning_rate": 0.0002, "epoch": 1.7556966753828913, "step": 2350}, {"loss": 1.6853, "grad_norm": 0.3615196645259857, "learning_rate": 0.0002, "epoch": 1.7631677250653717, "step": 2360}, {"loss": 1.7612, "grad_norm": 0.4118947684764862, "learning_rate": 0.0002, "epoch": 1.770638774747852, "step": 2370}, {"loss": 1.6946, "grad_norm": 0.4067276120185852, "learning_rate": 0.0002, "epoch": 1.7781098244303326, "step": 2380}, {"loss": 1.712, "grad_norm": 0.3979823887348175, "learning_rate": 0.0002, "epoch": 1.7855808741128127, "step": 2390}, {"loss": 1.7644, "grad_norm": 0.44045883417129517, "learning_rate": 0.0002, "epoch": 1.7930519237952933, "step": 2400}, {"loss": 1.7251, "grad_norm": 0.3998069167137146, "learning_rate": 0.0002, "epoch": 1.8005229734777737, "step": 2410}, {"loss": 1.7354, "grad_norm": 0.3450094759464264, "learning_rate": 0.0002, "epoch": 1.807994023160254, "step": 2420}, {"loss": 1.6998, "grad_norm": 0.3759009838104248, "learning_rate": 0.0002, "epoch": 1.8154650728427344, "step": 2430}, {"loss": 1.7706, "grad_norm": 0.34347015619277954, "learning_rate": 0.0002, "epoch": 1.8229361225252148, "step": 2440}, {"loss": 1.7345, "grad_norm": 0.3511228859424591, "learning_rate": 0.0002, "epoch": 1.8304071722076953, "step": 2450}, {"loss": 1.6909, "grad_norm": 0.36853715777397156, "learning_rate": 0.0002, "epoch": 1.8378782218901755, "step": 2460}, {"loss": 1.6931, "grad_norm": 0.40659376978874207, "learning_rate": 0.0002, "epoch": 1.845349271572656, "step": 2470}, {"loss": 1.7626, "grad_norm": 0.39621320366859436, "learning_rate": 0.0002, "epoch": 1.8528203212551362, "step": 2480}, {"loss": 1.7427, "grad_norm": 0.3753979504108429, "learning_rate": 0.0002, "epoch": 1.8602913709376168, "step": 2490}, {"loss": 1.6622, "grad_norm": 0.3811938464641571, "learning_rate": 0.0002, "epoch": 1.8677624206200971, "step": 2500}, {"loss": 1.7718, "grad_norm": 0.3432596027851105, "learning_rate": 0.0002, "epoch": 1.8752334703025775, "step": 2510}, {"loss": 1.7488, "grad_norm": 0.3670712113380432, "learning_rate": 0.0002, "epoch": 1.882704519985058, "step": 2520}, {"loss": 1.705, "grad_norm": 0.40907177329063416, "learning_rate": 0.0002, "epoch": 1.8901755696675382, "step": 2530}, {"loss": 1.7148, "grad_norm": 0.3821999728679657, "learning_rate": 0.0002, "epoch": 1.8976466193500188, "step": 2540}, {"loss": 1.7934, "grad_norm": 0.36173978447914124, "learning_rate": 0.0002, "epoch": 1.905117669032499, "step": 2550}, {"loss": 1.6939, "grad_norm": 0.38990336656570435, "learning_rate": 0.0002, "epoch": 1.9125887187149795, "step": 2560}, {"loss": 1.6893, "grad_norm": 0.35242322087287903, "learning_rate": 0.0002, "epoch": 1.9200597683974598, "step": 2570}, {"loss": 1.7268, "grad_norm": 0.3506428003311157, "learning_rate": 0.0002, "epoch": 1.9275308180799402, "step": 2580}, {"loss": 1.6953, "grad_norm": 0.39540135860443115, "learning_rate": 0.0002, "epoch": 1.9350018677624208, "step": 2590}, {"loss": 1.6511, "grad_norm": 0.3444725573062897, "learning_rate": 0.0002, "epoch": 1.942472917444901, "step": 2600}, {"loss": 1.7259, "grad_norm": 0.3963521718978882, "learning_rate": 0.0002, "epoch": 1.9499439671273815, "step": 2610}, {"loss": 1.6946, "grad_norm": 0.3689815402030945, "learning_rate": 0.0002, "epoch": 1.9574150168098616, "step": 2620}, {"loss": 1.7384, "grad_norm": 0.3482626676559448, "learning_rate": 0.0002, "epoch": 1.9648860664923422, "step": 2630}, {"loss": 1.7048, "grad_norm": 0.35832616686820984, "learning_rate": 0.0002, "epoch": 1.9723571161748226, "step": 2640}, {"loss": 1.6681, "grad_norm": 0.4776208996772766, "learning_rate": 0.0002, "epoch": 1.979828165857303, "step": 2650}, {"loss": 1.6696, "grad_norm": 0.32570165395736694, "learning_rate": 0.0002, "epoch": 1.9872992155397835, "step": 2660}, {"loss": 1.7232, "grad_norm": 0.3380725085735321, "learning_rate": 0.0002, "epoch": 1.9947702652222636, "step": 2670}, {"eval_loss": 1.8046749830245972, "eval_runtime": 38.5096, "eval_samples_per_second": 13.373, "eval_steps_per_second": 1.688, "epoch": 2.0, "step": 2677}, {"loss": 1.7265, "grad_norm": 0.36817631125450134, "learning_rate": 0.0002, "epoch": 2.002241314904744, "step": 2680}, {"loss": 1.548, "grad_norm": 0.4056456685066223, "learning_rate": 0.0002, "epoch": 2.0097123645872244, "step": 2690}, {"loss": 1.5515, "grad_norm": 0.37416863441467285, "learning_rate": 0.0002, "epoch": 2.017183414269705, "step": 2700}, {"loss": 1.5895, "grad_norm": 0.4273638427257538, "learning_rate": 0.0002, "epoch": 2.024654463952185, "step": 2710}, {"loss": 1.5884, "grad_norm": 0.36497923731803894, "learning_rate": 0.0002, "epoch": 2.0321255136346656, "step": 2720}, {"loss": 1.6999, "grad_norm": 0.5021994113922119, "learning_rate": 0.0002, "epoch": 2.0395965633171462, "step": 2730}, {"loss": 1.6655, "grad_norm": 0.45896220207214355, "learning_rate": 0.0002, "epoch": 2.0470676129996264, "step": 2740}, {"loss": 1.6305, "grad_norm": 0.3973815143108368, "learning_rate": 0.0002, "epoch": 2.054538662682107, "step": 2750}, {"loss": 1.6301, "grad_norm": 0.4521815776824951, "learning_rate": 0.0002, "epoch": 2.062009712364587, "step": 2760}, {"loss": 1.6189, "grad_norm": 0.42775002121925354, "learning_rate": 0.0002, "epoch": 2.0694807620470677, "step": 2770}, {"loss": 1.6491, "grad_norm": 0.48158586025238037, "learning_rate": 0.0002, "epoch": 2.076951811729548, "step": 2780}, {"loss": 1.6301, "grad_norm": 0.4612371623516083, "learning_rate": 0.0002, "epoch": 2.0844228614120284, "step": 2790}, {"loss": 1.6327, "grad_norm": 0.42536866664886475, "learning_rate": 0.0002, "epoch": 2.091893911094509, "step": 2800}, {"loss": 1.651, "grad_norm": 0.48515772819519043, "learning_rate": 0.0002, "epoch": 2.099364960776989, "step": 2810}, {"loss": 1.6829, "grad_norm": 0.41418662667274475, "learning_rate": 0.0002, "epoch": 2.1068360104594697, "step": 2820}, {"loss": 1.6266, "grad_norm": 0.4683697819709778, "learning_rate": 0.0002, "epoch": 2.11430706014195, "step": 2830}, {"loss": 1.6586, "grad_norm": 0.4484657049179077, "learning_rate": 0.0002, "epoch": 2.1217781098244304, "step": 2840}, {"loss": 1.6483, "grad_norm": 0.6621400713920593, "learning_rate": 0.0002, "epoch": 2.1292491595069105, "step": 2850}, {"loss": 1.5755, "grad_norm": 0.45074811577796936, "learning_rate": 0.0002, "epoch": 2.136720209189391, "step": 2860}, {"loss": 1.6456, "grad_norm": 0.3513113558292389, "learning_rate": 0.0002, "epoch": 2.1441912588718717, "step": 2870}, {"loss": 1.6081, "grad_norm": 0.40411314368247986, "learning_rate": 0.0002, "epoch": 2.151662308554352, "step": 2880}, {"loss": 1.6323, "grad_norm": 0.4121065139770508, "learning_rate": 0.0002, "epoch": 2.1591333582368324, "step": 2890}, {"loss": 1.6324, "grad_norm": 0.44723689556121826, "learning_rate": 0.0002, "epoch": 2.1666044079193125, "step": 2900}, {"loss": 1.5699, "grad_norm": 0.4226122498512268, "learning_rate": 0.0002, "epoch": 2.174075457601793, "step": 2910}, {"loss": 1.5652, "grad_norm": 0.46617650985717773, "learning_rate": 0.0002, "epoch": 2.1815465072842732, "step": 2920}, {"loss": 1.6378, "grad_norm": 0.4506422281265259, "learning_rate": 0.0002, "epoch": 2.189017556966754, "step": 2930}, {"loss": 1.6112, "grad_norm": 0.4892672896385193, "learning_rate": 0.0002, "epoch": 2.1964886066492344, "step": 2940}, {"loss": 1.6176, "grad_norm": 0.44095516204833984, "learning_rate": 0.0002, "epoch": 2.2039596563317145, "step": 2950}, {"loss": 1.6058, "grad_norm": 0.41522109508514404, "learning_rate": 0.0002, "epoch": 2.211430706014195, "step": 2960}, {"loss": 1.5964, "grad_norm": 0.4860858917236328, "learning_rate": 0.0002, "epoch": 2.2189017556966752, "step": 2970}, {"loss": 1.6427, "grad_norm": 0.42662516236305237, "learning_rate": 0.0002, "epoch": 2.226372805379156, "step": 2980}, {"loss": 1.6313, "grad_norm": 0.4390648305416107, "learning_rate": 0.0002, "epoch": 2.233843855061636, "step": 2990}, {"loss": 1.5992, "grad_norm": 0.47515565156936646, "learning_rate": 0.0002, "epoch": 2.2413149047441165, "step": 3000}, {"loss": 1.5563, "grad_norm": 0.4104543924331665, "learning_rate": 0.0002, "epoch": 2.248785954426597, "step": 3010}, {"loss": 1.6895, "grad_norm": 0.4404028654098511, "learning_rate": 0.0002, "epoch": 2.2562570041090773, "step": 3020}, {"loss": 1.6088, "grad_norm": 0.4717366695404053, "learning_rate": 0.0002, "epoch": 2.263728053791558, "step": 3030}, {"loss": 1.7287, "grad_norm": 0.48345857858657837, "learning_rate": 0.0002, "epoch": 2.271199103474038, "step": 3040}, {"loss": 1.681, "grad_norm": 0.5312452912330627, "learning_rate": 0.0002, "epoch": 2.2786701531565186, "step": 3050}, {"loss": 1.5901, "grad_norm": 0.5073099732398987, "learning_rate": 0.0002, "epoch": 2.2861412028389987, "step": 3060}, {"loss": 1.6914, "grad_norm": 0.5027463436126709, "learning_rate": 0.0002, "epoch": 2.2936122525214793, "step": 3070}, {"loss": 1.5862, "grad_norm": 0.5436304807662964, "learning_rate": 0.0002, "epoch": 2.30108330220396, "step": 3080}, {"loss": 1.5763, "grad_norm": 0.4701065123081207, "learning_rate": 0.0002, "epoch": 2.30855435188644, "step": 3090}, {"loss": 1.6177, "grad_norm": 0.46988746523857117, "learning_rate": 0.0002, "epoch": 2.3160254015689206, "step": 3100}, {"loss": 1.6502, "grad_norm": 0.45112869143486023, "learning_rate": 0.0002, "epoch": 2.3234964512514007, "step": 3110}, {"loss": 1.6291, "grad_norm": 0.5173566937446594, "learning_rate": 0.0002, "epoch": 2.3309675009338813, "step": 3120}, {"loss": 1.6743, "grad_norm": 0.40345850586891174, "learning_rate": 0.0002, "epoch": 2.3384385506163614, "step": 3130}, {"loss": 1.621, "grad_norm": 0.4218924939632416, "learning_rate": 0.0002, "epoch": 2.345909600298842, "step": 3140}, {"loss": 1.6341, "grad_norm": 0.41857317090034485, "learning_rate": 0.0002, "epoch": 2.3533806499813226, "step": 3150}, {"loss": 1.6087, "grad_norm": 0.4197218418121338, "learning_rate": 0.0002, "epoch": 2.3608516996638027, "step": 3160}, {"loss": 1.6572, "grad_norm": 0.4260677397251129, "learning_rate": 0.0002, "epoch": 2.3683227493462833, "step": 3170}, {"loss": 1.6376, "grad_norm": 0.4209042191505432, "learning_rate": 0.0002, "epoch": 2.3757937990287634, "step": 3180}, {"loss": 1.634, "grad_norm": 0.4092234969139099, "learning_rate": 0.0002, "epoch": 2.383264848711244, "step": 3190}, {"loss": 1.6339, "grad_norm": 0.4928431510925293, "learning_rate": 0.0002, "epoch": 2.390735898393724, "step": 3200}, {"loss": 1.6015, "grad_norm": 0.49252402782440186, "learning_rate": 0.0002, "epoch": 2.3982069480762047, "step": 3210}, {"loss": 1.5773, "grad_norm": 0.4368397295475006, "learning_rate": 0.0002, "epoch": 2.4056779977586853, "step": 3220}, {"loss": 1.6629, "grad_norm": 0.46122390031814575, "learning_rate": 0.0002, "epoch": 2.4131490474411654, "step": 3230}, {"loss": 1.6224, "grad_norm": 0.4272301197052002, "learning_rate": 0.0002, "epoch": 2.420620097123646, "step": 3240}, {"loss": 1.5961, "grad_norm": 0.41480937600135803, "learning_rate": 0.0002, "epoch": 2.428091146806126, "step": 3250}, {"loss": 1.6281, "grad_norm": 0.48911941051483154, "learning_rate": 0.0002, "epoch": 2.4355621964886067, "step": 3260}, {"loss": 1.6846, "grad_norm": 0.4444098472595215, "learning_rate": 0.0002, "epoch": 2.443033246171087, "step": 3270}, {"loss": 1.6961, "grad_norm": 0.5111684799194336, "learning_rate": 0.0002, "epoch": 2.4505042958535674, "step": 3280}, {"loss": 1.6152, "grad_norm": 0.5058825016021729, "learning_rate": 0.0002, "epoch": 2.457975345536048, "step": 3290}, {"loss": 1.625, "grad_norm": 0.44173210859298706, "learning_rate": 0.0002, "epoch": 2.465446395218528, "step": 3300}, {"loss": 1.6491, "grad_norm": 0.4659745991230011, "learning_rate": 0.0002, "epoch": 2.4729174449010087, "step": 3310}, {"loss": 1.6114, "grad_norm": 0.47237497568130493, "learning_rate": 0.0002, "epoch": 2.480388494583489, "step": 3320}, {"loss": 1.6193, "grad_norm": 0.47303131222724915, "learning_rate": 0.0002, "epoch": 2.4878595442659694, "step": 3330}, {"loss": 1.7256, "grad_norm": 0.4522389769554138, "learning_rate": 0.0002, "epoch": 2.4953305939484496, "step": 3340}, {"loss": 1.6834, "grad_norm": 0.4467332363128662, "learning_rate": 0.0002, "epoch": 2.50280164363093, "step": 3350}, {"loss": 1.6108, "grad_norm": 0.4413762092590332, "learning_rate": 0.0002, "epoch": 2.5102726933134107, "step": 3360}, {"loss": 1.537, "grad_norm": 0.495514452457428, "learning_rate": 0.0002, "epoch": 2.517743742995891, "step": 3370}, {"loss": 1.5839, "grad_norm": 0.4429773986339569, "learning_rate": 0.0002, "epoch": 2.5252147926783715, "step": 3380}, {"loss": 1.6522, "grad_norm": 0.4589079022407532, "learning_rate": 0.0002, "epoch": 2.5326858423608516, "step": 3390}, {"loss": 1.6529, "grad_norm": 0.4683997333049774, "learning_rate": 0.0002, "epoch": 2.540156892043332, "step": 3400}, {"loss": 1.6745, "grad_norm": 0.4651731252670288, "learning_rate": 0.0002, "epoch": 2.5476279417258123, "step": 3410}, {"loss": 1.5918, "grad_norm": 0.45818084478378296, "learning_rate": 0.0002, "epoch": 2.555098991408293, "step": 3420}, {"loss": 1.6326, "grad_norm": 0.45209529995918274, "learning_rate": 0.0002, "epoch": 2.5625700410907735, "step": 3430}, {"loss": 1.5606, "grad_norm": 0.4344733655452728, "learning_rate": 0.0002, "epoch": 2.5700410907732536, "step": 3440}, {"loss": 1.6748, "grad_norm": 0.47435566782951355, "learning_rate": 0.0002, "epoch": 2.577512140455734, "step": 3450}, {"loss": 1.6237, "grad_norm": 0.43841999769210815, "learning_rate": 0.0002, "epoch": 2.5849831901382143, "step": 3460}, {"loss": 1.7207, "grad_norm": 0.4323869049549103, "learning_rate": 0.0002, "epoch": 2.592454239820695, "step": 3470}, {"loss": 1.5494, "grad_norm": 0.44355881214141846, "learning_rate": 0.0002, "epoch": 2.599925289503175, "step": 3480}, {"loss": 1.665, "grad_norm": 0.45847779512405396, "learning_rate": 0.0002, "epoch": 2.6073963391856556, "step": 3490}, {"loss": 1.6006, "grad_norm": 0.4411061704158783, "learning_rate": 0.0002, "epoch": 2.614867388868136, "step": 3500}, {"loss": 1.5868, "grad_norm": 0.4446796178817749, "learning_rate": 0.0002, "epoch": 2.6223384385506163, "step": 3510}, {"loss": 1.5946, "grad_norm": 0.41969653964042664, "learning_rate": 0.0002, "epoch": 2.629809488233097, "step": 3520}, {"loss": 1.6798, "grad_norm": 0.5263747572898865, "learning_rate": 0.0002, "epoch": 2.637280537915577, "step": 3530}, {"loss": 1.6309, "grad_norm": 0.47719451785087585, "learning_rate": 0.0002, "epoch": 2.6447515875980576, "step": 3540}, {"loss": 1.7024, "grad_norm": 0.46574118733406067, "learning_rate": 0.0002, "epoch": 2.6522226372805378, "step": 3550}, {"loss": 1.618, "grad_norm": 0.46867135167121887, "learning_rate": 0.0002, "epoch": 2.6596936869630183, "step": 3560}, {"loss": 1.5885, "grad_norm": 0.4441198706626892, "learning_rate": 0.0002, "epoch": 2.667164736645499, "step": 3570}, {"loss": 1.6426, "grad_norm": 0.4871319830417633, "learning_rate": 0.0002, "epoch": 2.674635786327979, "step": 3580}, {"loss": 1.6575, "grad_norm": 0.43900373578071594, "learning_rate": 0.0002, "epoch": 2.6821068360104596, "step": 3590}, {"loss": 1.6071, "grad_norm": 0.42509549856185913, "learning_rate": 0.0002, "epoch": 2.6895778856929398, "step": 3600}, {"loss": 1.5651, "grad_norm": 0.4691086709499359, "learning_rate": 0.0002, "epoch": 2.6970489353754203, "step": 3610}, {"loss": 1.5491, "grad_norm": 0.46318942308425903, "learning_rate": 0.0002, "epoch": 2.7045199850579005, "step": 3620}, {"loss": 1.5422, "grad_norm": 0.44631096720695496, "learning_rate": 0.0002, "epoch": 2.711991034740381, "step": 3630}, {"loss": 1.6831, "grad_norm": 0.42315489053726196, "learning_rate": 0.0002, "epoch": 2.7194620844228616, "step": 3640}, {"loss": 1.6008, "grad_norm": 0.4971241056919098, "learning_rate": 0.0002, "epoch": 2.7269331341053418, "step": 3650}, {"loss": 1.6042, "grad_norm": 0.4578486382961273, "learning_rate": 0.0002, "epoch": 2.7344041837878224, "step": 3660}, {"loss": 1.6076, "grad_norm": 0.46584776043891907, "learning_rate": 0.0002, "epoch": 2.7418752334703025, "step": 3670}, {"loss": 1.6809, "grad_norm": 0.4951731264591217, "learning_rate": 0.0002, "epoch": 2.749346283152783, "step": 3680}, {"loss": 1.6226, "grad_norm": 0.4935225546360016, "learning_rate": 0.0002, "epoch": 2.756817332835263, "step": 3690}, {"loss": 1.5878, "grad_norm": 0.41805586218833923, "learning_rate": 0.0002, "epoch": 2.764288382517744, "step": 3700}, {"loss": 1.7173, "grad_norm": 0.4417555630207062, "learning_rate": 0.0002, "epoch": 2.7717594322002244, "step": 3710}, {"loss": 1.6398, "grad_norm": 0.48229655623435974, "learning_rate": 0.0002, "epoch": 2.7792304818827045, "step": 3720}, {"loss": 1.6074, "grad_norm": 0.48562315106391907, "learning_rate": 0.0002, "epoch": 2.786701531565185, "step": 3730}, {"loss": 1.607, "grad_norm": 0.4473940432071686, "learning_rate": 0.0002, "epoch": 2.794172581247665, "step": 3740}, {"loss": 1.6065, "grad_norm": 0.4626813232898712, "learning_rate": 0.0002, "epoch": 2.801643630930146, "step": 3750}, {"loss": 1.6296, "grad_norm": 0.4339792728424072, "learning_rate": 0.0002, "epoch": 2.809114680612626, "step": 3760}, {"loss": 1.6815, "grad_norm": 0.5250858068466187, "learning_rate": 0.0002, "epoch": 2.8165857302951065, "step": 3770}, {"loss": 1.6644, "grad_norm": 0.4537523090839386, "learning_rate": 0.0002, "epoch": 2.824056779977587, "step": 3780}, {"loss": 1.6535, "grad_norm": 0.5646113157272339, "learning_rate": 0.0002, "epoch": 2.831527829660067, "step": 3790}, {"loss": 1.5712, "grad_norm": 0.44243332743644714, "learning_rate": 0.0002, "epoch": 2.8389988793425474, "step": 3800}, {"loss": 1.6478, "grad_norm": 0.4585791826248169, "learning_rate": 0.0002, "epoch": 2.846469929025028, "step": 3810}, {"loss": 1.6854, "grad_norm": 0.489702045917511, "learning_rate": 0.0002, "epoch": 2.8539409787075085, "step": 3820}, {"loss": 1.7066, "grad_norm": 0.502470850944519, "learning_rate": 0.0002, "epoch": 2.8614120283899886, "step": 3830}, {"loss": 1.5785, "grad_norm": 0.4395960867404938, "learning_rate": 0.0002, "epoch": 2.8688830780724692, "step": 3840}, {"loss": 1.6434, "grad_norm": 0.4348670244216919, "learning_rate": 0.0002, "epoch": 2.87635412775495, "step": 3850}, {"loss": 1.6163, "grad_norm": 0.48852720856666565, "learning_rate": 0.0002, "epoch": 2.88382517743743, "step": 3860}, {"loss": 1.5916, "grad_norm": 0.45317450165748596, "learning_rate": 0.0002, "epoch": 2.89129622711991, "step": 3870}, {"loss": 1.6486, "grad_norm": 0.4732758700847626, "learning_rate": 0.0002, "epoch": 2.8987672768023907, "step": 3880}, {"loss": 1.6758, "grad_norm": 0.45238012075424194, "learning_rate": 0.0002, "epoch": 2.9062383264848712, "step": 3890}, {"loss": 1.6228, "grad_norm": 0.48838064074516296, "learning_rate": 0.0002, "epoch": 2.9137093761673514, "step": 3900}, {"loss": 1.658, "grad_norm": 0.43496349453926086, "learning_rate": 0.0002, "epoch": 2.921180425849832, "step": 3910}, {"loss": 1.7063, "grad_norm": 0.47963935136795044, "learning_rate": 0.0002, "epoch": 2.9286514755323125, "step": 3920}, {"loss": 1.6553, "grad_norm": 0.4544987976551056, "learning_rate": 0.0002, "epoch": 2.9361225252147927, "step": 3930}, {"loss": 1.6192, "grad_norm": 0.4622892141342163, "learning_rate": 0.0002, "epoch": 2.943593574897273, "step": 3940}, {"loss": 1.6178, "grad_norm": 0.47026222944259644, "learning_rate": 0.0002, "epoch": 2.9510646245797534, "step": 3950}, {"loss": 1.6612, "grad_norm": 0.4549552798271179, "learning_rate": 0.0002, "epoch": 2.958535674262234, "step": 3960}, {"loss": 1.6458, "grad_norm": 0.46647515892982483, "learning_rate": 0.0002, "epoch": 2.966006723944714, "step": 3970}, {"loss": 1.6051, "grad_norm": 0.45095112919807434, "learning_rate": 0.0002, "epoch": 2.9734777736271947, "step": 3980}, {"loss": 1.6471, "grad_norm": 0.4690017104148865, "learning_rate": 0.0002, "epoch": 2.9809488233096753, "step": 3990}, {"loss": 1.6061, "grad_norm": 0.4603444039821625, "learning_rate": 0.0002, "epoch": 2.9884198729921554, "step": 4000}, {"loss": 1.6431, "grad_norm": 0.4743294417858124, "learning_rate": 0.0002, "epoch": 2.9958909226746355, "step": 4010}, {"eval_loss": 1.8252571821212769, "eval_runtime": 38.7853, "eval_samples_per_second": 13.278, "eval_steps_per_second": 1.676, "epoch": 2.999626447515876, "step": 4015}, {"loss": 1.6512, "grad_norm": 0.4919724464416504, "learning_rate": 0.0002, "epoch": 3.003361972357116, "step": 4020}, {"loss": 1.5354, "grad_norm": 0.4747185707092285, "learning_rate": 0.0002, "epoch": 3.0108330220395967, "step": 4030}, {"loss": 1.568, "grad_norm": 0.4797595143318176, "learning_rate": 0.0002, "epoch": 3.018304071722077, "step": 4040}, {"loss": 1.5194, "grad_norm": 0.5450999140739441, "learning_rate": 0.0002, "epoch": 3.0257751214045574, "step": 4050}, {"loss": 1.5065, "grad_norm": 0.49058812856674194, "learning_rate": 0.0002, "epoch": 3.0332461710870375, "step": 4060}, {"loss": 1.4884, "grad_norm": 0.5219563841819763, "learning_rate": 0.0002, "epoch": 3.040717220769518, "step": 4070}, {"loss": 1.4742, "grad_norm": 0.515628457069397, "learning_rate": 0.0002, "epoch": 3.0481882704519987, "step": 4080}, {"loss": 1.5313, "grad_norm": 0.6145984530448914, "learning_rate": 0.0002, "epoch": 3.055659320134479, "step": 4090}, {"loss": 1.4989, "grad_norm": 0.6067144274711609, "learning_rate": 0.0002, "epoch": 3.0631303698169594, "step": 4100}, {"loss": 1.528, "grad_norm": 0.5773133039474487, "learning_rate": 0.0002, "epoch": 3.0706014194994395, "step": 4110}, {"loss": 1.5374, "grad_norm": 0.6894241571426392, "learning_rate": 0.0002, "epoch": 3.07807246918192, "step": 4120}, {"loss": 1.5422, "grad_norm": 0.6422514915466309, "learning_rate": 0.0002, "epoch": 3.0855435188644003, "step": 4130}, {"loss": 1.4724, "grad_norm": 0.6119855046272278, "learning_rate": 0.0002, "epoch": 3.093014568546881, "step": 4140}, {"loss": 1.5361, "grad_norm": 0.5847280025482178, "learning_rate": 0.0002, "epoch": 3.1004856182293614, "step": 4150}, {"loss": 1.5151, "grad_norm": 0.5401515960693359, "learning_rate": 0.0002, "epoch": 3.1079566679118416, "step": 4160}, {"loss": 1.502, "grad_norm": 0.6501587629318237, "learning_rate": 0.0002, "epoch": 3.115427717594322, "step": 4170}, {"loss": 1.4952, "grad_norm": 0.5988039374351501, "learning_rate": 0.0002, "epoch": 3.1228987672768023, "step": 4180}, {"loss": 1.5287, "grad_norm": 0.4982665181159973, "learning_rate": 0.0002, "epoch": 3.130369816959283, "step": 4190}, {"loss": 1.5078, "grad_norm": 0.5548039078712463, "learning_rate": 0.0002, "epoch": 3.137840866641763, "step": 4200}, {"loss": 1.4904, "grad_norm": 0.5920777320861816, "learning_rate": 0.0002, "epoch": 3.1453119163242436, "step": 4210}, {"loss": 1.442, "grad_norm": 0.6965190172195435, "learning_rate": 0.0002, "epoch": 3.152782966006724, "step": 4220}, {"loss": 1.557, "grad_norm": 0.5196244716644287, "learning_rate": 0.0002, "epoch": 3.1602540156892043, "step": 4230}, {"loss": 1.5706, "grad_norm": 0.6942682266235352, "learning_rate": 0.0002, "epoch": 3.167725065371685, "step": 4240}, {"loss": 1.5407, "grad_norm": 0.5765156149864197, "learning_rate": 0.0002, "epoch": 3.175196115054165, "step": 4250}, {"loss": 1.4963, "grad_norm": 0.5801976919174194, "learning_rate": 0.0002, "epoch": 3.1826671647366456, "step": 4260}, {"loss": 1.4988, "grad_norm": 0.6260752081871033, "learning_rate": 0.0002, "epoch": 3.1901382144191257, "step": 4270}, {"loss": 1.5074, "grad_norm": 0.6610770225524902, "learning_rate": 0.0002, "epoch": 3.1976092641016063, "step": 4280}, {"loss": 1.4657, "grad_norm": 0.5762143135070801, "learning_rate": 0.0002, "epoch": 3.205080313784087, "step": 4290}, {"loss": 1.5181, "grad_norm": 0.5926990509033203, "learning_rate": 0.0002, "epoch": 3.212551363466567, "step": 4300}, {"loss": 1.5492, "grad_norm": 0.7373854517936707, "learning_rate": 0.0002, "epoch": 3.2200224131490476, "step": 4310}, {"loss": 1.4648, "grad_norm": 0.5963311195373535, "learning_rate": 0.0002, "epoch": 3.2274934628315277, "step": 4320}, {"loss": 1.5262, "grad_norm": 0.5754616856575012, "learning_rate": 0.0002, "epoch": 3.2349645125140083, "step": 4330}, {"loss": 1.4767, "grad_norm": 0.6116095781326294, "learning_rate": 0.0002, "epoch": 3.2424355621964884, "step": 4340}, {"loss": 1.5008, "grad_norm": 0.6001536846160889, "learning_rate": 0.0002, "epoch": 3.249906611878969, "step": 4350}, {"loss": 1.5738, "grad_norm": 0.5270227789878845, "learning_rate": 0.0002, "epoch": 3.257377661561449, "step": 4360}, {"loss": 1.5235, "grad_norm": 0.6666602492332458, "learning_rate": 0.0002, "epoch": 3.2648487112439297, "step": 4370}, {"loss": 1.5665, "grad_norm": 0.520310640335083, "learning_rate": 0.0002, "epoch": 3.2723197609264103, "step": 4380}, {"loss": 1.542, "grad_norm": 0.5165975093841553, "learning_rate": 0.0002, "epoch": 3.2797908106088904, "step": 4390}, {"loss": 1.4746, "grad_norm": 0.6080228686332703, "learning_rate": 0.0002, "epoch": 3.287261860291371, "step": 4400}, {"loss": 1.4901, "grad_norm": 0.670122504234314, "learning_rate": 0.0002, "epoch": 3.294732909973851, "step": 4410}, {"loss": 1.4677, "grad_norm": 0.6019457578659058, "learning_rate": 0.0002, "epoch": 3.3022039596563317, "step": 4420}, {"loss": 1.4249, "grad_norm": 0.5519300103187561, "learning_rate": 0.0002, "epoch": 3.309675009338812, "step": 4430}, {"loss": 1.555, "grad_norm": 0.5958521962165833, "learning_rate": 0.0002, "epoch": 3.3171460590212924, "step": 4440}, {"loss": 1.5067, "grad_norm": 0.5552705526351929, "learning_rate": 0.0002, "epoch": 3.324617108703773, "step": 4450}, {"loss": 1.5926, "grad_norm": 0.6583784818649292, "learning_rate": 0.0002, "epoch": 3.332088158386253, "step": 4460}, {"loss": 1.4206, "grad_norm": 0.5815939903259277, "learning_rate": 0.0002, "epoch": 3.3395592080687337, "step": 4470}, {"loss": 1.5942, "grad_norm": 1.3342205286026, "learning_rate": 0.0002, "epoch": 3.347030257751214, "step": 4480}, {"loss": 1.484, "grad_norm": 0.6341500878334045, "learning_rate": 0.0002, "epoch": 3.3545013074336945, "step": 4490}, {"loss": 1.5219, "grad_norm": 0.6384079456329346, "learning_rate": 0.0002, "epoch": 3.3619723571161746, "step": 4500}, {"loss": 1.5222, "grad_norm": 0.6098346710205078, "learning_rate": 0.0002, "epoch": 3.369443406798655, "step": 4510}, {"loss": 1.5475, "grad_norm": 0.5958296656608582, "learning_rate": 0.0002, "epoch": 3.3769144564811358, "step": 4520}, {"loss": 1.5171, "grad_norm": 0.6157881617546082, "learning_rate": 0.0002, "epoch": 3.384385506163616, "step": 4530}, {"loss": 1.569, "grad_norm": 0.5671007037162781, "learning_rate": 0.0002, "epoch": 3.3918565558460965, "step": 4540}, {"loss": 1.604, "grad_norm": 0.6203294992446899, "learning_rate": 0.0002, "epoch": 3.3993276055285766, "step": 4550}, {"loss": 1.5364, "grad_norm": 0.6743317246437073, "learning_rate": 0.0002, "epoch": 3.406798655211057, "step": 4560}, {"loss": 1.5034, "grad_norm": 0.731765627861023, "learning_rate": 0.0002, "epoch": 3.4142697048935373, "step": 4570}, {"loss": 1.4585, "grad_norm": 0.6285187602043152, "learning_rate": 0.0002, "epoch": 3.421740754576018, "step": 4580}, {"loss": 1.5296, "grad_norm": 0.612680196762085, "learning_rate": 0.0002, "epoch": 3.4292118042584985, "step": 4590}, {"loss": 1.5577, "grad_norm": 0.6413681507110596, "learning_rate": 0.0002, "epoch": 3.4366828539409786, "step": 4600}, {"loss": 1.5026, "grad_norm": 0.6240990161895752, "learning_rate": 0.0002, "epoch": 3.444153903623459, "step": 4610}, {"loss": 1.5887, "grad_norm": 0.5095735192298889, "learning_rate": 0.0002, "epoch": 3.4516249533059393, "step": 4620}, {"loss": 1.4906, "grad_norm": 0.5699611902236938, "learning_rate": 0.0002, "epoch": 3.45909600298842, "step": 4630}, {"loss": 1.5176, "grad_norm": 0.7289775609970093, "learning_rate": 0.0002, "epoch": 3.4665670526709, "step": 4640}, {"loss": 1.5467, "grad_norm": 0.6211609840393066, "learning_rate": 0.0002, "epoch": 3.4740381023533806, "step": 4650}, {"loss": 1.533, "grad_norm": 0.5714802145957947, "learning_rate": 0.0002, "epoch": 3.481509152035861, "step": 4660}, {"loss": 1.5096, "grad_norm": 0.6287049651145935, "learning_rate": 0.0002, "epoch": 3.4889802017183413, "step": 4670}, {"loss": 1.4212, "grad_norm": 0.5480595827102661, "learning_rate": 0.0002, "epoch": 3.496451251400822, "step": 4680}, {"loss": 1.4746, "grad_norm": 0.5683253407478333, "learning_rate": 0.0002, "epoch": 3.503922301083302, "step": 4690}, {"loss": 1.5012, "grad_norm": 0.601140558719635, "learning_rate": 0.0002, "epoch": 3.5113933507657826, "step": 4700}, {"loss": 1.5383, "grad_norm": 0.5344498157501221, "learning_rate": 0.0002, "epoch": 3.5188644004482628, "step": 4710}, {"loss": 1.5428, "grad_norm": 0.5739690661430359, "learning_rate": 0.0002, "epoch": 3.5263354501307433, "step": 4720}, {"loss": 1.5589, "grad_norm": 0.5640085935592651, "learning_rate": 0.0002, "epoch": 3.533806499813224, "step": 4730}, {"loss": 1.487, "grad_norm": 0.5967805981636047, "learning_rate": 0.0002, "epoch": 3.541277549495704, "step": 4740}, {"loss": 1.5461, "grad_norm": 0.6138835549354553, "learning_rate": 0.0002, "epoch": 3.5487485991781846, "step": 4750}, {"loss": 1.5502, "grad_norm": 0.6779900193214417, "learning_rate": 0.0002, "epoch": 3.5562196488606648, "step": 4760}, {"loss": 1.4917, "grad_norm": 0.6122010350227356, "learning_rate": 0.0002, "epoch": 3.5636906985431454, "step": 4770}, {"loss": 1.5405, "grad_norm": 0.5685241222381592, "learning_rate": 0.0002, "epoch": 3.5711617482256255, "step": 4780}, {"loss": 1.5427, "grad_norm": 0.604583203792572, "learning_rate": 0.0002, "epoch": 3.578632797908106, "step": 4790}, {"loss": 1.4514, "grad_norm": 0.651165246963501, "learning_rate": 0.0002, "epoch": 3.5861038475905866, "step": 4800}, {"loss": 1.4109, "grad_norm": 0.6398511528968811, "learning_rate": 0.0002, "epoch": 3.593574897273067, "step": 4810}, {"loss": 1.4261, "grad_norm": 0.6444641351699829, "learning_rate": 0.0002, "epoch": 3.6010459469555474, "step": 4820}, {"loss": 1.5274, "grad_norm": 0.6018481850624084, "learning_rate": 0.0002, "epoch": 3.6085169966380275, "step": 4830}, {"loss": 1.4647, "grad_norm": 0.6025291085243225, "learning_rate": 0.0002, "epoch": 3.615988046320508, "step": 4840}, {"loss": 1.5609, "grad_norm": 0.6810156106948853, "learning_rate": 0.0002, "epoch": 3.623459096002988, "step": 4850}, {"loss": 1.5299, "grad_norm": 0.6408044695854187, "learning_rate": 0.0002, "epoch": 3.630930145685469, "step": 4860}, {"loss": 1.5366, "grad_norm": 0.5608272552490234, "learning_rate": 0.0002, "epoch": 3.6384011953679494, "step": 4870}, {"loss": 1.5188, "grad_norm": 0.6136814951896667, "learning_rate": 0.0002, "epoch": 3.6458722450504295, "step": 4880}, {"loss": 1.5021, "grad_norm": 0.5927900075912476, "learning_rate": 0.0002, "epoch": 3.65334329473291, "step": 4890}, {"loss": 1.6084, "grad_norm": 0.5336901545524597, "learning_rate": 0.0002, "epoch": 3.66081434441539, "step": 4900}, {"loss": 1.5701, "grad_norm": 0.7823320627212524, "learning_rate": 0.0002, "epoch": 3.668285394097871, "step": 4910}, {"loss": 1.4881, "grad_norm": 0.6703504323959351, "learning_rate": 0.0002, "epoch": 3.675756443780351, "step": 4920}, {"loss": 1.5332, "grad_norm": 0.6061160564422607, "learning_rate": 0.0002, "epoch": 3.6832274934628315, "step": 4930}, {"loss": 1.5405, "grad_norm": 0.6237227916717529, "learning_rate": 0.0002, "epoch": 3.690698543145312, "step": 4940}, {"loss": 1.497, "grad_norm": 0.5985278487205505, "learning_rate": 0.0002, "epoch": 3.6981695928277922, "step": 4950}, {"loss": 1.5132, "grad_norm": 0.6483839750289917, "learning_rate": 0.0002, "epoch": 3.705640642510273, "step": 4960}, {"loss": 1.5338, "grad_norm": 0.5788805484771729, "learning_rate": 0.0002, "epoch": 3.713111692192753, "step": 4970}, {"loss": 1.5258, "grad_norm": 0.5609974265098572, "learning_rate": 0.0002, "epoch": 3.7205827418752335, "step": 4980}, {"loss": 1.4759, "grad_norm": 0.5681300759315491, "learning_rate": 0.0002, "epoch": 3.7280537915577137, "step": 4990}, {"loss": 1.6018, "grad_norm": 0.5860186219215393, "learning_rate": 0.0002, "epoch": 3.7355248412401942, "step": 5000}, {"loss": 1.58, "grad_norm": 0.5718157291412354, "learning_rate": 0.0002, "epoch": 3.742995890922675, "step": 5010}, {"loss": 1.5834, "grad_norm": 0.6173721551895142, "learning_rate": 0.0002, "epoch": 3.750466940605155, "step": 5020}, {"loss": 1.5617, "grad_norm": 0.629152238368988, "learning_rate": 0.0002, "epoch": 3.7579379902876355, "step": 5030}, {"loss": 1.519, "grad_norm": 0.5666284561157227, "learning_rate": 0.0002, "epoch": 3.7654090399701157, "step": 5040}, {"loss": 1.5329, "grad_norm": 0.6053005456924438, "learning_rate": 0.0002, "epoch": 3.7728800896525962, "step": 5050}, {"loss": 1.5404, "grad_norm": 0.5870583057403564, "learning_rate": 0.0002, "epoch": 3.7803511393350764, "step": 5060}, {"loss": 1.4444, "grad_norm": 0.5422009229660034, "learning_rate": 0.0002, "epoch": 3.787822189017557, "step": 5070}, {"loss": 1.5308, "grad_norm": 0.5396918058395386, "learning_rate": 0.0002, "epoch": 3.7952932387000375, "step": 5080}, {"loss": 1.464, "grad_norm": 0.5544713139533997, "learning_rate": 0.0002, "epoch": 3.8027642883825177, "step": 5090}, {"loss": 1.4752, "grad_norm": 0.5983749628067017, "learning_rate": 0.0002, "epoch": 3.8102353380649983, "step": 5100}, {"loss": 1.4972, "grad_norm": 0.5702024102210999, "learning_rate": 0.0002, "epoch": 3.8177063877474784, "step": 5110}, {"loss": 1.5471, "grad_norm": 0.5436882376670837, "learning_rate": 0.0002, "epoch": 3.825177437429959, "step": 5120}, {"loss": 1.5118, "grad_norm": 0.5453617572784424, "learning_rate": 0.0002, "epoch": 3.832648487112439, "step": 5130}, {"loss": 1.5732, "grad_norm": 0.6269069314002991, "learning_rate": 0.0002, "epoch": 3.8401195367949197, "step": 5140}, {"loss": 1.4959, "grad_norm": 0.6189185380935669, "learning_rate": 0.0002, "epoch": 3.8475905864774003, "step": 5150}, {"loss": 1.4999, "grad_norm": 0.6653388142585754, "learning_rate": 0.0002, "epoch": 3.8550616361598804, "step": 5160}, {"loss": 1.5075, "grad_norm": 0.5771768689155579, "learning_rate": 0.0002, "epoch": 3.862532685842361, "step": 5170}, {"loss": 1.5545, "grad_norm": 0.6052790880203247, "learning_rate": 0.0002, "epoch": 3.870003735524841, "step": 5180}, {"loss": 1.4987, "grad_norm": 0.6572316884994507, "learning_rate": 0.0002, "epoch": 3.8774747852073217, "step": 5190}, {"loss": 1.5241, "grad_norm": 0.670576810836792, "learning_rate": 0.0002, "epoch": 3.884945834889802, "step": 5200}, {"loss": 1.4777, "grad_norm": 0.5728798508644104, "learning_rate": 0.0002, "epoch": 3.8924168845722824, "step": 5210}, {"loss": 1.5351, "grad_norm": 0.6340774297714233, "learning_rate": 0.0002, "epoch": 3.899887934254763, "step": 5220}, {"loss": 1.5081, "grad_norm": 0.5981315970420837, "learning_rate": 0.0002, "epoch": 3.907358983937243, "step": 5230}, {"loss": 1.4875, "grad_norm": 0.6212025880813599, "learning_rate": 0.0002, "epoch": 3.9148300336197237, "step": 5240}, {"loss": 1.5545, "grad_norm": 0.6202296018600464, "learning_rate": 0.0002, "epoch": 3.922301083302204, "step": 5250}, {"loss": 1.5765, "grad_norm": 0.6159142255783081, "learning_rate": 0.0002, "epoch": 3.9297721329846844, "step": 5260}, {"loss": 1.4938, "grad_norm": 0.6519438624382019, "learning_rate": 0.0002, "epoch": 3.9372431826671646, "step": 5270}, {"loss": 1.4859, "grad_norm": 0.539813756942749, "learning_rate": 0.0002, "epoch": 3.944714232349645, "step": 5280}, {"loss": 1.5921, "grad_norm": 0.6443665027618408, "learning_rate": 0.0002, "epoch": 3.9521852820321257, "step": 5290}, {"loss": 1.5153, "grad_norm": 0.6635757684707642, "learning_rate": 0.0002, "epoch": 3.959656331714606, "step": 5300}, {"loss": 1.5485, "grad_norm": 0.589363157749176, "learning_rate": 0.0002, "epoch": 3.9671273813970864, "step": 5310}, {"loss": 1.5498, "grad_norm": 0.5788735747337341, "learning_rate": 0.0002, "epoch": 3.9745984310795666, "step": 5320}, {"loss": 1.5607, "grad_norm": 0.5976864695549011, "learning_rate": 0.0002, "epoch": 3.982069480762047, "step": 5330}, {"loss": 1.5302, "grad_norm": 0.6624067425727844, "learning_rate": 0.0002, "epoch": 3.9895405304445273, "step": 5340}, {"loss": 1.5904, "grad_norm": 0.6738956570625305, "learning_rate": 0.0002, "epoch": 3.997011580127008, "step": 5350}]}
+{"epoch": 4.999626447515876, "step": 6692, "epoch_duration": 1697.2771589756012, "total_accumulated_duration": 7520.807319164276, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7672.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-42/checkpoint-2677", "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": 2.6181, "grad_norm": 0.4912872612476349, "learning_rate": 0.0002, "epoch": 0.007471049682480389, "step": 10}, {"loss": 2.2606, "grad_norm": 0.4856316149234772, "learning_rate": 0.0002, "epoch": 0.014942099364960777, "step": 20}, {"loss": 2.0957, "grad_norm": 0.47683125734329224, "learning_rate": 0.0002, "epoch": 0.022413149047441166, "step": 30}, {"loss": 1.8908, "grad_norm": 0.515082597732544, "learning_rate": 0.0002, "epoch": 0.029884198729921554, "step": 40}, {"loss": 1.9704, "grad_norm": 0.5299215316772461, "learning_rate": 0.0002, "epoch": 0.03735524841240194, "step": 50}, {"loss": 1.9225, "grad_norm": 0.4951399862766266, "learning_rate": 0.0002, "epoch": 0.04482629809488233, "step": 60}, {"loss": 1.9742, "grad_norm": 0.48079821467399597, "learning_rate": 0.0002, "epoch": 0.05229734777736272, "step": 70}, {"loss": 1.9466, "grad_norm": 0.49402132630348206, "learning_rate": 0.0002, "epoch": 0.05976839745984311, "step": 80}, {"loss": 1.8691, "grad_norm": 0.4778193235397339, "learning_rate": 0.0002, "epoch": 0.0672394471423235, "step": 90}, {"loss": 1.8455, "grad_norm": 0.42472657561302185, "learning_rate": 0.0002, "epoch": 0.07471049682480388, "step": 100}, {"loss": 1.8744, "grad_norm": 0.4433092474937439, "learning_rate": 0.0002, "epoch": 0.08218154650728428, "step": 110}, {"loss": 1.865, "grad_norm": 0.4472862780094147, "learning_rate": 0.0002, "epoch": 0.08965259618976466, "step": 120}, {"loss": 1.9256, "grad_norm": 0.42596298456192017, "learning_rate": 0.0002, "epoch": 0.09712364587224505, "step": 130}, {"loss": 1.8015, "grad_norm": 0.46645811200141907, "learning_rate": 0.0002, "epoch": 0.10459469555472543, "step": 140}, {"loss": 1.8307, "grad_norm": 0.41041234135627747, "learning_rate": 0.0002, "epoch": 0.11206574523720583, "step": 150}, {"loss": 1.8276, "grad_norm": 0.5329819917678833, "learning_rate": 0.0002, "epoch": 0.11953679491968622, "step": 160}, {"loss": 1.8118, "grad_norm": 0.4065922200679779, "learning_rate": 0.0002, "epoch": 0.1270078446021666, "step": 170}, {"loss": 1.8559, "grad_norm": 0.38406994938850403, "learning_rate": 0.0002, "epoch": 0.134478894284647, "step": 180}, {"loss": 1.8647, "grad_norm": 0.4246881306171417, "learning_rate": 0.0002, "epoch": 0.14194994396712737, "step": 190}, {"loss": 1.8054, "grad_norm": 0.35136649012565613, "learning_rate": 0.0002, "epoch": 0.14942099364960776, "step": 200}, {"loss": 1.802, "grad_norm": 0.43252742290496826, "learning_rate": 0.0002, "epoch": 0.15689204333208817, "step": 210}, {"loss": 1.7823, "grad_norm": 0.39236941933631897, "learning_rate": 0.0002, "epoch": 0.16436309301456856, "step": 220}, {"loss": 1.818, "grad_norm": 0.3748249113559723, "learning_rate": 0.0002, "epoch": 0.17183414269704894, "step": 230}, {"loss": 1.866, "grad_norm": 0.6432855725288391, "learning_rate": 0.0002, "epoch": 0.17930519237952933, "step": 240}, {"loss": 1.8397, "grad_norm": 0.34874802827835083, "learning_rate": 0.0002, "epoch": 0.1867762420620097, "step": 250}, {"loss": 1.79, "grad_norm": 0.3721984326839447, "learning_rate": 0.0002, "epoch": 0.1942472917444901, "step": 260}, {"loss": 1.8464, "grad_norm": 0.4339311420917511, "learning_rate": 0.0002, "epoch": 0.20171834142697048, "step": 270}, {"loss": 1.8665, "grad_norm": 0.4018215537071228, "learning_rate": 0.0002, "epoch": 0.20918939110945087, "step": 280}, {"loss": 1.8048, "grad_norm": 0.3278839886188507, "learning_rate": 0.0002, "epoch": 0.21666044079193125, "step": 290}, {"loss": 1.7395, "grad_norm": 0.36146077513694763, "learning_rate": 0.0002, "epoch": 0.22413149047441167, "step": 300}, {"loss": 1.7916, "grad_norm": 0.38175010681152344, "learning_rate": 0.0002, "epoch": 0.23160254015689205, "step": 310}, {"loss": 1.8593, "grad_norm": 0.44776618480682373, "learning_rate": 0.0002, "epoch": 0.23907358983937244, "step": 320}, {"loss": 1.7824, "grad_norm": 0.3933652937412262, "learning_rate": 0.0002, "epoch": 0.24654463952185282, "step": 330}, {"loss": 1.8393, "grad_norm": 0.3515005111694336, "learning_rate": 0.0002, "epoch": 0.2540156892043332, "step": 340}, {"loss": 1.8653, "grad_norm": 0.6683304309844971, "learning_rate": 0.0002, "epoch": 0.2614867388868136, "step": 350}, {"loss": 1.8797, "grad_norm": 0.37093454599380493, "learning_rate": 0.0002, "epoch": 0.268957788569294, "step": 360}, {"loss": 1.8251, "grad_norm": 0.3450651168823242, "learning_rate": 0.0002, "epoch": 0.2764288382517744, "step": 370}, {"loss": 1.7435, "grad_norm": 0.5140917301177979, "learning_rate": 0.0002, "epoch": 0.28389988793425475, "step": 380}, {"loss": 1.8026, "grad_norm": 0.32885563373565674, "learning_rate": 0.0002, "epoch": 0.29137093761673516, "step": 390}, {"loss": 1.8174, "grad_norm": 0.33962297439575195, "learning_rate": 0.0002, "epoch": 0.2988419872992155, "step": 400}, {"loss": 1.7467, "grad_norm": 0.3723141849040985, "learning_rate": 0.0002, "epoch": 0.30631303698169593, "step": 410}, {"loss": 1.8459, "grad_norm": 0.37173134088516235, "learning_rate": 0.0002, "epoch": 0.31378408666417634, "step": 420}, {"loss": 1.8876, "grad_norm": 0.33736956119537354, "learning_rate": 0.0002, "epoch": 0.3212551363466567, "step": 430}, {"loss": 1.8367, "grad_norm": 0.3602448105812073, "learning_rate": 0.0002, "epoch": 0.3287261860291371, "step": 440}, {"loss": 1.8058, "grad_norm": 0.3569699227809906, "learning_rate": 0.0002, "epoch": 0.33619723571161747, "step": 450}, {"loss": 1.8086, "grad_norm": 0.31009167432785034, "learning_rate": 0.0002, "epoch": 0.3436682853940979, "step": 460}, {"loss": 1.8876, "grad_norm": 0.5278693437576294, "learning_rate": 0.0002, "epoch": 0.35113933507657824, "step": 470}, {"loss": 1.8534, "grad_norm": 0.3587537109851837, "learning_rate": 0.0002, "epoch": 0.35861038475905865, "step": 480}, {"loss": 1.8046, "grad_norm": 0.3859670162200928, "learning_rate": 0.0002, "epoch": 0.366081434441539, "step": 490}, {"loss": 1.8287, "grad_norm": 0.395913690328598, "learning_rate": 0.0002, "epoch": 0.3735524841240194, "step": 500}, {"loss": 1.7619, "grad_norm": 0.35052940249443054, "learning_rate": 0.0002, "epoch": 0.38102353380649984, "step": 510}, {"loss": 1.7824, "grad_norm": 0.2979494333267212, "learning_rate": 0.0002, "epoch": 0.3884945834889802, "step": 520}, {"loss": 1.8641, "grad_norm": 0.3062683343887329, "learning_rate": 0.0002, "epoch": 0.3959656331714606, "step": 530}, {"loss": 1.7651, "grad_norm": 0.3172847330570221, "learning_rate": 0.0002, "epoch": 0.40343668285394096, "step": 540}, {"loss": 1.806, "grad_norm": 0.360435426235199, "learning_rate": 0.0002, "epoch": 0.4109077325364214, "step": 550}, {"loss": 1.9054, "grad_norm": 0.3427872359752655, "learning_rate": 0.0002, "epoch": 0.41837878221890173, "step": 560}, {"loss": 1.7562, "grad_norm": 0.34036558866500854, "learning_rate": 0.0002, "epoch": 0.42584983190138215, "step": 570}, {"loss": 1.7254, "grad_norm": 0.3365345299243927, "learning_rate": 0.0002, "epoch": 0.4333208815838625, "step": 580}, {"loss": 1.8328, "grad_norm": 0.35619041323661804, "learning_rate": 0.0002, "epoch": 0.4407919312663429, "step": 590}, {"loss": 1.8114, "grad_norm": 0.3569088280200958, "learning_rate": 0.0002, "epoch": 0.44826298094882333, "step": 600}, {"loss": 1.8599, "grad_norm": 0.3581278622150421, "learning_rate": 0.0002, "epoch": 0.4557340306313037, "step": 610}, {"loss": 1.7078, "grad_norm": 0.43197110295295715, "learning_rate": 0.0002, "epoch": 0.4632050803137841, "step": 620}, {"loss": 1.8257, "grad_norm": 0.33966198563575745, "learning_rate": 0.0002, "epoch": 0.47067612999626446, "step": 630}, {"loss": 1.7528, "grad_norm": 0.3343866467475891, "learning_rate": 0.0002, "epoch": 0.47814717967874487, "step": 640}, {"loss": 1.8191, "grad_norm": 0.33878564834594727, "learning_rate": 0.0002, "epoch": 0.48561822936122523, "step": 650}, {"loss": 1.8801, "grad_norm": 0.387195885181427, "learning_rate": 0.0002, "epoch": 0.49308927904370564, "step": 660}, {"loss": 1.7559, "grad_norm": 0.3755440413951874, "learning_rate": 0.0002, "epoch": 0.500560328726186, "step": 670}, {"loss": 1.8057, "grad_norm": 0.3272816836833954, "learning_rate": 0.0002, "epoch": 0.5080313784086664, "step": 680}, {"loss": 1.8156, "grad_norm": 0.36063864827156067, "learning_rate": 0.0002, "epoch": 0.5155024280911468, "step": 690}, {"loss": 1.8397, "grad_norm": 0.35317373275756836, "learning_rate": 0.0002, "epoch": 0.5229734777736272, "step": 700}, {"loss": 1.7603, "grad_norm": 0.3561195433139801, "learning_rate": 0.0002, "epoch": 0.5304445274561076, "step": 710}, {"loss": 1.8149, "grad_norm": 0.31124624609947205, "learning_rate": 0.0002, "epoch": 0.537915577138588, "step": 720}, {"loss": 1.7434, "grad_norm": 0.3294544517993927, "learning_rate": 0.0002, "epoch": 0.5453866268210683, "step": 730}, {"loss": 1.8027, "grad_norm": 0.31933900713920593, "learning_rate": 0.0002, "epoch": 0.5528576765035488, "step": 740}, {"loss": 1.7601, "grad_norm": 0.3226020634174347, "learning_rate": 0.0002, "epoch": 0.5603287261860291, "step": 750}, {"loss": 1.7862, "grad_norm": 0.3147525489330292, "learning_rate": 0.0002, "epoch": 0.5677997758685095, "step": 760}, {"loss": 1.9028, "grad_norm": 0.32234328985214233, "learning_rate": 0.0002, "epoch": 0.57527082555099, "step": 770}, {"loss": 1.7623, "grad_norm": 0.3258664309978485, "learning_rate": 0.0002, "epoch": 0.5827418752334703, "step": 780}, {"loss": 1.7384, "grad_norm": 0.3166961967945099, "learning_rate": 0.0002, "epoch": 0.5902129249159507, "step": 790}, {"loss": 1.8799, "grad_norm": 0.35621458292007446, "learning_rate": 0.0002, "epoch": 0.597683974598431, "step": 800}, {"loss": 1.8313, "grad_norm": 0.3236999213695526, "learning_rate": 0.0002, "epoch": 0.6051550242809115, "step": 810}, {"loss": 1.7132, "grad_norm": 0.2892923653125763, "learning_rate": 0.0002, "epoch": 0.6126260739633919, "step": 820}, {"loss": 1.8709, "grad_norm": 0.4098321497440338, "learning_rate": 0.0002, "epoch": 0.6200971236458722, "step": 830}, {"loss": 1.7637, "grad_norm": 0.3337118923664093, "learning_rate": 0.0002, "epoch": 0.6275681733283527, "step": 840}, {"loss": 1.7375, "grad_norm": 0.30416029691696167, "learning_rate": 0.0002, "epoch": 0.635039223010833, "step": 850}, {"loss": 1.7419, "grad_norm": 0.3361026346683502, "learning_rate": 0.0002, "epoch": 0.6425102726933134, "step": 860}, {"loss": 1.732, "grad_norm": 0.3537365198135376, "learning_rate": 0.0002, "epoch": 0.6499813223757938, "step": 870}, {"loss": 1.7825, "grad_norm": 0.33854469656944275, "learning_rate": 0.0002, "epoch": 0.6574523720582742, "step": 880}, {"loss": 1.7561, "grad_norm": 0.3332272469997406, "learning_rate": 0.0002, "epoch": 0.6649234217407546, "step": 890}, {"loss": 1.7247, "grad_norm": 0.34954726696014404, "learning_rate": 0.0002, "epoch": 0.6723944714232349, "step": 900}, {"loss": 1.7917, "grad_norm": 0.2921750247478485, "learning_rate": 0.0002, "epoch": 0.6798655211057153, "step": 910}, {"loss": 1.7807, "grad_norm": 0.30508682131767273, "learning_rate": 0.0002, "epoch": 0.6873365707881958, "step": 920}, {"loss": 1.8082, "grad_norm": 0.32268425822257996, "learning_rate": 0.0002, "epoch": 0.6948076204706761, "step": 930}, {"loss": 1.8283, "grad_norm": 0.2844390869140625, "learning_rate": 0.0002, "epoch": 0.7022786701531565, "step": 940}, {"loss": 1.7363, "grad_norm": 0.31263890862464905, "learning_rate": 0.0002, "epoch": 0.709749719835637, "step": 950}, {"loss": 1.8081, "grad_norm": 0.3626808822154999, "learning_rate": 0.0002, "epoch": 0.7172207695181173, "step": 960}, {"loss": 1.853, "grad_norm": 0.3322749733924866, "learning_rate": 0.0002, "epoch": 0.7246918192005977, "step": 970}, {"loss": 1.7912, "grad_norm": 0.29177871346473694, "learning_rate": 0.0002, "epoch": 0.732162868883078, "step": 980}, {"loss": 1.8447, "grad_norm": 0.35405513644218445, "learning_rate": 0.0002, "epoch": 0.7396339185655585, "step": 990}, {"loss": 1.7008, "grad_norm": 0.39318400621414185, "learning_rate": 0.0002, "epoch": 0.7471049682480388, "step": 1000}, {"loss": 1.7803, "grad_norm": 0.29401418566703796, "learning_rate": 0.0002, "epoch": 0.7545760179305192, "step": 1010}, {"loss": 1.7649, "grad_norm": 0.3271748721599579, "learning_rate": 0.0002, "epoch": 0.7620470676129997, "step": 1020}, {"loss": 1.7266, "grad_norm": 0.30883970856666565, "learning_rate": 0.0002, "epoch": 0.76951811729548, "step": 1030}, {"loss": 1.7722, "grad_norm": 0.3411838412284851, "learning_rate": 0.0002, "epoch": 0.7769891669779604, "step": 1040}, {"loss": 1.829, "grad_norm": 0.30608129501342773, "learning_rate": 0.0002, "epoch": 0.7844602166604407, "step": 1050}, {"loss": 1.7815, "grad_norm": 0.30899080634117126, "learning_rate": 0.0002, "epoch": 0.7919312663429212, "step": 1060}, {"loss": 1.7625, "grad_norm": 0.3160453140735626, "learning_rate": 0.0002, "epoch": 0.7994023160254016, "step": 1070}, {"loss": 1.8452, "grad_norm": 0.30947187542915344, "learning_rate": 0.0002, "epoch": 0.8068733657078819, "step": 1080}, {"loss": 1.7418, "grad_norm": 0.3103134036064148, "learning_rate": 0.0002, "epoch": 0.8143444153903624, "step": 1090}, {"loss": 1.842, "grad_norm": 0.31771138310432434, "learning_rate": 0.0002, "epoch": 0.8218154650728428, "step": 1100}, {"loss": 1.7918, "grad_norm": 0.5860997438430786, "learning_rate": 0.0002, "epoch": 0.8292865147553231, "step": 1110}, {"loss": 1.8443, "grad_norm": 0.3230148255825043, "learning_rate": 0.0002, "epoch": 0.8367575644378035, "step": 1120}, {"loss": 1.8478, "grad_norm": 0.29611510038375854, "learning_rate": 0.0002, "epoch": 0.8442286141202839, "step": 1130}, {"loss": 1.7673, "grad_norm": 0.3373654782772064, "learning_rate": 0.0002, "epoch": 0.8516996638027643, "step": 1140}, {"loss": 1.7997, "grad_norm": 0.3474279046058655, "learning_rate": 0.0002, "epoch": 0.8591707134852447, "step": 1150}, {"loss": 1.75, "grad_norm": 0.35057875514030457, "learning_rate": 0.0002, "epoch": 0.866641763167725, "step": 1160}, {"loss": 1.8273, "grad_norm": 0.39537495374679565, "learning_rate": 0.0002, "epoch": 0.8741128128502055, "step": 1170}, {"loss": 1.7682, "grad_norm": 0.3714233636856079, "learning_rate": 0.0002, "epoch": 0.8815838625326858, "step": 1180}, {"loss": 1.7549, "grad_norm": 0.2950296998023987, "learning_rate": 0.0002, "epoch": 0.8890549122151662, "step": 1190}, {"loss": 1.7612, "grad_norm": 0.38182979822158813, "learning_rate": 0.0002, "epoch": 0.8965259618976467, "step": 1200}, {"loss": 1.827, "grad_norm": 0.27883678674697876, "learning_rate": 0.0002, "epoch": 0.903997011580127, "step": 1210}, {"loss": 1.7623, "grad_norm": 0.33874374628067017, "learning_rate": 0.0002, "epoch": 0.9114680612626074, "step": 1220}, {"loss": 1.7334, "grad_norm": 0.3014272153377533, "learning_rate": 0.0002, "epoch": 0.9189391109450877, "step": 1230}, {"loss": 1.8235, "grad_norm": 0.3194271922111511, "learning_rate": 0.0002, "epoch": 0.9264101606275682, "step": 1240}, {"loss": 1.7924, "grad_norm": 0.3049403429031372, "learning_rate": 0.0002, "epoch": 0.9338812103100486, "step": 1250}, {"loss": 1.7535, "grad_norm": 0.30621254444122314, "learning_rate": 0.0002, "epoch": 0.9413522599925289, "step": 1260}, {"loss": 1.8287, "grad_norm": 0.28675132989883423, "learning_rate": 0.0002, "epoch": 0.9488233096750094, "step": 1270}, {"loss": 1.7586, "grad_norm": 0.3322032690048218, "learning_rate": 0.0002, "epoch": 0.9562943593574897, "step": 1280}, {"loss": 1.8054, "grad_norm": 0.35408294200897217, "learning_rate": 0.0002, "epoch": 0.9637654090399701, "step": 1290}, {"loss": 1.7343, "grad_norm": 0.36386919021606445, "learning_rate": 0.0002, "epoch": 0.9712364587224505, "step": 1300}, {"loss": 1.8633, "grad_norm": 0.32338324189186096, "learning_rate": 0.0002, "epoch": 0.9787075084049309, "step": 1310}, {"loss": 1.7724, "grad_norm": 0.3714013993740082, "learning_rate": 0.0002, "epoch": 0.9861785580874113, "step": 1320}, {"loss": 1.7766, "grad_norm": 0.3133082389831543, "learning_rate": 0.0002, "epoch": 0.9936496077698916, "step": 1330}, {"eval_loss": 1.8051470518112183, "eval_runtime": 38.6332, "eval_samples_per_second": 13.331, "eval_steps_per_second": 1.682, "epoch": 0.9996264475158759, "step": 1338}, {"loss": 1.8035, "grad_norm": 0.31595754623413086, "learning_rate": 0.0002, "epoch": 1.001120657452372, "step": 1340}, {"loss": 1.7486, "grad_norm": 0.3095700144767761, "learning_rate": 0.0002, "epoch": 1.0085917071348525, "step": 1350}, {"loss": 1.6981, "grad_norm": 0.34677496552467346, "learning_rate": 0.0002, "epoch": 1.0160627568173328, "step": 1360}, {"loss": 1.7377, "grad_norm": 0.29108840227127075, "learning_rate": 0.0002, "epoch": 1.0235338064998132, "step": 1370}, {"loss": 1.7194, "grad_norm": 0.32356950640678406, "learning_rate": 0.0002, "epoch": 1.0310048561822935, "step": 1380}, {"loss": 1.7593, "grad_norm": 0.4200669229030609, "learning_rate": 0.0002, "epoch": 1.038475905864774, "step": 1390}, {"loss": 1.797, "grad_norm": 0.3283711373806, "learning_rate": 0.0002, "epoch": 1.0459469555472545, "step": 1400}, {"loss": 1.7163, "grad_norm": 0.32898256182670593, "learning_rate": 0.0002, "epoch": 1.0534180052297348, "step": 1410}, {"loss": 1.7559, "grad_norm": 0.38790300488471985, "learning_rate": 0.0002, "epoch": 1.0608890549122152, "step": 1420}, {"loss": 1.6922, "grad_norm": 0.339800089597702, "learning_rate": 0.0002, "epoch": 1.0683601045946955, "step": 1430}, {"loss": 1.7076, "grad_norm": 0.3548751175403595, "learning_rate": 0.0002, "epoch": 1.075831154277176, "step": 1440}, {"loss": 1.6985, "grad_norm": 0.35114359855651855, "learning_rate": 0.0002, "epoch": 1.0833022039596563, "step": 1450}, {"loss": 1.7217, "grad_norm": 0.35226720571517944, "learning_rate": 0.0002, "epoch": 1.0907732536421366, "step": 1460}, {"loss": 1.6822, "grad_norm": 0.33665576577186584, "learning_rate": 0.0002, "epoch": 1.0982443033246172, "step": 1470}, {"loss": 1.6699, "grad_norm": 0.363889217376709, "learning_rate": 0.0002, "epoch": 1.1057153530070976, "step": 1480}, {"loss": 1.7933, "grad_norm": 0.3826201856136322, "learning_rate": 0.0002, "epoch": 1.113186402689578, "step": 1490}, {"loss": 1.7022, "grad_norm": 0.34058740735054016, "learning_rate": 0.0002, "epoch": 1.1206574523720583, "step": 1500}, {"loss": 1.6375, "grad_norm": 0.3462134301662445, "learning_rate": 0.0002, "epoch": 1.1281285020545386, "step": 1510}, {"loss": 1.7147, "grad_norm": 0.3396756052970886, "learning_rate": 0.0002, "epoch": 1.135599551737019, "step": 1520}, {"loss": 1.7219, "grad_norm": 0.32004743814468384, "learning_rate": 0.0002, "epoch": 1.1430706014194993, "step": 1530}, {"loss": 1.743, "grad_norm": 0.3397733271121979, "learning_rate": 0.0002, "epoch": 1.15054165110198, "step": 1540}, {"loss": 1.7333, "grad_norm": 0.3783262073993683, "learning_rate": 0.0002, "epoch": 1.1580127007844603, "step": 1550}, {"loss": 1.6075, "grad_norm": 0.35121291875839233, "learning_rate": 0.0002, "epoch": 1.1654837504669406, "step": 1560}, {"loss": 1.678, "grad_norm": 0.35816895961761475, "learning_rate": 0.0002, "epoch": 1.172954800149421, "step": 1570}, {"loss": 1.7143, "grad_norm": 0.33843839168548584, "learning_rate": 0.0002, "epoch": 1.1804258498319014, "step": 1580}, {"loss": 1.7434, "grad_norm": 0.3371972143650055, "learning_rate": 0.0002, "epoch": 1.1878968995143817, "step": 1590}, {"loss": 1.7671, "grad_norm": 0.36016878485679626, "learning_rate": 0.0002, "epoch": 1.195367949196862, "step": 1600}, {"loss": 1.6914, "grad_norm": 0.40879473090171814, "learning_rate": 0.0002, "epoch": 1.2028389988793426, "step": 1610}, {"loss": 1.6955, "grad_norm": 0.3216715455055237, "learning_rate": 0.0002, "epoch": 1.210310048561823, "step": 1620}, {"loss": 1.632, "grad_norm": 0.4482610821723938, "learning_rate": 0.0002, "epoch": 1.2177810982443034, "step": 1630}, {"loss": 1.6999, "grad_norm": 0.3257700502872467, "learning_rate": 0.0002, "epoch": 1.2252521479267837, "step": 1640}, {"loss": 1.7177, "grad_norm": 0.38646459579467773, "learning_rate": 0.0002, "epoch": 1.232723197609264, "step": 1650}, {"loss": 1.7081, "grad_norm": 0.4081360697746277, "learning_rate": 0.0002, "epoch": 1.2401942472917444, "step": 1660}, {"loss": 1.7519, "grad_norm": 0.4326848089694977, "learning_rate": 0.0002, "epoch": 1.2476652969742248, "step": 1670}, {"loss": 1.6752, "grad_norm": 0.346401572227478, "learning_rate": 0.0002, "epoch": 1.2551363466567054, "step": 1680}, {"loss": 1.7425, "grad_norm": 0.34536251425743103, "learning_rate": 0.0002, "epoch": 1.2626073963391857, "step": 1690}, {"loss": 1.7061, "grad_norm": 0.41359591484069824, "learning_rate": 0.0002, "epoch": 1.270078446021666, "step": 1700}, {"loss": 1.7906, "grad_norm": 0.3530874252319336, "learning_rate": 0.0002, "epoch": 1.2775494957041464, "step": 1710}, {"loss": 1.7357, "grad_norm": 0.3702719211578369, "learning_rate": 0.0002, "epoch": 1.2850205453866268, "step": 1720}, {"loss": 1.766, "grad_norm": 0.3703329563140869, "learning_rate": 0.0002, "epoch": 1.2924915950691072, "step": 1730}, {"loss": 1.7221, "grad_norm": 0.37919729948043823, "learning_rate": 0.0002, "epoch": 1.2999626447515875, "step": 1740}, {"loss": 1.7859, "grad_norm": 0.32526856660842896, "learning_rate": 0.0002, "epoch": 1.307433694434068, "step": 1750}, {"loss": 1.7117, "grad_norm": 0.36752620339393616, "learning_rate": 0.0002, "epoch": 1.3149047441165485, "step": 1760}, {"loss": 1.7335, "grad_norm": 0.3398192524909973, "learning_rate": 0.0002, "epoch": 1.3223757937990288, "step": 1770}, {"loss": 1.7492, "grad_norm": 0.37435585260391235, "learning_rate": 0.0002, "epoch": 1.3298468434815092, "step": 1780}, {"loss": 1.7393, "grad_norm": 0.35793280601501465, "learning_rate": 0.0002, "epoch": 1.3373178931639895, "step": 1790}, {"loss": 1.7266, "grad_norm": 0.35481882095336914, "learning_rate": 0.0002, "epoch": 1.3447889428464699, "step": 1800}, {"loss": 1.7456, "grad_norm": 0.3786393105983734, "learning_rate": 0.0002, "epoch": 1.3522599925289502, "step": 1810}, {"loss": 1.7169, "grad_norm": 0.33245593309402466, "learning_rate": 0.0002, "epoch": 1.3597310422114308, "step": 1820}, {"loss": 1.7577, "grad_norm": 0.35388344526290894, "learning_rate": 0.0002, "epoch": 1.3672020918939112, "step": 1830}, {"loss": 1.6968, "grad_norm": 0.3695325553417206, "learning_rate": 0.0002, "epoch": 1.3746731415763915, "step": 1840}, {"loss": 1.7086, "grad_norm": 0.3683604598045349, "learning_rate": 0.0002, "epoch": 1.382144191258872, "step": 1850}, {"loss": 1.7878, "grad_norm": 0.3753012418746948, "learning_rate": 0.0002, "epoch": 1.3896152409413522, "step": 1860}, {"loss": 1.6969, "grad_norm": 0.3331069350242615, "learning_rate": 0.0002, "epoch": 1.3970862906238326, "step": 1870}, {"loss": 1.6644, "grad_norm": 0.3877500295639038, "learning_rate": 0.0002, "epoch": 1.404557340306313, "step": 1880}, {"loss": 1.7586, "grad_norm": 0.33525151014328003, "learning_rate": 0.0002, "epoch": 1.4120283899887935, "step": 1890}, {"loss": 1.7031, "grad_norm": 0.3697299659252167, "learning_rate": 0.0002, "epoch": 1.4194994396712737, "step": 1900}, {"loss": 1.6956, "grad_norm": 0.4029286205768585, "learning_rate": 0.0002, "epoch": 1.4269704893537543, "step": 1910}, {"loss": 1.6897, "grad_norm": 0.3596203029155731, "learning_rate": 0.0002, "epoch": 1.4344415390362346, "step": 1920}, {"loss": 1.7139, "grad_norm": 0.450783908367157, "learning_rate": 0.0002, "epoch": 1.441912588718715, "step": 1930}, {"loss": 1.7243, "grad_norm": 0.3651481866836548, "learning_rate": 0.0002, "epoch": 1.4493836384011953, "step": 1940}, {"loss": 1.6637, "grad_norm": 0.3608424663543701, "learning_rate": 0.0002, "epoch": 1.4568546880836757, "step": 1950}, {"loss": 1.8285, "grad_norm": 0.39684420824050903, "learning_rate": 0.0002, "epoch": 1.4643257377661563, "step": 1960}, {"loss": 1.7514, "grad_norm": 0.34618663787841797, "learning_rate": 0.0002, "epoch": 1.4717967874486364, "step": 1970}, {"loss": 1.6655, "grad_norm": 0.4150386452674866, "learning_rate": 0.0002, "epoch": 1.479267837131117, "step": 1980}, {"loss": 1.7021, "grad_norm": 0.35500776767730713, "learning_rate": 0.0002, "epoch": 1.4867388868135973, "step": 1990}, {"loss": 1.7322, "grad_norm": 0.344144344329834, "learning_rate": 0.0002, "epoch": 1.4942099364960777, "step": 2000}, {"loss": 1.6998, "grad_norm": 0.3340149223804474, "learning_rate": 0.0002, "epoch": 1.501680986178558, "step": 2010}, {"loss": 1.7508, "grad_norm": 0.37685006856918335, "learning_rate": 0.0002, "epoch": 1.5091520358610384, "step": 2020}, {"loss": 1.8299, "grad_norm": 0.3699876368045807, "learning_rate": 0.0002, "epoch": 1.516623085543519, "step": 2030}, {"loss": 1.7357, "grad_norm": 0.3370307385921478, "learning_rate": 0.0002, "epoch": 1.5240941352259991, "step": 2040}, {"loss": 1.8044, "grad_norm": 0.37780630588531494, "learning_rate": 0.0002, "epoch": 1.5315651849084797, "step": 2050}, {"loss": 1.7408, "grad_norm": 0.370259165763855, "learning_rate": 0.0002, "epoch": 1.53903623459096, "step": 2060}, {"loss": 1.7398, "grad_norm": 0.3440011441707611, "learning_rate": 0.0002, "epoch": 1.5465072842734404, "step": 2070}, {"loss": 1.7105, "grad_norm": 0.40382063388824463, "learning_rate": 0.0002, "epoch": 1.5539783339559208, "step": 2080}, {"loss": 1.7071, "grad_norm": 0.38002029061317444, "learning_rate": 0.0002, "epoch": 1.5614493836384011, "step": 2090}, {"loss": 1.6815, "grad_norm": 0.3658451437950134, "learning_rate": 0.0002, "epoch": 1.5689204333208817, "step": 2100}, {"loss": 1.7598, "grad_norm": 0.354842871427536, "learning_rate": 0.0002, "epoch": 1.5763914830033618, "step": 2110}, {"loss": 1.6898, "grad_norm": 0.34735530614852905, "learning_rate": 0.0002, "epoch": 1.5838625326858424, "step": 2120}, {"loss": 1.7363, "grad_norm": 0.377581924200058, "learning_rate": 0.0002, "epoch": 1.5913335823683228, "step": 2130}, {"loss": 1.7789, "grad_norm": 0.41254034638404846, "learning_rate": 0.0002, "epoch": 1.5988046320508031, "step": 2140}, {"loss": 1.6782, "grad_norm": 0.3630715310573578, "learning_rate": 0.0002, "epoch": 1.6062756817332835, "step": 2150}, {"loss": 1.7531, "grad_norm": 0.36980143189430237, "learning_rate": 0.0002, "epoch": 1.6137467314157639, "step": 2160}, {"loss": 1.6847, "grad_norm": 0.3634769320487976, "learning_rate": 0.0002, "epoch": 1.6212177810982444, "step": 2170}, {"loss": 1.6367, "grad_norm": 0.3794139623641968, "learning_rate": 0.0002, "epoch": 1.6286888307807246, "step": 2180}, {"loss": 1.7064, "grad_norm": 0.359742134809494, "learning_rate": 0.0002, "epoch": 1.6361598804632052, "step": 2190}, {"loss": 1.7027, "grad_norm": 0.3770543932914734, "learning_rate": 0.0002, "epoch": 1.6436309301456855, "step": 2200}, {"loss": 1.784, "grad_norm": 0.3797036409378052, "learning_rate": 0.0002, "epoch": 1.6511019798281659, "step": 2210}, {"loss": 1.7875, "grad_norm": 0.35622093081474304, "learning_rate": 0.0002, "epoch": 1.6585730295106462, "step": 2220}, {"loss": 1.6615, "grad_norm": 0.34552520513534546, "learning_rate": 0.0002, "epoch": 1.6660440791931266, "step": 2230}, {"loss": 1.7522, "grad_norm": 0.379926860332489, "learning_rate": 0.0002, "epoch": 1.6735151288756072, "step": 2240}, {"loss": 1.7953, "grad_norm": 0.37083810567855835, "learning_rate": 0.0002, "epoch": 1.6809861785580873, "step": 2250}, {"loss": 1.7485, "grad_norm": 0.42746543884277344, "learning_rate": 0.0002, "epoch": 1.6884572282405679, "step": 2260}, {"loss": 1.776, "grad_norm": 0.3372884690761566, "learning_rate": 0.0002, "epoch": 1.6959282779230482, "step": 2270}, {"loss": 1.7604, "grad_norm": 0.35220256447792053, "learning_rate": 0.0002, "epoch": 1.7033993276055286, "step": 2280}, {"loss": 1.7154, "grad_norm": 0.3659130930900574, "learning_rate": 0.0002, "epoch": 1.710870377288009, "step": 2290}, {"loss": 1.6953, "grad_norm": 0.37629297375679016, "learning_rate": 0.0002, "epoch": 1.7183414269704893, "step": 2300}, {"loss": 1.7212, "grad_norm": 0.36312398314476013, "learning_rate": 0.0002, "epoch": 1.7258124766529699, "step": 2310}, {"loss": 1.7903, "grad_norm": 0.467709481716156, "learning_rate": 0.0002, "epoch": 1.73328352633545, "step": 2320}, {"loss": 1.696, "grad_norm": 0.38685527443885803, "learning_rate": 0.0002, "epoch": 1.7407545760179306, "step": 2330}, {"loss": 1.7041, "grad_norm": 0.3578338325023651, "learning_rate": 0.0002, "epoch": 1.748225625700411, "step": 2340}, {"loss": 1.6456, "grad_norm": 0.36057502031326294, "learning_rate": 0.0002, "epoch": 1.7556966753828913, "step": 2350}, {"loss": 1.6853, "grad_norm": 0.3615196645259857, "learning_rate": 0.0002, "epoch": 1.7631677250653717, "step": 2360}, {"loss": 1.7612, "grad_norm": 0.4118947684764862, "learning_rate": 0.0002, "epoch": 1.770638774747852, "step": 2370}, {"loss": 1.6946, "grad_norm": 0.4067276120185852, "learning_rate": 0.0002, "epoch": 1.7781098244303326, "step": 2380}, {"loss": 1.712, "grad_norm": 0.3979823887348175, "learning_rate": 0.0002, "epoch": 1.7855808741128127, "step": 2390}, {"loss": 1.7644, "grad_norm": 0.44045883417129517, "learning_rate": 0.0002, "epoch": 1.7930519237952933, "step": 2400}, {"loss": 1.7251, "grad_norm": 0.3998069167137146, "learning_rate": 0.0002, "epoch": 1.8005229734777737, "step": 2410}, {"loss": 1.7354, "grad_norm": 0.3450094759464264, "learning_rate": 0.0002, "epoch": 1.807994023160254, "step": 2420}, {"loss": 1.6998, "grad_norm": 0.3759009838104248, "learning_rate": 0.0002, "epoch": 1.8154650728427344, "step": 2430}, {"loss": 1.7706, "grad_norm": 0.34347015619277954, "learning_rate": 0.0002, "epoch": 1.8229361225252148, "step": 2440}, {"loss": 1.7345, "grad_norm": 0.3511228859424591, "learning_rate": 0.0002, "epoch": 1.8304071722076953, "step": 2450}, {"loss": 1.6909, "grad_norm": 0.36853715777397156, "learning_rate": 0.0002, "epoch": 1.8378782218901755, "step": 2460}, {"loss": 1.6931, "grad_norm": 0.40659376978874207, "learning_rate": 0.0002, "epoch": 1.845349271572656, "step": 2470}, {"loss": 1.7626, "grad_norm": 0.39621320366859436, "learning_rate": 0.0002, "epoch": 1.8528203212551362, "step": 2480}, {"loss": 1.7427, "grad_norm": 0.3753979504108429, "learning_rate": 0.0002, "epoch": 1.8602913709376168, "step": 2490}, {"loss": 1.6622, "grad_norm": 0.3811938464641571, "learning_rate": 0.0002, "epoch": 1.8677624206200971, "step": 2500}, {"loss": 1.7718, "grad_norm": 0.3432596027851105, "learning_rate": 0.0002, "epoch": 1.8752334703025775, "step": 2510}, {"loss": 1.7488, "grad_norm": 0.3670712113380432, "learning_rate": 0.0002, "epoch": 1.882704519985058, "step": 2520}, {"loss": 1.705, "grad_norm": 0.40907177329063416, "learning_rate": 0.0002, "epoch": 1.8901755696675382, "step": 2530}, {"loss": 1.7148, "grad_norm": 0.3821999728679657, "learning_rate": 0.0002, "epoch": 1.8976466193500188, "step": 2540}, {"loss": 1.7934, "grad_norm": 0.36173978447914124, "learning_rate": 0.0002, "epoch": 1.905117669032499, "step": 2550}, {"loss": 1.6939, "grad_norm": 0.38990336656570435, "learning_rate": 0.0002, "epoch": 1.9125887187149795, "step": 2560}, {"loss": 1.6893, "grad_norm": 0.35242322087287903, "learning_rate": 0.0002, "epoch": 1.9200597683974598, "step": 2570}, {"loss": 1.7268, "grad_norm": 0.3506428003311157, "learning_rate": 0.0002, "epoch": 1.9275308180799402, "step": 2580}, {"loss": 1.6953, "grad_norm": 0.39540135860443115, "learning_rate": 0.0002, "epoch": 1.9350018677624208, "step": 2590}, {"loss": 1.6511, "grad_norm": 0.3444725573062897, "learning_rate": 0.0002, "epoch": 1.942472917444901, "step": 2600}, {"loss": 1.7259, "grad_norm": 0.3963521718978882, "learning_rate": 0.0002, "epoch": 1.9499439671273815, "step": 2610}, {"loss": 1.6946, "grad_norm": 0.3689815402030945, "learning_rate": 0.0002, "epoch": 1.9574150168098616, "step": 2620}, {"loss": 1.7384, "grad_norm": 0.3482626676559448, "learning_rate": 0.0002, "epoch": 1.9648860664923422, "step": 2630}, {"loss": 1.7048, "grad_norm": 0.35832616686820984, "learning_rate": 0.0002, "epoch": 1.9723571161748226, "step": 2640}, {"loss": 1.6681, "grad_norm": 0.4776208996772766, "learning_rate": 0.0002, "epoch": 1.979828165857303, "step": 2650}, {"loss": 1.6696, "grad_norm": 0.32570165395736694, "learning_rate": 0.0002, "epoch": 1.9872992155397835, "step": 2660}, {"loss": 1.7232, "grad_norm": 0.3380725085735321, "learning_rate": 0.0002, "epoch": 1.9947702652222636, "step": 2670}, {"eval_loss": 1.8046749830245972, "eval_runtime": 38.5096, "eval_samples_per_second": 13.373, "eval_steps_per_second": 1.688, "epoch": 2.0, "step": 2677}, {"loss": 1.7265, "grad_norm": 0.36817631125450134, "learning_rate": 0.0002, "epoch": 2.002241314904744, "step": 2680}, {"loss": 1.548, "grad_norm": 0.4056456685066223, "learning_rate": 0.0002, "epoch": 2.0097123645872244, "step": 2690}, {"loss": 1.5515, "grad_norm": 0.37416863441467285, "learning_rate": 0.0002, "epoch": 2.017183414269705, "step": 2700}, {"loss": 1.5895, "grad_norm": 0.4273638427257538, "learning_rate": 0.0002, "epoch": 2.024654463952185, "step": 2710}, {"loss": 1.5884, "grad_norm": 0.36497923731803894, "learning_rate": 0.0002, "epoch": 2.0321255136346656, "step": 2720}, {"loss": 1.6999, "grad_norm": 0.5021994113922119, "learning_rate": 0.0002, "epoch": 2.0395965633171462, "step": 2730}, {"loss": 1.6655, "grad_norm": 0.45896220207214355, "learning_rate": 0.0002, "epoch": 2.0470676129996264, "step": 2740}, {"loss": 1.6305, "grad_norm": 0.3973815143108368, "learning_rate": 0.0002, "epoch": 2.054538662682107, "step": 2750}, {"loss": 1.6301, "grad_norm": 0.4521815776824951, "learning_rate": 0.0002, "epoch": 2.062009712364587, "step": 2760}, {"loss": 1.6189, "grad_norm": 0.42775002121925354, "learning_rate": 0.0002, "epoch": 2.0694807620470677, "step": 2770}, {"loss": 1.6491, "grad_norm": 0.48158586025238037, "learning_rate": 0.0002, "epoch": 2.076951811729548, "step": 2780}, {"loss": 1.6301, "grad_norm": 0.4612371623516083, "learning_rate": 0.0002, "epoch": 2.0844228614120284, "step": 2790}, {"loss": 1.6327, "grad_norm": 0.42536866664886475, "learning_rate": 0.0002, "epoch": 2.091893911094509, "step": 2800}, {"loss": 1.651, "grad_norm": 0.48515772819519043, "learning_rate": 0.0002, "epoch": 2.099364960776989, "step": 2810}, {"loss": 1.6829, "grad_norm": 0.41418662667274475, "learning_rate": 0.0002, "epoch": 2.1068360104594697, "step": 2820}, {"loss": 1.6266, "grad_norm": 0.4683697819709778, "learning_rate": 0.0002, "epoch": 2.11430706014195, "step": 2830}, {"loss": 1.6586, "grad_norm": 0.4484657049179077, "learning_rate": 0.0002, "epoch": 2.1217781098244304, "step": 2840}, {"loss": 1.6483, "grad_norm": 0.6621400713920593, "learning_rate": 0.0002, "epoch": 2.1292491595069105, "step": 2850}, {"loss": 1.5755, "grad_norm": 0.45074811577796936, "learning_rate": 0.0002, "epoch": 2.136720209189391, "step": 2860}, {"loss": 1.6456, "grad_norm": 0.3513113558292389, "learning_rate": 0.0002, "epoch": 2.1441912588718717, "step": 2870}, {"loss": 1.6081, "grad_norm": 0.40411314368247986, "learning_rate": 0.0002, "epoch": 2.151662308554352, "step": 2880}, {"loss": 1.6323, "grad_norm": 0.4121065139770508, "learning_rate": 0.0002, "epoch": 2.1591333582368324, "step": 2890}, {"loss": 1.6324, "grad_norm": 0.44723689556121826, "learning_rate": 0.0002, "epoch": 2.1666044079193125, "step": 2900}, {"loss": 1.5699, "grad_norm": 0.4226122498512268, "learning_rate": 0.0002, "epoch": 2.174075457601793, "step": 2910}, {"loss": 1.5652, "grad_norm": 0.46617650985717773, "learning_rate": 0.0002, "epoch": 2.1815465072842732, "step": 2920}, {"loss": 1.6378, "grad_norm": 0.4506422281265259, "learning_rate": 0.0002, "epoch": 2.189017556966754, "step": 2930}, {"loss": 1.6112, "grad_norm": 0.4892672896385193, "learning_rate": 0.0002, "epoch": 2.1964886066492344, "step": 2940}, {"loss": 1.6176, "grad_norm": 0.44095516204833984, "learning_rate": 0.0002, "epoch": 2.2039596563317145, "step": 2950}, {"loss": 1.6058, "grad_norm": 0.41522109508514404, "learning_rate": 0.0002, "epoch": 2.211430706014195, "step": 2960}, {"loss": 1.5964, "grad_norm": 0.4860858917236328, "learning_rate": 0.0002, "epoch": 2.2189017556966752, "step": 2970}, {"loss": 1.6427, "grad_norm": 0.42662516236305237, "learning_rate": 0.0002, "epoch": 2.226372805379156, "step": 2980}, {"loss": 1.6313, "grad_norm": 0.4390648305416107, "learning_rate": 0.0002, "epoch": 2.233843855061636, "step": 2990}, {"loss": 1.5992, "grad_norm": 0.47515565156936646, "learning_rate": 0.0002, "epoch": 2.2413149047441165, "step": 3000}, {"loss": 1.5563, "grad_norm": 0.4104543924331665, "learning_rate": 0.0002, "epoch": 2.248785954426597, "step": 3010}, {"loss": 1.6895, "grad_norm": 0.4404028654098511, "learning_rate": 0.0002, "epoch": 2.2562570041090773, "step": 3020}, {"loss": 1.6088, "grad_norm": 0.4717366695404053, "learning_rate": 0.0002, "epoch": 2.263728053791558, "step": 3030}, {"loss": 1.7287, "grad_norm": 0.48345857858657837, "learning_rate": 0.0002, "epoch": 2.271199103474038, "step": 3040}, {"loss": 1.681, "grad_norm": 0.5312452912330627, "learning_rate": 0.0002, "epoch": 2.2786701531565186, "step": 3050}, {"loss": 1.5901, "grad_norm": 0.5073099732398987, "learning_rate": 0.0002, "epoch": 2.2861412028389987, "step": 3060}, {"loss": 1.6914, "grad_norm": 0.5027463436126709, "learning_rate": 0.0002, "epoch": 2.2936122525214793, "step": 3070}, {"loss": 1.5862, "grad_norm": 0.5436304807662964, "learning_rate": 0.0002, "epoch": 2.30108330220396, "step": 3080}, {"loss": 1.5763, "grad_norm": 0.4701065123081207, "learning_rate": 0.0002, "epoch": 2.30855435188644, "step": 3090}, {"loss": 1.6177, "grad_norm": 0.46988746523857117, "learning_rate": 0.0002, "epoch": 2.3160254015689206, "step": 3100}, {"loss": 1.6502, "grad_norm": 0.45112869143486023, "learning_rate": 0.0002, "epoch": 2.3234964512514007, "step": 3110}, {"loss": 1.6291, "grad_norm": 0.5173566937446594, "learning_rate": 0.0002, "epoch": 2.3309675009338813, "step": 3120}, {"loss": 1.6743, "grad_norm": 0.40345850586891174, "learning_rate": 0.0002, "epoch": 2.3384385506163614, "step": 3130}, {"loss": 1.621, "grad_norm": 0.4218924939632416, "learning_rate": 0.0002, "epoch": 2.345909600298842, "step": 3140}, {"loss": 1.6341, "grad_norm": 0.41857317090034485, "learning_rate": 0.0002, "epoch": 2.3533806499813226, "step": 3150}, {"loss": 1.6087, "grad_norm": 0.4197218418121338, "learning_rate": 0.0002, "epoch": 2.3608516996638027, "step": 3160}, {"loss": 1.6572, "grad_norm": 0.4260677397251129, "learning_rate": 0.0002, "epoch": 2.3683227493462833, "step": 3170}, {"loss": 1.6376, "grad_norm": 0.4209042191505432, "learning_rate": 0.0002, "epoch": 2.3757937990287634, "step": 3180}, {"loss": 1.634, "grad_norm": 0.4092234969139099, "learning_rate": 0.0002, "epoch": 2.383264848711244, "step": 3190}, {"loss": 1.6339, "grad_norm": 0.4928431510925293, "learning_rate": 0.0002, "epoch": 2.390735898393724, "step": 3200}, {"loss": 1.6015, "grad_norm": 0.49252402782440186, "learning_rate": 0.0002, "epoch": 2.3982069480762047, "step": 3210}, {"loss": 1.5773, "grad_norm": 0.4368397295475006, "learning_rate": 0.0002, "epoch": 2.4056779977586853, "step": 3220}, {"loss": 1.6629, "grad_norm": 0.46122390031814575, "learning_rate": 0.0002, "epoch": 2.4131490474411654, "step": 3230}, {"loss": 1.6224, "grad_norm": 0.4272301197052002, "learning_rate": 0.0002, "epoch": 2.420620097123646, "step": 3240}, {"loss": 1.5961, "grad_norm": 0.41480937600135803, "learning_rate": 0.0002, "epoch": 2.428091146806126, "step": 3250}, {"loss": 1.6281, "grad_norm": 0.48911941051483154, "learning_rate": 0.0002, "epoch": 2.4355621964886067, "step": 3260}, {"loss": 1.6846, "grad_norm": 0.4444098472595215, "learning_rate": 0.0002, "epoch": 2.443033246171087, "step": 3270}, {"loss": 1.6961, "grad_norm": 0.5111684799194336, "learning_rate": 0.0002, "epoch": 2.4505042958535674, "step": 3280}, {"loss": 1.6152, "grad_norm": 0.5058825016021729, "learning_rate": 0.0002, "epoch": 2.457975345536048, "step": 3290}, {"loss": 1.625, "grad_norm": 0.44173210859298706, "learning_rate": 0.0002, "epoch": 2.465446395218528, "step": 3300}, {"loss": 1.6491, "grad_norm": 0.4659745991230011, "learning_rate": 0.0002, "epoch": 2.4729174449010087, "step": 3310}, {"loss": 1.6114, "grad_norm": 0.47237497568130493, "learning_rate": 0.0002, "epoch": 2.480388494583489, "step": 3320}, {"loss": 1.6193, "grad_norm": 0.47303131222724915, "learning_rate": 0.0002, "epoch": 2.4878595442659694, "step": 3330}, {"loss": 1.7256, "grad_norm": 0.4522389769554138, "learning_rate": 0.0002, "epoch": 2.4953305939484496, "step": 3340}, {"loss": 1.6834, "grad_norm": 0.4467332363128662, "learning_rate": 0.0002, "epoch": 2.50280164363093, "step": 3350}, {"loss": 1.6108, "grad_norm": 0.4413762092590332, "learning_rate": 0.0002, "epoch": 2.5102726933134107, "step": 3360}, {"loss": 1.537, "grad_norm": 0.495514452457428, "learning_rate": 0.0002, "epoch": 2.517743742995891, "step": 3370}, {"loss": 1.5839, "grad_norm": 0.4429773986339569, "learning_rate": 0.0002, "epoch": 2.5252147926783715, "step": 3380}, {"loss": 1.6522, "grad_norm": 0.4589079022407532, "learning_rate": 0.0002, "epoch": 2.5326858423608516, "step": 3390}, {"loss": 1.6529, "grad_norm": 0.4683997333049774, "learning_rate": 0.0002, "epoch": 2.540156892043332, "step": 3400}, {"loss": 1.6745, "grad_norm": 0.4651731252670288, "learning_rate": 0.0002, "epoch": 2.5476279417258123, "step": 3410}, {"loss": 1.5918, "grad_norm": 0.45818084478378296, "learning_rate": 0.0002, "epoch": 2.555098991408293, "step": 3420}, {"loss": 1.6326, "grad_norm": 0.45209529995918274, "learning_rate": 0.0002, "epoch": 2.5625700410907735, "step": 3430}, {"loss": 1.5606, "grad_norm": 0.4344733655452728, "learning_rate": 0.0002, "epoch": 2.5700410907732536, "step": 3440}, {"loss": 1.6748, "grad_norm": 0.47435566782951355, "learning_rate": 0.0002, "epoch": 2.577512140455734, "step": 3450}, {"loss": 1.6237, "grad_norm": 0.43841999769210815, "learning_rate": 0.0002, "epoch": 2.5849831901382143, "step": 3460}, {"loss": 1.7207, "grad_norm": 0.4323869049549103, "learning_rate": 0.0002, "epoch": 2.592454239820695, "step": 3470}, {"loss": 1.5494, "grad_norm": 0.44355881214141846, "learning_rate": 0.0002, "epoch": 2.599925289503175, "step": 3480}, {"loss": 1.665, "grad_norm": 0.45847779512405396, "learning_rate": 0.0002, "epoch": 2.6073963391856556, "step": 3490}, {"loss": 1.6006, "grad_norm": 0.4411061704158783, "learning_rate": 0.0002, "epoch": 2.614867388868136, "step": 3500}, {"loss": 1.5868, "grad_norm": 0.4446796178817749, "learning_rate": 0.0002, "epoch": 2.6223384385506163, "step": 3510}, {"loss": 1.5946, "grad_norm": 0.41969653964042664, "learning_rate": 0.0002, "epoch": 2.629809488233097, "step": 3520}, {"loss": 1.6798, "grad_norm": 0.5263747572898865, "learning_rate": 0.0002, "epoch": 2.637280537915577, "step": 3530}, {"loss": 1.6309, "grad_norm": 0.47719451785087585, "learning_rate": 0.0002, "epoch": 2.6447515875980576, "step": 3540}, {"loss": 1.7024, "grad_norm": 0.46574118733406067, "learning_rate": 0.0002, "epoch": 2.6522226372805378, "step": 3550}, {"loss": 1.618, "grad_norm": 0.46867135167121887, "learning_rate": 0.0002, "epoch": 2.6596936869630183, "step": 3560}, {"loss": 1.5885, "grad_norm": 0.4441198706626892, "learning_rate": 0.0002, "epoch": 2.667164736645499, "step": 3570}, {"loss": 1.6426, "grad_norm": 0.4871319830417633, "learning_rate": 0.0002, "epoch": 2.674635786327979, "step": 3580}, {"loss": 1.6575, "grad_norm": 0.43900373578071594, "learning_rate": 0.0002, "epoch": 2.6821068360104596, "step": 3590}, {"loss": 1.6071, "grad_norm": 0.42509549856185913, "learning_rate": 0.0002, "epoch": 2.6895778856929398, "step": 3600}, {"loss": 1.5651, "grad_norm": 0.4691086709499359, "learning_rate": 0.0002, "epoch": 2.6970489353754203, "step": 3610}, {"loss": 1.5491, "grad_norm": 0.46318942308425903, "learning_rate": 0.0002, "epoch": 2.7045199850579005, "step": 3620}, {"loss": 1.5422, "grad_norm": 0.44631096720695496, "learning_rate": 0.0002, "epoch": 2.711991034740381, "step": 3630}, {"loss": 1.6831, "grad_norm": 0.42315489053726196, "learning_rate": 0.0002, "epoch": 2.7194620844228616, "step": 3640}, {"loss": 1.6008, "grad_norm": 0.4971241056919098, "learning_rate": 0.0002, "epoch": 2.7269331341053418, "step": 3650}, {"loss": 1.6042, "grad_norm": 0.4578486382961273, "learning_rate": 0.0002, "epoch": 2.7344041837878224, "step": 3660}, {"loss": 1.6076, "grad_norm": 0.46584776043891907, "learning_rate": 0.0002, "epoch": 2.7418752334703025, "step": 3670}, {"loss": 1.6809, "grad_norm": 0.4951731264591217, "learning_rate": 0.0002, "epoch": 2.749346283152783, "step": 3680}, {"loss": 1.6226, "grad_norm": 0.4935225546360016, "learning_rate": 0.0002, "epoch": 2.756817332835263, "step": 3690}, {"loss": 1.5878, "grad_norm": 0.41805586218833923, "learning_rate": 0.0002, "epoch": 2.764288382517744, "step": 3700}, {"loss": 1.7173, "grad_norm": 0.4417555630207062, "learning_rate": 0.0002, "epoch": 2.7717594322002244, "step": 3710}, {"loss": 1.6398, "grad_norm": 0.48229655623435974, "learning_rate": 0.0002, "epoch": 2.7792304818827045, "step": 3720}, {"loss": 1.6074, "grad_norm": 0.48562315106391907, "learning_rate": 0.0002, "epoch": 2.786701531565185, "step": 3730}, {"loss": 1.607, "grad_norm": 0.4473940432071686, "learning_rate": 0.0002, "epoch": 2.794172581247665, "step": 3740}, {"loss": 1.6065, "grad_norm": 0.4626813232898712, "learning_rate": 0.0002, "epoch": 2.801643630930146, "step": 3750}, {"loss": 1.6296, "grad_norm": 0.4339792728424072, "learning_rate": 0.0002, "epoch": 2.809114680612626, "step": 3760}, {"loss": 1.6815, "grad_norm": 0.5250858068466187, "learning_rate": 0.0002, "epoch": 2.8165857302951065, "step": 3770}, {"loss": 1.6644, "grad_norm": 0.4537523090839386, "learning_rate": 0.0002, "epoch": 2.824056779977587, "step": 3780}, {"loss": 1.6535, "grad_norm": 0.5646113157272339, "learning_rate": 0.0002, "epoch": 2.831527829660067, "step": 3790}, {"loss": 1.5712, "grad_norm": 0.44243332743644714, "learning_rate": 0.0002, "epoch": 2.8389988793425474, "step": 3800}, {"loss": 1.6478, "grad_norm": 0.4585791826248169, "learning_rate": 0.0002, "epoch": 2.846469929025028, "step": 3810}, {"loss": 1.6854, "grad_norm": 0.489702045917511, "learning_rate": 0.0002, "epoch": 2.8539409787075085, "step": 3820}, {"loss": 1.7066, "grad_norm": 0.502470850944519, "learning_rate": 0.0002, "epoch": 2.8614120283899886, "step": 3830}, {"loss": 1.5785, "grad_norm": 0.4395960867404938, "learning_rate": 0.0002, "epoch": 2.8688830780724692, "step": 3840}, {"loss": 1.6434, "grad_norm": 0.4348670244216919, "learning_rate": 0.0002, "epoch": 2.87635412775495, "step": 3850}, {"loss": 1.6163, "grad_norm": 0.48852720856666565, "learning_rate": 0.0002, "epoch": 2.88382517743743, "step": 3860}, {"loss": 1.5916, "grad_norm": 0.45317450165748596, "learning_rate": 0.0002, "epoch": 2.89129622711991, "step": 3870}, {"loss": 1.6486, "grad_norm": 0.4732758700847626, "learning_rate": 0.0002, "epoch": 2.8987672768023907, "step": 3880}, {"loss": 1.6758, "grad_norm": 0.45238012075424194, "learning_rate": 0.0002, "epoch": 2.9062383264848712, "step": 3890}, {"loss": 1.6228, "grad_norm": 0.48838064074516296, "learning_rate": 0.0002, "epoch": 2.9137093761673514, "step": 3900}, {"loss": 1.658, "grad_norm": 0.43496349453926086, "learning_rate": 0.0002, "epoch": 2.921180425849832, "step": 3910}, {"loss": 1.7063, "grad_norm": 0.47963935136795044, "learning_rate": 0.0002, "epoch": 2.9286514755323125, "step": 3920}, {"loss": 1.6553, "grad_norm": 0.4544987976551056, "learning_rate": 0.0002, "epoch": 2.9361225252147927, "step": 3930}, {"loss": 1.6192, "grad_norm": 0.4622892141342163, "learning_rate": 0.0002, "epoch": 2.943593574897273, "step": 3940}, {"loss": 1.6178, "grad_norm": 0.47026222944259644, "learning_rate": 0.0002, "epoch": 2.9510646245797534, "step": 3950}, {"loss": 1.6612, "grad_norm": 0.4549552798271179, "learning_rate": 0.0002, "epoch": 2.958535674262234, "step": 3960}, {"loss": 1.6458, "grad_norm": 0.46647515892982483, "learning_rate": 0.0002, "epoch": 2.966006723944714, "step": 3970}, {"loss": 1.6051, "grad_norm": 0.45095112919807434, "learning_rate": 0.0002, "epoch": 2.9734777736271947, "step": 3980}, {"loss": 1.6471, "grad_norm": 0.4690017104148865, "learning_rate": 0.0002, "epoch": 2.9809488233096753, "step": 3990}, {"loss": 1.6061, "grad_norm": 0.4603444039821625, "learning_rate": 0.0002, "epoch": 2.9884198729921554, "step": 4000}, {"loss": 1.6431, "grad_norm": 0.4743294417858124, "learning_rate": 0.0002, "epoch": 2.9958909226746355, "step": 4010}, {"eval_loss": 1.8252571821212769, "eval_runtime": 38.7853, "eval_samples_per_second": 13.278, "eval_steps_per_second": 1.676, "epoch": 2.999626447515876, "step": 4015}, {"loss": 1.6512, "grad_norm": 0.4919724464416504, "learning_rate": 0.0002, "epoch": 3.003361972357116, "step": 4020}, {"loss": 1.5354, "grad_norm": 0.4747185707092285, "learning_rate": 0.0002, "epoch": 3.0108330220395967, "step": 4030}, {"loss": 1.568, "grad_norm": 0.4797595143318176, "learning_rate": 0.0002, "epoch": 3.018304071722077, "step": 4040}, {"loss": 1.5194, "grad_norm": 0.5450999140739441, "learning_rate": 0.0002, "epoch": 3.0257751214045574, "step": 4050}, {"loss": 1.5065, "grad_norm": 0.49058812856674194, "learning_rate": 0.0002, "epoch": 3.0332461710870375, "step": 4060}, {"loss": 1.4884, "grad_norm": 0.5219563841819763, "learning_rate": 0.0002, "epoch": 3.040717220769518, "step": 4070}, {"loss": 1.4742, "grad_norm": 0.515628457069397, "learning_rate": 0.0002, "epoch": 3.0481882704519987, "step": 4080}, {"loss": 1.5313, "grad_norm": 0.6145984530448914, "learning_rate": 0.0002, "epoch": 3.055659320134479, "step": 4090}, {"loss": 1.4989, "grad_norm": 0.6067144274711609, "learning_rate": 0.0002, "epoch": 3.0631303698169594, "step": 4100}, {"loss": 1.528, "grad_norm": 0.5773133039474487, "learning_rate": 0.0002, "epoch": 3.0706014194994395, "step": 4110}, {"loss": 1.5374, "grad_norm": 0.6894241571426392, "learning_rate": 0.0002, "epoch": 3.07807246918192, "step": 4120}, {"loss": 1.5422, "grad_norm": 0.6422514915466309, "learning_rate": 0.0002, "epoch": 3.0855435188644003, "step": 4130}, {"loss": 1.4724, "grad_norm": 0.6119855046272278, "learning_rate": 0.0002, "epoch": 3.093014568546881, "step": 4140}, {"loss": 1.5361, "grad_norm": 0.5847280025482178, "learning_rate": 0.0002, "epoch": 3.1004856182293614, "step": 4150}, {"loss": 1.5151, "grad_norm": 0.5401515960693359, "learning_rate": 0.0002, "epoch": 3.1079566679118416, "step": 4160}, {"loss": 1.502, "grad_norm": 0.6501587629318237, "learning_rate": 0.0002, "epoch": 3.115427717594322, "step": 4170}, {"loss": 1.4952, "grad_norm": 0.5988039374351501, "learning_rate": 0.0002, "epoch": 3.1228987672768023, "step": 4180}, {"loss": 1.5287, "grad_norm": 0.4982665181159973, "learning_rate": 0.0002, "epoch": 3.130369816959283, "step": 4190}, {"loss": 1.5078, "grad_norm": 0.5548039078712463, "learning_rate": 0.0002, "epoch": 3.137840866641763, "step": 4200}, {"loss": 1.4904, "grad_norm": 0.5920777320861816, "learning_rate": 0.0002, "epoch": 3.1453119163242436, "step": 4210}, {"loss": 1.442, "grad_norm": 0.6965190172195435, "learning_rate": 0.0002, "epoch": 3.152782966006724, "step": 4220}, {"loss": 1.557, "grad_norm": 0.5196244716644287, "learning_rate": 0.0002, "epoch": 3.1602540156892043, "step": 4230}, {"loss": 1.5706, "grad_norm": 0.6942682266235352, "learning_rate": 0.0002, "epoch": 3.167725065371685, "step": 4240}, {"loss": 1.5407, "grad_norm": 0.5765156149864197, "learning_rate": 0.0002, "epoch": 3.175196115054165, "step": 4250}, {"loss": 1.4963, "grad_norm": 0.5801976919174194, "learning_rate": 0.0002, "epoch": 3.1826671647366456, "step": 4260}, {"loss": 1.4988, "grad_norm": 0.6260752081871033, "learning_rate": 0.0002, "epoch": 3.1901382144191257, "step": 4270}, {"loss": 1.5074, "grad_norm": 0.6610770225524902, "learning_rate": 0.0002, "epoch": 3.1976092641016063, "step": 4280}, {"loss": 1.4657, "grad_norm": 0.5762143135070801, "learning_rate": 0.0002, "epoch": 3.205080313784087, "step": 4290}, {"loss": 1.5181, "grad_norm": 0.5926990509033203, "learning_rate": 0.0002, "epoch": 3.212551363466567, "step": 4300}, {"loss": 1.5492, "grad_norm": 0.7373854517936707, "learning_rate": 0.0002, "epoch": 3.2200224131490476, "step": 4310}, {"loss": 1.4648, "grad_norm": 0.5963311195373535, "learning_rate": 0.0002, "epoch": 3.2274934628315277, "step": 4320}, {"loss": 1.5262, "grad_norm": 0.5754616856575012, "learning_rate": 0.0002, "epoch": 3.2349645125140083, "step": 4330}, {"loss": 1.4767, "grad_norm": 0.6116095781326294, "learning_rate": 0.0002, "epoch": 3.2424355621964884, "step": 4340}, {"loss": 1.5008, "grad_norm": 0.6001536846160889, "learning_rate": 0.0002, "epoch": 3.249906611878969, "step": 4350}, {"loss": 1.5738, "grad_norm": 0.5270227789878845, "learning_rate": 0.0002, "epoch": 3.257377661561449, "step": 4360}, {"loss": 1.5235, "grad_norm": 0.6666602492332458, "learning_rate": 0.0002, "epoch": 3.2648487112439297, "step": 4370}, {"loss": 1.5665, "grad_norm": 0.520310640335083, "learning_rate": 0.0002, "epoch": 3.2723197609264103, "step": 4380}, {"loss": 1.542, "grad_norm": 0.5165975093841553, "learning_rate": 0.0002, "epoch": 3.2797908106088904, "step": 4390}, {"loss": 1.4746, "grad_norm": 0.6080228686332703, "learning_rate": 0.0002, "epoch": 3.287261860291371, "step": 4400}, {"loss": 1.4901, "grad_norm": 0.670122504234314, "learning_rate": 0.0002, "epoch": 3.294732909973851, "step": 4410}, {"loss": 1.4677, "grad_norm": 0.6019457578659058, "learning_rate": 0.0002, "epoch": 3.3022039596563317, "step": 4420}, {"loss": 1.4249, "grad_norm": 0.5519300103187561, "learning_rate": 0.0002, "epoch": 3.309675009338812, "step": 4430}, {"loss": 1.555, "grad_norm": 0.5958521962165833, "learning_rate": 0.0002, "epoch": 3.3171460590212924, "step": 4440}, {"loss": 1.5067, "grad_norm": 0.5552705526351929, "learning_rate": 0.0002, "epoch": 3.324617108703773, "step": 4450}, {"loss": 1.5926, "grad_norm": 0.6583784818649292, "learning_rate": 0.0002, "epoch": 3.332088158386253, "step": 4460}, {"loss": 1.4206, "grad_norm": 0.5815939903259277, "learning_rate": 0.0002, "epoch": 3.3395592080687337, "step": 4470}, {"loss": 1.5942, "grad_norm": 1.3342205286026, "learning_rate": 0.0002, "epoch": 3.347030257751214, "step": 4480}, {"loss": 1.484, "grad_norm": 0.6341500878334045, "learning_rate": 0.0002, "epoch": 3.3545013074336945, "step": 4490}, {"loss": 1.5219, "grad_norm": 0.6384079456329346, "learning_rate": 0.0002, "epoch": 3.3619723571161746, "step": 4500}, {"loss": 1.5222, "grad_norm": 0.6098346710205078, "learning_rate": 0.0002, "epoch": 3.369443406798655, "step": 4510}, {"loss": 1.5475, "grad_norm": 0.5958296656608582, "learning_rate": 0.0002, "epoch": 3.3769144564811358, "step": 4520}, {"loss": 1.5171, "grad_norm": 0.6157881617546082, "learning_rate": 0.0002, "epoch": 3.384385506163616, "step": 4530}, {"loss": 1.569, "grad_norm": 0.5671007037162781, "learning_rate": 0.0002, "epoch": 3.3918565558460965, "step": 4540}, {"loss": 1.604, "grad_norm": 0.6203294992446899, "learning_rate": 0.0002, "epoch": 3.3993276055285766, "step": 4550}, {"loss": 1.5364, "grad_norm": 0.6743317246437073, "learning_rate": 0.0002, "epoch": 3.406798655211057, "step": 4560}, {"loss": 1.5034, "grad_norm": 0.731765627861023, "learning_rate": 0.0002, "epoch": 3.4142697048935373, "step": 4570}, {"loss": 1.4585, "grad_norm": 0.6285187602043152, "learning_rate": 0.0002, "epoch": 3.421740754576018, "step": 4580}, {"loss": 1.5296, "grad_norm": 0.612680196762085, "learning_rate": 0.0002, "epoch": 3.4292118042584985, "step": 4590}, {"loss": 1.5577, "grad_norm": 0.6413681507110596, "learning_rate": 0.0002, "epoch": 3.4366828539409786, "step": 4600}, {"loss": 1.5026, "grad_norm": 0.6240990161895752, "learning_rate": 0.0002, "epoch": 3.444153903623459, "step": 4610}, {"loss": 1.5887, "grad_norm": 0.5095735192298889, "learning_rate": 0.0002, "epoch": 3.4516249533059393, "step": 4620}, {"loss": 1.4906, "grad_norm": 0.5699611902236938, "learning_rate": 0.0002, "epoch": 3.45909600298842, "step": 4630}, {"loss": 1.5176, "grad_norm": 0.7289775609970093, "learning_rate": 0.0002, "epoch": 3.4665670526709, "step": 4640}, {"loss": 1.5467, "grad_norm": 0.6211609840393066, "learning_rate": 0.0002, "epoch": 3.4740381023533806, "step": 4650}, {"loss": 1.533, "grad_norm": 0.5714802145957947, "learning_rate": 0.0002, "epoch": 3.481509152035861, "step": 4660}, {"loss": 1.5096, "grad_norm": 0.6287049651145935, "learning_rate": 0.0002, "epoch": 3.4889802017183413, "step": 4670}, {"loss": 1.4212, "grad_norm": 0.5480595827102661, "learning_rate": 0.0002, "epoch": 3.496451251400822, "step": 4680}, {"loss": 1.4746, "grad_norm": 0.5683253407478333, "learning_rate": 0.0002, "epoch": 3.503922301083302, "step": 4690}, {"loss": 1.5012, "grad_norm": 0.601140558719635, "learning_rate": 0.0002, "epoch": 3.5113933507657826, "step": 4700}, {"loss": 1.5383, "grad_norm": 0.5344498157501221, "learning_rate": 0.0002, "epoch": 3.5188644004482628, "step": 4710}, {"loss": 1.5428, "grad_norm": 0.5739690661430359, "learning_rate": 0.0002, "epoch": 3.5263354501307433, "step": 4720}, {"loss": 1.5589, "grad_norm": 0.5640085935592651, "learning_rate": 0.0002, "epoch": 3.533806499813224, "step": 4730}, {"loss": 1.487, "grad_norm": 0.5967805981636047, "learning_rate": 0.0002, "epoch": 3.541277549495704, "step": 4740}, {"loss": 1.5461, "grad_norm": 0.6138835549354553, "learning_rate": 0.0002, "epoch": 3.5487485991781846, "step": 4750}, {"loss": 1.5502, "grad_norm": 0.6779900193214417, "learning_rate": 0.0002, "epoch": 3.5562196488606648, "step": 4760}, {"loss": 1.4917, "grad_norm": 0.6122010350227356, "learning_rate": 0.0002, "epoch": 3.5636906985431454, "step": 4770}, {"loss": 1.5405, "grad_norm": 0.5685241222381592, "learning_rate": 0.0002, "epoch": 3.5711617482256255, "step": 4780}, {"loss": 1.5427, "grad_norm": 0.604583203792572, "learning_rate": 0.0002, "epoch": 3.578632797908106, "step": 4790}, {"loss": 1.4514, "grad_norm": 0.651165246963501, "learning_rate": 0.0002, "epoch": 3.5861038475905866, "step": 4800}, {"loss": 1.4109, "grad_norm": 0.6398511528968811, "learning_rate": 0.0002, "epoch": 3.593574897273067, "step": 4810}, {"loss": 1.4261, "grad_norm": 0.6444641351699829, "learning_rate": 0.0002, "epoch": 3.6010459469555474, "step": 4820}, {"loss": 1.5274, "grad_norm": 0.6018481850624084, "learning_rate": 0.0002, "epoch": 3.6085169966380275, "step": 4830}, {"loss": 1.4647, "grad_norm": 0.6025291085243225, "learning_rate": 0.0002, "epoch": 3.615988046320508, "step": 4840}, {"loss": 1.5609, "grad_norm": 0.6810156106948853, "learning_rate": 0.0002, "epoch": 3.623459096002988, "step": 4850}, {"loss": 1.5299, "grad_norm": 0.6408044695854187, "learning_rate": 0.0002, "epoch": 3.630930145685469, "step": 4860}, {"loss": 1.5366, "grad_norm": 0.5608272552490234, "learning_rate": 0.0002, "epoch": 3.6384011953679494, "step": 4870}, {"loss": 1.5188, "grad_norm": 0.6136814951896667, "learning_rate": 0.0002, "epoch": 3.6458722450504295, "step": 4880}, {"loss": 1.5021, "grad_norm": 0.5927900075912476, "learning_rate": 0.0002, "epoch": 3.65334329473291, "step": 4890}, {"loss": 1.6084, "grad_norm": 0.5336901545524597, "learning_rate": 0.0002, "epoch": 3.66081434441539, "step": 4900}, {"loss": 1.5701, "grad_norm": 0.7823320627212524, "learning_rate": 0.0002, "epoch": 3.668285394097871, "step": 4910}, {"loss": 1.4881, "grad_norm": 0.6703504323959351, "learning_rate": 0.0002, "epoch": 3.675756443780351, "step": 4920}, {"loss": 1.5332, "grad_norm": 0.6061160564422607, "learning_rate": 0.0002, "epoch": 3.6832274934628315, "step": 4930}, {"loss": 1.5405, "grad_norm": 0.6237227916717529, "learning_rate": 0.0002, "epoch": 3.690698543145312, "step": 4940}, {"loss": 1.497, "grad_norm": 0.5985278487205505, "learning_rate": 0.0002, "epoch": 3.6981695928277922, "step": 4950}, {"loss": 1.5132, "grad_norm": 0.6483839750289917, "learning_rate": 0.0002, "epoch": 3.705640642510273, "step": 4960}, {"loss": 1.5338, "grad_norm": 0.5788805484771729, "learning_rate": 0.0002, "epoch": 3.713111692192753, "step": 4970}, {"loss": 1.5258, "grad_norm": 0.5609974265098572, "learning_rate": 0.0002, "epoch": 3.7205827418752335, "step": 4980}, {"loss": 1.4759, "grad_norm": 0.5681300759315491, "learning_rate": 0.0002, "epoch": 3.7280537915577137, "step": 4990}, {"loss": 1.6018, "grad_norm": 0.5860186219215393, "learning_rate": 0.0002, "epoch": 3.7355248412401942, "step": 5000}, {"loss": 1.58, "grad_norm": 0.5718157291412354, "learning_rate": 0.0002, "epoch": 3.742995890922675, "step": 5010}, {"loss": 1.5834, "grad_norm": 0.6173721551895142, "learning_rate": 0.0002, "epoch": 3.750466940605155, "step": 5020}, {"loss": 1.5617, "grad_norm": 0.629152238368988, "learning_rate": 0.0002, "epoch": 3.7579379902876355, "step": 5030}, {"loss": 1.519, "grad_norm": 0.5666284561157227, "learning_rate": 0.0002, "epoch": 3.7654090399701157, "step": 5040}, {"loss": 1.5329, "grad_norm": 0.6053005456924438, "learning_rate": 0.0002, "epoch": 3.7728800896525962, "step": 5050}, {"loss": 1.5404, "grad_norm": 0.5870583057403564, "learning_rate": 0.0002, "epoch": 3.7803511393350764, "step": 5060}, {"loss": 1.4444, "grad_norm": 0.5422009229660034, "learning_rate": 0.0002, "epoch": 3.787822189017557, "step": 5070}, {"loss": 1.5308, "grad_norm": 0.5396918058395386, "learning_rate": 0.0002, "epoch": 3.7952932387000375, "step": 5080}, {"loss": 1.464, "grad_norm": 0.5544713139533997, "learning_rate": 0.0002, "epoch": 3.8027642883825177, "step": 5090}, {"loss": 1.4752, "grad_norm": 0.5983749628067017, "learning_rate": 0.0002, "epoch": 3.8102353380649983, "step": 5100}, {"loss": 1.4972, "grad_norm": 0.5702024102210999, "learning_rate": 0.0002, "epoch": 3.8177063877474784, "step": 5110}, {"loss": 1.5471, "grad_norm": 0.5436882376670837, "learning_rate": 0.0002, "epoch": 3.825177437429959, "step": 5120}, {"loss": 1.5118, "grad_norm": 0.5453617572784424, "learning_rate": 0.0002, "epoch": 3.832648487112439, "step": 5130}, {"loss": 1.5732, "grad_norm": 0.6269069314002991, "learning_rate": 0.0002, "epoch": 3.8401195367949197, "step": 5140}, {"loss": 1.4959, "grad_norm": 0.6189185380935669, "learning_rate": 0.0002, "epoch": 3.8475905864774003, "step": 5150}, {"loss": 1.4999, "grad_norm": 0.6653388142585754, "learning_rate": 0.0002, "epoch": 3.8550616361598804, "step": 5160}, {"loss": 1.5075, "grad_norm": 0.5771768689155579, "learning_rate": 0.0002, "epoch": 3.862532685842361, "step": 5170}, {"loss": 1.5545, "grad_norm": 0.6052790880203247, "learning_rate": 0.0002, "epoch": 3.870003735524841, "step": 5180}, {"loss": 1.4987, "grad_norm": 0.6572316884994507, "learning_rate": 0.0002, "epoch": 3.8774747852073217, "step": 5190}, {"loss": 1.5241, "grad_norm": 0.670576810836792, "learning_rate": 0.0002, "epoch": 3.884945834889802, "step": 5200}, {"loss": 1.4777, "grad_norm": 0.5728798508644104, "learning_rate": 0.0002, "epoch": 3.8924168845722824, "step": 5210}, {"loss": 1.5351, "grad_norm": 0.6340774297714233, "learning_rate": 0.0002, "epoch": 3.899887934254763, "step": 5220}, {"loss": 1.5081, "grad_norm": 0.5981315970420837, "learning_rate": 0.0002, "epoch": 3.907358983937243, "step": 5230}, {"loss": 1.4875, "grad_norm": 0.6212025880813599, "learning_rate": 0.0002, "epoch": 3.9148300336197237, "step": 5240}, {"loss": 1.5545, "grad_norm": 0.6202296018600464, "learning_rate": 0.0002, "epoch": 3.922301083302204, "step": 5250}, {"loss": 1.5765, "grad_norm": 0.6159142255783081, "learning_rate": 0.0002, "epoch": 3.9297721329846844, "step": 5260}, {"loss": 1.4938, "grad_norm": 0.6519438624382019, "learning_rate": 0.0002, "epoch": 3.9372431826671646, "step": 5270}, {"loss": 1.4859, "grad_norm": 0.539813756942749, "learning_rate": 0.0002, "epoch": 3.944714232349645, "step": 5280}, {"loss": 1.5921, "grad_norm": 0.6443665027618408, "learning_rate": 0.0002, "epoch": 3.9521852820321257, "step": 5290}, {"loss": 1.5153, "grad_norm": 0.6635757684707642, "learning_rate": 0.0002, "epoch": 3.959656331714606, "step": 5300}, {"loss": 1.5485, "grad_norm": 0.589363157749176, "learning_rate": 0.0002, "epoch": 3.9671273813970864, "step": 5310}, {"loss": 1.5498, "grad_norm": 0.5788735747337341, "learning_rate": 0.0002, "epoch": 3.9745984310795666, "step": 5320}, {"loss": 1.5607, "grad_norm": 0.5976864695549011, "learning_rate": 0.0002, "epoch": 3.982069480762047, "step": 5330}, {"loss": 1.5302, "grad_norm": 0.6624067425727844, "learning_rate": 0.0002, "epoch": 3.9895405304445273, "step": 5340}, {"loss": 1.5904, "grad_norm": 0.6738956570625305, "learning_rate": 0.0002, "epoch": 3.997011580127008, "step": 5350}, {"eval_loss": 1.868006944656372, "eval_runtime": 38.5153, "eval_samples_per_second": 13.371, "eval_steps_per_second": 1.688, "epoch": 4.0, "step": 5354}, {"loss": 1.4535, "grad_norm": 0.6023468971252441, "learning_rate": 0.0002, "epoch": 4.004482629809488, "step": 5360}, {"loss": 1.3987, "grad_norm": 0.8589285612106323, "learning_rate": 0.0002, "epoch": 4.011953679491969, "step": 5370}, {"loss": 1.3952, "grad_norm": 0.7477491497993469, "learning_rate": 0.0002, "epoch": 4.019424729174449, "step": 5380}, {"loss": 1.3745, "grad_norm": 0.7601922154426575, "learning_rate": 0.0002, "epoch": 4.02689577885693, "step": 5390}, {"loss": 1.4133, "grad_norm": 0.8115614056587219, "learning_rate": 0.0002, "epoch": 4.03436682853941, "step": 5400}, {"loss": 1.3748, "grad_norm": 0.669925332069397, "learning_rate": 0.0002, "epoch": 4.04183787822189, "step": 5410}, {"loss": 1.2835, "grad_norm": 0.8091904520988464, "learning_rate": 0.0002, "epoch": 4.04930892790437, "step": 5420}, {"loss": 1.3615, "grad_norm": 0.709405779838562, "learning_rate": 0.0002, "epoch": 4.056779977586851, "step": 5430}, {"loss": 1.3558, "grad_norm": 1.0006179809570312, "learning_rate": 0.0002, "epoch": 4.064251027269331, "step": 5440}, {"loss": 1.3491, "grad_norm": 0.7017965912818909, "learning_rate": 0.0002, "epoch": 4.071722076951811, "step": 5450}, {"loss": 1.3642, "grad_norm": 0.8991572260856628, "learning_rate": 0.0002, "epoch": 4.0791931266342925, "step": 5460}, {"loss": 1.392, "grad_norm": 0.9064797759056091, "learning_rate": 0.0002, "epoch": 4.086664176316773, "step": 5470}, {"loss": 1.3425, "grad_norm": 0.7981749176979065, "learning_rate": 0.0002, "epoch": 4.094135225999253, "step": 5480}, {"loss": 1.3826, "grad_norm": 0.7280883193016052, "learning_rate": 0.0002, "epoch": 4.101606275681733, "step": 5490}, {"loss": 1.3275, "grad_norm": 0.7419600486755371, "learning_rate": 0.0002, "epoch": 4.109077325364214, "step": 5500}, {"loss": 1.3199, "grad_norm": 0.8019949197769165, "learning_rate": 0.0002, "epoch": 4.116548375046694, "step": 5510}, {"loss": 1.3133, "grad_norm": 0.7501229047775269, "learning_rate": 0.0002, "epoch": 4.124019424729174, "step": 5520}, {"loss": 1.4432, "grad_norm": 0.8166249990463257, "learning_rate": 0.0002, "epoch": 4.131490474411655, "step": 5530}, {"loss": 1.3901, "grad_norm": 0.9728496074676514, "learning_rate": 0.0002, "epoch": 4.138961524094135, "step": 5540}, {"loss": 1.3538, "grad_norm": 0.7590922117233276, "learning_rate": 0.0002, "epoch": 4.1464325737766154, "step": 5550}, {"loss": 1.4368, "grad_norm": 0.7759010791778564, "learning_rate": 0.0002, "epoch": 4.153903623459096, "step": 5560}, {"loss": 1.3635, "grad_norm": 0.9057986736297607, "learning_rate": 0.0002, "epoch": 4.161374673141577, "step": 5570}, {"loss": 1.4152, "grad_norm": 0.8853937983512878, "learning_rate": 0.0002, "epoch": 4.168845722824057, "step": 5580}, {"loss": 1.3633, "grad_norm": 0.7070684432983398, "learning_rate": 0.0002, "epoch": 4.176316772506537, "step": 5590}, {"loss": 1.3218, "grad_norm": 0.7649410963058472, "learning_rate": 0.0002, "epoch": 4.183787822189018, "step": 5600}, {"loss": 1.3857, "grad_norm": 1.2048029899597168, "learning_rate": 0.0002, "epoch": 4.191258871871498, "step": 5610}, {"loss": 1.3629, "grad_norm": 0.7986605763435364, "learning_rate": 0.0002, "epoch": 4.198729921553978, "step": 5620}, {"loss": 1.3995, "grad_norm": 0.8151885867118835, "learning_rate": 0.0002, "epoch": 4.206200971236458, "step": 5630}, {"loss": 1.3782, "grad_norm": 0.7719064354896545, "learning_rate": 0.0002, "epoch": 4.213672020918939, "step": 5640}, {"loss": 1.3852, "grad_norm": 0.8422448039054871, "learning_rate": 0.0002, "epoch": 4.2211430706014195, "step": 5650}, {"loss": 1.3321, "grad_norm": 0.7017164826393127, "learning_rate": 0.0002, "epoch": 4.2286141202839, "step": 5660}, {"loss": 1.4105, "grad_norm": 0.8559677600860596, "learning_rate": 0.0002, "epoch": 4.236085169966381, "step": 5670}, {"loss": 1.3701, "grad_norm": 0.8216157555580139, "learning_rate": 0.0002, "epoch": 4.243556219648861, "step": 5680}, {"loss": 1.3565, "grad_norm": 0.7681755423545837, "learning_rate": 0.0002, "epoch": 4.251027269331341, "step": 5690}, {"loss": 1.3806, "grad_norm": 0.811665952205658, "learning_rate": 0.0002, "epoch": 4.258498319013821, "step": 5700}, {"loss": 1.4161, "grad_norm": 0.7242204546928406, "learning_rate": 0.0002, "epoch": 4.265969368696302, "step": 5710}, {"loss": 1.2958, "grad_norm": 0.7570181488990784, "learning_rate": 0.0002, "epoch": 4.273440418378782, "step": 5720}, {"loss": 1.4265, "grad_norm": 0.8951969146728516, "learning_rate": 0.0002, "epoch": 4.280911468061262, "step": 5730}, {"loss": 1.3895, "grad_norm": 0.7222902178764343, "learning_rate": 0.0002, "epoch": 4.288382517743743, "step": 5740}, {"loss": 1.4155, "grad_norm": 0.8508469462394714, "learning_rate": 0.0002, "epoch": 4.2958535674262235, "step": 5750}, {"loss": 1.365, "grad_norm": 0.7215430736541748, "learning_rate": 0.0002, "epoch": 4.303324617108704, "step": 5760}, {"loss": 1.4472, "grad_norm": 0.8774884939193726, "learning_rate": 0.0002, "epoch": 4.310795666791184, "step": 5770}, {"loss": 1.427, "grad_norm": 0.8354552984237671, "learning_rate": 0.0002, "epoch": 4.318266716473665, "step": 5780}, {"loss": 1.3222, "grad_norm": 0.6938814520835876, "learning_rate": 0.0002, "epoch": 4.325737766156145, "step": 5790}, {"loss": 1.3589, "grad_norm": 0.78675377368927, "learning_rate": 0.0002, "epoch": 4.333208815838625, "step": 5800}, {"loss": 1.3662, "grad_norm": 0.7147697806358337, "learning_rate": 0.0002, "epoch": 4.340679865521106, "step": 5810}, {"loss": 1.3597, "grad_norm": 0.7693623304367065, "learning_rate": 0.0002, "epoch": 4.348150915203586, "step": 5820}, {"loss": 1.2944, "grad_norm": 0.856517493724823, "learning_rate": 0.0002, "epoch": 4.355621964886066, "step": 5830}, {"loss": 1.4307, "grad_norm": 0.7200973033905029, "learning_rate": 0.0002, "epoch": 4.3630930145685465, "step": 5840}, {"loss": 1.442, "grad_norm": 0.743281364440918, "learning_rate": 0.0002, "epoch": 4.3705640642510275, "step": 5850}, {"loss": 1.3999, "grad_norm": 0.7627727389335632, "learning_rate": 0.0002, "epoch": 4.378035113933508, "step": 5860}, {"loss": 1.4082, "grad_norm": 0.7238836884498596, "learning_rate": 0.0002, "epoch": 4.385506163615988, "step": 5870}, {"loss": 1.4292, "grad_norm": 0.7253410816192627, "learning_rate": 0.0002, "epoch": 4.392977213298469, "step": 5880}, {"loss": 1.3774, "grad_norm": 0.8232238292694092, "learning_rate": 0.0002, "epoch": 4.400448262980949, "step": 5890}, {"loss": 1.3757, "grad_norm": 0.8778504729270935, "learning_rate": 0.0002, "epoch": 4.407919312663429, "step": 5900}, {"loss": 1.387, "grad_norm": 0.7639474868774414, "learning_rate": 0.0002, "epoch": 4.415390362345909, "step": 5910}, {"loss": 1.3862, "grad_norm": 0.7666519284248352, "learning_rate": 0.0002, "epoch": 4.42286141202839, "step": 5920}, {"loss": 1.4168, "grad_norm": 0.867132842540741, "learning_rate": 0.0002, "epoch": 4.43033246171087, "step": 5930}, {"loss": 1.4772, "grad_norm": 0.7571166753768921, "learning_rate": 0.0002, "epoch": 4.4378035113933505, "step": 5940}, {"loss": 1.4401, "grad_norm": 0.7911370992660522, "learning_rate": 0.0002, "epoch": 4.4452745610758315, "step": 5950}, {"loss": 1.4516, "grad_norm": 0.8844250440597534, "learning_rate": 0.0002, "epoch": 4.452745610758312, "step": 5960}, {"loss": 1.4109, "grad_norm": 0.7336231470108032, "learning_rate": 0.0002, "epoch": 4.460216660440792, "step": 5970}, {"loss": 1.3891, "grad_norm": 0.8162738084793091, "learning_rate": 0.0002, "epoch": 4.467687710123272, "step": 5980}, {"loss": 1.393, "grad_norm": 0.7413017153739929, "learning_rate": 0.0002, "epoch": 4.475158759805753, "step": 5990}, {"loss": 1.3712, "grad_norm": 0.7215432524681091, "learning_rate": 0.0002, "epoch": 4.482629809488233, "step": 6000}, {"loss": 1.3521, "grad_norm": 0.8943389058113098, "learning_rate": 0.0002, "epoch": 4.490100859170713, "step": 6010}, {"loss": 1.4172, "grad_norm": 0.7850823998451233, "learning_rate": 0.0002, "epoch": 4.497571908853194, "step": 6020}, {"loss": 1.3582, "grad_norm": 0.8117504119873047, "learning_rate": 0.0002, "epoch": 4.505042958535674, "step": 6030}, {"loss": 1.4272, "grad_norm": 0.8381605744361877, "learning_rate": 0.0002, "epoch": 4.5125140082181545, "step": 6040}, {"loss": 1.3829, "grad_norm": 0.7964059710502625, "learning_rate": 0.0002, "epoch": 4.519985057900635, "step": 6050}, {"loss": 1.3555, "grad_norm": 0.7935128211975098, "learning_rate": 0.0002, "epoch": 4.527456107583116, "step": 6060}, {"loss": 1.3994, "grad_norm": 0.8725124597549438, "learning_rate": 0.0002, "epoch": 4.534927157265596, "step": 6070}, {"loss": 1.3923, "grad_norm": 0.880325198173523, "learning_rate": 0.0002, "epoch": 4.542398206948076, "step": 6080}, {"loss": 1.4459, "grad_norm": 0.7220637202262878, "learning_rate": 0.0002, "epoch": 4.549869256630557, "step": 6090}, {"loss": 1.3281, "grad_norm": 0.6908547878265381, "learning_rate": 0.0002, "epoch": 4.557340306313037, "step": 6100}, {"loss": 1.437, "grad_norm": 0.797931969165802, "learning_rate": 0.0002, "epoch": 4.564811355995517, "step": 6110}, {"loss": 1.4023, "grad_norm": 0.7056134343147278, "learning_rate": 0.0002, "epoch": 4.572282405677997, "step": 6120}, {"loss": 1.3814, "grad_norm": 0.7850478887557983, "learning_rate": 0.0002, "epoch": 4.579753455360478, "step": 6130}, {"loss": 1.3579, "grad_norm": 0.8112621307373047, "learning_rate": 0.0002, "epoch": 4.5872245050429585, "step": 6140}, {"loss": 1.3523, "grad_norm": 0.7040849328041077, "learning_rate": 0.0002, "epoch": 4.594695554725439, "step": 6150}, {"loss": 1.3526, "grad_norm": 0.7214553952217102, "learning_rate": 0.0002, "epoch": 4.60216660440792, "step": 6160}, {"loss": 1.3932, "grad_norm": 0.8616511821746826, "learning_rate": 0.0002, "epoch": 4.6096376540904, "step": 6170}, {"loss": 1.4622, "grad_norm": 0.8374658226966858, "learning_rate": 0.0002, "epoch": 4.61710870377288, "step": 6180}, {"loss": 1.3703, "grad_norm": 0.6761606931686401, "learning_rate": 0.0002, "epoch": 4.62457975345536, "step": 6190}, {"loss": 1.3977, "grad_norm": 0.768028199672699, "learning_rate": 0.0002, "epoch": 4.632050803137841, "step": 6200}, {"loss": 1.3772, "grad_norm": 0.9372717142105103, "learning_rate": 0.0002, "epoch": 4.639521852820321, "step": 6210}, {"loss": 1.4098, "grad_norm": 0.7906546592712402, "learning_rate": 0.0002, "epoch": 4.646992902502801, "step": 6220}, {"loss": 1.3962, "grad_norm": 0.7376723289489746, "learning_rate": 0.0002, "epoch": 4.654463952185282, "step": 6230}, {"loss": 1.4529, "grad_norm": 0.8972630500793457, "learning_rate": 0.0002, "epoch": 4.6619350018677626, "step": 6240}, {"loss": 1.4668, "grad_norm": 0.8261756300926208, "learning_rate": 0.0002, "epoch": 4.669406051550243, "step": 6250}, {"loss": 1.3267, "grad_norm": 0.7512393593788147, "learning_rate": 0.0002, "epoch": 4.676877101232723, "step": 6260}, {"loss": 1.4278, "grad_norm": 0.7132362127304077, "learning_rate": 0.0002, "epoch": 4.684348150915204, "step": 6270}, {"loss": 1.4299, "grad_norm": 0.7690575122833252, "learning_rate": 0.0002, "epoch": 4.691819200597684, "step": 6280}, {"loss": 1.4769, "grad_norm": 0.9886258840560913, "learning_rate": 0.0002, "epoch": 4.699290250280164, "step": 6290}, {"loss": 1.4005, "grad_norm": 0.9502435922622681, "learning_rate": 0.0002, "epoch": 4.706761299962645, "step": 6300}, {"loss": 1.4319, "grad_norm": 0.702255129814148, "learning_rate": 0.0002, "epoch": 4.714232349645125, "step": 6310}, {"loss": 1.4447, "grad_norm": 0.7713103890419006, "learning_rate": 0.0002, "epoch": 4.721703399327605, "step": 6320}, {"loss": 1.4392, "grad_norm": 0.7778580784797668, "learning_rate": 0.0002, "epoch": 4.7291744490100855, "step": 6330}, {"loss": 1.4169, "grad_norm": 0.7275111079216003, "learning_rate": 0.0002, "epoch": 4.736645498692567, "step": 6340}, {"loss": 1.4429, "grad_norm": 0.7728744149208069, "learning_rate": 0.0002, "epoch": 4.744116548375047, "step": 6350}, {"loss": 1.3756, "grad_norm": 0.9724260568618774, "learning_rate": 0.0002, "epoch": 4.751587598057527, "step": 6360}, {"loss": 1.3358, "grad_norm": 0.7505622506141663, "learning_rate": 0.0002, "epoch": 4.759058647740007, "step": 6370}, {"loss": 1.379, "grad_norm": 0.7994682788848877, "learning_rate": 0.0002, "epoch": 4.766529697422488, "step": 6380}, {"loss": 1.4275, "grad_norm": 0.8432038426399231, "learning_rate": 0.0002, "epoch": 4.774000747104968, "step": 6390}, {"loss": 1.4606, "grad_norm": 0.7436022758483887, "learning_rate": 0.0002, "epoch": 4.781471796787448, "step": 6400}, {"loss": 1.3461, "grad_norm": 0.7709194421768188, "learning_rate": 0.0002, "epoch": 4.788942846469929, "step": 6410}, {"loss": 1.3715, "grad_norm": 0.8798436522483826, "learning_rate": 0.0002, "epoch": 4.796413896152409, "step": 6420}, {"loss": 1.3761, "grad_norm": 0.790189266204834, "learning_rate": 0.0002, "epoch": 4.80388494583489, "step": 6430}, {"loss": 1.4109, "grad_norm": 0.6824303865432739, "learning_rate": 0.0002, "epoch": 4.811355995517371, "step": 6440}, {"loss": 1.3877, "grad_norm": 0.7501044869422913, "learning_rate": 0.0002, "epoch": 4.818827045199851, "step": 6450}, {"loss": 1.4458, "grad_norm": 0.8840398192405701, "learning_rate": 0.0002, "epoch": 4.826298094882331, "step": 6460}, {"loss": 1.4412, "grad_norm": 0.7812688946723938, "learning_rate": 0.0002, "epoch": 4.833769144564811, "step": 6470}, {"loss": 1.4299, "grad_norm": 0.7429926991462708, "learning_rate": 0.0002, "epoch": 4.841240194247292, "step": 6480}, {"loss": 1.5062, "grad_norm": 0.7778021693229675, "learning_rate": 0.0002, "epoch": 4.848711243929772, "step": 6490}, {"loss": 1.4589, "grad_norm": 0.8270702362060547, "learning_rate": 0.0002, "epoch": 4.856182293612252, "step": 6500}, {"loss": 1.4091, "grad_norm": 0.6960513591766357, "learning_rate": 0.0002, "epoch": 4.863653343294732, "step": 6510}, {"loss": 1.376, "grad_norm": 0.7728942632675171, "learning_rate": 0.0002, "epoch": 4.8711243929772134, "step": 6520}, {"loss": 1.4852, "grad_norm": 0.7377303838729858, "learning_rate": 0.0002, "epoch": 4.878595442659694, "step": 6530}, {"loss": 1.3846, "grad_norm": 0.7257253527641296, "learning_rate": 0.0002, "epoch": 4.886066492342174, "step": 6540}, {"loss": 1.4166, "grad_norm": 0.7875821590423584, "learning_rate": 0.0002, "epoch": 4.893537542024655, "step": 6550}, {"loss": 1.357, "grad_norm": 0.8346304297447205, "learning_rate": 0.0002, "epoch": 4.901008591707135, "step": 6560}, {"loss": 1.4522, "grad_norm": 0.7710739374160767, "learning_rate": 0.0002, "epoch": 4.908479641389615, "step": 6570}, {"loss": 1.4465, "grad_norm": 0.7015138268470764, "learning_rate": 0.0002, "epoch": 4.915950691072096, "step": 6580}, {"loss": 1.435, "grad_norm": 0.8707432150840759, "learning_rate": 0.0002, "epoch": 4.923421740754576, "step": 6590}, {"loss": 1.2968, "grad_norm": 0.786601185798645, "learning_rate": 0.0002, "epoch": 4.930892790437056, "step": 6600}, {"loss": 1.4385, "grad_norm": 0.978519082069397, "learning_rate": 0.0002, "epoch": 4.938363840119536, "step": 6610}, {"loss": 1.3997, "grad_norm": 0.8102927207946777, "learning_rate": 0.0002, "epoch": 4.9458348898020175, "step": 6620}, {"loss": 1.4859, "grad_norm": 0.7628704309463501, "learning_rate": 0.0002, "epoch": 4.953305939484498, "step": 6630}, {"loss": 1.3774, "grad_norm": 0.8053455352783203, "learning_rate": 0.0002, "epoch": 4.960776989166978, "step": 6640}, {"loss": 1.5092, "grad_norm": 0.8680412173271179, "learning_rate": 0.0002, "epoch": 4.968248038849458, "step": 6650}, {"loss": 1.3978, "grad_norm": 0.7415758371353149, "learning_rate": 0.0002, "epoch": 4.975719088531939, "step": 6660}, {"loss": 1.3793, "grad_norm": 0.7730312347412109, "learning_rate": 0.0002, "epoch": 4.983190138214419, "step": 6670}, {"loss": 1.4863, "grad_norm": 0.7924041152000427, "learning_rate": 0.0002, "epoch": 4.990661187896899, "step": 6680}, {"loss": 1.4137, "grad_norm": 0.8677893877029419, "learning_rate": 0.0002, "epoch": 4.99813223757938, "step": 6690}]}
+{"epoch": 6.0, "step": 8031, "epoch_duration": 1900.1275906562805, "total_accumulated_duration": 9420.934909820557, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7568.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-42/checkpoint-2677", "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": 2.6181, "grad_norm": 0.4912872612476349, "learning_rate": 0.0002, "epoch": 0.007471049682480389, "step": 10}, {"loss": 2.2606, "grad_norm": 0.4856316149234772, "learning_rate": 0.0002, "epoch": 0.014942099364960777, "step": 20}, {"loss": 2.0957, "grad_norm": 0.47683125734329224, "learning_rate": 0.0002, "epoch": 0.022413149047441166, "step": 30}, {"loss": 1.8908, "grad_norm": 0.515082597732544, "learning_rate": 0.0002, "epoch": 0.029884198729921554, "step": 40}, {"loss": 1.9704, "grad_norm": 0.5299215316772461, "learning_rate": 0.0002, "epoch": 0.03735524841240194, "step": 50}, {"loss": 1.9225, "grad_norm": 0.4951399862766266, "learning_rate": 0.0002, "epoch": 0.04482629809488233, "step": 60}, {"loss": 1.9742, "grad_norm": 0.48079821467399597, "learning_rate": 0.0002, "epoch": 0.05229734777736272, "step": 70}, {"loss": 1.9466, "grad_norm": 0.49402132630348206, "learning_rate": 0.0002, "epoch": 0.05976839745984311, "step": 80}, {"loss": 1.8691, "grad_norm": 0.4778193235397339, "learning_rate": 0.0002, "epoch": 0.0672394471423235, "step": 90}, {"loss": 1.8455, "grad_norm": 0.42472657561302185, "learning_rate": 0.0002, "epoch": 0.07471049682480388, "step": 100}, {"loss": 1.8744, "grad_norm": 0.4433092474937439, "learning_rate": 0.0002, "epoch": 0.08218154650728428, "step": 110}, {"loss": 1.865, "grad_norm": 0.4472862780094147, "learning_rate": 0.0002, "epoch": 0.08965259618976466, "step": 120}, {"loss": 1.9256, "grad_norm": 0.42596298456192017, "learning_rate": 0.0002, "epoch": 0.09712364587224505, "step": 130}, {"loss": 1.8015, "grad_norm": 0.46645811200141907, "learning_rate": 0.0002, "epoch": 0.10459469555472543, "step": 140}, {"loss": 1.8307, "grad_norm": 0.41041234135627747, "learning_rate": 0.0002, "epoch": 0.11206574523720583, "step": 150}, {"loss": 1.8276, "grad_norm": 0.5329819917678833, "learning_rate": 0.0002, "epoch": 0.11953679491968622, "step": 160}, {"loss": 1.8118, "grad_norm": 0.4065922200679779, "learning_rate": 0.0002, "epoch": 0.1270078446021666, "step": 170}, {"loss": 1.8559, "grad_norm": 0.38406994938850403, "learning_rate": 0.0002, "epoch": 0.134478894284647, "step": 180}, {"loss": 1.8647, "grad_norm": 0.4246881306171417, "learning_rate": 0.0002, "epoch": 0.14194994396712737, "step": 190}, {"loss": 1.8054, "grad_norm": 0.35136649012565613, "learning_rate": 0.0002, "epoch": 0.14942099364960776, "step": 200}, {"loss": 1.802, "grad_norm": 0.43252742290496826, "learning_rate": 0.0002, "epoch": 0.15689204333208817, "step": 210}, {"loss": 1.7823, "grad_norm": 0.39236941933631897, "learning_rate": 0.0002, "epoch": 0.16436309301456856, "step": 220}, {"loss": 1.818, "grad_norm": 0.3748249113559723, "learning_rate": 0.0002, "epoch": 0.17183414269704894, "step": 230}, {"loss": 1.866, "grad_norm": 0.6432855725288391, "learning_rate": 0.0002, "epoch": 0.17930519237952933, "step": 240}, {"loss": 1.8397, "grad_norm": 0.34874802827835083, "learning_rate": 0.0002, "epoch": 0.1867762420620097, "step": 250}, {"loss": 1.79, "grad_norm": 0.3721984326839447, "learning_rate": 0.0002, "epoch": 0.1942472917444901, "step": 260}, {"loss": 1.8464, "grad_norm": 0.4339311420917511, "learning_rate": 0.0002, "epoch": 0.20171834142697048, "step": 270}, {"loss": 1.8665, "grad_norm": 0.4018215537071228, "learning_rate": 0.0002, "epoch": 0.20918939110945087, "step": 280}, {"loss": 1.8048, "grad_norm": 0.3278839886188507, "learning_rate": 0.0002, "epoch": 0.21666044079193125, "step": 290}, {"loss": 1.7395, "grad_norm": 0.36146077513694763, "learning_rate": 0.0002, "epoch": 0.22413149047441167, "step": 300}, {"loss": 1.7916, "grad_norm": 0.38175010681152344, "learning_rate": 0.0002, "epoch": 0.23160254015689205, "step": 310}, {"loss": 1.8593, "grad_norm": 0.44776618480682373, "learning_rate": 0.0002, "epoch": 0.23907358983937244, "step": 320}, {"loss": 1.7824, "grad_norm": 0.3933652937412262, "learning_rate": 0.0002, "epoch": 0.24654463952185282, "step": 330}, {"loss": 1.8393, "grad_norm": 0.3515005111694336, "learning_rate": 0.0002, "epoch": 0.2540156892043332, "step": 340}, {"loss": 1.8653, "grad_norm": 0.6683304309844971, "learning_rate": 0.0002, "epoch": 0.2614867388868136, "step": 350}, {"loss": 1.8797, "grad_norm": 0.37093454599380493, "learning_rate": 0.0002, "epoch": 0.268957788569294, "step": 360}, {"loss": 1.8251, "grad_norm": 0.3450651168823242, "learning_rate": 0.0002, "epoch": 0.2764288382517744, "step": 370}, {"loss": 1.7435, "grad_norm": 0.5140917301177979, "learning_rate": 0.0002, "epoch": 0.28389988793425475, "step": 380}, {"loss": 1.8026, "grad_norm": 0.32885563373565674, "learning_rate": 0.0002, "epoch": 0.29137093761673516, "step": 390}, {"loss": 1.8174, "grad_norm": 0.33962297439575195, "learning_rate": 0.0002, "epoch": 0.2988419872992155, "step": 400}, {"loss": 1.7467, "grad_norm": 0.3723141849040985, "learning_rate": 0.0002, "epoch": 0.30631303698169593, "step": 410}, {"loss": 1.8459, "grad_norm": 0.37173134088516235, "learning_rate": 0.0002, "epoch": 0.31378408666417634, "step": 420}, {"loss": 1.8876, "grad_norm": 0.33736956119537354, "learning_rate": 0.0002, "epoch": 0.3212551363466567, "step": 430}, {"loss": 1.8367, "grad_norm": 0.3602448105812073, "learning_rate": 0.0002, "epoch": 0.3287261860291371, "step": 440}, {"loss": 1.8058, "grad_norm": 0.3569699227809906, "learning_rate": 0.0002, "epoch": 0.33619723571161747, "step": 450}, {"loss": 1.8086, "grad_norm": 0.31009167432785034, "learning_rate": 0.0002, "epoch": 0.3436682853940979, "step": 460}, {"loss": 1.8876, "grad_norm": 0.5278693437576294, "learning_rate": 0.0002, "epoch": 0.35113933507657824, "step": 470}, {"loss": 1.8534, "grad_norm": 0.3587537109851837, "learning_rate": 0.0002, "epoch": 0.35861038475905865, "step": 480}, {"loss": 1.8046, "grad_norm": 0.3859670162200928, "learning_rate": 0.0002, "epoch": 0.366081434441539, "step": 490}, {"loss": 1.8287, "grad_norm": 0.395913690328598, "learning_rate": 0.0002, "epoch": 0.3735524841240194, "step": 500}, {"loss": 1.7619, "grad_norm": 0.35052940249443054, "learning_rate": 0.0002, "epoch": 0.38102353380649984, "step": 510}, {"loss": 1.7824, "grad_norm": 0.2979494333267212, "learning_rate": 0.0002, "epoch": 0.3884945834889802, "step": 520}, {"loss": 1.8641, "grad_norm": 0.3062683343887329, "learning_rate": 0.0002, "epoch": 0.3959656331714606, "step": 530}, {"loss": 1.7651, "grad_norm": 0.3172847330570221, "learning_rate": 0.0002, "epoch": 0.40343668285394096, "step": 540}, {"loss": 1.806, "grad_norm": 0.360435426235199, "learning_rate": 0.0002, "epoch": 0.4109077325364214, "step": 550}, {"loss": 1.9054, "grad_norm": 0.3427872359752655, "learning_rate": 0.0002, "epoch": 0.41837878221890173, "step": 560}, {"loss": 1.7562, "grad_norm": 0.34036558866500854, "learning_rate": 0.0002, "epoch": 0.42584983190138215, "step": 570}, {"loss": 1.7254, "grad_norm": 0.3365345299243927, "learning_rate": 0.0002, "epoch": 0.4333208815838625, "step": 580}, {"loss": 1.8328, "grad_norm": 0.35619041323661804, "learning_rate": 0.0002, "epoch": 0.4407919312663429, "step": 590}, {"loss": 1.8114, "grad_norm": 0.3569088280200958, "learning_rate": 0.0002, "epoch": 0.44826298094882333, "step": 600}, {"loss": 1.8599, "grad_norm": 0.3581278622150421, "learning_rate": 0.0002, "epoch": 0.4557340306313037, "step": 610}, {"loss": 1.7078, "grad_norm": 0.43197110295295715, "learning_rate": 0.0002, "epoch": 0.4632050803137841, "step": 620}, {"loss": 1.8257, "grad_norm": 0.33966198563575745, "learning_rate": 0.0002, "epoch": 0.47067612999626446, "step": 630}, {"loss": 1.7528, "grad_norm": 0.3343866467475891, "learning_rate": 0.0002, "epoch": 0.47814717967874487, "step": 640}, {"loss": 1.8191, "grad_norm": 0.33878564834594727, "learning_rate": 0.0002, "epoch": 0.48561822936122523, "step": 650}, {"loss": 1.8801, "grad_norm": 0.387195885181427, "learning_rate": 0.0002, "epoch": 0.49308927904370564, "step": 660}, {"loss": 1.7559, "grad_norm": 0.3755440413951874, "learning_rate": 0.0002, "epoch": 0.500560328726186, "step": 670}, {"loss": 1.8057, "grad_norm": 0.3272816836833954, "learning_rate": 0.0002, "epoch": 0.5080313784086664, "step": 680}, {"loss": 1.8156, "grad_norm": 0.36063864827156067, "learning_rate": 0.0002, "epoch": 0.5155024280911468, "step": 690}, {"loss": 1.8397, "grad_norm": 0.35317373275756836, "learning_rate": 0.0002, "epoch": 0.5229734777736272, "step": 700}, {"loss": 1.7603, "grad_norm": 0.3561195433139801, "learning_rate": 0.0002, "epoch": 0.5304445274561076, "step": 710}, {"loss": 1.8149, "grad_norm": 0.31124624609947205, "learning_rate": 0.0002, "epoch": 0.537915577138588, "step": 720}, {"loss": 1.7434, "grad_norm": 0.3294544517993927, "learning_rate": 0.0002, "epoch": 0.5453866268210683, "step": 730}, {"loss": 1.8027, "grad_norm": 0.31933900713920593, "learning_rate": 0.0002, "epoch": 0.5528576765035488, "step": 740}, {"loss": 1.7601, "grad_norm": 0.3226020634174347, "learning_rate": 0.0002, "epoch": 0.5603287261860291, "step": 750}, {"loss": 1.7862, "grad_norm": 0.3147525489330292, "learning_rate": 0.0002, "epoch": 0.5677997758685095, "step": 760}, {"loss": 1.9028, "grad_norm": 0.32234328985214233, "learning_rate": 0.0002, "epoch": 0.57527082555099, "step": 770}, {"loss": 1.7623, "grad_norm": 0.3258664309978485, "learning_rate": 0.0002, "epoch": 0.5827418752334703, "step": 780}, {"loss": 1.7384, "grad_norm": 0.3166961967945099, "learning_rate": 0.0002, "epoch": 0.5902129249159507, "step": 790}, {"loss": 1.8799, "grad_norm": 0.35621458292007446, "learning_rate": 0.0002, "epoch": 0.597683974598431, "step": 800}, {"loss": 1.8313, "grad_norm": 0.3236999213695526, "learning_rate": 0.0002, "epoch": 0.6051550242809115, "step": 810}, {"loss": 1.7132, "grad_norm": 0.2892923653125763, "learning_rate": 0.0002, "epoch": 0.6126260739633919, "step": 820}, {"loss": 1.8709, "grad_norm": 0.4098321497440338, "learning_rate": 0.0002, "epoch": 0.6200971236458722, "step": 830}, {"loss": 1.7637, "grad_norm": 0.3337118923664093, "learning_rate": 0.0002, "epoch": 0.6275681733283527, "step": 840}, {"loss": 1.7375, "grad_norm": 0.30416029691696167, "learning_rate": 0.0002, "epoch": 0.635039223010833, "step": 850}, {"loss": 1.7419, "grad_norm": 0.3361026346683502, "learning_rate": 0.0002, "epoch": 0.6425102726933134, "step": 860}, {"loss": 1.732, "grad_norm": 0.3537365198135376, "learning_rate": 0.0002, "epoch": 0.6499813223757938, "step": 870}, {"loss": 1.7825, "grad_norm": 0.33854469656944275, "learning_rate": 0.0002, "epoch": 0.6574523720582742, "step": 880}, {"loss": 1.7561, "grad_norm": 0.3332272469997406, "learning_rate": 0.0002, "epoch": 0.6649234217407546, "step": 890}, {"loss": 1.7247, "grad_norm": 0.34954726696014404, "learning_rate": 0.0002, "epoch": 0.6723944714232349, "step": 900}, {"loss": 1.7917, "grad_norm": 0.2921750247478485, "learning_rate": 0.0002, "epoch": 0.6798655211057153, "step": 910}, {"loss": 1.7807, "grad_norm": 0.30508682131767273, "learning_rate": 0.0002, "epoch": 0.6873365707881958, "step": 920}, {"loss": 1.8082, "grad_norm": 0.32268425822257996, "learning_rate": 0.0002, "epoch": 0.6948076204706761, "step": 930}, {"loss": 1.8283, "grad_norm": 0.2844390869140625, "learning_rate": 0.0002, "epoch": 0.7022786701531565, "step": 940}, {"loss": 1.7363, "grad_norm": 0.31263890862464905, "learning_rate": 0.0002, "epoch": 0.709749719835637, "step": 950}, {"loss": 1.8081, "grad_norm": 0.3626808822154999, "learning_rate": 0.0002, "epoch": 0.7172207695181173, "step": 960}, {"loss": 1.853, "grad_norm": 0.3322749733924866, "learning_rate": 0.0002, "epoch": 0.7246918192005977, "step": 970}, {"loss": 1.7912, "grad_norm": 0.29177871346473694, "learning_rate": 0.0002, "epoch": 0.732162868883078, "step": 980}, {"loss": 1.8447, "grad_norm": 0.35405513644218445, "learning_rate": 0.0002, "epoch": 0.7396339185655585, "step": 990}, {"loss": 1.7008, "grad_norm": 0.39318400621414185, "learning_rate": 0.0002, "epoch": 0.7471049682480388, "step": 1000}, {"loss": 1.7803, "grad_norm": 0.29401418566703796, "learning_rate": 0.0002, "epoch": 0.7545760179305192, "step": 1010}, {"loss": 1.7649, "grad_norm": 0.3271748721599579, "learning_rate": 0.0002, "epoch": 0.7620470676129997, "step": 1020}, {"loss": 1.7266, "grad_norm": 0.30883970856666565, "learning_rate": 0.0002, "epoch": 0.76951811729548, "step": 1030}, {"loss": 1.7722, "grad_norm": 0.3411838412284851, "learning_rate": 0.0002, "epoch": 0.7769891669779604, "step": 1040}, {"loss": 1.829, "grad_norm": 0.30608129501342773, "learning_rate": 0.0002, "epoch": 0.7844602166604407, "step": 1050}, {"loss": 1.7815, "grad_norm": 0.30899080634117126, "learning_rate": 0.0002, "epoch": 0.7919312663429212, "step": 1060}, {"loss": 1.7625, "grad_norm": 0.3160453140735626, "learning_rate": 0.0002, "epoch": 0.7994023160254016, "step": 1070}, {"loss": 1.8452, "grad_norm": 0.30947187542915344, "learning_rate": 0.0002, "epoch": 0.8068733657078819, "step": 1080}, {"loss": 1.7418, "grad_norm": 0.3103134036064148, "learning_rate": 0.0002, "epoch": 0.8143444153903624, "step": 1090}, {"loss": 1.842, "grad_norm": 0.31771138310432434, "learning_rate": 0.0002, "epoch": 0.8218154650728428, "step": 1100}, {"loss": 1.7918, "grad_norm": 0.5860997438430786, "learning_rate": 0.0002, "epoch": 0.8292865147553231, "step": 1110}, {"loss": 1.8443, "grad_norm": 0.3230148255825043, "learning_rate": 0.0002, "epoch": 0.8367575644378035, "step": 1120}, {"loss": 1.8478, "grad_norm": 0.29611510038375854, "learning_rate": 0.0002, "epoch": 0.8442286141202839, "step": 1130}, {"loss": 1.7673, "grad_norm": 0.3373654782772064, "learning_rate": 0.0002, "epoch": 0.8516996638027643, "step": 1140}, {"loss": 1.7997, "grad_norm": 0.3474279046058655, "learning_rate": 0.0002, "epoch": 0.8591707134852447, "step": 1150}, {"loss": 1.75, "grad_norm": 0.35057875514030457, "learning_rate": 0.0002, "epoch": 0.866641763167725, "step": 1160}, {"loss": 1.8273, "grad_norm": 0.39537495374679565, "learning_rate": 0.0002, "epoch": 0.8741128128502055, "step": 1170}, {"loss": 1.7682, "grad_norm": 0.3714233636856079, "learning_rate": 0.0002, "epoch": 0.8815838625326858, "step": 1180}, {"loss": 1.7549, "grad_norm": 0.2950296998023987, "learning_rate": 0.0002, "epoch": 0.8890549122151662, "step": 1190}, {"loss": 1.7612, "grad_norm": 0.38182979822158813, "learning_rate": 0.0002, "epoch": 0.8965259618976467, "step": 1200}, {"loss": 1.827, "grad_norm": 0.27883678674697876, "learning_rate": 0.0002, "epoch": 0.903997011580127, "step": 1210}, {"loss": 1.7623, "grad_norm": 0.33874374628067017, "learning_rate": 0.0002, "epoch": 0.9114680612626074, "step": 1220}, {"loss": 1.7334, "grad_norm": 0.3014272153377533, "learning_rate": 0.0002, "epoch": 0.9189391109450877, "step": 1230}, {"loss": 1.8235, "grad_norm": 0.3194271922111511, "learning_rate": 0.0002, "epoch": 0.9264101606275682, "step": 1240}, {"loss": 1.7924, "grad_norm": 0.3049403429031372, "learning_rate": 0.0002, "epoch": 0.9338812103100486, "step": 1250}, {"loss": 1.7535, "grad_norm": 0.30621254444122314, "learning_rate": 0.0002, "epoch": 0.9413522599925289, "step": 1260}, {"loss": 1.8287, "grad_norm": 0.28675132989883423, "learning_rate": 0.0002, "epoch": 0.9488233096750094, "step": 1270}, {"loss": 1.7586, "grad_norm": 0.3322032690048218, "learning_rate": 0.0002, "epoch": 0.9562943593574897, "step": 1280}, {"loss": 1.8054, "grad_norm": 0.35408294200897217, "learning_rate": 0.0002, "epoch": 0.9637654090399701, "step": 1290}, {"loss": 1.7343, "grad_norm": 0.36386919021606445, "learning_rate": 0.0002, "epoch": 0.9712364587224505, "step": 1300}, {"loss": 1.8633, "grad_norm": 0.32338324189186096, "learning_rate": 0.0002, "epoch": 0.9787075084049309, "step": 1310}, {"loss": 1.7724, "grad_norm": 0.3714013993740082, "learning_rate": 0.0002, "epoch": 0.9861785580874113, "step": 1320}, {"loss": 1.7766, "grad_norm": 0.3133082389831543, "learning_rate": 0.0002, "epoch": 0.9936496077698916, "step": 1330}, {"eval_loss": 1.8051470518112183, "eval_runtime": 38.6332, "eval_samples_per_second": 13.331, "eval_steps_per_second": 1.682, "epoch": 0.9996264475158759, "step": 1338}, {"loss": 1.8035, "grad_norm": 0.31595754623413086, "learning_rate": 0.0002, "epoch": 1.001120657452372, "step": 1340}, {"loss": 1.7486, "grad_norm": 0.3095700144767761, "learning_rate": 0.0002, "epoch": 1.0085917071348525, "step": 1350}, {"loss": 1.6981, "grad_norm": 0.34677496552467346, "learning_rate": 0.0002, "epoch": 1.0160627568173328, "step": 1360}, {"loss": 1.7377, "grad_norm": 0.29108840227127075, "learning_rate": 0.0002, "epoch": 1.0235338064998132, "step": 1370}, {"loss": 1.7194, "grad_norm": 0.32356950640678406, "learning_rate": 0.0002, "epoch": 1.0310048561822935, "step": 1380}, {"loss": 1.7593, "grad_norm": 0.4200669229030609, "learning_rate": 0.0002, "epoch": 1.038475905864774, "step": 1390}, {"loss": 1.797, "grad_norm": 0.3283711373806, "learning_rate": 0.0002, "epoch": 1.0459469555472545, "step": 1400}, {"loss": 1.7163, "grad_norm": 0.32898256182670593, "learning_rate": 0.0002, "epoch": 1.0534180052297348, "step": 1410}, {"loss": 1.7559, "grad_norm": 0.38790300488471985, "learning_rate": 0.0002, "epoch": 1.0608890549122152, "step": 1420}, {"loss": 1.6922, "grad_norm": 0.339800089597702, "learning_rate": 0.0002, "epoch": 1.0683601045946955, "step": 1430}, {"loss": 1.7076, "grad_norm": 0.3548751175403595, "learning_rate": 0.0002, "epoch": 1.075831154277176, "step": 1440}, {"loss": 1.6985, "grad_norm": 0.35114359855651855, "learning_rate": 0.0002, "epoch": 1.0833022039596563, "step": 1450}, {"loss": 1.7217, "grad_norm": 0.35226720571517944, "learning_rate": 0.0002, "epoch": 1.0907732536421366, "step": 1460}, {"loss": 1.6822, "grad_norm": 0.33665576577186584, "learning_rate": 0.0002, "epoch": 1.0982443033246172, "step": 1470}, {"loss": 1.6699, "grad_norm": 0.363889217376709, "learning_rate": 0.0002, "epoch": 1.1057153530070976, "step": 1480}, {"loss": 1.7933, "grad_norm": 0.3826201856136322, "learning_rate": 0.0002, "epoch": 1.113186402689578, "step": 1490}, {"loss": 1.7022, "grad_norm": 0.34058740735054016, "learning_rate": 0.0002, "epoch": 1.1206574523720583, "step": 1500}, {"loss": 1.6375, "grad_norm": 0.3462134301662445, "learning_rate": 0.0002, "epoch": 1.1281285020545386, "step": 1510}, {"loss": 1.7147, "grad_norm": 0.3396756052970886, "learning_rate": 0.0002, "epoch": 1.135599551737019, "step": 1520}, {"loss": 1.7219, "grad_norm": 0.32004743814468384, "learning_rate": 0.0002, "epoch": 1.1430706014194993, "step": 1530}, {"loss": 1.743, "grad_norm": 0.3397733271121979, "learning_rate": 0.0002, "epoch": 1.15054165110198, "step": 1540}, {"loss": 1.7333, "grad_norm": 0.3783262073993683, "learning_rate": 0.0002, "epoch": 1.1580127007844603, "step": 1550}, {"loss": 1.6075, "grad_norm": 0.35121291875839233, "learning_rate": 0.0002, "epoch": 1.1654837504669406, "step": 1560}, {"loss": 1.678, "grad_norm": 0.35816895961761475, "learning_rate": 0.0002, "epoch": 1.172954800149421, "step": 1570}, {"loss": 1.7143, "grad_norm": 0.33843839168548584, "learning_rate": 0.0002, "epoch": 1.1804258498319014, "step": 1580}, {"loss": 1.7434, "grad_norm": 0.3371972143650055, "learning_rate": 0.0002, "epoch": 1.1878968995143817, "step": 1590}, {"loss": 1.7671, "grad_norm": 0.36016878485679626, "learning_rate": 0.0002, "epoch": 1.195367949196862, "step": 1600}, {"loss": 1.6914, "grad_norm": 0.40879473090171814, "learning_rate": 0.0002, "epoch": 1.2028389988793426, "step": 1610}, {"loss": 1.6955, "grad_norm": 0.3216715455055237, "learning_rate": 0.0002, "epoch": 1.210310048561823, "step": 1620}, {"loss": 1.632, "grad_norm": 0.4482610821723938, "learning_rate": 0.0002, "epoch": 1.2177810982443034, "step": 1630}, {"loss": 1.6999, "grad_norm": 0.3257700502872467, "learning_rate": 0.0002, "epoch": 1.2252521479267837, "step": 1640}, {"loss": 1.7177, "grad_norm": 0.38646459579467773, "learning_rate": 0.0002, "epoch": 1.232723197609264, "step": 1650}, {"loss": 1.7081, "grad_norm": 0.4081360697746277, "learning_rate": 0.0002, "epoch": 1.2401942472917444, "step": 1660}, {"loss": 1.7519, "grad_norm": 0.4326848089694977, "learning_rate": 0.0002, "epoch": 1.2476652969742248, "step": 1670}, {"loss": 1.6752, "grad_norm": 0.346401572227478, "learning_rate": 0.0002, "epoch": 1.2551363466567054, "step": 1680}, {"loss": 1.7425, "grad_norm": 0.34536251425743103, "learning_rate": 0.0002, "epoch": 1.2626073963391857, "step": 1690}, {"loss": 1.7061, "grad_norm": 0.41359591484069824, "learning_rate": 0.0002, "epoch": 1.270078446021666, "step": 1700}, {"loss": 1.7906, "grad_norm": 0.3530874252319336, "learning_rate": 0.0002, "epoch": 1.2775494957041464, "step": 1710}, {"loss": 1.7357, "grad_norm": 0.3702719211578369, "learning_rate": 0.0002, "epoch": 1.2850205453866268, "step": 1720}, {"loss": 1.766, "grad_norm": 0.3703329563140869, "learning_rate": 0.0002, "epoch": 1.2924915950691072, "step": 1730}, {"loss": 1.7221, "grad_norm": 0.37919729948043823, "learning_rate": 0.0002, "epoch": 1.2999626447515875, "step": 1740}, {"loss": 1.7859, "grad_norm": 0.32526856660842896, "learning_rate": 0.0002, "epoch": 1.307433694434068, "step": 1750}, {"loss": 1.7117, "grad_norm": 0.36752620339393616, "learning_rate": 0.0002, "epoch": 1.3149047441165485, "step": 1760}, {"loss": 1.7335, "grad_norm": 0.3398192524909973, "learning_rate": 0.0002, "epoch": 1.3223757937990288, "step": 1770}, {"loss": 1.7492, "grad_norm": 0.37435585260391235, "learning_rate": 0.0002, "epoch": 1.3298468434815092, "step": 1780}, {"loss": 1.7393, "grad_norm": 0.35793280601501465, "learning_rate": 0.0002, "epoch": 1.3373178931639895, "step": 1790}, {"loss": 1.7266, "grad_norm": 0.35481882095336914, "learning_rate": 0.0002, "epoch": 1.3447889428464699, "step": 1800}, {"loss": 1.7456, "grad_norm": 0.3786393105983734, "learning_rate": 0.0002, "epoch": 1.3522599925289502, "step": 1810}, {"loss": 1.7169, "grad_norm": 0.33245593309402466, "learning_rate": 0.0002, "epoch": 1.3597310422114308, "step": 1820}, {"loss": 1.7577, "grad_norm": 0.35388344526290894, "learning_rate": 0.0002, "epoch": 1.3672020918939112, "step": 1830}, {"loss": 1.6968, "grad_norm": 0.3695325553417206, "learning_rate": 0.0002, "epoch": 1.3746731415763915, "step": 1840}, {"loss": 1.7086, "grad_norm": 0.3683604598045349, "learning_rate": 0.0002, "epoch": 1.382144191258872, "step": 1850}, {"loss": 1.7878, "grad_norm": 0.3753012418746948, "learning_rate": 0.0002, "epoch": 1.3896152409413522, "step": 1860}, {"loss": 1.6969, "grad_norm": 0.3331069350242615, "learning_rate": 0.0002, "epoch": 1.3970862906238326, "step": 1870}, {"loss": 1.6644, "grad_norm": 0.3877500295639038, "learning_rate": 0.0002, "epoch": 1.404557340306313, "step": 1880}, {"loss": 1.7586, "grad_norm": 0.33525151014328003, "learning_rate": 0.0002, "epoch": 1.4120283899887935, "step": 1890}, {"loss": 1.7031, "grad_norm": 0.3697299659252167, "learning_rate": 0.0002, "epoch": 1.4194994396712737, "step": 1900}, {"loss": 1.6956, "grad_norm": 0.4029286205768585, "learning_rate": 0.0002, "epoch": 1.4269704893537543, "step": 1910}, {"loss": 1.6897, "grad_norm": 0.3596203029155731, "learning_rate": 0.0002, "epoch": 1.4344415390362346, "step": 1920}, {"loss": 1.7139, "grad_norm": 0.450783908367157, "learning_rate": 0.0002, "epoch": 1.441912588718715, "step": 1930}, {"loss": 1.7243, "grad_norm": 0.3651481866836548, "learning_rate": 0.0002, "epoch": 1.4493836384011953, "step": 1940}, {"loss": 1.6637, "grad_norm": 0.3608424663543701, "learning_rate": 0.0002, "epoch": 1.4568546880836757, "step": 1950}, {"loss": 1.8285, "grad_norm": 0.39684420824050903, "learning_rate": 0.0002, "epoch": 1.4643257377661563, "step": 1960}, {"loss": 1.7514, "grad_norm": 0.34618663787841797, "learning_rate": 0.0002, "epoch": 1.4717967874486364, "step": 1970}, {"loss": 1.6655, "grad_norm": 0.4150386452674866, "learning_rate": 0.0002, "epoch": 1.479267837131117, "step": 1980}, {"loss": 1.7021, "grad_norm": 0.35500776767730713, "learning_rate": 0.0002, "epoch": 1.4867388868135973, "step": 1990}, {"loss": 1.7322, "grad_norm": 0.344144344329834, "learning_rate": 0.0002, "epoch": 1.4942099364960777, "step": 2000}, {"loss": 1.6998, "grad_norm": 0.3340149223804474, "learning_rate": 0.0002, "epoch": 1.501680986178558, "step": 2010}, {"loss": 1.7508, "grad_norm": 0.37685006856918335, "learning_rate": 0.0002, "epoch": 1.5091520358610384, "step": 2020}, {"loss": 1.8299, "grad_norm": 0.3699876368045807, "learning_rate": 0.0002, "epoch": 1.516623085543519, "step": 2030}, {"loss": 1.7357, "grad_norm": 0.3370307385921478, "learning_rate": 0.0002, "epoch": 1.5240941352259991, "step": 2040}, {"loss": 1.8044, "grad_norm": 0.37780630588531494, "learning_rate": 0.0002, "epoch": 1.5315651849084797, "step": 2050}, {"loss": 1.7408, "grad_norm": 0.370259165763855, "learning_rate": 0.0002, "epoch": 1.53903623459096, "step": 2060}, {"loss": 1.7398, "grad_norm": 0.3440011441707611, "learning_rate": 0.0002, "epoch": 1.5465072842734404, "step": 2070}, {"loss": 1.7105, "grad_norm": 0.40382063388824463, "learning_rate": 0.0002, "epoch": 1.5539783339559208, "step": 2080}, {"loss": 1.7071, "grad_norm": 0.38002029061317444, "learning_rate": 0.0002, "epoch": 1.5614493836384011, "step": 2090}, {"loss": 1.6815, "grad_norm": 0.3658451437950134, "learning_rate": 0.0002, "epoch": 1.5689204333208817, "step": 2100}, {"loss": 1.7598, "grad_norm": 0.354842871427536, "learning_rate": 0.0002, "epoch": 1.5763914830033618, "step": 2110}, {"loss": 1.6898, "grad_norm": 0.34735530614852905, "learning_rate": 0.0002, "epoch": 1.5838625326858424, "step": 2120}, {"loss": 1.7363, "grad_norm": 0.377581924200058, "learning_rate": 0.0002, "epoch": 1.5913335823683228, "step": 2130}, {"loss": 1.7789, "grad_norm": 0.41254034638404846, "learning_rate": 0.0002, "epoch": 1.5988046320508031, "step": 2140}, {"loss": 1.6782, "grad_norm": 0.3630715310573578, "learning_rate": 0.0002, "epoch": 1.6062756817332835, "step": 2150}, {"loss": 1.7531, "grad_norm": 0.36980143189430237, "learning_rate": 0.0002, "epoch": 1.6137467314157639, "step": 2160}, {"loss": 1.6847, "grad_norm": 0.3634769320487976, "learning_rate": 0.0002, "epoch": 1.6212177810982444, "step": 2170}, {"loss": 1.6367, "grad_norm": 0.3794139623641968, "learning_rate": 0.0002, "epoch": 1.6286888307807246, "step": 2180}, {"loss": 1.7064, "grad_norm": 0.359742134809494, "learning_rate": 0.0002, "epoch": 1.6361598804632052, "step": 2190}, {"loss": 1.7027, "grad_norm": 0.3770543932914734, "learning_rate": 0.0002, "epoch": 1.6436309301456855, "step": 2200}, {"loss": 1.784, "grad_norm": 0.3797036409378052, "learning_rate": 0.0002, "epoch": 1.6511019798281659, "step": 2210}, {"loss": 1.7875, "grad_norm": 0.35622093081474304, "learning_rate": 0.0002, "epoch": 1.6585730295106462, "step": 2220}, {"loss": 1.6615, "grad_norm": 0.34552520513534546, "learning_rate": 0.0002, "epoch": 1.6660440791931266, "step": 2230}, {"loss": 1.7522, "grad_norm": 0.379926860332489, "learning_rate": 0.0002, "epoch": 1.6735151288756072, "step": 2240}, {"loss": 1.7953, "grad_norm": 0.37083810567855835, "learning_rate": 0.0002, "epoch": 1.6809861785580873, "step": 2250}, {"loss": 1.7485, "grad_norm": 0.42746543884277344, "learning_rate": 0.0002, "epoch": 1.6884572282405679, "step": 2260}, {"loss": 1.776, "grad_norm": 0.3372884690761566, "learning_rate": 0.0002, "epoch": 1.6959282779230482, "step": 2270}, {"loss": 1.7604, "grad_norm": 0.35220256447792053, "learning_rate": 0.0002, "epoch": 1.7033993276055286, "step": 2280}, {"loss": 1.7154, "grad_norm": 0.3659130930900574, "learning_rate": 0.0002, "epoch": 1.710870377288009, "step": 2290}, {"loss": 1.6953, "grad_norm": 0.37629297375679016, "learning_rate": 0.0002, "epoch": 1.7183414269704893, "step": 2300}, {"loss": 1.7212, "grad_norm": 0.36312398314476013, "learning_rate": 0.0002, "epoch": 1.7258124766529699, "step": 2310}, {"loss": 1.7903, "grad_norm": 0.467709481716156, "learning_rate": 0.0002, "epoch": 1.73328352633545, "step": 2320}, {"loss": 1.696, "grad_norm": 0.38685527443885803, "learning_rate": 0.0002, "epoch": 1.7407545760179306, "step": 2330}, {"loss": 1.7041, "grad_norm": 0.3578338325023651, "learning_rate": 0.0002, "epoch": 1.748225625700411, "step": 2340}, {"loss": 1.6456, "grad_norm": 0.36057502031326294, "learning_rate": 0.0002, "epoch": 1.7556966753828913, "step": 2350}, {"loss": 1.6853, "grad_norm": 0.3615196645259857, "learning_rate": 0.0002, "epoch": 1.7631677250653717, "step": 2360}, {"loss": 1.7612, "grad_norm": 0.4118947684764862, "learning_rate": 0.0002, "epoch": 1.770638774747852, "step": 2370}, {"loss": 1.6946, "grad_norm": 0.4067276120185852, "learning_rate": 0.0002, "epoch": 1.7781098244303326, "step": 2380}, {"loss": 1.712, "grad_norm": 0.3979823887348175, "learning_rate": 0.0002, "epoch": 1.7855808741128127, "step": 2390}, {"loss": 1.7644, "grad_norm": 0.44045883417129517, "learning_rate": 0.0002, "epoch": 1.7930519237952933, "step": 2400}, {"loss": 1.7251, "grad_norm": 0.3998069167137146, "learning_rate": 0.0002, "epoch": 1.8005229734777737, "step": 2410}, {"loss": 1.7354, "grad_norm": 0.3450094759464264, "learning_rate": 0.0002, "epoch": 1.807994023160254, "step": 2420}, {"loss": 1.6998, "grad_norm": 0.3759009838104248, "learning_rate": 0.0002, "epoch": 1.8154650728427344, "step": 2430}, {"loss": 1.7706, "grad_norm": 0.34347015619277954, "learning_rate": 0.0002, "epoch": 1.8229361225252148, "step": 2440}, {"loss": 1.7345, "grad_norm": 0.3511228859424591, "learning_rate": 0.0002, "epoch": 1.8304071722076953, "step": 2450}, {"loss": 1.6909, "grad_norm": 0.36853715777397156, "learning_rate": 0.0002, "epoch": 1.8378782218901755, "step": 2460}, {"loss": 1.6931, "grad_norm": 0.40659376978874207, "learning_rate": 0.0002, "epoch": 1.845349271572656, "step": 2470}, {"loss": 1.7626, "grad_norm": 0.39621320366859436, "learning_rate": 0.0002, "epoch": 1.8528203212551362, "step": 2480}, {"loss": 1.7427, "grad_norm": 0.3753979504108429, "learning_rate": 0.0002, "epoch": 1.8602913709376168, "step": 2490}, {"loss": 1.6622, "grad_norm": 0.3811938464641571, "learning_rate": 0.0002, "epoch": 1.8677624206200971, "step": 2500}, {"loss": 1.7718, "grad_norm": 0.3432596027851105, "learning_rate": 0.0002, "epoch": 1.8752334703025775, "step": 2510}, {"loss": 1.7488, "grad_norm": 0.3670712113380432, "learning_rate": 0.0002, "epoch": 1.882704519985058, "step": 2520}, {"loss": 1.705, "grad_norm": 0.40907177329063416, "learning_rate": 0.0002, "epoch": 1.8901755696675382, "step": 2530}, {"loss": 1.7148, "grad_norm": 0.3821999728679657, "learning_rate": 0.0002, "epoch": 1.8976466193500188, "step": 2540}, {"loss": 1.7934, "grad_norm": 0.36173978447914124, "learning_rate": 0.0002, "epoch": 1.905117669032499, "step": 2550}, {"loss": 1.6939, "grad_norm": 0.38990336656570435, "learning_rate": 0.0002, "epoch": 1.9125887187149795, "step": 2560}, {"loss": 1.6893, "grad_norm": 0.35242322087287903, "learning_rate": 0.0002, "epoch": 1.9200597683974598, "step": 2570}, {"loss": 1.7268, "grad_norm": 0.3506428003311157, "learning_rate": 0.0002, "epoch": 1.9275308180799402, "step": 2580}, {"loss": 1.6953, "grad_norm": 0.39540135860443115, "learning_rate": 0.0002, "epoch": 1.9350018677624208, "step": 2590}, {"loss": 1.6511, "grad_norm": 0.3444725573062897, "learning_rate": 0.0002, "epoch": 1.942472917444901, "step": 2600}, {"loss": 1.7259, "grad_norm": 0.3963521718978882, "learning_rate": 0.0002, "epoch": 1.9499439671273815, "step": 2610}, {"loss": 1.6946, "grad_norm": 0.3689815402030945, "learning_rate": 0.0002, "epoch": 1.9574150168098616, "step": 2620}, {"loss": 1.7384, "grad_norm": 0.3482626676559448, "learning_rate": 0.0002, "epoch": 1.9648860664923422, "step": 2630}, {"loss": 1.7048, "grad_norm": 0.35832616686820984, "learning_rate": 0.0002, "epoch": 1.9723571161748226, "step": 2640}, {"loss": 1.6681, "grad_norm": 0.4776208996772766, "learning_rate": 0.0002, "epoch": 1.979828165857303, "step": 2650}, {"loss": 1.6696, "grad_norm": 0.32570165395736694, "learning_rate": 0.0002, "epoch": 1.9872992155397835, "step": 2660}, {"loss": 1.7232, "grad_norm": 0.3380725085735321, "learning_rate": 0.0002, "epoch": 1.9947702652222636, "step": 2670}, {"eval_loss": 1.8046749830245972, "eval_runtime": 38.5096, "eval_samples_per_second": 13.373, "eval_steps_per_second": 1.688, "epoch": 2.0, "step": 2677}, {"loss": 1.7265, "grad_norm": 0.36817631125450134, "learning_rate": 0.0002, "epoch": 2.002241314904744, "step": 2680}, {"loss": 1.548, "grad_norm": 0.4056456685066223, "learning_rate": 0.0002, "epoch": 2.0097123645872244, "step": 2690}, {"loss": 1.5515, "grad_norm": 0.37416863441467285, "learning_rate": 0.0002, "epoch": 2.017183414269705, "step": 2700}, {"loss": 1.5895, "grad_norm": 0.4273638427257538, "learning_rate": 0.0002, "epoch": 2.024654463952185, "step": 2710}, {"loss": 1.5884, "grad_norm": 0.36497923731803894, "learning_rate": 0.0002, "epoch": 2.0321255136346656, "step": 2720}, {"loss": 1.6999, "grad_norm": 0.5021994113922119, "learning_rate": 0.0002, "epoch": 2.0395965633171462, "step": 2730}, {"loss": 1.6655, "grad_norm": 0.45896220207214355, "learning_rate": 0.0002, "epoch": 2.0470676129996264, "step": 2740}, {"loss": 1.6305, "grad_norm": 0.3973815143108368, "learning_rate": 0.0002, "epoch": 2.054538662682107, "step": 2750}, {"loss": 1.6301, "grad_norm": 0.4521815776824951, "learning_rate": 0.0002, "epoch": 2.062009712364587, "step": 2760}, {"loss": 1.6189, "grad_norm": 0.42775002121925354, "learning_rate": 0.0002, "epoch": 2.0694807620470677, "step": 2770}, {"loss": 1.6491, "grad_norm": 0.48158586025238037, "learning_rate": 0.0002, "epoch": 2.076951811729548, "step": 2780}, {"loss": 1.6301, "grad_norm": 0.4612371623516083, "learning_rate": 0.0002, "epoch": 2.0844228614120284, "step": 2790}, {"loss": 1.6327, "grad_norm": 0.42536866664886475, "learning_rate": 0.0002, "epoch": 2.091893911094509, "step": 2800}, {"loss": 1.651, "grad_norm": 0.48515772819519043, "learning_rate": 0.0002, "epoch": 2.099364960776989, "step": 2810}, {"loss": 1.6829, "grad_norm": 0.41418662667274475, "learning_rate": 0.0002, "epoch": 2.1068360104594697, "step": 2820}, {"loss": 1.6266, "grad_norm": 0.4683697819709778, "learning_rate": 0.0002, "epoch": 2.11430706014195, "step": 2830}, {"loss": 1.6586, "grad_norm": 0.4484657049179077, "learning_rate": 0.0002, "epoch": 2.1217781098244304, "step": 2840}, {"loss": 1.6483, "grad_norm": 0.6621400713920593, "learning_rate": 0.0002, "epoch": 2.1292491595069105, "step": 2850}, {"loss": 1.5755, "grad_norm": 0.45074811577796936, "learning_rate": 0.0002, "epoch": 2.136720209189391, "step": 2860}, {"loss": 1.6456, "grad_norm": 0.3513113558292389, "learning_rate": 0.0002, "epoch": 2.1441912588718717, "step": 2870}, {"loss": 1.6081, "grad_norm": 0.40411314368247986, "learning_rate": 0.0002, "epoch": 2.151662308554352, "step": 2880}, {"loss": 1.6323, "grad_norm": 0.4121065139770508, "learning_rate": 0.0002, "epoch": 2.1591333582368324, "step": 2890}, {"loss": 1.6324, "grad_norm": 0.44723689556121826, "learning_rate": 0.0002, "epoch": 2.1666044079193125, "step": 2900}, {"loss": 1.5699, "grad_norm": 0.4226122498512268, "learning_rate": 0.0002, "epoch": 2.174075457601793, "step": 2910}, {"loss": 1.5652, "grad_norm": 0.46617650985717773, "learning_rate": 0.0002, "epoch": 2.1815465072842732, "step": 2920}, {"loss": 1.6378, "grad_norm": 0.4506422281265259, "learning_rate": 0.0002, "epoch": 2.189017556966754, "step": 2930}, {"loss": 1.6112, "grad_norm": 0.4892672896385193, "learning_rate": 0.0002, "epoch": 2.1964886066492344, "step": 2940}, {"loss": 1.6176, "grad_norm": 0.44095516204833984, "learning_rate": 0.0002, "epoch": 2.2039596563317145, "step": 2950}, {"loss": 1.6058, "grad_norm": 0.41522109508514404, "learning_rate": 0.0002, "epoch": 2.211430706014195, "step": 2960}, {"loss": 1.5964, "grad_norm": 0.4860858917236328, "learning_rate": 0.0002, "epoch": 2.2189017556966752, "step": 2970}, {"loss": 1.6427, "grad_norm": 0.42662516236305237, "learning_rate": 0.0002, "epoch": 2.226372805379156, "step": 2980}, {"loss": 1.6313, "grad_norm": 0.4390648305416107, "learning_rate": 0.0002, "epoch": 2.233843855061636, "step": 2990}, {"loss": 1.5992, "grad_norm": 0.47515565156936646, "learning_rate": 0.0002, "epoch": 2.2413149047441165, "step": 3000}, {"loss": 1.5563, "grad_norm": 0.4104543924331665, "learning_rate": 0.0002, "epoch": 2.248785954426597, "step": 3010}, {"loss": 1.6895, "grad_norm": 0.4404028654098511, "learning_rate": 0.0002, "epoch": 2.2562570041090773, "step": 3020}, {"loss": 1.6088, "grad_norm": 0.4717366695404053, "learning_rate": 0.0002, "epoch": 2.263728053791558, "step": 3030}, {"loss": 1.7287, "grad_norm": 0.48345857858657837, "learning_rate": 0.0002, "epoch": 2.271199103474038, "step": 3040}, {"loss": 1.681, "grad_norm": 0.5312452912330627, "learning_rate": 0.0002, "epoch": 2.2786701531565186, "step": 3050}, {"loss": 1.5901, "grad_norm": 0.5073099732398987, "learning_rate": 0.0002, "epoch": 2.2861412028389987, "step": 3060}, {"loss": 1.6914, "grad_norm": 0.5027463436126709, "learning_rate": 0.0002, "epoch": 2.2936122525214793, "step": 3070}, {"loss": 1.5862, "grad_norm": 0.5436304807662964, "learning_rate": 0.0002, "epoch": 2.30108330220396, "step": 3080}, {"loss": 1.5763, "grad_norm": 0.4701065123081207, "learning_rate": 0.0002, "epoch": 2.30855435188644, "step": 3090}, {"loss": 1.6177, "grad_norm": 0.46988746523857117, "learning_rate": 0.0002, "epoch": 2.3160254015689206, "step": 3100}, {"loss": 1.6502, "grad_norm": 0.45112869143486023, "learning_rate": 0.0002, "epoch": 2.3234964512514007, "step": 3110}, {"loss": 1.6291, "grad_norm": 0.5173566937446594, "learning_rate": 0.0002, "epoch": 2.3309675009338813, "step": 3120}, {"loss": 1.6743, "grad_norm": 0.40345850586891174, "learning_rate": 0.0002, "epoch": 2.3384385506163614, "step": 3130}, {"loss": 1.621, "grad_norm": 0.4218924939632416, "learning_rate": 0.0002, "epoch": 2.345909600298842, "step": 3140}, {"loss": 1.6341, "grad_norm": 0.41857317090034485, "learning_rate": 0.0002, "epoch": 2.3533806499813226, "step": 3150}, {"loss": 1.6087, "grad_norm": 0.4197218418121338, "learning_rate": 0.0002, "epoch": 2.3608516996638027, "step": 3160}, {"loss": 1.6572, "grad_norm": 0.4260677397251129, "learning_rate": 0.0002, "epoch": 2.3683227493462833, "step": 3170}, {"loss": 1.6376, "grad_norm": 0.4209042191505432, "learning_rate": 0.0002, "epoch": 2.3757937990287634, "step": 3180}, {"loss": 1.634, "grad_norm": 0.4092234969139099, "learning_rate": 0.0002, "epoch": 2.383264848711244, "step": 3190}, {"loss": 1.6339, "grad_norm": 0.4928431510925293, "learning_rate": 0.0002, "epoch": 2.390735898393724, "step": 3200}, {"loss": 1.6015, "grad_norm": 0.49252402782440186, "learning_rate": 0.0002, "epoch": 2.3982069480762047, "step": 3210}, {"loss": 1.5773, "grad_norm": 0.4368397295475006, "learning_rate": 0.0002, "epoch": 2.4056779977586853, "step": 3220}, {"loss": 1.6629, "grad_norm": 0.46122390031814575, "learning_rate": 0.0002, "epoch": 2.4131490474411654, "step": 3230}, {"loss": 1.6224, "grad_norm": 0.4272301197052002, "learning_rate": 0.0002, "epoch": 2.420620097123646, "step": 3240}, {"loss": 1.5961, "grad_norm": 0.41480937600135803, "learning_rate": 0.0002, "epoch": 2.428091146806126, "step": 3250}, {"loss": 1.6281, "grad_norm": 0.48911941051483154, "learning_rate": 0.0002, "epoch": 2.4355621964886067, "step": 3260}, {"loss": 1.6846, "grad_norm": 0.4444098472595215, "learning_rate": 0.0002, "epoch": 2.443033246171087, "step": 3270}, {"loss": 1.6961, "grad_norm": 0.5111684799194336, "learning_rate": 0.0002, "epoch": 2.4505042958535674, "step": 3280}, {"loss": 1.6152, "grad_norm": 0.5058825016021729, "learning_rate": 0.0002, "epoch": 2.457975345536048, "step": 3290}, {"loss": 1.625, "grad_norm": 0.44173210859298706, "learning_rate": 0.0002, "epoch": 2.465446395218528, "step": 3300}, {"loss": 1.6491, "grad_norm": 0.4659745991230011, "learning_rate": 0.0002, "epoch": 2.4729174449010087, "step": 3310}, {"loss": 1.6114, "grad_norm": 0.47237497568130493, "learning_rate": 0.0002, "epoch": 2.480388494583489, "step": 3320}, {"loss": 1.6193, "grad_norm": 0.47303131222724915, "learning_rate": 0.0002, "epoch": 2.4878595442659694, "step": 3330}, {"loss": 1.7256, "grad_norm": 0.4522389769554138, "learning_rate": 0.0002, "epoch": 2.4953305939484496, "step": 3340}, {"loss": 1.6834, "grad_norm": 0.4467332363128662, "learning_rate": 0.0002, "epoch": 2.50280164363093, "step": 3350}, {"loss": 1.6108, "grad_norm": 0.4413762092590332, "learning_rate": 0.0002, "epoch": 2.5102726933134107, "step": 3360}, {"loss": 1.537, "grad_norm": 0.495514452457428, "learning_rate": 0.0002, "epoch": 2.517743742995891, "step": 3370}, {"loss": 1.5839, "grad_norm": 0.4429773986339569, "learning_rate": 0.0002, "epoch": 2.5252147926783715, "step": 3380}, {"loss": 1.6522, "grad_norm": 0.4589079022407532, "learning_rate": 0.0002, "epoch": 2.5326858423608516, "step": 3390}, {"loss": 1.6529, "grad_norm": 0.4683997333049774, "learning_rate": 0.0002, "epoch": 2.540156892043332, "step": 3400}, {"loss": 1.6745, "grad_norm": 0.4651731252670288, "learning_rate": 0.0002, "epoch": 2.5476279417258123, "step": 3410}, {"loss": 1.5918, "grad_norm": 0.45818084478378296, "learning_rate": 0.0002, "epoch": 2.555098991408293, "step": 3420}, {"loss": 1.6326, "grad_norm": 0.45209529995918274, "learning_rate": 0.0002, "epoch": 2.5625700410907735, "step": 3430}, {"loss": 1.5606, "grad_norm": 0.4344733655452728, "learning_rate": 0.0002, "epoch": 2.5700410907732536, "step": 3440}, {"loss": 1.6748, "grad_norm": 0.47435566782951355, "learning_rate": 0.0002, "epoch": 2.577512140455734, "step": 3450}, {"loss": 1.6237, "grad_norm": 0.43841999769210815, "learning_rate": 0.0002, "epoch": 2.5849831901382143, "step": 3460}, {"loss": 1.7207, "grad_norm": 0.4323869049549103, "learning_rate": 0.0002, "epoch": 2.592454239820695, "step": 3470}, {"loss": 1.5494, "grad_norm": 0.44355881214141846, "learning_rate": 0.0002, "epoch": 2.599925289503175, "step": 3480}, {"loss": 1.665, "grad_norm": 0.45847779512405396, "learning_rate": 0.0002, "epoch": 2.6073963391856556, "step": 3490}, {"loss": 1.6006, "grad_norm": 0.4411061704158783, "learning_rate": 0.0002, "epoch": 2.614867388868136, "step": 3500}, {"loss": 1.5868, "grad_norm": 0.4446796178817749, "learning_rate": 0.0002, "epoch": 2.6223384385506163, "step": 3510}, {"loss": 1.5946, "grad_norm": 0.41969653964042664, "learning_rate": 0.0002, "epoch": 2.629809488233097, "step": 3520}, {"loss": 1.6798, "grad_norm": 0.5263747572898865, "learning_rate": 0.0002, "epoch": 2.637280537915577, "step": 3530}, {"loss": 1.6309, "grad_norm": 0.47719451785087585, "learning_rate": 0.0002, "epoch": 2.6447515875980576, "step": 3540}, {"loss": 1.7024, "grad_norm": 0.46574118733406067, "learning_rate": 0.0002, "epoch": 2.6522226372805378, "step": 3550}, {"loss": 1.618, "grad_norm": 0.46867135167121887, "learning_rate": 0.0002, "epoch": 2.6596936869630183, "step": 3560}, {"loss": 1.5885, "grad_norm": 0.4441198706626892, "learning_rate": 0.0002, "epoch": 2.667164736645499, "step": 3570}, {"loss": 1.6426, "grad_norm": 0.4871319830417633, "learning_rate": 0.0002, "epoch": 2.674635786327979, "step": 3580}, {"loss": 1.6575, "grad_norm": 0.43900373578071594, "learning_rate": 0.0002, "epoch": 2.6821068360104596, "step": 3590}, {"loss": 1.6071, "grad_norm": 0.42509549856185913, "learning_rate": 0.0002, "epoch": 2.6895778856929398, "step": 3600}, {"loss": 1.5651, "grad_norm": 0.4691086709499359, "learning_rate": 0.0002, "epoch": 2.6970489353754203, "step": 3610}, {"loss": 1.5491, "grad_norm": 0.46318942308425903, "learning_rate": 0.0002, "epoch": 2.7045199850579005, "step": 3620}, {"loss": 1.5422, "grad_norm": 0.44631096720695496, "learning_rate": 0.0002, "epoch": 2.711991034740381, "step": 3630}, {"loss": 1.6831, "grad_norm": 0.42315489053726196, "learning_rate": 0.0002, "epoch": 2.7194620844228616, "step": 3640}, {"loss": 1.6008, "grad_norm": 0.4971241056919098, "learning_rate": 0.0002, "epoch": 2.7269331341053418, "step": 3650}, {"loss": 1.6042, "grad_norm": 0.4578486382961273, "learning_rate": 0.0002, "epoch": 2.7344041837878224, "step": 3660}, {"loss": 1.6076, "grad_norm": 0.46584776043891907, "learning_rate": 0.0002, "epoch": 2.7418752334703025, "step": 3670}, {"loss": 1.6809, "grad_norm": 0.4951731264591217, "learning_rate": 0.0002, "epoch": 2.749346283152783, "step": 3680}, {"loss": 1.6226, "grad_norm": 0.4935225546360016, "learning_rate": 0.0002, "epoch": 2.756817332835263, "step": 3690}, {"loss": 1.5878, "grad_norm": 0.41805586218833923, "learning_rate": 0.0002, "epoch": 2.764288382517744, "step": 3700}, {"loss": 1.7173, "grad_norm": 0.4417555630207062, "learning_rate": 0.0002, "epoch": 2.7717594322002244, "step": 3710}, {"loss": 1.6398, "grad_norm": 0.48229655623435974, "learning_rate": 0.0002, "epoch": 2.7792304818827045, "step": 3720}, {"loss": 1.6074, "grad_norm": 0.48562315106391907, "learning_rate": 0.0002, "epoch": 2.786701531565185, "step": 3730}, {"loss": 1.607, "grad_norm": 0.4473940432071686, "learning_rate": 0.0002, "epoch": 2.794172581247665, "step": 3740}, {"loss": 1.6065, "grad_norm": 0.4626813232898712, "learning_rate": 0.0002, "epoch": 2.801643630930146, "step": 3750}, {"loss": 1.6296, "grad_norm": 0.4339792728424072, "learning_rate": 0.0002, "epoch": 2.809114680612626, "step": 3760}, {"loss": 1.6815, "grad_norm": 0.5250858068466187, "learning_rate": 0.0002, "epoch": 2.8165857302951065, "step": 3770}, {"loss": 1.6644, "grad_norm": 0.4537523090839386, "learning_rate": 0.0002, "epoch": 2.824056779977587, "step": 3780}, {"loss": 1.6535, "grad_norm": 0.5646113157272339, "learning_rate": 0.0002, "epoch": 2.831527829660067, "step": 3790}, {"loss": 1.5712, "grad_norm": 0.44243332743644714, "learning_rate": 0.0002, "epoch": 2.8389988793425474, "step": 3800}, {"loss": 1.6478, "grad_norm": 0.4585791826248169, "learning_rate": 0.0002, "epoch": 2.846469929025028, "step": 3810}, {"loss": 1.6854, "grad_norm": 0.489702045917511, "learning_rate": 0.0002, "epoch": 2.8539409787075085, "step": 3820}, {"loss": 1.7066, "grad_norm": 0.502470850944519, "learning_rate": 0.0002, "epoch": 2.8614120283899886, "step": 3830}, {"loss": 1.5785, "grad_norm": 0.4395960867404938, "learning_rate": 0.0002, "epoch": 2.8688830780724692, "step": 3840}, {"loss": 1.6434, "grad_norm": 0.4348670244216919, "learning_rate": 0.0002, "epoch": 2.87635412775495, "step": 3850}, {"loss": 1.6163, "grad_norm": 0.48852720856666565, "learning_rate": 0.0002, "epoch": 2.88382517743743, "step": 3860}, {"loss": 1.5916, "grad_norm": 0.45317450165748596, "learning_rate": 0.0002, "epoch": 2.89129622711991, "step": 3870}, {"loss": 1.6486, "grad_norm": 0.4732758700847626, "learning_rate": 0.0002, "epoch": 2.8987672768023907, "step": 3880}, {"loss": 1.6758, "grad_norm": 0.45238012075424194, "learning_rate": 0.0002, "epoch": 2.9062383264848712, "step": 3890}, {"loss": 1.6228, "grad_norm": 0.48838064074516296, "learning_rate": 0.0002, "epoch": 2.9137093761673514, "step": 3900}, {"loss": 1.658, "grad_norm": 0.43496349453926086, "learning_rate": 0.0002, "epoch": 2.921180425849832, "step": 3910}, {"loss": 1.7063, "grad_norm": 0.47963935136795044, "learning_rate": 0.0002, "epoch": 2.9286514755323125, "step": 3920}, {"loss": 1.6553, "grad_norm": 0.4544987976551056, "learning_rate": 0.0002, "epoch": 2.9361225252147927, "step": 3930}, {"loss": 1.6192, "grad_norm": 0.4622892141342163, "learning_rate": 0.0002, "epoch": 2.943593574897273, "step": 3940}, {"loss": 1.6178, "grad_norm": 0.47026222944259644, "learning_rate": 0.0002, "epoch": 2.9510646245797534, "step": 3950}, {"loss": 1.6612, "grad_norm": 0.4549552798271179, "learning_rate": 0.0002, "epoch": 2.958535674262234, "step": 3960}, {"loss": 1.6458, "grad_norm": 0.46647515892982483, "learning_rate": 0.0002, "epoch": 2.966006723944714, "step": 3970}, {"loss": 1.6051, "grad_norm": 0.45095112919807434, "learning_rate": 0.0002, "epoch": 2.9734777736271947, "step": 3980}, {"loss": 1.6471, "grad_norm": 0.4690017104148865, "learning_rate": 0.0002, "epoch": 2.9809488233096753, "step": 3990}, {"loss": 1.6061, "grad_norm": 0.4603444039821625, "learning_rate": 0.0002, "epoch": 2.9884198729921554, "step": 4000}, {"loss": 1.6431, "grad_norm": 0.4743294417858124, "learning_rate": 0.0002, "epoch": 2.9958909226746355, "step": 4010}, {"eval_loss": 1.8252571821212769, "eval_runtime": 38.7853, "eval_samples_per_second": 13.278, "eval_steps_per_second": 1.676, "epoch": 2.999626447515876, "step": 4015}, {"loss": 1.6512, "grad_norm": 0.4919724464416504, "learning_rate": 0.0002, "epoch": 3.003361972357116, "step": 4020}, {"loss": 1.5354, "grad_norm": 0.4747185707092285, "learning_rate": 0.0002, "epoch": 3.0108330220395967, "step": 4030}, {"loss": 1.568, "grad_norm": 0.4797595143318176, "learning_rate": 0.0002, "epoch": 3.018304071722077, "step": 4040}, {"loss": 1.5194, "grad_norm": 0.5450999140739441, "learning_rate": 0.0002, "epoch": 3.0257751214045574, "step": 4050}, {"loss": 1.5065, "grad_norm": 0.49058812856674194, "learning_rate": 0.0002, "epoch": 3.0332461710870375, "step": 4060}, {"loss": 1.4884, "grad_norm": 0.5219563841819763, "learning_rate": 0.0002, "epoch": 3.040717220769518, "step": 4070}, {"loss": 1.4742, "grad_norm": 0.515628457069397, "learning_rate": 0.0002, "epoch": 3.0481882704519987, "step": 4080}, {"loss": 1.5313, "grad_norm": 0.6145984530448914, "learning_rate": 0.0002, "epoch": 3.055659320134479, "step": 4090}, {"loss": 1.4989, "grad_norm": 0.6067144274711609, "learning_rate": 0.0002, "epoch": 3.0631303698169594, "step": 4100}, {"loss": 1.528, "grad_norm": 0.5773133039474487, "learning_rate": 0.0002, "epoch": 3.0706014194994395, "step": 4110}, {"loss": 1.5374, "grad_norm": 0.6894241571426392, "learning_rate": 0.0002, "epoch": 3.07807246918192, "step": 4120}, {"loss": 1.5422, "grad_norm": 0.6422514915466309, "learning_rate": 0.0002, "epoch": 3.0855435188644003, "step": 4130}, {"loss": 1.4724, "grad_norm": 0.6119855046272278, "learning_rate": 0.0002, "epoch": 3.093014568546881, "step": 4140}, {"loss": 1.5361, "grad_norm": 0.5847280025482178, "learning_rate": 0.0002, "epoch": 3.1004856182293614, "step": 4150}, {"loss": 1.5151, "grad_norm": 0.5401515960693359, "learning_rate": 0.0002, "epoch": 3.1079566679118416, "step": 4160}, {"loss": 1.502, "grad_norm": 0.6501587629318237, "learning_rate": 0.0002, "epoch": 3.115427717594322, "step": 4170}, {"loss": 1.4952, "grad_norm": 0.5988039374351501, "learning_rate": 0.0002, "epoch": 3.1228987672768023, "step": 4180}, {"loss": 1.5287, "grad_norm": 0.4982665181159973, "learning_rate": 0.0002, "epoch": 3.130369816959283, "step": 4190}, {"loss": 1.5078, "grad_norm": 0.5548039078712463, "learning_rate": 0.0002, "epoch": 3.137840866641763, "step": 4200}, {"loss": 1.4904, "grad_norm": 0.5920777320861816, "learning_rate": 0.0002, "epoch": 3.1453119163242436, "step": 4210}, {"loss": 1.442, "grad_norm": 0.6965190172195435, "learning_rate": 0.0002, "epoch": 3.152782966006724, "step": 4220}, {"loss": 1.557, "grad_norm": 0.5196244716644287, "learning_rate": 0.0002, "epoch": 3.1602540156892043, "step": 4230}, {"loss": 1.5706, "grad_norm": 0.6942682266235352, "learning_rate": 0.0002, "epoch": 3.167725065371685, "step": 4240}, {"loss": 1.5407, "grad_norm": 0.5765156149864197, "learning_rate": 0.0002, "epoch": 3.175196115054165, "step": 4250}, {"loss": 1.4963, "grad_norm": 0.5801976919174194, "learning_rate": 0.0002, "epoch": 3.1826671647366456, "step": 4260}, {"loss": 1.4988, "grad_norm": 0.6260752081871033, "learning_rate": 0.0002, "epoch": 3.1901382144191257, "step": 4270}, {"loss": 1.5074, "grad_norm": 0.6610770225524902, "learning_rate": 0.0002, "epoch": 3.1976092641016063, "step": 4280}, {"loss": 1.4657, "grad_norm": 0.5762143135070801, "learning_rate": 0.0002, "epoch": 3.205080313784087, "step": 4290}, {"loss": 1.5181, "grad_norm": 0.5926990509033203, "learning_rate": 0.0002, "epoch": 3.212551363466567, "step": 4300}, {"loss": 1.5492, "grad_norm": 0.7373854517936707, "learning_rate": 0.0002, "epoch": 3.2200224131490476, "step": 4310}, {"loss": 1.4648, "grad_norm": 0.5963311195373535, "learning_rate": 0.0002, "epoch": 3.2274934628315277, "step": 4320}, {"loss": 1.5262, "grad_norm": 0.5754616856575012, "learning_rate": 0.0002, "epoch": 3.2349645125140083, "step": 4330}, {"loss": 1.4767, "grad_norm": 0.6116095781326294, "learning_rate": 0.0002, "epoch": 3.2424355621964884, "step": 4340}, {"loss": 1.5008, "grad_norm": 0.6001536846160889, "learning_rate": 0.0002, "epoch": 3.249906611878969, "step": 4350}, {"loss": 1.5738, "grad_norm": 0.5270227789878845, "learning_rate": 0.0002, "epoch": 3.257377661561449, "step": 4360}, {"loss": 1.5235, "grad_norm": 0.6666602492332458, "learning_rate": 0.0002, "epoch": 3.2648487112439297, "step": 4370}, {"loss": 1.5665, "grad_norm": 0.520310640335083, "learning_rate": 0.0002, "epoch": 3.2723197609264103, "step": 4380}, {"loss": 1.542, "grad_norm": 0.5165975093841553, "learning_rate": 0.0002, "epoch": 3.2797908106088904, "step": 4390}, {"loss": 1.4746, "grad_norm": 0.6080228686332703, "learning_rate": 0.0002, "epoch": 3.287261860291371, "step": 4400}, {"loss": 1.4901, "grad_norm": 0.670122504234314, "learning_rate": 0.0002, "epoch": 3.294732909973851, "step": 4410}, {"loss": 1.4677, "grad_norm": 0.6019457578659058, "learning_rate": 0.0002, "epoch": 3.3022039596563317, "step": 4420}, {"loss": 1.4249, "grad_norm": 0.5519300103187561, "learning_rate": 0.0002, "epoch": 3.309675009338812, "step": 4430}, {"loss": 1.555, "grad_norm": 0.5958521962165833, "learning_rate": 0.0002, "epoch": 3.3171460590212924, "step": 4440}, {"loss": 1.5067, "grad_norm": 0.5552705526351929, "learning_rate": 0.0002, "epoch": 3.324617108703773, "step": 4450}, {"loss": 1.5926, "grad_norm": 0.6583784818649292, "learning_rate": 0.0002, "epoch": 3.332088158386253, "step": 4460}, {"loss": 1.4206, "grad_norm": 0.5815939903259277, "learning_rate": 0.0002, "epoch": 3.3395592080687337, "step": 4470}, {"loss": 1.5942, "grad_norm": 1.3342205286026, "learning_rate": 0.0002, "epoch": 3.347030257751214, "step": 4480}, {"loss": 1.484, "grad_norm": 0.6341500878334045, "learning_rate": 0.0002, "epoch": 3.3545013074336945, "step": 4490}, {"loss": 1.5219, "grad_norm": 0.6384079456329346, "learning_rate": 0.0002, "epoch": 3.3619723571161746, "step": 4500}, {"loss": 1.5222, "grad_norm": 0.6098346710205078, "learning_rate": 0.0002, "epoch": 3.369443406798655, "step": 4510}, {"loss": 1.5475, "grad_norm": 0.5958296656608582, "learning_rate": 0.0002, "epoch": 3.3769144564811358, "step": 4520}, {"loss": 1.5171, "grad_norm": 0.6157881617546082, "learning_rate": 0.0002, "epoch": 3.384385506163616, "step": 4530}, {"loss": 1.569, "grad_norm": 0.5671007037162781, "learning_rate": 0.0002, "epoch": 3.3918565558460965, "step": 4540}, {"loss": 1.604, "grad_norm": 0.6203294992446899, "learning_rate": 0.0002, "epoch": 3.3993276055285766, "step": 4550}, {"loss": 1.5364, "grad_norm": 0.6743317246437073, "learning_rate": 0.0002, "epoch": 3.406798655211057, "step": 4560}, {"loss": 1.5034, "grad_norm": 0.731765627861023, "learning_rate": 0.0002, "epoch": 3.4142697048935373, "step": 4570}, {"loss": 1.4585, "grad_norm": 0.6285187602043152, "learning_rate": 0.0002, "epoch": 3.421740754576018, "step": 4580}, {"loss": 1.5296, "grad_norm": 0.612680196762085, "learning_rate": 0.0002, "epoch": 3.4292118042584985, "step": 4590}, {"loss": 1.5577, "grad_norm": 0.6413681507110596, "learning_rate": 0.0002, "epoch": 3.4366828539409786, "step": 4600}, {"loss": 1.5026, "grad_norm": 0.6240990161895752, "learning_rate": 0.0002, "epoch": 3.444153903623459, "step": 4610}, {"loss": 1.5887, "grad_norm": 0.5095735192298889, "learning_rate": 0.0002, "epoch": 3.4516249533059393, "step": 4620}, {"loss": 1.4906, "grad_norm": 0.5699611902236938, "learning_rate": 0.0002, "epoch": 3.45909600298842, "step": 4630}, {"loss": 1.5176, "grad_norm": 0.7289775609970093, "learning_rate": 0.0002, "epoch": 3.4665670526709, "step": 4640}, {"loss": 1.5467, "grad_norm": 0.6211609840393066, "learning_rate": 0.0002, "epoch": 3.4740381023533806, "step": 4650}, {"loss": 1.533, "grad_norm": 0.5714802145957947, "learning_rate": 0.0002, "epoch": 3.481509152035861, "step": 4660}, {"loss": 1.5096, "grad_norm": 0.6287049651145935, "learning_rate": 0.0002, "epoch": 3.4889802017183413, "step": 4670}, {"loss": 1.4212, "grad_norm": 0.5480595827102661, "learning_rate": 0.0002, "epoch": 3.496451251400822, "step": 4680}, {"loss": 1.4746, "grad_norm": 0.5683253407478333, "learning_rate": 0.0002, "epoch": 3.503922301083302, "step": 4690}, {"loss": 1.5012, "grad_norm": 0.601140558719635, "learning_rate": 0.0002, "epoch": 3.5113933507657826, "step": 4700}, {"loss": 1.5383, "grad_norm": 0.5344498157501221, "learning_rate": 0.0002, "epoch": 3.5188644004482628, "step": 4710}, {"loss": 1.5428, "grad_norm": 0.5739690661430359, "learning_rate": 0.0002, "epoch": 3.5263354501307433, "step": 4720}, {"loss": 1.5589, "grad_norm": 0.5640085935592651, "learning_rate": 0.0002, "epoch": 3.533806499813224, "step": 4730}, {"loss": 1.487, "grad_norm": 0.5967805981636047, "learning_rate": 0.0002, "epoch": 3.541277549495704, "step": 4740}, {"loss": 1.5461, "grad_norm": 0.6138835549354553, "learning_rate": 0.0002, "epoch": 3.5487485991781846, "step": 4750}, {"loss": 1.5502, "grad_norm": 0.6779900193214417, "learning_rate": 0.0002, "epoch": 3.5562196488606648, "step": 4760}, {"loss": 1.4917, "grad_norm": 0.6122010350227356, "learning_rate": 0.0002, "epoch": 3.5636906985431454, "step": 4770}, {"loss": 1.5405, "grad_norm": 0.5685241222381592, "learning_rate": 0.0002, "epoch": 3.5711617482256255, "step": 4780}, {"loss": 1.5427, "grad_norm": 0.604583203792572, "learning_rate": 0.0002, "epoch": 3.578632797908106, "step": 4790}, {"loss": 1.4514, "grad_norm": 0.651165246963501, "learning_rate": 0.0002, "epoch": 3.5861038475905866, "step": 4800}, {"loss": 1.4109, "grad_norm": 0.6398511528968811, "learning_rate": 0.0002, "epoch": 3.593574897273067, "step": 4810}, {"loss": 1.4261, "grad_norm": 0.6444641351699829, "learning_rate": 0.0002, "epoch": 3.6010459469555474, "step": 4820}, {"loss": 1.5274, "grad_norm": 0.6018481850624084, "learning_rate": 0.0002, "epoch": 3.6085169966380275, "step": 4830}, {"loss": 1.4647, "grad_norm": 0.6025291085243225, "learning_rate": 0.0002, "epoch": 3.615988046320508, "step": 4840}, {"loss": 1.5609, "grad_norm": 0.6810156106948853, "learning_rate": 0.0002, "epoch": 3.623459096002988, "step": 4850}, {"loss": 1.5299, "grad_norm": 0.6408044695854187, "learning_rate": 0.0002, "epoch": 3.630930145685469, "step": 4860}, {"loss": 1.5366, "grad_norm": 0.5608272552490234, "learning_rate": 0.0002, "epoch": 3.6384011953679494, "step": 4870}, {"loss": 1.5188, "grad_norm": 0.6136814951896667, "learning_rate": 0.0002, "epoch": 3.6458722450504295, "step": 4880}, {"loss": 1.5021, "grad_norm": 0.5927900075912476, "learning_rate": 0.0002, "epoch": 3.65334329473291, "step": 4890}, {"loss": 1.6084, "grad_norm": 0.5336901545524597, "learning_rate": 0.0002, "epoch": 3.66081434441539, "step": 4900}, {"loss": 1.5701, "grad_norm": 0.7823320627212524, "learning_rate": 0.0002, "epoch": 3.668285394097871, "step": 4910}, {"loss": 1.4881, "grad_norm": 0.6703504323959351, "learning_rate": 0.0002, "epoch": 3.675756443780351, "step": 4920}, {"loss": 1.5332, "grad_norm": 0.6061160564422607, "learning_rate": 0.0002, "epoch": 3.6832274934628315, "step": 4930}, {"loss": 1.5405, "grad_norm": 0.6237227916717529, "learning_rate": 0.0002, "epoch": 3.690698543145312, "step": 4940}, {"loss": 1.497, "grad_norm": 0.5985278487205505, "learning_rate": 0.0002, "epoch": 3.6981695928277922, "step": 4950}, {"loss": 1.5132, "grad_norm": 0.6483839750289917, "learning_rate": 0.0002, "epoch": 3.705640642510273, "step": 4960}, {"loss": 1.5338, "grad_norm": 0.5788805484771729, "learning_rate": 0.0002, "epoch": 3.713111692192753, "step": 4970}, {"loss": 1.5258, "grad_norm": 0.5609974265098572, "learning_rate": 0.0002, "epoch": 3.7205827418752335, "step": 4980}, {"loss": 1.4759, "grad_norm": 0.5681300759315491, "learning_rate": 0.0002, "epoch": 3.7280537915577137, "step": 4990}, {"loss": 1.6018, "grad_norm": 0.5860186219215393, "learning_rate": 0.0002, "epoch": 3.7355248412401942, "step": 5000}, {"loss": 1.58, "grad_norm": 0.5718157291412354, "learning_rate": 0.0002, "epoch": 3.742995890922675, "step": 5010}, {"loss": 1.5834, "grad_norm": 0.6173721551895142, "learning_rate": 0.0002, "epoch": 3.750466940605155, "step": 5020}, {"loss": 1.5617, "grad_norm": 0.629152238368988, "learning_rate": 0.0002, "epoch": 3.7579379902876355, "step": 5030}, {"loss": 1.519, "grad_norm": 0.5666284561157227, "learning_rate": 0.0002, "epoch": 3.7654090399701157, "step": 5040}, {"loss": 1.5329, "grad_norm": 0.6053005456924438, "learning_rate": 0.0002, "epoch": 3.7728800896525962, "step": 5050}, {"loss": 1.5404, "grad_norm": 0.5870583057403564, "learning_rate": 0.0002, "epoch": 3.7803511393350764, "step": 5060}, {"loss": 1.4444, "grad_norm": 0.5422009229660034, "learning_rate": 0.0002, "epoch": 3.787822189017557, "step": 5070}, {"loss": 1.5308, "grad_norm": 0.5396918058395386, "learning_rate": 0.0002, "epoch": 3.7952932387000375, "step": 5080}, {"loss": 1.464, "grad_norm": 0.5544713139533997, "learning_rate": 0.0002, "epoch": 3.8027642883825177, "step": 5090}, {"loss": 1.4752, "grad_norm": 0.5983749628067017, "learning_rate": 0.0002, "epoch": 3.8102353380649983, "step": 5100}, {"loss": 1.4972, "grad_norm": 0.5702024102210999, "learning_rate": 0.0002, "epoch": 3.8177063877474784, "step": 5110}, {"loss": 1.5471, "grad_norm": 0.5436882376670837, "learning_rate": 0.0002, "epoch": 3.825177437429959, "step": 5120}, {"loss": 1.5118, "grad_norm": 0.5453617572784424, "learning_rate": 0.0002, "epoch": 3.832648487112439, "step": 5130}, {"loss": 1.5732, "grad_norm": 0.6269069314002991, "learning_rate": 0.0002, "epoch": 3.8401195367949197, "step": 5140}, {"loss": 1.4959, "grad_norm": 0.6189185380935669, "learning_rate": 0.0002, "epoch": 3.8475905864774003, "step": 5150}, {"loss": 1.4999, "grad_norm": 0.6653388142585754, "learning_rate": 0.0002, "epoch": 3.8550616361598804, "step": 5160}, {"loss": 1.5075, "grad_norm": 0.5771768689155579, "learning_rate": 0.0002, "epoch": 3.862532685842361, "step": 5170}, {"loss": 1.5545, "grad_norm": 0.6052790880203247, "learning_rate": 0.0002, "epoch": 3.870003735524841, "step": 5180}, {"loss": 1.4987, "grad_norm": 0.6572316884994507, "learning_rate": 0.0002, "epoch": 3.8774747852073217, "step": 5190}, {"loss": 1.5241, "grad_norm": 0.670576810836792, "learning_rate": 0.0002, "epoch": 3.884945834889802, "step": 5200}, {"loss": 1.4777, "grad_norm": 0.5728798508644104, "learning_rate": 0.0002, "epoch": 3.8924168845722824, "step": 5210}, {"loss": 1.5351, "grad_norm": 0.6340774297714233, "learning_rate": 0.0002, "epoch": 3.899887934254763, "step": 5220}, {"loss": 1.5081, "grad_norm": 0.5981315970420837, "learning_rate": 0.0002, "epoch": 3.907358983937243, "step": 5230}, {"loss": 1.4875, "grad_norm": 0.6212025880813599, "learning_rate": 0.0002, "epoch": 3.9148300336197237, "step": 5240}, {"loss": 1.5545, "grad_norm": 0.6202296018600464, "learning_rate": 0.0002, "epoch": 3.922301083302204, "step": 5250}, {"loss": 1.5765, "grad_norm": 0.6159142255783081, "learning_rate": 0.0002, "epoch": 3.9297721329846844, "step": 5260}, {"loss": 1.4938, "grad_norm": 0.6519438624382019, "learning_rate": 0.0002, "epoch": 3.9372431826671646, "step": 5270}, {"loss": 1.4859, "grad_norm": 0.539813756942749, "learning_rate": 0.0002, "epoch": 3.944714232349645, "step": 5280}, {"loss": 1.5921, "grad_norm": 0.6443665027618408, "learning_rate": 0.0002, "epoch": 3.9521852820321257, "step": 5290}, {"loss": 1.5153, "grad_norm": 0.6635757684707642, "learning_rate": 0.0002, "epoch": 3.959656331714606, "step": 5300}, {"loss": 1.5485, "grad_norm": 0.589363157749176, "learning_rate": 0.0002, "epoch": 3.9671273813970864, "step": 5310}, {"loss": 1.5498, "grad_norm": 0.5788735747337341, "learning_rate": 0.0002, "epoch": 3.9745984310795666, "step": 5320}, {"loss": 1.5607, "grad_norm": 0.5976864695549011, "learning_rate": 0.0002, "epoch": 3.982069480762047, "step": 5330}, {"loss": 1.5302, "grad_norm": 0.6624067425727844, "learning_rate": 0.0002, "epoch": 3.9895405304445273, "step": 5340}, {"loss": 1.5904, "grad_norm": 0.6738956570625305, "learning_rate": 0.0002, "epoch": 3.997011580127008, "step": 5350}, {"eval_loss": 1.868006944656372, "eval_runtime": 38.5153, "eval_samples_per_second": 13.371, "eval_steps_per_second": 1.688, "epoch": 4.0, "step": 5354}, {"loss": 1.4535, "grad_norm": 0.6023468971252441, "learning_rate": 0.0002, "epoch": 4.004482629809488, "step": 5360}, {"loss": 1.3987, "grad_norm": 0.8589285612106323, "learning_rate": 0.0002, "epoch": 4.011953679491969, "step": 5370}, {"loss": 1.3952, "grad_norm": 0.7477491497993469, "learning_rate": 0.0002, "epoch": 4.019424729174449, "step": 5380}, {"loss": 1.3745, "grad_norm": 0.7601922154426575, "learning_rate": 0.0002, "epoch": 4.02689577885693, "step": 5390}, {"loss": 1.4133, "grad_norm": 0.8115614056587219, "learning_rate": 0.0002, "epoch": 4.03436682853941, "step": 5400}, {"loss": 1.3748, "grad_norm": 0.669925332069397, "learning_rate": 0.0002, "epoch": 4.04183787822189, "step": 5410}, {"loss": 1.2835, "grad_norm": 0.8091904520988464, "learning_rate": 0.0002, "epoch": 4.04930892790437, "step": 5420}, {"loss": 1.3615, "grad_norm": 0.709405779838562, "learning_rate": 0.0002, "epoch": 4.056779977586851, "step": 5430}, {"loss": 1.3558, "grad_norm": 1.0006179809570312, "learning_rate": 0.0002, "epoch": 4.064251027269331, "step": 5440}, {"loss": 1.3491, "grad_norm": 0.7017965912818909, "learning_rate": 0.0002, "epoch": 4.071722076951811, "step": 5450}, {"loss": 1.3642, "grad_norm": 0.8991572260856628, "learning_rate": 0.0002, "epoch": 4.0791931266342925, "step": 5460}, {"loss": 1.392, "grad_norm": 0.9064797759056091, "learning_rate": 0.0002, "epoch": 4.086664176316773, "step": 5470}, {"loss": 1.3425, "grad_norm": 0.7981749176979065, "learning_rate": 0.0002, "epoch": 4.094135225999253, "step": 5480}, {"loss": 1.3826, "grad_norm": 0.7280883193016052, "learning_rate": 0.0002, "epoch": 4.101606275681733, "step": 5490}, {"loss": 1.3275, "grad_norm": 0.7419600486755371, "learning_rate": 0.0002, "epoch": 4.109077325364214, "step": 5500}, {"loss": 1.3199, "grad_norm": 0.8019949197769165, "learning_rate": 0.0002, "epoch": 4.116548375046694, "step": 5510}, {"loss": 1.3133, "grad_norm": 0.7501229047775269, "learning_rate": 0.0002, "epoch": 4.124019424729174, "step": 5520}, {"loss": 1.4432, "grad_norm": 0.8166249990463257, "learning_rate": 0.0002, "epoch": 4.131490474411655, "step": 5530}, {"loss": 1.3901, "grad_norm": 0.9728496074676514, "learning_rate": 0.0002, "epoch": 4.138961524094135, "step": 5540}, {"loss": 1.3538, "grad_norm": 0.7590922117233276, "learning_rate": 0.0002, "epoch": 4.1464325737766154, "step": 5550}, {"loss": 1.4368, "grad_norm": 0.7759010791778564, "learning_rate": 0.0002, "epoch": 4.153903623459096, "step": 5560}, {"loss": 1.3635, "grad_norm": 0.9057986736297607, "learning_rate": 0.0002, "epoch": 4.161374673141577, "step": 5570}, {"loss": 1.4152, "grad_norm": 0.8853937983512878, "learning_rate": 0.0002, "epoch": 4.168845722824057, "step": 5580}, {"loss": 1.3633, "grad_norm": 0.7070684432983398, "learning_rate": 0.0002, "epoch": 4.176316772506537, "step": 5590}, {"loss": 1.3218, "grad_norm": 0.7649410963058472, "learning_rate": 0.0002, "epoch": 4.183787822189018, "step": 5600}, {"loss": 1.3857, "grad_norm": 1.2048029899597168, "learning_rate": 0.0002, "epoch": 4.191258871871498, "step": 5610}, {"loss": 1.3629, "grad_norm": 0.7986605763435364, "learning_rate": 0.0002, "epoch": 4.198729921553978, "step": 5620}, {"loss": 1.3995, "grad_norm": 0.8151885867118835, "learning_rate": 0.0002, "epoch": 4.206200971236458, "step": 5630}, {"loss": 1.3782, "grad_norm": 0.7719064354896545, "learning_rate": 0.0002, "epoch": 4.213672020918939, "step": 5640}, {"loss": 1.3852, "grad_norm": 0.8422448039054871, "learning_rate": 0.0002, "epoch": 4.2211430706014195, "step": 5650}, {"loss": 1.3321, "grad_norm": 0.7017164826393127, "learning_rate": 0.0002, "epoch": 4.2286141202839, "step": 5660}, {"loss": 1.4105, "grad_norm": 0.8559677600860596, "learning_rate": 0.0002, "epoch": 4.236085169966381, "step": 5670}, {"loss": 1.3701, "grad_norm": 0.8216157555580139, "learning_rate": 0.0002, "epoch": 4.243556219648861, "step": 5680}, {"loss": 1.3565, "grad_norm": 0.7681755423545837, "learning_rate": 0.0002, "epoch": 4.251027269331341, "step": 5690}, {"loss": 1.3806, "grad_norm": 0.811665952205658, "learning_rate": 0.0002, "epoch": 4.258498319013821, "step": 5700}, {"loss": 1.4161, "grad_norm": 0.7242204546928406, "learning_rate": 0.0002, "epoch": 4.265969368696302, "step": 5710}, {"loss": 1.2958, "grad_norm": 0.7570181488990784, "learning_rate": 0.0002, "epoch": 4.273440418378782, "step": 5720}, {"loss": 1.4265, "grad_norm": 0.8951969146728516, "learning_rate": 0.0002, "epoch": 4.280911468061262, "step": 5730}, {"loss": 1.3895, "grad_norm": 0.7222902178764343, "learning_rate": 0.0002, "epoch": 4.288382517743743, "step": 5740}, {"loss": 1.4155, "grad_norm": 0.8508469462394714, "learning_rate": 0.0002, "epoch": 4.2958535674262235, "step": 5750}, {"loss": 1.365, "grad_norm": 0.7215430736541748, "learning_rate": 0.0002, "epoch": 4.303324617108704, "step": 5760}, {"loss": 1.4472, "grad_norm": 0.8774884939193726, "learning_rate": 0.0002, "epoch": 4.310795666791184, "step": 5770}, {"loss": 1.427, "grad_norm": 0.8354552984237671, "learning_rate": 0.0002, "epoch": 4.318266716473665, "step": 5780}, {"loss": 1.3222, "grad_norm": 0.6938814520835876, "learning_rate": 0.0002, "epoch": 4.325737766156145, "step": 5790}, {"loss": 1.3589, "grad_norm": 0.78675377368927, "learning_rate": 0.0002, "epoch": 4.333208815838625, "step": 5800}, {"loss": 1.3662, "grad_norm": 0.7147697806358337, "learning_rate": 0.0002, "epoch": 4.340679865521106, "step": 5810}, {"loss": 1.3597, "grad_norm": 0.7693623304367065, "learning_rate": 0.0002, "epoch": 4.348150915203586, "step": 5820}, {"loss": 1.2944, "grad_norm": 0.856517493724823, "learning_rate": 0.0002, "epoch": 4.355621964886066, "step": 5830}, {"loss": 1.4307, "grad_norm": 0.7200973033905029, "learning_rate": 0.0002, "epoch": 4.3630930145685465, "step": 5840}, {"loss": 1.442, "grad_norm": 0.743281364440918, "learning_rate": 0.0002, "epoch": 4.3705640642510275, "step": 5850}, {"loss": 1.3999, "grad_norm": 0.7627727389335632, "learning_rate": 0.0002, "epoch": 4.378035113933508, "step": 5860}, {"loss": 1.4082, "grad_norm": 0.7238836884498596, "learning_rate": 0.0002, "epoch": 4.385506163615988, "step": 5870}, {"loss": 1.4292, "grad_norm": 0.7253410816192627, "learning_rate": 0.0002, "epoch": 4.392977213298469, "step": 5880}, {"loss": 1.3774, "grad_norm": 0.8232238292694092, "learning_rate": 0.0002, "epoch": 4.400448262980949, "step": 5890}, {"loss": 1.3757, "grad_norm": 0.8778504729270935, "learning_rate": 0.0002, "epoch": 4.407919312663429, "step": 5900}, {"loss": 1.387, "grad_norm": 0.7639474868774414, "learning_rate": 0.0002, "epoch": 4.415390362345909, "step": 5910}, {"loss": 1.3862, "grad_norm": 0.7666519284248352, "learning_rate": 0.0002, "epoch": 4.42286141202839, "step": 5920}, {"loss": 1.4168, "grad_norm": 0.867132842540741, "learning_rate": 0.0002, "epoch": 4.43033246171087, "step": 5930}, {"loss": 1.4772, "grad_norm": 0.7571166753768921, "learning_rate": 0.0002, "epoch": 4.4378035113933505, "step": 5940}, {"loss": 1.4401, "grad_norm": 0.7911370992660522, "learning_rate": 0.0002, "epoch": 4.4452745610758315, "step": 5950}, {"loss": 1.4516, "grad_norm": 0.8844250440597534, "learning_rate": 0.0002, "epoch": 4.452745610758312, "step": 5960}, {"loss": 1.4109, "grad_norm": 0.7336231470108032, "learning_rate": 0.0002, "epoch": 4.460216660440792, "step": 5970}, {"loss": 1.3891, "grad_norm": 0.8162738084793091, "learning_rate": 0.0002, "epoch": 4.467687710123272, "step": 5980}, {"loss": 1.393, "grad_norm": 0.7413017153739929, "learning_rate": 0.0002, "epoch": 4.475158759805753, "step": 5990}, {"loss": 1.3712, "grad_norm": 0.7215432524681091, "learning_rate": 0.0002, "epoch": 4.482629809488233, "step": 6000}, {"loss": 1.3521, "grad_norm": 0.8943389058113098, "learning_rate": 0.0002, "epoch": 4.490100859170713, "step": 6010}, {"loss": 1.4172, "grad_norm": 0.7850823998451233, "learning_rate": 0.0002, "epoch": 4.497571908853194, "step": 6020}, {"loss": 1.3582, "grad_norm": 0.8117504119873047, "learning_rate": 0.0002, "epoch": 4.505042958535674, "step": 6030}, {"loss": 1.4272, "grad_norm": 0.8381605744361877, "learning_rate": 0.0002, "epoch": 4.5125140082181545, "step": 6040}, {"loss": 1.3829, "grad_norm": 0.7964059710502625, "learning_rate": 0.0002, "epoch": 4.519985057900635, "step": 6050}, {"loss": 1.3555, "grad_norm": 0.7935128211975098, "learning_rate": 0.0002, "epoch": 4.527456107583116, "step": 6060}, {"loss": 1.3994, "grad_norm": 0.8725124597549438, "learning_rate": 0.0002, "epoch": 4.534927157265596, "step": 6070}, {"loss": 1.3923, "grad_norm": 0.880325198173523, "learning_rate": 0.0002, "epoch": 4.542398206948076, "step": 6080}, {"loss": 1.4459, "grad_norm": 0.7220637202262878, "learning_rate": 0.0002, "epoch": 4.549869256630557, "step": 6090}, {"loss": 1.3281, "grad_norm": 0.6908547878265381, "learning_rate": 0.0002, "epoch": 4.557340306313037, "step": 6100}, {"loss": 1.437, "grad_norm": 0.797931969165802, "learning_rate": 0.0002, "epoch": 4.564811355995517, "step": 6110}, {"loss": 1.4023, "grad_norm": 0.7056134343147278, "learning_rate": 0.0002, "epoch": 4.572282405677997, "step": 6120}, {"loss": 1.3814, "grad_norm": 0.7850478887557983, "learning_rate": 0.0002, "epoch": 4.579753455360478, "step": 6130}, {"loss": 1.3579, "grad_norm": 0.8112621307373047, "learning_rate": 0.0002, "epoch": 4.5872245050429585, "step": 6140}, {"loss": 1.3523, "grad_norm": 0.7040849328041077, "learning_rate": 0.0002, "epoch": 4.594695554725439, "step": 6150}, {"loss": 1.3526, "grad_norm": 0.7214553952217102, "learning_rate": 0.0002, "epoch": 4.60216660440792, "step": 6160}, {"loss": 1.3932, "grad_norm": 0.8616511821746826, "learning_rate": 0.0002, "epoch": 4.6096376540904, "step": 6170}, {"loss": 1.4622, "grad_norm": 0.8374658226966858, "learning_rate": 0.0002, "epoch": 4.61710870377288, "step": 6180}, {"loss": 1.3703, "grad_norm": 0.6761606931686401, "learning_rate": 0.0002, "epoch": 4.62457975345536, "step": 6190}, {"loss": 1.3977, "grad_norm": 0.768028199672699, "learning_rate": 0.0002, "epoch": 4.632050803137841, "step": 6200}, {"loss": 1.3772, "grad_norm": 0.9372717142105103, "learning_rate": 0.0002, "epoch": 4.639521852820321, "step": 6210}, {"loss": 1.4098, "grad_norm": 0.7906546592712402, "learning_rate": 0.0002, "epoch": 4.646992902502801, "step": 6220}, {"loss": 1.3962, "grad_norm": 0.7376723289489746, "learning_rate": 0.0002, "epoch": 4.654463952185282, "step": 6230}, {"loss": 1.4529, "grad_norm": 0.8972630500793457, "learning_rate": 0.0002, "epoch": 4.6619350018677626, "step": 6240}, {"loss": 1.4668, "grad_norm": 0.8261756300926208, "learning_rate": 0.0002, "epoch": 4.669406051550243, "step": 6250}, {"loss": 1.3267, "grad_norm": 0.7512393593788147, "learning_rate": 0.0002, "epoch": 4.676877101232723, "step": 6260}, {"loss": 1.4278, "grad_norm": 0.7132362127304077, "learning_rate": 0.0002, "epoch": 4.684348150915204, "step": 6270}, {"loss": 1.4299, "grad_norm": 0.7690575122833252, "learning_rate": 0.0002, "epoch": 4.691819200597684, "step": 6280}, {"loss": 1.4769, "grad_norm": 0.9886258840560913, "learning_rate": 0.0002, "epoch": 4.699290250280164, "step": 6290}, {"loss": 1.4005, "grad_norm": 0.9502435922622681, "learning_rate": 0.0002, "epoch": 4.706761299962645, "step": 6300}, {"loss": 1.4319, "grad_norm": 0.702255129814148, "learning_rate": 0.0002, "epoch": 4.714232349645125, "step": 6310}, {"loss": 1.4447, "grad_norm": 0.7713103890419006, "learning_rate": 0.0002, "epoch": 4.721703399327605, "step": 6320}, {"loss": 1.4392, "grad_norm": 0.7778580784797668, "learning_rate": 0.0002, "epoch": 4.7291744490100855, "step": 6330}, {"loss": 1.4169, "grad_norm": 0.7275111079216003, "learning_rate": 0.0002, "epoch": 4.736645498692567, "step": 6340}, {"loss": 1.4429, "grad_norm": 0.7728744149208069, "learning_rate": 0.0002, "epoch": 4.744116548375047, "step": 6350}, {"loss": 1.3756, "grad_norm": 0.9724260568618774, "learning_rate": 0.0002, "epoch": 4.751587598057527, "step": 6360}, {"loss": 1.3358, "grad_norm": 0.7505622506141663, "learning_rate": 0.0002, "epoch": 4.759058647740007, "step": 6370}, {"loss": 1.379, "grad_norm": 0.7994682788848877, "learning_rate": 0.0002, "epoch": 4.766529697422488, "step": 6380}, {"loss": 1.4275, "grad_norm": 0.8432038426399231, "learning_rate": 0.0002, "epoch": 4.774000747104968, "step": 6390}, {"loss": 1.4606, "grad_norm": 0.7436022758483887, "learning_rate": 0.0002, "epoch": 4.781471796787448, "step": 6400}, {"loss": 1.3461, "grad_norm": 0.7709194421768188, "learning_rate": 0.0002, "epoch": 4.788942846469929, "step": 6410}, {"loss": 1.3715, "grad_norm": 0.8798436522483826, "learning_rate": 0.0002, "epoch": 4.796413896152409, "step": 6420}, {"loss": 1.3761, "grad_norm": 0.790189266204834, "learning_rate": 0.0002, "epoch": 4.80388494583489, "step": 6430}, {"loss": 1.4109, "grad_norm": 0.6824303865432739, "learning_rate": 0.0002, "epoch": 4.811355995517371, "step": 6440}, {"loss": 1.3877, "grad_norm": 0.7501044869422913, "learning_rate": 0.0002, "epoch": 4.818827045199851, "step": 6450}, {"loss": 1.4458, "grad_norm": 0.8840398192405701, "learning_rate": 0.0002, "epoch": 4.826298094882331, "step": 6460}, {"loss": 1.4412, "grad_norm": 0.7812688946723938, "learning_rate": 0.0002, "epoch": 4.833769144564811, "step": 6470}, {"loss": 1.4299, "grad_norm": 0.7429926991462708, "learning_rate": 0.0002, "epoch": 4.841240194247292, "step": 6480}, {"loss": 1.5062, "grad_norm": 0.7778021693229675, "learning_rate": 0.0002, "epoch": 4.848711243929772, "step": 6490}, {"loss": 1.4589, "grad_norm": 0.8270702362060547, "learning_rate": 0.0002, "epoch": 4.856182293612252, "step": 6500}, {"loss": 1.4091, "grad_norm": 0.6960513591766357, "learning_rate": 0.0002, "epoch": 4.863653343294732, "step": 6510}, {"loss": 1.376, "grad_norm": 0.7728942632675171, "learning_rate": 0.0002, "epoch": 4.8711243929772134, "step": 6520}, {"loss": 1.4852, "grad_norm": 0.7377303838729858, "learning_rate": 0.0002, "epoch": 4.878595442659694, "step": 6530}, {"loss": 1.3846, "grad_norm": 0.7257253527641296, "learning_rate": 0.0002, "epoch": 4.886066492342174, "step": 6540}, {"loss": 1.4166, "grad_norm": 0.7875821590423584, "learning_rate": 0.0002, "epoch": 4.893537542024655, "step": 6550}, {"loss": 1.357, "grad_norm": 0.8346304297447205, "learning_rate": 0.0002, "epoch": 4.901008591707135, "step": 6560}, {"loss": 1.4522, "grad_norm": 0.7710739374160767, "learning_rate": 0.0002, "epoch": 4.908479641389615, "step": 6570}, {"loss": 1.4465, "grad_norm": 0.7015138268470764, "learning_rate": 0.0002, "epoch": 4.915950691072096, "step": 6580}, {"loss": 1.435, "grad_norm": 0.8707432150840759, "learning_rate": 0.0002, "epoch": 4.923421740754576, "step": 6590}, {"loss": 1.2968, "grad_norm": 0.786601185798645, "learning_rate": 0.0002, "epoch": 4.930892790437056, "step": 6600}, {"loss": 1.4385, "grad_norm": 0.978519082069397, "learning_rate": 0.0002, "epoch": 4.938363840119536, "step": 6610}, {"loss": 1.3997, "grad_norm": 0.8102927207946777, "learning_rate": 0.0002, "epoch": 4.9458348898020175, "step": 6620}, {"loss": 1.4859, "grad_norm": 0.7628704309463501, "learning_rate": 0.0002, "epoch": 4.953305939484498, "step": 6630}, {"loss": 1.3774, "grad_norm": 0.8053455352783203, "learning_rate": 0.0002, "epoch": 4.960776989166978, "step": 6640}, {"loss": 1.5092, "grad_norm": 0.8680412173271179, "learning_rate": 0.0002, "epoch": 4.968248038849458, "step": 6650}, {"loss": 1.3978, "grad_norm": 0.7415758371353149, "learning_rate": 0.0002, "epoch": 4.975719088531939, "step": 6660}, {"loss": 1.3793, "grad_norm": 0.7730312347412109, "learning_rate": 0.0002, "epoch": 4.983190138214419, "step": 6670}, {"loss": 1.4863, "grad_norm": 0.7924041152000427, "learning_rate": 0.0002, "epoch": 4.990661187896899, "step": 6680}, {"loss": 1.4137, "grad_norm": 0.8677893877029419, "learning_rate": 0.0002, "epoch": 4.99813223757938, "step": 6690}, {"eval_loss": 1.9444633722305298, "eval_runtime": 39.3488, "eval_samples_per_second": 13.088, "eval_steps_per_second": 1.652, "epoch": 4.999626447515876, "step": 6692}, {"loss": 1.3076, "grad_norm": 0.7102245092391968, "learning_rate": 0.0002, "epoch": 5.00560328726186, "step": 6700}, {"loss": 1.2714, "grad_norm": 1.0425463914871216, "learning_rate": 0.0002, "epoch": 5.0130743369443405, "step": 6710}, {"loss": 1.181, "grad_norm": 0.9320756793022156, "learning_rate": 0.0002, "epoch": 5.0205453866268215, "step": 6720}, {"loss": 1.1786, "grad_norm": 0.8797217607498169, "learning_rate": 0.0002, "epoch": 5.028016436309302, "step": 6730}, {"loss": 1.2097, "grad_norm": 2.135707139968872, "learning_rate": 0.0002, "epoch": 5.035487485991782, "step": 6740}, {"loss": 1.1761, "grad_norm": 0.8747734427452087, "learning_rate": 0.0002, "epoch": 5.042958535674262, "step": 6750}, {"loss": 1.1675, "grad_norm": 0.9981076717376709, "learning_rate": 0.0002, "epoch": 5.050429585356743, "step": 6760}, {"loss": 1.1976, "grad_norm": 0.985078752040863, "learning_rate": 0.0002, "epoch": 5.057900635039223, "step": 6770}, {"loss": 1.2688, "grad_norm": 1.0974019765853882, "learning_rate": 0.0002, "epoch": 5.065371684721703, "step": 6780}, {"loss": 1.1982, "grad_norm": 0.9823219180107117, "learning_rate": 0.0002, "epoch": 5.072842734404184, "step": 6790}, {"loss": 1.2586, "grad_norm": 1.122605562210083, "learning_rate": 0.0002, "epoch": 5.080313784086664, "step": 6800}, {"loss": 1.2069, "grad_norm": 0.8556802272796631, "learning_rate": 0.0002, "epoch": 5.0877848337691445, "step": 6810}, {"loss": 1.1908, "grad_norm": 1.1699262857437134, "learning_rate": 0.0002, "epoch": 5.095255883451625, "step": 6820}, {"loss": 1.1869, "grad_norm": 1.0440590381622314, "learning_rate": 0.0002, "epoch": 5.102726933134106, "step": 6830}, {"loss": 1.1655, "grad_norm": 1.0445300340652466, "learning_rate": 0.0002, "epoch": 5.110197982816586, "step": 6840}, {"loss": 1.2392, "grad_norm": 0.8289563059806824, "learning_rate": 0.0002, "epoch": 5.117669032499066, "step": 6850}, {"loss": 1.1687, "grad_norm": 1.1051193475723267, "learning_rate": 0.0002, "epoch": 5.125140082181547, "step": 6860}, {"loss": 1.2737, "grad_norm": 0.9345614910125732, "learning_rate": 0.0002, "epoch": 5.132611131864027, "step": 6870}, {"loss": 1.3021, "grad_norm": 1.1222996711730957, "learning_rate": 0.0002, "epoch": 5.140082181546507, "step": 6880}, {"loss": 1.2408, "grad_norm": 0.9405338764190674, "learning_rate": 0.0002, "epoch": 5.147553231228987, "step": 6890}, {"loss": 1.2367, "grad_norm": 1.0935171842575073, "learning_rate": 0.0002, "epoch": 5.155024280911468, "step": 6900}, {"loss": 1.2458, "grad_norm": 1.0438612699508667, "learning_rate": 0.0002, "epoch": 5.1624953305939485, "step": 6910}, {"loss": 1.2562, "grad_norm": 1.1189004182815552, "learning_rate": 0.0002, "epoch": 5.169966380276429, "step": 6920}, {"loss": 1.25, "grad_norm": 1.0533215999603271, "learning_rate": 0.0002, "epoch": 5.17743742995891, "step": 6930}, {"loss": 1.2974, "grad_norm": 0.9779648780822754, "learning_rate": 0.0002, "epoch": 5.18490847964139, "step": 6940}, {"loss": 1.1965, "grad_norm": 0.8920868635177612, "learning_rate": 0.0002, "epoch": 5.19237952932387, "step": 6950}, {"loss": 1.283, "grad_norm": 0.8374548554420471, "learning_rate": 0.0002, "epoch": 5.19985057900635, "step": 6960}, {"loss": 1.2775, "grad_norm": 1.0490682125091553, "learning_rate": 0.0002, "epoch": 5.207321628688831, "step": 6970}, {"loss": 1.1826, "grad_norm": 0.9658287167549133, "learning_rate": 0.0002, "epoch": 5.214792678371311, "step": 6980}, {"loss": 1.2647, "grad_norm": 0.9652056097984314, "learning_rate": 0.0002, "epoch": 5.222263728053791, "step": 6990}, {"loss": 1.3023, "grad_norm": 0.9141794443130493, "learning_rate": 0.0002, "epoch": 5.229734777736272, "step": 7000}, {"loss": 1.2456, "grad_norm": 0.9831376671791077, "learning_rate": 0.0002, "epoch": 5.2372058274187525, "step": 7010}, {"loss": 1.2176, "grad_norm": 1.0198718309402466, "learning_rate": 0.0002, "epoch": 5.244676877101233, "step": 7020}, {"loss": 1.2643, "grad_norm": 0.9647888541221619, "learning_rate": 0.0002, "epoch": 5.252147926783713, "step": 7030}, {"loss": 1.2106, "grad_norm": 1.3941649198532104, "learning_rate": 0.0002, "epoch": 5.259618976466194, "step": 7040}, {"loss": 1.2885, "grad_norm": 1.0305466651916504, "learning_rate": 0.0002, "epoch": 5.267090026148674, "step": 7050}, {"loss": 1.2362, "grad_norm": 0.9577859044075012, "learning_rate": 0.0002, "epoch": 5.274561075831154, "step": 7060}, {"loss": 1.2231, "grad_norm": 1.149092197418213, "learning_rate": 0.0002, "epoch": 5.282032125513634, "step": 7070}, {"loss": 1.2986, "grad_norm": 1.2582733631134033, "learning_rate": 0.0002, "epoch": 5.289503175196115, "step": 7080}, {"loss": 1.2307, "grad_norm": 1.1777442693710327, "learning_rate": 0.0002, "epoch": 5.296974224878595, "step": 7090}, {"loss": 1.24, "grad_norm": 1.0076404809951782, "learning_rate": 0.0002, "epoch": 5.3044452745610755, "step": 7100}, {"loss": 1.1407, "grad_norm": 0.9037365913391113, "learning_rate": 0.0002, "epoch": 5.3119163242435565, "step": 7110}, {"loss": 1.238, "grad_norm": 0.9428724646568298, "learning_rate": 0.0002, "epoch": 5.319387373926037, "step": 7120}, {"loss": 1.2571, "grad_norm": 0.9935154318809509, "learning_rate": 0.0002, "epoch": 5.326858423608517, "step": 7130}, {"loss": 1.2833, "grad_norm": 1.087500810623169, "learning_rate": 0.0002, "epoch": 5.334329473290998, "step": 7140}, {"loss": 1.2304, "grad_norm": 0.8543072938919067, "learning_rate": 0.0002, "epoch": 5.341800522973478, "step": 7150}, {"loss": 1.2755, "grad_norm": 0.9323700070381165, "learning_rate": 0.0002, "epoch": 5.349271572655958, "step": 7160}, {"loss": 1.2769, "grad_norm": 1.0037827491760254, "learning_rate": 0.0002, "epoch": 5.356742622338438, "step": 7170}, {"loss": 1.3204, "grad_norm": 0.8746469616889954, "learning_rate": 0.0002, "epoch": 5.364213672020919, "step": 7180}, {"loss": 1.2759, "grad_norm": 0.9516328573226929, "learning_rate": 0.0002, "epoch": 5.371684721703399, "step": 7190}, {"loss": 1.2428, "grad_norm": 0.9395177364349365, "learning_rate": 0.0002, "epoch": 5.3791557713858795, "step": 7200}, {"loss": 1.3214, "grad_norm": 1.000369906425476, "learning_rate": 0.0002, "epoch": 5.38662682106836, "step": 7210}, {"loss": 1.2337, "grad_norm": 1.0845502614974976, "learning_rate": 0.0002, "epoch": 5.394097870750841, "step": 7220}, {"loss": 1.2776, "grad_norm": 0.8975145220756531, "learning_rate": 0.0002, "epoch": 5.401568920433321, "step": 7230}, {"loss": 1.2306, "grad_norm": 1.040077805519104, "learning_rate": 0.0002, "epoch": 5.409039970115801, "step": 7240}, {"loss": 1.2277, "grad_norm": 1.0729942321777344, "learning_rate": 0.0002, "epoch": 5.416511019798282, "step": 7250}, {"loss": 1.2714, "grad_norm": 0.8322232961654663, "learning_rate": 0.0002, "epoch": 5.423982069480762, "step": 7260}, {"loss": 1.3036, "grad_norm": 1.0654641389846802, "learning_rate": 0.0002, "epoch": 5.431453119163242, "step": 7270}, {"loss": 1.268, "grad_norm": 1.0445852279663086, "learning_rate": 0.0002, "epoch": 5.438924168845723, "step": 7280}, {"loss": 1.2743, "grad_norm": 1.0762956142425537, "learning_rate": 0.0002, "epoch": 5.446395218528203, "step": 7290}, {"loss": 1.2887, "grad_norm": 0.9721953868865967, "learning_rate": 0.0002, "epoch": 5.4538662682106835, "step": 7300}, {"loss": 1.2833, "grad_norm": 0.9238539338111877, "learning_rate": 0.0002, "epoch": 5.461337317893164, "step": 7310}, {"loss": 1.255, "grad_norm": 0.9912874102592468, "learning_rate": 0.0002, "epoch": 5.468808367575645, "step": 7320}, {"loss": 1.2557, "grad_norm": 1.0727077722549438, "learning_rate": 0.0002, "epoch": 5.476279417258125, "step": 7330}, {"loss": 1.3471, "grad_norm": 0.8633865118026733, "learning_rate": 0.0002, "epoch": 5.483750466940605, "step": 7340}, {"loss": 1.3155, "grad_norm": 0.9396262764930725, "learning_rate": 0.0002, "epoch": 5.491221516623085, "step": 7350}, {"loss": 1.3146, "grad_norm": 1.0253715515136719, "learning_rate": 0.0002, "epoch": 5.498692566305566, "step": 7360}, {"loss": 1.3156, "grad_norm": 1.006047010421753, "learning_rate": 0.0002, "epoch": 5.506163615988046, "step": 7370}, {"loss": 1.3107, "grad_norm": 0.9781233072280884, "learning_rate": 0.0002, "epoch": 5.513634665670526, "step": 7380}, {"loss": 1.2703, "grad_norm": 0.9945126175880432, "learning_rate": 0.0002, "epoch": 5.521105715353007, "step": 7390}, {"loss": 1.1936, "grad_norm": 0.9081175327301025, "learning_rate": 0.0002, "epoch": 5.528576765035488, "step": 7400}, {"loss": 1.2651, "grad_norm": 1.2215938568115234, "learning_rate": 0.0002, "epoch": 5.536047814717968, "step": 7410}, {"loss": 1.2484, "grad_norm": 1.0724077224731445, "learning_rate": 0.0002, "epoch": 5.543518864400449, "step": 7420}, {"loss": 1.3083, "grad_norm": 1.106955885887146, "learning_rate": 0.0002, "epoch": 5.550989914082929, "step": 7430}, {"loss": 1.2125, "grad_norm": 1.0657650232315063, "learning_rate": 0.0002, "epoch": 5.558460963765409, "step": 7440}, {"loss": 1.2576, "grad_norm": 0.9725455641746521, "learning_rate": 0.0002, "epoch": 5.565932013447889, "step": 7450}, {"loss": 1.3297, "grad_norm": 0.8604224324226379, "learning_rate": 0.0002, "epoch": 5.57340306313037, "step": 7460}, {"loss": 1.3084, "grad_norm": 0.9913371205329895, "learning_rate": 0.0002, "epoch": 5.58087411281285, "step": 7470}, {"loss": 1.3371, "grad_norm": 1.012073040008545, "learning_rate": 0.0002, "epoch": 5.58834516249533, "step": 7480}, {"loss": 1.2526, "grad_norm": 1.1003159284591675, "learning_rate": 0.0002, "epoch": 5.5958162121778106, "step": 7490}, {"loss": 1.2577, "grad_norm": 0.9104593992233276, "learning_rate": 0.0002, "epoch": 5.603287261860292, "step": 7500}, {"loss": 1.2578, "grad_norm": 0.9480831623077393, "learning_rate": 0.0002, "epoch": 5.610758311542772, "step": 7510}, {"loss": 1.3056, "grad_norm": 1.0826456546783447, "learning_rate": 0.0002, "epoch": 5.618229361225252, "step": 7520}, {"loss": 1.2931, "grad_norm": 0.8286259174346924, "learning_rate": 0.0002, "epoch": 5.625700410907733, "step": 7530}, {"loss": 1.2918, "grad_norm": 0.9145061373710632, "learning_rate": 0.0002, "epoch": 5.633171460590213, "step": 7540}, {"loss": 1.1736, "grad_norm": 0.9363601803779602, "learning_rate": 0.0002, "epoch": 5.640642510272693, "step": 7550}, {"loss": 1.2265, "grad_norm": 0.9553244709968567, "learning_rate": 0.0002, "epoch": 5.648113559955174, "step": 7560}, {"loss": 1.2356, "grad_norm": 1.0343557596206665, "learning_rate": 0.0002, "epoch": 5.655584609637654, "step": 7570}, {"loss": 1.3171, "grad_norm": 0.8734238743782043, "learning_rate": 0.0002, "epoch": 5.663055659320134, "step": 7580}, {"loss": 1.2785, "grad_norm": 1.0230586528778076, "learning_rate": 0.0002, "epoch": 5.670526709002615, "step": 7590}, {"loss": 1.2936, "grad_norm": 1.0063409805297852, "learning_rate": 0.0002, "epoch": 5.677997758685096, "step": 7600}, {"loss": 1.2396, "grad_norm": 1.0104626417160034, "learning_rate": 0.0002, "epoch": 5.685468808367576, "step": 7610}, {"loss": 1.2581, "grad_norm": 0.9528168439865112, "learning_rate": 0.0002, "epoch": 5.692939858050056, "step": 7620}, {"loss": 1.3116, "grad_norm": 0.9799878597259521, "learning_rate": 0.0002, "epoch": 5.700410907732536, "step": 7630}, {"loss": 1.2632, "grad_norm": 0.969351589679718, "learning_rate": 0.0002, "epoch": 5.707881957415017, "step": 7640}, {"loss": 1.3055, "grad_norm": 1.3037652969360352, "learning_rate": 0.0002, "epoch": 5.715353007097497, "step": 7650}, {"loss": 1.3126, "grad_norm": 1.0640486478805542, "learning_rate": 0.0002, "epoch": 5.722824056779977, "step": 7660}, {"loss": 1.3325, "grad_norm": 1.0416420698165894, "learning_rate": 0.0002, "epoch": 5.730295106462458, "step": 7670}, {"loss": 1.25, "grad_norm": 0.8893619775772095, "learning_rate": 0.0002, "epoch": 5.7377661561449385, "step": 7680}, {"loss": 1.319, "grad_norm": 0.8512844443321228, "learning_rate": 0.0002, "epoch": 5.745237205827419, "step": 7690}, {"loss": 1.3328, "grad_norm": 0.9955748319625854, "learning_rate": 0.0002, "epoch": 5.7527082555099, "step": 7700}, {"loss": 1.294, "grad_norm": 1.0409910678863525, "learning_rate": 0.0002, "epoch": 5.76017930519238, "step": 7710}, {"loss": 1.3518, "grad_norm": 1.010097861289978, "learning_rate": 0.0002, "epoch": 5.76765035487486, "step": 7720}, {"loss": 1.2106, "grad_norm": 0.8974892497062683, "learning_rate": 0.0002, "epoch": 5.77512140455734, "step": 7730}, {"loss": 1.2743, "grad_norm": 0.972835123538971, "learning_rate": 0.0002, "epoch": 5.782592454239821, "step": 7740}, {"loss": 1.3549, "grad_norm": 0.9607440829277039, "learning_rate": 0.0002, "epoch": 5.790063503922301, "step": 7750}, {"loss": 1.29, "grad_norm": 0.9426500797271729, "learning_rate": 0.0002, "epoch": 5.797534553604781, "step": 7760}, {"loss": 1.274, "grad_norm": 0.8745320439338684, "learning_rate": 0.0002, "epoch": 5.8050056032872615, "step": 7770}, {"loss": 1.3009, "grad_norm": 1.0117204189300537, "learning_rate": 0.0002, "epoch": 5.8124766529697425, "step": 7780}, {"loss": 1.3135, "grad_norm": 1.0387755632400513, "learning_rate": 0.0002, "epoch": 5.819947702652223, "step": 7790}, {"loss": 1.2709, "grad_norm": 1.0709784030914307, "learning_rate": 0.0002, "epoch": 5.827418752334703, "step": 7800}, {"loss": 1.225, "grad_norm": 0.9512667655944824, "learning_rate": 0.0002, "epoch": 5.834889802017184, "step": 7810}, {"loss": 1.3284, "grad_norm": 1.021094560623169, "learning_rate": 0.0002, "epoch": 5.842360851699664, "step": 7820}, {"loss": 1.2794, "grad_norm": 1.117491364479065, "learning_rate": 0.0002, "epoch": 5.849831901382144, "step": 7830}, {"loss": 1.3646, "grad_norm": 0.9252554178237915, "learning_rate": 0.0002, "epoch": 5.857302951064625, "step": 7840}, {"loss": 1.2976, "grad_norm": 1.1416207551956177, "learning_rate": 0.0002, "epoch": 5.864774000747105, "step": 7850}, {"loss": 1.3293, "grad_norm": 1.1219907999038696, "learning_rate": 0.0002, "epoch": 5.872245050429585, "step": 7860}, {"loss": 1.2334, "grad_norm": 0.8300467729568481, "learning_rate": 0.0002, "epoch": 5.8797161001120655, "step": 7870}, {"loss": 1.3132, "grad_norm": 1.00551438331604, "learning_rate": 0.0002, "epoch": 5.8871871497945465, "step": 7880}, {"loss": 1.2609, "grad_norm": 0.8981153964996338, "learning_rate": 0.0002, "epoch": 5.894658199477027, "step": 7890}, {"loss": 1.2817, "grad_norm": 1.0247976779937744, "learning_rate": 0.0002, "epoch": 5.902129249159507, "step": 7900}, {"loss": 1.2866, "grad_norm": 1.0820319652557373, "learning_rate": 0.0002, "epoch": 5.909600298841987, "step": 7910}, {"loss": 1.2941, "grad_norm": 0.952675461769104, "learning_rate": 0.0002, "epoch": 5.917071348524468, "step": 7920}, {"loss": 1.307, "grad_norm": 0.8666740655899048, "learning_rate": 0.0002, "epoch": 5.924542398206948, "step": 7930}, {"loss": 1.2752, "grad_norm": 0.8640421032905579, "learning_rate": 0.0002, "epoch": 5.932013447889428, "step": 7940}, {"loss": 1.2386, "grad_norm": 1.2343276739120483, "learning_rate": 0.0002, "epoch": 5.939484497571909, "step": 7950}, {"loss": 1.2333, "grad_norm": 0.958046555519104, "learning_rate": 0.0002, "epoch": 5.946955547254389, "step": 7960}, {"loss": 1.2352, "grad_norm": 1.0538510084152222, "learning_rate": 0.0002, "epoch": 5.9544265969368695, "step": 7970}, {"loss": 1.3233, "grad_norm": 1.2681571245193481, "learning_rate": 0.0002, "epoch": 5.9618976466193505, "step": 7980}, {"loss": 1.2514, "grad_norm": 0.8171183466911316, "learning_rate": 0.0002, "epoch": 5.969368696301831, "step": 7990}, {"loss": 1.3412, "grad_norm": 0.9109523892402649, "learning_rate": 0.0002, "epoch": 5.976839745984311, "step": 8000}, {"loss": 1.3497, "grad_norm": 1.0040639638900757, "learning_rate": 0.0002, "epoch": 5.984310795666791, "step": 8010}, {"loss": 1.3299, "grad_norm": 0.9596554040908813, "learning_rate": 0.0002, "epoch": 5.991781845349272, "step": 8020}, {"loss": 1.3109, "grad_norm": 0.9782963991165161, "learning_rate": 0.0002, "epoch": 5.999252895031752, "step": 8030}]}
+{"epoch": 6.999626447515876, "step": 9369, "epoch_duration": 1874.6490201950073, "total_accumulated_duration": 11295.583930015564, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7672.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-42/checkpoint-2677", "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": 2.6181, "grad_norm": 0.4912872612476349, "learning_rate": 0.0002, "epoch": 0.007471049682480389, "step": 10}, {"loss": 2.2606, "grad_norm": 0.4856316149234772, "learning_rate": 0.0002, "epoch": 0.014942099364960777, "step": 20}, {"loss": 2.0957, "grad_norm": 0.47683125734329224, "learning_rate": 0.0002, "epoch": 0.022413149047441166, "step": 30}, {"loss": 1.8908, "grad_norm": 0.515082597732544, "learning_rate": 0.0002, "epoch": 0.029884198729921554, "step": 40}, {"loss": 1.9704, "grad_norm": 0.5299215316772461, "learning_rate": 0.0002, "epoch": 0.03735524841240194, "step": 50}, {"loss": 1.9225, "grad_norm": 0.4951399862766266, "learning_rate": 0.0002, "epoch": 0.04482629809488233, "step": 60}, {"loss": 1.9742, "grad_norm": 0.48079821467399597, "learning_rate": 0.0002, "epoch": 0.05229734777736272, "step": 70}, {"loss": 1.9466, "grad_norm": 0.49402132630348206, "learning_rate": 0.0002, "epoch": 0.05976839745984311, "step": 80}, {"loss": 1.8691, "grad_norm": 0.4778193235397339, "learning_rate": 0.0002, "epoch": 0.0672394471423235, "step": 90}, {"loss": 1.8455, "grad_norm": 0.42472657561302185, "learning_rate": 0.0002, "epoch": 0.07471049682480388, "step": 100}, {"loss": 1.8744, "grad_norm": 0.4433092474937439, "learning_rate": 0.0002, "epoch": 0.08218154650728428, "step": 110}, {"loss": 1.865, "grad_norm": 0.4472862780094147, "learning_rate": 0.0002, "epoch": 0.08965259618976466, "step": 120}, {"loss": 1.9256, "grad_norm": 0.42596298456192017, "learning_rate": 0.0002, "epoch": 0.09712364587224505, "step": 130}, {"loss": 1.8015, "grad_norm": 0.46645811200141907, "learning_rate": 0.0002, "epoch": 0.10459469555472543, "step": 140}, {"loss": 1.8307, "grad_norm": 0.41041234135627747, "learning_rate": 0.0002, "epoch": 0.11206574523720583, "step": 150}, {"loss": 1.8276, "grad_norm": 0.5329819917678833, "learning_rate": 0.0002, "epoch": 0.11953679491968622, "step": 160}, {"loss": 1.8118, "grad_norm": 0.4065922200679779, "learning_rate": 0.0002, "epoch": 0.1270078446021666, "step": 170}, {"loss": 1.8559, "grad_norm": 0.38406994938850403, "learning_rate": 0.0002, "epoch": 0.134478894284647, "step": 180}, {"loss": 1.8647, "grad_norm": 0.4246881306171417, "learning_rate": 0.0002, "epoch": 0.14194994396712737, "step": 190}, {"loss": 1.8054, "grad_norm": 0.35136649012565613, "learning_rate": 0.0002, "epoch": 0.14942099364960776, "step": 200}, {"loss": 1.802, "grad_norm": 0.43252742290496826, "learning_rate": 0.0002, "epoch": 0.15689204333208817, "step": 210}, {"loss": 1.7823, "grad_norm": 0.39236941933631897, "learning_rate": 0.0002, "epoch": 0.16436309301456856, "step": 220}, {"loss": 1.818, "grad_norm": 0.3748249113559723, "learning_rate": 0.0002, "epoch": 0.17183414269704894, "step": 230}, {"loss": 1.866, "grad_norm": 0.6432855725288391, "learning_rate": 0.0002, "epoch": 0.17930519237952933, "step": 240}, {"loss": 1.8397, "grad_norm": 0.34874802827835083, "learning_rate": 0.0002, "epoch": 0.1867762420620097, "step": 250}, {"loss": 1.79, "grad_norm": 0.3721984326839447, "learning_rate": 0.0002, "epoch": 0.1942472917444901, "step": 260}, {"loss": 1.8464, "grad_norm": 0.4339311420917511, "learning_rate": 0.0002, "epoch": 0.20171834142697048, "step": 270}, {"loss": 1.8665, "grad_norm": 0.4018215537071228, "learning_rate": 0.0002, "epoch": 0.20918939110945087, "step": 280}, {"loss": 1.8048, "grad_norm": 0.3278839886188507, "learning_rate": 0.0002, "epoch": 0.21666044079193125, "step": 290}, {"loss": 1.7395, "grad_norm": 0.36146077513694763, "learning_rate": 0.0002, "epoch": 0.22413149047441167, "step": 300}, {"loss": 1.7916, "grad_norm": 0.38175010681152344, "learning_rate": 0.0002, "epoch": 0.23160254015689205, "step": 310}, {"loss": 1.8593, "grad_norm": 0.44776618480682373, "learning_rate": 0.0002, "epoch": 0.23907358983937244, "step": 320}, {"loss": 1.7824, "grad_norm": 0.3933652937412262, "learning_rate": 0.0002, "epoch": 0.24654463952185282, "step": 330}, {"loss": 1.8393, "grad_norm": 0.3515005111694336, "learning_rate": 0.0002, "epoch": 0.2540156892043332, "step": 340}, {"loss": 1.8653, "grad_norm": 0.6683304309844971, "learning_rate": 0.0002, "epoch": 0.2614867388868136, "step": 350}, {"loss": 1.8797, "grad_norm": 0.37093454599380493, "learning_rate": 0.0002, "epoch": 0.268957788569294, "step": 360}, {"loss": 1.8251, "grad_norm": 0.3450651168823242, "learning_rate": 0.0002, "epoch": 0.2764288382517744, "step": 370}, {"loss": 1.7435, "grad_norm": 0.5140917301177979, "learning_rate": 0.0002, "epoch": 0.28389988793425475, "step": 380}, {"loss": 1.8026, "grad_norm": 0.32885563373565674, "learning_rate": 0.0002, "epoch": 0.29137093761673516, "step": 390}, {"loss": 1.8174, "grad_norm": 0.33962297439575195, "learning_rate": 0.0002, "epoch": 0.2988419872992155, "step": 400}, {"loss": 1.7467, "grad_norm": 0.3723141849040985, "learning_rate": 0.0002, "epoch": 0.30631303698169593, "step": 410}, {"loss": 1.8459, "grad_norm": 0.37173134088516235, "learning_rate": 0.0002, "epoch": 0.31378408666417634, "step": 420}, {"loss": 1.8876, "grad_norm": 0.33736956119537354, "learning_rate": 0.0002, "epoch": 0.3212551363466567, "step": 430}, {"loss": 1.8367, "grad_norm": 0.3602448105812073, "learning_rate": 0.0002, "epoch": 0.3287261860291371, "step": 440}, {"loss": 1.8058, "grad_norm": 0.3569699227809906, "learning_rate": 0.0002, "epoch": 0.33619723571161747, "step": 450}, {"loss": 1.8086, "grad_norm": 0.31009167432785034, "learning_rate": 0.0002, "epoch": 0.3436682853940979, "step": 460}, {"loss": 1.8876, "grad_norm": 0.5278693437576294, "learning_rate": 0.0002, "epoch": 0.35113933507657824, "step": 470}, {"loss": 1.8534, "grad_norm": 0.3587537109851837, "learning_rate": 0.0002, "epoch": 0.35861038475905865, "step": 480}, {"loss": 1.8046, "grad_norm": 0.3859670162200928, "learning_rate": 0.0002, "epoch": 0.366081434441539, "step": 490}, {"loss": 1.8287, "grad_norm": 0.395913690328598, "learning_rate": 0.0002, "epoch": 0.3735524841240194, "step": 500}, {"loss": 1.7619, "grad_norm": 0.35052940249443054, "learning_rate": 0.0002, "epoch": 0.38102353380649984, "step": 510}, {"loss": 1.7824, "grad_norm": 0.2979494333267212, "learning_rate": 0.0002, "epoch": 0.3884945834889802, "step": 520}, {"loss": 1.8641, "grad_norm": 0.3062683343887329, "learning_rate": 0.0002, "epoch": 0.3959656331714606, "step": 530}, {"loss": 1.7651, "grad_norm": 0.3172847330570221, "learning_rate": 0.0002, "epoch": 0.40343668285394096, "step": 540}, {"loss": 1.806, "grad_norm": 0.360435426235199, "learning_rate": 0.0002, "epoch": 0.4109077325364214, "step": 550}, {"loss": 1.9054, "grad_norm": 0.3427872359752655, "learning_rate": 0.0002, "epoch": 0.41837878221890173, "step": 560}, {"loss": 1.7562, "grad_norm": 0.34036558866500854, "learning_rate": 0.0002, "epoch": 0.42584983190138215, "step": 570}, {"loss": 1.7254, "grad_norm": 0.3365345299243927, "learning_rate": 0.0002, "epoch": 0.4333208815838625, "step": 580}, {"loss": 1.8328, "grad_norm": 0.35619041323661804, "learning_rate": 0.0002, "epoch": 0.4407919312663429, "step": 590}, {"loss": 1.8114, "grad_norm": 0.3569088280200958, "learning_rate": 0.0002, "epoch": 0.44826298094882333, "step": 600}, {"loss": 1.8599, "grad_norm": 0.3581278622150421, "learning_rate": 0.0002, "epoch": 0.4557340306313037, "step": 610}, {"loss": 1.7078, "grad_norm": 0.43197110295295715, "learning_rate": 0.0002, "epoch": 0.4632050803137841, "step": 620}, {"loss": 1.8257, "grad_norm": 0.33966198563575745, "learning_rate": 0.0002, "epoch": 0.47067612999626446, "step": 630}, {"loss": 1.7528, "grad_norm": 0.3343866467475891, "learning_rate": 0.0002, "epoch": 0.47814717967874487, "step": 640}, {"loss": 1.8191, "grad_norm": 0.33878564834594727, "learning_rate": 0.0002, "epoch": 0.48561822936122523, "step": 650}, {"loss": 1.8801, "grad_norm": 0.387195885181427, "learning_rate": 0.0002, "epoch": 0.49308927904370564, "step": 660}, {"loss": 1.7559, "grad_norm": 0.3755440413951874, "learning_rate": 0.0002, "epoch": 0.500560328726186, "step": 670}, {"loss": 1.8057, "grad_norm": 0.3272816836833954, "learning_rate": 0.0002, "epoch": 0.5080313784086664, "step": 680}, {"loss": 1.8156, "grad_norm": 0.36063864827156067, "learning_rate": 0.0002, "epoch": 0.5155024280911468, "step": 690}, {"loss": 1.8397, "grad_norm": 0.35317373275756836, "learning_rate": 0.0002, "epoch": 0.5229734777736272, "step": 700}, {"loss": 1.7603, "grad_norm": 0.3561195433139801, "learning_rate": 0.0002, "epoch": 0.5304445274561076, "step": 710}, {"loss": 1.8149, "grad_norm": 0.31124624609947205, "learning_rate": 0.0002, "epoch": 0.537915577138588, "step": 720}, {"loss": 1.7434, "grad_norm": 0.3294544517993927, "learning_rate": 0.0002, "epoch": 0.5453866268210683, "step": 730}, {"loss": 1.8027, "grad_norm": 0.31933900713920593, "learning_rate": 0.0002, "epoch": 0.5528576765035488, "step": 740}, {"loss": 1.7601, "grad_norm": 0.3226020634174347, "learning_rate": 0.0002, "epoch": 0.5603287261860291, "step": 750}, {"loss": 1.7862, "grad_norm": 0.3147525489330292, "learning_rate": 0.0002, "epoch": 0.5677997758685095, "step": 760}, {"loss": 1.9028, "grad_norm": 0.32234328985214233, "learning_rate": 0.0002, "epoch": 0.57527082555099, "step": 770}, {"loss": 1.7623, "grad_norm": 0.3258664309978485, "learning_rate": 0.0002, "epoch": 0.5827418752334703, "step": 780}, {"loss": 1.7384, "grad_norm": 0.3166961967945099, "learning_rate": 0.0002, "epoch": 0.5902129249159507, "step": 790}, {"loss": 1.8799, "grad_norm": 0.35621458292007446, "learning_rate": 0.0002, "epoch": 0.597683974598431, "step": 800}, {"loss": 1.8313, "grad_norm": 0.3236999213695526, "learning_rate": 0.0002, "epoch": 0.6051550242809115, "step": 810}, {"loss": 1.7132, "grad_norm": 0.2892923653125763, "learning_rate": 0.0002, "epoch": 0.6126260739633919, "step": 820}, {"loss": 1.8709, "grad_norm": 0.4098321497440338, "learning_rate": 0.0002, "epoch": 0.6200971236458722, "step": 830}, {"loss": 1.7637, "grad_norm": 0.3337118923664093, "learning_rate": 0.0002, "epoch": 0.6275681733283527, "step": 840}, {"loss": 1.7375, "grad_norm": 0.30416029691696167, "learning_rate": 0.0002, "epoch": 0.635039223010833, "step": 850}, {"loss": 1.7419, "grad_norm": 0.3361026346683502, "learning_rate": 0.0002, "epoch": 0.6425102726933134, "step": 860}, {"loss": 1.732, "grad_norm": 0.3537365198135376, "learning_rate": 0.0002, "epoch": 0.6499813223757938, "step": 870}, {"loss": 1.7825, "grad_norm": 0.33854469656944275, "learning_rate": 0.0002, "epoch": 0.6574523720582742, "step": 880}, {"loss": 1.7561, "grad_norm": 0.3332272469997406, "learning_rate": 0.0002, "epoch": 0.6649234217407546, "step": 890}, {"loss": 1.7247, "grad_norm": 0.34954726696014404, "learning_rate": 0.0002, "epoch": 0.6723944714232349, "step": 900}, {"loss": 1.7917, "grad_norm": 0.2921750247478485, "learning_rate": 0.0002, "epoch": 0.6798655211057153, "step": 910}, {"loss": 1.7807, "grad_norm": 0.30508682131767273, "learning_rate": 0.0002, "epoch": 0.6873365707881958, "step": 920}, {"loss": 1.8082, "grad_norm": 0.32268425822257996, "learning_rate": 0.0002, "epoch": 0.6948076204706761, "step": 930}, {"loss": 1.8283, "grad_norm": 0.2844390869140625, "learning_rate": 0.0002, "epoch": 0.7022786701531565, "step": 940}, {"loss": 1.7363, "grad_norm": 0.31263890862464905, "learning_rate": 0.0002, "epoch": 0.709749719835637, "step": 950}, {"loss": 1.8081, "grad_norm": 0.3626808822154999, "learning_rate": 0.0002, "epoch": 0.7172207695181173, "step": 960}, {"loss": 1.853, "grad_norm": 0.3322749733924866, "learning_rate": 0.0002, "epoch": 0.7246918192005977, "step": 970}, {"loss": 1.7912, "grad_norm": 0.29177871346473694, "learning_rate": 0.0002, "epoch": 0.732162868883078, "step": 980}, {"loss": 1.8447, "grad_norm": 0.35405513644218445, "learning_rate": 0.0002, "epoch": 0.7396339185655585, "step": 990}, {"loss": 1.7008, "grad_norm": 0.39318400621414185, "learning_rate": 0.0002, "epoch": 0.7471049682480388, "step": 1000}, {"loss": 1.7803, "grad_norm": 0.29401418566703796, "learning_rate": 0.0002, "epoch": 0.7545760179305192, "step": 1010}, {"loss": 1.7649, "grad_norm": 0.3271748721599579, "learning_rate": 0.0002, "epoch": 0.7620470676129997, "step": 1020}, {"loss": 1.7266, "grad_norm": 0.30883970856666565, "learning_rate": 0.0002, "epoch": 0.76951811729548, "step": 1030}, {"loss": 1.7722, "grad_norm": 0.3411838412284851, "learning_rate": 0.0002, "epoch": 0.7769891669779604, "step": 1040}, {"loss": 1.829, "grad_norm": 0.30608129501342773, "learning_rate": 0.0002, "epoch": 0.7844602166604407, "step": 1050}, {"loss": 1.7815, "grad_norm": 0.30899080634117126, "learning_rate": 0.0002, "epoch": 0.7919312663429212, "step": 1060}, {"loss": 1.7625, "grad_norm": 0.3160453140735626, "learning_rate": 0.0002, "epoch": 0.7994023160254016, "step": 1070}, {"loss": 1.8452, "grad_norm": 0.30947187542915344, "learning_rate": 0.0002, "epoch": 0.8068733657078819, "step": 1080}, {"loss": 1.7418, "grad_norm": 0.3103134036064148, "learning_rate": 0.0002, "epoch": 0.8143444153903624, "step": 1090}, {"loss": 1.842, "grad_norm": 0.31771138310432434, "learning_rate": 0.0002, "epoch": 0.8218154650728428, "step": 1100}, {"loss": 1.7918, "grad_norm": 0.5860997438430786, "learning_rate": 0.0002, "epoch": 0.8292865147553231, "step": 1110}, {"loss": 1.8443, "grad_norm": 0.3230148255825043, "learning_rate": 0.0002, "epoch": 0.8367575644378035, "step": 1120}, {"loss": 1.8478, "grad_norm": 0.29611510038375854, "learning_rate": 0.0002, "epoch": 0.8442286141202839, "step": 1130}, {"loss": 1.7673, "grad_norm": 0.3373654782772064, "learning_rate": 0.0002, "epoch": 0.8516996638027643, "step": 1140}, {"loss": 1.7997, "grad_norm": 0.3474279046058655, "learning_rate": 0.0002, "epoch": 0.8591707134852447, "step": 1150}, {"loss": 1.75, "grad_norm": 0.35057875514030457, "learning_rate": 0.0002, "epoch": 0.866641763167725, "step": 1160}, {"loss": 1.8273, "grad_norm": 0.39537495374679565, "learning_rate": 0.0002, "epoch": 0.8741128128502055, "step": 1170}, {"loss": 1.7682, "grad_norm": 0.3714233636856079, "learning_rate": 0.0002, "epoch": 0.8815838625326858, "step": 1180}, {"loss": 1.7549, "grad_norm": 0.2950296998023987, "learning_rate": 0.0002, "epoch": 0.8890549122151662, "step": 1190}, {"loss": 1.7612, "grad_norm": 0.38182979822158813, "learning_rate": 0.0002, "epoch": 0.8965259618976467, "step": 1200}, {"loss": 1.827, "grad_norm": 0.27883678674697876, "learning_rate": 0.0002, "epoch": 0.903997011580127, "step": 1210}, {"loss": 1.7623, "grad_norm": 0.33874374628067017, "learning_rate": 0.0002, "epoch": 0.9114680612626074, "step": 1220}, {"loss": 1.7334, "grad_norm": 0.3014272153377533, "learning_rate": 0.0002, "epoch": 0.9189391109450877, "step": 1230}, {"loss": 1.8235, "grad_norm": 0.3194271922111511, "learning_rate": 0.0002, "epoch": 0.9264101606275682, "step": 1240}, {"loss": 1.7924, "grad_norm": 0.3049403429031372, "learning_rate": 0.0002, "epoch": 0.9338812103100486, "step": 1250}, {"loss": 1.7535, "grad_norm": 0.30621254444122314, "learning_rate": 0.0002, "epoch": 0.9413522599925289, "step": 1260}, {"loss": 1.8287, "grad_norm": 0.28675132989883423, "learning_rate": 0.0002, "epoch": 0.9488233096750094, "step": 1270}, {"loss": 1.7586, "grad_norm": 0.3322032690048218, "learning_rate": 0.0002, "epoch": 0.9562943593574897, "step": 1280}, {"loss": 1.8054, "grad_norm": 0.35408294200897217, "learning_rate": 0.0002, "epoch": 0.9637654090399701, "step": 1290}, {"loss": 1.7343, "grad_norm": 0.36386919021606445, "learning_rate": 0.0002, "epoch": 0.9712364587224505, "step": 1300}, {"loss": 1.8633, "grad_norm": 0.32338324189186096, "learning_rate": 0.0002, "epoch": 0.9787075084049309, "step": 1310}, {"loss": 1.7724, "grad_norm": 0.3714013993740082, "learning_rate": 0.0002, "epoch": 0.9861785580874113, "step": 1320}, {"loss": 1.7766, "grad_norm": 0.3133082389831543, "learning_rate": 0.0002, "epoch": 0.9936496077698916, "step": 1330}, {"eval_loss": 1.8051470518112183, "eval_runtime": 38.6332, "eval_samples_per_second": 13.331, "eval_steps_per_second": 1.682, "epoch": 0.9996264475158759, "step": 1338}, {"loss": 1.8035, "grad_norm": 0.31595754623413086, "learning_rate": 0.0002, "epoch": 1.001120657452372, "step": 1340}, {"loss": 1.7486, "grad_norm": 0.3095700144767761, "learning_rate": 0.0002, "epoch": 1.0085917071348525, "step": 1350}, {"loss": 1.6981, "grad_norm": 0.34677496552467346, "learning_rate": 0.0002, "epoch": 1.0160627568173328, "step": 1360}, {"loss": 1.7377, "grad_norm": 0.29108840227127075, "learning_rate": 0.0002, "epoch": 1.0235338064998132, "step": 1370}, {"loss": 1.7194, "grad_norm": 0.32356950640678406, "learning_rate": 0.0002, "epoch": 1.0310048561822935, "step": 1380}, {"loss": 1.7593, "grad_norm": 0.4200669229030609, "learning_rate": 0.0002, "epoch": 1.038475905864774, "step": 1390}, {"loss": 1.797, "grad_norm": 0.3283711373806, "learning_rate": 0.0002, "epoch": 1.0459469555472545, "step": 1400}, {"loss": 1.7163, "grad_norm": 0.32898256182670593, "learning_rate": 0.0002, "epoch": 1.0534180052297348, "step": 1410}, {"loss": 1.7559, "grad_norm": 0.38790300488471985, "learning_rate": 0.0002, "epoch": 1.0608890549122152, "step": 1420}, {"loss": 1.6922, "grad_norm": 0.339800089597702, "learning_rate": 0.0002, "epoch": 1.0683601045946955, "step": 1430}, {"loss": 1.7076, "grad_norm": 0.3548751175403595, "learning_rate": 0.0002, "epoch": 1.075831154277176, "step": 1440}, {"loss": 1.6985, "grad_norm": 0.35114359855651855, "learning_rate": 0.0002, "epoch": 1.0833022039596563, "step": 1450}, {"loss": 1.7217, "grad_norm": 0.35226720571517944, "learning_rate": 0.0002, "epoch": 1.0907732536421366, "step": 1460}, {"loss": 1.6822, "grad_norm": 0.33665576577186584, "learning_rate": 0.0002, "epoch": 1.0982443033246172, "step": 1470}, {"loss": 1.6699, "grad_norm": 0.363889217376709, "learning_rate": 0.0002, "epoch": 1.1057153530070976, "step": 1480}, {"loss": 1.7933, "grad_norm": 0.3826201856136322, "learning_rate": 0.0002, "epoch": 1.113186402689578, "step": 1490}, {"loss": 1.7022, "grad_norm": 0.34058740735054016, "learning_rate": 0.0002, "epoch": 1.1206574523720583, "step": 1500}, {"loss": 1.6375, "grad_norm": 0.3462134301662445, "learning_rate": 0.0002, "epoch": 1.1281285020545386, "step": 1510}, {"loss": 1.7147, "grad_norm": 0.3396756052970886, "learning_rate": 0.0002, "epoch": 1.135599551737019, "step": 1520}, {"loss": 1.7219, "grad_norm": 0.32004743814468384, "learning_rate": 0.0002, "epoch": 1.1430706014194993, "step": 1530}, {"loss": 1.743, "grad_norm": 0.3397733271121979, "learning_rate": 0.0002, "epoch": 1.15054165110198, "step": 1540}, {"loss": 1.7333, "grad_norm": 0.3783262073993683, "learning_rate": 0.0002, "epoch": 1.1580127007844603, "step": 1550}, {"loss": 1.6075, "grad_norm": 0.35121291875839233, "learning_rate": 0.0002, "epoch": 1.1654837504669406, "step": 1560}, {"loss": 1.678, "grad_norm": 0.35816895961761475, "learning_rate": 0.0002, "epoch": 1.172954800149421, "step": 1570}, {"loss": 1.7143, "grad_norm": 0.33843839168548584, "learning_rate": 0.0002, "epoch": 1.1804258498319014, "step": 1580}, {"loss": 1.7434, "grad_norm": 0.3371972143650055, "learning_rate": 0.0002, "epoch": 1.1878968995143817, "step": 1590}, {"loss": 1.7671, "grad_norm": 0.36016878485679626, "learning_rate": 0.0002, "epoch": 1.195367949196862, "step": 1600}, {"loss": 1.6914, "grad_norm": 0.40879473090171814, "learning_rate": 0.0002, "epoch": 1.2028389988793426, "step": 1610}, {"loss": 1.6955, "grad_norm": 0.3216715455055237, "learning_rate": 0.0002, "epoch": 1.210310048561823, "step": 1620}, {"loss": 1.632, "grad_norm": 0.4482610821723938, "learning_rate": 0.0002, "epoch": 1.2177810982443034, "step": 1630}, {"loss": 1.6999, "grad_norm": 0.3257700502872467, "learning_rate": 0.0002, "epoch": 1.2252521479267837, "step": 1640}, {"loss": 1.7177, "grad_norm": 0.38646459579467773, "learning_rate": 0.0002, "epoch": 1.232723197609264, "step": 1650}, {"loss": 1.7081, "grad_norm": 0.4081360697746277, "learning_rate": 0.0002, "epoch": 1.2401942472917444, "step": 1660}, {"loss": 1.7519, "grad_norm": 0.4326848089694977, "learning_rate": 0.0002, "epoch": 1.2476652969742248, "step": 1670}, {"loss": 1.6752, "grad_norm": 0.346401572227478, "learning_rate": 0.0002, "epoch": 1.2551363466567054, "step": 1680}, {"loss": 1.7425, "grad_norm": 0.34536251425743103, "learning_rate": 0.0002, "epoch": 1.2626073963391857, "step": 1690}, {"loss": 1.7061, "grad_norm": 0.41359591484069824, "learning_rate": 0.0002, "epoch": 1.270078446021666, "step": 1700}, {"loss": 1.7906, "grad_norm": 0.3530874252319336, "learning_rate": 0.0002, "epoch": 1.2775494957041464, "step": 1710}, {"loss": 1.7357, "grad_norm": 0.3702719211578369, "learning_rate": 0.0002, "epoch": 1.2850205453866268, "step": 1720}, {"loss": 1.766, "grad_norm": 0.3703329563140869, "learning_rate": 0.0002, "epoch": 1.2924915950691072, "step": 1730}, {"loss": 1.7221, "grad_norm": 0.37919729948043823, "learning_rate": 0.0002, "epoch": 1.2999626447515875, "step": 1740}, {"loss": 1.7859, "grad_norm": 0.32526856660842896, "learning_rate": 0.0002, "epoch": 1.307433694434068, "step": 1750}, {"loss": 1.7117, "grad_norm": 0.36752620339393616, "learning_rate": 0.0002, "epoch": 1.3149047441165485, "step": 1760}, {"loss": 1.7335, "grad_norm": 0.3398192524909973, "learning_rate": 0.0002, "epoch": 1.3223757937990288, "step": 1770}, {"loss": 1.7492, "grad_norm": 0.37435585260391235, "learning_rate": 0.0002, "epoch": 1.3298468434815092, "step": 1780}, {"loss": 1.7393, "grad_norm": 0.35793280601501465, "learning_rate": 0.0002, "epoch": 1.3373178931639895, "step": 1790}, {"loss": 1.7266, "grad_norm": 0.35481882095336914, "learning_rate": 0.0002, "epoch": 1.3447889428464699, "step": 1800}, {"loss": 1.7456, "grad_norm": 0.3786393105983734, "learning_rate": 0.0002, "epoch": 1.3522599925289502, "step": 1810}, {"loss": 1.7169, "grad_norm": 0.33245593309402466, "learning_rate": 0.0002, "epoch": 1.3597310422114308, "step": 1820}, {"loss": 1.7577, "grad_norm": 0.35388344526290894, "learning_rate": 0.0002, "epoch": 1.3672020918939112, "step": 1830}, {"loss": 1.6968, "grad_norm": 0.3695325553417206, "learning_rate": 0.0002, "epoch": 1.3746731415763915, "step": 1840}, {"loss": 1.7086, "grad_norm": 0.3683604598045349, "learning_rate": 0.0002, "epoch": 1.382144191258872, "step": 1850}, {"loss": 1.7878, "grad_norm": 0.3753012418746948, "learning_rate": 0.0002, "epoch": 1.3896152409413522, "step": 1860}, {"loss": 1.6969, "grad_norm": 0.3331069350242615, "learning_rate": 0.0002, "epoch": 1.3970862906238326, "step": 1870}, {"loss": 1.6644, "grad_norm": 0.3877500295639038, "learning_rate": 0.0002, "epoch": 1.404557340306313, "step": 1880}, {"loss": 1.7586, "grad_norm": 0.33525151014328003, "learning_rate": 0.0002, "epoch": 1.4120283899887935, "step": 1890}, {"loss": 1.7031, "grad_norm": 0.3697299659252167, "learning_rate": 0.0002, "epoch": 1.4194994396712737, "step": 1900}, {"loss": 1.6956, "grad_norm": 0.4029286205768585, "learning_rate": 0.0002, "epoch": 1.4269704893537543, "step": 1910}, {"loss": 1.6897, "grad_norm": 0.3596203029155731, "learning_rate": 0.0002, "epoch": 1.4344415390362346, "step": 1920}, {"loss": 1.7139, "grad_norm": 0.450783908367157, "learning_rate": 0.0002, "epoch": 1.441912588718715, "step": 1930}, {"loss": 1.7243, "grad_norm": 0.3651481866836548, "learning_rate": 0.0002, "epoch": 1.4493836384011953, "step": 1940}, {"loss": 1.6637, "grad_norm": 0.3608424663543701, "learning_rate": 0.0002, "epoch": 1.4568546880836757, "step": 1950}, {"loss": 1.8285, "grad_norm": 0.39684420824050903, "learning_rate": 0.0002, "epoch": 1.4643257377661563, "step": 1960}, {"loss": 1.7514, "grad_norm": 0.34618663787841797, "learning_rate": 0.0002, "epoch": 1.4717967874486364, "step": 1970}, {"loss": 1.6655, "grad_norm": 0.4150386452674866, "learning_rate": 0.0002, "epoch": 1.479267837131117, "step": 1980}, {"loss": 1.7021, "grad_norm": 0.35500776767730713, "learning_rate": 0.0002, "epoch": 1.4867388868135973, "step": 1990}, {"loss": 1.7322, "grad_norm": 0.344144344329834, "learning_rate": 0.0002, "epoch": 1.4942099364960777, "step": 2000}, {"loss": 1.6998, "grad_norm": 0.3340149223804474, "learning_rate": 0.0002, "epoch": 1.501680986178558, "step": 2010}, {"loss": 1.7508, "grad_norm": 0.37685006856918335, "learning_rate": 0.0002, "epoch": 1.5091520358610384, "step": 2020}, {"loss": 1.8299, "grad_norm": 0.3699876368045807, "learning_rate": 0.0002, "epoch": 1.516623085543519, "step": 2030}, {"loss": 1.7357, "grad_norm": 0.3370307385921478, "learning_rate": 0.0002, "epoch": 1.5240941352259991, "step": 2040}, {"loss": 1.8044, "grad_norm": 0.37780630588531494, "learning_rate": 0.0002, "epoch": 1.5315651849084797, "step": 2050}, {"loss": 1.7408, "grad_norm": 0.370259165763855, "learning_rate": 0.0002, "epoch": 1.53903623459096, "step": 2060}, {"loss": 1.7398, "grad_norm": 0.3440011441707611, "learning_rate": 0.0002, "epoch": 1.5465072842734404, "step": 2070}, {"loss": 1.7105, "grad_norm": 0.40382063388824463, "learning_rate": 0.0002, "epoch": 1.5539783339559208, "step": 2080}, {"loss": 1.7071, "grad_norm": 0.38002029061317444, "learning_rate": 0.0002, "epoch": 1.5614493836384011, "step": 2090}, {"loss": 1.6815, "grad_norm": 0.3658451437950134, "learning_rate": 0.0002, "epoch": 1.5689204333208817, "step": 2100}, {"loss": 1.7598, "grad_norm": 0.354842871427536, "learning_rate": 0.0002, "epoch": 1.5763914830033618, "step": 2110}, {"loss": 1.6898, "grad_norm": 0.34735530614852905, "learning_rate": 0.0002, "epoch": 1.5838625326858424, "step": 2120}, {"loss": 1.7363, "grad_norm": 0.377581924200058, "learning_rate": 0.0002, "epoch": 1.5913335823683228, "step": 2130}, {"loss": 1.7789, "grad_norm": 0.41254034638404846, "learning_rate": 0.0002, "epoch": 1.5988046320508031, "step": 2140}, {"loss": 1.6782, "grad_norm": 0.3630715310573578, "learning_rate": 0.0002, "epoch": 1.6062756817332835, "step": 2150}, {"loss": 1.7531, "grad_norm": 0.36980143189430237, "learning_rate": 0.0002, "epoch": 1.6137467314157639, "step": 2160}, {"loss": 1.6847, "grad_norm": 0.3634769320487976, "learning_rate": 0.0002, "epoch": 1.6212177810982444, "step": 2170}, {"loss": 1.6367, "grad_norm": 0.3794139623641968, "learning_rate": 0.0002, "epoch": 1.6286888307807246, "step": 2180}, {"loss": 1.7064, "grad_norm": 0.359742134809494, "learning_rate": 0.0002, "epoch": 1.6361598804632052, "step": 2190}, {"loss": 1.7027, "grad_norm": 0.3770543932914734, "learning_rate": 0.0002, "epoch": 1.6436309301456855, "step": 2200}, {"loss": 1.784, "grad_norm": 0.3797036409378052, "learning_rate": 0.0002, "epoch": 1.6511019798281659, "step": 2210}, {"loss": 1.7875, "grad_norm": 0.35622093081474304, "learning_rate": 0.0002, "epoch": 1.6585730295106462, "step": 2220}, {"loss": 1.6615, "grad_norm": 0.34552520513534546, "learning_rate": 0.0002, "epoch": 1.6660440791931266, "step": 2230}, {"loss": 1.7522, "grad_norm": 0.379926860332489, "learning_rate": 0.0002, "epoch": 1.6735151288756072, "step": 2240}, {"loss": 1.7953, "grad_norm": 0.37083810567855835, "learning_rate": 0.0002, "epoch": 1.6809861785580873, "step": 2250}, {"loss": 1.7485, "grad_norm": 0.42746543884277344, "learning_rate": 0.0002, "epoch": 1.6884572282405679, "step": 2260}, {"loss": 1.776, "grad_norm": 0.3372884690761566, "learning_rate": 0.0002, "epoch": 1.6959282779230482, "step": 2270}, {"loss": 1.7604, "grad_norm": 0.35220256447792053, "learning_rate": 0.0002, "epoch": 1.7033993276055286, "step": 2280}, {"loss": 1.7154, "grad_norm": 0.3659130930900574, "learning_rate": 0.0002, "epoch": 1.710870377288009, "step": 2290}, {"loss": 1.6953, "grad_norm": 0.37629297375679016, "learning_rate": 0.0002, "epoch": 1.7183414269704893, "step": 2300}, {"loss": 1.7212, "grad_norm": 0.36312398314476013, "learning_rate": 0.0002, "epoch": 1.7258124766529699, "step": 2310}, {"loss": 1.7903, "grad_norm": 0.467709481716156, "learning_rate": 0.0002, "epoch": 1.73328352633545, "step": 2320}, {"loss": 1.696, "grad_norm": 0.38685527443885803, "learning_rate": 0.0002, "epoch": 1.7407545760179306, "step": 2330}, {"loss": 1.7041, "grad_norm": 0.3578338325023651, "learning_rate": 0.0002, "epoch": 1.748225625700411, "step": 2340}, {"loss": 1.6456, "grad_norm": 0.36057502031326294, "learning_rate": 0.0002, "epoch": 1.7556966753828913, "step": 2350}, {"loss": 1.6853, "grad_norm": 0.3615196645259857, "learning_rate": 0.0002, "epoch": 1.7631677250653717, "step": 2360}, {"loss": 1.7612, "grad_norm": 0.4118947684764862, "learning_rate": 0.0002, "epoch": 1.770638774747852, "step": 2370}, {"loss": 1.6946, "grad_norm": 0.4067276120185852, "learning_rate": 0.0002, "epoch": 1.7781098244303326, "step": 2380}, {"loss": 1.712, "grad_norm": 0.3979823887348175, "learning_rate": 0.0002, "epoch": 1.7855808741128127, "step": 2390}, {"loss": 1.7644, "grad_norm": 0.44045883417129517, "learning_rate": 0.0002, "epoch": 1.7930519237952933, "step": 2400}, {"loss": 1.7251, "grad_norm": 0.3998069167137146, "learning_rate": 0.0002, "epoch": 1.8005229734777737, "step": 2410}, {"loss": 1.7354, "grad_norm": 0.3450094759464264, "learning_rate": 0.0002, "epoch": 1.807994023160254, "step": 2420}, {"loss": 1.6998, "grad_norm": 0.3759009838104248, "learning_rate": 0.0002, "epoch": 1.8154650728427344, "step": 2430}, {"loss": 1.7706, "grad_norm": 0.34347015619277954, "learning_rate": 0.0002, "epoch": 1.8229361225252148, "step": 2440}, {"loss": 1.7345, "grad_norm": 0.3511228859424591, "learning_rate": 0.0002, "epoch": 1.8304071722076953, "step": 2450}, {"loss": 1.6909, "grad_norm": 0.36853715777397156, "learning_rate": 0.0002, "epoch": 1.8378782218901755, "step": 2460}, {"loss": 1.6931, "grad_norm": 0.40659376978874207, "learning_rate": 0.0002, "epoch": 1.845349271572656, "step": 2470}, {"loss": 1.7626, "grad_norm": 0.39621320366859436, "learning_rate": 0.0002, "epoch": 1.8528203212551362, "step": 2480}, {"loss": 1.7427, "grad_norm": 0.3753979504108429, "learning_rate": 0.0002, "epoch": 1.8602913709376168, "step": 2490}, {"loss": 1.6622, "grad_norm": 0.3811938464641571, "learning_rate": 0.0002, "epoch": 1.8677624206200971, "step": 2500}, {"loss": 1.7718, "grad_norm": 0.3432596027851105, "learning_rate": 0.0002, "epoch": 1.8752334703025775, "step": 2510}, {"loss": 1.7488, "grad_norm": 0.3670712113380432, "learning_rate": 0.0002, "epoch": 1.882704519985058, "step": 2520}, {"loss": 1.705, "grad_norm": 0.40907177329063416, "learning_rate": 0.0002, "epoch": 1.8901755696675382, "step": 2530}, {"loss": 1.7148, "grad_norm": 0.3821999728679657, "learning_rate": 0.0002, "epoch": 1.8976466193500188, "step": 2540}, {"loss": 1.7934, "grad_norm": 0.36173978447914124, "learning_rate": 0.0002, "epoch": 1.905117669032499, "step": 2550}, {"loss": 1.6939, "grad_norm": 0.38990336656570435, "learning_rate": 0.0002, "epoch": 1.9125887187149795, "step": 2560}, {"loss": 1.6893, "grad_norm": 0.35242322087287903, "learning_rate": 0.0002, "epoch": 1.9200597683974598, "step": 2570}, {"loss": 1.7268, "grad_norm": 0.3506428003311157, "learning_rate": 0.0002, "epoch": 1.9275308180799402, "step": 2580}, {"loss": 1.6953, "grad_norm": 0.39540135860443115, "learning_rate": 0.0002, "epoch": 1.9350018677624208, "step": 2590}, {"loss": 1.6511, "grad_norm": 0.3444725573062897, "learning_rate": 0.0002, "epoch": 1.942472917444901, "step": 2600}, {"loss": 1.7259, "grad_norm": 0.3963521718978882, "learning_rate": 0.0002, "epoch": 1.9499439671273815, "step": 2610}, {"loss": 1.6946, "grad_norm": 0.3689815402030945, "learning_rate": 0.0002, "epoch": 1.9574150168098616, "step": 2620}, {"loss": 1.7384, "grad_norm": 0.3482626676559448, "learning_rate": 0.0002, "epoch": 1.9648860664923422, "step": 2630}, {"loss": 1.7048, "grad_norm": 0.35832616686820984, "learning_rate": 0.0002, "epoch": 1.9723571161748226, "step": 2640}, {"loss": 1.6681, "grad_norm": 0.4776208996772766, "learning_rate": 0.0002, "epoch": 1.979828165857303, "step": 2650}, {"loss": 1.6696, "grad_norm": 0.32570165395736694, "learning_rate": 0.0002, "epoch": 1.9872992155397835, "step": 2660}, {"loss": 1.7232, "grad_norm": 0.3380725085735321, "learning_rate": 0.0002, "epoch": 1.9947702652222636, "step": 2670}, {"eval_loss": 1.8046749830245972, "eval_runtime": 38.5096, "eval_samples_per_second": 13.373, "eval_steps_per_second": 1.688, "epoch": 2.0, "step": 2677}, {"loss": 1.7265, "grad_norm": 0.36817631125450134, "learning_rate": 0.0002, "epoch": 2.002241314904744, "step": 2680}, {"loss": 1.548, "grad_norm": 0.4056456685066223, "learning_rate": 0.0002, "epoch": 2.0097123645872244, "step": 2690}, {"loss": 1.5515, "grad_norm": 0.37416863441467285, "learning_rate": 0.0002, "epoch": 2.017183414269705, "step": 2700}, {"loss": 1.5895, "grad_norm": 0.4273638427257538, "learning_rate": 0.0002, "epoch": 2.024654463952185, "step": 2710}, {"loss": 1.5884, "grad_norm": 0.36497923731803894, "learning_rate": 0.0002, "epoch": 2.0321255136346656, "step": 2720}, {"loss": 1.6999, "grad_norm": 0.5021994113922119, "learning_rate": 0.0002, "epoch": 2.0395965633171462, "step": 2730}, {"loss": 1.6655, "grad_norm": 0.45896220207214355, "learning_rate": 0.0002, "epoch": 2.0470676129996264, "step": 2740}, {"loss": 1.6305, "grad_norm": 0.3973815143108368, "learning_rate": 0.0002, "epoch": 2.054538662682107, "step": 2750}, {"loss": 1.6301, "grad_norm": 0.4521815776824951, "learning_rate": 0.0002, "epoch": 2.062009712364587, "step": 2760}, {"loss": 1.6189, "grad_norm": 0.42775002121925354, "learning_rate": 0.0002, "epoch": 2.0694807620470677, "step": 2770}, {"loss": 1.6491, "grad_norm": 0.48158586025238037, "learning_rate": 0.0002, "epoch": 2.076951811729548, "step": 2780}, {"loss": 1.6301, "grad_norm": 0.4612371623516083, "learning_rate": 0.0002, "epoch": 2.0844228614120284, "step": 2790}, {"loss": 1.6327, "grad_norm": 0.42536866664886475, "learning_rate": 0.0002, "epoch": 2.091893911094509, "step": 2800}, {"loss": 1.651, "grad_norm": 0.48515772819519043, "learning_rate": 0.0002, "epoch": 2.099364960776989, "step": 2810}, {"loss": 1.6829, "grad_norm": 0.41418662667274475, "learning_rate": 0.0002, "epoch": 2.1068360104594697, "step": 2820}, {"loss": 1.6266, "grad_norm": 0.4683697819709778, "learning_rate": 0.0002, "epoch": 2.11430706014195, "step": 2830}, {"loss": 1.6586, "grad_norm": 0.4484657049179077, "learning_rate": 0.0002, "epoch": 2.1217781098244304, "step": 2840}, {"loss": 1.6483, "grad_norm": 0.6621400713920593, "learning_rate": 0.0002, "epoch": 2.1292491595069105, "step": 2850}, {"loss": 1.5755, "grad_norm": 0.45074811577796936, "learning_rate": 0.0002, "epoch": 2.136720209189391, "step": 2860}, {"loss": 1.6456, "grad_norm": 0.3513113558292389, "learning_rate": 0.0002, "epoch": 2.1441912588718717, "step": 2870}, {"loss": 1.6081, "grad_norm": 0.40411314368247986, "learning_rate": 0.0002, "epoch": 2.151662308554352, "step": 2880}, {"loss": 1.6323, "grad_norm": 0.4121065139770508, "learning_rate": 0.0002, "epoch": 2.1591333582368324, "step": 2890}, {"loss": 1.6324, "grad_norm": 0.44723689556121826, "learning_rate": 0.0002, "epoch": 2.1666044079193125, "step": 2900}, {"loss": 1.5699, "grad_norm": 0.4226122498512268, "learning_rate": 0.0002, "epoch": 2.174075457601793, "step": 2910}, {"loss": 1.5652, "grad_norm": 0.46617650985717773, "learning_rate": 0.0002, "epoch": 2.1815465072842732, "step": 2920}, {"loss": 1.6378, "grad_norm": 0.4506422281265259, "learning_rate": 0.0002, "epoch": 2.189017556966754, "step": 2930}, {"loss": 1.6112, "grad_norm": 0.4892672896385193, "learning_rate": 0.0002, "epoch": 2.1964886066492344, "step": 2940}, {"loss": 1.6176, "grad_norm": 0.44095516204833984, "learning_rate": 0.0002, "epoch": 2.2039596563317145, "step": 2950}, {"loss": 1.6058, "grad_norm": 0.41522109508514404, "learning_rate": 0.0002, "epoch": 2.211430706014195, "step": 2960}, {"loss": 1.5964, "grad_norm": 0.4860858917236328, "learning_rate": 0.0002, "epoch": 2.2189017556966752, "step": 2970}, {"loss": 1.6427, "grad_norm": 0.42662516236305237, "learning_rate": 0.0002, "epoch": 2.226372805379156, "step": 2980}, {"loss": 1.6313, "grad_norm": 0.4390648305416107, "learning_rate": 0.0002, "epoch": 2.233843855061636, "step": 2990}, {"loss": 1.5992, "grad_norm": 0.47515565156936646, "learning_rate": 0.0002, "epoch": 2.2413149047441165, "step": 3000}, {"loss": 1.5563, "grad_norm": 0.4104543924331665, "learning_rate": 0.0002, "epoch": 2.248785954426597, "step": 3010}, {"loss": 1.6895, "grad_norm": 0.4404028654098511, "learning_rate": 0.0002, "epoch": 2.2562570041090773, "step": 3020}, {"loss": 1.6088, "grad_norm": 0.4717366695404053, "learning_rate": 0.0002, "epoch": 2.263728053791558, "step": 3030}, {"loss": 1.7287, "grad_norm": 0.48345857858657837, "learning_rate": 0.0002, "epoch": 2.271199103474038, "step": 3040}, {"loss": 1.681, "grad_norm": 0.5312452912330627, "learning_rate": 0.0002, "epoch": 2.2786701531565186, "step": 3050}, {"loss": 1.5901, "grad_norm": 0.5073099732398987, "learning_rate": 0.0002, "epoch": 2.2861412028389987, "step": 3060}, {"loss": 1.6914, "grad_norm": 0.5027463436126709, "learning_rate": 0.0002, "epoch": 2.2936122525214793, "step": 3070}, {"loss": 1.5862, "grad_norm": 0.5436304807662964, "learning_rate": 0.0002, "epoch": 2.30108330220396, "step": 3080}, {"loss": 1.5763, "grad_norm": 0.4701065123081207, "learning_rate": 0.0002, "epoch": 2.30855435188644, "step": 3090}, {"loss": 1.6177, "grad_norm": 0.46988746523857117, "learning_rate": 0.0002, "epoch": 2.3160254015689206, "step": 3100}, {"loss": 1.6502, "grad_norm": 0.45112869143486023, "learning_rate": 0.0002, "epoch": 2.3234964512514007, "step": 3110}, {"loss": 1.6291, "grad_norm": 0.5173566937446594, "learning_rate": 0.0002, "epoch": 2.3309675009338813, "step": 3120}, {"loss": 1.6743, "grad_norm": 0.40345850586891174, "learning_rate": 0.0002, "epoch": 2.3384385506163614, "step": 3130}, {"loss": 1.621, "grad_norm": 0.4218924939632416, "learning_rate": 0.0002, "epoch": 2.345909600298842, "step": 3140}, {"loss": 1.6341, "grad_norm": 0.41857317090034485, "learning_rate": 0.0002, "epoch": 2.3533806499813226, "step": 3150}, {"loss": 1.6087, "grad_norm": 0.4197218418121338, "learning_rate": 0.0002, "epoch": 2.3608516996638027, "step": 3160}, {"loss": 1.6572, "grad_norm": 0.4260677397251129, "learning_rate": 0.0002, "epoch": 2.3683227493462833, "step": 3170}, {"loss": 1.6376, "grad_norm": 0.4209042191505432, "learning_rate": 0.0002, "epoch": 2.3757937990287634, "step": 3180}, {"loss": 1.634, "grad_norm": 0.4092234969139099, "learning_rate": 0.0002, "epoch": 2.383264848711244, "step": 3190}, {"loss": 1.6339, "grad_norm": 0.4928431510925293, "learning_rate": 0.0002, "epoch": 2.390735898393724, "step": 3200}, {"loss": 1.6015, "grad_norm": 0.49252402782440186, "learning_rate": 0.0002, "epoch": 2.3982069480762047, "step": 3210}, {"loss": 1.5773, "grad_norm": 0.4368397295475006, "learning_rate": 0.0002, "epoch": 2.4056779977586853, "step": 3220}, {"loss": 1.6629, "grad_norm": 0.46122390031814575, "learning_rate": 0.0002, "epoch": 2.4131490474411654, "step": 3230}, {"loss": 1.6224, "grad_norm": 0.4272301197052002, "learning_rate": 0.0002, "epoch": 2.420620097123646, "step": 3240}, {"loss": 1.5961, "grad_norm": 0.41480937600135803, "learning_rate": 0.0002, "epoch": 2.428091146806126, "step": 3250}, {"loss": 1.6281, "grad_norm": 0.48911941051483154, "learning_rate": 0.0002, "epoch": 2.4355621964886067, "step": 3260}, {"loss": 1.6846, "grad_norm": 0.4444098472595215, "learning_rate": 0.0002, "epoch": 2.443033246171087, "step": 3270}, {"loss": 1.6961, "grad_norm": 0.5111684799194336, "learning_rate": 0.0002, "epoch": 2.4505042958535674, "step": 3280}, {"loss": 1.6152, "grad_norm": 0.5058825016021729, "learning_rate": 0.0002, "epoch": 2.457975345536048, "step": 3290}, {"loss": 1.625, "grad_norm": 0.44173210859298706, "learning_rate": 0.0002, "epoch": 2.465446395218528, "step": 3300}, {"loss": 1.6491, "grad_norm": 0.4659745991230011, "learning_rate": 0.0002, "epoch": 2.4729174449010087, "step": 3310}, {"loss": 1.6114, "grad_norm": 0.47237497568130493, "learning_rate": 0.0002, "epoch": 2.480388494583489, "step": 3320}, {"loss": 1.6193, "grad_norm": 0.47303131222724915, "learning_rate": 0.0002, "epoch": 2.4878595442659694, "step": 3330}, {"loss": 1.7256, "grad_norm": 0.4522389769554138, "learning_rate": 0.0002, "epoch": 2.4953305939484496, "step": 3340}, {"loss": 1.6834, "grad_norm": 0.4467332363128662, "learning_rate": 0.0002, "epoch": 2.50280164363093, "step": 3350}, {"loss": 1.6108, "grad_norm": 0.4413762092590332, "learning_rate": 0.0002, "epoch": 2.5102726933134107, "step": 3360}, {"loss": 1.537, "grad_norm": 0.495514452457428, "learning_rate": 0.0002, "epoch": 2.517743742995891, "step": 3370}, {"loss": 1.5839, "grad_norm": 0.4429773986339569, "learning_rate": 0.0002, "epoch": 2.5252147926783715, "step": 3380}, {"loss": 1.6522, "grad_norm": 0.4589079022407532, "learning_rate": 0.0002, "epoch": 2.5326858423608516, "step": 3390}, {"loss": 1.6529, "grad_norm": 0.4683997333049774, "learning_rate": 0.0002, "epoch": 2.540156892043332, "step": 3400}, {"loss": 1.6745, "grad_norm": 0.4651731252670288, "learning_rate": 0.0002, "epoch": 2.5476279417258123, "step": 3410}, {"loss": 1.5918, "grad_norm": 0.45818084478378296, "learning_rate": 0.0002, "epoch": 2.555098991408293, "step": 3420}, {"loss": 1.6326, "grad_norm": 0.45209529995918274, "learning_rate": 0.0002, "epoch": 2.5625700410907735, "step": 3430}, {"loss": 1.5606, "grad_norm": 0.4344733655452728, "learning_rate": 0.0002, "epoch": 2.5700410907732536, "step": 3440}, {"loss": 1.6748, "grad_norm": 0.47435566782951355, "learning_rate": 0.0002, "epoch": 2.577512140455734, "step": 3450}, {"loss": 1.6237, "grad_norm": 0.43841999769210815, "learning_rate": 0.0002, "epoch": 2.5849831901382143, "step": 3460}, {"loss": 1.7207, "grad_norm": 0.4323869049549103, "learning_rate": 0.0002, "epoch": 2.592454239820695, "step": 3470}, {"loss": 1.5494, "grad_norm": 0.44355881214141846, "learning_rate": 0.0002, "epoch": 2.599925289503175, "step": 3480}, {"loss": 1.665, "grad_norm": 0.45847779512405396, "learning_rate": 0.0002, "epoch": 2.6073963391856556, "step": 3490}, {"loss": 1.6006, "grad_norm": 0.4411061704158783, "learning_rate": 0.0002, "epoch": 2.614867388868136, "step": 3500}, {"loss": 1.5868, "grad_norm": 0.4446796178817749, "learning_rate": 0.0002, "epoch": 2.6223384385506163, "step": 3510}, {"loss": 1.5946, "grad_norm": 0.41969653964042664, "learning_rate": 0.0002, "epoch": 2.629809488233097, "step": 3520}, {"loss": 1.6798, "grad_norm": 0.5263747572898865, "learning_rate": 0.0002, "epoch": 2.637280537915577, "step": 3530}, {"loss": 1.6309, "grad_norm": 0.47719451785087585, "learning_rate": 0.0002, "epoch": 2.6447515875980576, "step": 3540}, {"loss": 1.7024, "grad_norm": 0.46574118733406067, "learning_rate": 0.0002, "epoch": 2.6522226372805378, "step": 3550}, {"loss": 1.618, "grad_norm": 0.46867135167121887, "learning_rate": 0.0002, "epoch": 2.6596936869630183, "step": 3560}, {"loss": 1.5885, "grad_norm": 0.4441198706626892, "learning_rate": 0.0002, "epoch": 2.667164736645499, "step": 3570}, {"loss": 1.6426, "grad_norm": 0.4871319830417633, "learning_rate": 0.0002, "epoch": 2.674635786327979, "step": 3580}, {"loss": 1.6575, "grad_norm": 0.43900373578071594, "learning_rate": 0.0002, "epoch": 2.6821068360104596, "step": 3590}, {"loss": 1.6071, "grad_norm": 0.42509549856185913, "learning_rate": 0.0002, "epoch": 2.6895778856929398, "step": 3600}, {"loss": 1.5651, "grad_norm": 0.4691086709499359, "learning_rate": 0.0002, "epoch": 2.6970489353754203, "step": 3610}, {"loss": 1.5491, "grad_norm": 0.46318942308425903, "learning_rate": 0.0002, "epoch": 2.7045199850579005, "step": 3620}, {"loss": 1.5422, "grad_norm": 0.44631096720695496, "learning_rate": 0.0002, "epoch": 2.711991034740381, "step": 3630}, {"loss": 1.6831, "grad_norm": 0.42315489053726196, "learning_rate": 0.0002, "epoch": 2.7194620844228616, "step": 3640}, {"loss": 1.6008, "grad_norm": 0.4971241056919098, "learning_rate": 0.0002, "epoch": 2.7269331341053418, "step": 3650}, {"loss": 1.6042, "grad_norm": 0.4578486382961273, "learning_rate": 0.0002, "epoch": 2.7344041837878224, "step": 3660}, {"loss": 1.6076, "grad_norm": 0.46584776043891907, "learning_rate": 0.0002, "epoch": 2.7418752334703025, "step": 3670}, {"loss": 1.6809, "grad_norm": 0.4951731264591217, "learning_rate": 0.0002, "epoch": 2.749346283152783, "step": 3680}, {"loss": 1.6226, "grad_norm": 0.4935225546360016, "learning_rate": 0.0002, "epoch": 2.756817332835263, "step": 3690}, {"loss": 1.5878, "grad_norm": 0.41805586218833923, "learning_rate": 0.0002, "epoch": 2.764288382517744, "step": 3700}, {"loss": 1.7173, "grad_norm": 0.4417555630207062, "learning_rate": 0.0002, "epoch": 2.7717594322002244, "step": 3710}, {"loss": 1.6398, "grad_norm": 0.48229655623435974, "learning_rate": 0.0002, "epoch": 2.7792304818827045, "step": 3720}, {"loss": 1.6074, "grad_norm": 0.48562315106391907, "learning_rate": 0.0002, "epoch": 2.786701531565185, "step": 3730}, {"loss": 1.607, "grad_norm": 0.4473940432071686, "learning_rate": 0.0002, "epoch": 2.794172581247665, "step": 3740}, {"loss": 1.6065, "grad_norm": 0.4626813232898712, "learning_rate": 0.0002, "epoch": 2.801643630930146, "step": 3750}, {"loss": 1.6296, "grad_norm": 0.4339792728424072, "learning_rate": 0.0002, "epoch": 2.809114680612626, "step": 3760}, {"loss": 1.6815, "grad_norm": 0.5250858068466187, "learning_rate": 0.0002, "epoch": 2.8165857302951065, "step": 3770}, {"loss": 1.6644, "grad_norm": 0.4537523090839386, "learning_rate": 0.0002, "epoch": 2.824056779977587, "step": 3780}, {"loss": 1.6535, "grad_norm": 0.5646113157272339, "learning_rate": 0.0002, "epoch": 2.831527829660067, "step": 3790}, {"loss": 1.5712, "grad_norm": 0.44243332743644714, "learning_rate": 0.0002, "epoch": 2.8389988793425474, "step": 3800}, {"loss": 1.6478, "grad_norm": 0.4585791826248169, "learning_rate": 0.0002, "epoch": 2.846469929025028, "step": 3810}, {"loss": 1.6854, "grad_norm": 0.489702045917511, "learning_rate": 0.0002, "epoch": 2.8539409787075085, "step": 3820}, {"loss": 1.7066, "grad_norm": 0.502470850944519, "learning_rate": 0.0002, "epoch": 2.8614120283899886, "step": 3830}, {"loss": 1.5785, "grad_norm": 0.4395960867404938, "learning_rate": 0.0002, "epoch": 2.8688830780724692, "step": 3840}, {"loss": 1.6434, "grad_norm": 0.4348670244216919, "learning_rate": 0.0002, "epoch": 2.87635412775495, "step": 3850}, {"loss": 1.6163, "grad_norm": 0.48852720856666565, "learning_rate": 0.0002, "epoch": 2.88382517743743, "step": 3860}, {"loss": 1.5916, "grad_norm": 0.45317450165748596, "learning_rate": 0.0002, "epoch": 2.89129622711991, "step": 3870}, {"loss": 1.6486, "grad_norm": 0.4732758700847626, "learning_rate": 0.0002, "epoch": 2.8987672768023907, "step": 3880}, {"loss": 1.6758, "grad_norm": 0.45238012075424194, "learning_rate": 0.0002, "epoch": 2.9062383264848712, "step": 3890}, {"loss": 1.6228, "grad_norm": 0.48838064074516296, "learning_rate": 0.0002, "epoch": 2.9137093761673514, "step": 3900}, {"loss": 1.658, "grad_norm": 0.43496349453926086, "learning_rate": 0.0002, "epoch": 2.921180425849832, "step": 3910}, {"loss": 1.7063, "grad_norm": 0.47963935136795044, "learning_rate": 0.0002, "epoch": 2.9286514755323125, "step": 3920}, {"loss": 1.6553, "grad_norm": 0.4544987976551056, "learning_rate": 0.0002, "epoch": 2.9361225252147927, "step": 3930}, {"loss": 1.6192, "grad_norm": 0.4622892141342163, "learning_rate": 0.0002, "epoch": 2.943593574897273, "step": 3940}, {"loss": 1.6178, "grad_norm": 0.47026222944259644, "learning_rate": 0.0002, "epoch": 2.9510646245797534, "step": 3950}, {"loss": 1.6612, "grad_norm": 0.4549552798271179, "learning_rate": 0.0002, "epoch": 2.958535674262234, "step": 3960}, {"loss": 1.6458, "grad_norm": 0.46647515892982483, "learning_rate": 0.0002, "epoch": 2.966006723944714, "step": 3970}, {"loss": 1.6051, "grad_norm": 0.45095112919807434, "learning_rate": 0.0002, "epoch": 2.9734777736271947, "step": 3980}, {"loss": 1.6471, "grad_norm": 0.4690017104148865, "learning_rate": 0.0002, "epoch": 2.9809488233096753, "step": 3990}, {"loss": 1.6061, "grad_norm": 0.4603444039821625, "learning_rate": 0.0002, "epoch": 2.9884198729921554, "step": 4000}, {"loss": 1.6431, "grad_norm": 0.4743294417858124, "learning_rate": 0.0002, "epoch": 2.9958909226746355, "step": 4010}, {"eval_loss": 1.8252571821212769, "eval_runtime": 38.7853, "eval_samples_per_second": 13.278, "eval_steps_per_second": 1.676, "epoch": 2.999626447515876, "step": 4015}, {"loss": 1.6512, "grad_norm": 0.4919724464416504, "learning_rate": 0.0002, "epoch": 3.003361972357116, "step": 4020}, {"loss": 1.5354, "grad_norm": 0.4747185707092285, "learning_rate": 0.0002, "epoch": 3.0108330220395967, "step": 4030}, {"loss": 1.568, "grad_norm": 0.4797595143318176, "learning_rate": 0.0002, "epoch": 3.018304071722077, "step": 4040}, {"loss": 1.5194, "grad_norm": 0.5450999140739441, "learning_rate": 0.0002, "epoch": 3.0257751214045574, "step": 4050}, {"loss": 1.5065, "grad_norm": 0.49058812856674194, "learning_rate": 0.0002, "epoch": 3.0332461710870375, "step": 4060}, {"loss": 1.4884, "grad_norm": 0.5219563841819763, "learning_rate": 0.0002, "epoch": 3.040717220769518, "step": 4070}, {"loss": 1.4742, "grad_norm": 0.515628457069397, "learning_rate": 0.0002, "epoch": 3.0481882704519987, "step": 4080}, {"loss": 1.5313, "grad_norm": 0.6145984530448914, "learning_rate": 0.0002, "epoch": 3.055659320134479, "step": 4090}, {"loss": 1.4989, "grad_norm": 0.6067144274711609, "learning_rate": 0.0002, "epoch": 3.0631303698169594, "step": 4100}, {"loss": 1.528, "grad_norm": 0.5773133039474487, "learning_rate": 0.0002, "epoch": 3.0706014194994395, "step": 4110}, {"loss": 1.5374, "grad_norm": 0.6894241571426392, "learning_rate": 0.0002, "epoch": 3.07807246918192, "step": 4120}, {"loss": 1.5422, "grad_norm": 0.6422514915466309, "learning_rate": 0.0002, "epoch": 3.0855435188644003, "step": 4130}, {"loss": 1.4724, "grad_norm": 0.6119855046272278, "learning_rate": 0.0002, "epoch": 3.093014568546881, "step": 4140}, {"loss": 1.5361, "grad_norm": 0.5847280025482178, "learning_rate": 0.0002, "epoch": 3.1004856182293614, "step": 4150}, {"loss": 1.5151, "grad_norm": 0.5401515960693359, "learning_rate": 0.0002, "epoch": 3.1079566679118416, "step": 4160}, {"loss": 1.502, "grad_norm": 0.6501587629318237, "learning_rate": 0.0002, "epoch": 3.115427717594322, "step": 4170}, {"loss": 1.4952, "grad_norm": 0.5988039374351501, "learning_rate": 0.0002, "epoch": 3.1228987672768023, "step": 4180}, {"loss": 1.5287, "grad_norm": 0.4982665181159973, "learning_rate": 0.0002, "epoch": 3.130369816959283, "step": 4190}, {"loss": 1.5078, "grad_norm": 0.5548039078712463, "learning_rate": 0.0002, "epoch": 3.137840866641763, "step": 4200}, {"loss": 1.4904, "grad_norm": 0.5920777320861816, "learning_rate": 0.0002, "epoch": 3.1453119163242436, "step": 4210}, {"loss": 1.442, "grad_norm": 0.6965190172195435, "learning_rate": 0.0002, "epoch": 3.152782966006724, "step": 4220}, {"loss": 1.557, "grad_norm": 0.5196244716644287, "learning_rate": 0.0002, "epoch": 3.1602540156892043, "step": 4230}, {"loss": 1.5706, "grad_norm": 0.6942682266235352, "learning_rate": 0.0002, "epoch": 3.167725065371685, "step": 4240}, {"loss": 1.5407, "grad_norm": 0.5765156149864197, "learning_rate": 0.0002, "epoch": 3.175196115054165, "step": 4250}, {"loss": 1.4963, "grad_norm": 0.5801976919174194, "learning_rate": 0.0002, "epoch": 3.1826671647366456, "step": 4260}, {"loss": 1.4988, "grad_norm": 0.6260752081871033, "learning_rate": 0.0002, "epoch": 3.1901382144191257, "step": 4270}, {"loss": 1.5074, "grad_norm": 0.6610770225524902, "learning_rate": 0.0002, "epoch": 3.1976092641016063, "step": 4280}, {"loss": 1.4657, "grad_norm": 0.5762143135070801, "learning_rate": 0.0002, "epoch": 3.205080313784087, "step": 4290}, {"loss": 1.5181, "grad_norm": 0.5926990509033203, "learning_rate": 0.0002, "epoch": 3.212551363466567, "step": 4300}, {"loss": 1.5492, "grad_norm": 0.7373854517936707, "learning_rate": 0.0002, "epoch": 3.2200224131490476, "step": 4310}, {"loss": 1.4648, "grad_norm": 0.5963311195373535, "learning_rate": 0.0002, "epoch": 3.2274934628315277, "step": 4320}, {"loss": 1.5262, "grad_norm": 0.5754616856575012, "learning_rate": 0.0002, "epoch": 3.2349645125140083, "step": 4330}, {"loss": 1.4767, "grad_norm": 0.6116095781326294, "learning_rate": 0.0002, "epoch": 3.2424355621964884, "step": 4340}, {"loss": 1.5008, "grad_norm": 0.6001536846160889, "learning_rate": 0.0002, "epoch": 3.249906611878969, "step": 4350}, {"loss": 1.5738, "grad_norm": 0.5270227789878845, "learning_rate": 0.0002, "epoch": 3.257377661561449, "step": 4360}, {"loss": 1.5235, "grad_norm": 0.6666602492332458, "learning_rate": 0.0002, "epoch": 3.2648487112439297, "step": 4370}, {"loss": 1.5665, "grad_norm": 0.520310640335083, "learning_rate": 0.0002, "epoch": 3.2723197609264103, "step": 4380}, {"loss": 1.542, "grad_norm": 0.5165975093841553, "learning_rate": 0.0002, "epoch": 3.2797908106088904, "step": 4390}, {"loss": 1.4746, "grad_norm": 0.6080228686332703, "learning_rate": 0.0002, "epoch": 3.287261860291371, "step": 4400}, {"loss": 1.4901, "grad_norm": 0.670122504234314, "learning_rate": 0.0002, "epoch": 3.294732909973851, "step": 4410}, {"loss": 1.4677, "grad_norm": 0.6019457578659058, "learning_rate": 0.0002, "epoch": 3.3022039596563317, "step": 4420}, {"loss": 1.4249, "grad_norm": 0.5519300103187561, "learning_rate": 0.0002, "epoch": 3.309675009338812, "step": 4430}, {"loss": 1.555, "grad_norm": 0.5958521962165833, "learning_rate": 0.0002, "epoch": 3.3171460590212924, "step": 4440}, {"loss": 1.5067, "grad_norm": 0.5552705526351929, "learning_rate": 0.0002, "epoch": 3.324617108703773, "step": 4450}, {"loss": 1.5926, "grad_norm": 0.6583784818649292, "learning_rate": 0.0002, "epoch": 3.332088158386253, "step": 4460}, {"loss": 1.4206, "grad_norm": 0.5815939903259277, "learning_rate": 0.0002, "epoch": 3.3395592080687337, "step": 4470}, {"loss": 1.5942, "grad_norm": 1.3342205286026, "learning_rate": 0.0002, "epoch": 3.347030257751214, "step": 4480}, {"loss": 1.484, "grad_norm": 0.6341500878334045, "learning_rate": 0.0002, "epoch": 3.3545013074336945, "step": 4490}, {"loss": 1.5219, "grad_norm": 0.6384079456329346, "learning_rate": 0.0002, "epoch": 3.3619723571161746, "step": 4500}, {"loss": 1.5222, "grad_norm": 0.6098346710205078, "learning_rate": 0.0002, "epoch": 3.369443406798655, "step": 4510}, {"loss": 1.5475, "grad_norm": 0.5958296656608582, "learning_rate": 0.0002, "epoch": 3.3769144564811358, "step": 4520}, {"loss": 1.5171, "grad_norm": 0.6157881617546082, "learning_rate": 0.0002, "epoch": 3.384385506163616, "step": 4530}, {"loss": 1.569, "grad_norm": 0.5671007037162781, "learning_rate": 0.0002, "epoch": 3.3918565558460965, "step": 4540}, {"loss": 1.604, "grad_norm": 0.6203294992446899, "learning_rate": 0.0002, "epoch": 3.3993276055285766, "step": 4550}, {"loss": 1.5364, "grad_norm": 0.6743317246437073, "learning_rate": 0.0002, "epoch": 3.406798655211057, "step": 4560}, {"loss": 1.5034, "grad_norm": 0.731765627861023, "learning_rate": 0.0002, "epoch": 3.4142697048935373, "step": 4570}, {"loss": 1.4585, "grad_norm": 0.6285187602043152, "learning_rate": 0.0002, "epoch": 3.421740754576018, "step": 4580}, {"loss": 1.5296, "grad_norm": 0.612680196762085, "learning_rate": 0.0002, "epoch": 3.4292118042584985, "step": 4590}, {"loss": 1.5577, "grad_norm": 0.6413681507110596, "learning_rate": 0.0002, "epoch": 3.4366828539409786, "step": 4600}, {"loss": 1.5026, "grad_norm": 0.6240990161895752, "learning_rate": 0.0002, "epoch": 3.444153903623459, "step": 4610}, {"loss": 1.5887, "grad_norm": 0.5095735192298889, "learning_rate": 0.0002, "epoch": 3.4516249533059393, "step": 4620}, {"loss": 1.4906, "grad_norm": 0.5699611902236938, "learning_rate": 0.0002, "epoch": 3.45909600298842, "step": 4630}, {"loss": 1.5176, "grad_norm": 0.7289775609970093, "learning_rate": 0.0002, "epoch": 3.4665670526709, "step": 4640}, {"loss": 1.5467, "grad_norm": 0.6211609840393066, "learning_rate": 0.0002, "epoch": 3.4740381023533806, "step": 4650}, {"loss": 1.533, "grad_norm": 0.5714802145957947, "learning_rate": 0.0002, "epoch": 3.481509152035861, "step": 4660}, {"loss": 1.5096, "grad_norm": 0.6287049651145935, "learning_rate": 0.0002, "epoch": 3.4889802017183413, "step": 4670}, {"loss": 1.4212, "grad_norm": 0.5480595827102661, "learning_rate": 0.0002, "epoch": 3.496451251400822, "step": 4680}, {"loss": 1.4746, "grad_norm": 0.5683253407478333, "learning_rate": 0.0002, "epoch": 3.503922301083302, "step": 4690}, {"loss": 1.5012, "grad_norm": 0.601140558719635, "learning_rate": 0.0002, "epoch": 3.5113933507657826, "step": 4700}, {"loss": 1.5383, "grad_norm": 0.5344498157501221, "learning_rate": 0.0002, "epoch": 3.5188644004482628, "step": 4710}, {"loss": 1.5428, "grad_norm": 0.5739690661430359, "learning_rate": 0.0002, "epoch": 3.5263354501307433, "step": 4720}, {"loss": 1.5589, "grad_norm": 0.5640085935592651, "learning_rate": 0.0002, "epoch": 3.533806499813224, "step": 4730}, {"loss": 1.487, "grad_norm": 0.5967805981636047, "learning_rate": 0.0002, "epoch": 3.541277549495704, "step": 4740}, {"loss": 1.5461, "grad_norm": 0.6138835549354553, "learning_rate": 0.0002, "epoch": 3.5487485991781846, "step": 4750}, {"loss": 1.5502, "grad_norm": 0.6779900193214417, "learning_rate": 0.0002, "epoch": 3.5562196488606648, "step": 4760}, {"loss": 1.4917, "grad_norm": 0.6122010350227356, "learning_rate": 0.0002, "epoch": 3.5636906985431454, "step": 4770}, {"loss": 1.5405, "grad_norm": 0.5685241222381592, "learning_rate": 0.0002, "epoch": 3.5711617482256255, "step": 4780}, {"loss": 1.5427, "grad_norm": 0.604583203792572, "learning_rate": 0.0002, "epoch": 3.578632797908106, "step": 4790}, {"loss": 1.4514, "grad_norm": 0.651165246963501, "learning_rate": 0.0002, "epoch": 3.5861038475905866, "step": 4800}, {"loss": 1.4109, "grad_norm": 0.6398511528968811, "learning_rate": 0.0002, "epoch": 3.593574897273067, "step": 4810}, {"loss": 1.4261, "grad_norm": 0.6444641351699829, "learning_rate": 0.0002, "epoch": 3.6010459469555474, "step": 4820}, {"loss": 1.5274, "grad_norm": 0.6018481850624084, "learning_rate": 0.0002, "epoch": 3.6085169966380275, "step": 4830}, {"loss": 1.4647, "grad_norm": 0.6025291085243225, "learning_rate": 0.0002, "epoch": 3.615988046320508, "step": 4840}, {"loss": 1.5609, "grad_norm": 0.6810156106948853, "learning_rate": 0.0002, "epoch": 3.623459096002988, "step": 4850}, {"loss": 1.5299, "grad_norm": 0.6408044695854187, "learning_rate": 0.0002, "epoch": 3.630930145685469, "step": 4860}, {"loss": 1.5366, "grad_norm": 0.5608272552490234, "learning_rate": 0.0002, "epoch": 3.6384011953679494, "step": 4870}, {"loss": 1.5188, "grad_norm": 0.6136814951896667, "learning_rate": 0.0002, "epoch": 3.6458722450504295, "step": 4880}, {"loss": 1.5021, "grad_norm": 0.5927900075912476, "learning_rate": 0.0002, "epoch": 3.65334329473291, "step": 4890}, {"loss": 1.6084, "grad_norm": 0.5336901545524597, "learning_rate": 0.0002, "epoch": 3.66081434441539, "step": 4900}, {"loss": 1.5701, "grad_norm": 0.7823320627212524, "learning_rate": 0.0002, "epoch": 3.668285394097871, "step": 4910}, {"loss": 1.4881, "grad_norm": 0.6703504323959351, "learning_rate": 0.0002, "epoch": 3.675756443780351, "step": 4920}, {"loss": 1.5332, "grad_norm": 0.6061160564422607, "learning_rate": 0.0002, "epoch": 3.6832274934628315, "step": 4930}, {"loss": 1.5405, "grad_norm": 0.6237227916717529, "learning_rate": 0.0002, "epoch": 3.690698543145312, "step": 4940}, {"loss": 1.497, "grad_norm": 0.5985278487205505, "learning_rate": 0.0002, "epoch": 3.6981695928277922, "step": 4950}, {"loss": 1.5132, "grad_norm": 0.6483839750289917, "learning_rate": 0.0002, "epoch": 3.705640642510273, "step": 4960}, {"loss": 1.5338, "grad_norm": 0.5788805484771729, "learning_rate": 0.0002, "epoch": 3.713111692192753, "step": 4970}, {"loss": 1.5258, "grad_norm": 0.5609974265098572, "learning_rate": 0.0002, "epoch": 3.7205827418752335, "step": 4980}, {"loss": 1.4759, "grad_norm": 0.5681300759315491, "learning_rate": 0.0002, "epoch": 3.7280537915577137, "step": 4990}, {"loss": 1.6018, "grad_norm": 0.5860186219215393, "learning_rate": 0.0002, "epoch": 3.7355248412401942, "step": 5000}, {"loss": 1.58, "grad_norm": 0.5718157291412354, "learning_rate": 0.0002, "epoch": 3.742995890922675, "step": 5010}, {"loss": 1.5834, "grad_norm": 0.6173721551895142, "learning_rate": 0.0002, "epoch": 3.750466940605155, "step": 5020}, {"loss": 1.5617, "grad_norm": 0.629152238368988, "learning_rate": 0.0002, "epoch": 3.7579379902876355, "step": 5030}, {"loss": 1.519, "grad_norm": 0.5666284561157227, "learning_rate": 0.0002, "epoch": 3.7654090399701157, "step": 5040}, {"loss": 1.5329, "grad_norm": 0.6053005456924438, "learning_rate": 0.0002, "epoch": 3.7728800896525962, "step": 5050}, {"loss": 1.5404, "grad_norm": 0.5870583057403564, "learning_rate": 0.0002, "epoch": 3.7803511393350764, "step": 5060}, {"loss": 1.4444, "grad_norm": 0.5422009229660034, "learning_rate": 0.0002, "epoch": 3.787822189017557, "step": 5070}, {"loss": 1.5308, "grad_norm": 0.5396918058395386, "learning_rate": 0.0002, "epoch": 3.7952932387000375, "step": 5080}, {"loss": 1.464, "grad_norm": 0.5544713139533997, "learning_rate": 0.0002, "epoch": 3.8027642883825177, "step": 5090}, {"loss": 1.4752, "grad_norm": 0.5983749628067017, "learning_rate": 0.0002, "epoch": 3.8102353380649983, "step": 5100}, {"loss": 1.4972, "grad_norm": 0.5702024102210999, "learning_rate": 0.0002, "epoch": 3.8177063877474784, "step": 5110}, {"loss": 1.5471, "grad_norm": 0.5436882376670837, "learning_rate": 0.0002, "epoch": 3.825177437429959, "step": 5120}, {"loss": 1.5118, "grad_norm": 0.5453617572784424, "learning_rate": 0.0002, "epoch": 3.832648487112439, "step": 5130}, {"loss": 1.5732, "grad_norm": 0.6269069314002991, "learning_rate": 0.0002, "epoch": 3.8401195367949197, "step": 5140}, {"loss": 1.4959, "grad_norm": 0.6189185380935669, "learning_rate": 0.0002, "epoch": 3.8475905864774003, "step": 5150}, {"loss": 1.4999, "grad_norm": 0.6653388142585754, "learning_rate": 0.0002, "epoch": 3.8550616361598804, "step": 5160}, {"loss": 1.5075, "grad_norm": 0.5771768689155579, "learning_rate": 0.0002, "epoch": 3.862532685842361, "step": 5170}, {"loss": 1.5545, "grad_norm": 0.6052790880203247, "learning_rate": 0.0002, "epoch": 3.870003735524841, "step": 5180}, {"loss": 1.4987, "grad_norm": 0.6572316884994507, "learning_rate": 0.0002, "epoch": 3.8774747852073217, "step": 5190}, {"loss": 1.5241, "grad_norm": 0.670576810836792, "learning_rate": 0.0002, "epoch": 3.884945834889802, "step": 5200}, {"loss": 1.4777, "grad_norm": 0.5728798508644104, "learning_rate": 0.0002, "epoch": 3.8924168845722824, "step": 5210}, {"loss": 1.5351, "grad_norm": 0.6340774297714233, "learning_rate": 0.0002, "epoch": 3.899887934254763, "step": 5220}, {"loss": 1.5081, "grad_norm": 0.5981315970420837, "learning_rate": 0.0002, "epoch": 3.907358983937243, "step": 5230}, {"loss": 1.4875, "grad_norm": 0.6212025880813599, "learning_rate": 0.0002, "epoch": 3.9148300336197237, "step": 5240}, {"loss": 1.5545, "grad_norm": 0.6202296018600464, "learning_rate": 0.0002, "epoch": 3.922301083302204, "step": 5250}, {"loss": 1.5765, "grad_norm": 0.6159142255783081, "learning_rate": 0.0002, "epoch": 3.9297721329846844, "step": 5260}, {"loss": 1.4938, "grad_norm": 0.6519438624382019, "learning_rate": 0.0002, "epoch": 3.9372431826671646, "step": 5270}, {"loss": 1.4859, "grad_norm": 0.539813756942749, "learning_rate": 0.0002, "epoch": 3.944714232349645, "step": 5280}, {"loss": 1.5921, "grad_norm": 0.6443665027618408, "learning_rate": 0.0002, "epoch": 3.9521852820321257, "step": 5290}, {"loss": 1.5153, "grad_norm": 0.6635757684707642, "learning_rate": 0.0002, "epoch": 3.959656331714606, "step": 5300}, {"loss": 1.5485, "grad_norm": 0.589363157749176, "learning_rate": 0.0002, "epoch": 3.9671273813970864, "step": 5310}, {"loss": 1.5498, "grad_norm": 0.5788735747337341, "learning_rate": 0.0002, "epoch": 3.9745984310795666, "step": 5320}, {"loss": 1.5607, "grad_norm": 0.5976864695549011, "learning_rate": 0.0002, "epoch": 3.982069480762047, "step": 5330}, {"loss": 1.5302, "grad_norm": 0.6624067425727844, "learning_rate": 0.0002, "epoch": 3.9895405304445273, "step": 5340}, {"loss": 1.5904, "grad_norm": 0.6738956570625305, "learning_rate": 0.0002, "epoch": 3.997011580127008, "step": 5350}, {"eval_loss": 1.868006944656372, "eval_runtime": 38.5153, "eval_samples_per_second": 13.371, "eval_steps_per_second": 1.688, "epoch": 4.0, "step": 5354}, {"loss": 1.4535, "grad_norm": 0.6023468971252441, "learning_rate": 0.0002, "epoch": 4.004482629809488, "step": 5360}, {"loss": 1.3987, "grad_norm": 0.8589285612106323, "learning_rate": 0.0002, "epoch": 4.011953679491969, "step": 5370}, {"loss": 1.3952, "grad_norm": 0.7477491497993469, "learning_rate": 0.0002, "epoch": 4.019424729174449, "step": 5380}, {"loss": 1.3745, "grad_norm": 0.7601922154426575, "learning_rate": 0.0002, "epoch": 4.02689577885693, "step": 5390}, {"loss": 1.4133, "grad_norm": 0.8115614056587219, "learning_rate": 0.0002, "epoch": 4.03436682853941, "step": 5400}, {"loss": 1.3748, "grad_norm": 0.669925332069397, "learning_rate": 0.0002, "epoch": 4.04183787822189, "step": 5410}, {"loss": 1.2835, "grad_norm": 0.8091904520988464, "learning_rate": 0.0002, "epoch": 4.04930892790437, "step": 5420}, {"loss": 1.3615, "grad_norm": 0.709405779838562, "learning_rate": 0.0002, "epoch": 4.056779977586851, "step": 5430}, {"loss": 1.3558, "grad_norm": 1.0006179809570312, "learning_rate": 0.0002, "epoch": 4.064251027269331, "step": 5440}, {"loss": 1.3491, "grad_norm": 0.7017965912818909, "learning_rate": 0.0002, "epoch": 4.071722076951811, "step": 5450}, {"loss": 1.3642, "grad_norm": 0.8991572260856628, "learning_rate": 0.0002, "epoch": 4.0791931266342925, "step": 5460}, {"loss": 1.392, "grad_norm": 0.9064797759056091, "learning_rate": 0.0002, "epoch": 4.086664176316773, "step": 5470}, {"loss": 1.3425, "grad_norm": 0.7981749176979065, "learning_rate": 0.0002, "epoch": 4.094135225999253, "step": 5480}, {"loss": 1.3826, "grad_norm": 0.7280883193016052, "learning_rate": 0.0002, "epoch": 4.101606275681733, "step": 5490}, {"loss": 1.3275, "grad_norm": 0.7419600486755371, "learning_rate": 0.0002, "epoch": 4.109077325364214, "step": 5500}, {"loss": 1.3199, "grad_norm": 0.8019949197769165, "learning_rate": 0.0002, "epoch": 4.116548375046694, "step": 5510}, {"loss": 1.3133, "grad_norm": 0.7501229047775269, "learning_rate": 0.0002, "epoch": 4.124019424729174, "step": 5520}, {"loss": 1.4432, "grad_norm": 0.8166249990463257, "learning_rate": 0.0002, "epoch": 4.131490474411655, "step": 5530}, {"loss": 1.3901, "grad_norm": 0.9728496074676514, "learning_rate": 0.0002, "epoch": 4.138961524094135, "step": 5540}, {"loss": 1.3538, "grad_norm": 0.7590922117233276, "learning_rate": 0.0002, "epoch": 4.1464325737766154, "step": 5550}, {"loss": 1.4368, "grad_norm": 0.7759010791778564, "learning_rate": 0.0002, "epoch": 4.153903623459096, "step": 5560}, {"loss": 1.3635, "grad_norm": 0.9057986736297607, "learning_rate": 0.0002, "epoch": 4.161374673141577, "step": 5570}, {"loss": 1.4152, "grad_norm": 0.8853937983512878, "learning_rate": 0.0002, "epoch": 4.168845722824057, "step": 5580}, {"loss": 1.3633, "grad_norm": 0.7070684432983398, "learning_rate": 0.0002, "epoch": 4.176316772506537, "step": 5590}, {"loss": 1.3218, "grad_norm": 0.7649410963058472, "learning_rate": 0.0002, "epoch": 4.183787822189018, "step": 5600}, {"loss": 1.3857, "grad_norm": 1.2048029899597168, "learning_rate": 0.0002, "epoch": 4.191258871871498, "step": 5610}, {"loss": 1.3629, "grad_norm": 0.7986605763435364, "learning_rate": 0.0002, "epoch": 4.198729921553978, "step": 5620}, {"loss": 1.3995, "grad_norm": 0.8151885867118835, "learning_rate": 0.0002, "epoch": 4.206200971236458, "step": 5630}, {"loss": 1.3782, "grad_norm": 0.7719064354896545, "learning_rate": 0.0002, "epoch": 4.213672020918939, "step": 5640}, {"loss": 1.3852, "grad_norm": 0.8422448039054871, "learning_rate": 0.0002, "epoch": 4.2211430706014195, "step": 5650}, {"loss": 1.3321, "grad_norm": 0.7017164826393127, "learning_rate": 0.0002, "epoch": 4.2286141202839, "step": 5660}, {"loss": 1.4105, "grad_norm": 0.8559677600860596, "learning_rate": 0.0002, "epoch": 4.236085169966381, "step": 5670}, {"loss": 1.3701, "grad_norm": 0.8216157555580139, "learning_rate": 0.0002, "epoch": 4.243556219648861, "step": 5680}, {"loss": 1.3565, "grad_norm": 0.7681755423545837, "learning_rate": 0.0002, "epoch": 4.251027269331341, "step": 5690}, {"loss": 1.3806, "grad_norm": 0.811665952205658, "learning_rate": 0.0002, "epoch": 4.258498319013821, "step": 5700}, {"loss": 1.4161, "grad_norm": 0.7242204546928406, "learning_rate": 0.0002, "epoch": 4.265969368696302, "step": 5710}, {"loss": 1.2958, "grad_norm": 0.7570181488990784, "learning_rate": 0.0002, "epoch": 4.273440418378782, "step": 5720}, {"loss": 1.4265, "grad_norm": 0.8951969146728516, "learning_rate": 0.0002, "epoch": 4.280911468061262, "step": 5730}, {"loss": 1.3895, "grad_norm": 0.7222902178764343, "learning_rate": 0.0002, "epoch": 4.288382517743743, "step": 5740}, {"loss": 1.4155, "grad_norm": 0.8508469462394714, "learning_rate": 0.0002, "epoch": 4.2958535674262235, "step": 5750}, {"loss": 1.365, "grad_norm": 0.7215430736541748, "learning_rate": 0.0002, "epoch": 4.303324617108704, "step": 5760}, {"loss": 1.4472, "grad_norm": 0.8774884939193726, "learning_rate": 0.0002, "epoch": 4.310795666791184, "step": 5770}, {"loss": 1.427, "grad_norm": 0.8354552984237671, "learning_rate": 0.0002, "epoch": 4.318266716473665, "step": 5780}, {"loss": 1.3222, "grad_norm": 0.6938814520835876, "learning_rate": 0.0002, "epoch": 4.325737766156145, "step": 5790}, {"loss": 1.3589, "grad_norm": 0.78675377368927, "learning_rate": 0.0002, "epoch": 4.333208815838625, "step": 5800}, {"loss": 1.3662, "grad_norm": 0.7147697806358337, "learning_rate": 0.0002, "epoch": 4.340679865521106, "step": 5810}, {"loss": 1.3597, "grad_norm": 0.7693623304367065, "learning_rate": 0.0002, "epoch": 4.348150915203586, "step": 5820}, {"loss": 1.2944, "grad_norm": 0.856517493724823, "learning_rate": 0.0002, "epoch": 4.355621964886066, "step": 5830}, {"loss": 1.4307, "grad_norm": 0.7200973033905029, "learning_rate": 0.0002, "epoch": 4.3630930145685465, "step": 5840}, {"loss": 1.442, "grad_norm": 0.743281364440918, "learning_rate": 0.0002, "epoch": 4.3705640642510275, "step": 5850}, {"loss": 1.3999, "grad_norm": 0.7627727389335632, "learning_rate": 0.0002, "epoch": 4.378035113933508, "step": 5860}, {"loss": 1.4082, "grad_norm": 0.7238836884498596, "learning_rate": 0.0002, "epoch": 4.385506163615988, "step": 5870}, {"loss": 1.4292, "grad_norm": 0.7253410816192627, "learning_rate": 0.0002, "epoch": 4.392977213298469, "step": 5880}, {"loss": 1.3774, "grad_norm": 0.8232238292694092, "learning_rate": 0.0002, "epoch": 4.400448262980949, "step": 5890}, {"loss": 1.3757, "grad_norm": 0.8778504729270935, "learning_rate": 0.0002, "epoch": 4.407919312663429, "step": 5900}, {"loss": 1.387, "grad_norm": 0.7639474868774414, "learning_rate": 0.0002, "epoch": 4.415390362345909, "step": 5910}, {"loss": 1.3862, "grad_norm": 0.7666519284248352, "learning_rate": 0.0002, "epoch": 4.42286141202839, "step": 5920}, {"loss": 1.4168, "grad_norm": 0.867132842540741, "learning_rate": 0.0002, "epoch": 4.43033246171087, "step": 5930}, {"loss": 1.4772, "grad_norm": 0.7571166753768921, "learning_rate": 0.0002, "epoch": 4.4378035113933505, "step": 5940}, {"loss": 1.4401, "grad_norm": 0.7911370992660522, "learning_rate": 0.0002, "epoch": 4.4452745610758315, "step": 5950}, {"loss": 1.4516, "grad_norm": 0.8844250440597534, "learning_rate": 0.0002, "epoch": 4.452745610758312, "step": 5960}, {"loss": 1.4109, "grad_norm": 0.7336231470108032, "learning_rate": 0.0002, "epoch": 4.460216660440792, "step": 5970}, {"loss": 1.3891, "grad_norm": 0.8162738084793091, "learning_rate": 0.0002, "epoch": 4.467687710123272, "step": 5980}, {"loss": 1.393, "grad_norm": 0.7413017153739929, "learning_rate": 0.0002, "epoch": 4.475158759805753, "step": 5990}, {"loss": 1.3712, "grad_norm": 0.7215432524681091, "learning_rate": 0.0002, "epoch": 4.482629809488233, "step": 6000}, {"loss": 1.3521, "grad_norm": 0.8943389058113098, "learning_rate": 0.0002, "epoch": 4.490100859170713, "step": 6010}, {"loss": 1.4172, "grad_norm": 0.7850823998451233, "learning_rate": 0.0002, "epoch": 4.497571908853194, "step": 6020}, {"loss": 1.3582, "grad_norm": 0.8117504119873047, "learning_rate": 0.0002, "epoch": 4.505042958535674, "step": 6030}, {"loss": 1.4272, "grad_norm": 0.8381605744361877, "learning_rate": 0.0002, "epoch": 4.5125140082181545, "step": 6040}, {"loss": 1.3829, "grad_norm": 0.7964059710502625, "learning_rate": 0.0002, "epoch": 4.519985057900635, "step": 6050}, {"loss": 1.3555, "grad_norm": 0.7935128211975098, "learning_rate": 0.0002, "epoch": 4.527456107583116, "step": 6060}, {"loss": 1.3994, "grad_norm": 0.8725124597549438, "learning_rate": 0.0002, "epoch": 4.534927157265596, "step": 6070}, {"loss": 1.3923, "grad_norm": 0.880325198173523, "learning_rate": 0.0002, "epoch": 4.542398206948076, "step": 6080}, {"loss": 1.4459, "grad_norm": 0.7220637202262878, "learning_rate": 0.0002, "epoch": 4.549869256630557, "step": 6090}, {"loss": 1.3281, "grad_norm": 0.6908547878265381, "learning_rate": 0.0002, "epoch": 4.557340306313037, "step": 6100}, {"loss": 1.437, "grad_norm": 0.797931969165802, "learning_rate": 0.0002, "epoch": 4.564811355995517, "step": 6110}, {"loss": 1.4023, "grad_norm": 0.7056134343147278, "learning_rate": 0.0002, "epoch": 4.572282405677997, "step": 6120}, {"loss": 1.3814, "grad_norm": 0.7850478887557983, "learning_rate": 0.0002, "epoch": 4.579753455360478, "step": 6130}, {"loss": 1.3579, "grad_norm": 0.8112621307373047, "learning_rate": 0.0002, "epoch": 4.5872245050429585, "step": 6140}, {"loss": 1.3523, "grad_norm": 0.7040849328041077, "learning_rate": 0.0002, "epoch": 4.594695554725439, "step": 6150}, {"loss": 1.3526, "grad_norm": 0.7214553952217102, "learning_rate": 0.0002, "epoch": 4.60216660440792, "step": 6160}, {"loss": 1.3932, "grad_norm": 0.8616511821746826, "learning_rate": 0.0002, "epoch": 4.6096376540904, "step": 6170}, {"loss": 1.4622, "grad_norm": 0.8374658226966858, "learning_rate": 0.0002, "epoch": 4.61710870377288, "step": 6180}, {"loss": 1.3703, "grad_norm": 0.6761606931686401, "learning_rate": 0.0002, "epoch": 4.62457975345536, "step": 6190}, {"loss": 1.3977, "grad_norm": 0.768028199672699, "learning_rate": 0.0002, "epoch": 4.632050803137841, "step": 6200}, {"loss": 1.3772, "grad_norm": 0.9372717142105103, "learning_rate": 0.0002, "epoch": 4.639521852820321, "step": 6210}, {"loss": 1.4098, "grad_norm": 0.7906546592712402, "learning_rate": 0.0002, "epoch": 4.646992902502801, "step": 6220}, {"loss": 1.3962, "grad_norm": 0.7376723289489746, "learning_rate": 0.0002, "epoch": 4.654463952185282, "step": 6230}, {"loss": 1.4529, "grad_norm": 0.8972630500793457, "learning_rate": 0.0002, "epoch": 4.6619350018677626, "step": 6240}, {"loss": 1.4668, "grad_norm": 0.8261756300926208, "learning_rate": 0.0002, "epoch": 4.669406051550243, "step": 6250}, {"loss": 1.3267, "grad_norm": 0.7512393593788147, "learning_rate": 0.0002, "epoch": 4.676877101232723, "step": 6260}, {"loss": 1.4278, "grad_norm": 0.7132362127304077, "learning_rate": 0.0002, "epoch": 4.684348150915204, "step": 6270}, {"loss": 1.4299, "grad_norm": 0.7690575122833252, "learning_rate": 0.0002, "epoch": 4.691819200597684, "step": 6280}, {"loss": 1.4769, "grad_norm": 0.9886258840560913, "learning_rate": 0.0002, "epoch": 4.699290250280164, "step": 6290}, {"loss": 1.4005, "grad_norm": 0.9502435922622681, "learning_rate": 0.0002, "epoch": 4.706761299962645, "step": 6300}, {"loss": 1.4319, "grad_norm": 0.702255129814148, "learning_rate": 0.0002, "epoch": 4.714232349645125, "step": 6310}, {"loss": 1.4447, "grad_norm": 0.7713103890419006, "learning_rate": 0.0002, "epoch": 4.721703399327605, "step": 6320}, {"loss": 1.4392, "grad_norm": 0.7778580784797668, "learning_rate": 0.0002, "epoch": 4.7291744490100855, "step": 6330}, {"loss": 1.4169, "grad_norm": 0.7275111079216003, "learning_rate": 0.0002, "epoch": 4.736645498692567, "step": 6340}, {"loss": 1.4429, "grad_norm": 0.7728744149208069, "learning_rate": 0.0002, "epoch": 4.744116548375047, "step": 6350}, {"loss": 1.3756, "grad_norm": 0.9724260568618774, "learning_rate": 0.0002, "epoch": 4.751587598057527, "step": 6360}, {"loss": 1.3358, "grad_norm": 0.7505622506141663, "learning_rate": 0.0002, "epoch": 4.759058647740007, "step": 6370}, {"loss": 1.379, "grad_norm": 0.7994682788848877, "learning_rate": 0.0002, "epoch": 4.766529697422488, "step": 6380}, {"loss": 1.4275, "grad_norm": 0.8432038426399231, "learning_rate": 0.0002, "epoch": 4.774000747104968, "step": 6390}, {"loss": 1.4606, "grad_norm": 0.7436022758483887, "learning_rate": 0.0002, "epoch": 4.781471796787448, "step": 6400}, {"loss": 1.3461, "grad_norm": 0.7709194421768188, "learning_rate": 0.0002, "epoch": 4.788942846469929, "step": 6410}, {"loss": 1.3715, "grad_norm": 0.8798436522483826, "learning_rate": 0.0002, "epoch": 4.796413896152409, "step": 6420}, {"loss": 1.3761, "grad_norm": 0.790189266204834, "learning_rate": 0.0002, "epoch": 4.80388494583489, "step": 6430}, {"loss": 1.4109, "grad_norm": 0.6824303865432739, "learning_rate": 0.0002, "epoch": 4.811355995517371, "step": 6440}, {"loss": 1.3877, "grad_norm": 0.7501044869422913, "learning_rate": 0.0002, "epoch": 4.818827045199851, "step": 6450}, {"loss": 1.4458, "grad_norm": 0.8840398192405701, "learning_rate": 0.0002, "epoch": 4.826298094882331, "step": 6460}, {"loss": 1.4412, "grad_norm": 0.7812688946723938, "learning_rate": 0.0002, "epoch": 4.833769144564811, "step": 6470}, {"loss": 1.4299, "grad_norm": 0.7429926991462708, "learning_rate": 0.0002, "epoch": 4.841240194247292, "step": 6480}, {"loss": 1.5062, "grad_norm": 0.7778021693229675, "learning_rate": 0.0002, "epoch": 4.848711243929772, "step": 6490}, {"loss": 1.4589, "grad_norm": 0.8270702362060547, "learning_rate": 0.0002, "epoch": 4.856182293612252, "step": 6500}, {"loss": 1.4091, "grad_norm": 0.6960513591766357, "learning_rate": 0.0002, "epoch": 4.863653343294732, "step": 6510}, {"loss": 1.376, "grad_norm": 0.7728942632675171, "learning_rate": 0.0002, "epoch": 4.8711243929772134, "step": 6520}, {"loss": 1.4852, "grad_norm": 0.7377303838729858, "learning_rate": 0.0002, "epoch": 4.878595442659694, "step": 6530}, {"loss": 1.3846, "grad_norm": 0.7257253527641296, "learning_rate": 0.0002, "epoch": 4.886066492342174, "step": 6540}, {"loss": 1.4166, "grad_norm": 0.7875821590423584, "learning_rate": 0.0002, "epoch": 4.893537542024655, "step": 6550}, {"loss": 1.357, "grad_norm": 0.8346304297447205, "learning_rate": 0.0002, "epoch": 4.901008591707135, "step": 6560}, {"loss": 1.4522, "grad_norm": 0.7710739374160767, "learning_rate": 0.0002, "epoch": 4.908479641389615, "step": 6570}, {"loss": 1.4465, "grad_norm": 0.7015138268470764, "learning_rate": 0.0002, "epoch": 4.915950691072096, "step": 6580}, {"loss": 1.435, "grad_norm": 0.8707432150840759, "learning_rate": 0.0002, "epoch": 4.923421740754576, "step": 6590}, {"loss": 1.2968, "grad_norm": 0.786601185798645, "learning_rate": 0.0002, "epoch": 4.930892790437056, "step": 6600}, {"loss": 1.4385, "grad_norm": 0.978519082069397, "learning_rate": 0.0002, "epoch": 4.938363840119536, "step": 6610}, {"loss": 1.3997, "grad_norm": 0.8102927207946777, "learning_rate": 0.0002, "epoch": 4.9458348898020175, "step": 6620}, {"loss": 1.4859, "grad_norm": 0.7628704309463501, "learning_rate": 0.0002, "epoch": 4.953305939484498, "step": 6630}, {"loss": 1.3774, "grad_norm": 0.8053455352783203, "learning_rate": 0.0002, "epoch": 4.960776989166978, "step": 6640}, {"loss": 1.5092, "grad_norm": 0.8680412173271179, "learning_rate": 0.0002, "epoch": 4.968248038849458, "step": 6650}, {"loss": 1.3978, "grad_norm": 0.7415758371353149, "learning_rate": 0.0002, "epoch": 4.975719088531939, "step": 6660}, {"loss": 1.3793, "grad_norm": 0.7730312347412109, "learning_rate": 0.0002, "epoch": 4.983190138214419, "step": 6670}, {"loss": 1.4863, "grad_norm": 0.7924041152000427, "learning_rate": 0.0002, "epoch": 4.990661187896899, "step": 6680}, {"loss": 1.4137, "grad_norm": 0.8677893877029419, "learning_rate": 0.0002, "epoch": 4.99813223757938, "step": 6690}, {"eval_loss": 1.9444633722305298, "eval_runtime": 39.3488, "eval_samples_per_second": 13.088, "eval_steps_per_second": 1.652, "epoch": 4.999626447515876, "step": 6692}, {"loss": 1.3076, "grad_norm": 0.7102245092391968, "learning_rate": 0.0002, "epoch": 5.00560328726186, "step": 6700}, {"loss": 1.2714, "grad_norm": 1.0425463914871216, "learning_rate": 0.0002, "epoch": 5.0130743369443405, "step": 6710}, {"loss": 1.181, "grad_norm": 0.9320756793022156, "learning_rate": 0.0002, "epoch": 5.0205453866268215, "step": 6720}, {"loss": 1.1786, "grad_norm": 0.8797217607498169, "learning_rate": 0.0002, "epoch": 5.028016436309302, "step": 6730}, {"loss": 1.2097, "grad_norm": 2.135707139968872, "learning_rate": 0.0002, "epoch": 5.035487485991782, "step": 6740}, {"loss": 1.1761, "grad_norm": 0.8747734427452087, "learning_rate": 0.0002, "epoch": 5.042958535674262, "step": 6750}, {"loss": 1.1675, "grad_norm": 0.9981076717376709, "learning_rate": 0.0002, "epoch": 5.050429585356743, "step": 6760}, {"loss": 1.1976, "grad_norm": 0.985078752040863, "learning_rate": 0.0002, "epoch": 5.057900635039223, "step": 6770}, {"loss": 1.2688, "grad_norm": 1.0974019765853882, "learning_rate": 0.0002, "epoch": 5.065371684721703, "step": 6780}, {"loss": 1.1982, "grad_norm": 0.9823219180107117, "learning_rate": 0.0002, "epoch": 5.072842734404184, "step": 6790}, {"loss": 1.2586, "grad_norm": 1.122605562210083, "learning_rate": 0.0002, "epoch": 5.080313784086664, "step": 6800}, {"loss": 1.2069, "grad_norm": 0.8556802272796631, "learning_rate": 0.0002, "epoch": 5.0877848337691445, "step": 6810}, {"loss": 1.1908, "grad_norm": 1.1699262857437134, "learning_rate": 0.0002, "epoch": 5.095255883451625, "step": 6820}, {"loss": 1.1869, "grad_norm": 1.0440590381622314, "learning_rate": 0.0002, "epoch": 5.102726933134106, "step": 6830}, {"loss": 1.1655, "grad_norm": 1.0445300340652466, "learning_rate": 0.0002, "epoch": 5.110197982816586, "step": 6840}, {"loss": 1.2392, "grad_norm": 0.8289563059806824, "learning_rate": 0.0002, "epoch": 5.117669032499066, "step": 6850}, {"loss": 1.1687, "grad_norm": 1.1051193475723267, "learning_rate": 0.0002, "epoch": 5.125140082181547, "step": 6860}, {"loss": 1.2737, "grad_norm": 0.9345614910125732, "learning_rate": 0.0002, "epoch": 5.132611131864027, "step": 6870}, {"loss": 1.3021, "grad_norm": 1.1222996711730957, "learning_rate": 0.0002, "epoch": 5.140082181546507, "step": 6880}, {"loss": 1.2408, "grad_norm": 0.9405338764190674, "learning_rate": 0.0002, "epoch": 5.147553231228987, "step": 6890}, {"loss": 1.2367, "grad_norm": 1.0935171842575073, "learning_rate": 0.0002, "epoch": 5.155024280911468, "step": 6900}, {"loss": 1.2458, "grad_norm": 1.0438612699508667, "learning_rate": 0.0002, "epoch": 5.1624953305939485, "step": 6910}, {"loss": 1.2562, "grad_norm": 1.1189004182815552, "learning_rate": 0.0002, "epoch": 5.169966380276429, "step": 6920}, {"loss": 1.25, "grad_norm": 1.0533215999603271, "learning_rate": 0.0002, "epoch": 5.17743742995891, "step": 6930}, {"loss": 1.2974, "grad_norm": 0.9779648780822754, "learning_rate": 0.0002, "epoch": 5.18490847964139, "step": 6940}, {"loss": 1.1965, "grad_norm": 0.8920868635177612, "learning_rate": 0.0002, "epoch": 5.19237952932387, "step": 6950}, {"loss": 1.283, "grad_norm": 0.8374548554420471, "learning_rate": 0.0002, "epoch": 5.19985057900635, "step": 6960}, {"loss": 1.2775, "grad_norm": 1.0490682125091553, "learning_rate": 0.0002, "epoch": 5.207321628688831, "step": 6970}, {"loss": 1.1826, "grad_norm": 0.9658287167549133, "learning_rate": 0.0002, "epoch": 5.214792678371311, "step": 6980}, {"loss": 1.2647, "grad_norm": 0.9652056097984314, "learning_rate": 0.0002, "epoch": 5.222263728053791, "step": 6990}, {"loss": 1.3023, "grad_norm": 0.9141794443130493, "learning_rate": 0.0002, "epoch": 5.229734777736272, "step": 7000}, {"loss": 1.2456, "grad_norm": 0.9831376671791077, "learning_rate": 0.0002, "epoch": 5.2372058274187525, "step": 7010}, {"loss": 1.2176, "grad_norm": 1.0198718309402466, "learning_rate": 0.0002, "epoch": 5.244676877101233, "step": 7020}, {"loss": 1.2643, "grad_norm": 0.9647888541221619, "learning_rate": 0.0002, "epoch": 5.252147926783713, "step": 7030}, {"loss": 1.2106, "grad_norm": 1.3941649198532104, "learning_rate": 0.0002, "epoch": 5.259618976466194, "step": 7040}, {"loss": 1.2885, "grad_norm": 1.0305466651916504, "learning_rate": 0.0002, "epoch": 5.267090026148674, "step": 7050}, {"loss": 1.2362, "grad_norm": 0.9577859044075012, "learning_rate": 0.0002, "epoch": 5.274561075831154, "step": 7060}, {"loss": 1.2231, "grad_norm": 1.149092197418213, "learning_rate": 0.0002, "epoch": 5.282032125513634, "step": 7070}, {"loss": 1.2986, "grad_norm": 1.2582733631134033, "learning_rate": 0.0002, "epoch": 5.289503175196115, "step": 7080}, {"loss": 1.2307, "grad_norm": 1.1777442693710327, "learning_rate": 0.0002, "epoch": 5.296974224878595, "step": 7090}, {"loss": 1.24, "grad_norm": 1.0076404809951782, "learning_rate": 0.0002, "epoch": 5.3044452745610755, "step": 7100}, {"loss": 1.1407, "grad_norm": 0.9037365913391113, "learning_rate": 0.0002, "epoch": 5.3119163242435565, "step": 7110}, {"loss": 1.238, "grad_norm": 0.9428724646568298, "learning_rate": 0.0002, "epoch": 5.319387373926037, "step": 7120}, {"loss": 1.2571, "grad_norm": 0.9935154318809509, "learning_rate": 0.0002, "epoch": 5.326858423608517, "step": 7130}, {"loss": 1.2833, "grad_norm": 1.087500810623169, "learning_rate": 0.0002, "epoch": 5.334329473290998, "step": 7140}, {"loss": 1.2304, "grad_norm": 0.8543072938919067, "learning_rate": 0.0002, "epoch": 5.341800522973478, "step": 7150}, {"loss": 1.2755, "grad_norm": 0.9323700070381165, "learning_rate": 0.0002, "epoch": 5.349271572655958, "step": 7160}, {"loss": 1.2769, "grad_norm": 1.0037827491760254, "learning_rate": 0.0002, "epoch": 5.356742622338438, "step": 7170}, {"loss": 1.3204, "grad_norm": 0.8746469616889954, "learning_rate": 0.0002, "epoch": 5.364213672020919, "step": 7180}, {"loss": 1.2759, "grad_norm": 0.9516328573226929, "learning_rate": 0.0002, "epoch": 5.371684721703399, "step": 7190}, {"loss": 1.2428, "grad_norm": 0.9395177364349365, "learning_rate": 0.0002, "epoch": 5.3791557713858795, "step": 7200}, {"loss": 1.3214, "grad_norm": 1.000369906425476, "learning_rate": 0.0002, "epoch": 5.38662682106836, "step": 7210}, {"loss": 1.2337, "grad_norm": 1.0845502614974976, "learning_rate": 0.0002, "epoch": 5.394097870750841, "step": 7220}, {"loss": 1.2776, "grad_norm": 0.8975145220756531, "learning_rate": 0.0002, "epoch": 5.401568920433321, "step": 7230}, {"loss": 1.2306, "grad_norm": 1.040077805519104, "learning_rate": 0.0002, "epoch": 5.409039970115801, "step": 7240}, {"loss": 1.2277, "grad_norm": 1.0729942321777344, "learning_rate": 0.0002, "epoch": 5.416511019798282, "step": 7250}, {"loss": 1.2714, "grad_norm": 0.8322232961654663, "learning_rate": 0.0002, "epoch": 5.423982069480762, "step": 7260}, {"loss": 1.3036, "grad_norm": 1.0654641389846802, "learning_rate": 0.0002, "epoch": 5.431453119163242, "step": 7270}, {"loss": 1.268, "grad_norm": 1.0445852279663086, "learning_rate": 0.0002, "epoch": 5.438924168845723, "step": 7280}, {"loss": 1.2743, "grad_norm": 1.0762956142425537, "learning_rate": 0.0002, "epoch": 5.446395218528203, "step": 7290}, {"loss": 1.2887, "grad_norm": 0.9721953868865967, "learning_rate": 0.0002, "epoch": 5.4538662682106835, "step": 7300}, {"loss": 1.2833, "grad_norm": 0.9238539338111877, "learning_rate": 0.0002, "epoch": 5.461337317893164, "step": 7310}, {"loss": 1.255, "grad_norm": 0.9912874102592468, "learning_rate": 0.0002, "epoch": 5.468808367575645, "step": 7320}, {"loss": 1.2557, "grad_norm": 1.0727077722549438, "learning_rate": 0.0002, "epoch": 5.476279417258125, "step": 7330}, {"loss": 1.3471, "grad_norm": 0.8633865118026733, "learning_rate": 0.0002, "epoch": 5.483750466940605, "step": 7340}, {"loss": 1.3155, "grad_norm": 0.9396262764930725, "learning_rate": 0.0002, "epoch": 5.491221516623085, "step": 7350}, {"loss": 1.3146, "grad_norm": 1.0253715515136719, "learning_rate": 0.0002, "epoch": 5.498692566305566, "step": 7360}, {"loss": 1.3156, "grad_norm": 1.006047010421753, "learning_rate": 0.0002, "epoch": 5.506163615988046, "step": 7370}, {"loss": 1.3107, "grad_norm": 0.9781233072280884, "learning_rate": 0.0002, "epoch": 5.513634665670526, "step": 7380}, {"loss": 1.2703, "grad_norm": 0.9945126175880432, "learning_rate": 0.0002, "epoch": 5.521105715353007, "step": 7390}, {"loss": 1.1936, "grad_norm": 0.9081175327301025, "learning_rate": 0.0002, "epoch": 5.528576765035488, "step": 7400}, {"loss": 1.2651, "grad_norm": 1.2215938568115234, "learning_rate": 0.0002, "epoch": 5.536047814717968, "step": 7410}, {"loss": 1.2484, "grad_norm": 1.0724077224731445, "learning_rate": 0.0002, "epoch": 5.543518864400449, "step": 7420}, {"loss": 1.3083, "grad_norm": 1.106955885887146, "learning_rate": 0.0002, "epoch": 5.550989914082929, "step": 7430}, {"loss": 1.2125, "grad_norm": 1.0657650232315063, "learning_rate": 0.0002, "epoch": 5.558460963765409, "step": 7440}, {"loss": 1.2576, "grad_norm": 0.9725455641746521, "learning_rate": 0.0002, "epoch": 5.565932013447889, "step": 7450}, {"loss": 1.3297, "grad_norm": 0.8604224324226379, "learning_rate": 0.0002, "epoch": 5.57340306313037, "step": 7460}, {"loss": 1.3084, "grad_norm": 0.9913371205329895, "learning_rate": 0.0002, "epoch": 5.58087411281285, "step": 7470}, {"loss": 1.3371, "grad_norm": 1.012073040008545, "learning_rate": 0.0002, "epoch": 5.58834516249533, "step": 7480}, {"loss": 1.2526, "grad_norm": 1.1003159284591675, "learning_rate": 0.0002, "epoch": 5.5958162121778106, "step": 7490}, {"loss": 1.2577, "grad_norm": 0.9104593992233276, "learning_rate": 0.0002, "epoch": 5.603287261860292, "step": 7500}, {"loss": 1.2578, "grad_norm": 0.9480831623077393, "learning_rate": 0.0002, "epoch": 5.610758311542772, "step": 7510}, {"loss": 1.3056, "grad_norm": 1.0826456546783447, "learning_rate": 0.0002, "epoch": 5.618229361225252, "step": 7520}, {"loss": 1.2931, "grad_norm": 0.8286259174346924, "learning_rate": 0.0002, "epoch": 5.625700410907733, "step": 7530}, {"loss": 1.2918, "grad_norm": 0.9145061373710632, "learning_rate": 0.0002, "epoch": 5.633171460590213, "step": 7540}, {"loss": 1.1736, "grad_norm": 0.9363601803779602, "learning_rate": 0.0002, "epoch": 5.640642510272693, "step": 7550}, {"loss": 1.2265, "grad_norm": 0.9553244709968567, "learning_rate": 0.0002, "epoch": 5.648113559955174, "step": 7560}, {"loss": 1.2356, "grad_norm": 1.0343557596206665, "learning_rate": 0.0002, "epoch": 5.655584609637654, "step": 7570}, {"loss": 1.3171, "grad_norm": 0.8734238743782043, "learning_rate": 0.0002, "epoch": 5.663055659320134, "step": 7580}, {"loss": 1.2785, "grad_norm": 1.0230586528778076, "learning_rate": 0.0002, "epoch": 5.670526709002615, "step": 7590}, {"loss": 1.2936, "grad_norm": 1.0063409805297852, "learning_rate": 0.0002, "epoch": 5.677997758685096, "step": 7600}, {"loss": 1.2396, "grad_norm": 1.0104626417160034, "learning_rate": 0.0002, "epoch": 5.685468808367576, "step": 7610}, {"loss": 1.2581, "grad_norm": 0.9528168439865112, "learning_rate": 0.0002, "epoch": 5.692939858050056, "step": 7620}, {"loss": 1.3116, "grad_norm": 0.9799878597259521, "learning_rate": 0.0002, "epoch": 5.700410907732536, "step": 7630}, {"loss": 1.2632, "grad_norm": 0.969351589679718, "learning_rate": 0.0002, "epoch": 5.707881957415017, "step": 7640}, {"loss": 1.3055, "grad_norm": 1.3037652969360352, "learning_rate": 0.0002, "epoch": 5.715353007097497, "step": 7650}, {"loss": 1.3126, "grad_norm": 1.0640486478805542, "learning_rate": 0.0002, "epoch": 5.722824056779977, "step": 7660}, {"loss": 1.3325, "grad_norm": 1.0416420698165894, "learning_rate": 0.0002, "epoch": 5.730295106462458, "step": 7670}, {"loss": 1.25, "grad_norm": 0.8893619775772095, "learning_rate": 0.0002, "epoch": 5.7377661561449385, "step": 7680}, {"loss": 1.319, "grad_norm": 0.8512844443321228, "learning_rate": 0.0002, "epoch": 5.745237205827419, "step": 7690}, {"loss": 1.3328, "grad_norm": 0.9955748319625854, "learning_rate": 0.0002, "epoch": 5.7527082555099, "step": 7700}, {"loss": 1.294, "grad_norm": 1.0409910678863525, "learning_rate": 0.0002, "epoch": 5.76017930519238, "step": 7710}, {"loss": 1.3518, "grad_norm": 1.010097861289978, "learning_rate": 0.0002, "epoch": 5.76765035487486, "step": 7720}, {"loss": 1.2106, "grad_norm": 0.8974892497062683, "learning_rate": 0.0002, "epoch": 5.77512140455734, "step": 7730}, {"loss": 1.2743, "grad_norm": 0.972835123538971, "learning_rate": 0.0002, "epoch": 5.782592454239821, "step": 7740}, {"loss": 1.3549, "grad_norm": 0.9607440829277039, "learning_rate": 0.0002, "epoch": 5.790063503922301, "step": 7750}, {"loss": 1.29, "grad_norm": 0.9426500797271729, "learning_rate": 0.0002, "epoch": 5.797534553604781, "step": 7760}, {"loss": 1.274, "grad_norm": 0.8745320439338684, "learning_rate": 0.0002, "epoch": 5.8050056032872615, "step": 7770}, {"loss": 1.3009, "grad_norm": 1.0117204189300537, "learning_rate": 0.0002, "epoch": 5.8124766529697425, "step": 7780}, {"loss": 1.3135, "grad_norm": 1.0387755632400513, "learning_rate": 0.0002, "epoch": 5.819947702652223, "step": 7790}, {"loss": 1.2709, "grad_norm": 1.0709784030914307, "learning_rate": 0.0002, "epoch": 5.827418752334703, "step": 7800}, {"loss": 1.225, "grad_norm": 0.9512667655944824, "learning_rate": 0.0002, "epoch": 5.834889802017184, "step": 7810}, {"loss": 1.3284, "grad_norm": 1.021094560623169, "learning_rate": 0.0002, "epoch": 5.842360851699664, "step": 7820}, {"loss": 1.2794, "grad_norm": 1.117491364479065, "learning_rate": 0.0002, "epoch": 5.849831901382144, "step": 7830}, {"loss": 1.3646, "grad_norm": 0.9252554178237915, "learning_rate": 0.0002, "epoch": 5.857302951064625, "step": 7840}, {"loss": 1.2976, "grad_norm": 1.1416207551956177, "learning_rate": 0.0002, "epoch": 5.864774000747105, "step": 7850}, {"loss": 1.3293, "grad_norm": 1.1219907999038696, "learning_rate": 0.0002, "epoch": 5.872245050429585, "step": 7860}, {"loss": 1.2334, "grad_norm": 0.8300467729568481, "learning_rate": 0.0002, "epoch": 5.8797161001120655, "step": 7870}, {"loss": 1.3132, "grad_norm": 1.00551438331604, "learning_rate": 0.0002, "epoch": 5.8871871497945465, "step": 7880}, {"loss": 1.2609, "grad_norm": 0.8981153964996338, "learning_rate": 0.0002, "epoch": 5.894658199477027, "step": 7890}, {"loss": 1.2817, "grad_norm": 1.0247976779937744, "learning_rate": 0.0002, "epoch": 5.902129249159507, "step": 7900}, {"loss": 1.2866, "grad_norm": 1.0820319652557373, "learning_rate": 0.0002, "epoch": 5.909600298841987, "step": 7910}, {"loss": 1.2941, "grad_norm": 0.952675461769104, "learning_rate": 0.0002, "epoch": 5.917071348524468, "step": 7920}, {"loss": 1.307, "grad_norm": 0.8666740655899048, "learning_rate": 0.0002, "epoch": 5.924542398206948, "step": 7930}, {"loss": 1.2752, "grad_norm": 0.8640421032905579, "learning_rate": 0.0002, "epoch": 5.932013447889428, "step": 7940}, {"loss": 1.2386, "grad_norm": 1.2343276739120483, "learning_rate": 0.0002, "epoch": 5.939484497571909, "step": 7950}, {"loss": 1.2333, "grad_norm": 0.958046555519104, "learning_rate": 0.0002, "epoch": 5.946955547254389, "step": 7960}, {"loss": 1.2352, "grad_norm": 1.0538510084152222, "learning_rate": 0.0002, "epoch": 5.9544265969368695, "step": 7970}, {"loss": 1.3233, "grad_norm": 1.2681571245193481, "learning_rate": 0.0002, "epoch": 5.9618976466193505, "step": 7980}, {"loss": 1.2514, "grad_norm": 0.8171183466911316, "learning_rate": 0.0002, "epoch": 5.969368696301831, "step": 7990}, {"loss": 1.3412, "grad_norm": 0.9109523892402649, "learning_rate": 0.0002, "epoch": 5.976839745984311, "step": 8000}, {"loss": 1.3497, "grad_norm": 1.0040639638900757, "learning_rate": 0.0002, "epoch": 5.984310795666791, "step": 8010}, {"loss": 1.3299, "grad_norm": 0.9596554040908813, "learning_rate": 0.0002, "epoch": 5.991781845349272, "step": 8020}, {"loss": 1.3109, "grad_norm": 0.9782963991165161, "learning_rate": 0.0002, "epoch": 5.999252895031752, "step": 8030}, {"eval_loss": 2.0417845249176025, "eval_runtime": 38.8465, "eval_samples_per_second": 13.257, "eval_steps_per_second": 1.673, "epoch": 6.0, "step": 8031}, {"loss": 1.0886, "grad_norm": 1.380823016166687, "learning_rate": 0.0002, "epoch": 6.006723944714232, "step": 8040}, {"loss": 1.0413, "grad_norm": 1.067636251449585, "learning_rate": 0.0002, "epoch": 6.014194994396712, "step": 8050}, {"loss": 1.0686, "grad_norm": 1.363402009010315, "learning_rate": 0.0002, "epoch": 6.021666044079193, "step": 8060}, {"loss": 1.0762, "grad_norm": 0.9901054501533508, "learning_rate": 0.0002, "epoch": 6.0291370937616735, "step": 8070}, {"loss": 1.1182, "grad_norm": 1.1545379161834717, "learning_rate": 0.0002, "epoch": 6.036608143444154, "step": 8080}, {"loss": 1.0644, "grad_norm": 1.2259265184402466, "learning_rate": 0.0002, "epoch": 6.044079193126635, "step": 8090}, {"loss": 1.1273, "grad_norm": 1.1237425804138184, "learning_rate": 0.0002, "epoch": 6.051550242809115, "step": 8100}, {"loss": 1.1001, "grad_norm": 1.2805622816085815, "learning_rate": 0.0002, "epoch": 6.059021292491595, "step": 8110}, {"loss": 1.0731, "grad_norm": 1.2270452976226807, "learning_rate": 0.0002, "epoch": 6.066492342174075, "step": 8120}, {"loss": 1.0692, "grad_norm": 1.1924101114273071, "learning_rate": 0.0002, "epoch": 6.073963391856556, "step": 8130}, {"loss": 1.1698, "grad_norm": 1.2543894052505493, "learning_rate": 0.0002, "epoch": 6.081434441539036, "step": 8140}, {"loss": 1.069, "grad_norm": 1.1821149587631226, "learning_rate": 0.0002, "epoch": 6.088905491221516, "step": 8150}, {"loss": 1.109, "grad_norm": 1.2202836275100708, "learning_rate": 0.0002, "epoch": 6.096376540903997, "step": 8160}, {"loss": 1.136, "grad_norm": 1.0576019287109375, "learning_rate": 0.0002, "epoch": 6.1038475905864775, "step": 8170}, {"loss": 1.1395, "grad_norm": 1.31708824634552, "learning_rate": 0.0002, "epoch": 6.111318640268958, "step": 8180}, {"loss": 1.0887, "grad_norm": 1.0479495525360107, "learning_rate": 0.0002, "epoch": 6.118789689951438, "step": 8190}, {"loss": 1.0764, "grad_norm": 1.285003423690796, "learning_rate": 0.0002, "epoch": 6.126260739633919, "step": 8200}, {"loss": 1.0642, "grad_norm": 1.0989165306091309, "learning_rate": 0.0002, "epoch": 6.133731789316399, "step": 8210}, {"loss": 1.0981, "grad_norm": 1.1659013032913208, "learning_rate": 0.0002, "epoch": 6.141202838998879, "step": 8220}, {"loss": 1.1138, "grad_norm": 1.2796376943588257, "learning_rate": 0.0002, "epoch": 6.14867388868136, "step": 8230}, {"loss": 1.1116, "grad_norm": 1.060564637184143, "learning_rate": 0.0002, "epoch": 6.15614493836384, "step": 8240}, {"loss": 1.1493, "grad_norm": 1.3884605169296265, "learning_rate": 0.0002, "epoch": 6.16361598804632, "step": 8250}, {"loss": 1.0504, "grad_norm": 1.1570569276809692, "learning_rate": 0.0002, "epoch": 6.1710870377288005, "step": 8260}, {"loss": 1.0386, "grad_norm": 1.4136502742767334, "learning_rate": 0.0002, "epoch": 6.1785580874112815, "step": 8270}, {"loss": 1.0882, "grad_norm": 1.3396095037460327, "learning_rate": 0.0002, "epoch": 6.186029137093762, "step": 8280}, {"loss": 1.133, "grad_norm": 1.2549997568130493, "learning_rate": 0.0002, "epoch": 6.193500186776242, "step": 8290}, {"loss": 1.0626, "grad_norm": 1.3629751205444336, "learning_rate": 0.0002, "epoch": 6.200971236458723, "step": 8300}, {"loss": 1.1343, "grad_norm": 1.1029163599014282, "learning_rate": 0.0002, "epoch": 6.208442286141203, "step": 8310}, {"loss": 1.0895, "grad_norm": 1.1992450952529907, "learning_rate": 0.0002, "epoch": 6.215913335823683, "step": 8320}, {"loss": 1.1417, "grad_norm": 1.3317986726760864, "learning_rate": 0.0002, "epoch": 6.223384385506163, "step": 8330}, {"loss": 1.0958, "grad_norm": 1.0538336038589478, "learning_rate": 0.0002, "epoch": 6.230855435188644, "step": 8340}, {"loss": 1.1557, "grad_norm": 1.1767704486846924, "learning_rate": 0.0002, "epoch": 6.238326484871124, "step": 8350}, {"loss": 1.1038, "grad_norm": 1.1213016510009766, "learning_rate": 0.0002, "epoch": 6.2457975345536045, "step": 8360}, {"loss": 1.1241, "grad_norm": 1.1895716190338135, "learning_rate": 0.0002, "epoch": 6.253268584236086, "step": 8370}, {"loss": 1.1171, "grad_norm": 1.1078153848648071, "learning_rate": 0.0002, "epoch": 6.260739633918566, "step": 8380}, {"loss": 1.1124, "grad_norm": 1.1662801504135132, "learning_rate": 0.0002, "epoch": 6.268210683601046, "step": 8390}, {"loss": 1.125, "grad_norm": 1.2071197032928467, "learning_rate": 0.0002, "epoch": 6.275681733283526, "step": 8400}, {"loss": 1.0625, "grad_norm": 1.2653778791427612, "learning_rate": 0.0002, "epoch": 6.283152782966007, "step": 8410}, {"loss": 1.0565, "grad_norm": 1.6128872632980347, "learning_rate": 0.0002, "epoch": 6.290623832648487, "step": 8420}, {"loss": 1.1212, "grad_norm": 1.4993070363998413, "learning_rate": 0.0002, "epoch": 6.298094882330967, "step": 8430}, {"loss": 1.1516, "grad_norm": 1.16339910030365, "learning_rate": 0.0002, "epoch": 6.305565932013448, "step": 8440}, {"loss": 1.0662, "grad_norm": 1.256822943687439, "learning_rate": 0.0002, "epoch": 6.313036981695928, "step": 8450}, {"loss": 1.1566, "grad_norm": 1.1352964639663696, "learning_rate": 0.0002, "epoch": 6.3205080313784086, "step": 8460}, {"loss": 1.1297, "grad_norm": 1.0061070919036865, "learning_rate": 0.0002, "epoch": 6.327979081060889, "step": 8470}, {"loss": 1.0967, "grad_norm": 1.1901768445968628, "learning_rate": 0.0002, "epoch": 6.33545013074337, "step": 8480}, {"loss": 1.1463, "grad_norm": 1.2715139389038086, "learning_rate": 0.0002, "epoch": 6.34292118042585, "step": 8490}, {"loss": 1.2143, "grad_norm": 1.1583346128463745, "learning_rate": 0.0002, "epoch": 6.35039223010833, "step": 8500}, {"loss": 1.1072, "grad_norm": 1.1427477598190308, "learning_rate": 0.0002, "epoch": 6.357863279790811, "step": 8510}, {"loss": 1.1119, "grad_norm": 1.1952263116836548, "learning_rate": 0.0002, "epoch": 6.365334329473291, "step": 8520}, {"loss": 1.0797, "grad_norm": 1.0599623918533325, "learning_rate": 0.0002, "epoch": 6.372805379155771, "step": 8530}, {"loss": 1.1091, "grad_norm": 1.3511574268341064, "learning_rate": 0.0002, "epoch": 6.380276428838251, "step": 8540}, {"loss": 1.1272, "grad_norm": 1.171126127243042, "learning_rate": 0.0002, "epoch": 6.387747478520732, "step": 8550}, {"loss": 1.1615, "grad_norm": 1.285474419593811, "learning_rate": 0.0002, "epoch": 6.395218528203213, "step": 8560}, {"loss": 1.1505, "grad_norm": 0.9751279950141907, "learning_rate": 0.0002, "epoch": 6.402689577885693, "step": 8570}, {"loss": 1.1502, "grad_norm": 1.2194149494171143, "learning_rate": 0.0002, "epoch": 6.410160627568174, "step": 8580}, {"loss": 1.138, "grad_norm": 1.255888819694519, "learning_rate": 0.0002, "epoch": 6.417631677250654, "step": 8590}, {"loss": 1.1308, "grad_norm": 1.1636122465133667, "learning_rate": 0.0002, "epoch": 6.425102726933134, "step": 8600}, {"loss": 1.1398, "grad_norm": 1.0769859552383423, "learning_rate": 0.0002, "epoch": 6.432573776615614, "step": 8610}, {"loss": 1.1183, "grad_norm": 1.151778221130371, "learning_rate": 0.0002, "epoch": 6.440044826298095, "step": 8620}, {"loss": 1.0706, "grad_norm": 1.2749944925308228, "learning_rate": 0.0002, "epoch": 6.447515875980575, "step": 8630}, {"loss": 1.1011, "grad_norm": 1.1925828456878662, "learning_rate": 0.0002, "epoch": 6.454986925663055, "step": 8640}, {"loss": 1.1581, "grad_norm": 1.166107416152954, "learning_rate": 0.0002, "epoch": 6.4624579753455365, "step": 8650}, {"loss": 1.105, "grad_norm": 1.0372248888015747, "learning_rate": 0.0002, "epoch": 6.469929025028017, "step": 8660}, {"loss": 1.1546, "grad_norm": 1.26933753490448, "learning_rate": 0.0002, "epoch": 6.477400074710497, "step": 8670}, {"loss": 1.2362, "grad_norm": 1.2154223918914795, "learning_rate": 0.0002, "epoch": 6.484871124392977, "step": 8680}, {"loss": 1.1096, "grad_norm": 1.09475839138031, "learning_rate": 0.0002, "epoch": 6.492342174075458, "step": 8690}, {"loss": 1.1168, "grad_norm": 1.0763037204742432, "learning_rate": 0.0002, "epoch": 6.499813223757938, "step": 8700}, {"loss": 1.1993, "grad_norm": 1.1882896423339844, "learning_rate": 0.0002, "epoch": 6.507284273440418, "step": 8710}, {"loss": 1.1498, "grad_norm": 1.1662089824676514, "learning_rate": 0.0002, "epoch": 6.514755323122898, "step": 8720}, {"loss": 1.2008, "grad_norm": 1.3259495496749878, "learning_rate": 0.0002, "epoch": 6.522226372805379, "step": 8730}, {"loss": 1.1289, "grad_norm": 1.0858017206192017, "learning_rate": 0.0002, "epoch": 6.5296974224878594, "step": 8740}, {"loss": 1.1335, "grad_norm": 1.240337610244751, "learning_rate": 0.0002, "epoch": 6.53716847217034, "step": 8750}, {"loss": 1.1479, "grad_norm": 1.1381462812423706, "learning_rate": 0.0002, "epoch": 6.544639521852821, "step": 8760}, {"loss": 1.0991, "grad_norm": 1.2220063209533691, "learning_rate": 0.0002, "epoch": 6.552110571535301, "step": 8770}, {"loss": 1.159, "grad_norm": 1.1553083658218384, "learning_rate": 0.0002, "epoch": 6.559581621217781, "step": 8780}, {"loss": 1.0996, "grad_norm": 1.1383219957351685, "learning_rate": 0.0002, "epoch": 6.567052670900262, "step": 8790}, {"loss": 1.1355, "grad_norm": 1.0379676818847656, "learning_rate": 0.0002, "epoch": 6.574523720582742, "step": 8800}, {"loss": 1.1704, "grad_norm": 1.376488447189331, "learning_rate": 0.0002, "epoch": 6.581994770265222, "step": 8810}, {"loss": 1.1265, "grad_norm": 1.1586211919784546, "learning_rate": 0.0002, "epoch": 6.589465819947702, "step": 8820}, {"loss": 1.1904, "grad_norm": 1.28152334690094, "learning_rate": 0.0002, "epoch": 6.596936869630183, "step": 8830}, {"loss": 1.1646, "grad_norm": 1.2656810283660889, "learning_rate": 0.0002, "epoch": 6.6044079193126635, "step": 8840}, {"loss": 1.1865, "grad_norm": 1.0636502504348755, "learning_rate": 0.0002, "epoch": 6.611878968995144, "step": 8850}, {"loss": 1.125, "grad_norm": 1.273239254951477, "learning_rate": 0.0002, "epoch": 6.619350018677624, "step": 8860}, {"loss": 1.1443, "grad_norm": 1.1055482625961304, "learning_rate": 0.0002, "epoch": 6.626821068360105, "step": 8870}, {"loss": 1.0877, "grad_norm": 1.1934176683425903, "learning_rate": 0.0002, "epoch": 6.634292118042585, "step": 8880}, {"loss": 1.194, "grad_norm": 1.2248114347457886, "learning_rate": 0.0002, "epoch": 6.641763167725065, "step": 8890}, {"loss": 1.1609, "grad_norm": 1.1950982809066772, "learning_rate": 0.0002, "epoch": 6.649234217407546, "step": 8900}, {"loss": 1.169, "grad_norm": 1.0821784734725952, "learning_rate": 0.0002, "epoch": 6.656705267090026, "step": 8910}, {"loss": 1.1337, "grad_norm": 1.0062463283538818, "learning_rate": 0.0002, "epoch": 6.664176316772506, "step": 8920}, {"loss": 1.1403, "grad_norm": 1.2373089790344238, "learning_rate": 0.0002, "epoch": 6.671647366454987, "step": 8930}, {"loss": 1.2051, "grad_norm": 1.1821746826171875, "learning_rate": 0.0002, "epoch": 6.6791184161374675, "step": 8940}, {"loss": 1.1214, "grad_norm": 1.2350659370422363, "learning_rate": 0.0002, "epoch": 6.686589465819948, "step": 8950}, {"loss": 1.225, "grad_norm": 1.1012883186340332, "learning_rate": 0.0002, "epoch": 6.694060515502428, "step": 8960}, {"loss": 1.2111, "grad_norm": 1.2008943557739258, "learning_rate": 0.0002, "epoch": 6.701531565184909, "step": 8970}, {"loss": 1.1769, "grad_norm": 1.2355504035949707, "learning_rate": 0.0002, "epoch": 6.709002614867389, "step": 8980}, {"loss": 1.1323, "grad_norm": 1.2367502450942993, "learning_rate": 0.0002, "epoch": 6.716473664549869, "step": 8990}, {"loss": 1.1235, "grad_norm": 1.1075866222381592, "learning_rate": 0.0002, "epoch": 6.723944714232349, "step": 9000}, {"loss": 1.1239, "grad_norm": 1.246480941772461, "learning_rate": 0.0002, "epoch": 6.73141576391483, "step": 9010}, {"loss": 1.2154, "grad_norm": 1.1252824068069458, "learning_rate": 0.0002, "epoch": 6.73888681359731, "step": 9020}, {"loss": 1.1762, "grad_norm": 1.0706887245178223, "learning_rate": 0.0002, "epoch": 6.7463578632797905, "step": 9030}, {"loss": 1.1961, "grad_norm": 1.0874755382537842, "learning_rate": 0.0002, "epoch": 6.7538289129622715, "step": 9040}, {"loss": 1.0889, "grad_norm": 1.121434211730957, "learning_rate": 0.0002, "epoch": 6.761299962644752, "step": 9050}, {"loss": 1.2018, "grad_norm": 1.1517996788024902, "learning_rate": 0.0002, "epoch": 6.768771012327232, "step": 9060}, {"loss": 1.1593, "grad_norm": 1.2484540939331055, "learning_rate": 0.0002, "epoch": 6.776242062009713, "step": 9070}, {"loss": 1.13, "grad_norm": 1.023059368133545, "learning_rate": 0.0002, "epoch": 6.783713111692193, "step": 9080}, {"loss": 1.1929, "grad_norm": 1.1334631443023682, "learning_rate": 0.0002, "epoch": 6.791184161374673, "step": 9090}, {"loss": 1.18, "grad_norm": 1.2991816997528076, "learning_rate": 0.0002, "epoch": 6.798655211057153, "step": 9100}, {"loss": 1.2398, "grad_norm": 1.4147199392318726, "learning_rate": 0.0002, "epoch": 6.806126260739634, "step": 9110}, {"loss": 1.0958, "grad_norm": 1.1353832483291626, "learning_rate": 0.0002, "epoch": 6.813597310422114, "step": 9120}, {"loss": 1.1379, "grad_norm": 1.0332539081573486, "learning_rate": 0.0002, "epoch": 6.8210683601045945, "step": 9130}, {"loss": 1.1652, "grad_norm": 1.2208142280578613, "learning_rate": 0.0002, "epoch": 6.828539409787075, "step": 9140}, {"loss": 1.1463, "grad_norm": 1.3033398389816284, "learning_rate": 0.0002, "epoch": 6.836010459469556, "step": 9150}, {"loss": 1.1834, "grad_norm": 1.2676737308502197, "learning_rate": 0.0002, "epoch": 6.843481509152036, "step": 9160}, {"loss": 1.1786, "grad_norm": 1.1668603420257568, "learning_rate": 0.0002, "epoch": 6.850952558834516, "step": 9170}, {"loss": 1.1801, "grad_norm": 1.1994788646697998, "learning_rate": 0.0002, "epoch": 6.858423608516997, "step": 9180}, {"loss": 1.2131, "grad_norm": 1.231873869895935, "learning_rate": 0.0002, "epoch": 6.865894658199477, "step": 9190}, {"loss": 1.2109, "grad_norm": 0.9981484413146973, "learning_rate": 0.0002, "epoch": 6.873365707881957, "step": 9200}, {"loss": 1.1084, "grad_norm": 1.2799428701400757, "learning_rate": 0.0002, "epoch": 6.880836757564438, "step": 9210}, {"loss": 1.2004, "grad_norm": 1.2042057514190674, "learning_rate": 0.0002, "epoch": 6.888307807246918, "step": 9220}, {"loss": 1.1567, "grad_norm": 1.070420265197754, "learning_rate": 0.0002, "epoch": 6.8957788569293985, "step": 9230}, {"loss": 1.1353, "grad_norm": 1.327160358428955, "learning_rate": 0.0002, "epoch": 6.903249906611879, "step": 9240}, {"loss": 1.1945, "grad_norm": 1.1109007596969604, "learning_rate": 0.0002, "epoch": 6.91072095629436, "step": 9250}, {"loss": 1.1701, "grad_norm": 1.1669930219650269, "learning_rate": 0.0002, "epoch": 6.91819200597684, "step": 9260}, {"loss": 1.1854, "grad_norm": 1.034532904624939, "learning_rate": 0.0002, "epoch": 6.92566305565932, "step": 9270}, {"loss": 1.1712, "grad_norm": 1.1035540103912354, "learning_rate": 0.0002, "epoch": 6.9331341053418, "step": 9280}, {"loss": 1.1767, "grad_norm": 1.366254448890686, "learning_rate": 0.0002, "epoch": 6.940605155024281, "step": 9290}, {"loss": 1.1591, "grad_norm": 1.094214677810669, "learning_rate": 0.0002, "epoch": 6.948076204706761, "step": 9300}, {"loss": 1.18, "grad_norm": 1.131238579750061, "learning_rate": 0.0002, "epoch": 6.955547254389241, "step": 9310}, {"loss": 1.2513, "grad_norm": 1.202369213104248, "learning_rate": 0.0002, "epoch": 6.963018304071722, "step": 9320}, {"loss": 1.1922, "grad_norm": 1.1067225933074951, "learning_rate": 0.0002, "epoch": 6.9704893537542025, "step": 9330}, {"loss": 1.1965, "grad_norm": 1.0258643627166748, "learning_rate": 0.0002, "epoch": 6.977960403436683, "step": 9340}, {"loss": 1.2053, "grad_norm": 1.3311655521392822, "learning_rate": 0.0002, "epoch": 6.985431453119164, "step": 9350}, {"loss": 1.1778, "grad_norm": 1.1245559453964233, "learning_rate": 0.0002, "epoch": 6.992902502801644, "step": 9360}]}
+{"epoch": 7.997011580127008, "step": 10704, "epoch_duration": 1871.7742564678192, "total_accumulated_duration": 13167.358186483383, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7568.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-6326-sd-42/checkpoint-2677", "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": 2.6181, "grad_norm": 0.4912872612476349, "learning_rate": 0.0002, "epoch": 0.007471049682480389, "step": 10}, {"loss": 2.2606, "grad_norm": 0.4856316149234772, "learning_rate": 0.0002, "epoch": 0.014942099364960777, "step": 20}, {"loss": 2.0957, "grad_norm": 0.47683125734329224, "learning_rate": 0.0002, "epoch": 0.022413149047441166, "step": 30}, {"loss": 1.8908, "grad_norm": 0.515082597732544, "learning_rate": 0.0002, "epoch": 0.029884198729921554, "step": 40}, {"loss": 1.9704, "grad_norm": 0.5299215316772461, "learning_rate": 0.0002, "epoch": 0.03735524841240194, "step": 50}, {"loss": 1.9225, "grad_norm": 0.4951399862766266, "learning_rate": 0.0002, "epoch": 0.04482629809488233, "step": 60}, {"loss": 1.9742, "grad_norm": 0.48079821467399597, "learning_rate": 0.0002, "epoch": 0.05229734777736272, "step": 70}, {"loss": 1.9466, "grad_norm": 0.49402132630348206, "learning_rate": 0.0002, "epoch": 0.05976839745984311, "step": 80}, {"loss": 1.8691, "grad_norm": 0.4778193235397339, "learning_rate": 0.0002, "epoch": 0.0672394471423235, "step": 90}, {"loss": 1.8455, "grad_norm": 0.42472657561302185, "learning_rate": 0.0002, "epoch": 0.07471049682480388, "step": 100}, {"loss": 1.8744, "grad_norm": 0.4433092474937439, "learning_rate": 0.0002, "epoch": 0.08218154650728428, "step": 110}, {"loss": 1.865, "grad_norm": 0.4472862780094147, "learning_rate": 0.0002, "epoch": 0.08965259618976466, "step": 120}, {"loss": 1.9256, "grad_norm": 0.42596298456192017, "learning_rate": 0.0002, "epoch": 0.09712364587224505, "step": 130}, {"loss": 1.8015, "grad_norm": 0.46645811200141907, "learning_rate": 0.0002, "epoch": 0.10459469555472543, "step": 140}, {"loss": 1.8307, "grad_norm": 0.41041234135627747, "learning_rate": 0.0002, "epoch": 0.11206574523720583, "step": 150}, {"loss": 1.8276, "grad_norm": 0.5329819917678833, "learning_rate": 0.0002, "epoch": 0.11953679491968622, "step": 160}, {"loss": 1.8118, "grad_norm": 0.4065922200679779, "learning_rate": 0.0002, "epoch": 0.1270078446021666, "step": 170}, {"loss": 1.8559, "grad_norm": 0.38406994938850403, "learning_rate": 0.0002, "epoch": 0.134478894284647, "step": 180}, {"loss": 1.8647, "grad_norm": 0.4246881306171417, "learning_rate": 0.0002, "epoch": 0.14194994396712737, "step": 190}, {"loss": 1.8054, "grad_norm": 0.35136649012565613, "learning_rate": 0.0002, "epoch": 0.14942099364960776, "step": 200}, {"loss": 1.802, "grad_norm": 0.43252742290496826, "learning_rate": 0.0002, "epoch": 0.15689204333208817, "step": 210}, {"loss": 1.7823, "grad_norm": 0.39236941933631897, "learning_rate": 0.0002, "epoch": 0.16436309301456856, "step": 220}, {"loss": 1.818, "grad_norm": 0.3748249113559723, "learning_rate": 0.0002, "epoch": 0.17183414269704894, "step": 230}, {"loss": 1.866, "grad_norm": 0.6432855725288391, "learning_rate": 0.0002, "epoch": 0.17930519237952933, "step": 240}, {"loss": 1.8397, "grad_norm": 0.34874802827835083, "learning_rate": 0.0002, "epoch": 0.1867762420620097, "step": 250}, {"loss": 1.79, "grad_norm": 0.3721984326839447, "learning_rate": 0.0002, "epoch": 0.1942472917444901, "step": 260}, {"loss": 1.8464, "grad_norm": 0.4339311420917511, "learning_rate": 0.0002, "epoch": 0.20171834142697048, "step": 270}, {"loss": 1.8665, "grad_norm": 0.4018215537071228, "learning_rate": 0.0002, "epoch": 0.20918939110945087, "step": 280}, {"loss": 1.8048, "grad_norm": 0.3278839886188507, "learning_rate": 0.0002, "epoch": 0.21666044079193125, "step": 290}, {"loss": 1.7395, "grad_norm": 0.36146077513694763, "learning_rate": 0.0002, "epoch": 0.22413149047441167, "step": 300}, {"loss": 1.7916, "grad_norm": 0.38175010681152344, "learning_rate": 0.0002, "epoch": 0.23160254015689205, "step": 310}, {"loss": 1.8593, "grad_norm": 0.44776618480682373, "learning_rate": 0.0002, "epoch": 0.23907358983937244, "step": 320}, {"loss": 1.7824, "grad_norm": 0.3933652937412262, "learning_rate": 0.0002, "epoch": 0.24654463952185282, "step": 330}, {"loss": 1.8393, "grad_norm": 0.3515005111694336, "learning_rate": 0.0002, "epoch": 0.2540156892043332, "step": 340}, {"loss": 1.8653, "grad_norm": 0.6683304309844971, "learning_rate": 0.0002, "epoch": 0.2614867388868136, "step": 350}, {"loss": 1.8797, "grad_norm": 0.37093454599380493, "learning_rate": 0.0002, "epoch": 0.268957788569294, "step": 360}, {"loss": 1.8251, "grad_norm": 0.3450651168823242, "learning_rate": 0.0002, "epoch": 0.2764288382517744, "step": 370}, {"loss": 1.7435, "grad_norm": 0.5140917301177979, "learning_rate": 0.0002, "epoch": 0.28389988793425475, "step": 380}, {"loss": 1.8026, "grad_norm": 0.32885563373565674, "learning_rate": 0.0002, "epoch": 0.29137093761673516, "step": 390}, {"loss": 1.8174, "grad_norm": 0.33962297439575195, "learning_rate": 0.0002, "epoch": 0.2988419872992155, "step": 400}, {"loss": 1.7467, "grad_norm": 0.3723141849040985, "learning_rate": 0.0002, "epoch": 0.30631303698169593, "step": 410}, {"loss": 1.8459, "grad_norm": 0.37173134088516235, "learning_rate": 0.0002, "epoch": 0.31378408666417634, "step": 420}, {"loss": 1.8876, "grad_norm": 0.33736956119537354, "learning_rate": 0.0002, "epoch": 0.3212551363466567, "step": 430}, {"loss": 1.8367, "grad_norm": 0.3602448105812073, "learning_rate": 0.0002, "epoch": 0.3287261860291371, "step": 440}, {"loss": 1.8058, "grad_norm": 0.3569699227809906, "learning_rate": 0.0002, "epoch": 0.33619723571161747, "step": 450}, {"loss": 1.8086, "grad_norm": 0.31009167432785034, "learning_rate": 0.0002, "epoch": 0.3436682853940979, "step": 460}, {"loss": 1.8876, "grad_norm": 0.5278693437576294, "learning_rate": 0.0002, "epoch": 0.35113933507657824, "step": 470}, {"loss": 1.8534, "grad_norm": 0.3587537109851837, "learning_rate": 0.0002, "epoch": 0.35861038475905865, "step": 480}, {"loss": 1.8046, "grad_norm": 0.3859670162200928, "learning_rate": 0.0002, "epoch": 0.366081434441539, "step": 490}, {"loss": 1.8287, "grad_norm": 0.395913690328598, "learning_rate": 0.0002, "epoch": 0.3735524841240194, "step": 500}, {"loss": 1.7619, "grad_norm": 0.35052940249443054, "learning_rate": 0.0002, "epoch": 0.38102353380649984, "step": 510}, {"loss": 1.7824, "grad_norm": 0.2979494333267212, "learning_rate": 0.0002, "epoch": 0.3884945834889802, "step": 520}, {"loss": 1.8641, "grad_norm": 0.3062683343887329, "learning_rate": 0.0002, "epoch": 0.3959656331714606, "step": 530}, {"loss": 1.7651, "grad_norm": 0.3172847330570221, "learning_rate": 0.0002, "epoch": 0.40343668285394096, "step": 540}, {"loss": 1.806, "grad_norm": 0.360435426235199, "learning_rate": 0.0002, "epoch": 0.4109077325364214, "step": 550}, {"loss": 1.9054, "grad_norm": 0.3427872359752655, "learning_rate": 0.0002, "epoch": 0.41837878221890173, "step": 560}, {"loss": 1.7562, "grad_norm": 0.34036558866500854, "learning_rate": 0.0002, "epoch": 0.42584983190138215, "step": 570}, {"loss": 1.7254, "grad_norm": 0.3365345299243927, "learning_rate": 0.0002, "epoch": 0.4333208815838625, "step": 580}, {"loss": 1.8328, "grad_norm": 0.35619041323661804, "learning_rate": 0.0002, "epoch": 0.4407919312663429, "step": 590}, {"loss": 1.8114, "grad_norm": 0.3569088280200958, "learning_rate": 0.0002, "epoch": 0.44826298094882333, "step": 600}, {"loss": 1.8599, "grad_norm": 0.3581278622150421, "learning_rate": 0.0002, "epoch": 0.4557340306313037, "step": 610}, {"loss": 1.7078, "grad_norm": 0.43197110295295715, "learning_rate": 0.0002, "epoch": 0.4632050803137841, "step": 620}, {"loss": 1.8257, "grad_norm": 0.33966198563575745, "learning_rate": 0.0002, "epoch": 0.47067612999626446, "step": 630}, {"loss": 1.7528, "grad_norm": 0.3343866467475891, "learning_rate": 0.0002, "epoch": 0.47814717967874487, "step": 640}, {"loss": 1.8191, "grad_norm": 0.33878564834594727, "learning_rate": 0.0002, "epoch": 0.48561822936122523, "step": 650}, {"loss": 1.8801, "grad_norm": 0.387195885181427, "learning_rate": 0.0002, "epoch": 0.49308927904370564, "step": 660}, {"loss": 1.7559, "grad_norm": 0.3755440413951874, "learning_rate": 0.0002, "epoch": 0.500560328726186, "step": 670}, {"loss": 1.8057, "grad_norm": 0.3272816836833954, "learning_rate": 0.0002, "epoch": 0.5080313784086664, "step": 680}, {"loss": 1.8156, "grad_norm": 0.36063864827156067, "learning_rate": 0.0002, "epoch": 0.5155024280911468, "step": 690}, {"loss": 1.8397, "grad_norm": 0.35317373275756836, "learning_rate": 0.0002, "epoch": 0.5229734777736272, "step": 700}, {"loss": 1.7603, "grad_norm": 0.3561195433139801, "learning_rate": 0.0002, "epoch": 0.5304445274561076, "step": 710}, {"loss": 1.8149, "grad_norm": 0.31124624609947205, "learning_rate": 0.0002, "epoch": 0.537915577138588, "step": 720}, {"loss": 1.7434, "grad_norm": 0.3294544517993927, "learning_rate": 0.0002, "epoch": 0.5453866268210683, "step": 730}, {"loss": 1.8027, "grad_norm": 0.31933900713920593, "learning_rate": 0.0002, "epoch": 0.5528576765035488, "step": 740}, {"loss": 1.7601, "grad_norm": 0.3226020634174347, "learning_rate": 0.0002, "epoch": 0.5603287261860291, "step": 750}, {"loss": 1.7862, "grad_norm": 0.3147525489330292, "learning_rate": 0.0002, "epoch": 0.5677997758685095, "step": 760}, {"loss": 1.9028, "grad_norm": 0.32234328985214233, "learning_rate": 0.0002, "epoch": 0.57527082555099, "step": 770}, {"loss": 1.7623, "grad_norm": 0.3258664309978485, "learning_rate": 0.0002, "epoch": 0.5827418752334703, "step": 780}, {"loss": 1.7384, "grad_norm": 0.3166961967945099, "learning_rate": 0.0002, "epoch": 0.5902129249159507, "step": 790}, {"loss": 1.8799, "grad_norm": 0.35621458292007446, "learning_rate": 0.0002, "epoch": 0.597683974598431, "step": 800}, {"loss": 1.8313, "grad_norm": 0.3236999213695526, "learning_rate": 0.0002, "epoch": 0.6051550242809115, "step": 810}, {"loss": 1.7132, "grad_norm": 0.2892923653125763, "learning_rate": 0.0002, "epoch": 0.6126260739633919, "step": 820}, {"loss": 1.8709, "grad_norm": 0.4098321497440338, "learning_rate": 0.0002, "epoch": 0.6200971236458722, "step": 830}, {"loss": 1.7637, "grad_norm": 0.3337118923664093, "learning_rate": 0.0002, "epoch": 0.6275681733283527, "step": 840}, {"loss": 1.7375, "grad_norm": 0.30416029691696167, "learning_rate": 0.0002, "epoch": 0.635039223010833, "step": 850}, {"loss": 1.7419, "grad_norm": 0.3361026346683502, "learning_rate": 0.0002, "epoch": 0.6425102726933134, "step": 860}, {"loss": 1.732, "grad_norm": 0.3537365198135376, "learning_rate": 0.0002, "epoch": 0.6499813223757938, "step": 870}, {"loss": 1.7825, "grad_norm": 0.33854469656944275, "learning_rate": 0.0002, "epoch": 0.6574523720582742, "step": 880}, {"loss": 1.7561, "grad_norm": 0.3332272469997406, "learning_rate": 0.0002, "epoch": 0.6649234217407546, "step": 890}, {"loss": 1.7247, "grad_norm": 0.34954726696014404, "learning_rate": 0.0002, "epoch": 0.6723944714232349, "step": 900}, {"loss": 1.7917, "grad_norm": 0.2921750247478485, "learning_rate": 0.0002, "epoch": 0.6798655211057153, "step": 910}, {"loss": 1.7807, "grad_norm": 0.30508682131767273, "learning_rate": 0.0002, "epoch": 0.6873365707881958, "step": 920}, {"loss": 1.8082, "grad_norm": 0.32268425822257996, "learning_rate": 0.0002, "epoch": 0.6948076204706761, "step": 930}, {"loss": 1.8283, "grad_norm": 0.2844390869140625, "learning_rate": 0.0002, "epoch": 0.7022786701531565, "step": 940}, {"loss": 1.7363, "grad_norm": 0.31263890862464905, "learning_rate": 0.0002, "epoch": 0.709749719835637, "step": 950}, {"loss": 1.8081, "grad_norm": 0.3626808822154999, "learning_rate": 0.0002, "epoch": 0.7172207695181173, "step": 960}, {"loss": 1.853, "grad_norm": 0.3322749733924866, "learning_rate": 0.0002, "epoch": 0.7246918192005977, "step": 970}, {"loss": 1.7912, "grad_norm": 0.29177871346473694, "learning_rate": 0.0002, "epoch": 0.732162868883078, "step": 980}, {"loss": 1.8447, "grad_norm": 0.35405513644218445, "learning_rate": 0.0002, "epoch": 0.7396339185655585, "step": 990}, {"loss": 1.7008, "grad_norm": 0.39318400621414185, "learning_rate": 0.0002, "epoch": 0.7471049682480388, "step": 1000}, {"loss": 1.7803, "grad_norm": 0.29401418566703796, "learning_rate": 0.0002, "epoch": 0.7545760179305192, "step": 1010}, {"loss": 1.7649, "grad_norm": 0.3271748721599579, "learning_rate": 0.0002, "epoch": 0.7620470676129997, "step": 1020}, {"loss": 1.7266, "grad_norm": 0.30883970856666565, "learning_rate": 0.0002, "epoch": 0.76951811729548, "step": 1030}, {"loss": 1.7722, "grad_norm": 0.3411838412284851, "learning_rate": 0.0002, "epoch": 0.7769891669779604, "step": 1040}, {"loss": 1.829, "grad_norm": 0.30608129501342773, "learning_rate": 0.0002, "epoch": 0.7844602166604407, "step": 1050}, {"loss": 1.7815, "grad_norm": 0.30899080634117126, "learning_rate": 0.0002, "epoch": 0.7919312663429212, "step": 1060}, {"loss": 1.7625, "grad_norm": 0.3160453140735626, "learning_rate": 0.0002, "epoch": 0.7994023160254016, "step": 1070}, {"loss": 1.8452, "grad_norm": 0.30947187542915344, "learning_rate": 0.0002, "epoch": 0.8068733657078819, "step": 1080}, {"loss": 1.7418, "grad_norm": 0.3103134036064148, "learning_rate": 0.0002, "epoch": 0.8143444153903624, "step": 1090}, {"loss": 1.842, "grad_norm": 0.31771138310432434, "learning_rate": 0.0002, "epoch": 0.8218154650728428, "step": 1100}, {"loss": 1.7918, "grad_norm": 0.5860997438430786, "learning_rate": 0.0002, "epoch": 0.8292865147553231, "step": 1110}, {"loss": 1.8443, "grad_norm": 0.3230148255825043, "learning_rate": 0.0002, "epoch": 0.8367575644378035, "step": 1120}, {"loss": 1.8478, "grad_norm": 0.29611510038375854, "learning_rate": 0.0002, "epoch": 0.8442286141202839, "step": 1130}, {"loss": 1.7673, "grad_norm": 0.3373654782772064, "learning_rate": 0.0002, "epoch": 0.8516996638027643, "step": 1140}, {"loss": 1.7997, "grad_norm": 0.3474279046058655, "learning_rate": 0.0002, "epoch": 0.8591707134852447, "step": 1150}, {"loss": 1.75, "grad_norm": 0.35057875514030457, "learning_rate": 0.0002, "epoch": 0.866641763167725, "step": 1160}, {"loss": 1.8273, "grad_norm": 0.39537495374679565, "learning_rate": 0.0002, "epoch": 0.8741128128502055, "step": 1170}, {"loss": 1.7682, "grad_norm": 0.3714233636856079, "learning_rate": 0.0002, "epoch": 0.8815838625326858, "step": 1180}, {"loss": 1.7549, "grad_norm": 0.2950296998023987, "learning_rate": 0.0002, "epoch": 0.8890549122151662, "step": 1190}, {"loss": 1.7612, "grad_norm": 0.38182979822158813, "learning_rate": 0.0002, "epoch": 0.8965259618976467, "step": 1200}, {"loss": 1.827, "grad_norm": 0.27883678674697876, "learning_rate": 0.0002, "epoch": 0.903997011580127, "step": 1210}, {"loss": 1.7623, "grad_norm": 0.33874374628067017, "learning_rate": 0.0002, "epoch": 0.9114680612626074, "step": 1220}, {"loss": 1.7334, "grad_norm": 0.3014272153377533, "learning_rate": 0.0002, "epoch": 0.9189391109450877, "step": 1230}, {"loss": 1.8235, "grad_norm": 0.3194271922111511, "learning_rate": 0.0002, "epoch": 0.9264101606275682, "step": 1240}, {"loss": 1.7924, "grad_norm": 0.3049403429031372, "learning_rate": 0.0002, "epoch": 0.9338812103100486, "step": 1250}, {"loss": 1.7535, "grad_norm": 0.30621254444122314, "learning_rate": 0.0002, "epoch": 0.9413522599925289, "step": 1260}, {"loss": 1.8287, "grad_norm": 0.28675132989883423, "learning_rate": 0.0002, "epoch": 0.9488233096750094, "step": 1270}, {"loss": 1.7586, "grad_norm": 0.3322032690048218, "learning_rate": 0.0002, "epoch": 0.9562943593574897, "step": 1280}, {"loss": 1.8054, "grad_norm": 0.35408294200897217, "learning_rate": 0.0002, "epoch": 0.9637654090399701, "step": 1290}, {"loss": 1.7343, "grad_norm": 0.36386919021606445, "learning_rate": 0.0002, "epoch": 0.9712364587224505, "step": 1300}, {"loss": 1.8633, "grad_norm": 0.32338324189186096, "learning_rate": 0.0002, "epoch": 0.9787075084049309, "step": 1310}, {"loss": 1.7724, "grad_norm": 0.3714013993740082, "learning_rate": 0.0002, "epoch": 0.9861785580874113, "step": 1320}, {"loss": 1.7766, "grad_norm": 0.3133082389831543, "learning_rate": 0.0002, "epoch": 0.9936496077698916, "step": 1330}, {"eval_loss": 1.8051470518112183, "eval_runtime": 38.6332, "eval_samples_per_second": 13.331, "eval_steps_per_second": 1.682, "epoch": 0.9996264475158759, "step": 1338}, {"loss": 1.8035, "grad_norm": 0.31595754623413086, "learning_rate": 0.0002, "epoch": 1.001120657452372, "step": 1340}, {"loss": 1.7486, "grad_norm": 0.3095700144767761, "learning_rate": 0.0002, "epoch": 1.0085917071348525, "step": 1350}, {"loss": 1.6981, "grad_norm": 0.34677496552467346, "learning_rate": 0.0002, "epoch": 1.0160627568173328, "step": 1360}, {"loss": 1.7377, "grad_norm": 0.29108840227127075, "learning_rate": 0.0002, "epoch": 1.0235338064998132, "step": 1370}, {"loss": 1.7194, "grad_norm": 0.32356950640678406, "learning_rate": 0.0002, "epoch": 1.0310048561822935, "step": 1380}, {"loss": 1.7593, "grad_norm": 0.4200669229030609, "learning_rate": 0.0002, "epoch": 1.038475905864774, "step": 1390}, {"loss": 1.797, "grad_norm": 0.3283711373806, "learning_rate": 0.0002, "epoch": 1.0459469555472545, "step": 1400}, {"loss": 1.7163, "grad_norm": 0.32898256182670593, "learning_rate": 0.0002, "epoch": 1.0534180052297348, "step": 1410}, {"loss": 1.7559, "grad_norm": 0.38790300488471985, "learning_rate": 0.0002, "epoch": 1.0608890549122152, "step": 1420}, {"loss": 1.6922, "grad_norm": 0.339800089597702, "learning_rate": 0.0002, "epoch": 1.0683601045946955, "step": 1430}, {"loss": 1.7076, "grad_norm": 0.3548751175403595, "learning_rate": 0.0002, "epoch": 1.075831154277176, "step": 1440}, {"loss": 1.6985, "grad_norm": 0.35114359855651855, "learning_rate": 0.0002, "epoch": 1.0833022039596563, "step": 1450}, {"loss": 1.7217, "grad_norm": 0.35226720571517944, "learning_rate": 0.0002, "epoch": 1.0907732536421366, "step": 1460}, {"loss": 1.6822, "grad_norm": 0.33665576577186584, "learning_rate": 0.0002, "epoch": 1.0982443033246172, "step": 1470}, {"loss": 1.6699, "grad_norm": 0.363889217376709, "learning_rate": 0.0002, "epoch": 1.1057153530070976, "step": 1480}, {"loss": 1.7933, "grad_norm": 0.3826201856136322, "learning_rate": 0.0002, "epoch": 1.113186402689578, "step": 1490}, {"loss": 1.7022, "grad_norm": 0.34058740735054016, "learning_rate": 0.0002, "epoch": 1.1206574523720583, "step": 1500}, {"loss": 1.6375, "grad_norm": 0.3462134301662445, "learning_rate": 0.0002, "epoch": 1.1281285020545386, "step": 1510}, {"loss": 1.7147, "grad_norm": 0.3396756052970886, "learning_rate": 0.0002, "epoch": 1.135599551737019, "step": 1520}, {"loss": 1.7219, "grad_norm": 0.32004743814468384, "learning_rate": 0.0002, "epoch": 1.1430706014194993, "step": 1530}, {"loss": 1.743, "grad_norm": 0.3397733271121979, "learning_rate": 0.0002, "epoch": 1.15054165110198, "step": 1540}, {"loss": 1.7333, "grad_norm": 0.3783262073993683, "learning_rate": 0.0002, "epoch": 1.1580127007844603, "step": 1550}, {"loss": 1.6075, "grad_norm": 0.35121291875839233, "learning_rate": 0.0002, "epoch": 1.1654837504669406, "step": 1560}, {"loss": 1.678, "grad_norm": 0.35816895961761475, "learning_rate": 0.0002, "epoch": 1.172954800149421, "step": 1570}, {"loss": 1.7143, "grad_norm": 0.33843839168548584, "learning_rate": 0.0002, "epoch": 1.1804258498319014, "step": 1580}, {"loss": 1.7434, "grad_norm": 0.3371972143650055, "learning_rate": 0.0002, "epoch": 1.1878968995143817, "step": 1590}, {"loss": 1.7671, "grad_norm": 0.36016878485679626, "learning_rate": 0.0002, "epoch": 1.195367949196862, "step": 1600}, {"loss": 1.6914, "grad_norm": 0.40879473090171814, "learning_rate": 0.0002, "epoch": 1.2028389988793426, "step": 1610}, {"loss": 1.6955, "grad_norm": 0.3216715455055237, "learning_rate": 0.0002, "epoch": 1.210310048561823, "step": 1620}, {"loss": 1.632, "grad_norm": 0.4482610821723938, "learning_rate": 0.0002, "epoch": 1.2177810982443034, "step": 1630}, {"loss": 1.6999, "grad_norm": 0.3257700502872467, "learning_rate": 0.0002, "epoch": 1.2252521479267837, "step": 1640}, {"loss": 1.7177, "grad_norm": 0.38646459579467773, "learning_rate": 0.0002, "epoch": 1.232723197609264, "step": 1650}, {"loss": 1.7081, "grad_norm": 0.4081360697746277, "learning_rate": 0.0002, "epoch": 1.2401942472917444, "step": 1660}, {"loss": 1.7519, "grad_norm": 0.4326848089694977, "learning_rate": 0.0002, "epoch": 1.2476652969742248, "step": 1670}, {"loss": 1.6752, "grad_norm": 0.346401572227478, "learning_rate": 0.0002, "epoch": 1.2551363466567054, "step": 1680}, {"loss": 1.7425, "grad_norm": 0.34536251425743103, "learning_rate": 0.0002, "epoch": 1.2626073963391857, "step": 1690}, {"loss": 1.7061, "grad_norm": 0.41359591484069824, "learning_rate": 0.0002, "epoch": 1.270078446021666, "step": 1700}, {"loss": 1.7906, "grad_norm": 0.3530874252319336, "learning_rate": 0.0002, "epoch": 1.2775494957041464, "step": 1710}, {"loss": 1.7357, "grad_norm": 0.3702719211578369, "learning_rate": 0.0002, "epoch": 1.2850205453866268, "step": 1720}, {"loss": 1.766, "grad_norm": 0.3703329563140869, "learning_rate": 0.0002, "epoch": 1.2924915950691072, "step": 1730}, {"loss": 1.7221, "grad_norm": 0.37919729948043823, "learning_rate": 0.0002, "epoch": 1.2999626447515875, "step": 1740}, {"loss": 1.7859, "grad_norm": 0.32526856660842896, "learning_rate": 0.0002, "epoch": 1.307433694434068, "step": 1750}, {"loss": 1.7117, "grad_norm": 0.36752620339393616, "learning_rate": 0.0002, "epoch": 1.3149047441165485, "step": 1760}, {"loss": 1.7335, "grad_norm": 0.3398192524909973, "learning_rate": 0.0002, "epoch": 1.3223757937990288, "step": 1770}, {"loss": 1.7492, "grad_norm": 0.37435585260391235, "learning_rate": 0.0002, "epoch": 1.3298468434815092, "step": 1780}, {"loss": 1.7393, "grad_norm": 0.35793280601501465, "learning_rate": 0.0002, "epoch": 1.3373178931639895, "step": 1790}, {"loss": 1.7266, "grad_norm": 0.35481882095336914, "learning_rate": 0.0002, "epoch": 1.3447889428464699, "step": 1800}, {"loss": 1.7456, "grad_norm": 0.3786393105983734, "learning_rate": 0.0002, "epoch": 1.3522599925289502, "step": 1810}, {"loss": 1.7169, "grad_norm": 0.33245593309402466, "learning_rate": 0.0002, "epoch": 1.3597310422114308, "step": 1820}, {"loss": 1.7577, "grad_norm": 0.35388344526290894, "learning_rate": 0.0002, "epoch": 1.3672020918939112, "step": 1830}, {"loss": 1.6968, "grad_norm": 0.3695325553417206, "learning_rate": 0.0002, "epoch": 1.3746731415763915, "step": 1840}, {"loss": 1.7086, "grad_norm": 0.3683604598045349, "learning_rate": 0.0002, "epoch": 1.382144191258872, "step": 1850}, {"loss": 1.7878, "grad_norm": 0.3753012418746948, "learning_rate": 0.0002, "epoch": 1.3896152409413522, "step": 1860}, {"loss": 1.6969, "grad_norm": 0.3331069350242615, "learning_rate": 0.0002, "epoch": 1.3970862906238326, "step": 1870}, {"loss": 1.6644, "grad_norm": 0.3877500295639038, "learning_rate": 0.0002, "epoch": 1.404557340306313, "step": 1880}, {"loss": 1.7586, "grad_norm": 0.33525151014328003, "learning_rate": 0.0002, "epoch": 1.4120283899887935, "step": 1890}, {"loss": 1.7031, "grad_norm": 0.3697299659252167, "learning_rate": 0.0002, "epoch": 1.4194994396712737, "step": 1900}, {"loss": 1.6956, "grad_norm": 0.4029286205768585, "learning_rate": 0.0002, "epoch": 1.4269704893537543, "step": 1910}, {"loss": 1.6897, "grad_norm": 0.3596203029155731, "learning_rate": 0.0002, "epoch": 1.4344415390362346, "step": 1920}, {"loss": 1.7139, "grad_norm": 0.450783908367157, "learning_rate": 0.0002, "epoch": 1.441912588718715, "step": 1930}, {"loss": 1.7243, "grad_norm": 0.3651481866836548, "learning_rate": 0.0002, "epoch": 1.4493836384011953, "step": 1940}, {"loss": 1.6637, "grad_norm": 0.3608424663543701, "learning_rate": 0.0002, "epoch": 1.4568546880836757, "step": 1950}, {"loss": 1.8285, "grad_norm": 0.39684420824050903, "learning_rate": 0.0002, "epoch": 1.4643257377661563, "step": 1960}, {"loss": 1.7514, "grad_norm": 0.34618663787841797, "learning_rate": 0.0002, "epoch": 1.4717967874486364, "step": 1970}, {"loss": 1.6655, "grad_norm": 0.4150386452674866, "learning_rate": 0.0002, "epoch": 1.479267837131117, "step": 1980}, {"loss": 1.7021, "grad_norm": 0.35500776767730713, "learning_rate": 0.0002, "epoch": 1.4867388868135973, "step": 1990}, {"loss": 1.7322, "grad_norm": 0.344144344329834, "learning_rate": 0.0002, "epoch": 1.4942099364960777, "step": 2000}, {"loss": 1.6998, "grad_norm": 0.3340149223804474, "learning_rate": 0.0002, "epoch": 1.501680986178558, "step": 2010}, {"loss": 1.7508, "grad_norm": 0.37685006856918335, "learning_rate": 0.0002, "epoch": 1.5091520358610384, "step": 2020}, {"loss": 1.8299, "grad_norm": 0.3699876368045807, "learning_rate": 0.0002, "epoch": 1.516623085543519, "step": 2030}, {"loss": 1.7357, "grad_norm": 0.3370307385921478, "learning_rate": 0.0002, "epoch": 1.5240941352259991, "step": 2040}, {"loss": 1.8044, "grad_norm": 0.37780630588531494, "learning_rate": 0.0002, "epoch": 1.5315651849084797, "step": 2050}, {"loss": 1.7408, "grad_norm": 0.370259165763855, "learning_rate": 0.0002, "epoch": 1.53903623459096, "step": 2060}, {"loss": 1.7398, "grad_norm": 0.3440011441707611, "learning_rate": 0.0002, "epoch": 1.5465072842734404, "step": 2070}, {"loss": 1.7105, "grad_norm": 0.40382063388824463, "learning_rate": 0.0002, "epoch": 1.5539783339559208, "step": 2080}, {"loss": 1.7071, "grad_norm": 0.38002029061317444, "learning_rate": 0.0002, "epoch": 1.5614493836384011, "step": 2090}, {"loss": 1.6815, "grad_norm": 0.3658451437950134, "learning_rate": 0.0002, "epoch": 1.5689204333208817, "step": 2100}, {"loss": 1.7598, "grad_norm": 0.354842871427536, "learning_rate": 0.0002, "epoch": 1.5763914830033618, "step": 2110}, {"loss": 1.6898, "grad_norm": 0.34735530614852905, "learning_rate": 0.0002, "epoch": 1.5838625326858424, "step": 2120}, {"loss": 1.7363, "grad_norm": 0.377581924200058, "learning_rate": 0.0002, "epoch": 1.5913335823683228, "step": 2130}, {"loss": 1.7789, "grad_norm": 0.41254034638404846, "learning_rate": 0.0002, "epoch": 1.5988046320508031, "step": 2140}, {"loss": 1.6782, "grad_norm": 0.3630715310573578, "learning_rate": 0.0002, "epoch": 1.6062756817332835, "step": 2150}, {"loss": 1.7531, "grad_norm": 0.36980143189430237, "learning_rate": 0.0002, "epoch": 1.6137467314157639, "step": 2160}, {"loss": 1.6847, "grad_norm": 0.3634769320487976, "learning_rate": 0.0002, "epoch": 1.6212177810982444, "step": 2170}, {"loss": 1.6367, "grad_norm": 0.3794139623641968, "learning_rate": 0.0002, "epoch": 1.6286888307807246, "step": 2180}, {"loss": 1.7064, "grad_norm": 0.359742134809494, "learning_rate": 0.0002, "epoch": 1.6361598804632052, "step": 2190}, {"loss": 1.7027, "grad_norm": 0.3770543932914734, "learning_rate": 0.0002, "epoch": 1.6436309301456855, "step": 2200}, {"loss": 1.784, "grad_norm": 0.3797036409378052, "learning_rate": 0.0002, "epoch": 1.6511019798281659, "step": 2210}, {"loss": 1.7875, "grad_norm": 0.35622093081474304, "learning_rate": 0.0002, "epoch": 1.6585730295106462, "step": 2220}, {"loss": 1.6615, "grad_norm": 0.34552520513534546, "learning_rate": 0.0002, "epoch": 1.6660440791931266, "step": 2230}, {"loss": 1.7522, "grad_norm": 0.379926860332489, "learning_rate": 0.0002, "epoch": 1.6735151288756072, "step": 2240}, {"loss": 1.7953, "grad_norm": 0.37083810567855835, "learning_rate": 0.0002, "epoch": 1.6809861785580873, "step": 2250}, {"loss": 1.7485, "grad_norm": 0.42746543884277344, "learning_rate": 0.0002, "epoch": 1.6884572282405679, "step": 2260}, {"loss": 1.776, "grad_norm": 0.3372884690761566, "learning_rate": 0.0002, "epoch": 1.6959282779230482, "step": 2270}, {"loss": 1.7604, "grad_norm": 0.35220256447792053, "learning_rate": 0.0002, "epoch": 1.7033993276055286, "step": 2280}, {"loss": 1.7154, "grad_norm": 0.3659130930900574, "learning_rate": 0.0002, "epoch": 1.710870377288009, "step": 2290}, {"loss": 1.6953, "grad_norm": 0.37629297375679016, "learning_rate": 0.0002, "epoch": 1.7183414269704893, "step": 2300}, {"loss": 1.7212, "grad_norm": 0.36312398314476013, "learning_rate": 0.0002, "epoch": 1.7258124766529699, "step": 2310}, {"loss": 1.7903, "grad_norm": 0.467709481716156, "learning_rate": 0.0002, "epoch": 1.73328352633545, "step": 2320}, {"loss": 1.696, "grad_norm": 0.38685527443885803, "learning_rate": 0.0002, "epoch": 1.7407545760179306, "step": 2330}, {"loss": 1.7041, "grad_norm": 0.3578338325023651, "learning_rate": 0.0002, "epoch": 1.748225625700411, "step": 2340}, {"loss": 1.6456, "grad_norm": 0.36057502031326294, "learning_rate": 0.0002, "epoch": 1.7556966753828913, "step": 2350}, {"loss": 1.6853, "grad_norm": 0.3615196645259857, "learning_rate": 0.0002, "epoch": 1.7631677250653717, "step": 2360}, {"loss": 1.7612, "grad_norm": 0.4118947684764862, "learning_rate": 0.0002, "epoch": 1.770638774747852, "step": 2370}, {"loss": 1.6946, "grad_norm": 0.4067276120185852, "learning_rate": 0.0002, "epoch": 1.7781098244303326, "step": 2380}, {"loss": 1.712, "grad_norm": 0.3979823887348175, "learning_rate": 0.0002, "epoch": 1.7855808741128127, "step": 2390}, {"loss": 1.7644, "grad_norm": 0.44045883417129517, "learning_rate": 0.0002, "epoch": 1.7930519237952933, "step": 2400}, {"loss": 1.7251, "grad_norm": 0.3998069167137146, "learning_rate": 0.0002, "epoch": 1.8005229734777737, "step": 2410}, {"loss": 1.7354, "grad_norm": 0.3450094759464264, "learning_rate": 0.0002, "epoch": 1.807994023160254, "step": 2420}, {"loss": 1.6998, "grad_norm": 0.3759009838104248, "learning_rate": 0.0002, "epoch": 1.8154650728427344, "step": 2430}, {"loss": 1.7706, "grad_norm": 0.34347015619277954, "learning_rate": 0.0002, "epoch": 1.8229361225252148, "step": 2440}, {"loss": 1.7345, "grad_norm": 0.3511228859424591, "learning_rate": 0.0002, "epoch": 1.8304071722076953, "step": 2450}, {"loss": 1.6909, "grad_norm": 0.36853715777397156, "learning_rate": 0.0002, "epoch": 1.8378782218901755, "step": 2460}, {"loss": 1.6931, "grad_norm": 0.40659376978874207, "learning_rate": 0.0002, "epoch": 1.845349271572656, "step": 2470}, {"loss": 1.7626, "grad_norm": 0.39621320366859436, "learning_rate": 0.0002, "epoch": 1.8528203212551362, "step": 2480}, {"loss": 1.7427, "grad_norm": 0.3753979504108429, "learning_rate": 0.0002, "epoch": 1.8602913709376168, "step": 2490}, {"loss": 1.6622, "grad_norm": 0.3811938464641571, "learning_rate": 0.0002, "epoch": 1.8677624206200971, "step": 2500}, {"loss": 1.7718, "grad_norm": 0.3432596027851105, "learning_rate": 0.0002, "epoch": 1.8752334703025775, "step": 2510}, {"loss": 1.7488, "grad_norm": 0.3670712113380432, "learning_rate": 0.0002, "epoch": 1.882704519985058, "step": 2520}, {"loss": 1.705, "grad_norm": 0.40907177329063416, "learning_rate": 0.0002, "epoch": 1.8901755696675382, "step": 2530}, {"loss": 1.7148, "grad_norm": 0.3821999728679657, "learning_rate": 0.0002, "epoch": 1.8976466193500188, "step": 2540}, {"loss": 1.7934, "grad_norm": 0.36173978447914124, "learning_rate": 0.0002, "epoch": 1.905117669032499, "step": 2550}, {"loss": 1.6939, "grad_norm": 0.38990336656570435, "learning_rate": 0.0002, "epoch": 1.9125887187149795, "step": 2560}, {"loss": 1.6893, "grad_norm": 0.35242322087287903, "learning_rate": 0.0002, "epoch": 1.9200597683974598, "step": 2570}, {"loss": 1.7268, "grad_norm": 0.3506428003311157, "learning_rate": 0.0002, "epoch": 1.9275308180799402, "step": 2580}, {"loss": 1.6953, "grad_norm": 0.39540135860443115, "learning_rate": 0.0002, "epoch": 1.9350018677624208, "step": 2590}, {"loss": 1.6511, "grad_norm": 0.3444725573062897, "learning_rate": 0.0002, "epoch": 1.942472917444901, "step": 2600}, {"loss": 1.7259, "grad_norm": 0.3963521718978882, "learning_rate": 0.0002, "epoch": 1.9499439671273815, "step": 2610}, {"loss": 1.6946, "grad_norm": 0.3689815402030945, "learning_rate": 0.0002, "epoch": 1.9574150168098616, "step": 2620}, {"loss": 1.7384, "grad_norm": 0.3482626676559448, "learning_rate": 0.0002, "epoch": 1.9648860664923422, "step": 2630}, {"loss": 1.7048, "grad_norm": 0.35832616686820984, "learning_rate": 0.0002, "epoch": 1.9723571161748226, "step": 2640}, {"loss": 1.6681, "grad_norm": 0.4776208996772766, "learning_rate": 0.0002, "epoch": 1.979828165857303, "step": 2650}, {"loss": 1.6696, "grad_norm": 0.32570165395736694, "learning_rate": 0.0002, "epoch": 1.9872992155397835, "step": 2660}, {"loss": 1.7232, "grad_norm": 0.3380725085735321, "learning_rate": 0.0002, "epoch": 1.9947702652222636, "step": 2670}, {"eval_loss": 1.8046749830245972, "eval_runtime": 38.5096, "eval_samples_per_second": 13.373, "eval_steps_per_second": 1.688, "epoch": 2.0, "step": 2677}, {"loss": 1.7265, "grad_norm": 0.36817631125450134, "learning_rate": 0.0002, "epoch": 2.002241314904744, "step": 2680}, {"loss": 1.548, "grad_norm": 0.4056456685066223, "learning_rate": 0.0002, "epoch": 2.0097123645872244, "step": 2690}, {"loss": 1.5515, "grad_norm": 0.37416863441467285, "learning_rate": 0.0002, "epoch": 2.017183414269705, "step": 2700}, {"loss": 1.5895, "grad_norm": 0.4273638427257538, "learning_rate": 0.0002, "epoch": 2.024654463952185, "step": 2710}, {"loss": 1.5884, "grad_norm": 0.36497923731803894, "learning_rate": 0.0002, "epoch": 2.0321255136346656, "step": 2720}, {"loss": 1.6999, "grad_norm": 0.5021994113922119, "learning_rate": 0.0002, "epoch": 2.0395965633171462, "step": 2730}, {"loss": 1.6655, "grad_norm": 0.45896220207214355, "learning_rate": 0.0002, "epoch": 2.0470676129996264, "step": 2740}, {"loss": 1.6305, "grad_norm": 0.3973815143108368, "learning_rate": 0.0002, "epoch": 2.054538662682107, "step": 2750}, {"loss": 1.6301, "grad_norm": 0.4521815776824951, "learning_rate": 0.0002, "epoch": 2.062009712364587, "step": 2760}, {"loss": 1.6189, "grad_norm": 0.42775002121925354, "learning_rate": 0.0002, "epoch": 2.0694807620470677, "step": 2770}, {"loss": 1.6491, "grad_norm": 0.48158586025238037, "learning_rate": 0.0002, "epoch": 2.076951811729548, "step": 2780}, {"loss": 1.6301, "grad_norm": 0.4612371623516083, "learning_rate": 0.0002, "epoch": 2.0844228614120284, "step": 2790}, {"loss": 1.6327, "grad_norm": 0.42536866664886475, "learning_rate": 0.0002, "epoch": 2.091893911094509, "step": 2800}, {"loss": 1.651, "grad_norm": 0.48515772819519043, "learning_rate": 0.0002, "epoch": 2.099364960776989, "step": 2810}, {"loss": 1.6829, "grad_norm": 0.41418662667274475, "learning_rate": 0.0002, "epoch": 2.1068360104594697, "step": 2820}, {"loss": 1.6266, "grad_norm": 0.4683697819709778, "learning_rate": 0.0002, "epoch": 2.11430706014195, "step": 2830}, {"loss": 1.6586, "grad_norm": 0.4484657049179077, "learning_rate": 0.0002, "epoch": 2.1217781098244304, "step": 2840}, {"loss": 1.6483, "grad_norm": 0.6621400713920593, "learning_rate": 0.0002, "epoch": 2.1292491595069105, "step": 2850}, {"loss": 1.5755, "grad_norm": 0.45074811577796936, "learning_rate": 0.0002, "epoch": 2.136720209189391, "step": 2860}, {"loss": 1.6456, "grad_norm": 0.3513113558292389, "learning_rate": 0.0002, "epoch": 2.1441912588718717, "step": 2870}, {"loss": 1.6081, "grad_norm": 0.40411314368247986, "learning_rate": 0.0002, "epoch": 2.151662308554352, "step": 2880}, {"loss": 1.6323, "grad_norm": 0.4121065139770508, "learning_rate": 0.0002, "epoch": 2.1591333582368324, "step": 2890}, {"loss": 1.6324, "grad_norm": 0.44723689556121826, "learning_rate": 0.0002, "epoch": 2.1666044079193125, "step": 2900}, {"loss": 1.5699, "grad_norm": 0.4226122498512268, "learning_rate": 0.0002, "epoch": 2.174075457601793, "step": 2910}, {"loss": 1.5652, "grad_norm": 0.46617650985717773, "learning_rate": 0.0002, "epoch": 2.1815465072842732, "step": 2920}, {"loss": 1.6378, "grad_norm": 0.4506422281265259, "learning_rate": 0.0002, "epoch": 2.189017556966754, "step": 2930}, {"loss": 1.6112, "grad_norm": 0.4892672896385193, "learning_rate": 0.0002, "epoch": 2.1964886066492344, "step": 2940}, {"loss": 1.6176, "grad_norm": 0.44095516204833984, "learning_rate": 0.0002, "epoch": 2.2039596563317145, "step": 2950}, {"loss": 1.6058, "grad_norm": 0.41522109508514404, "learning_rate": 0.0002, "epoch": 2.211430706014195, "step": 2960}, {"loss": 1.5964, "grad_norm": 0.4860858917236328, "learning_rate": 0.0002, "epoch": 2.2189017556966752, "step": 2970}, {"loss": 1.6427, "grad_norm": 0.42662516236305237, "learning_rate": 0.0002, "epoch": 2.226372805379156, "step": 2980}, {"loss": 1.6313, "grad_norm": 0.4390648305416107, "learning_rate": 0.0002, "epoch": 2.233843855061636, "step": 2990}, {"loss": 1.5992, "grad_norm": 0.47515565156936646, "learning_rate": 0.0002, "epoch": 2.2413149047441165, "step": 3000}, {"loss": 1.5563, "grad_norm": 0.4104543924331665, "learning_rate": 0.0002, "epoch": 2.248785954426597, "step": 3010}, {"loss": 1.6895, "grad_norm": 0.4404028654098511, "learning_rate": 0.0002, "epoch": 2.2562570041090773, "step": 3020}, {"loss": 1.6088, "grad_norm": 0.4717366695404053, "learning_rate": 0.0002, "epoch": 2.263728053791558, "step": 3030}, {"loss": 1.7287, "grad_norm": 0.48345857858657837, "learning_rate": 0.0002, "epoch": 2.271199103474038, "step": 3040}, {"loss": 1.681, "grad_norm": 0.5312452912330627, "learning_rate": 0.0002, "epoch": 2.2786701531565186, "step": 3050}, {"loss": 1.5901, "grad_norm": 0.5073099732398987, "learning_rate": 0.0002, "epoch": 2.2861412028389987, "step": 3060}, {"loss": 1.6914, "grad_norm": 0.5027463436126709, "learning_rate": 0.0002, "epoch": 2.2936122525214793, "step": 3070}, {"loss": 1.5862, "grad_norm": 0.5436304807662964, "learning_rate": 0.0002, "epoch": 2.30108330220396, "step": 3080}, {"loss": 1.5763, "grad_norm": 0.4701065123081207, "learning_rate": 0.0002, "epoch": 2.30855435188644, "step": 3090}, {"loss": 1.6177, "grad_norm": 0.46988746523857117, "learning_rate": 0.0002, "epoch": 2.3160254015689206, "step": 3100}, {"loss": 1.6502, "grad_norm": 0.45112869143486023, "learning_rate": 0.0002, "epoch": 2.3234964512514007, "step": 3110}, {"loss": 1.6291, "grad_norm": 0.5173566937446594, "learning_rate": 0.0002, "epoch": 2.3309675009338813, "step": 3120}, {"loss": 1.6743, "grad_norm": 0.40345850586891174, "learning_rate": 0.0002, "epoch": 2.3384385506163614, "step": 3130}, {"loss": 1.621, "grad_norm": 0.4218924939632416, "learning_rate": 0.0002, "epoch": 2.345909600298842, "step": 3140}, {"loss": 1.6341, "grad_norm": 0.41857317090034485, "learning_rate": 0.0002, "epoch": 2.3533806499813226, "step": 3150}, {"loss": 1.6087, "grad_norm": 0.4197218418121338, "learning_rate": 0.0002, "epoch": 2.3608516996638027, "step": 3160}, {"loss": 1.6572, "grad_norm": 0.4260677397251129, "learning_rate": 0.0002, "epoch": 2.3683227493462833, "step": 3170}, {"loss": 1.6376, "grad_norm": 0.4209042191505432, "learning_rate": 0.0002, "epoch": 2.3757937990287634, "step": 3180}, {"loss": 1.634, "grad_norm": 0.4092234969139099, "learning_rate": 0.0002, "epoch": 2.383264848711244, "step": 3190}, {"loss": 1.6339, "grad_norm": 0.4928431510925293, "learning_rate": 0.0002, "epoch": 2.390735898393724, "step": 3200}, {"loss": 1.6015, "grad_norm": 0.49252402782440186, "learning_rate": 0.0002, "epoch": 2.3982069480762047, "step": 3210}, {"loss": 1.5773, "grad_norm": 0.4368397295475006, "learning_rate": 0.0002, "epoch": 2.4056779977586853, "step": 3220}, {"loss": 1.6629, "grad_norm": 0.46122390031814575, "learning_rate": 0.0002, "epoch": 2.4131490474411654, "step": 3230}, {"loss": 1.6224, "grad_norm": 0.4272301197052002, "learning_rate": 0.0002, "epoch": 2.420620097123646, "step": 3240}, {"loss": 1.5961, "grad_norm": 0.41480937600135803, "learning_rate": 0.0002, "epoch": 2.428091146806126, "step": 3250}, {"loss": 1.6281, "grad_norm": 0.48911941051483154, "learning_rate": 0.0002, "epoch": 2.4355621964886067, "step": 3260}, {"loss": 1.6846, "grad_norm": 0.4444098472595215, "learning_rate": 0.0002, "epoch": 2.443033246171087, "step": 3270}, {"loss": 1.6961, "grad_norm": 0.5111684799194336, "learning_rate": 0.0002, "epoch": 2.4505042958535674, "step": 3280}, {"loss": 1.6152, "grad_norm": 0.5058825016021729, "learning_rate": 0.0002, "epoch": 2.457975345536048, "step": 3290}, {"loss": 1.625, "grad_norm": 0.44173210859298706, "learning_rate": 0.0002, "epoch": 2.465446395218528, "step": 3300}, {"loss": 1.6491, "grad_norm": 0.4659745991230011, "learning_rate": 0.0002, "epoch": 2.4729174449010087, "step": 3310}, {"loss": 1.6114, "grad_norm": 0.47237497568130493, "learning_rate": 0.0002, "epoch": 2.480388494583489, "step": 3320}, {"loss": 1.6193, "grad_norm": 0.47303131222724915, "learning_rate": 0.0002, "epoch": 2.4878595442659694, "step": 3330}, {"loss": 1.7256, "grad_norm": 0.4522389769554138, "learning_rate": 0.0002, "epoch": 2.4953305939484496, "step": 3340}, {"loss": 1.6834, "grad_norm": 0.4467332363128662, "learning_rate": 0.0002, "epoch": 2.50280164363093, "step": 3350}, {"loss": 1.6108, "grad_norm": 0.4413762092590332, "learning_rate": 0.0002, "epoch": 2.5102726933134107, "step": 3360}, {"loss": 1.537, "grad_norm": 0.495514452457428, "learning_rate": 0.0002, "epoch": 2.517743742995891, "step": 3370}, {"loss": 1.5839, "grad_norm": 0.4429773986339569, "learning_rate": 0.0002, "epoch": 2.5252147926783715, "step": 3380}, {"loss": 1.6522, "grad_norm": 0.4589079022407532, "learning_rate": 0.0002, "epoch": 2.5326858423608516, "step": 3390}, {"loss": 1.6529, "grad_norm": 0.4683997333049774, "learning_rate": 0.0002, "epoch": 2.540156892043332, "step": 3400}, {"loss": 1.6745, "grad_norm": 0.4651731252670288, "learning_rate": 0.0002, "epoch": 2.5476279417258123, "step": 3410}, {"loss": 1.5918, "grad_norm": 0.45818084478378296, "learning_rate": 0.0002, "epoch": 2.555098991408293, "step": 3420}, {"loss": 1.6326, "grad_norm": 0.45209529995918274, "learning_rate": 0.0002, "epoch": 2.5625700410907735, "step": 3430}, {"loss": 1.5606, "grad_norm": 0.4344733655452728, "learning_rate": 0.0002, "epoch": 2.5700410907732536, "step": 3440}, {"loss": 1.6748, "grad_norm": 0.47435566782951355, "learning_rate": 0.0002, "epoch": 2.577512140455734, "step": 3450}, {"loss": 1.6237, "grad_norm": 0.43841999769210815, "learning_rate": 0.0002, "epoch": 2.5849831901382143, "step": 3460}, {"loss": 1.7207, "grad_norm": 0.4323869049549103, "learning_rate": 0.0002, "epoch": 2.592454239820695, "step": 3470}, {"loss": 1.5494, "grad_norm": 0.44355881214141846, "learning_rate": 0.0002, "epoch": 2.599925289503175, "step": 3480}, {"loss": 1.665, "grad_norm": 0.45847779512405396, "learning_rate": 0.0002, "epoch": 2.6073963391856556, "step": 3490}, {"loss": 1.6006, "grad_norm": 0.4411061704158783, "learning_rate": 0.0002, "epoch": 2.614867388868136, "step": 3500}, {"loss": 1.5868, "grad_norm": 0.4446796178817749, "learning_rate": 0.0002, "epoch": 2.6223384385506163, "step": 3510}, {"loss": 1.5946, "grad_norm": 0.41969653964042664, "learning_rate": 0.0002, "epoch": 2.629809488233097, "step": 3520}, {"loss": 1.6798, "grad_norm": 0.5263747572898865, "learning_rate": 0.0002, "epoch": 2.637280537915577, "step": 3530}, {"loss": 1.6309, "grad_norm": 0.47719451785087585, "learning_rate": 0.0002, "epoch": 2.6447515875980576, "step": 3540}, {"loss": 1.7024, "grad_norm": 0.46574118733406067, "learning_rate": 0.0002, "epoch": 2.6522226372805378, "step": 3550}, {"loss": 1.618, "grad_norm": 0.46867135167121887, "learning_rate": 0.0002, "epoch": 2.6596936869630183, "step": 3560}, {"loss": 1.5885, "grad_norm": 0.4441198706626892, "learning_rate": 0.0002, "epoch": 2.667164736645499, "step": 3570}, {"loss": 1.6426, "grad_norm": 0.4871319830417633, "learning_rate": 0.0002, "epoch": 2.674635786327979, "step": 3580}, {"loss": 1.6575, "grad_norm": 0.43900373578071594, "learning_rate": 0.0002, "epoch": 2.6821068360104596, "step": 3590}, {"loss": 1.6071, "grad_norm": 0.42509549856185913, "learning_rate": 0.0002, "epoch": 2.6895778856929398, "step": 3600}, {"loss": 1.5651, "grad_norm": 0.4691086709499359, "learning_rate": 0.0002, "epoch": 2.6970489353754203, "step": 3610}, {"loss": 1.5491, "grad_norm": 0.46318942308425903, "learning_rate": 0.0002, "epoch": 2.7045199850579005, "step": 3620}, {"loss": 1.5422, "grad_norm": 0.44631096720695496, "learning_rate": 0.0002, "epoch": 2.711991034740381, "step": 3630}, {"loss": 1.6831, "grad_norm": 0.42315489053726196, "learning_rate": 0.0002, "epoch": 2.7194620844228616, "step": 3640}, {"loss": 1.6008, "grad_norm": 0.4971241056919098, "learning_rate": 0.0002, "epoch": 2.7269331341053418, "step": 3650}, {"loss": 1.6042, "grad_norm": 0.4578486382961273, "learning_rate": 0.0002, "epoch": 2.7344041837878224, "step": 3660}, {"loss": 1.6076, "grad_norm": 0.46584776043891907, "learning_rate": 0.0002, "epoch": 2.7418752334703025, "step": 3670}, {"loss": 1.6809, "grad_norm": 0.4951731264591217, "learning_rate": 0.0002, "epoch": 2.749346283152783, "step": 3680}, {"loss": 1.6226, "grad_norm": 0.4935225546360016, "learning_rate": 0.0002, "epoch": 2.756817332835263, "step": 3690}, {"loss": 1.5878, "grad_norm": 0.41805586218833923, "learning_rate": 0.0002, "epoch": 2.764288382517744, "step": 3700}, {"loss": 1.7173, "grad_norm": 0.4417555630207062, "learning_rate": 0.0002, "epoch": 2.7717594322002244, "step": 3710}, {"loss": 1.6398, "grad_norm": 0.48229655623435974, "learning_rate": 0.0002, "epoch": 2.7792304818827045, "step": 3720}, {"loss": 1.6074, "grad_norm": 0.48562315106391907, "learning_rate": 0.0002, "epoch": 2.786701531565185, "step": 3730}, {"loss": 1.607, "grad_norm": 0.4473940432071686, "learning_rate": 0.0002, "epoch": 2.794172581247665, "step": 3740}, {"loss": 1.6065, "grad_norm": 0.4626813232898712, "learning_rate": 0.0002, "epoch": 2.801643630930146, "step": 3750}, {"loss": 1.6296, "grad_norm": 0.4339792728424072, "learning_rate": 0.0002, "epoch": 2.809114680612626, "step": 3760}, {"loss": 1.6815, "grad_norm": 0.5250858068466187, "learning_rate": 0.0002, "epoch": 2.8165857302951065, "step": 3770}, {"loss": 1.6644, "grad_norm": 0.4537523090839386, "learning_rate": 0.0002, "epoch": 2.824056779977587, "step": 3780}, {"loss": 1.6535, "grad_norm": 0.5646113157272339, "learning_rate": 0.0002, "epoch": 2.831527829660067, "step": 3790}, {"loss": 1.5712, "grad_norm": 0.44243332743644714, "learning_rate": 0.0002, "epoch": 2.8389988793425474, "step": 3800}, {"loss": 1.6478, "grad_norm": 0.4585791826248169, "learning_rate": 0.0002, "epoch": 2.846469929025028, "step": 3810}, {"loss": 1.6854, "grad_norm": 0.489702045917511, "learning_rate": 0.0002, "epoch": 2.8539409787075085, "step": 3820}, {"loss": 1.7066, "grad_norm": 0.502470850944519, "learning_rate": 0.0002, "epoch": 2.8614120283899886, "step": 3830}, {"loss": 1.5785, "grad_norm": 0.4395960867404938, "learning_rate": 0.0002, "epoch": 2.8688830780724692, "step": 3840}, {"loss": 1.6434, "grad_norm": 0.4348670244216919, "learning_rate": 0.0002, "epoch": 2.87635412775495, "step": 3850}, {"loss": 1.6163, "grad_norm": 0.48852720856666565, "learning_rate": 0.0002, "epoch": 2.88382517743743, "step": 3860}, {"loss": 1.5916, "grad_norm": 0.45317450165748596, "learning_rate": 0.0002, "epoch": 2.89129622711991, "step": 3870}, {"loss": 1.6486, "grad_norm": 0.4732758700847626, "learning_rate": 0.0002, "epoch": 2.8987672768023907, "step": 3880}, {"loss": 1.6758, "grad_norm": 0.45238012075424194, "learning_rate": 0.0002, "epoch": 2.9062383264848712, "step": 3890}, {"loss": 1.6228, "grad_norm": 0.48838064074516296, "learning_rate": 0.0002, "epoch": 2.9137093761673514, "step": 3900}, {"loss": 1.658, "grad_norm": 0.43496349453926086, "learning_rate": 0.0002, "epoch": 2.921180425849832, "step": 3910}, {"loss": 1.7063, "grad_norm": 0.47963935136795044, "learning_rate": 0.0002, "epoch": 2.9286514755323125, "step": 3920}, {"loss": 1.6553, "grad_norm": 0.4544987976551056, "learning_rate": 0.0002, "epoch": 2.9361225252147927, "step": 3930}, {"loss": 1.6192, "grad_norm": 0.4622892141342163, "learning_rate": 0.0002, "epoch": 2.943593574897273, "step": 3940}, {"loss": 1.6178, "grad_norm": 0.47026222944259644, "learning_rate": 0.0002, "epoch": 2.9510646245797534, "step": 3950}, {"loss": 1.6612, "grad_norm": 0.4549552798271179, "learning_rate": 0.0002, "epoch": 2.958535674262234, "step": 3960}, {"loss": 1.6458, "grad_norm": 0.46647515892982483, "learning_rate": 0.0002, "epoch": 2.966006723944714, "step": 3970}, {"loss": 1.6051, "grad_norm": 0.45095112919807434, "learning_rate": 0.0002, "epoch": 2.9734777736271947, "step": 3980}, {"loss": 1.6471, "grad_norm": 0.4690017104148865, "learning_rate": 0.0002, "epoch": 2.9809488233096753, "step": 3990}, {"loss": 1.6061, "grad_norm": 0.4603444039821625, "learning_rate": 0.0002, "epoch": 2.9884198729921554, "step": 4000}, {"loss": 1.6431, "grad_norm": 0.4743294417858124, "learning_rate": 0.0002, "epoch": 2.9958909226746355, "step": 4010}, {"eval_loss": 1.8252571821212769, "eval_runtime": 38.7853, "eval_samples_per_second": 13.278, "eval_steps_per_second": 1.676, "epoch": 2.999626447515876, "step": 4015}, {"loss": 1.6512, "grad_norm": 0.4919724464416504, "learning_rate": 0.0002, "epoch": 3.003361972357116, "step": 4020}, {"loss": 1.5354, "grad_norm": 0.4747185707092285, "learning_rate": 0.0002, "epoch": 3.0108330220395967, "step": 4030}, {"loss": 1.568, "grad_norm": 0.4797595143318176, "learning_rate": 0.0002, "epoch": 3.018304071722077, "step": 4040}, {"loss": 1.5194, "grad_norm": 0.5450999140739441, "learning_rate": 0.0002, "epoch": 3.0257751214045574, "step": 4050}, {"loss": 1.5065, "grad_norm": 0.49058812856674194, "learning_rate": 0.0002, "epoch": 3.0332461710870375, "step": 4060}, {"loss": 1.4884, "grad_norm": 0.5219563841819763, "learning_rate": 0.0002, "epoch": 3.040717220769518, "step": 4070}, {"loss": 1.4742, "grad_norm": 0.515628457069397, "learning_rate": 0.0002, "epoch": 3.0481882704519987, "step": 4080}, {"loss": 1.5313, "grad_norm": 0.6145984530448914, "learning_rate": 0.0002, "epoch": 3.055659320134479, "step": 4090}, {"loss": 1.4989, "grad_norm": 0.6067144274711609, "learning_rate": 0.0002, "epoch": 3.0631303698169594, "step": 4100}, {"loss": 1.528, "grad_norm": 0.5773133039474487, "learning_rate": 0.0002, "epoch": 3.0706014194994395, "step": 4110}, {"loss": 1.5374, "grad_norm": 0.6894241571426392, "learning_rate": 0.0002, "epoch": 3.07807246918192, "step": 4120}, {"loss": 1.5422, "grad_norm": 0.6422514915466309, "learning_rate": 0.0002, "epoch": 3.0855435188644003, "step": 4130}, {"loss": 1.4724, "grad_norm": 0.6119855046272278, "learning_rate": 0.0002, "epoch": 3.093014568546881, "step": 4140}, {"loss": 1.5361, "grad_norm": 0.5847280025482178, "learning_rate": 0.0002, "epoch": 3.1004856182293614, "step": 4150}, {"loss": 1.5151, "grad_norm": 0.5401515960693359, "learning_rate": 0.0002, "epoch": 3.1079566679118416, "step": 4160}, {"loss": 1.502, "grad_norm": 0.6501587629318237, "learning_rate": 0.0002, "epoch": 3.115427717594322, "step": 4170}, {"loss": 1.4952, "grad_norm": 0.5988039374351501, "learning_rate": 0.0002, "epoch": 3.1228987672768023, "step": 4180}, {"loss": 1.5287, "grad_norm": 0.4982665181159973, "learning_rate": 0.0002, "epoch": 3.130369816959283, "step": 4190}, {"loss": 1.5078, "grad_norm": 0.5548039078712463, "learning_rate": 0.0002, "epoch": 3.137840866641763, "step": 4200}, {"loss": 1.4904, "grad_norm": 0.5920777320861816, "learning_rate": 0.0002, "epoch": 3.1453119163242436, "step": 4210}, {"loss": 1.442, "grad_norm": 0.6965190172195435, "learning_rate": 0.0002, "epoch": 3.152782966006724, "step": 4220}, {"loss": 1.557, "grad_norm": 0.5196244716644287, "learning_rate": 0.0002, "epoch": 3.1602540156892043, "step": 4230}, {"loss": 1.5706, "grad_norm": 0.6942682266235352, "learning_rate": 0.0002, "epoch": 3.167725065371685, "step": 4240}, {"loss": 1.5407, "grad_norm": 0.5765156149864197, "learning_rate": 0.0002, "epoch": 3.175196115054165, "step": 4250}, {"loss": 1.4963, "grad_norm": 0.5801976919174194, "learning_rate": 0.0002, "epoch": 3.1826671647366456, "step": 4260}, {"loss": 1.4988, "grad_norm": 0.6260752081871033, "learning_rate": 0.0002, "epoch": 3.1901382144191257, "step": 4270}, {"loss": 1.5074, "grad_norm": 0.6610770225524902, "learning_rate": 0.0002, "epoch": 3.1976092641016063, "step": 4280}, {"loss": 1.4657, "grad_norm": 0.5762143135070801, "learning_rate": 0.0002, "epoch": 3.205080313784087, "step": 4290}, {"loss": 1.5181, "grad_norm": 0.5926990509033203, "learning_rate": 0.0002, "epoch": 3.212551363466567, "step": 4300}, {"loss": 1.5492, "grad_norm": 0.7373854517936707, "learning_rate": 0.0002, "epoch": 3.2200224131490476, "step": 4310}, {"loss": 1.4648, "grad_norm": 0.5963311195373535, "learning_rate": 0.0002, "epoch": 3.2274934628315277, "step": 4320}, {"loss": 1.5262, "grad_norm": 0.5754616856575012, "learning_rate": 0.0002, "epoch": 3.2349645125140083, "step": 4330}, {"loss": 1.4767, "grad_norm": 0.6116095781326294, "learning_rate": 0.0002, "epoch": 3.2424355621964884, "step": 4340}, {"loss": 1.5008, "grad_norm": 0.6001536846160889, "learning_rate": 0.0002, "epoch": 3.249906611878969, "step": 4350}, {"loss": 1.5738, "grad_norm": 0.5270227789878845, "learning_rate": 0.0002, "epoch": 3.257377661561449, "step": 4360}, {"loss": 1.5235, "grad_norm": 0.6666602492332458, "learning_rate": 0.0002, "epoch": 3.2648487112439297, "step": 4370}, {"loss": 1.5665, "grad_norm": 0.520310640335083, "learning_rate": 0.0002, "epoch": 3.2723197609264103, "step": 4380}, {"loss": 1.542, "grad_norm": 0.5165975093841553, "learning_rate": 0.0002, "epoch": 3.2797908106088904, "step": 4390}, {"loss": 1.4746, "grad_norm": 0.6080228686332703, "learning_rate": 0.0002, "epoch": 3.287261860291371, "step": 4400}, {"loss": 1.4901, "grad_norm": 0.670122504234314, "learning_rate": 0.0002, "epoch": 3.294732909973851, "step": 4410}, {"loss": 1.4677, "grad_norm": 0.6019457578659058, "learning_rate": 0.0002, "epoch": 3.3022039596563317, "step": 4420}, {"loss": 1.4249, "grad_norm": 0.5519300103187561, "learning_rate": 0.0002, "epoch": 3.309675009338812, "step": 4430}, {"loss": 1.555, "grad_norm": 0.5958521962165833, "learning_rate": 0.0002, "epoch": 3.3171460590212924, "step": 4440}, {"loss": 1.5067, "grad_norm": 0.5552705526351929, "learning_rate": 0.0002, "epoch": 3.324617108703773, "step": 4450}, {"loss": 1.5926, "grad_norm": 0.6583784818649292, "learning_rate": 0.0002, "epoch": 3.332088158386253, "step": 4460}, {"loss": 1.4206, "grad_norm": 0.5815939903259277, "learning_rate": 0.0002, "epoch": 3.3395592080687337, "step": 4470}, {"loss": 1.5942, "grad_norm": 1.3342205286026, "learning_rate": 0.0002, "epoch": 3.347030257751214, "step": 4480}, {"loss": 1.484, "grad_norm": 0.6341500878334045, "learning_rate": 0.0002, "epoch": 3.3545013074336945, "step": 4490}, {"loss": 1.5219, "grad_norm": 0.6384079456329346, "learning_rate": 0.0002, "epoch": 3.3619723571161746, "step": 4500}, {"loss": 1.5222, "grad_norm": 0.6098346710205078, "learning_rate": 0.0002, "epoch": 3.369443406798655, "step": 4510}, {"loss": 1.5475, "grad_norm": 0.5958296656608582, "learning_rate": 0.0002, "epoch": 3.3769144564811358, "step": 4520}, {"loss": 1.5171, "grad_norm": 0.6157881617546082, "learning_rate": 0.0002, "epoch": 3.384385506163616, "step": 4530}, {"loss": 1.569, "grad_norm": 0.5671007037162781, "learning_rate": 0.0002, "epoch": 3.3918565558460965, "step": 4540}, {"loss": 1.604, "grad_norm": 0.6203294992446899, "learning_rate": 0.0002, "epoch": 3.3993276055285766, "step": 4550}, {"loss": 1.5364, "grad_norm": 0.6743317246437073, "learning_rate": 0.0002, "epoch": 3.406798655211057, "step": 4560}, {"loss": 1.5034, "grad_norm": 0.731765627861023, "learning_rate": 0.0002, "epoch": 3.4142697048935373, "step": 4570}, {"loss": 1.4585, "grad_norm": 0.6285187602043152, "learning_rate": 0.0002, "epoch": 3.421740754576018, "step": 4580}, {"loss": 1.5296, "grad_norm": 0.612680196762085, "learning_rate": 0.0002, "epoch": 3.4292118042584985, "step": 4590}, {"loss": 1.5577, "grad_norm": 0.6413681507110596, "learning_rate": 0.0002, "epoch": 3.4366828539409786, "step": 4600}, {"loss": 1.5026, "grad_norm": 0.6240990161895752, "learning_rate": 0.0002, "epoch": 3.444153903623459, "step": 4610}, {"loss": 1.5887, "grad_norm": 0.5095735192298889, "learning_rate": 0.0002, "epoch": 3.4516249533059393, "step": 4620}, {"loss": 1.4906, "grad_norm": 0.5699611902236938, "learning_rate": 0.0002, "epoch": 3.45909600298842, "step": 4630}, {"loss": 1.5176, "grad_norm": 0.7289775609970093, "learning_rate": 0.0002, "epoch": 3.4665670526709, "step": 4640}, {"loss": 1.5467, "grad_norm": 0.6211609840393066, "learning_rate": 0.0002, "epoch": 3.4740381023533806, "step": 4650}, {"loss": 1.533, "grad_norm": 0.5714802145957947, "learning_rate": 0.0002, "epoch": 3.481509152035861, "step": 4660}, {"loss": 1.5096, "grad_norm": 0.6287049651145935, "learning_rate": 0.0002, "epoch": 3.4889802017183413, "step": 4670}, {"loss": 1.4212, "grad_norm": 0.5480595827102661, "learning_rate": 0.0002, "epoch": 3.496451251400822, "step": 4680}, {"loss": 1.4746, "grad_norm": 0.5683253407478333, "learning_rate": 0.0002, "epoch": 3.503922301083302, "step": 4690}, {"loss": 1.5012, "grad_norm": 0.601140558719635, "learning_rate": 0.0002, "epoch": 3.5113933507657826, "step": 4700}, {"loss": 1.5383, "grad_norm": 0.5344498157501221, "learning_rate": 0.0002, "epoch": 3.5188644004482628, "step": 4710}, {"loss": 1.5428, "grad_norm": 0.5739690661430359, "learning_rate": 0.0002, "epoch": 3.5263354501307433, "step": 4720}, {"loss": 1.5589, "grad_norm": 0.5640085935592651, "learning_rate": 0.0002, "epoch": 3.533806499813224, "step": 4730}, {"loss": 1.487, "grad_norm": 0.5967805981636047, "learning_rate": 0.0002, "epoch": 3.541277549495704, "step": 4740}, {"loss": 1.5461, "grad_norm": 0.6138835549354553, "learning_rate": 0.0002, "epoch": 3.5487485991781846, "step": 4750}, {"loss": 1.5502, "grad_norm": 0.6779900193214417, "learning_rate": 0.0002, "epoch": 3.5562196488606648, "step": 4760}, {"loss": 1.4917, "grad_norm": 0.6122010350227356, "learning_rate": 0.0002, "epoch": 3.5636906985431454, "step": 4770}, {"loss": 1.5405, "grad_norm": 0.5685241222381592, "learning_rate": 0.0002, "epoch": 3.5711617482256255, "step": 4780}, {"loss": 1.5427, "grad_norm": 0.604583203792572, "learning_rate": 0.0002, "epoch": 3.578632797908106, "step": 4790}, {"loss": 1.4514, "grad_norm": 0.651165246963501, "learning_rate": 0.0002, "epoch": 3.5861038475905866, "step": 4800}, {"loss": 1.4109, "grad_norm": 0.6398511528968811, "learning_rate": 0.0002, "epoch": 3.593574897273067, "step": 4810}, {"loss": 1.4261, "grad_norm": 0.6444641351699829, "learning_rate": 0.0002, "epoch": 3.6010459469555474, "step": 4820}, {"loss": 1.5274, "grad_norm": 0.6018481850624084, "learning_rate": 0.0002, "epoch": 3.6085169966380275, "step": 4830}, {"loss": 1.4647, "grad_norm": 0.6025291085243225, "learning_rate": 0.0002, "epoch": 3.615988046320508, "step": 4840}, {"loss": 1.5609, "grad_norm": 0.6810156106948853, "learning_rate": 0.0002, "epoch": 3.623459096002988, "step": 4850}, {"loss": 1.5299, "grad_norm": 0.6408044695854187, "learning_rate": 0.0002, "epoch": 3.630930145685469, "step": 4860}, {"loss": 1.5366, "grad_norm": 0.5608272552490234, "learning_rate": 0.0002, "epoch": 3.6384011953679494, "step": 4870}, {"loss": 1.5188, "grad_norm": 0.6136814951896667, "learning_rate": 0.0002, "epoch": 3.6458722450504295, "step": 4880}, {"loss": 1.5021, "grad_norm": 0.5927900075912476, "learning_rate": 0.0002, "epoch": 3.65334329473291, "step": 4890}, {"loss": 1.6084, "grad_norm": 0.5336901545524597, "learning_rate": 0.0002, "epoch": 3.66081434441539, "step": 4900}, {"loss": 1.5701, "grad_norm": 0.7823320627212524, "learning_rate": 0.0002, "epoch": 3.668285394097871, "step": 4910}, {"loss": 1.4881, "grad_norm": 0.6703504323959351, "learning_rate": 0.0002, "epoch": 3.675756443780351, "step": 4920}, {"loss": 1.5332, "grad_norm": 0.6061160564422607, "learning_rate": 0.0002, "epoch": 3.6832274934628315, "step": 4930}, {"loss": 1.5405, "grad_norm": 0.6237227916717529, "learning_rate": 0.0002, "epoch": 3.690698543145312, "step": 4940}, {"loss": 1.497, "grad_norm": 0.5985278487205505, "learning_rate": 0.0002, "epoch": 3.6981695928277922, "step": 4950}, {"loss": 1.5132, "grad_norm": 0.6483839750289917, "learning_rate": 0.0002, "epoch": 3.705640642510273, "step": 4960}, {"loss": 1.5338, "grad_norm": 0.5788805484771729, "learning_rate": 0.0002, "epoch": 3.713111692192753, "step": 4970}, {"loss": 1.5258, "grad_norm": 0.5609974265098572, "learning_rate": 0.0002, "epoch": 3.7205827418752335, "step": 4980}, {"loss": 1.4759, "grad_norm": 0.5681300759315491, "learning_rate": 0.0002, "epoch": 3.7280537915577137, "step": 4990}, {"loss": 1.6018, "grad_norm": 0.5860186219215393, "learning_rate": 0.0002, "epoch": 3.7355248412401942, "step": 5000}, {"loss": 1.58, "grad_norm": 0.5718157291412354, "learning_rate": 0.0002, "epoch": 3.742995890922675, "step": 5010}, {"loss": 1.5834, "grad_norm": 0.6173721551895142, "learning_rate": 0.0002, "epoch": 3.750466940605155, "step": 5020}, {"loss": 1.5617, "grad_norm": 0.629152238368988, "learning_rate": 0.0002, "epoch": 3.7579379902876355, "step": 5030}, {"loss": 1.519, "grad_norm": 0.5666284561157227, "learning_rate": 0.0002, "epoch": 3.7654090399701157, "step": 5040}, {"loss": 1.5329, "grad_norm": 0.6053005456924438, "learning_rate": 0.0002, "epoch": 3.7728800896525962, "step": 5050}, {"loss": 1.5404, "grad_norm": 0.5870583057403564, "learning_rate": 0.0002, "epoch": 3.7803511393350764, "step": 5060}, {"loss": 1.4444, "grad_norm": 0.5422009229660034, "learning_rate": 0.0002, "epoch": 3.787822189017557, "step": 5070}, {"loss": 1.5308, "grad_norm": 0.5396918058395386, "learning_rate": 0.0002, "epoch": 3.7952932387000375, "step": 5080}, {"loss": 1.464, "grad_norm": 0.5544713139533997, "learning_rate": 0.0002, "epoch": 3.8027642883825177, "step": 5090}, {"loss": 1.4752, "grad_norm": 0.5983749628067017, "learning_rate": 0.0002, "epoch": 3.8102353380649983, "step": 5100}, {"loss": 1.4972, "grad_norm": 0.5702024102210999, "learning_rate": 0.0002, "epoch": 3.8177063877474784, "step": 5110}, {"loss": 1.5471, "grad_norm": 0.5436882376670837, "learning_rate": 0.0002, "epoch": 3.825177437429959, "step": 5120}, {"loss": 1.5118, "grad_norm": 0.5453617572784424, "learning_rate": 0.0002, "epoch": 3.832648487112439, "step": 5130}, {"loss": 1.5732, "grad_norm": 0.6269069314002991, "learning_rate": 0.0002, "epoch": 3.8401195367949197, "step": 5140}, {"loss": 1.4959, "grad_norm": 0.6189185380935669, "learning_rate": 0.0002, "epoch": 3.8475905864774003, "step": 5150}, {"loss": 1.4999, "grad_norm": 0.6653388142585754, "learning_rate": 0.0002, "epoch": 3.8550616361598804, "step": 5160}, {"loss": 1.5075, "grad_norm": 0.5771768689155579, "learning_rate": 0.0002, "epoch": 3.862532685842361, "step": 5170}, {"loss": 1.5545, "grad_norm": 0.6052790880203247, "learning_rate": 0.0002, "epoch": 3.870003735524841, "step": 5180}, {"loss": 1.4987, "grad_norm": 0.6572316884994507, "learning_rate": 0.0002, "epoch": 3.8774747852073217, "step": 5190}, {"loss": 1.5241, "grad_norm": 0.670576810836792, "learning_rate": 0.0002, "epoch": 3.884945834889802, "step": 5200}, {"loss": 1.4777, "grad_norm": 0.5728798508644104, "learning_rate": 0.0002, "epoch": 3.8924168845722824, "step": 5210}, {"loss": 1.5351, "grad_norm": 0.6340774297714233, "learning_rate": 0.0002, "epoch": 3.899887934254763, "step": 5220}, {"loss": 1.5081, "grad_norm": 0.5981315970420837, "learning_rate": 0.0002, "epoch": 3.907358983937243, "step": 5230}, {"loss": 1.4875, "grad_norm": 0.6212025880813599, "learning_rate": 0.0002, "epoch": 3.9148300336197237, "step": 5240}, {"loss": 1.5545, "grad_norm": 0.6202296018600464, "learning_rate": 0.0002, "epoch": 3.922301083302204, "step": 5250}, {"loss": 1.5765, "grad_norm": 0.6159142255783081, "learning_rate": 0.0002, "epoch": 3.9297721329846844, "step": 5260}, {"loss": 1.4938, "grad_norm": 0.6519438624382019, "learning_rate": 0.0002, "epoch": 3.9372431826671646, "step": 5270}, {"loss": 1.4859, "grad_norm": 0.539813756942749, "learning_rate": 0.0002, "epoch": 3.944714232349645, "step": 5280}, {"loss": 1.5921, "grad_norm": 0.6443665027618408, "learning_rate": 0.0002, "epoch": 3.9521852820321257, "step": 5290}, {"loss": 1.5153, "grad_norm": 0.6635757684707642, "learning_rate": 0.0002, "epoch": 3.959656331714606, "step": 5300}, {"loss": 1.5485, "grad_norm": 0.589363157749176, "learning_rate": 0.0002, "epoch": 3.9671273813970864, "step": 5310}, {"loss": 1.5498, "grad_norm": 0.5788735747337341, "learning_rate": 0.0002, "epoch": 3.9745984310795666, "step": 5320}, {"loss": 1.5607, "grad_norm": 0.5976864695549011, "learning_rate": 0.0002, "epoch": 3.982069480762047, "step": 5330}, {"loss": 1.5302, "grad_norm": 0.6624067425727844, "learning_rate": 0.0002, "epoch": 3.9895405304445273, "step": 5340}, {"loss": 1.5904, "grad_norm": 0.6738956570625305, "learning_rate": 0.0002, "epoch": 3.997011580127008, "step": 5350}, {"eval_loss": 1.868006944656372, "eval_runtime": 38.5153, "eval_samples_per_second": 13.371, "eval_steps_per_second": 1.688, "epoch": 4.0, "step": 5354}, {"loss": 1.4535, "grad_norm": 0.6023468971252441, "learning_rate": 0.0002, "epoch": 4.004482629809488, "step": 5360}, {"loss": 1.3987, "grad_norm": 0.8589285612106323, "learning_rate": 0.0002, "epoch": 4.011953679491969, "step": 5370}, {"loss": 1.3952, "grad_norm": 0.7477491497993469, "learning_rate": 0.0002, "epoch": 4.019424729174449, "step": 5380}, {"loss": 1.3745, "grad_norm": 0.7601922154426575, "learning_rate": 0.0002, "epoch": 4.02689577885693, "step": 5390}, {"loss": 1.4133, "grad_norm": 0.8115614056587219, "learning_rate": 0.0002, "epoch": 4.03436682853941, "step": 5400}, {"loss": 1.3748, "grad_norm": 0.669925332069397, "learning_rate": 0.0002, "epoch": 4.04183787822189, "step": 5410}, {"loss": 1.2835, "grad_norm": 0.8091904520988464, "learning_rate": 0.0002, "epoch": 4.04930892790437, "step": 5420}, {"loss": 1.3615, "grad_norm": 0.709405779838562, "learning_rate": 0.0002, "epoch": 4.056779977586851, "step": 5430}, {"loss": 1.3558, "grad_norm": 1.0006179809570312, "learning_rate": 0.0002, "epoch": 4.064251027269331, "step": 5440}, {"loss": 1.3491, "grad_norm": 0.7017965912818909, "learning_rate": 0.0002, "epoch": 4.071722076951811, "step": 5450}, {"loss": 1.3642, "grad_norm": 0.8991572260856628, "learning_rate": 0.0002, "epoch": 4.0791931266342925, "step": 5460}, {"loss": 1.392, "grad_norm": 0.9064797759056091, "learning_rate": 0.0002, "epoch": 4.086664176316773, "step": 5470}, {"loss": 1.3425, "grad_norm": 0.7981749176979065, "learning_rate": 0.0002, "epoch": 4.094135225999253, "step": 5480}, {"loss": 1.3826, "grad_norm": 0.7280883193016052, "learning_rate": 0.0002, "epoch": 4.101606275681733, "step": 5490}, {"loss": 1.3275, "grad_norm": 0.7419600486755371, "learning_rate": 0.0002, "epoch": 4.109077325364214, "step": 5500}, {"loss": 1.3199, "grad_norm": 0.8019949197769165, "learning_rate": 0.0002, "epoch": 4.116548375046694, "step": 5510}, {"loss": 1.3133, "grad_norm": 0.7501229047775269, "learning_rate": 0.0002, "epoch": 4.124019424729174, "step": 5520}, {"loss": 1.4432, "grad_norm": 0.8166249990463257, "learning_rate": 0.0002, "epoch": 4.131490474411655, "step": 5530}, {"loss": 1.3901, "grad_norm": 0.9728496074676514, "learning_rate": 0.0002, "epoch": 4.138961524094135, "step": 5540}, {"loss": 1.3538, "grad_norm": 0.7590922117233276, "learning_rate": 0.0002, "epoch": 4.1464325737766154, "step": 5550}, {"loss": 1.4368, "grad_norm": 0.7759010791778564, "learning_rate": 0.0002, "epoch": 4.153903623459096, "step": 5560}, {"loss": 1.3635, "grad_norm": 0.9057986736297607, "learning_rate": 0.0002, "epoch": 4.161374673141577, "step": 5570}, {"loss": 1.4152, "grad_norm": 0.8853937983512878, "learning_rate": 0.0002, "epoch": 4.168845722824057, "step": 5580}, {"loss": 1.3633, "grad_norm": 0.7070684432983398, "learning_rate": 0.0002, "epoch": 4.176316772506537, "step": 5590}, {"loss": 1.3218, "grad_norm": 0.7649410963058472, "learning_rate": 0.0002, "epoch": 4.183787822189018, "step": 5600}, {"loss": 1.3857, "grad_norm": 1.2048029899597168, "learning_rate": 0.0002, "epoch": 4.191258871871498, "step": 5610}, {"loss": 1.3629, "grad_norm": 0.7986605763435364, "learning_rate": 0.0002, "epoch": 4.198729921553978, "step": 5620}, {"loss": 1.3995, "grad_norm": 0.8151885867118835, "learning_rate": 0.0002, "epoch": 4.206200971236458, "step": 5630}, {"loss": 1.3782, "grad_norm": 0.7719064354896545, "learning_rate": 0.0002, "epoch": 4.213672020918939, "step": 5640}, {"loss": 1.3852, "grad_norm": 0.8422448039054871, "learning_rate": 0.0002, "epoch": 4.2211430706014195, "step": 5650}, {"loss": 1.3321, "grad_norm": 0.7017164826393127, "learning_rate": 0.0002, "epoch": 4.2286141202839, "step": 5660}, {"loss": 1.4105, "grad_norm": 0.8559677600860596, "learning_rate": 0.0002, "epoch": 4.236085169966381, "step": 5670}, {"loss": 1.3701, "grad_norm": 0.8216157555580139, "learning_rate": 0.0002, "epoch": 4.243556219648861, "step": 5680}, {"loss": 1.3565, "grad_norm": 0.7681755423545837, "learning_rate": 0.0002, "epoch": 4.251027269331341, "step": 5690}, {"loss": 1.3806, "grad_norm": 0.811665952205658, "learning_rate": 0.0002, "epoch": 4.258498319013821, "step": 5700}, {"loss": 1.4161, "grad_norm": 0.7242204546928406, "learning_rate": 0.0002, "epoch": 4.265969368696302, "step": 5710}, {"loss": 1.2958, "grad_norm": 0.7570181488990784, "learning_rate": 0.0002, "epoch": 4.273440418378782, "step": 5720}, {"loss": 1.4265, "grad_norm": 0.8951969146728516, "learning_rate": 0.0002, "epoch": 4.280911468061262, "step": 5730}, {"loss": 1.3895, "grad_norm": 0.7222902178764343, "learning_rate": 0.0002, "epoch": 4.288382517743743, "step": 5740}, {"loss": 1.4155, "grad_norm": 0.8508469462394714, "learning_rate": 0.0002, "epoch": 4.2958535674262235, "step": 5750}, {"loss": 1.365, "grad_norm": 0.7215430736541748, "learning_rate": 0.0002, "epoch": 4.303324617108704, "step": 5760}, {"loss": 1.4472, "grad_norm": 0.8774884939193726, "learning_rate": 0.0002, "epoch": 4.310795666791184, "step": 5770}, {"loss": 1.427, "grad_norm": 0.8354552984237671, "learning_rate": 0.0002, "epoch": 4.318266716473665, "step": 5780}, {"loss": 1.3222, "grad_norm": 0.6938814520835876, "learning_rate": 0.0002, "epoch": 4.325737766156145, "step": 5790}, {"loss": 1.3589, "grad_norm": 0.78675377368927, "learning_rate": 0.0002, "epoch": 4.333208815838625, "step": 5800}, {"loss": 1.3662, "grad_norm": 0.7147697806358337, "learning_rate": 0.0002, "epoch": 4.340679865521106, "step": 5810}, {"loss": 1.3597, "grad_norm": 0.7693623304367065, "learning_rate": 0.0002, "epoch": 4.348150915203586, "step": 5820}, {"loss": 1.2944, "grad_norm": 0.856517493724823, "learning_rate": 0.0002, "epoch": 4.355621964886066, "step": 5830}, {"loss": 1.4307, "grad_norm": 0.7200973033905029, "learning_rate": 0.0002, "epoch": 4.3630930145685465, "step": 5840}, {"loss": 1.442, "grad_norm": 0.743281364440918, "learning_rate": 0.0002, "epoch": 4.3705640642510275, "step": 5850}, {"loss": 1.3999, "grad_norm": 0.7627727389335632, "learning_rate": 0.0002, "epoch": 4.378035113933508, "step": 5860}, {"loss": 1.4082, "grad_norm": 0.7238836884498596, "learning_rate": 0.0002, "epoch": 4.385506163615988, "step": 5870}, {"loss": 1.4292, "grad_norm": 0.7253410816192627, "learning_rate": 0.0002, "epoch": 4.392977213298469, "step": 5880}, {"loss": 1.3774, "grad_norm": 0.8232238292694092, "learning_rate": 0.0002, "epoch": 4.400448262980949, "step": 5890}, {"loss": 1.3757, "grad_norm": 0.8778504729270935, "learning_rate": 0.0002, "epoch": 4.407919312663429, "step": 5900}, {"loss": 1.387, "grad_norm": 0.7639474868774414, "learning_rate": 0.0002, "epoch": 4.415390362345909, "step": 5910}, {"loss": 1.3862, "grad_norm": 0.7666519284248352, "learning_rate": 0.0002, "epoch": 4.42286141202839, "step": 5920}, {"loss": 1.4168, "grad_norm": 0.867132842540741, "learning_rate": 0.0002, "epoch": 4.43033246171087, "step": 5930}, {"loss": 1.4772, "grad_norm": 0.7571166753768921, "learning_rate": 0.0002, "epoch": 4.4378035113933505, "step": 5940}, {"loss": 1.4401, "grad_norm": 0.7911370992660522, "learning_rate": 0.0002, "epoch": 4.4452745610758315, "step": 5950}, {"loss": 1.4516, "grad_norm": 0.8844250440597534, "learning_rate": 0.0002, "epoch": 4.452745610758312, "step": 5960}, {"loss": 1.4109, "grad_norm": 0.7336231470108032, "learning_rate": 0.0002, "epoch": 4.460216660440792, "step": 5970}, {"loss": 1.3891, "grad_norm": 0.8162738084793091, "learning_rate": 0.0002, "epoch": 4.467687710123272, "step": 5980}, {"loss": 1.393, "grad_norm": 0.7413017153739929, "learning_rate": 0.0002, "epoch": 4.475158759805753, "step": 5990}, {"loss": 1.3712, "grad_norm": 0.7215432524681091, "learning_rate": 0.0002, "epoch": 4.482629809488233, "step": 6000}, {"loss": 1.3521, "grad_norm": 0.8943389058113098, "learning_rate": 0.0002, "epoch": 4.490100859170713, "step": 6010}, {"loss": 1.4172, "grad_norm": 0.7850823998451233, "learning_rate": 0.0002, "epoch": 4.497571908853194, "step": 6020}, {"loss": 1.3582, "grad_norm": 0.8117504119873047, "learning_rate": 0.0002, "epoch": 4.505042958535674, "step": 6030}, {"loss": 1.4272, "grad_norm": 0.8381605744361877, "learning_rate": 0.0002, "epoch": 4.5125140082181545, "step": 6040}, {"loss": 1.3829, "grad_norm": 0.7964059710502625, "learning_rate": 0.0002, "epoch": 4.519985057900635, "step": 6050}, {"loss": 1.3555, "grad_norm": 0.7935128211975098, "learning_rate": 0.0002, "epoch": 4.527456107583116, "step": 6060}, {"loss": 1.3994, "grad_norm": 0.8725124597549438, "learning_rate": 0.0002, "epoch": 4.534927157265596, "step": 6070}, {"loss": 1.3923, "grad_norm": 0.880325198173523, "learning_rate": 0.0002, "epoch": 4.542398206948076, "step": 6080}, {"loss": 1.4459, "grad_norm": 0.7220637202262878, "learning_rate": 0.0002, "epoch": 4.549869256630557, "step": 6090}, {"loss": 1.3281, "grad_norm": 0.6908547878265381, "learning_rate": 0.0002, "epoch": 4.557340306313037, "step": 6100}, {"loss": 1.437, "grad_norm": 0.797931969165802, "learning_rate": 0.0002, "epoch": 4.564811355995517, "step": 6110}, {"loss": 1.4023, "grad_norm": 0.7056134343147278, "learning_rate": 0.0002, "epoch": 4.572282405677997, "step": 6120}, {"loss": 1.3814, "grad_norm": 0.7850478887557983, "learning_rate": 0.0002, "epoch": 4.579753455360478, "step": 6130}, {"loss": 1.3579, "grad_norm": 0.8112621307373047, "learning_rate": 0.0002, "epoch": 4.5872245050429585, "step": 6140}, {"loss": 1.3523, "grad_norm": 0.7040849328041077, "learning_rate": 0.0002, "epoch": 4.594695554725439, "step": 6150}, {"loss": 1.3526, "grad_norm": 0.7214553952217102, "learning_rate": 0.0002, "epoch": 4.60216660440792, "step": 6160}, {"loss": 1.3932, "grad_norm": 0.8616511821746826, "learning_rate": 0.0002, "epoch": 4.6096376540904, "step": 6170}, {"loss": 1.4622, "grad_norm": 0.8374658226966858, "learning_rate": 0.0002, "epoch": 4.61710870377288, "step": 6180}, {"loss": 1.3703, "grad_norm": 0.6761606931686401, "learning_rate": 0.0002, "epoch": 4.62457975345536, "step": 6190}, {"loss": 1.3977, "grad_norm": 0.768028199672699, "learning_rate": 0.0002, "epoch": 4.632050803137841, "step": 6200}, {"loss": 1.3772, "grad_norm": 0.9372717142105103, "learning_rate": 0.0002, "epoch": 4.639521852820321, "step": 6210}, {"loss": 1.4098, "grad_norm": 0.7906546592712402, "learning_rate": 0.0002, "epoch": 4.646992902502801, "step": 6220}, {"loss": 1.3962, "grad_norm": 0.7376723289489746, "learning_rate": 0.0002, "epoch": 4.654463952185282, "step": 6230}, {"loss": 1.4529, "grad_norm": 0.8972630500793457, "learning_rate": 0.0002, "epoch": 4.6619350018677626, "step": 6240}, {"loss": 1.4668, "grad_norm": 0.8261756300926208, "learning_rate": 0.0002, "epoch": 4.669406051550243, "step": 6250}, {"loss": 1.3267, "grad_norm": 0.7512393593788147, "learning_rate": 0.0002, "epoch": 4.676877101232723, "step": 6260}, {"loss": 1.4278, "grad_norm": 0.7132362127304077, "learning_rate": 0.0002, "epoch": 4.684348150915204, "step": 6270}, {"loss": 1.4299, "grad_norm": 0.7690575122833252, "learning_rate": 0.0002, "epoch": 4.691819200597684, "step": 6280}, {"loss": 1.4769, "grad_norm": 0.9886258840560913, "learning_rate": 0.0002, "epoch": 4.699290250280164, "step": 6290}, {"loss": 1.4005, "grad_norm": 0.9502435922622681, "learning_rate": 0.0002, "epoch": 4.706761299962645, "step": 6300}, {"loss": 1.4319, "grad_norm": 0.702255129814148, "learning_rate": 0.0002, "epoch": 4.714232349645125, "step": 6310}, {"loss": 1.4447, "grad_norm": 0.7713103890419006, "learning_rate": 0.0002, "epoch": 4.721703399327605, "step": 6320}, {"loss": 1.4392, "grad_norm": 0.7778580784797668, "learning_rate": 0.0002, "epoch": 4.7291744490100855, "step": 6330}, {"loss": 1.4169, "grad_norm": 0.7275111079216003, "learning_rate": 0.0002, "epoch": 4.736645498692567, "step": 6340}, {"loss": 1.4429, "grad_norm": 0.7728744149208069, "learning_rate": 0.0002, "epoch": 4.744116548375047, "step": 6350}, {"loss": 1.3756, "grad_norm": 0.9724260568618774, "learning_rate": 0.0002, "epoch": 4.751587598057527, "step": 6360}, {"loss": 1.3358, "grad_norm": 0.7505622506141663, "learning_rate": 0.0002, "epoch": 4.759058647740007, "step": 6370}, {"loss": 1.379, "grad_norm": 0.7994682788848877, "learning_rate": 0.0002, "epoch": 4.766529697422488, "step": 6380}, {"loss": 1.4275, "grad_norm": 0.8432038426399231, "learning_rate": 0.0002, "epoch": 4.774000747104968, "step": 6390}, {"loss": 1.4606, "grad_norm": 0.7436022758483887, "learning_rate": 0.0002, "epoch": 4.781471796787448, "step": 6400}, {"loss": 1.3461, "grad_norm": 0.7709194421768188, "learning_rate": 0.0002, "epoch": 4.788942846469929, "step": 6410}, {"loss": 1.3715, "grad_norm": 0.8798436522483826, "learning_rate": 0.0002, "epoch": 4.796413896152409, "step": 6420}, {"loss": 1.3761, "grad_norm": 0.790189266204834, "learning_rate": 0.0002, "epoch": 4.80388494583489, "step": 6430}, {"loss": 1.4109, "grad_norm": 0.6824303865432739, "learning_rate": 0.0002, "epoch": 4.811355995517371, "step": 6440}, {"loss": 1.3877, "grad_norm": 0.7501044869422913, "learning_rate": 0.0002, "epoch": 4.818827045199851, "step": 6450}, {"loss": 1.4458, "grad_norm": 0.8840398192405701, "learning_rate": 0.0002, "epoch": 4.826298094882331, "step": 6460}, {"loss": 1.4412, "grad_norm": 0.7812688946723938, "learning_rate": 0.0002, "epoch": 4.833769144564811, "step": 6470}, {"loss": 1.4299, "grad_norm": 0.7429926991462708, "learning_rate": 0.0002, "epoch": 4.841240194247292, "step": 6480}, {"loss": 1.5062, "grad_norm": 0.7778021693229675, "learning_rate": 0.0002, "epoch": 4.848711243929772, "step": 6490}, {"loss": 1.4589, "grad_norm": 0.8270702362060547, "learning_rate": 0.0002, "epoch": 4.856182293612252, "step": 6500}, {"loss": 1.4091, "grad_norm": 0.6960513591766357, "learning_rate": 0.0002, "epoch": 4.863653343294732, "step": 6510}, {"loss": 1.376, "grad_norm": 0.7728942632675171, "learning_rate": 0.0002, "epoch": 4.8711243929772134, "step": 6520}, {"loss": 1.4852, "grad_norm": 0.7377303838729858, "learning_rate": 0.0002, "epoch": 4.878595442659694, "step": 6530}, {"loss": 1.3846, "grad_norm": 0.7257253527641296, "learning_rate": 0.0002, "epoch": 4.886066492342174, "step": 6540}, {"loss": 1.4166, "grad_norm": 0.7875821590423584, "learning_rate": 0.0002, "epoch": 4.893537542024655, "step": 6550}, {"loss": 1.357, "grad_norm": 0.8346304297447205, "learning_rate": 0.0002, "epoch": 4.901008591707135, "step": 6560}, {"loss": 1.4522, "grad_norm": 0.7710739374160767, "learning_rate": 0.0002, "epoch": 4.908479641389615, "step": 6570}, {"loss": 1.4465, "grad_norm": 0.7015138268470764, "learning_rate": 0.0002, "epoch": 4.915950691072096, "step": 6580}, {"loss": 1.435, "grad_norm": 0.8707432150840759, "learning_rate": 0.0002, "epoch": 4.923421740754576, "step": 6590}, {"loss": 1.2968, "grad_norm": 0.786601185798645, "learning_rate": 0.0002, "epoch": 4.930892790437056, "step": 6600}, {"loss": 1.4385, "grad_norm": 0.978519082069397, "learning_rate": 0.0002, "epoch": 4.938363840119536, "step": 6610}, {"loss": 1.3997, "grad_norm": 0.8102927207946777, "learning_rate": 0.0002, "epoch": 4.9458348898020175, "step": 6620}, {"loss": 1.4859, "grad_norm": 0.7628704309463501, "learning_rate": 0.0002, "epoch": 4.953305939484498, "step": 6630}, {"loss": 1.3774, "grad_norm": 0.8053455352783203, "learning_rate": 0.0002, "epoch": 4.960776989166978, "step": 6640}, {"loss": 1.5092, "grad_norm": 0.8680412173271179, "learning_rate": 0.0002, "epoch": 4.968248038849458, "step": 6650}, {"loss": 1.3978, "grad_norm": 0.7415758371353149, "learning_rate": 0.0002, "epoch": 4.975719088531939, "step": 6660}, {"loss": 1.3793, "grad_norm": 0.7730312347412109, "learning_rate": 0.0002, "epoch": 4.983190138214419, "step": 6670}, {"loss": 1.4863, "grad_norm": 0.7924041152000427, "learning_rate": 0.0002, "epoch": 4.990661187896899, "step": 6680}, {"loss": 1.4137, "grad_norm": 0.8677893877029419, "learning_rate": 0.0002, "epoch": 4.99813223757938, "step": 6690}, {"eval_loss": 1.9444633722305298, "eval_runtime": 39.3488, "eval_samples_per_second": 13.088, "eval_steps_per_second": 1.652, "epoch": 4.999626447515876, "step": 6692}, {"loss": 1.3076, "grad_norm": 0.7102245092391968, "learning_rate": 0.0002, "epoch": 5.00560328726186, "step": 6700}, {"loss": 1.2714, "grad_norm": 1.0425463914871216, "learning_rate": 0.0002, "epoch": 5.0130743369443405, "step": 6710}, {"loss": 1.181, "grad_norm": 0.9320756793022156, "learning_rate": 0.0002, "epoch": 5.0205453866268215, "step": 6720}, {"loss": 1.1786, "grad_norm": 0.8797217607498169, "learning_rate": 0.0002, "epoch": 5.028016436309302, "step": 6730}, {"loss": 1.2097, "grad_norm": 2.135707139968872, "learning_rate": 0.0002, "epoch": 5.035487485991782, "step": 6740}, {"loss": 1.1761, "grad_norm": 0.8747734427452087, "learning_rate": 0.0002, "epoch": 5.042958535674262, "step": 6750}, {"loss": 1.1675, "grad_norm": 0.9981076717376709, "learning_rate": 0.0002, "epoch": 5.050429585356743, "step": 6760}, {"loss": 1.1976, "grad_norm": 0.985078752040863, "learning_rate": 0.0002, "epoch": 5.057900635039223, "step": 6770}, {"loss": 1.2688, "grad_norm": 1.0974019765853882, "learning_rate": 0.0002, "epoch": 5.065371684721703, "step": 6780}, {"loss": 1.1982, "grad_norm": 0.9823219180107117, "learning_rate": 0.0002, "epoch": 5.072842734404184, "step": 6790}, {"loss": 1.2586, "grad_norm": 1.122605562210083, "learning_rate": 0.0002, "epoch": 5.080313784086664, "step": 6800}, {"loss": 1.2069, "grad_norm": 0.8556802272796631, "learning_rate": 0.0002, "epoch": 5.0877848337691445, "step": 6810}, {"loss": 1.1908, "grad_norm": 1.1699262857437134, "learning_rate": 0.0002, "epoch": 5.095255883451625, "step": 6820}, {"loss": 1.1869, "grad_norm": 1.0440590381622314, "learning_rate": 0.0002, "epoch": 5.102726933134106, "step": 6830}, {"loss": 1.1655, "grad_norm": 1.0445300340652466, "learning_rate": 0.0002, "epoch": 5.110197982816586, "step": 6840}, {"loss": 1.2392, "grad_norm": 0.8289563059806824, "learning_rate": 0.0002, "epoch": 5.117669032499066, "step": 6850}, {"loss": 1.1687, "grad_norm": 1.1051193475723267, "learning_rate": 0.0002, "epoch": 5.125140082181547, "step": 6860}, {"loss": 1.2737, "grad_norm": 0.9345614910125732, "learning_rate": 0.0002, "epoch": 5.132611131864027, "step": 6870}, {"loss": 1.3021, "grad_norm": 1.1222996711730957, "learning_rate": 0.0002, "epoch": 5.140082181546507, "step": 6880}, {"loss": 1.2408, "grad_norm": 0.9405338764190674, "learning_rate": 0.0002, "epoch": 5.147553231228987, "step": 6890}, {"loss": 1.2367, "grad_norm": 1.0935171842575073, "learning_rate": 0.0002, "epoch": 5.155024280911468, "step": 6900}, {"loss": 1.2458, "grad_norm": 1.0438612699508667, "learning_rate": 0.0002, "epoch": 5.1624953305939485, "step": 6910}, {"loss": 1.2562, "grad_norm": 1.1189004182815552, "learning_rate": 0.0002, "epoch": 5.169966380276429, "step": 6920}, {"loss": 1.25, "grad_norm": 1.0533215999603271, "learning_rate": 0.0002, "epoch": 5.17743742995891, "step": 6930}, {"loss": 1.2974, "grad_norm": 0.9779648780822754, "learning_rate": 0.0002, "epoch": 5.18490847964139, "step": 6940}, {"loss": 1.1965, "grad_norm": 0.8920868635177612, "learning_rate": 0.0002, "epoch": 5.19237952932387, "step": 6950}, {"loss": 1.283, "grad_norm": 0.8374548554420471, "learning_rate": 0.0002, "epoch": 5.19985057900635, "step": 6960}, {"loss": 1.2775, "grad_norm": 1.0490682125091553, "learning_rate": 0.0002, "epoch": 5.207321628688831, "step": 6970}, {"loss": 1.1826, "grad_norm": 0.9658287167549133, "learning_rate": 0.0002, "epoch": 5.214792678371311, "step": 6980}, {"loss": 1.2647, "grad_norm": 0.9652056097984314, "learning_rate": 0.0002, "epoch": 5.222263728053791, "step": 6990}, {"loss": 1.3023, "grad_norm": 0.9141794443130493, "learning_rate": 0.0002, "epoch": 5.229734777736272, "step": 7000}, {"loss": 1.2456, "grad_norm": 0.9831376671791077, "learning_rate": 0.0002, "epoch": 5.2372058274187525, "step": 7010}, {"loss": 1.2176, "grad_norm": 1.0198718309402466, "learning_rate": 0.0002, "epoch": 5.244676877101233, "step": 7020}, {"loss": 1.2643, "grad_norm": 0.9647888541221619, "learning_rate": 0.0002, "epoch": 5.252147926783713, "step": 7030}, {"loss": 1.2106, "grad_norm": 1.3941649198532104, "learning_rate": 0.0002, "epoch": 5.259618976466194, "step": 7040}, {"loss": 1.2885, "grad_norm": 1.0305466651916504, "learning_rate": 0.0002, "epoch": 5.267090026148674, "step": 7050}, {"loss": 1.2362, "grad_norm": 0.9577859044075012, "learning_rate": 0.0002, "epoch": 5.274561075831154, "step": 7060}, {"loss": 1.2231, "grad_norm": 1.149092197418213, "learning_rate": 0.0002, "epoch": 5.282032125513634, "step": 7070}, {"loss": 1.2986, "grad_norm": 1.2582733631134033, "learning_rate": 0.0002, "epoch": 5.289503175196115, "step": 7080}, {"loss": 1.2307, "grad_norm": 1.1777442693710327, "learning_rate": 0.0002, "epoch": 5.296974224878595, "step": 7090}, {"loss": 1.24, "grad_norm": 1.0076404809951782, "learning_rate": 0.0002, "epoch": 5.3044452745610755, "step": 7100}, {"loss": 1.1407, "grad_norm": 0.9037365913391113, "learning_rate": 0.0002, "epoch": 5.3119163242435565, "step": 7110}, {"loss": 1.238, "grad_norm": 0.9428724646568298, "learning_rate": 0.0002, "epoch": 5.319387373926037, "step": 7120}, {"loss": 1.2571, "grad_norm": 0.9935154318809509, "learning_rate": 0.0002, "epoch": 5.326858423608517, "step": 7130}, {"loss": 1.2833, "grad_norm": 1.087500810623169, "learning_rate": 0.0002, "epoch": 5.334329473290998, "step": 7140}, {"loss": 1.2304, "grad_norm": 0.8543072938919067, "learning_rate": 0.0002, "epoch": 5.341800522973478, "step": 7150}, {"loss": 1.2755, "grad_norm": 0.9323700070381165, "learning_rate": 0.0002, "epoch": 5.349271572655958, "step": 7160}, {"loss": 1.2769, "grad_norm": 1.0037827491760254, "learning_rate": 0.0002, "epoch": 5.356742622338438, "step": 7170}, {"loss": 1.3204, "grad_norm": 0.8746469616889954, "learning_rate": 0.0002, "epoch": 5.364213672020919, "step": 7180}, {"loss": 1.2759, "grad_norm": 0.9516328573226929, "learning_rate": 0.0002, "epoch": 5.371684721703399, "step": 7190}, {"loss": 1.2428, "grad_norm": 0.9395177364349365, "learning_rate": 0.0002, "epoch": 5.3791557713858795, "step": 7200}, {"loss": 1.3214, "grad_norm": 1.000369906425476, "learning_rate": 0.0002, "epoch": 5.38662682106836, "step": 7210}, {"loss": 1.2337, "grad_norm": 1.0845502614974976, "learning_rate": 0.0002, "epoch": 5.394097870750841, "step": 7220}, {"loss": 1.2776, "grad_norm": 0.8975145220756531, "learning_rate": 0.0002, "epoch": 5.401568920433321, "step": 7230}, {"loss": 1.2306, "grad_norm": 1.040077805519104, "learning_rate": 0.0002, "epoch": 5.409039970115801, "step": 7240}, {"loss": 1.2277, "grad_norm": 1.0729942321777344, "learning_rate": 0.0002, "epoch": 5.416511019798282, "step": 7250}, {"loss": 1.2714, "grad_norm": 0.8322232961654663, "learning_rate": 0.0002, "epoch": 5.423982069480762, "step": 7260}, {"loss": 1.3036, "grad_norm": 1.0654641389846802, "learning_rate": 0.0002, "epoch": 5.431453119163242, "step": 7270}, {"loss": 1.268, "grad_norm": 1.0445852279663086, "learning_rate": 0.0002, "epoch": 5.438924168845723, "step": 7280}, {"loss": 1.2743, "grad_norm": 1.0762956142425537, "learning_rate": 0.0002, "epoch": 5.446395218528203, "step": 7290}, {"loss": 1.2887, "grad_norm": 0.9721953868865967, "learning_rate": 0.0002, "epoch": 5.4538662682106835, "step": 7300}, {"loss": 1.2833, "grad_norm": 0.9238539338111877, "learning_rate": 0.0002, "epoch": 5.461337317893164, "step": 7310}, {"loss": 1.255, "grad_norm": 0.9912874102592468, "learning_rate": 0.0002, "epoch": 5.468808367575645, "step": 7320}, {"loss": 1.2557, "grad_norm": 1.0727077722549438, "learning_rate": 0.0002, "epoch": 5.476279417258125, "step": 7330}, {"loss": 1.3471, "grad_norm": 0.8633865118026733, "learning_rate": 0.0002, "epoch": 5.483750466940605, "step": 7340}, {"loss": 1.3155, "grad_norm": 0.9396262764930725, "learning_rate": 0.0002, "epoch": 5.491221516623085, "step": 7350}, {"loss": 1.3146, "grad_norm": 1.0253715515136719, "learning_rate": 0.0002, "epoch": 5.498692566305566, "step": 7360}, {"loss": 1.3156, "grad_norm": 1.006047010421753, "learning_rate": 0.0002, "epoch": 5.506163615988046, "step": 7370}, {"loss": 1.3107, "grad_norm": 0.9781233072280884, "learning_rate": 0.0002, "epoch": 5.513634665670526, "step": 7380}, {"loss": 1.2703, "grad_norm": 0.9945126175880432, "learning_rate": 0.0002, "epoch": 5.521105715353007, "step": 7390}, {"loss": 1.1936, "grad_norm": 0.9081175327301025, "learning_rate": 0.0002, "epoch": 5.528576765035488, "step": 7400}, {"loss": 1.2651, "grad_norm": 1.2215938568115234, "learning_rate": 0.0002, "epoch": 5.536047814717968, "step": 7410}, {"loss": 1.2484, "grad_norm": 1.0724077224731445, "learning_rate": 0.0002, "epoch": 5.543518864400449, "step": 7420}, {"loss": 1.3083, "grad_norm": 1.106955885887146, "learning_rate": 0.0002, "epoch": 5.550989914082929, "step": 7430}, {"loss": 1.2125, "grad_norm": 1.0657650232315063, "learning_rate": 0.0002, "epoch": 5.558460963765409, "step": 7440}, {"loss": 1.2576, "grad_norm": 0.9725455641746521, "learning_rate": 0.0002, "epoch": 5.565932013447889, "step": 7450}, {"loss": 1.3297, "grad_norm": 0.8604224324226379, "learning_rate": 0.0002, "epoch": 5.57340306313037, "step": 7460}, {"loss": 1.3084, "grad_norm": 0.9913371205329895, "learning_rate": 0.0002, "epoch": 5.58087411281285, "step": 7470}, {"loss": 1.3371, "grad_norm": 1.012073040008545, "learning_rate": 0.0002, "epoch": 5.58834516249533, "step": 7480}, {"loss": 1.2526, "grad_norm": 1.1003159284591675, "learning_rate": 0.0002, "epoch": 5.5958162121778106, "step": 7490}, {"loss": 1.2577, "grad_norm": 0.9104593992233276, "learning_rate": 0.0002, "epoch": 5.603287261860292, "step": 7500}, {"loss": 1.2578, "grad_norm": 0.9480831623077393, "learning_rate": 0.0002, "epoch": 5.610758311542772, "step": 7510}, {"loss": 1.3056, "grad_norm": 1.0826456546783447, "learning_rate": 0.0002, "epoch": 5.618229361225252, "step": 7520}, {"loss": 1.2931, "grad_norm": 0.8286259174346924, "learning_rate": 0.0002, "epoch": 5.625700410907733, "step": 7530}, {"loss": 1.2918, "grad_norm": 0.9145061373710632, "learning_rate": 0.0002, "epoch": 5.633171460590213, "step": 7540}, {"loss": 1.1736, "grad_norm": 0.9363601803779602, "learning_rate": 0.0002, "epoch": 5.640642510272693, "step": 7550}, {"loss": 1.2265, "grad_norm": 0.9553244709968567, "learning_rate": 0.0002, "epoch": 5.648113559955174, "step": 7560}, {"loss": 1.2356, "grad_norm": 1.0343557596206665, "learning_rate": 0.0002, "epoch": 5.655584609637654, "step": 7570}, {"loss": 1.3171, "grad_norm": 0.8734238743782043, "learning_rate": 0.0002, "epoch": 5.663055659320134, "step": 7580}, {"loss": 1.2785, "grad_norm": 1.0230586528778076, "learning_rate": 0.0002, "epoch": 5.670526709002615, "step": 7590}, {"loss": 1.2936, "grad_norm": 1.0063409805297852, "learning_rate": 0.0002, "epoch": 5.677997758685096, "step": 7600}, {"loss": 1.2396, "grad_norm": 1.0104626417160034, "learning_rate": 0.0002, "epoch": 5.685468808367576, "step": 7610}, {"loss": 1.2581, "grad_norm": 0.9528168439865112, "learning_rate": 0.0002, "epoch": 5.692939858050056, "step": 7620}, {"loss": 1.3116, "grad_norm": 0.9799878597259521, "learning_rate": 0.0002, "epoch": 5.700410907732536, "step": 7630}, {"loss": 1.2632, "grad_norm": 0.969351589679718, "learning_rate": 0.0002, "epoch": 5.707881957415017, "step": 7640}, {"loss": 1.3055, "grad_norm": 1.3037652969360352, "learning_rate": 0.0002, "epoch": 5.715353007097497, "step": 7650}, {"loss": 1.3126, "grad_norm": 1.0640486478805542, "learning_rate": 0.0002, "epoch": 5.722824056779977, "step": 7660}, {"loss": 1.3325, "grad_norm": 1.0416420698165894, "learning_rate": 0.0002, "epoch": 5.730295106462458, "step": 7670}, {"loss": 1.25, "grad_norm": 0.8893619775772095, "learning_rate": 0.0002, "epoch": 5.7377661561449385, "step": 7680}, {"loss": 1.319, "grad_norm": 0.8512844443321228, "learning_rate": 0.0002, "epoch": 5.745237205827419, "step": 7690}, {"loss": 1.3328, "grad_norm": 0.9955748319625854, "learning_rate": 0.0002, "epoch": 5.7527082555099, "step": 7700}, {"loss": 1.294, "grad_norm": 1.0409910678863525, "learning_rate": 0.0002, "epoch": 5.76017930519238, "step": 7710}, {"loss": 1.3518, "grad_norm": 1.010097861289978, "learning_rate": 0.0002, "epoch": 5.76765035487486, "step": 7720}, {"loss": 1.2106, "grad_norm": 0.8974892497062683, "learning_rate": 0.0002, "epoch": 5.77512140455734, "step": 7730}, {"loss": 1.2743, "grad_norm": 0.972835123538971, "learning_rate": 0.0002, "epoch": 5.782592454239821, "step": 7740}, {"loss": 1.3549, "grad_norm": 0.9607440829277039, "learning_rate": 0.0002, "epoch": 5.790063503922301, "step": 7750}, {"loss": 1.29, "grad_norm": 0.9426500797271729, "learning_rate": 0.0002, "epoch": 5.797534553604781, "step": 7760}, {"loss": 1.274, "grad_norm": 0.8745320439338684, "learning_rate": 0.0002, "epoch": 5.8050056032872615, "step": 7770}, {"loss": 1.3009, "grad_norm": 1.0117204189300537, "learning_rate": 0.0002, "epoch": 5.8124766529697425, "step": 7780}, {"loss": 1.3135, "grad_norm": 1.0387755632400513, "learning_rate": 0.0002, "epoch": 5.819947702652223, "step": 7790}, {"loss": 1.2709, "grad_norm": 1.0709784030914307, "learning_rate": 0.0002, "epoch": 5.827418752334703, "step": 7800}, {"loss": 1.225, "grad_norm": 0.9512667655944824, "learning_rate": 0.0002, "epoch": 5.834889802017184, "step": 7810}, {"loss": 1.3284, "grad_norm": 1.021094560623169, "learning_rate": 0.0002, "epoch": 5.842360851699664, "step": 7820}, {"loss": 1.2794, "grad_norm": 1.117491364479065, "learning_rate": 0.0002, "epoch": 5.849831901382144, "step": 7830}, {"loss": 1.3646, "grad_norm": 0.9252554178237915, "learning_rate": 0.0002, "epoch": 5.857302951064625, "step": 7840}, {"loss": 1.2976, "grad_norm": 1.1416207551956177, "learning_rate": 0.0002, "epoch": 5.864774000747105, "step": 7850}, {"loss": 1.3293, "grad_norm": 1.1219907999038696, "learning_rate": 0.0002, "epoch": 5.872245050429585, "step": 7860}, {"loss": 1.2334, "grad_norm": 0.8300467729568481, "learning_rate": 0.0002, "epoch": 5.8797161001120655, "step": 7870}, {"loss": 1.3132, "grad_norm": 1.00551438331604, "learning_rate": 0.0002, "epoch": 5.8871871497945465, "step": 7880}, {"loss": 1.2609, "grad_norm": 0.8981153964996338, "learning_rate": 0.0002, "epoch": 5.894658199477027, "step": 7890}, {"loss": 1.2817, "grad_norm": 1.0247976779937744, "learning_rate": 0.0002, "epoch": 5.902129249159507, "step": 7900}, {"loss": 1.2866, "grad_norm": 1.0820319652557373, "learning_rate": 0.0002, "epoch": 5.909600298841987, "step": 7910}, {"loss": 1.2941, "grad_norm": 0.952675461769104, "learning_rate": 0.0002, "epoch": 5.917071348524468, "step": 7920}, {"loss": 1.307, "grad_norm": 0.8666740655899048, "learning_rate": 0.0002, "epoch": 5.924542398206948, "step": 7930}, {"loss": 1.2752, "grad_norm": 0.8640421032905579, "learning_rate": 0.0002, "epoch": 5.932013447889428, "step": 7940}, {"loss": 1.2386, "grad_norm": 1.2343276739120483, "learning_rate": 0.0002, "epoch": 5.939484497571909, "step": 7950}, {"loss": 1.2333, "grad_norm": 0.958046555519104, "learning_rate": 0.0002, "epoch": 5.946955547254389, "step": 7960}, {"loss": 1.2352, "grad_norm": 1.0538510084152222, "learning_rate": 0.0002, "epoch": 5.9544265969368695, "step": 7970}, {"loss": 1.3233, "grad_norm": 1.2681571245193481, "learning_rate": 0.0002, "epoch": 5.9618976466193505, "step": 7980}, {"loss": 1.2514, "grad_norm": 0.8171183466911316, "learning_rate": 0.0002, "epoch": 5.969368696301831, "step": 7990}, {"loss": 1.3412, "grad_norm": 0.9109523892402649, "learning_rate": 0.0002, "epoch": 5.976839745984311, "step": 8000}, {"loss": 1.3497, "grad_norm": 1.0040639638900757, "learning_rate": 0.0002, "epoch": 5.984310795666791, "step": 8010}, {"loss": 1.3299, "grad_norm": 0.9596554040908813, "learning_rate": 0.0002, "epoch": 5.991781845349272, "step": 8020}, {"loss": 1.3109, "grad_norm": 0.9782963991165161, "learning_rate": 0.0002, "epoch": 5.999252895031752, "step": 8030}, {"eval_loss": 2.0417845249176025, "eval_runtime": 38.8465, "eval_samples_per_second": 13.257, "eval_steps_per_second": 1.673, "epoch": 6.0, "step": 8031}, {"loss": 1.0886, "grad_norm": 1.380823016166687, "learning_rate": 0.0002, "epoch": 6.006723944714232, "step": 8040}, {"loss": 1.0413, "grad_norm": 1.067636251449585, "learning_rate": 0.0002, "epoch": 6.014194994396712, "step": 8050}, {"loss": 1.0686, "grad_norm": 1.363402009010315, "learning_rate": 0.0002, "epoch": 6.021666044079193, "step": 8060}, {"loss": 1.0762, "grad_norm": 0.9901054501533508, "learning_rate": 0.0002, "epoch": 6.0291370937616735, "step": 8070}, {"loss": 1.1182, "grad_norm": 1.1545379161834717, "learning_rate": 0.0002, "epoch": 6.036608143444154, "step": 8080}, {"loss": 1.0644, "grad_norm": 1.2259265184402466, "learning_rate": 0.0002, "epoch": 6.044079193126635, "step": 8090}, {"loss": 1.1273, "grad_norm": 1.1237425804138184, "learning_rate": 0.0002, "epoch": 6.051550242809115, "step": 8100}, {"loss": 1.1001, "grad_norm": 1.2805622816085815, "learning_rate": 0.0002, "epoch": 6.059021292491595, "step": 8110}, {"loss": 1.0731, "grad_norm": 1.2270452976226807, "learning_rate": 0.0002, "epoch": 6.066492342174075, "step": 8120}, {"loss": 1.0692, "grad_norm": 1.1924101114273071, "learning_rate": 0.0002, "epoch": 6.073963391856556, "step": 8130}, {"loss": 1.1698, "grad_norm": 1.2543894052505493, "learning_rate": 0.0002, "epoch": 6.081434441539036, "step": 8140}, {"loss": 1.069, "grad_norm": 1.1821149587631226, "learning_rate": 0.0002, "epoch": 6.088905491221516, "step": 8150}, {"loss": 1.109, "grad_norm": 1.2202836275100708, "learning_rate": 0.0002, "epoch": 6.096376540903997, "step": 8160}, {"loss": 1.136, "grad_norm": 1.0576019287109375, "learning_rate": 0.0002, "epoch": 6.1038475905864775, "step": 8170}, {"loss": 1.1395, "grad_norm": 1.31708824634552, "learning_rate": 0.0002, "epoch": 6.111318640268958, "step": 8180}, {"loss": 1.0887, "grad_norm": 1.0479495525360107, "learning_rate": 0.0002, "epoch": 6.118789689951438, "step": 8190}, {"loss": 1.0764, "grad_norm": 1.285003423690796, "learning_rate": 0.0002, "epoch": 6.126260739633919, "step": 8200}, {"loss": 1.0642, "grad_norm": 1.0989165306091309, "learning_rate": 0.0002, "epoch": 6.133731789316399, "step": 8210}, {"loss": 1.0981, "grad_norm": 1.1659013032913208, "learning_rate": 0.0002, "epoch": 6.141202838998879, "step": 8220}, {"loss": 1.1138, "grad_norm": 1.2796376943588257, "learning_rate": 0.0002, "epoch": 6.14867388868136, "step": 8230}, {"loss": 1.1116, "grad_norm": 1.060564637184143, "learning_rate": 0.0002, "epoch": 6.15614493836384, "step": 8240}, {"loss": 1.1493, "grad_norm": 1.3884605169296265, "learning_rate": 0.0002, "epoch": 6.16361598804632, "step": 8250}, {"loss": 1.0504, "grad_norm": 1.1570569276809692, "learning_rate": 0.0002, "epoch": 6.1710870377288005, "step": 8260}, {"loss": 1.0386, "grad_norm": 1.4136502742767334, "learning_rate": 0.0002, "epoch": 6.1785580874112815, "step": 8270}, {"loss": 1.0882, "grad_norm": 1.3396095037460327, "learning_rate": 0.0002, "epoch": 6.186029137093762, "step": 8280}, {"loss": 1.133, "grad_norm": 1.2549997568130493, "learning_rate": 0.0002, "epoch": 6.193500186776242, "step": 8290}, {"loss": 1.0626, "grad_norm": 1.3629751205444336, "learning_rate": 0.0002, "epoch": 6.200971236458723, "step": 8300}, {"loss": 1.1343, "grad_norm": 1.1029163599014282, "learning_rate": 0.0002, "epoch": 6.208442286141203, "step": 8310}, {"loss": 1.0895, "grad_norm": 1.1992450952529907, "learning_rate": 0.0002, "epoch": 6.215913335823683, "step": 8320}, {"loss": 1.1417, "grad_norm": 1.3317986726760864, "learning_rate": 0.0002, "epoch": 6.223384385506163, "step": 8330}, {"loss": 1.0958, "grad_norm": 1.0538336038589478, "learning_rate": 0.0002, "epoch": 6.230855435188644, "step": 8340}, {"loss": 1.1557, "grad_norm": 1.1767704486846924, "learning_rate": 0.0002, "epoch": 6.238326484871124, "step": 8350}, {"loss": 1.1038, "grad_norm": 1.1213016510009766, "learning_rate": 0.0002, "epoch": 6.2457975345536045, "step": 8360}, {"loss": 1.1241, "grad_norm": 1.1895716190338135, "learning_rate": 0.0002, "epoch": 6.253268584236086, "step": 8370}, {"loss": 1.1171, "grad_norm": 1.1078153848648071, "learning_rate": 0.0002, "epoch": 6.260739633918566, "step": 8380}, {"loss": 1.1124, "grad_norm": 1.1662801504135132, "learning_rate": 0.0002, "epoch": 6.268210683601046, "step": 8390}, {"loss": 1.125, "grad_norm": 1.2071197032928467, "learning_rate": 0.0002, "epoch": 6.275681733283526, "step": 8400}, {"loss": 1.0625, "grad_norm": 1.2653778791427612, "learning_rate": 0.0002, "epoch": 6.283152782966007, "step": 8410}, {"loss": 1.0565, "grad_norm": 1.6128872632980347, "learning_rate": 0.0002, "epoch": 6.290623832648487, "step": 8420}, {"loss": 1.1212, "grad_norm": 1.4993070363998413, "learning_rate": 0.0002, "epoch": 6.298094882330967, "step": 8430}, {"loss": 1.1516, "grad_norm": 1.16339910030365, "learning_rate": 0.0002, "epoch": 6.305565932013448, "step": 8440}, {"loss": 1.0662, "grad_norm": 1.256822943687439, "learning_rate": 0.0002, "epoch": 6.313036981695928, "step": 8450}, {"loss": 1.1566, "grad_norm": 1.1352964639663696, "learning_rate": 0.0002, "epoch": 6.3205080313784086, "step": 8460}, {"loss": 1.1297, "grad_norm": 1.0061070919036865, "learning_rate": 0.0002, "epoch": 6.327979081060889, "step": 8470}, {"loss": 1.0967, "grad_norm": 1.1901768445968628, "learning_rate": 0.0002, "epoch": 6.33545013074337, "step": 8480}, {"loss": 1.1463, "grad_norm": 1.2715139389038086, "learning_rate": 0.0002, "epoch": 6.34292118042585, "step": 8490}, {"loss": 1.2143, "grad_norm": 1.1583346128463745, "learning_rate": 0.0002, "epoch": 6.35039223010833, "step": 8500}, {"loss": 1.1072, "grad_norm": 1.1427477598190308, "learning_rate": 0.0002, "epoch": 6.357863279790811, "step": 8510}, {"loss": 1.1119, "grad_norm": 1.1952263116836548, "learning_rate": 0.0002, "epoch": 6.365334329473291, "step": 8520}, {"loss": 1.0797, "grad_norm": 1.0599623918533325, "learning_rate": 0.0002, "epoch": 6.372805379155771, "step": 8530}, {"loss": 1.1091, "grad_norm": 1.3511574268341064, "learning_rate": 0.0002, "epoch": 6.380276428838251, "step": 8540}, {"loss": 1.1272, "grad_norm": 1.171126127243042, "learning_rate": 0.0002, "epoch": 6.387747478520732, "step": 8550}, {"loss": 1.1615, "grad_norm": 1.285474419593811, "learning_rate": 0.0002, "epoch": 6.395218528203213, "step": 8560}, {"loss": 1.1505, "grad_norm": 0.9751279950141907, "learning_rate": 0.0002, "epoch": 6.402689577885693, "step": 8570}, {"loss": 1.1502, "grad_norm": 1.2194149494171143, "learning_rate": 0.0002, "epoch": 6.410160627568174, "step": 8580}, {"loss": 1.138, "grad_norm": 1.255888819694519, "learning_rate": 0.0002, "epoch": 6.417631677250654, "step": 8590}, {"loss": 1.1308, "grad_norm": 1.1636122465133667, "learning_rate": 0.0002, "epoch": 6.425102726933134, "step": 8600}, {"loss": 1.1398, "grad_norm": 1.0769859552383423, "learning_rate": 0.0002, "epoch": 6.432573776615614, "step": 8610}, {"loss": 1.1183, "grad_norm": 1.151778221130371, "learning_rate": 0.0002, "epoch": 6.440044826298095, "step": 8620}, {"loss": 1.0706, "grad_norm": 1.2749944925308228, "learning_rate": 0.0002, "epoch": 6.447515875980575, "step": 8630}, {"loss": 1.1011, "grad_norm": 1.1925828456878662, "learning_rate": 0.0002, "epoch": 6.454986925663055, "step": 8640}, {"loss": 1.1581, "grad_norm": 1.166107416152954, "learning_rate": 0.0002, "epoch": 6.4624579753455365, "step": 8650}, {"loss": 1.105, "grad_norm": 1.0372248888015747, "learning_rate": 0.0002, "epoch": 6.469929025028017, "step": 8660}, {"loss": 1.1546, "grad_norm": 1.26933753490448, "learning_rate": 0.0002, "epoch": 6.477400074710497, "step": 8670}, {"loss": 1.2362, "grad_norm": 1.2154223918914795, "learning_rate": 0.0002, "epoch": 6.484871124392977, "step": 8680}, {"loss": 1.1096, "grad_norm": 1.09475839138031, "learning_rate": 0.0002, "epoch": 6.492342174075458, "step": 8690}, {"loss": 1.1168, "grad_norm": 1.0763037204742432, "learning_rate": 0.0002, "epoch": 6.499813223757938, "step": 8700}, {"loss": 1.1993, "grad_norm": 1.1882896423339844, "learning_rate": 0.0002, "epoch": 6.507284273440418, "step": 8710}, {"loss": 1.1498, "grad_norm": 1.1662089824676514, "learning_rate": 0.0002, "epoch": 6.514755323122898, "step": 8720}, {"loss": 1.2008, "grad_norm": 1.3259495496749878, "learning_rate": 0.0002, "epoch": 6.522226372805379, "step": 8730}, {"loss": 1.1289, "grad_norm": 1.0858017206192017, "learning_rate": 0.0002, "epoch": 6.5296974224878594, "step": 8740}, {"loss": 1.1335, "grad_norm": 1.240337610244751, "learning_rate": 0.0002, "epoch": 6.53716847217034, "step": 8750}, {"loss": 1.1479, "grad_norm": 1.1381462812423706, "learning_rate": 0.0002, "epoch": 6.544639521852821, "step": 8760}, {"loss": 1.0991, "grad_norm": 1.2220063209533691, "learning_rate": 0.0002, "epoch": 6.552110571535301, "step": 8770}, {"loss": 1.159, "grad_norm": 1.1553083658218384, "learning_rate": 0.0002, "epoch": 6.559581621217781, "step": 8780}, {"loss": 1.0996, "grad_norm": 1.1383219957351685, "learning_rate": 0.0002, "epoch": 6.567052670900262, "step": 8790}, {"loss": 1.1355, "grad_norm": 1.0379676818847656, "learning_rate": 0.0002, "epoch": 6.574523720582742, "step": 8800}, {"loss": 1.1704, "grad_norm": 1.376488447189331, "learning_rate": 0.0002, "epoch": 6.581994770265222, "step": 8810}, {"loss": 1.1265, "grad_norm": 1.1586211919784546, "learning_rate": 0.0002, "epoch": 6.589465819947702, "step": 8820}, {"loss": 1.1904, "grad_norm": 1.28152334690094, "learning_rate": 0.0002, "epoch": 6.596936869630183, "step": 8830}, {"loss": 1.1646, "grad_norm": 1.2656810283660889, "learning_rate": 0.0002, "epoch": 6.6044079193126635, "step": 8840}, {"loss": 1.1865, "grad_norm": 1.0636502504348755, "learning_rate": 0.0002, "epoch": 6.611878968995144, "step": 8850}, {"loss": 1.125, "grad_norm": 1.273239254951477, "learning_rate": 0.0002, "epoch": 6.619350018677624, "step": 8860}, {"loss": 1.1443, "grad_norm": 1.1055482625961304, "learning_rate": 0.0002, "epoch": 6.626821068360105, "step": 8870}, {"loss": 1.0877, "grad_norm": 1.1934176683425903, "learning_rate": 0.0002, "epoch": 6.634292118042585, "step": 8880}, {"loss": 1.194, "grad_norm": 1.2248114347457886, "learning_rate": 0.0002, "epoch": 6.641763167725065, "step": 8890}, {"loss": 1.1609, "grad_norm": 1.1950982809066772, "learning_rate": 0.0002, "epoch": 6.649234217407546, "step": 8900}, {"loss": 1.169, "grad_norm": 1.0821784734725952, "learning_rate": 0.0002, "epoch": 6.656705267090026, "step": 8910}, {"loss": 1.1337, "grad_norm": 1.0062463283538818, "learning_rate": 0.0002, "epoch": 6.664176316772506, "step": 8920}, {"loss": 1.1403, "grad_norm": 1.2373089790344238, "learning_rate": 0.0002, "epoch": 6.671647366454987, "step": 8930}, {"loss": 1.2051, "grad_norm": 1.1821746826171875, "learning_rate": 0.0002, "epoch": 6.6791184161374675, "step": 8940}, {"loss": 1.1214, "grad_norm": 1.2350659370422363, "learning_rate": 0.0002, "epoch": 6.686589465819948, "step": 8950}, {"loss": 1.225, "grad_norm": 1.1012883186340332, "learning_rate": 0.0002, "epoch": 6.694060515502428, "step": 8960}, {"loss": 1.2111, "grad_norm": 1.2008943557739258, "learning_rate": 0.0002, "epoch": 6.701531565184909, "step": 8970}, {"loss": 1.1769, "grad_norm": 1.2355504035949707, "learning_rate": 0.0002, "epoch": 6.709002614867389, "step": 8980}, {"loss": 1.1323, "grad_norm": 1.2367502450942993, "learning_rate": 0.0002, "epoch": 6.716473664549869, "step": 8990}, {"loss": 1.1235, "grad_norm": 1.1075866222381592, "learning_rate": 0.0002, "epoch": 6.723944714232349, "step": 9000}, {"loss": 1.1239, "grad_norm": 1.246480941772461, "learning_rate": 0.0002, "epoch": 6.73141576391483, "step": 9010}, {"loss": 1.2154, "grad_norm": 1.1252824068069458, "learning_rate": 0.0002, "epoch": 6.73888681359731, "step": 9020}, {"loss": 1.1762, "grad_norm": 1.0706887245178223, "learning_rate": 0.0002, "epoch": 6.7463578632797905, "step": 9030}, {"loss": 1.1961, "grad_norm": 1.0874755382537842, "learning_rate": 0.0002, "epoch": 6.7538289129622715, "step": 9040}, {"loss": 1.0889, "grad_norm": 1.121434211730957, "learning_rate": 0.0002, "epoch": 6.761299962644752, "step": 9050}, {"loss": 1.2018, "grad_norm": 1.1517996788024902, "learning_rate": 0.0002, "epoch": 6.768771012327232, "step": 9060}, {"loss": 1.1593, "grad_norm": 1.2484540939331055, "learning_rate": 0.0002, "epoch": 6.776242062009713, "step": 9070}, {"loss": 1.13, "grad_norm": 1.023059368133545, "learning_rate": 0.0002, "epoch": 6.783713111692193, "step": 9080}, {"loss": 1.1929, "grad_norm": 1.1334631443023682, "learning_rate": 0.0002, "epoch": 6.791184161374673, "step": 9090}, {"loss": 1.18, "grad_norm": 1.2991816997528076, "learning_rate": 0.0002, "epoch": 6.798655211057153, "step": 9100}, {"loss": 1.2398, "grad_norm": 1.4147199392318726, "learning_rate": 0.0002, "epoch": 6.806126260739634, "step": 9110}, {"loss": 1.0958, "grad_norm": 1.1353832483291626, "learning_rate": 0.0002, "epoch": 6.813597310422114, "step": 9120}, {"loss": 1.1379, "grad_norm": 1.0332539081573486, "learning_rate": 0.0002, "epoch": 6.8210683601045945, "step": 9130}, {"loss": 1.1652, "grad_norm": 1.2208142280578613, "learning_rate": 0.0002, "epoch": 6.828539409787075, "step": 9140}, {"loss": 1.1463, "grad_norm": 1.3033398389816284, "learning_rate": 0.0002, "epoch": 6.836010459469556, "step": 9150}, {"loss": 1.1834, "grad_norm": 1.2676737308502197, "learning_rate": 0.0002, "epoch": 6.843481509152036, "step": 9160}, {"loss": 1.1786, "grad_norm": 1.1668603420257568, "learning_rate": 0.0002, "epoch": 6.850952558834516, "step": 9170}, {"loss": 1.1801, "grad_norm": 1.1994788646697998, "learning_rate": 0.0002, "epoch": 6.858423608516997, "step": 9180}, {"loss": 1.2131, "grad_norm": 1.231873869895935, "learning_rate": 0.0002, "epoch": 6.865894658199477, "step": 9190}, {"loss": 1.2109, "grad_norm": 0.9981484413146973, "learning_rate": 0.0002, "epoch": 6.873365707881957, "step": 9200}, {"loss": 1.1084, "grad_norm": 1.2799428701400757, "learning_rate": 0.0002, "epoch": 6.880836757564438, "step": 9210}, {"loss": 1.2004, "grad_norm": 1.2042057514190674, "learning_rate": 0.0002, "epoch": 6.888307807246918, "step": 9220}, {"loss": 1.1567, "grad_norm": 1.070420265197754, "learning_rate": 0.0002, "epoch": 6.8957788569293985, "step": 9230}, {"loss": 1.1353, "grad_norm": 1.327160358428955, "learning_rate": 0.0002, "epoch": 6.903249906611879, "step": 9240}, {"loss": 1.1945, "grad_norm": 1.1109007596969604, "learning_rate": 0.0002, "epoch": 6.91072095629436, "step": 9250}, {"loss": 1.1701, "grad_norm": 1.1669930219650269, "learning_rate": 0.0002, "epoch": 6.91819200597684, "step": 9260}, {"loss": 1.1854, "grad_norm": 1.034532904624939, "learning_rate": 0.0002, "epoch": 6.92566305565932, "step": 9270}, {"loss": 1.1712, "grad_norm": 1.1035540103912354, "learning_rate": 0.0002, "epoch": 6.9331341053418, "step": 9280}, {"loss": 1.1767, "grad_norm": 1.366254448890686, "learning_rate": 0.0002, "epoch": 6.940605155024281, "step": 9290}, {"loss": 1.1591, "grad_norm": 1.094214677810669, "learning_rate": 0.0002, "epoch": 6.948076204706761, "step": 9300}, {"loss": 1.18, "grad_norm": 1.131238579750061, "learning_rate": 0.0002, "epoch": 6.955547254389241, "step": 9310}, {"loss": 1.2513, "grad_norm": 1.202369213104248, "learning_rate": 0.0002, "epoch": 6.963018304071722, "step": 9320}, {"loss": 1.1922, "grad_norm": 1.1067225933074951, "learning_rate": 0.0002, "epoch": 6.9704893537542025, "step": 9330}, {"loss": 1.1965, "grad_norm": 1.0258643627166748, "learning_rate": 0.0002, "epoch": 6.977960403436683, "step": 9340}, {"loss": 1.2053, "grad_norm": 1.3311655521392822, "learning_rate": 0.0002, "epoch": 6.985431453119164, "step": 9350}, {"loss": 1.1778, "grad_norm": 1.1245559453964233, "learning_rate": 0.0002, "epoch": 6.992902502801644, "step": 9360}, {"eval_loss": 2.128103017807007, "eval_runtime": 39.1339, "eval_samples_per_second": 13.16, "eval_steps_per_second": 1.661, "epoch": 6.999626447515876, "step": 9369}, {"loss": 1.1782, "grad_norm": 1.0868251323699951, "learning_rate": 0.0002, "epoch": 7.000373552484124, "step": 9370}, {"loss": 1.0133, "grad_norm": 1.5252128839492798, "learning_rate": 0.0002, "epoch": 7.007844602166604, "step": 9380}, {"loss": 0.9364, "grad_norm": 1.1230034828186035, "learning_rate": 0.0002, "epoch": 7.015315651849085, "step": 9390}, {"loss": 0.9702, "grad_norm": 1.275871992111206, "learning_rate": 0.0002, "epoch": 7.022786701531565, "step": 9400}, {"loss": 0.9305, "grad_norm": 1.462963342666626, "learning_rate": 0.0002, "epoch": 7.030257751214045, "step": 9410}, {"loss": 0.9329, "grad_norm": 1.0506054162979126, "learning_rate": 0.0002, "epoch": 7.0377288008965255, "step": 9420}, {"loss": 0.9398, "grad_norm": 1.4315128326416016, "learning_rate": 0.0002, "epoch": 7.0451998505790066, "step": 9430}, {"loss": 0.9086, "grad_norm": 1.5143473148345947, "learning_rate": 0.0002, "epoch": 7.052670900261487, "step": 9440}, {"loss": 0.9712, "grad_norm": 1.2537293434143066, "learning_rate": 0.0002, "epoch": 7.060141949943967, "step": 9450}, {"loss": 0.9591, "grad_norm": 1.36807382106781, "learning_rate": 0.0002, "epoch": 7.067612999626448, "step": 9460}, {"loss": 1.0046, "grad_norm": 1.5365028381347656, "learning_rate": 0.0002, "epoch": 7.075084049308928, "step": 9470}, {"loss": 1.0045, "grad_norm": 1.227250576019287, "learning_rate": 0.0002, "epoch": 7.082555098991408, "step": 9480}, {"loss": 0.9745, "grad_norm": 1.6941372156143188, "learning_rate": 0.0002, "epoch": 7.090026148673888, "step": 9490}, {"loss": 0.9203, "grad_norm": 1.587410569190979, "learning_rate": 0.0002, "epoch": 7.097497198356369, "step": 9500}, {"loss": 0.9713, "grad_norm": 1.481272578239441, "learning_rate": 0.0002, "epoch": 7.104968248038849, "step": 9510}, {"loss": 1.0066, "grad_norm": 1.2331953048706055, "learning_rate": 0.0002, "epoch": 7.1124392977213295, "step": 9520}, {"loss": 0.9422, "grad_norm": 1.6446775197982788, "learning_rate": 0.0002, "epoch": 7.119910347403811, "step": 9530}, {"loss": 0.901, "grad_norm": 1.2055929899215698, "learning_rate": 0.0002, "epoch": 7.127381397086291, "step": 9540}, {"loss": 0.8959, "grad_norm": 1.119033932685852, "learning_rate": 0.0002, "epoch": 7.134852446768771, "step": 9550}, {"loss": 0.9586, "grad_norm": 1.712833046913147, "learning_rate": 0.0002, "epoch": 7.142323496451251, "step": 9560}, {"loss": 0.9326, "grad_norm": 1.2007980346679688, "learning_rate": 0.0002, "epoch": 7.149794546133732, "step": 9570}, {"loss": 0.9926, "grad_norm": 1.3251731395721436, "learning_rate": 0.0002, "epoch": 7.157265595816212, "step": 9580}, {"loss": 0.973, "grad_norm": 1.1897934675216675, "learning_rate": 0.0002, "epoch": 7.164736645498692, "step": 9590}, {"loss": 0.9401, "grad_norm": 1.348583698272705, "learning_rate": 0.0002, "epoch": 7.172207695181173, "step": 9600}, {"loss": 0.9931, "grad_norm": 1.1588937044143677, "learning_rate": 0.0002, "epoch": 7.179678744863653, "step": 9610}, {"loss": 0.9913, "grad_norm": 1.3808276653289795, "learning_rate": 0.0002, "epoch": 7.187149794546134, "step": 9620}, {"loss": 0.9955, "grad_norm": 1.552425503730774, "learning_rate": 0.0002, "epoch": 7.194620844228614, "step": 9630}, {"loss": 1.0255, "grad_norm": 1.3649828433990479, "learning_rate": 0.0002, "epoch": 7.202091893911095, "step": 9640}, {"loss": 0.973, "grad_norm": 1.3196533918380737, "learning_rate": 0.0002, "epoch": 7.209562943593575, "step": 9650}, {"loss": 1.0119, "grad_norm": 1.4874017238616943, "learning_rate": 0.0002, "epoch": 7.217033993276055, "step": 9660}, {"loss": 0.9778, "grad_norm": 1.2448325157165527, "learning_rate": 0.0002, "epoch": 7.224505042958536, "step": 9670}, {"loss": 0.9697, "grad_norm": 1.4631818532943726, "learning_rate": 0.0002, "epoch": 7.231976092641016, "step": 9680}, {"loss": 0.9827, "grad_norm": 1.2041361331939697, "learning_rate": 0.0002, "epoch": 7.239447142323496, "step": 9690}, {"loss": 0.9417, "grad_norm": 1.559156060218811, "learning_rate": 0.0002, "epoch": 7.246918192005976, "step": 9700}, {"loss": 1.0232, "grad_norm": 1.3939464092254639, "learning_rate": 0.0002, "epoch": 7.2543892416884574, "step": 9710}, {"loss": 1.005, "grad_norm": 1.347583293914795, "learning_rate": 0.0002, "epoch": 7.261860291370938, "step": 9720}, {"loss": 0.9256, "grad_norm": 1.302850365638733, "learning_rate": 0.0002, "epoch": 7.269331341053418, "step": 9730}, {"loss": 0.956, "grad_norm": 1.1425062417984009, "learning_rate": 0.0002, "epoch": 7.276802390735899, "step": 9740}, {"loss": 0.9978, "grad_norm": 1.2865869998931885, "learning_rate": 0.0002, "epoch": 7.284273440418379, "step": 9750}, {"loss": 0.9841, "grad_norm": 1.3773187398910522, "learning_rate": 0.0002, "epoch": 7.291744490100859, "step": 9760}, {"loss": 1.0063, "grad_norm": 1.2692701816558838, "learning_rate": 0.0002, "epoch": 7.299215539783339, "step": 9770}, {"loss": 1.0347, "grad_norm": 1.38542902469635, "learning_rate": 0.0002, "epoch": 7.30668658946582, "step": 9780}, {"loss": 0.9606, "grad_norm": 1.2204844951629639, "learning_rate": 0.0002, "epoch": 7.3141576391483, "step": 9790}, {"loss": 1.0225, "grad_norm": 1.4863795042037964, "learning_rate": 0.0002, "epoch": 7.32162868883078, "step": 9800}, {"loss": 0.9933, "grad_norm": 1.2458586692810059, "learning_rate": 0.0002, "epoch": 7.3290997385132615, "step": 9810}, {"loss": 1.0336, "grad_norm": 1.3530622720718384, "learning_rate": 0.0002, "epoch": 7.336570788195742, "step": 9820}, {"loss": 0.9319, "grad_norm": 1.2571991682052612, "learning_rate": 0.0002, "epoch": 7.344041837878222, "step": 9830}, {"loss": 1.0042, "grad_norm": 1.3074439764022827, "learning_rate": 0.0002, "epoch": 7.351512887560702, "step": 9840}, {"loss": 1.0433, "grad_norm": 1.2986950874328613, "learning_rate": 0.0002, "epoch": 7.358983937243183, "step": 9850}, {"loss": 1.0078, "grad_norm": 1.4233403205871582, "learning_rate": 0.0002, "epoch": 7.366454986925663, "step": 9860}, {"loss": 0.9359, "grad_norm": 1.468161702156067, "learning_rate": 0.0002, "epoch": 7.373926036608143, "step": 9870}, {"loss": 1.1074, "grad_norm": 1.354690432548523, "learning_rate": 0.0002, "epoch": 7.381397086290624, "step": 9880}, {"loss": 1.0153, "grad_norm": 1.4891324043273926, "learning_rate": 0.0002, "epoch": 7.388868135973104, "step": 9890}, {"loss": 1.0234, "grad_norm": 1.3470090627670288, "learning_rate": 0.0002, "epoch": 7.3963391856555845, "step": 9900}, {"loss": 1.063, "grad_norm": 1.373061180114746, "learning_rate": 0.0002, "epoch": 7.403810235338065, "step": 9910}, {"loss": 1.0109, "grad_norm": 1.4181641340255737, "learning_rate": 0.0002, "epoch": 7.411281285020546, "step": 9920}, {"loss": 0.9801, "grad_norm": 1.3284671306610107, "learning_rate": 0.0002, "epoch": 7.418752334703026, "step": 9930}, {"loss": 1.0859, "grad_norm": 1.333896517753601, "learning_rate": 0.0002, "epoch": 7.426223384385506, "step": 9940}, {"loss": 1.0056, "grad_norm": 1.6348158121109009, "learning_rate": 0.0002, "epoch": 7.433694434067987, "step": 9950}, {"loss": 1.0173, "grad_norm": 1.364643931388855, "learning_rate": 0.0002, "epoch": 7.441165483750467, "step": 9960}, {"loss": 1.0076, "grad_norm": 1.3974874019622803, "learning_rate": 0.0002, "epoch": 7.448636533432947, "step": 9970}, {"loss": 0.9918, "grad_norm": 1.5207233428955078, "learning_rate": 0.0002, "epoch": 7.456107583115427, "step": 9980}, {"loss": 1.019, "grad_norm": 1.541517734527588, "learning_rate": 0.0002, "epoch": 7.463578632797908, "step": 9990}, {"loss": 0.9904, "grad_norm": 1.3563939332962036, "learning_rate": 0.0002, "epoch": 7.4710496824803885, "step": 10000}, {"loss": 1.0285, "grad_norm": 1.3443987369537354, "learning_rate": 0.0002, "epoch": 7.478520732162869, "step": 10010}, {"loss": 1.0028, "grad_norm": 1.2904508113861084, "learning_rate": 0.0002, "epoch": 7.48599178184535, "step": 10020}, {"loss": 0.9949, "grad_norm": 1.434145450592041, "learning_rate": 0.0002, "epoch": 7.49346283152783, "step": 10030}, {"loss": 1.0837, "grad_norm": 1.4659384489059448, "learning_rate": 0.0002, "epoch": 7.50093388121031, "step": 10040}, {"loss": 1.0063, "grad_norm": 1.3430006504058838, "learning_rate": 0.0002, "epoch": 7.508404930892791, "step": 10050}, {"loss": 1.0168, "grad_norm": 1.3595343828201294, "learning_rate": 0.0002, "epoch": 7.515875980575271, "step": 10060}, {"loss": 0.9609, "grad_norm": 1.7456434965133667, "learning_rate": 0.0002, "epoch": 7.523347030257751, "step": 10070}, {"loss": 1.0843, "grad_norm": 1.329853892326355, "learning_rate": 0.0002, "epoch": 7.530818079940231, "step": 10080}, {"loss": 1.0879, "grad_norm": 1.548466682434082, "learning_rate": 0.0002, "epoch": 7.538289129622712, "step": 10090}, {"loss": 1.0221, "grad_norm": 1.2951644659042358, "learning_rate": 0.0002, "epoch": 7.5457601793051925, "step": 10100}, {"loss": 1.007, "grad_norm": 1.3988010883331299, "learning_rate": 0.0002, "epoch": 7.553231228987673, "step": 10110}, {"loss": 1.0353, "grad_norm": 1.211068868637085, "learning_rate": 0.0002, "epoch": 7.560702278670153, "step": 10120}, {"loss": 0.9962, "grad_norm": 1.2159098386764526, "learning_rate": 0.0002, "epoch": 7.568173328352634, "step": 10130}, {"loss": 0.9928, "grad_norm": 1.3533744812011719, "learning_rate": 0.0002, "epoch": 7.575644378035114, "step": 10140}, {"loss": 1.0386, "grad_norm": 1.3153362274169922, "learning_rate": 0.0002, "epoch": 7.583115427717594, "step": 10150}, {"loss": 1.0017, "grad_norm": 1.535762906074524, "learning_rate": 0.0002, "epoch": 7.590586477400075, "step": 10160}, {"loss": 1.0592, "grad_norm": 1.5531504154205322, "learning_rate": 0.0002, "epoch": 7.598057527082555, "step": 10170}, {"loss": 1.0684, "grad_norm": 1.3588606119155884, "learning_rate": 0.0002, "epoch": 7.605528576765035, "step": 10180}, {"loss": 1.0838, "grad_norm": 1.6648331880569458, "learning_rate": 0.0002, "epoch": 7.6129996264475155, "step": 10190}, {"loss": 1.04, "grad_norm": 1.250205159187317, "learning_rate": 0.0002, "epoch": 7.6204706761299965, "step": 10200}, {"loss": 1.0188, "grad_norm": 1.2442443370819092, "learning_rate": 0.0002, "epoch": 7.627941725812477, "step": 10210}, {"loss": 1.0297, "grad_norm": 1.386197805404663, "learning_rate": 0.0002, "epoch": 7.635412775494957, "step": 10220}, {"loss": 1.0233, "grad_norm": 1.3478381633758545, "learning_rate": 0.0002, "epoch": 7.642883825177438, "step": 10230}, {"loss": 1.0313, "grad_norm": 1.2800627946853638, "learning_rate": 0.0002, "epoch": 7.650354874859918, "step": 10240}, {"loss": 1.0252, "grad_norm": 1.4082499742507935, "learning_rate": 0.0002, "epoch": 7.657825924542398, "step": 10250}, {"loss": 1.0462, "grad_norm": 1.321529746055603, "learning_rate": 0.0002, "epoch": 7.665296974224878, "step": 10260}, {"loss": 1.0727, "grad_norm": 1.4213372468948364, "learning_rate": 0.0002, "epoch": 7.672768023907359, "step": 10270}, {"loss": 0.9938, "grad_norm": 1.5585565567016602, "learning_rate": 0.0002, "epoch": 7.680239073589839, "step": 10280}, {"loss": 1.021, "grad_norm": 1.4025108814239502, "learning_rate": 0.0002, "epoch": 7.6877101232723195, "step": 10290}, {"loss": 1.0136, "grad_norm": 1.344456434249878, "learning_rate": 0.0002, "epoch": 7.6951811729548005, "step": 10300}, {"loss": 1.0157, "grad_norm": 1.4962990283966064, "learning_rate": 0.0002, "epoch": 7.702652222637281, "step": 10310}, {"loss": 1.0621, "grad_norm": 1.4523862600326538, "learning_rate": 0.0002, "epoch": 7.710123272319761, "step": 10320}, {"loss": 1.0413, "grad_norm": 1.401842474937439, "learning_rate": 0.0002, "epoch": 7.717594322002241, "step": 10330}, {"loss": 1.0028, "grad_norm": 1.517730474472046, "learning_rate": 0.0002, "epoch": 7.725065371684722, "step": 10340}, {"loss": 1.0061, "grad_norm": 1.3876111507415771, "learning_rate": 0.0002, "epoch": 7.732536421367202, "step": 10350}, {"loss": 1.0071, "grad_norm": 1.5741353034973145, "learning_rate": 0.0002, "epoch": 7.740007471049682, "step": 10360}, {"loss": 1.0472, "grad_norm": 1.3465591669082642, "learning_rate": 0.0002, "epoch": 7.747478520732163, "step": 10370}, {"loss": 0.9961, "grad_norm": 1.3611412048339844, "learning_rate": 0.0002, "epoch": 7.754949570414643, "step": 10380}, {"loss": 1.0118, "grad_norm": 1.693565011024475, "learning_rate": 0.0002, "epoch": 7.7624206200971235, "step": 10390}, {"loss": 1.0981, "grad_norm": 1.4654128551483154, "learning_rate": 0.0002, "epoch": 7.769891669779604, "step": 10400}, {"loss": 1.029, "grad_norm": 1.417768955230713, "learning_rate": 0.0002, "epoch": 7.777362719462085, "step": 10410}, {"loss": 1.0218, "grad_norm": 1.3143322467803955, "learning_rate": 0.0002, "epoch": 7.784833769144565, "step": 10420}, {"loss": 1.0224, "grad_norm": 1.3467497825622559, "learning_rate": 0.0002, "epoch": 7.792304818827045, "step": 10430}, {"loss": 1.0555, "grad_norm": 1.223697543144226, "learning_rate": 0.0002, "epoch": 7.799775868509526, "step": 10440}, {"loss": 1.0198, "grad_norm": 1.3060917854309082, "learning_rate": 0.0002, "epoch": 7.807246918192006, "step": 10450}, {"loss": 1.0896, "grad_norm": 1.5561134815216064, "learning_rate": 0.0002, "epoch": 7.814717967874486, "step": 10460}, {"loss": 1.0981, "grad_norm": 1.2789647579193115, "learning_rate": 0.0002, "epoch": 7.822189017556966, "step": 10470}, {"loss": 1.0549, "grad_norm": 1.2422796487808228, "learning_rate": 0.0002, "epoch": 7.829660067239447, "step": 10480}, {"loss": 1.0255, "grad_norm": 1.377565622329712, "learning_rate": 0.0002, "epoch": 7.8371311169219275, "step": 10490}, {"loss": 1.0864, "grad_norm": 1.2221037149429321, "learning_rate": 0.0002, "epoch": 7.844602166604408, "step": 10500}, {"loss": 1.0944, "grad_norm": 1.3779186010360718, "learning_rate": 0.0002, "epoch": 7.852073216286888, "step": 10510}, {"loss": 1.0694, "grad_norm": 1.3062539100646973, "learning_rate": 0.0002, "epoch": 7.859544265969369, "step": 10520}, {"loss": 1.048, "grad_norm": 1.4066052436828613, "learning_rate": 0.0002, "epoch": 7.867015315651849, "step": 10530}, {"loss": 1.0674, "grad_norm": 1.6326613426208496, "learning_rate": 0.0002, "epoch": 7.874486365334329, "step": 10540}, {"loss": 1.1237, "grad_norm": 1.1732137203216553, "learning_rate": 0.0002, "epoch": 7.88195741501681, "step": 10550}, {"loss": 1.0682, "grad_norm": 1.303125023841858, "learning_rate": 0.0002, "epoch": 7.88942846469929, "step": 10560}, {"loss": 1.0524, "grad_norm": 1.294990062713623, "learning_rate": 0.0002, "epoch": 7.89689951438177, "step": 10570}, {"loss": 1.0577, "grad_norm": 1.4719983339309692, "learning_rate": 0.0002, "epoch": 7.904370564064251, "step": 10580}, {"loss": 1.0397, "grad_norm": 1.4117742776870728, "learning_rate": 0.0002, "epoch": 7.911841613746732, "step": 10590}, {"loss": 1.1129, "grad_norm": 1.384812355041504, "learning_rate": 0.0002, "epoch": 7.919312663429212, "step": 10600}, {"loss": 0.9994, "grad_norm": 1.5743740797042847, "learning_rate": 0.0002, "epoch": 7.926783713111692, "step": 10610}, {"loss": 1.0168, "grad_norm": 1.2799863815307617, "learning_rate": 0.0002, "epoch": 7.934254762794173, "step": 10620}, {"loss": 1.0893, "grad_norm": 1.4822591543197632, "learning_rate": 0.0002, "epoch": 7.941725812476653, "step": 10630}, {"loss": 1.0362, "grad_norm": 1.4634777307510376, "learning_rate": 0.0002, "epoch": 7.949196862159133, "step": 10640}, {"loss": 1.0116, "grad_norm": 1.5230964422225952, "learning_rate": 0.0002, "epoch": 7.956667911841613, "step": 10650}, {"loss": 1.0582, "grad_norm": 1.3622701168060303, "learning_rate": 0.0002, "epoch": 7.964138961524094, "step": 10660}, {"loss": 1.1127, "grad_norm": 1.2133928537368774, "learning_rate": 0.0002, "epoch": 7.971610011206574, "step": 10670}, {"loss": 1.096, "grad_norm": 1.2852206230163574, "learning_rate": 0.0002, "epoch": 7.9790810608890546, "step": 10680}, {"loss": 1.0286, "grad_norm": 1.243310570716858, "learning_rate": 0.0002, "epoch": 7.986552110571536, "step": 10690}, {"loss": 1.053, "grad_norm": 1.459757924079895, "learning_rate": 0.0002, "epoch": 7.994023160254016, "step": 10700}]}